Suchanek et al v. Sturm Foods, Inc. et al
Filing
324
ORDER DENYING 267 Motion to Exclude the Expert Testimony Of Candace Preston; DENYING 274 Motion to Exclude the Expert Testimony of Jeffrey Andrien; and DENYING 278 Motion to Decertify the Class. The joint submission regarding trial procedures described in the attached Order shall be filed with the Court on or before October 30, 2017. Signed by Judge Nancy J. Rosenstengel on 8/28/2017. (klh2)
IN THE UNITED STATES DISTRICT COURT
FOR THE SOUTHERN DISTRICT OF ILLINOIS
LINDA SUCHANEK,
RICHARD MCMANUS, CAROL CARR,
PAULA GLADSTONE,
EDNA AVAKIAN,
CHARLES CARDILLO, BEN CAPPS,
DEBORAH DIBENEDETTO, and
CAROL RITCHIE,
Plaintiffs,
vs.
STURM FOODS, INC. and
TREEHOUSE FOODS, INC.,
Defendants.
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
Case No. 11-CV-565-NJR-RJD
MEMORANDUM AND ORDER
ROSENSTENGEL, District Judge:
This matter is currently before the Court on (1) a motion filed by Defendants to
exclude the expert testimony of Candace Preston regarding damages pursuant to
Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993), and Federal Rule of Evidence 702
(Doc. 267); (2) a motion filed by Plaintiffs to exclude the expert testimony of Jeffrey
Andrien pursuant to Daubert and Rule 702 (Doc. 274); and (3) a motion for decertification
of the class filed by Defendants (Doc. 278).
For the reasons set forth below, these motions are denied.
PROCEDURAL BACKGROUND
On November 3, 2015, the Court entered an Order granting in part Plaintiffs’
renewed motion for class certification with respect to their claims under the consumer
Page 1 of 49
protection statutes of eight states (Doc. 247). The Court found that the requirements of
Rule 23(a)—numerosity, commonality, typicality, and adequacy—were satisfied
(Doc. 247). The Court further determined that the predominance and superiority
requirements of Rule 23(b)(3) were satisfied (Doc. 247). With respect to predominance,
the Court found, in pertinent part, that damages were susceptible to measurement on a
class-wide basis using the Retail Damages and Price Premium Damages models
presented by Plaintiffs’ expert Candace Preston (Doc. 247). This finding is now the
subject of Defendants’ motion to decertify (see Doc. 278).
Having concluded that Plaintiffs satisfied the requirements of Rule 23(a) and
23(b)(3), the Court certified a “liability” class comprised of all persons who purchased
Grove Square Coffee (“GSC”) “from September 2010 up through the date the case is
certified and notice is disseminated” in Alabama, California, Illinois, New Jersey, New
York, North Carolina, South Carolina, and Tennessee (Doc. 247).1
Following the class certification Order, the parties were permitted to do
additional discovery regarding damages (Doc. 254). Candace Preston submitted a
second supplemental report in March 2016 explaining her Price Premium Damages
model (Doc. 268-1). Defendants retained Jeffrey Andrien to rebut Ms. Preston’s Price
Premium Damages model as well as her Retail Damages model (see Doc. 268-2). While
the parties were conducting the additional discovery, notice went out to the class
(Doc. 254, 260).
Following the notice period and conclusion of discovery, Defendants filed their
“Liability” refers only to the question of whether the GSC packaging was likely to deceive a reasonable
consumer (see Doc. 247, 250). In order to be more clear from here on out, the Court will refer to that
question as the issue of “deception,” not “liability.”
1
Page 2 of 49
second Daubert motion seeking to exclude the expert testimony of Candace Preston
(Doc. 267; see also Doc. 194). Plaintiffs then filed a Daubert motion seeking to exclude the
testimony of Jeffrey Andrien (Doc. 274). Before either Daubert motion was ruled on,
Defendants filed another motion seeking to decertify the class (Doc. 278). A hearing on
these three motions was held on April 27, 2017 (Docs. 310, 311). Following the hearing,
both parties filed responses to the arguments made at the hearing.
As an initial matter, the Court wishes to comment on the quality of the briefing
and arguments advanced by both parties. The moving and opposing papers were
voluminous, but poorly written and poorly researched, and have needlessly complicated
the Court’s task of deciding the motions. To begin with, Defendants’ submissions are a
kitchen-sink collection of arguments that are not clearly delineated, poorly
organized—in fact, they are often split up between two sections of the briefs (the
“background” section and the “argument” section), and are frequently undeveloped.
Simply put, Defendants’ arguments are exceedingly difficult to digest. At times,
Defendants also misrepresented aspects of this case, ignored the Court’s previous
discussions, rehashed issues that were previously rejected, and squabbled over issues of
no consequence.
Plaintiffs’ submissions fare no better. They frequently rely on large block
quotations from previous Orders or case law, but make little to no attempt to connect the
quoted material back to issue at hand. Legal analysis is often absent; instead, they seem
to simply announce their position but make no attempt to flesh out their argument or
explain why their argument is of any significance. Perhaps most troubling is the
Page 3 of 49
complete dearth of supporting legal authority in Plaintiffs’ Daubert motion—none of
their arguments contain a single citation. Moreover, the parties frequently talk past one
another and fail to directly address one another’s arguments. As a result, the Court has
spent an extensive amount of time—too much time, without a doubt—struggling to sort
out and analyze the issues presented. In the future, the Court will be much less tolerant
of poor quality writing and other behavior that it perceives to fall below the minimum
standards expected of an attorney, to waste the Court’s time and resources, and to
needlessly protract the proceedings.
DAUBERT MOTIONS
The admissibility of expert testimony is governed by Federal Rule of Evidence
702 and the Supreme Court’s opinion in Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579
(1993). See Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147 (1999) (holding that the Daubert
analysis applies to all expert testimony under Rule 702, not just scientific testimony).
Under Rule 702 and Daubert, district courts have a “gatekeeping” obligation to ensure
that expert testimony is both relevant and reliable. Kumho Tire, 526 U.S. at 147. This
requires the district court to ensure the following before admitting expert testimony:
First, the expert must be qualified by knowledge, skill, experience,
training, or education; second, the proposed expert testimony must assist
the trier of fact in determining a relevant fact at issue in the case; third, the
expert’s testimony must be based on sufficient facts or data and reliable
principles and methods; and fourth, the expert must have reliably applied
the principles and methods to the facts of the case.
Lees v. Carthage Coll., 714 F.3d 516, 521–22 (7th Cir. 2013) (citing FED. R. EVID. 702 and
Smith v. Ford Motor Co., 215 F.3d 713, 717–19 (7th Cir. 2000)). The party offering the expert
testimony bears the burden of establishing that it meets these admissibility
Page 4 of 49
requirements. Brown v. Burlington N. Santa Fe Ry. Co., 765 F.3d 765, 772 (7th Cir. 2014);
Lewis v. CITGO Petroleum Corp., 561 F.3d 698, 705 (7th Cir. 2009).
Here, neither party challenges the experts’ qualifications or the relevance of the
experts’ opinions in their respective motions (see Docs. 268, 274). Instead, both parties
focus on the reliability of the experts’ opinions (see Docs. 268, 274). “A district court
enjoys broad latitude both in deciding how to determine reliability and in making the
ultimate reliability determination.” Higgins v. Koch Dev. Corp., 794 F.3d 697, 704 (7th Cir.
2015) (quoting Bryant v. City of Chicago, 200 F.3d 1092, 1098 (7th Cir. 2000)). In assessing
the reliability of expert testimony, courts can consider the non-exhaustive list of
guideposts set forth in Daubert: (1) whether the scientific theory can be or has
been tested; (2) whether the theory has been subjected to peer review and publication;
and (3) whether the theory has been generally accepted in the relevant scientific,
technical, or professional community. Am. Honda Motor Co. v. Allen, 600 F.3d 813, 817 (7th
Cir. 2010) (citing Daubert, 509 U.S. at 593–94). The 2000 Advisory Committee’s Notes to
Rule 702 also lists additional factors for gauging expert reliability, including whether:
(1) the testimony relates to matters growing naturally and directly out of research that
was conducted independently from the instant litigation; (2) the expert has unjustifiably
extrapolated from an accepted premise to an unfounded conclusion; (3) the expert has
adequately accounted for obvious alternative explanations; (4) the expert is being as
careful as she would be in her regular professional work outside of paid litigation
consulting; and (5) whether the expert’s field of expertise is known to reach reliable
results for the type of opinion the expert is giving. Am. Honda, 600 F.3d at 817 (quoting
Page 5 of 49
FED. R. EVID. 702, Advisory Committee’s Notes (2000 Amend.)).
These
factors
do
not,
however,
constitute
a
“definitive
checklist
or
test.” Daubert, 509 U.S. at 593. Rather, “[t]he inquiry envisioned by Rule 702 is . . . a
flexible one,” id. at 594, and “the gatekeeping inquiry must be tied to the facts of a
particular case,” Kumho Tire, 526 U.S. at 150 (internal quotation marks omitted).
A. OVERVIEW OF PRESTON AND ANDRIEN’S ANALYSES
Candace Preston is the damages expert hired by Plaintiffs. In her original report
submitted in August 2012, Preston provided, in relevant part, the Retail Damages model
(Doc. 101-12). 2 This model would provide class members with a full refund of the
purchase price, and it reflects Plaintiffs’ theory of liability that GSC had no value to
purchasers (Id.). Preston calculated that consumers spent a total of $6.273 million on GSC
during the class period in the eight states at issue (Doc. 268-1, pp. 3, 8).
In her second supplemental report submitted in March 2016, Preston provided an
alternative damages model: the Price Premium model (Doc. 268-1). This model would
provide class members with a partial refund, and it reflects Plaintiffs’ theory of liability
that consumers paid an inflated price for GSC (Id.). Specifically, the class members
would receive the difference between the inflated retail price they paid for GSC thinking
it was ground coffee and the actual value of GSC had it been marketed truthfully.
Suchanek v. Sturm Foods, Inc., 764 F.3d 750, 760 (7th Cir. 2014).
To determine the average retail price that class members paid for GSC, Candace
Preston obtained data from Information Resources, Inc. (“IRI”) for retail sales of GSC in
The original report also contained the Wholesale Damages model, which reflects the revenue Defendants
earned selling GSC at wholesale prices to retailers. This model was rejected by the Court because it did not
match either of Plaintiffs’ theories of liability (Doc. 247, p. 37).
2
Page 6 of 49
the relevant eight states from 2010 through 2014 (Doc. 268-1, p. 3). From that data,
Preston determined that the average retail price paid by consumers for GSC in the
relevant states was approximately $0.45 per cup (Doc. 268-1, p. 5). Defendants have no
objection to Preston’s methodology for determining the average retail price of GSC or
the $0.45 figure generated using that methodology (see Docs. 268, 278).
Preston then approximated the actual value of GSC by looking at the price per
cup of comparable products on the market whose packaging did not contain the
misrepresentations that were on the GSC packaging (Doc. 268-1, pp. 4, 5). Preston
identified 23 instant coffee products as comparable products and determined that the
average price per cup for those products was approximately $0.06 (Id. at pp. 6, 11). Thus,
she determined that class members paid a price premium of approximately $0.38 per
serving of GSC (Id. at pp. 7, 12). The price premium, multiplied by the number of units
sold, amounts to a total of $5.415 million in damages (Id. at pp. 7, 13).
Defendants retained Jeffrey Andrien to rebut Ms. Preston’s conclusions (see
Doc. 268-2). Andrien opined that the price premium model is the appropriate method for
quantifying damages, not the full refund model (Doc. 268-2). Andrien also believes,
however, that Preston’s application of the price premium methodology is fatally flawed
(Id.) In pertinent part, Andrien believes Preston should have used instant and
microground coffee products as comparables instead of instant-only products and that
she should have accounted for the value of single-serve packaging (Id.). Andrien
conducted his own analysis and determined that consumers are willing to pay more for
single-serve packaging and for products that contain both instant and microground
Page 7 of 49
coffee.
With respect to the single-serve packaging, Andrien determined that two of the
products used by Preston as comparables—Nescafe and Folgers—cost approximately
$0.10 more when it is packaged in single-serving packet than when it is packaged in jars
in bulk (Doc. 268-2, pp. 51–52). Andrien further determined that ground coffee costs
approximately $0.47 more on average when it is packaged in k-cups than when it is
packaged in bulk, tea costs approximately $0.52 more on average, and other non-coffee
products that do not require a filter to prepare (such as hot chocolate and cappuccino)
cost $0.50 more on average (Id. at pp. 21–22, 53–54, 56, 57). Thus, the premiums
associated with k-cup packaging are between $0.47 and $0.52, which is higher than the
$0.45 average retail price of GSC as calculated by Preston. Based on that, Andrien
extrapolated that the actual value of GSC is greater than what consumers paid for it
($0.45) (Id. at p. 23).
As for the value of a microground component, Andrien asserted that “instant and
microground coffee is a distinct segment of the coffee market, and . . . [is] more
expensive than . . . only instant coffee” (Doc. 268-2, p. 23). Andrien analyzed ten
comparator products that all contained instant and microground coffee packaged in
single-serving packets (Id. at p. 28). Eight of the ten products that Andrien used as
comparables were Starbucks VIA products (Id. at pp. 24–28). According to Andrien,
“Starbucks Via has one feature which alone makes it more comparable to GSC than any
of the products selected by Ms. Preston as comparable. Specifically, its formulation is
more similar to the GSC formulation than any of Ms. Preston’s 23 selected products” (Id.
Page 8 of 49
at p. 26). Additionally, like GSC, VIA products come in single-serve packages (although
it is a packet, not a k-cup), and 21 of the 23 products selected by Preston do not (Id. at
p. 26). Andrien also used two other instant and microground coffee products previously
sold in single-serve packets in the United States: Archer Farms Donut Shop and Café
Tastle Brucup Easy Brew (Id. at pp. 27, 28). Andrien calculated that the average price of
these instant and microground coffee products was $0.77 per cup, which once again is
higher than the $0.45 average retail price of GSC as calculated by Preston (Id. at p. 28).
This calculation reinforced Andrien’s previous assertion that the actual value of GSC is
greater than the $0.45 price that consumers paid for it (Id.).
B. DEFENDANTS’ MOTION TO EXCLUDE THE TESTIMONY OF CANDACE PRESTON
As previously indicated, Defendants challenge the reliability of Preston’s Retail
Damages Model and her Price Premium Model (Doc. 268). The Court analyzes the merits
of Defendants’ arguments as to each model in detail in the discussion that follows, but in
short, the Court finds them each unconvincing. The Court has no doubt that Preston
could have done more, and certain aspects of her methodology may be vulnerable on
cross-examination, but Defendants have not given the Court any reason to believe that
her damages models are so patently unreliable that they are inadmissible under Rule 702
and Daubert.
1. Retail Damages
The Retail Damages model, which provides consumers with a full refund of the
purchase price of GSC, is based on the notion that the product was worthless to
consumers and they would not have purchased GSC if not for the deceptive packaging
(Doc. 247, pp. 19, 35). Defendants argue that the Retail Damages model is “legally and
Page 9 of 49
factually insufficient” (Doc. 268, p. 23).
With respect to the legal validity of the damages model, Defendants once again
suggest that a full refund is not an appropriate model of damages in a case where a food
and beverage product is mislabeled (Doc. 268, pp. 12–13, 14, 23–24; see also Docs. 192,
247). In the class certification Order, however, the Court told Defendants that this
argument “was not a good fit for a Daubert motion,” and was more appropriately
considered with respect to the issue of predominance under Rule 23(b)(3) (Doc. 247,
p. 12). At any rate, the Court rejected the argument, finding that the four cases cited by
Defendants in support of their argument were “of very little persuasive value” (Doc. 247,
pp. 35–36; Doc. 268, p. 23).
Despite the Court’s previous statements, Defendants once again advance this
argument in a Daubert motion—but do not mention it in their motion for decertification.
Even more unbelieveable, Defendants cite to the same four cases the Court previously
found unpersuasive but do not offer any new analysis or explanation as to why the
Court should revisit this issue (see Doc. 268). Therefore, this argument can be summarily
denied as a basis for excluding Candace Preston’s expert testimony under Daubert.
That being said, the Court believes it is important to once again explain why a
full-refund model of damages may be appropriate in this matter. To begin, it is true that
the full refund model of damages is often rejected in cases involving mislabeled food
and beverages. A review of these cases demonstrates that they typically involve a
product label that does not misrepresent the entire essence of the product but instead
falsely claims the product possesses a particular premium quality. For example, in
Page 10 of 49
Khasin v. R. C. Bigelow, Inc., the green tea products at issue were, in fact, green tea, but
they did not contain the “powerful” or “healthy” antioxidants that the label claimed.
No. 12-CV-02204-WHO, 2016 WL 1213767 (N.D. Cal. Mar. 29, 2016). As another example,
in Werdebaugh v. Blue Diamond Growers, the almond milk products at issue were, in fact,
almond milk, but they were not “all natural” like the label claimed, because they
contained synthetic ingredients. No. 12-CV-02724-LHK, 2014 WL 7148923 (N.D. Cal.
Dec. 15, 2014). The usual story in these cases is that people purchased and consumed the
product (often repeatedly) and were generally satisfied until they later realized that the
product was missing the particular premium quality advertised on the label.
Furthermore, the premium quality was often not the driving force behind every
consumer’s purchase. In other words, some consumers would have purchased the
product for reasons unrelated to the promised premium quality, such as taste or brand
loyalty. Consequently, courts have held that the full refund model is not appropriate in
these cases, because the model fails to take into account that consumers still received
some benefit when they consumed the product—even though it did not have the
advertised premium quality—and it also fails to take into account that consumers may
have still purchased the product had they known the truth about the product.3
See, e.g., Khasin, 2016 WL 1213767, at *3 (in case where plaintiff alleged R.C. Bigelow green tea products
misleadingly claimed to contain “powerful antioxidants” full refund was not appropriate because it was
“too implausible to accept” that “consumers gain[ed] no benefit in the form of enjoyment, nutrition,
caffeine intake, or hydration from consuming the teas,” and the plaintiff himself admitted that he like the
taste of Bigelow teas and preferred it to Lipton); In re POM Wonderful, Case No. ML 10-02199(RZx), 2014
WL 1225184, at *11 (C.D. Cal. Mar. 25, 2014) and 2012 WL 4490860 (C.D. Cal. Sept. 28, 2012) (finding full
refund model invalid in case where Pom juices misleadingly claimed to provide health benefits because,
even though there was evidence this claim was the motivating factor behind consumers’ purchase of the
juice, plaintiffs could not “plausibly contend” that not a single consumer received a single benefit, be it
hydration, flavor, energy, or anything else of value, from the juice); Buetow v. A.L.S. Enters., Inc., 259 F.R.D.
187, 192 & n.4 (D. Minn. 2009) (finding no factual support for full refund model because, even if the
clothing at issue did not have the advertised odor-eliminating properties, “many class members may have
3
Page 11 of 49
This case is different. GSC wasn’t simply missing a particular premium quality
advertised on the package. Instead, the package misrepresented the very essence of what
was being purchased. The evidence also does not suggest that consumers were generally
satisfied with GSC and continued purchasing it until they realized that it was instant
coffee (see Doc. 247, p. 5 (citing Docs. 221, 100-12, 101-2, 101-9)). Rather, it suggests that
consumers were immediately displeased with the product after they made their first cup
of GSC, threw away the remainder of the product, and never purchased it again (see
Doc. 221).4 The evidence also suggests that GSC’s deceptive packaging, which implied
that GSC was ground coffee, was the driving force behind every consumer’s purchase.
While consumers may have ultimately chosen to buy GSC because they wanted to try
wanted this clothing for other reasons in addition to those pertaining to odor-elimination, such as warmth,
appearance, and waterproofing” and therefore the clothing “would likely have value outside of its
odor-eliminating capabilities”). See also Weiner v. Snapple Beverage Corp., No. 07 CIV. 8742 DLC, 2010 WL
3119452, at *2, 3 (S.D.N.Y. Aug. 5, 2010) (plaintiffs in putative class action alleging Snapple beverages
misleadingly claimed to be “All Natural” admitted that the “All Natural” label “was not the deciding
factor “in their purchasing decisions and they would have bought Snapple over other beverages even if it
was not labeled “All-Natural”).
Complaints from consumers to the company include statements such as GSC “is the worse [sic] coffee I
have ever tried to drink. (Both my wife and I poured it down the drain.) Total crap.”; “[W]e are very
disappointed, and now have 16 cups that we will not use.”; “I have never been so disappointed in any
other coffee like this one. It does not work well in the Keurig and tastes terrible. . . . I will warn everyone I
know not to buy this brand.”; “[I]t is the worst thing ever as far as coffee goes . . . I feel I’ve been swindled
and can only throw it out.”; “I recently purchased this coffee and wanted to tell you how disgusting it is.”;
“This coffee was AWFUL. How do you get by selling instant coffee under the guise of a K-cup????”; “All
you did was package INSTANT COFFEE!!!!! How dare you??? With the economy the way it is, a cup of
coffee from the Keurig is a treat. Generic instant coffee tastes better than this.”’; “[T]he flavor is
horrible. . . .Can you tell me how I can get a refund of this item because it is not usable.”; “We have become
accustomed to the excellent coffees for the Keurig, and yours was frankly awful. It tasted like a bad cup of
cheap instant coffee.”; “The coffee in the K-cup replica is nothing but instant coffee. . . . [It] was not much
better than dirty water. . . . I wasted $8.00 of my good earned money and now have to go purchase a non
replica of K-cup as I don’t drink instant coffee.”; “Instant coffee makes me sick, so I am going to have to
throw it out.”; “I’m sorry to say it’s the worst coffee I’ve ever had in my entire life it tasted terrible. I cannot
purchase this coffee again.”; “To say that it is worse than any generic coffee or instant coffee I ever tasted
would not be overstated. We regret purchasing this item and feel it was a waste of money.”; “This is
simply the worst tasting coffee I have experienced. . . . Be ashamed you cheated me out of $8.00.”
(Doc. 221).
4
Page 12 of 49
something new or because of its cheaper price, the only reason they ever considered
purchasing it in the first place was because they thought it was ground coffee. Had
consumers known that it was overwhelmingly instant coffee, they never would have
considered purchasing it.5 Consequently, Plaintiffs have marshaled evidence sufficient
to plausibly suggest that no one would have purchased GSC had they known it was
instant coffee and that no consumers received any benefit from the product.6 It is up to a
jury, however, to ultimately interpret the evidence and determine whether Plaintiffs
successfully made that showing.
As for the factual validity of the damages model, Defendants present an
argument that is a slightly refashioned version of the one that was already rejected by
this Court in the class certification Order. The first time around, Defendants argued that
the Retail Damages models should be stricken because it was based on an unreliable
As noted in the class certification Order, there is evidence that “the only consumers who would
potentially purchase GSC were those who owned a Keurig machine (or who were buying it for someone
else who owned a Keurig machine)” and that “when it came to coffee, Keurig machine owners wanted to
brew only premium, fresh, ground coffee” so Defendants “went to great lengths to disguise the fact that
GSC was not premium, fresh, ground coffee.” (Doc. 247, p. 36). See also Suchanek v. Sturm Foods, Inc., 764
F.3d 750, 753 (7th Cir. 2014) (finding that instant coffee “is not the kind of premium product that Keurig
customers expected, as Sturm’s marketing surveys confirmed. Indeed, Sturm’s consultants warned that
‘use of the word ‘instant’ is a real nono’ and should be avoided ‘if at all possible’ in marketing the GSC
product to the only people who would buy a KCup: Keurig machine owners.”) Customers indicated in
their complaints to the company that they felt disappointed, dissatisfied, displeased, disgusted, swindled,
robbed, cheated, ripped off, duped, and misled (Doc. 247, p. 5). Others said GSC was a hoax, deceptive, an
absolute fraud, a rip off, a sad joke, a gross misrepresentation, a clearly substandard instant coffee
disguised as a Keurig k-cup, and a waste of money (Id.).
5
See Makaeff v. Trump Univ., 309 F.R.D. 631, 636–640 (S.D. Cal. Sept. 18, 2015) (providing extensive
explanation on why full refund is appropriate in situations where consumers are deprived of the essence
of what they were promised). See also In re Morning Song Bird Food Litig., No. 12CV01592 JAH-AGS, 2017
WL 1191485, at *14 (S.D. Cal. Mar. 31, 2017) (approving full refund damages model where plaintiff alleged
no class members would have purchased the bird food if they knew it contained pesticides that made it
poisonous for birds even though defendants argued the product was not worthless and was safe for birds);
Mullins v. Premier Nutrition Corp., No. 13-CV-01271-RS, 2016 WL 1535057, at *6 (N.D. Cal. Apr. 15, 2016)
(accepting full refund damages model where plaintiff produced “some evidence” that no class members
would have bought Joint Juice had they known it would not benefit their joints, even though defendants
argued the juice had other benefits such as hydration, vitamins, antioxidants, taste, etc.).
6
Page 13 of 49
assumption that “if the product [GSC] had been marketed truthfully, consumers would
not have purchased the product at all” (Doc. 247, p. 12; Doc. 195, pp. 16–17). The Court
rejected this argument, finding the assumption underlying the Retail Damages model
was very clearly supported by the evidence (Doc. 247, pp. 12–13). This time around,
Defendants once again attack the assumption that GSC had no value. They argue the
Retail Damages models should be stricken because the assumption originated with
Plaintiffs’ attorneys, not Preston (Doc. 268, p. 23). Defendants further claim this
assumption is erroneous because the fact that “someone would not have purchased a
product if they were not misled does not mean they did not receive value” (Id.).
Defendants also claim the assumption is erroneous because there is evidence that the
ingredients in GSC and its packaging did, in fact, have value (Id. at p. 24).
The Court is wholly unconvinced by these arguments. First, the Court cannot
discern any reason why it matters that Plaintiffs’ counsel told Preston to assume that
GSC had no value. It is not as though Preston blindly accepted some baseless or
unfounded assumption. As discussed above, there is certainly evidence in the record
that supports the assumption that GSC had no value. And, as discussed in the class
certification Order, Candace Preston reviewed some of that evidence in preparing her
Retail Damages model (Doc. 247, pp. 12–13). So to the extent Preston was fed the
assumption that GSC had no value, it appears to the Court that she sufficiently verified
that the assumption was based on evidence. Consequently, the Court concludes that, if
anything, this criticism goes to the weight and credibility of Preston’s opinion, not its
admissibility. See Manpower, Inc. v. Ins. Co. of Pennsylvania, 732 F.3d 796, 808 (7th Cir.
Page 14 of 49
2013) (“The reliability of data and assumptions used in applying a methodology is tested
by the adversarial process and determined by the jury . . . .”) (citing Tuf Racing Products v.
Am. Suzuki Motor Corp., 223 F.3d 585, 591 (7th Cir. 2000)); Smith v. Ford Motor Co., 215 F.3d
713, 718 (7th Cir. 2000) (“The soundness of the factual underpinnings of the expert’s
analysis and the correctness of the expert’s conclusions based on that analysis are factual
matters to be determined by the trier of fact [.]”).
Second, the Court does not believe that evidence regarding value of the
ingredients and packaging of GSC has any bearing on the assumption that GSC had no
value. Of course the ingredients and the packaging of GSC were worth something—the
ingredients and packaging of any product always cost money and are never free. For
example, even a probiotic that provides no digestive benefits, grass seed that doesn’t
grow grass, bird seed that is unfit for consumption because it contains pesticides, and a
drinkable joint supplement that does not provide any joint health benefits, all contain
ingredients that have some monetary value. But the courts nevertheless approved a full
refund damages model in those cases.7 That’s because the full refund model does not
focus on the “value” of the product from the company’s perspective, such as the cost of
See Rikos v. Procter & Gamble Co., 799 F.3d 497, 523 (6th Cir. 2015), cert. denied, 136 S. Ct. 1493 (2016)
(approving full refund damages model where plaintiffs alleged probiotic was ineffective and did not
promote digestive health benefits for anyone because “there is no reason to buy [the product] except for its
purported digestive benefits”); In re Scotts EZ Seed Litig., 304 F.R.D. 397, 412 (S.D.N.Y. 2015) (approving full
refund damages model where consumers alleged that Scotts Turf Builder EZ Seed, an all-in-one-product
containing premium grass seed, mulch, and fertilizer, was defective and did not grow grass); In re Morning
Song Bird Food Litig., No. 12CV01592 JAH-AGS, 2017 WL 1191485, at *14 (S.D. Cal. Mar. 31, 2017)
(approving full refund damages model where plaintiff alleged no class members would have purchased
the bird food if they knew it contained pesticides that made it poisonous for birds even though defendants
argued the product was not worthless and was safe for birds); Mullins v. Premier Nutrition Corp.,
No. 13-CV-01271-RS, 2016 WL 1535057, at *6 (N.D. Cal. Apr. 15, 2016) (accepting full refund damages
model where plaintiff produced “some evidence” that no class members would have bought Joint Juice
had they known it would not benefit their joints, even though defendants argued the juice had other
benefits such as hydration, vitamins, antioxidants, taste, etc.).
7
Page 15 of 49
raw materials, of manufacturing, or of marketing the product. Instead, the focus is on the
retail “value” of the product from the consumers’ perspective, meaning whether anyone
be willing to buy the product (or what price people would be willing to pay) if they
knew they truth about the product. Contrary to Defendants’ assertion, pertinent case law
illustrates that the mere fact that the components or the packaging of a product have
value does not cut the legs out from under the assumption that a product is valueless or
the opinion that a full refund is an appropriate measure of damages.
For these reasons, the Court concludes that Defendants’ criticisms regarding the
factual basis of the Retail Damages model do not provide any grounds for finding
Preston’s opinion unreliable and inadmissible. The Court maintains its previous
conclusion that Candace Preston’s testimony regarding the Retail Damages should not
be excluded under Daubert or Rule 702.
2. Price Premium
The Price Premium model, which provides consumers with a partial refund of the
price they paid for GSC, is based on the notion that consumers paid an inflated price for
GSC. In other words, Plaintiffs believe that instant coffee is worth less than ground
coffee, and therefore the actual value of GSC is substantially less than what consumers
paid for it. The price premium was computed by taking “the difference between the
actual value of the package [consumers] purchased” and “the inflated price [consumers]
paid” thinking the product was as advertised. Suchanek v. Sturm Foods, Inc., 764 F.3d 750,
760 (7th Cir. 2014). In order to determine the “actual value” of GSC, Preston identified
comparable instant coffee products sold by competitors during the class period and
Page 16 of 49
calculated the average price of a serving of instant coffee (see Doc. 268-1). As Defendants’
expert Jeffrey Andrien put it, “[t]he intuition underlying a comparables analysis is that
that [sic] the comparable products share features and attributes so closely aligned with
the subject product that the value of the comparables can be used as a proxy for the
value of the subject product” (Doc. 268-2, p. 16). Andrien admitted that the use of
comparables to assess the value of a product is “a well-established and commonly used
methodology” (Doc. 268-2, p. 14). Defendants believe that Preston’s application of the
comparables methodology was highly flawed, however, and thus inadmissible under
Daubert for two reasons (see Doc. 268).
For their first argument, Defendants set the stage by noting that under Comcast
Corp. v. Behrend, 133 S. Ct. 1426 (2013), “class action plaintiffs must provide a damages
theory that measures only those damages attributable to the alleged misconduct and ‘[is]
consistent with [the plaintiff’s] liability case’” (Doc. 268, p. 15). Defendants go on to
argue that the Price Premium model is inadmissible under Daubert because it does not
“calculate the price premium due solely to the allegedly misleading label statements or
omissions,” and “it is not tied to the alleged wrongdoing” (Doc. 268, pp. 15, 16). For
instance, Defendants claim the model does not isolate the price premium associated with
the alleged misleading acts because Preston failed to distinguish between Plaintiffs’ two
theories of liability, failed to control for differences between GSC and the comparable
products that may contribute to pricing, and failed to consider the supply-side of
valuation (Doc. 268, pp. 15–18).
Defendants’ mention of Comcast, and their use of phrases and terminology
Page 17 of 49
associated with the requirements of Comcast, suggests that the issues they identified bear
on Rule 23(b)(3)’s predominance requirement and the propriety of class certification, not
the reliability of Preston’s methodology and the admissibility of the opinion under
Daubert and Rule 702. This is reinforced by the three cases Defendants cited to: In re POM
Wonderful, No. 10-cv-2199, 2014 WL 1225184 (C.D. Cal. Mar. 25, 2014); Lanovaz v.
Twinings North America, No. 12-cv-2646, 2014 WL 1652338 (C.D. Cal. July 24, 2014); and
Algarin v. Maybelline, 300 F.R.D. 444 (S.D. Cal. 2014)). Each of these cases discussed the
sufficiency of the damages model in the context of class certification and predominance
under Rule 23(b)(3). Therefore, these cases do not suggest Preston’s opinion is
inadmissible under Daubert; instead, they suggest that the issues identified by
Defendants should be evaluated under Comcast in the context of the motion for
decertification.
Consequently, the Court finds that Defendants’ first argument provides no basis
for excluding Preston’s Price Premium model under Daubert or Rule 702, and defers its
discussion of this argument and its subparts until later in this Order when it addresses
the motion for decertification. (Conveniently, Defendants repeat this same exact
argument in their motion for decertification (see Doc. 278)). See In re Fluidmaster, Inc.,
Water Connector Components Prod. Liab. Litig., No. 14-CV-5696, 2017 WL 1196990, at *28
(N.D. Ill. Mar. 31, 2017) (declining to consider arguments based on the requirements of
Comcast
in
the
context
of
a
Daubert
motion);
Dzielak v. Whirlpool Corp.,
No. CV2120089KMJBC, 2017 WL 1034197, at *13 (D.N.J. Mar. 17, 2017) (same, and also
discussing the difference between Comcast arguments and Daubert arguments); In re
Page 18 of 49
ConAgra Foods, Inc., 90 F. Supp. 3d 919, 946 (C.D. Cal. 2015) (same).
As for their second argument, Defendants contend that Preston’s application of
the comparables methodology is “so highly flawed” that the Price Premium model must
be excluded. Specifically, Defendants claim that Preston used inappropriate comparable
products by including only “low cost” instant coffees and excluding other instant and
microground coffees, particularly the Starbucks VIA products (see Doc. 268, pp. 11–12,
15, 18–20). Defendants argue that Preston’s comparables methodology was further
flawed because she failed to account for differences between GSC and the comparable
products she chose that may provide value to consumers, namely the single-serve
packaging and the inclusion of microground coffee (Doc. 268, pp. 11–12, 21–22).
For her part, Candace Preston explained that there is no product truly comparable
to GSC because no other companies have ever attempted to sell instant coffee, or instant
and microground coffee, in a k-cup (Doc. 268-1, p. 4; Doc. 272-3, p. 8). She chose to use
instant-only products as comparables without regard to how they were packaged
because she did not believe the microground component or the single-serve packaging
added any value to GSC (Doc. 272-3, pp. 5, 7, 9). In particular, she said the microground
coffee only made the product clumpy, which consumers complained about, and
therefore it detracted from the value of GSC rather than adding to it (Id. at p. 5). As for
the packaging, Preston said there is no added convenience with a k-cup of instant coffee
over a jar of instant coffee or a single-serve packet because neither of those products
requires the use of a special machine to prepare the product (Id. at pp. 5, 7, 9).
The Court finds that purported deficiencies in the Price Premium model
Page 19 of 49
identified by Defendants go to the weight and credibility of Preston’s opinion, not its
admissibility. Preston had her reasons for conducting her Price Premium analysis in the
manner that she did, and it is up to a jury, not the Court, to decide whether to credit
those reasons. Defendants will have the opportunity to criticize Preston’s choice of
comparables and the variables she included in her methodology through “[v]igorous
cross-examination . . . [and] presentation of contrary evidence.” Daubert v. Merrell Dow
Pharm., Inc., 509 U.S. 579, 596 (1993). See Smith v. Ford Motor Co., 215 F.3d 713, 718 (7th
Cir. 2000) (“The question of whether the expert is credible or whether his or her theories
are correct given the circumstances of a particular case is a factual one best left for the
jury to determine after opposing counsel has been provided the opportunity to
cross-examine the expert regarding his conclusions and the facts on which they are
based.” (citing Walker v. Soo Line R.R. Co., 208 F.3d 581, 590 (7th Cir. 2000))).
In sum, Defendants have failed to convince the Court that Candace Preston’s
Retail Damages analysis or her Price Premium Damages analysis are so unreliable that
they are inadmissible under Daubert and Rule 702. Consequently, Defendants’ motion to
exclude Preston’s testimony is denied.
C. PLAINTIFFS’ MOTION TO EXCLUDE THE TESTIMONY OF JEFFREY ANDRIEN8
Although Plaintiffs ask the Court to strike Jeffrey Andrien’s entire report
(Doc. 274, p. 10), Plaintiffs’ arguments actually pertain only to one portion of the report:
Andrien’s opinion that instant and microground products—particularly Starbucks Via,
Archer Farms Donut Shop, and Café Tastle Brucup Easy Brew—were the appropriate
Defendants ask the Court to summarily deny Plaintiffs’ motion because it is untimely (Doc. 276). The
Court rejected this argument at the hearing and indicated that it considered Plaintiffs’ motion timely
(Doc. 310).
8
Page 20 of 49
comparables to use, his related analysis that these products cost an average of $0.77 per
cup, and his conclusion that GSC was worth more than its actual cost of $0.45 per
serving (see Doc. 274). Plaintiffs argue that Andrien’s analysis is unreliable for the
following five reasons: (1) it ignores that Defendants abandoned their project to develop
a product to directly compete with VIA; (2) it is premised on the unsubstantiated belief
that the VIA products were similar in composition to GSC; (3) Andrien did not read and
was unaware of the legal rulings, factual findings, and evidentiary record in this case;
(4) two of Andrien’s comparator products were not available during the class period;
and (5) it leads to absurd conclusions (Doc. 268). A couple of these arguments, if
thoroughly developed and properly supported, potentially have merit. But in their
current state, they are simply too cursory to justify striking any portion of Andrien’s
testimony.
1. Starbucks VIA as a Comparable Product
Plaintiffs’ first two arguments relate to Andrien’s opinion that Starbucks VIA is
an appropriate comparable and that Candace Preston should have included it in her
analysis (Doc. 268-2, pp. 23–28).
Plaintiffs first argue that this opinion is unreliable because Defendants
abandoned their project to manufacture a microground and instant coffee product in
single-serving packets to directly compete with VIA; Defendants instead concentrated
solely on producing GSC in a Keurig-compatible format (Doc. 274, pp. 3–5). Plaintiffs
also argue that Andrien’s opinion is unreliable because Andrien has no direct
knowledge of VIA’s composition, meaning the amount of instant coffee versus the
Page 21 of 49
amount of microground coffee in VIA products (Doc. 274, pp. 5–6).9
In response, Defendants assert that their decision not to make an instant and
microground product in a single-serve packet in no way shows that the composition of
GSC is dissimilar to the composition of VIA (Doc. 276, p. 7). They claim the coffee
contained in GSC could still be compositionally similar to VIA—even though it was
packaged in a k-cup instead of a single-serve packet (see Doc. 276, p. 7). As for Andrien’s
knowledge about the composition of VIA, Defendants assert that Andrien “relie[d] on
the unrebutted testimony of one of the formulators of the GSC that it was developed to
be similar in composition to VIA” (Doc. 279, p. 7). While Defendants repeatedly asserted
that it was “undisputed” that VIA and GSC were compositionally similar, they never
offered any actual proof regarding the formulation of VIA (see, e.g., Doc. 268, p. 11; Doc.
276, pp. 6, 7)—that is until their briefing submitted after the hearing. In this brief, they
once again claimed that “VIA’s composition is undisputed: 97% soluble/instant and 3%
microground,” but this time they cited to a document at docket entry 300 (Doc. 312, p. 3).
That document is an internal memo from Sturm Foods in which Samantha Peskie writes
“Starbucks via is 3.3 grams of product which according to analysis is 0.1 grams of
microground coffee and 3.2 grams of a highly concentrated spray dried” (Doc. 312, p. 3
(citing to Doc. 300)).10
The Court notes that Plaintiffs’ argument also appears to apply to Andrien’s use of Archer Farms Donut
Shop and Café Tastle Brucup Easy Brew as comparables. The Court knows of no evidence in the record
regarding the composition of the Archer Farms product or the Café Tastle product. But given that neither
party addresses this issue, the Court also declines to address it.
9
The Court does not know what to make of the sudden appearance of this document, especially since
Defendants say very little about it (see Doc. 312). They do not indicate why it was not provided sooner.
They do not provide any context for the document or any explanation about the analysis that was
purportedly conducted. They do not indicate whether Samantha Peskie provided any testimony regarding
10
Page 22 of 49
Just like the Court concluded with respect to Candace Preston’s choice of
comparables, the Court believes that Plaintiffs’ criticisms regarding Andrien’s choice of
comparables go to the weight and credibility of his opinion, not its admissibility.
Plaintiffs are essentially arguing that Andrien ignored pertinent information, or did not
have adequate information, prior to opining that Starbucks VIA is an appropriate
comparable. These criticisms bear on “[t]he soundness of the factual underpinnings” and
the correctness of his opinion regarding which products are appropriate comparables.
Smith v. Ford Motor Co., 215 F.3d 713, 718 (7th Cir. 2000). The Court cannot decide these
factual matters on a Daubert motion. It is up to the jury to decide whether Defendants
knew the composition of VIA and whether VIA is more comparable to GSC than
instant-only products.
2. Archer Farms and Café Tastle as Comparable Products
Turning to Plaintiffs’ fourth argument, they challenge Andrien’s inclusion of
Archer Farms Donut Shop and Café Tastle Brucup Easy Brew as comparable products
because they were not sold during the class period (Doc. 274, pp. 8–9). Furthermore, the
Café Tastle product was only available on the internet and not sold in retail
establishments (Doc. 274, p. 9). While Plaintiffs fail to indicate what point they’re trying
to make (see Doc. 274), the Court believes they are questioning the extent to which the
price of products sold in 2015 and 2016 is indicative of the price GSC could have been
sold for between September 2010 and September 2014 if it had been marketed truthfully.
the document at her deposition. Defendants also do not make any attempt to reconcile this document with
Ken Noble and Harry Overly’s testimony that no one at Sturm ever determined what percentage of
microground coffee was contained in the VIA products (Doc. 108-2, p. 12; Doc. 108-3, p. 12). Additionally,
given the late submission of this document, Plaintiffs have not had an opportunity to provide their input
on the meaning or significance of the document.
Page 23 of 49
In response, Defendants provide little to nothing in the way of an explanation as
to why it was appropriate for Andrien to include the Archer Farms and Café Tastle
products in his analysis (see Doc. 276, p. 8). They state:
At least one of these products, Archer Farms, was sold during at least one
year, 2015, that GSC was sold. Mr. Andrien testified that whether the sales
overlap was during the “class period” was irrelevant, and Ms. Preston
testified she had no opinion of whether there would have been a price
difference between Archer Farms and GSC. Plaintiffs do not even attempt
to show how the absence of sales of a particular instant and microground
product during the class period renders Andrien’s conclusions unreliable
or unsupported and Plaintiffs cite no case or other authority that an expert
cannot rely on economic indicators outside of the class period.
(Doc. 276, p. 8).
The Court first notes that although Plaintiffs’ argument is indeed undeveloped
and unsupported, it is not their duty to establish that Andrien’s testimony is
unreliable—it is Defendants’ duty to establish its reliability. In other words, the onus
was on Defendants to establish that Andrien could rely on economic indicators outside
of the class period; it was not on Plaintiffs to establish that he couldn’t. The explanation
provided by Defendants is hardly convincing. The simple fact that both GSC and Archer
Farms were sold at some point in 2015 does not shed any light on how the price of the
Archer Farms product in 2015 is a reliable and valid indicator of what GSC could have
sold for one to four years earlier. And Andrien’s testimony does not provide the needed
connection. He simply repeated the same assertion (in a rather combative and obstinate
fashion) that it is irrelevant to whether GSC, the Archer Farms product, and the Café
Tastle product were sold during the same time period (see Doc. 320-1, pp. 22–23). The
only explanation he gave was that the purpose of his analysis was to determine whether
Page 24 of 49
consumers were willing to pay a premium for instant and microground coffee above
instant bulk coffee (Id.). That’s all well and good, but once again, it is not self-evident
how the existence of such a premium in 2015 and 2016 establishes that the same (or
similar) premium would have existed years earlier. Furthermore, the fact that Candace
Preston did not investigate and had no opinion on whether the premium would have
changed is illustrative of nothing. In other words, her lack of an opinion does not
somehow establish that the premium remained the same (Doc. 268-4, p. 10).
Given the shoddiness of both parties’ arguments, however, the Court is unable to
say with any degree of certainty whether it would be legally proper or warranted under
the circumstances to determine whether or not the Archer Farms and Café Tastle
products are appropriate comparables. Consequently, the Court will stick to its position
that the purported deficiencies in Andrien’s choice of comparables is a problem of
credibility, not admissibility. Plaintiffs will have the opportunity to criticize Andrien’s
choice of comparables on cross-examination, and the jury will ultimately have to decide
whether instant-only or instant and microground products are the most appropriate
comparables.
3. Unfamiliarity with Rulings and Evidence and Absurdity of Conclusions
In their third argument, Plaintiffs point out that Andrien never read the Seventh
Circuit’s opinion and was thus unaware of their “guidance on a methodology to
compute a price premium” (Doc. 274, p. 7). Plaintiffs also claim that Andrien “seemed
unaware of this Court’s factual findings regarding the potential viability of a full refund
method” (Id. at p. 8). Andrien was also unaware of the Seventh Circuit’s “factual
Page 25 of 49
findings” and the “voluminous record” regarding the public response to GSC, which
Plaintiffs claim totally undermine Andrien’s conclusion that GSC somehow provided
value above the $0.45 cost per serving (Id. at pp. 7–8). Plaintiffs assert that because
Andrien’s “value” conclusions are directly refuted by the factual findings of both this
Court and the Seventh Circuit, they should be excluded (Id. at p. 8). Relatedly, in their
fifth argument, Plaintiffs assert that Andrien’s analysis leads to the “absurd” conclusion
that the actual value of GSC had it been marketed truthfully was greater than what it
sold for, and therefore could not aid the jury as the trier of fact (Doc. 274, p. 10).
First, it certainly seems like it would have been helpful if Andrien was familiar
with the Seventh Circuit’s opinion or the class certification Order, but the Court does not
see why it was necessary or why it would disqualify Andrien’s testimony. And Plaintiffs
don’t cite to any legal authority indicating that it does, in fact, disqualify his testimony.
The Court thinks it’s also worth noting that Plaintiffs seem to be reading a bit too
much into some of the Orders and Opinions in this case. Yes, the Seventh Circuit
indicated that damages could be computed in this case “for example . . . by taking the
difference between the actual value of the package she purchased (instant coffee) and the
inflated price she paid (thinking the pods contained real coffee grounds).” Suchanek v.
Sturm Foods, Inc., 764 F.3d 750, 760 (7th Cir. 2014). But this language is by no means a
binding instruction on how the price premium must be calculated in this case. It is
merely an example, as the Seventh Circuit explicitly stated. Accordingly, the Court sees
no reason why Andrien cannot opine that the price premium should be calculated using
an alternate methodology, namely by taking the difference between the actual value of
Page 26 of 49
the package consumers purchased (instant and microground coffee) and the inflated
price they paid (thinking the pods contained real coffee grounds).
Also, this Court has indicated more than once that a full refund may be the proper
measure of damages based on the evidence in the case (see Doc. 247; see supra section
B(1)). But each time, the Court also has indicated that it is ultimately up to a jury to
decide whether the evidence supports that measure of damages (see Doc. 247; see supra
section B(1)). To that end, Defendants are certainly allowed to offer their interpretation
of the evidence (i.e., that some consumers intended to purchase instant coffee, liked GSC,
and repeatedly purchased GSC, and those who didn’t like it still received some benefit
from it), and to argue that a full refund is not the appropriate measure of damages.
As for Andrien’s unfamiliarity with the evidence regarding the public’s response
to GSC, the Court agrees that this is indeed problematic for him. But it’s a problem of
credibility, not admissibility. By opining that the actual value of GSC is greater than
$0.45, Andrien is in essence saying that consumers would have been willing to pay more
for instant coffee in a k-cup than they paid for ground coffee in a k-cup. This suggestion
seems comically inconsistent with the evidence presented in this case so far and the
realities of the coffee market in the United States. 11 But if that is the story that
Defendants want to present to the jury, it is their gamble to make. Plaintiffs’ counsel can
See, e.g., Instant Coffee Fights Back Against the Growing Challenge from Fresh, EUROMONITOR INT’L (Aug. 20,
2016), http://blog.euromonitor.com/2016/08/instant-coffee-fights-back-growing-challenge-fresh.html
(indicating that instant coffee sales in North America are “struggling” and “forecast to shrink”); Roberto
Ferdman, Almost Half of the World Actually Prefers Instant Coffee, WASH. POST, July 14, 2014,
https://www.washingtonpost.com/news/wonk/wp/2014/07/14/almost-half-of-the-world-actually-pr
efers-instant-coffee/?utm_term=.a8818f299baf (“Americans have proved pretty exceptional in their utter
disinterest in warming up to [instant coffee]. . . . [I]nstant coffee sales have barely budged since 2008, and
even fell marginally last year to just over $960 million. While that might sound like a lot, it’s actually a
paltry fraction of the $30-plus billion U.S. coffee market.”).
11
Page 27 of 49
certainly press Andrien about his counterintuitive conclusion that GSC is worth more
than its sale price on cross-examination. See Daubert v. Merrell Dow Pharm., Inc., 509 U.S.
579, 596 (1993); Smith v. Ford Motor Co., 215 F.3d 713, 718 (7th Cir. 2000).
In sum, Plaintiffs have failed to convince the Court that Jeffrey Andrien’s
opinions and analysis are so unreliable that they are inadmissible under Daubert and
Rule 702. Consequently, Plaintiffs’ motion to exclude Andrien’s testimony is denied.
DEFENDANTS’ MOTION TO DECERTIFY THE CLASS
An order regarding class certification is subject to alteration or amendment prior
to final judgment. FED. R. CIV. P. 23(c)(1)(C). The decision whether to decertify a class
action rests in the discretion of the district court. Retired Chi. Police Ass’n v. City of Chi.,
7 F.3d 584, 596 (7th Cir.1993) (“Under Federal Rule of Civil Procedure 23, a district court
has broad discretion to determine whether certification of a class is appropriate.”); see
also Phillips v. Sheriff of Cook Cty., 828 F.3d 541, 549 (7th Cir. 2016) (“We review a district
court’s decision regarding class certification for abuse of discretion.” (citing Suchanek v.
Sturm Foods, Inc., 764 F.3d 750, 755 (7th Cir. 2014)). In considering the appropriateness of
decertification, the standard of review is the same as for a motion for class certification:
whether the requirements of Rule 23 are met. See Phillips, 828 F.3d at 549.
In their motion to decertify the class, Defendants challenge only one aspect of
class certification: the measurement of damages, which bears on Rule 23(b)(3)’s
predominance requirement (see Doc. 278). The Court will begin by discussing the legal
standard for predominance and its previous predominance determination from the class
certification Order. The Court will then turn to the substance of Defendants’ specific
Page 28 of 49
arguments regarding predominance that are presented in the motion to decertify.
“The predominance inquiry tests whether proposed classes are sufficiently
cohesive to warrant adjudication by representation.” Tyson Foods, Inc. v. Bouaphakeo, 136
S. Ct. 1036, 1045 (2016) (quoting Amchem Products, Inc. v. Windsor, 521 U.S. 591, 623
(1997)). “This calls upon courts to give careful scrutiny to the relation between common
and individual questions in a case.” Tyson Foods, 136 S. Ct. at 1045. “An individual
question is one where ‘members of a proposed class will need to present evidence that
varies from member to member,’ while a common question is one where ‘the same
evidence will suffice for each member to make a prima facie showing [or] the issue is
susceptible to generalized, class-wide proof.” Id. (quoting 2 W. RUBENSTEIN, NEWBERG ON
CLASS ACTIONS § 4:50, pp. 196–197 (5th ed. 2012)). Predominance is satisfied when “the
common, aggregation-enabling issues in the case are more prevalent or important than
the non-common, aggregation-defeating, individual issues.” Tyson Foods, 136 S. Ct. at
1045. See also Kleen Prod. LLC v. Int’l Paper Co., 831 F.3d 919, 925 (7th Cir. 2016)
(“Predominance is satisfied when ‘common questions represent a significant aspect of a
case and . . . can be resolved for all members of a class in a single adjudication.’” (quoting
Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 815 (7th Cir. 2012))), cert. denied,
137 S. Ct. 1582 (2017).
The predominance analysis begins with the elements of the underlying cause of
action. Messner, 669 F.3d at 815 (citing Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S.
804, 809 (2011)). In the class certification Order, the Court looked at the three central
Page 29 of 49
elements of all (or at least most) of the consumer fraud claims at issue: deception,12
reliance/causation, and damages (see Doc. 247). The element of deception represents the
question common to the entire class: whether the GSC packaging was likely to deceive a
reasonable consumer (Doc. 247). The Court determined that this is the threshold
question when it comes to determining Defendants’ liability, and therefore it is the most
significant element of each class member’s claim (Doc. 247). This question can be
resolved with common proof (Docs. 247, 250). As for damages, the Court determined
that damages are capable of being measured on a class-wide basis using Candace
Preston’s Retail Damages model or her Price Premium Damages model (Doc. 247).
Reliance/proximate causation is the only element that possibly necessitates
individualized proof. But it is not a required element under the statutes of some states,
and other states permit a presumption of reliance/causation, meaning it could be proven
on a class-wide basis for those states, and individual inquiries would not be necessary
(Docs. 247, 250). Furthermore, the issue of reliance/causation is simpler than the issues
of deception and damages and does not require costly expert evidence to prove
(Doc. 247). With these things in mind, the Court concluded that common issues of law
and fact predominate, and proceeding as a class action was the superior form of
adjudication for this case (Id.).
In their motion to decertify the class, Defendants argue that, based on the
additional discovery developed by the parties, neither of Preston’s damages models is
As mentioned at the beginning of this Order, the Court previously referred to the element of
deception—meaning whether the GSC packaging was likely to deceive a reasonable consumer—as the
issue of “liability” (see Doc. 247, 250).
12
Page 30 of 49
adequate for measuring damages on a class-wide basis because, for a number of reasons,
neither model “provide[s] a measure of damages attributable only to the alleged
misconduct and that can be calculated on a classwide basis” (Doc. 278, p. 13). The idea is
that if Plaintiffs are left without a valid way to measure damages on a class-wide basis,
then individualized proof will be necessary to determine each class member’s damages.
And the individualized damage calculations, combined with the individualized
inquiries regarding reliance/causation, would mean that individual questions
overwhelm questions common to the class and therefore decertification is required (see
Doc. 278, p. 5).
An “essential component of the predominance inquiry” is whether there is a
reliable method for measuring damages on a class-wide basis that matches Plaintiffs’
theories of liability. 1 JOSEPH M. MCLAUGHLIN, MCLAUGHLIN ON CLASS ACTIONS § 5:23
(13th ed.) (discussing the broad implication of Comcast, 133 S. Ct. 1426); Suchanek, 764
F.3d at 760 (“In determining whether to certify a consumer fraud class, the court should
begin with a “rigorous analysis” into whether the plaintiffs’ “damages are susceptible of
measurement across the entire class.” (quoting Comcast, 133 S. Ct. at 1433)). See also
Mullins v. Direct Digital, LLC, 795 F.3d 654, 671 (7th Cir. 2015) (“[T]he method of
determining damages must match the plaintiff’s theory of liability and be sufficiently
reliable.” (citing Comcast, 133 S. Ct. at 1433)), cert. denied, 136 S. Ct. 1161 (2016); Rikos v.
Procter & Gamble Co., 799 F.3d 497, 523 (6th Cir. 2015) (Comcast requires courts to a
“conduct a ‘rigorous analysis’ to ensure at the class-certification stage that ‘any model
supporting a plaintiff’s damages case [is] consistent with its liability case,’ i.e., that the
Page 31 of 49
model ‘measure[s] only those damages attributable to that theory’ of liability.” (quoting
Comcast, 133 S. Ct. at 1433)), cert. denied, 136 S. Ct. 1493 (2016).
As an initial matter, the Court notes that even if Defendants succeed in showing
that neither of Candace Preston’s damages models meets the requirements of Comcast,
and Plaintiffs are left without a way to measure damages on a class-wide basis,
decertification is still not required. It is well-established that the need for individual
assessments regarding reliance/causation and damages does not preclude class
certification entirely; the Court still has the discretion to certify the case on the issues that
are common to the class, particularly the issue of deception. See In re IKO Roofing Shingle
Prod. Liab. Litig., 757 F.3d 599, 602, 603 (7th Cir. 2014); Butler v. Sears, Roebuck & Co., 727
F.3d 796, 800, 801 (7th Cir. 2013); Pella Corp. v. Saltzman, 606 F.3d 391, 393–94 (7th Cir.
2010).
A. STATUTORY DAMAGES
Defendants argue that Plaintiffs’ claims for statutory damages do not provide a
basis for class certification without “an independent and non-speculative measure of
damages” (Doc. 278, p. 20). In other words, Defendants seem to believe that before
statutory damages are available, class members must offer evidence that they not only
suffered actual harm, but they also must offer evidence that quantifies the amount of
that harm (see id.; Doc. 280, p. 5).13
The element of damages encompasses both the fact of damage (meaning proof of an actual injury) as
well as the measurement of that damage (meaning quantifying the value of the injury). In the class
certification Order, the Court spoke mainly in terms of measuring and quantifying the damage of each
class member (Doc. 247). See Suchanek, 764 F.3d at 760 (“In determining whether to certify a consumer
fraud class, the court should begin with a ‘rigorous analysis’ into whether the plaintiffs’ ‘damages are
susceptible of measurement across the entire class.’” (quoting Comcast Corp. v. Behrend, 133 S. Ct. 1426, 1433
(2013))). Candace Preston’s testimony is necessary when it comes to measuring damages.
13
Page 32 of 49
Two states provide for statutory damages: Alabama and New York. Alabama’s
Deceptive Trade Practices Act provides for the greater of $100 or actual damages. ALA.
CODE § 8-19-10(a)(1). New York’s Deceptive Act and Practices Law provides for the
greater of $50 or actual damages while the False Advertising law provides for the greater
of $500 or actual damages. N.Y. GEN. BUS. LAW §§ 349(h) and 350-e.
The Court agrees with Defendants that Plaintiffs must indeed prove an actual
injury before they can recover statutory damages under either the Alabama or New York
statutes. See ALA. CODE § 8-19-10(a) (“Any person who commits one of more of the acts
or practices declared unlawful under this chapter and thereby causes monetary damage to a
consumer . . . shall be liable to each consumer . . . . “) (emphasis added); Lynn v. Fort
McClellan Credit Union, No. 1:11-CV-2904-VEH, 2013 WL 5707372, at *7 (N.D. Ala. Oct.
21, 2013) (finding the plaintiff had no claim under the ADTPA because he proved no
monetary damages); Stutman v. Chem. Bank, 731 N.E.2d 608, 612 (N.Y. 2000) (“[A]
plaintiff must prove ‘actual’ injury to recover under the statute, though not necessarily
pecuniary harm.”).
But the Court is not convinced that Plaintiffs must prove the exact monetary
value of their injury. Kurtz v. Kimberly-Clark Corp., No. 14-CV-1142, 2017 WL 1155398, at
*57 (E.D.N.Y. Mar. 27, 2017) (“New York law does not require that the injury must be
proven with a specified degree of certitude; a plaintiff is only required to show ‘that
plaintiffs paid more than they would have for the good [because of] the deceptive
practices of the defendants-sellers.’” (quoting Orlander v. Staples, Inc., 802 F.3d 289, 302
(2d Cir. 2015))). Regardless of the amount of damage—whether it is a full refund or a
Page 33 of 49
partial refund—it will be, without question, less than $50. So it doesn’t seem to matter
what the exact amount of damage is when class members in Alabama and New York
will not be collecting that amount—they will be collecting the statutory damages
amount instead. Kurtz, 2017 WL 1155398, at *57 (holding that the “battle of experts” on
the amount of price premium that consumers paid for “flushable toilet wipes” that
allegedly were not actually flushable was “largely beside the point” because “[o]nce an
injury is established . . . New York General Business Law § 349(h) provides for statutory
damages of $50 to each class member”); Belfiore v. Procter & Gamble Co., 311 F.R.D. 29, 70
(E.D.N.Y. 2015), reconsideration denied, 140 F. Supp. 3d 241 (E.D.N.Y. 2015) (“Once the
injury is established, statutory damages can be calculated on a classwide basis. . . .
Although not necessary, expert Weir may be able to determine an average price paid for
the misrepresentation.”)
Thus, it appears to the Court that class members from Alabama and New York
only need to show the fact of damage. And when it comes to proving the fact of damage,
the Court does not believe that Candace Preston’s testimony is necessary. While it could
certainly be used to show that consumers bought a worthless product or paid an inflated
price, it seems like other evidence, such as consumer complaints, expert surveys, and
Defendants’ own internal documents, could just as easily show the same thing. If the
Court is correct, Plaintiffs’ ability to measure damages on a class-wide basis, or
Defendants’ criticisms of Candace Preston’s damages models, would not have any
bearing on the claims of class members from these states. In other words, these claims
could proceed even if Candace Preston’s damages models are thrown out.
Page 34 of 49
In the upcoming section titled “Management of Trial,” the Court invites the
parties’ input on the evidence needed to prove that fact of damage.
B. RETAIL DAMAGES MODEL
As previously indicated, the Retail Damages model, which provides consumers
with a full refund of the purchase price of GSC, is based on the notion that consumers
would not have purchased GSC if not for the deceptive packaging and the product was
worthless to them. Defendants argue that the Retail Damages model does not meet the
requirements of Comcast and is not a viable method for determining class-wide damages
because (1) it is contradicted by the evidence, (2) Preston did not develop the model
herself and instead relied on an assumption from Plaintiffs’ attorneys, and (3) it is
inconsistent with Plaintiffs’ omission theory of liability (Doc. 278, pp. 18–20). For the
reasons explained below, the Court is unpersuaded by these arguments.
1. Evidence Regarding the Value of GSC
Defendants repeat the argument from their Daubert motion that the Retail
Damages model is not a viable method for determining class-wide damages because
evidence in the record demonstrates that GSC was not worthless and that some
consumers received a benefit from GSC (Doc. 278, p. 18). Like they did in their Daubert
motion, Defendants point to Preston’s deposition testimony that the components of
GSC—the instant coffee and the single-serve k-cup packaging—had value (Id.). For the
reasons previously explained in this Order, the Court does not believe that the cost of
GSC’s ingredients and packaging has any bearing on whether a full refund is an
appropriate measure of damages.
Defendants also point to a comment from one participant in Dr. Bobby Calder’s
Page 35 of 49
consumer study, who said “I would recommend this coffee as having a very nice
medium rich flavor. It has nice color and body” (Doc. 278, pp. 18–19; Doc. 101-8, p. 82).
And Defendants point to favorable consumer reviews that they received about GSC,
which they claim show that some consumers liked GSC, appreciated it as a lower-cost
alternative and repeatedly purchased the product (Doc. 278, p. 19; Doc. 108-8, p. 9).
A jury could, however, easily reject this evidence. First, the passage from the
survey participant was cherry-picked, and Defendants ignore the rest of what the
participant had to say, including that GSC “is instant coffee” (Doc. 101-8, p. 82). She also
said
[I]t’s kind of like why even bother to have a Keurig with the Grove Square?
Add the water and call it a day. You’re paying for the separate tubs. I could
just take a spoon and drop it in if that’s all the Grove Square is. I’m kind of
sad wasting money. I want to make sure I get what I paid for. What’s the
point of Grove Square?
(Doc. 101-8, p. 82). As for the consumer reviews, there is evidence that Defendants had
their employees write fake, positive reviews (see Doc. 101-1, p. 28). Furthermore, the
positive reviews are only a very small fraction of the consumer reviews received by
Defendants. The overwhelming majority of the reviews involve consumers who were
very dissatisfied with the product, wanted their money back, and vowed never to
purchase it again.
Consequently, the evidence cited by Defendants does not make it indisputable, or
even relatively clear, that some consumers received a benefit from GSC. Defendants’
evidence simply shows that Plaintiffs ultimately may not be able to make the required
showing that GSC was of no value to consumers. But it must be left for a jury to decide
Page 36 of 49
whether Plaintiffs made that showing. See, e.g., Rikos v. Procter & Gamble Co., 799 F.3d 497,
521 (6th Cir. 2015), cert. denied, 136 S. Ct. 1493 (2016) (emphasizing that establishing
predominance requires plaintiffs to show only that “they will be able to prove injury
through common evidence, not that they have in fact proved that common injury.”)
2. Origination of the Model
Defendants repeat the argument from their Daubert motion that the Retail
Damages model is “unsupported” because instead of determining the value of GSC,
Preston adopted the assumption of Plaintiffs’ attorneys that GSC had no value (Doc. 278,
pp. 8, 19). According to Defendants, this is inappropriate because “determining the
value of a consumer product requires expert analysis” (Id. at p. 19). The Court does not
see how the case cited by Defendants stands for that proposition, however, nor is it clear
how the analysis of that case would apply to the facts here.14 Defendants also suggest
that Plaintiffs arrived at their position that GSC had no value through some
impermissible leap in logic (Doc. 278, p. 19). But as previously explained in this Order,
there is plenty of evidence to support that position. Consequently, Defendants’
argument provides no basis for concluding that the Retail Damages model is an invalid
method for measuring damages across the entire class.
3. Applicability to Omission Theory of Liability
Defendants’ final argument is that Preston’s Retail Damages model is not a viable
method for determining class-wide damages because it cannot apply to Plaintiffs’
omission theory of liability (Doc. 278, p. 20). This argument is part of Defendants’
Defendants cited to Weiner v. Snapple Beverage Corp., Case No. 07 CV 8742(DLC), 2010 WL 3119452, *10
(S.D. N.Y. Aug. 5, 2010).
14
Page 37 of 49
dogged attempt to convince the Court that Plaintiffs are proceeding on two separate,
“mutually exclusive” theories of liability: (1) the GSC packaging was deceptive by virtue
of affirmative misrepresentations set forth on the package, such as the use of the word
“soluble” instead of “instant,” and (2) the GSC packaging was deceptive by virtue of
omissions, such as not disclosing the percentage of microground coffee versus instant
coffee (see Doc. 268, pp. 9, 14-15, 16; Doc. 278, pp. 10, 13, 17).
Once again, however, the Court rejects Defendants’ characterization of Plaintiffs’
theories of liability (see Doc. 247, pp. 32–33, 35). Plaintiffs are, in fact, proceeding on two
different theories of liability, just not the two advanced by Defendants. Plaintiffs’ first
theory of liability is that because GSC was overwhelmingly instant coffee, its value is
nothing. That is, the deceptive packaging caused consumers to pay for a product they
never would have purchased had they known it was not ground coffee. Plaintiffs’
second theory of liability is that because GSC was overwhelmingly instant coffee, it is
worth something substantially less than what consumers paid for it thinking it was
ground coffee. In other words, the deceptive packaging caused consumers to pay an
inflated price for instant coffee.
Neither of these theories of liability is tied to any one particular aspect of the GSC
packaging. As the class certification Order previously highlighted, the totality of the GSC
package is at issue. Specifically, the Court explained that “[t]he named Plaintiffs all claim
that the overall GSC packaging was deceptive in that it created, and/or failed to correct, the
misimpression that the product was premium, ground coffee.” (Doc. 247, p. 32)
(emphasis added). The use of the word “soluble” instead of “instant” and the omission
Page 38 of 49
of the percentage of instant coffee were just two aspects of the package that contributed
to the deception that GSC was ground coffee. As the Seventh Circuit pointed out, there
were other aspects of the package that also contributed to the deception, such as “the
image of K–Cups with fresh roasted coffee beans and the admonition that the GSC
product was intended ‘[f]or use by owners of Keurig© coffee makers,’” the “quality
promise” indicating the coffee was “made with some of the world’s highest quality
Arabica beans, roasted and ground to ensure peak flavor, then packaged to lock in
optimum freshness,” and the “Coffee Lover’s Bill of Rights.” Suchanek v. Sturm Foods,
Inc., 764 F.3d 750, 753 (7th Cir. 2014)).
The Court further explained in the class certification Order that it “failed to see
how or why it makes a difference whether some of the named Plaintiffs were misled
solely by affirmative misrepresentations in statements, images, and descriptions set
forth on the packaging as opposed to a material omission or some combination of
affirmative misrepresentations and omissions” (Doc. 247, p. 32). Once again, Defendants
have not provided any explanation, or cited to any authority, showing that this
distinction actually matters, particularly with regard to calculating damages. The only
thing Defendants say is that there is no indication of “how a full refund can be the
proper measure of damages for consumers who understood that they were purchasing a
product comprised of instant and microground coffee but thought they were getting a
different ratio of instant and microground coffee” (Doc. 278, p. 20). Aside from being too
perfunctory for the Court to analyze, this explanation also appears to be entirely
divorced from the evidence in this case—the Court is unaware of any evidence in the
Page 39 of 49
record showing that some purchasers knew they were purchasing instant and
microground coffee.
In sum, neither this argument nor Defendants’ other two arguments provide any
basis for concluding that the Retail Damages model is an invalid method for measuring
damages across the entire class.
C. PRICE PREMIUM DAMAGES MODEL
The Price Premium Damages model reflects Plaintiffs’ theory of liability that
consumers paid an inflated price for GSC, and it would provide class members with a
partial refund. This model assumes that the value of GSC would have been substantially
less than what it sold for had it not been deceptively marketed as ground coffee. The
price premium was computed by taking “the difference between the actual value of the
package [consumers] purchased” and “the inflated price [consumers] paid” thinking the
product was as advertised. Suchanek v. Sturm Foods, Inc., 764 F.3d 750, 760 (7th Cir. 2014).
Defendants claim that deficiencies in Preston’s method for determining the
“actual value” of GSC mean that the Price Premium model does not satisfy the
requirements of Comcast and cannot be used to determine class-wide damages (Doc. 278,
pp. 7–9, 14–17). Specifically, Defendants argue that the model does not isolate the price
premium associated with the alleged misleading acts because Preston (1) failed to
distinguish between Plaintiffs’ two theories of liability, (2) failed to consider the
supply-side of valuation, and (3) failed to control for differences between GSC and the
comparable products that may contribute to pricing (Doc. 278, pp. 14–17). Before the
Court gets to these arguments, none of which it finds persuasive, it will first address
Page 40 of 49
Plaintiffs’ argument regarding the level of scrutiny required by Comcast.
1. Level of Scrutiny Required by Comcast
Plaintiffs argue that Defendants are trying to impose a level of refinement for the
damages models that is not required by Rule 23 or Comcast (Doc. 279). While this
argument holds some intrigue, the Court declines to address it because neither party
provided a worthwhile analysis. To begin with, Defendants simply assumed without
any discussion that the Seventh Circuit interprets Comcast to require the level of
specificity that the district courts in California do (see Doc. 268, pp. 14–17). Plaintiffs then
called into question whether that was true, but their entire “argument” consists of a
two-paragraph block quotation from a case that is eighty-five years old and predates
Rule 23 (Doc. 279, pp. 14–15). Plaintiffs make absolutely no effort to discuss the
requirements of Rule 23 as interpreted by the Seventh Circuit, let alone the impact
Comcast had on Rule 23 (see Doc. 279). Defendants came back and simply accused
Plaintiffs of failing to cite to any case law and failing to distinguish the California district
court cases that Defendants cited to, but once again make no effort to substantiate their
own argument in the first place (Doc. 280. p. 3). Quite simply, neither side did their job.
The Court will not conduct the necessary research and analysis to construct each side’s
argument and then debate itself to determine which argument is correct.
2. Failure to Distinguish Between Theories of Liability
Defendants argue that Preston’s Price Premium analysis “is not tied to the alleged
wrongdoing because it does not distinguish between Plaintiffs’ theories of liability”
(Doc. 278, p. 17; see also Doc. 268, p. 16). Defendants believe Preston needed to conduct a
separate analysis for consumers who were misled by affirmative misrepresentations and
Page 41 of 49
consumers who were misled by omissions (Doc. 278, p. 17). This argument is summarily
rejected for the same reasons stated above. See supra section (B)(3).
3. Failure to Control for Difference Between GSC and Comparable Products
This argument is Defendants’ most substantial one, and it relates to Preston’s
method of calculating the price premium by taking the difference between the retail
price of GSC and actual value of the package consumers purchased. Preston determined
the “actual value” of GSC by looking at the retail price of comparable instant coffees.
According to Defendants, simply taking the difference between the retail cost of GSC
and the retail cost of instant coffee is a “price differential” calculation, not a “price
premium” calculation (Doc. 268, pp. 16–17; Doc. 278, pp. 15–17). Defendants believe that
a proper price premium calculation must adjust for differences between GSC and the
comparable products, otherwise the model does not isolate the premium that is due
solely to the alleged deception (Doc. 278, p. 14).
The Court first notes that calculating the price premium is indisputably an
appropriate way to quantify damages on a class-wide basis (see Doc. 268-2, pp. 13– 14).
See also Suchanek v. Sturm Foods, Inc., 764 F.3d 750, 760 (7th Cir. 2014). Defendants simply
dispute the precision of Preston’s particular methodology of calculating the price
premium. They essentially argue that she needed to include several more variables in
order for her model to be an accurate measure of damages. Defendants have not,
however, convinced the Court that this purported deficiency means her damages model
is so deficient that it must be excluded and the class must be decertified.
Defendants refer the Court to four cases in which a “price differential” calculation
Page 42 of 49
was purportedly rejected: In re POM Wonderful, No. 10-cv-2199, 2014 WL 1225184 (C.D.
Cal. Mar. 25, 2014); (2) Lanovaz v. Twinings North America, No. 12-cv-2646, 2014 WL
1652338 (C.D. Cal. July 24, 2014); (3) Algarin v. Maybelline, 300 F.R.D. 444 (S.D. Cal. 2014);
and (4) Brazil v. Dole Packaged Foods, LLC, 2014 WL 5794873 (N.D. Cal. Nov. 6, 2014) (Doc.
at p. 15). Unfortunately, however, Defendants make absolutely no effort to apply the
analysis of those cases to the facts of this case (see Doc. 268, pp. 16–17; Doc. 278, pp. 14–
17). They simply make the conclusory statement that “Like the expert opinions rejected
In re POM Wonderful, Lanovaz, and Algarin, Preston offers only a price differential that is
not tied to the actual value of GSC and, therefore, to any alleged wrongdoing.”
(Doc. 278, p. 16). The Court will not review the cases and supply the legal analysis that
Defendants neglected to provide.
The Court also notes that there is case law that seemingly refutes Defendants’
suggestion that it is always legally unacceptable to use a “price differential” calculation
as a measure of damages in a false advertising case. See, e.g., Goldemberg v. Johnson &
Johnson Consumer Companies, Inc., 317 F.R.D. 374, 394 (S.D.N.Y. 2016) (“Calculating a
price premium can be as simple as computing the difference between the cost of the
second best product in the product class (without a deceiving label) and the cost of the
product at issue (with the label).”); Ebin v. Kangadis Food Inc., 297 F.R.D. 561, 571–72
(S.D.N.Y. 2014) (finding the price premium associated with the deceptive advertisement
was the difference between the market price of 100% olive oil and the market price of the
less expensive olive-pomace oil).
Furthermore, Candace Preston explained that she considered the differences
Page 43 of 49
between GSC and the instant-only products that she chose as comparables—such as,
branding, the microground component, and the k-cup packaging—and whether those
differences contributed to the difference in price (see Doc. 272-3). She concluded,
however, that those differences did not make the value of GSC higher than the value of
her chosen comparable (See Doc. 272-3). The Court believes that Preston’s explanations
have to be evaluated by a jury. Determining whether Preston’s model is accurate as it
stands, or whether it needed to include more variables in order to be accurate, would
require the Court to delve into the merits of the case and resolve issues of fact, which
would invade the province of the jury.
4. Failure to Consider Supply-Side of Valuation
Defendants’ final argument is that Preston’s Price Premium analysis “failed to
consider the supply-side of valuation, which also warrants decertification” (Doc. 278,
p 17; see also Doc. 268, p. 18). This argument appears to have been thrown in as an
afterthought because the entire argument is comprised of the following three sentences:
A determination of the price premium arising from an alleged deception
must consider whether manufacturer would even sell the product at the
alleged actual value. The court In re NJOY, Inc. Cons. Class Action Litig., No.
14-cv-428, 2016 WL 787415, at *7 (C.D. Cal. Feb. 2, 2016), excluded a price
premium testimony that ignored this question, stating: “Dr. Harris . . .
ignores the price at which NJOY, and other e-cigarette manufactures,
would be willing to sell their products.” Here too, Preston’s bulk instant as
“a comparable” theory ignores the price at which it would be feasible to
sell GSC and, in effect, concludes Sturm should have sold GSC at a loss.
(Dkt. 268-1, Preston 2d Supp. Rpt. Table 6 (2011 data); Dkt. 137-20 at Page
ID # 2193-94).
This argument is simply too undeveloped for the Court to properly analyze it.
Defendants do not provide any coherent explanation as to why an expert should
Page 44 of 49
consider the price at which the manufacturer would sell the product when calculating a
price premium. Instead, they simply rely on one sentence cherry-picked from the In re
NJOY case. Furthermore, Defendants insinuate that they would have sold GSC at a loss
if they charged only $0.06 per cup, but they do not provide any evidence to back this up.
Absent that evidence, this insinuation is not particularly credible because, after all, a
number of other manufacturers were willing to sell instant coffee at only $0.06 per cup
(see Doc. 268-1, p. 11).
For these reasons, the Court is unconvinced that Preston’s purported failure to
consider the supply side of valuation means that her Price Premium model of damages
fails the requirements of Comcast.
In sum, Defendants have failed to set forth an argument that justifies excluding
the Price Premium damages model as an inappropriate measure of damages. The Court
concludes the Retail Damages model and the Price Premium damages model are both
acceptable methods for calculating damages across the entire class. Consequently, the
Court stands by its previous determination that predominance is satisfied. Defendants’
motion to decertify is denied.
MANAGEMENT OF TRIAL
The class was certified as to liability only, and the Court envisioned a bifurcated
trial (Doc. 247; see also Doc. 250). The first phase would address the deception, meaning
the question of whether the GSC packaging was likely to mislead a reasonable consumer
(Docs. 247, 250, 254). If that question was answered in the affirmative, a second phase
would then address the issue of reliance/proximate causation with respect to each class
Page 45 of 49
member and damages (Doc. 254).
The time has come to replace this simplistic plan with something much more clear
and detailed. To that end, the Court notes that it always assumed, although it did not
explicitly state, that additional issues would be certified for the whole class, or at least
for certain subclasses, before trial arrived (see, e.g., Doc. 247, pp. 37–38 (“If a single class is
certified for the purposes of establishing damages, each class member will receive a full
refund or a partial refund. Alternatively, in the event subclasses are created, class
members in certain states will receive statutory damages.”). Moreover, it has become
clear to the Court that having the first jury answer only the question of deception would
not be an effective use of the jury or an effective use of the time and resources of the
parties and the Court. If the question of deception is answered in the affirmative, the jury
should also address all other pertinent issues that can be resolved using common proof,
as opposed to individualized proof. This will serve to minimize the overlap of evidence
between the two phases. And, as Plaintiffs suggested at the hearing, it could also
potentially resolve the claims of some class members, thereby paring down what must
be addressed during the second phase.
The Court acknowledges that there has already been some discussion and legal
analysis regarding additional common questions and proposals for managing trial. But
that information is scattered throughout the submissions and Orders in this case, and the
analysis is often deficient or devoid of input from one of the parties. The Court seeks to
have the parties now consolidate that information and supplement it with a more
thorough analysis than has been provided thus far. The Court aims to have this
Page 46 of 49
accomplished in one joint submission so that it does not have to sift through multiple
documents to identify and synthesize the parties’ respective positions. Both parties
should take particular care to adequately explain their position and provide sufficient
citations to legal authority. Vague, conclusory, or unsupported statements will not be
tolerated.
The Court would like Plaintiffs to begin by submitting a memorandum that not
only identifies additional questions common to the entire class or common to
sub-classes, but also contains something of a trial plan. More specifically, the Court
would like Plaintiffs to separately discuss each of the statutes under which they have
stated a claim. For each statute, Plaintiffs should identify:
1. the elements of a cause of action;
2. the elements that remain to be proven if the question of deception is
answered in the affirmative, including the fact of damage, see supra
n.13;
3. any additional issues that are not required elements of the claim, but
are nonetheless pertinent;15
4. whether the remaining elements and issues can be established using
common proof or whether individualized proof will be required, and
what evidence Plaintiffs intend to use to prove each remaining
elements/issues;
5. whether each remaining element or issue is relevant under any other
statute—in other words, whether the issue is relevant to the entire class
or a certain subclass; and
6. whether each remaining element or issue can be decided by the first
jury.
For example, the Court indicated in the class certification Order that Defendants’ intent was a required
element of any of the class members’ claims, but it was a relevant issue under some of the statutes
(Doc. 247, p. 25).
15
Page 47 of 49
Plaintiffs also should address whether the class definition needs to be amended.
The class was defined as “All persons or consumers that during the Class Period, from
September of 2010, up through the date the case is certified and notice is disseminated, who
purchased Defendants’ Grove Square Coffee (“GSC”) products in Alabama, California,
Illinois, New Jersey, New York, North Carolina, South Carolina, and Tennessee”
(Doc. 247) (emphasis added). It has been determined, however, that sales of GSC were
discontinued in September 2014 (Doc. 200), and Defendants indicated at the hearing that
they believed the class period ended in September 2014 (Doc. 310, p. 27). Thus the class
definition does not appear to accurately reflect the end of the class period.
For their response, Defendants should begin with Plaintiffs’ submission and add
their responses to that same document.16 They should respond point by point, explicitly
indicating whether they agree or disagree with Plaintiffs’ position. Where Defendants
disagree, they must provide a thorough explanation of their respective position and
include citations to relevant legal authority.
Plaintiffs should then take Defendants’ submission and add their own replies to
that same document. For points where the parties are in agreement, no reply is
necessary. For each point where there is a disagreement, however, Plaintiffs must
provide a reply to Defendants’ response.
Plaintiffs should send Defendants their submission in a format that can be edited, such as a Microsoft
Word document.
16
Page 48 of 49
CONCLUSION
For the reasons set forth above, Defendants’ Motion to Exclude the Expert
Testimony of Candace Preston Regarding Damages (Doc. 267) is DENIED. Defendants’
Motion to Decertify the Class (Doc. 278) is DENIED. Plaintiffs’ Motion to Exclude the
Expert Testimony of Jeffrey Andrien (Doc. 274) is DENIED.
The joint submission regarding trial procedures shall be filed with the Court on or
before October 30, 2017.
IT IS SO ORDERED.
DATED: August 28, 2017
NANCY J. ROSENSTENGEL
United States District Judge
Page 49 of 49
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?