Categories & Search

In Butter-Flavored Spray Case, Federal Labeling Preemption Sticks In the End

Federal food-labeling laws preempt state laws that impose requirements different from or in addition to those established by federal law.  In some cases, the FDA has spoken directly to a labeling issue by regulation, and if the food manufacturer is in compliance with that regulation, any state-law liability should be preempted.  Careful plaintiffs often try to draft their allegations to get around a federal regulation that would otherwise preempt their claims.  For instance, in challenging a defendant’s representations concerning honey in a cereal, a plaintiff avoided the defendant’s compliance with the federal labeling regulation on “flavoring” by alleging she was deceived about the relative amount of honey as a sweetener (which is not covered by a specific FDA regulation), rather than the relative amount of honey as a flavoring agent (which is covered).  When courts allow creative pleading to circumvent a preemption defense, defendants are deprived of the protections that Congress intended to provide them under federal labeling law, at least at the outset of the case.  But as a recent decision shows, defendants may be able to renew and succeed on a preemption defense after discovery shows plaintiff’s artful allegations were just that.


In Allen v. Conagra Foods, Inc. the plaintiffs took issue with ConAgra’s advertising for its butter-flavored spray, “Parkay Spray.”  The plaintiffs alleged that Parkay Spray was represented as “fat free” and “calorie free,” and that this was misleading because a 226-gram bottle actually contains 93 grams of fat.  2013 WL 4737421, *1 (N.D. Cal. Sept. 2, 2013).  All the way back in 2013, ConAgra moved to dismiss, arguing that federal labeling regulations allowed it to round the fat and calorie content of its product down to zero.  The relevant regulation, 21 C.F.R. § 101.12, sets forth the “reference amounts” customarily consumed per eating occasion (i.e., the serving sizes) for various categories of food products.  Under this regulation, ConAgra argued, a serving of Parkay Spray was so small that it could properly be declared fat-free and calorie-free – no matter how many calories and grams of fat were present in the whole bottle – and the plaintiffs’ claims were therefore preempted by the FDA’s labeling regulation. 

To evaluate this argument, the then-presiding judge first had to determine the proper serving size of Parkay Spray under FDA regulations.  ConAgra argued that Parkay Spray was a “spray type” “Fat and Oil” under those regulations, like Pam® cooking spray, and therefore the reference amount was 0.25 grams per serving, which would justify ConAgra’s “fat free” and “calorie free” representations.  By contrast, the plaintiffs argued that Parkay Spray was a “butter, margarine, oil, [or] shortening” under those regulations, which—since it is typically consumed in greater quantities than a cooking spray—had a larger serving size, rendering ConAgra’s representations non-compliant.

In denying ConAgra’s initial motion to dismiss in 2013, the court reasoned that the plaintiffs’ claims were not preempted because they had “marshal[ed] substantial support for [their] allegation that Parkay Spray should be categorized with other imitation butter products”—for example, the representation that the product was “a substitute for butter that shares butter’s ‘Fresh and Creamy Taste’ without the negative health consequences”; the product’s display of a picture of corn; its listing of “serving sizes…for cooking and for topping”; and an invitation on the packaging for “customers to visit the Parkay Spray website ‘for delicious recipes.’”  2013 WL 4737421, at *7.  The court also found support for the plaintiffs’ position in the FDA’s 1994 “Guide to Nutrition Labeling and Education (NLEA) Requirements,” which included in the “butter, margarine, oil, shortening” category “[a]ll types of butter and margarine …; spreads, oils; and shortenings,” and included under “spray types” “Nonstick cooking sprays (e.g., Pam).”  Id. 

In 2018, following class discovery and the plaintiffs’ filing of their second amended complaint, ConAgra moved to dismiss again, citing, among other things, changes in the relevant FDA guidance.  The court (with a new judge presiding) again held that the plaintiffs “adequately alleged that Parkay Spray is imitation butter and belongs in the same reference amount category as butter.”  2018 WL 6460451, *9 (Dec. 10, 2018).  Although the FDA had since broadened its guidance on the “spray types” category to include “all types of cooking sprays (e.g., cooking spray olive oil),” the court reasoned the “changes are not as dramatic as ConAgra suggests,” as the “butter,” etc. category “still include[d] ‘liquid’ as a possible form of the butter and margarine spread.”  Id. at *10 (emphasis added).  The court distinguished another case which found that “I Can’t Believe It’s Not Butter! Spray” belonged in the “spray type” category under the regulations.  The court reasoned that “[t]he plaintiffs here do not simply argue that some consumers choose to use Parkay Spray as a topping but rather that ConAgra markets it to be used in that way by communicating to consumers that it is a butter substitute,” and thus found that the plaintiffs plausibly alleged it was a butter “imitation product.”  Id. (distinguishing Pardini v. Unilever United States, Inc., 2014 WL 265663 (N.D. Cal. Jan. 22, 2014) (emphases added). 

In 2020, after merits discovery concluded, ConAgra moved for summary judgment, arguing yet again that the plaintiffs’ claims were preempted under the FDA’s labeling regulation.  The third time was the charm.  As before, the plaintiffs urged that Parkay Spray fell into the “butter/margarine category” because “(i) Conagra intends Parkay Spray to be used as a buttery topping, (ii) Conagra markets Parkay Spray as an alternative to butter for foods like corn and bread, and (iii) consumers in fact use Parkay Spray as a topping.”  2020 WL 4673914, *3 (Aug. 12, 2020) (emphases added).  In finally rejecting this argument, the court relied on the plaintiffs’ own expert, who cited “numerous characteristics [that] distinguish Parkay Spray from butter and margarine,” including that Parkay Spray is more a liquid oil than butter, a solid fat; “has a significantly lower fat content”; “is dispensed through a pump,” which would “probably not be possible for butter or margarine unless either was very warm”; and differs in color.  Id.  Moreover, the plaintiffs’ expert agreed that Parkay Spray “is unsuitable for use in food preparation, which was the ‘major intended use’ the FDA contemplated for butter, margarine, oil, and shortening when it set their reference amount.”  Id.  There was thus “no doubt that butter and Parkay Spray cannot be ‘used interchangeably.’”  Id. (quoting 21 C.F.R. § 101.13(d)).  Unlike before, the court on summary judgment did not delve into consumers’ alleged use or ConAgra’s marketing to decide the preemption question—apparently satisfied by the testimony of plaintiff’s expert that, as a matter of fact based on the products’ characteristics, Parkay Spray could not actually be used in the same manner as butter. 

Our Analysis

In one sense, Allen is a welcome outcome.  Although ConAgra was unable to rely on its compliance with food-labeling laws to avoid nearly a decade of litigation, it was at least able to prevail after discovery by proving that its product should be categorized as a “spray type,” and thus that its “fat free” and “calorie free” claims complied with federal labeling law. 

At the same time, the victory is bittersweet.  Federal labeling preemption exists in part to provide food manufacturers some sense of security and prevent costly litigation—yet undisputed compliance with federal law could not prevent that burden here.  As the court recognized, the seven-year-old case had a “long and complicated history” in which ConAgra had “argued vehemently since this case’s inception that Parkay Spray [was] a spray-type fat and oil under the federal regulations[.]”  Id. at *1.  And in the end, the court agreed with ConAgra, not because anything had changed about the product itself, but because the court shifted the focus of its analysis from the plaintiffs’ allegations about consumer use and ConAgra’s marketing to facts about the product’s actual characteristics.

A better approach—and the required approach under Supreme Court precedent—is for courts to treat preemption as a pure question of law, rather than one that requires intensive factual investigation.  See Merck, Sharp & Dohme Corp. v Albrecht, 139 S. Ct. 1668, 1679 (2019) (“[T]he question [of preemption] is a legal one for the judge, not a jury”). Courts should focus on concrete characteristics from the outset, and they should construe the FDA’s regulations in accordance with their plain text.  Such an approach not only can save years of litigation, but can provide fairer notice to food manufacturers and consumers as to how a product should be categorized.  This approach also would give less discretion to courts (and clever plaintiffs’ lawyers) and more to the FDA, where it ought to reside. 

Pardini, the case distinguished in Allen, shows such an analysis in action.  There, as in Allen, the plaintiffs alleged the product (I Can’t Believe It’s Not Butter!) was not a “spray type” because “consumers use” it as a topping and “don’t really use [it] as a lubricating cooking spray”; and because “images on [the product’s] label (corn on the cob) and recipes on Defendant’s website all support a conclusion that the product should be categorized as a topping, not a spray.”  2014 WL 265663, at *5.  The court found these arguments irrelevant to the categorization exercise, reasoning that the butter spray unambiguously fit within the more-specific “spray type” category, even if it could also fit within the more general “butter/margarine” type category.  Id.  And, critically, the court declined the plaintiffs’ proposed “state of regulatory affairs in which any consumer could overcome a motion to dismiss in a case like this one by insisting that people consume more (or less) of a product these days, rendering all sorts of products mislabeled at a consumer’s whim.”  Id.

In other words, from the beginning, the Pardini court focused on concrete product characteristics, and it assumed that the FDA’s regulations were intended to and did provide a clear answer as to how such a product should be labeled.  While Allen eventually reached the same result, the court’s ultimate shift—from considering plaintiffs’ allegations concerning consumer use and marketing to actual product characteristics—shows that Pardini had it right in the first place.  That approach not only better ensures notice to food manufacturers and consumers about the governing regulations and better ensures that responsibility for these determinations rests with the FDA, rather than courts.  It also better comports with Congress’ intent in passing federal labeling laws that food manufacturers in compliance can avoid costly litigation.