United States Court of Appeals For the Eighth Circuit ___________________________

No. 24-2830 ___________________________

In re: Folgers Coffee Marketing

------------------------------

Sharel Mawby

lllllllllllllllllllllPlaintiff

Marcia Sorin, Florida Southern, 9:20-cv-80897; Shelly Ashton, California Central, 5:20cv00992

lllllllllllllllllllllPlaintiffs - Appellees

Jay Schoener, California Central, 5:20-cv-00992

lllllllllllllllllllllPlaintiff

Frederick Tan, California Central, 2:20-cv-09370; Ellen Moser, Illinois Northern, 1:20-cv-07074; A. Kevin Fahey, District of Columbia, 1:20-cv-03620; Geoff Thomson, Texas Eastern, 1:21-cv-00009

lllllllllllllllllllllPlaintiffs - Appellees

Julie Marthaller; Rodger Smith, Florida Southern, 0:21-cv-60265; Deborah Bosso

lllllllllllllllllllllPlaintiffs

Mark Smith

lllllllllllllllllllllPlaintiff - Appellee Ramon Ibarra; Kimberley C. Clark, Eastern District of Texas 6:21-cv-00457; Trina Green, Missouri Western, 4:23-cv-00226-BP; Marcia Nupp

lllllllllllllllllllllPlaintiffs

v.

The Folger Coffee Company; The J.M. Smucker Company

lllllllllllllllllllllDefendants - Appellants

Does, 1 through 50, inclusive; Walmart, Inc., Florida Southern, 0:21-cv-60265

lllllllllllllllllllllDefendants ____________

Appeal from United States District Court for the Western District of Missouri - Kansas City ____________

Submitted: September 16, 2025 Filed: November 26, 2025 ____________

Before SMITH, ARNOLD, and SHEPHERD, Circuit Judges. ____________

ARNOLD, Circuit Judge.

Missouri resident Mark Smith bought containers of Folgers coffee that featured representations about the number of six-ounce cups that the coffee in each container could produce. Believing the representations to be misleading at best and downright false at worse, Smith, on behalf of himself and others similarly situated to him, sued the Folger Coffee Company and its corporate parent, the J.M. Smucker Company. For

-2- the sake of simplicity we will refer to the defendants as “Folgers.” Smith claimed that Folgers had violated the Missouri Merchandising Practices Act, see Mo. Rev. Stat. § 407.020.1, and claimed that Folgers was unjustly enriched at his expense. The district court, over Folgers’s objection, certified a class for Smith to represent, and Folgers, with our permission, see Fed. R. Civ. P. 23(f), appealed that decision. We agree with Folgers that the class was improperly certified, and so we reverse and remand.

Smith’s lawsuit is one of several similar suits brought against Folgers around the country that the Judicial Panel on Multidistrict Litigation transferred to the U.S. District Court for the Western District of Missouri for pretrial proceedings. See 28 U.S.C. § 1407(a). After the district court appointed interim class counsel at the parties’ behest, the parties filed a consolidated class complaint and later moved to certify six separate statewide classes led by different class representatives. They proposed that Smith represent a class of people who bought certain Folgers products “in Missouri for personal, family, or household purposes” after a specific date. The district court, without objection from the parties, decided that it would consider whether to certify the Missouri class before considering whether to certify classes involving purchases made in other states.

Folgers sells several different coffee products that vary by size, flavor, and roast. In seeking certification of the six state classes, class counsel identified over two dozen Folgers products that contained purported misrepresentations. Smith bought a few of these products, including, for example, Folgers’s Classic Coffee Roast weighing 30.5 ounces. The front of that product’s container says in all capital letters that it “[m]akes up to 240 6 fl oz cups.” The side of the container contains directions, telling consumers how to mix water and coffee “[f]or best brewing results.” If the consumer wants to make one serving of coffee, the container directs him to mix six fluid ounces of water with one tablespoon of coffee. The district court called this the “Single-Serving Method.” Or, if the consumer wants to make ten servings using the

-3- “Pot Method,” the container directs him to mix sixty fluid ounces of water with half a cup (eight tablespoons) of ground coffee. The Pot Method is more efficient than the Single-Serving Method because it can generate ten six-ounce cups of coffee using only eight tablespoons of ground coffee; to create ten six-ounce cups of coffee using the Single-Serving Method would take ten tablespoons of ground coffee. Beneath the directions, the container reiterates that it “makes up to 240 suggested strength 6 fl oz servings.” The other Folgers products class counsel target similarly state that the contents of the container make up to a specific number of cups of coffee.

Before certifying a class, a court must ensure that a plaintiff satisfies the requirements of Federal Rule of Civil Procedure 23. See Johannessohn v. Polaris Indus. Inc., 9 F.4th 981, 984 (8th Cir. 2021). One of those requirements for the type of class action Smith seeks to bring is that “the questions of law or fact common to class members predominate over any questions affecting only individual members.” See Fed. R. Civ. P. 23(b)(3). We’ve said that this “predominance inquiry tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation, and goes to the efficiency of a class action as an alternative to individual suits.” See Johannessohn, 9 F.4th at 985. Predominance isn’t satisfied when individual questions would overwhelm the questions common to the class. See Ebert v. Gen. Mills, Inc., 823 F.3d 472, 478–79 (8th Cir. 2016). Though district courts have discretion to determine whether certification is appropriate, see Johannessohn, 9 F.4th at 984, we’ve recognized that the predominance requirement is demanding and that courts must take a close look at whether common questions predominate over individual ones. See Hudock v. LG Elecs. U.S.A., Inc., 12 F.4th 773, 776 (8th Cir. 2021) (citing Comcast Corp. v. Behrend, 569 U.S. 27, 34 (2013)).

We begin with Smith’s MMPA claim. The MMPA creates a cause of action for damages for “[a]ny person who purchases . . . merchandise primarily for personal, family, or household purposes and thereby suffers an ascertainable loss of money or property, real or personal, as a result of the use or employment by another person” of

-4- an unlawful practice. See Mo. Rev. Stat. § 407.025.1(1). An unlawful practice includes “[t]he act, use or employment by any person of any deception, fraud, false pretense, false promise, misrepresentation, [or] unfair practice . . . in connection with the sale or advertisement of any merchandise in trade or commerce.” See id. § 407.020.1. Smith alleges that Folgers’s representations about the number of cups that could be brewed with its products violates the MMPA because they are “false, deceptive, and misleading.” In the operative complaint, he took exclusive aim at the Single-Serving Method, asserting that if one uses solely that method, a consumer can brew only about 70% of the cups promised. When responding to Folgers’s motion to dismiss and when moving for class certification, though, he arguably contended that the Pot Method also failed to yield the promised number of cups, but he explained that if the Pot Method actually produced the promised amount, consumers would still be deceived because they believe that both methods (and not just the Pot Method) should produce up to the promised number of cups. So, his argument runs, he should recover the benefit of the bargain he made with Folgers—the difference between the value of the product as represented and the actual value of the product as received. See Vitello v. Natrol, LLC, 50 F.4th 689, 693 (8th Cir. 2022).

Our court has recognized that fraud cases are typically unsuitable for class treatment. That’s because the proof required in such cases often varies with respect to what representations consumers received and whether those consumers relied on those representations, meaning that common questions do not predominate over individualized ones. See In re St. Jude Med., Inc., 522 F.3d 836, 838 (8th Cir. 2008). Smith points out, though, that his MMPA claim differs from a typical fraud claim in an important respect: “[T]he MMPA does not require individualized proof of reliance in order to state a claim,” and since all containers of the relevant Folgers products included the representations, individualized showings are unnecessary. We rejected a similar argument in St. Jude. There, the plaintiffs sued the producer of an allegedly defective prosthetic heart valve on the ground that the producer had violated Minnesota consumer-protection statutes, including one much like the MMPA. See

-5- Minn. Stat. § 325F.69, subd. 1. They argued that certification was appropriate because, though common-law fraud claims required individualized showings of reliance, the relevant Minnesota statutes do not. See St. Jude, 522 F.3d at 839. But we held that even though plaintiffs need not prove “traditional common law reliance,” they still had to prove a causal connection between the deceptive act and a harm they suffered. See id. And where “the plaintiffs allege that their damages were caused by deceptive, misleading, or fraudulent statements or conduct in violation of the misrepresentation in sales law, as a practical matter it is not possible that the damages could be caused by a violation without reliance on the statements or conduct alleged to violate the statutes.” See id. at 839–40 (quoting Grp. Health Plan, Inc. v. Philip Morris Inc., 621 N.W.2d 2, 13 (Minn. 2001)). Plus, we emphasized, excusing the plaintiffs from presenting direct proof of individual reliance would not foreclose the defendant from presenting evidence of a lack of causation or reliance, which too would raise individualized questions. See id. at 840. St. Jude wasn’t a one-off aberration. We’ve relied on it at least twice to hold that actions under the same Minnesota consumer-protection laws are unsuitable for class treatment. See Hudock, 12 F.4th at 776–77; Johannessohn, 9 F.4th at 985. The upshot is that, even though a plaintiff need not show traditional common-law reliance, individual issues will still likely swamp any efficiencies that a class action might offer, making certification inappropriate.

Smith attempts to distinguish Minnesota law from the MMPA, arguing that Minnesota law permits individualized inquiries while the MMPA does not, making class certification more appropriate under the MMPA. We don’t see it. While the MMPA also does not require common-law-fraud reliance, see Hess v. Chase Manhattan Bank, USA, N.A., 220 S.W.3d 758, 774 (Mo. banc 2007), the relevant Minnesota law didn’t either. See St. Jude, 522 F.3d at 839–40. And like the relevant Minnesota statute, “the plain language of the MMPA demands a causal connection between the ascertainable loss and the unfair or deceptive merchandising practice.” See Owen v. Gen. Motors Corp., 533 F.3d 913, 922 (8th Cir. 2008). The plaintiff must

-6- have bought or leased merchandise and “thereby” suffered “an ascertainable loss . . . as a result” of the deceptive practice. See Mo. Rev. Stat. § 407.025.1(1). From this language, “there is no denying that causation is a necessary element of an MMPA claim.” See Owen, 533 F.3d at 922.

The difficulty for Smith is that, for many people in his proposed class, the representations on the containers would not have caused them any ascertainable loss; instead, they got exactly what they had bargained for when they purchased the Folgers products at issue. See Vitello, 50 F.4th at 693–94. That’s because, though the representation appeared on each of these products, the conclusion seems ineluctable that a significant proportion of the proposed class did not read those representations or, if they did, did not care about them one way or the other. See State ex rel. Coca- Cola Co. v. Nixon, 249 S.W.3d 855, 862 (Mo. banc 2008); see also White v. Just Born, Inc., 2018 WL 3748405, at *3–4 (W.D. Mo. Aug. 7, 2018). And for those who read and cared about them, a significant proportion likely did not interpret the representations in the manner Smith suggests: A hefty percentage may well have read the representations and believed that the promised number of cups could be achieved only some of the time under certain conditions. Or perhaps they understood the representations to mean that other, non-recommended brewing methods could brew the promised number of cups. After all, some consumers prefer relatively weak cups of coffee. What matters is that many class members weren’t deceived, and figuring out who was and who wasn’t will require consumer-by-consumer inquiries into each class member’s individual tastes, interpretations, and circumstances, undermining the efficiencies class actions are intended to provide. So we do not think that common questions will predominate over individual ones here.

Smith contends that Folgers provides no evidence of any of this and that he has experts to support his view that some (though not all or nearly all) consumers viewed the representations as he does. To the contrary, Folgers does point to evidence in the record that purchasers did not notice or care about the representations, nor were they

-7- misled by them. Many of the proposed class representatives for the various statewide classes testified about their purchasing experiences, and Folgers directs us to several statements in their testimony indicating that the representations didn’t influence their purchasing decisions. For example, when one proposed representative was asked why she kept buying the products at issue even after she had sued Folgers, she said, “I like my coffee.” Cf. Hudock, 12 F.4th at 777. The district court discounted this kind of evidence because those proposed representatives did not include Smith or members of the proposed Missouri class. But in an order entered the same day as the certification order, the district court said that it was “not convinced consumers of coffee products in an individual state differ from those nationwide.” We agree, and so we think the district court was probably too quick to discount the testimony of the other proposed class representatives. We note that we were unswayed when the plaintiffs in Hudock made a similar argument about the defendants’ alleged evidentiary shortcomings with respect to whether individual issues would predominate. See Responsive Brief of Plaintiffs-Appellees at 47–52, Hudock, 12 F.4th 773 (No. 20-2317).

Smith also contends that individualized inquiries are unnecessary anyway because every member of the class suffered an ascertainable loss. His reasoning goes like this: The statements on the containers misled at least some purchasers and induced them to buy Folgers products. The representation thus spurred increased demand for Folgers products, and increased demand led to higher prices that every buyer had to pay. As one of Smith’s experts explained it, “[t]o the extent that the alleged misconduct in this case artificially inflated market demand, then each consumer who purchased a Contested Product paid a higher price compared to the price that consumer would have paid in the absence of the alleged deception.” As a result, the expert continued, “even if the willingness-to-pay for the Contested Product was not inflated for certain individuals (e.g., an individual [who] was not deceived by the alleged misconduct in this case and/or purchased the product for unrelated reasons), these consumers still paid an overcharge to the extent that market prices

-8- were artificially inflated due to the alleged misconduct.” At least one other circuit court appears to have accepted an argument like this one, albeit in a divided opinion that, though it applied the MMPA, did not discuss its ascertainable-loss requirement. See Tershakovec v. Ford Motor Co., 79 F.4th 1299, 1311–12 (11th Cir. 2023).

But we have already turned aside a similar price-premium argument. In Johannessohn, plaintiffs filed a putative class action against the maker of particular all-terrain vehicles on the ground that some, but not all, of the vehicles had overheated and even caught fire. See 9 F.4th at 984. We explained that only those injured by a heat defect could succeed on a claim. We rejected the plaintiffs’ contention that they could assert a claim because “they can show economic injury by the mere fact that they paid an inflated purchase price.” Their logic was as follows: “[B]ecause some buyers did not receive the benefit of the bargain, all buyers should have paid less for their ATVs. And because all class members paid sticker price, all class members suffered an economic injury.” We explained, though, that accepting that argument would allow purchasers who didn’t experience a defect in their vehicles to “piggyback” on the injuries suffered by others, which we said did “not create an Article III injury-in-fact.” See id. at 987–88. Similarly here, accepting Smith’s argument would allow those who suffered no ascertainable loss from Folgers’s representations to piggyback on the injuries of others to pursue a remedy.

The Missouri Supreme Court, moreover, hasn’t decided whether as a matter of state law it would allow consumers who weren’t deceived to piggyback on the injuries of others to assert an MMPA claim. Since there’s no ruling from the Missouri Supreme Court directly on point, our job is to predict how it would rule if confronted with the argument Smith advances here. See Academy Bank, N.A. v. AmGuard Ins. Co., 116 F.4th 768, 776 (8th Cir. 2024). We think the high court of Missouri would resolve this issue the same way the high court of New Jersey did in dealing with a New Jersey consumer-protection statute that we’ve already examined, see Hudock, 12 F.4th at 776–77, and that closely resembles the MMPA. The text of the relevant

-9- New Jersey statute also gives a cause of action to those who suffer an “ascertainable loss” as a result of an act or practice declared unlawful, including, as in Missouri, a “deception, fraud, false pretense, false promise, [or] misrepresentation.” See N.J. Stat. Ann. §§ 56:8–2, –19. The Supreme Court of New Jersey has confronted price- premium arguments like the one Smith makes and has rejected them, including in two cases we cited with approval in Hudock. See Dugan v. TGI Fridays, Inc., 171 A.3d 620, 639–41 (N.J. 2017); Int’l Union of Operating Eng’rs Local No. 68 Welfare Fund v. Merck & Co., 929 A.2d 1076, 1087–88 (N.J. 2007) (per curiam). In Dugan, for example, the Supreme Court of New Jersey said that “our case law has consistently rejected ‘price-inflation’ theories . . . as a substitute for proof of ascertainable loss or causation.” See 171 A.3d at 639; see also Harnish v. Widener Univ. Sch. of Law, 833 F.3d 298, 312–13 (3d Cir. 2016) (collecting cases from New Jersey and Delaware).

We think the Missouri Supreme Court would agree because accepting Smith’s price-inflation theory would write the ascertainable-loss requirement out of the statute and permit any purchaser of these Folgers products to bring an MMPA claim, even if the purchaser didn’t care about the representations or wasn’t misled by them. See Dugan, 171 A.3d at 640. Under Smith’s theory, a purchaser, fully aware of the representations and their potential shortcomings, could even buy the product for the sole purpose of bringing an MMPA claim on behalf of others. Surely someone with full knowledge of the facts and a willingness to make the purchase anyway cannot have suffered an ascertainable loss from the representations of the seller. See Coca- Cola Co., 249 S.W.3d at 862.

Since common questions would not predominate over individual questions with respect to the MMPA claim, we conclude that class certification was inappropriate.

Smith’s unjust-enrichment claim fares no better. To succeed on this claim, Smith must show that he conferred a benefit on Folgers, that Folgers appreciated the benefit, and that it accepted and retained the benefit under inequitable or unjust

-10- circumstances. See Hargis v. JLB Corp., 357 S.W.3d 574, 586 (Mo. banc 2011). Whether a particular transaction might be considered inequitable or unjust turns on the specific circumstances of each transaction, which again places individual inquiries front and center. We agree with the several courts and commentators who have said that, for this reason, unjust-enrichment claims are generally inappropriate for class treatment. See, e.g., Vega v. T-Mobile USA, Inc., 564 F.3d 1256, 1274 (11th Cir. 2009); see also 1 McLaughlin on Class Actions § 5:60 (21st ed. Oct. 2024 Update). There’s no reason that general rule shouldn’t apply here.

We therefore reverse the district court’s decision to certify a class and remand for further proceedings. ______________________________

-11-