less sodium soup should have less sodium than comparable soup

Smajlaj v. Campbell Soup Co. (D.N.J. Mar. 23, 2011)

Plaintiffs filed this putative class action alleging they were misled by labels on cans of Campbell’s less-sodium tomato soups and other marketing into buying those soups even though the sodium content was equal or nearly equal to that of Campbell’s regular tomato soup. Plaintiffs alleged that, during the class period, Campbell's regular tomato soup contained 480 mg of sodium per serving, which is the same amount of sodium that was contained in its 25% Less Sodium Tomato Soup, and only 2% more than the amount contained in the Healthy Request Soup. But, during the class period, some labels on both cans displayed comparisons to “OUR REGULAR PRODUCT.” Plaintiffs alleged that consumers were misled into thinking this was a comparison with the regular, 480-mg soup.

Some Less Sodium cans were labeled "25% LESS SODIUM THAN REGULAR CONDENSED SOUP," with a picture of a can of the regular tomato soup, together with the large-font words "The famous taste . . . with less salt!" This also had a small-font explanation that Campbell’s regular tomato soup uses sea salt, which results in less sodium. The large-font front-label claim on this label also had a fine-print footnote on the reverse side, saying "This product contains 480 mg of sodium versus an average of 830 mg for our regular condensed soup." Plaintiffs alleged similar misleading statements in other marketing, such as Campbell’s website, which displayed the 25% Less Sodium tomato soup together with the regular tomato soup, leading consumers to think that the comparison was to the regular tomato soup.

The court commented that, though the following is outside the pleadings and thus irrelevant to the motion to dismiss, it was helpful to note some facts alleged by Campbell: in September 2009, Campbell had only recently released the 480-mg regular tomato soup, replacing a 710-mg version. It simultaneously released new labels for the other soup lines, so the Healthy Request soup no longer made comparative statements, and the 25% Less began to compare to “regular condensed soup.” The comparisons to “our regular product,” Campbell alleged, were accurate comparisons to the old formulation of the regular product.

Plaintiffs alleged they were misled into paying more for the soups even though they were not lower-sodium than the cheaper alternative. They pled violations of the New Jersey Consumer Fraud act and breach of express warranty.

The court began by stating that it would rely on the can labels, which were part of the complaint, but not on Campbell’s additional statements about composition and timing. Boldly, Campbell argued that the court could consider this information under Iqbal, because the case endorsed reliance on “judicial experience and common sense.” In Iqbal, this led the court to rely on various facts about September 11, 2001, including that the attacks were perpetrated by 19 Arab Muslim hijackers who were part of al Qaeda. “Assuming for the sake of argument that Iqbal was endorsing the Court's consideration of certain facts outside the pleadings, the Court is not persuaded that the existence and timing of Campbell's release of a newly formulated tomato soup is akin to the general knowledge that the September 11, 2001 hijackers were Arab Muslims.”

Campbell argued preemption. The 1990 Nutrition Labeling and Education Act (NLEA) expressly
preempts inconsistent state nutritional content labeling requirements: a state may not impose any requirement respecting any claims of nutritional content on labels "that is not identical to the requirement" imposed by the Act. There is no implied preemption, and states may impose identical requirements to those of the FDCA.

FDA regulations allow “less sodium” claims when the food contains at least 25% less sodium per amount customarily consumed of an appropriate reference food and when the identity of the reference food and the percent/fraction is stated in immediate proximity to the most prominent less sodium claim. The key issue the FDA faced was to prevent misleading reference claims. The FDA therefore required that an appropriate reference food must be "a dissimilar food within a product category that can generally be substituted for one another in the diet (e.g., potato chips as reference for pretzels, orange juice as a reference for vitamin C tablets) or a similar food (e.g., potato chips as reference for potato chips, one brand of multivitamin as reference for another brand of multivitamin)."

To prevent misleading comparisons to specially-formulated or non-representative foods, the FDA concluded that an appropriate reference food is the manufacturer's regular brand, another manufacturer's regular brand, or a representative value for a broad base of foods of the particular type. If a manufacturer elects not to use a particular brand as a reference food, the generic comparator must "be representative of the type of food that includes the product that bears the claim." The FDA also required that "the nutrient value for the reference food shall be representative of a broad base of foods of that type," backed by data. If the manufacturer chooses to compare to its regular brand, that regular brand must have been “offered for sale to the public on a regular basis for a substantial period of time in the same geographic area by the same business entity or by one entitled to use its trade name." The regular product must be a "known specific food" that has been on the market long enough for consumers to be familiar with it.

Plaintiffs assumed that Campbell was permitted to use an average of unidentified condensed soups as a reference food. “This is a generous assumption since, contrary to what Campbell repeatedly and incorrectly asserts, the regulations do not expressly permit a reference food that is actually a category of foods.” So, “regular Italian salad dressing” or “creamy Italian salad dressing” can be an appropriate reference food, but there’s no guidance saying that “dressing” is. Thus, the court continued, “[t]here is no doubt that Campbell could have compared their less-sodium tomato soup to an average of condensed tomato soup …. That is very different from the position that Campbell can compare their less-sodium tomato soup to a hodge podge of their higher-sodium condensed soups.” The court was skeptical that this was appropriate: “The comparison to an average of condensed soups is akin to comparing a particular frozen meal to frozen meals generally, as if the fact that they are all frozen makes the comparison useful to the consumer.”

Still, plaintiffs accepted the selection of an average of condensed soups as a reference food, “except insofar as the unusual reference makes it less likely for a consumer to have read the labels as Campbell claims they intended.” Plaintiffs argued that the FDA regulation’s requirement that "[t]he label or labeling must state the identity of the reference food" requires a nonmisleading identification, especially given the FDCA’s background prohibition against labeling that is “false or misleading in any particular.”

The court agreed that plaintiffs were only trying to enforce rules consistent with FDA requirements. FDA regulations say a company can violate the regulations even if it uses the comparison language suggested by the FDA, if the company uses claims that are misleading because of their "overall context or presentation." Here, the allegation was that it is misleading to use a composite of condensed soups as the basis for a comparison with “regular condensed soup” juxtaposed with a picture of regular tomato soup and the phrase "The famous taste . . . with less salt!" “This cause of action is consistent with and requires nothing more than the FDA's requirement that the reference food be non-misleadingly identified.” A reasonable consumer might take away a false comparison to regular tomato soup from this and from the comparison to “our regular product.”

The court disregarded the footnote on the label referring to “an average of our regular condensed soup.” First, the FDA requires the identification of the reference food to appear in close proximity to the claim. Second, even the fine print “confusingly uses the singular ‘soup,’ when it refers to a composite of soups. It also states that the 25% less sodium soup has 480 mg of sodium compared to 830 mg for the ‘condensed soup,’ a difference of much greater than 25%, reasonably leading to confusion.”

Because Campbell’s allegations about its reformulated regular soup were not properly before the court, the court didn’t reach the question of whether the FDCA preempts a claim based on labeling misleading solely as a consequence of the release of a new “regular” product while old labels continued to make comparisons to the product being phased out. The court didn’t think preemption was obvious, and anyway plaintiffs claimed that even the new label was misleading.

The FDCA did preempt plaintiffs’ omission claims based on the idea that failure to disclose a comparison with regular tomato soup was misleading. The regulations require only a non-misleading identification of a proper reference food, not any additional comparisons.

The New Jersey Consumer Fraud Act requires plaintiffs to allege: (1) unlawful conduct; (2) an ascertainable loss; and (3) a causal relationship between the unlawful conduct and the ascertainable loss. While some NJCFA claims may not require pleading that satisfies Rule 9(b)—not every such claim involves an affirmative misrepresentation or material omission—the court applied Rule 9(b) here because it is meant to protect the reputations of manufacturers in circumstances such as these, where the gravamen of the claim is that the consumer was induced to buy a product because of a false or misleading representation. Neither intent nor reliance are required to show a violation of the law, however.

Campbell argued that its labels were literally true, either with respect to the old formulation of the regular tomato soup or to an average of condensed soups, and thus could not be unlawful. The court disagreed. The court disagreed. First, the key factual premises rested on facts outside the pleadings. Second, literal truth doesn’t necessarily avoid misleadingness to a reasonable consumer.

Ascertainable loss: this term isn’t defined in the statute or the legislative history. Campbell argued that using a misrepresentation to cause a consumer to buy a product doesn’t cause an ascertainable loss sufficient to constitute injury under Article III, much less under the NJCFA. The court begged to differ. Being fraudulently induced to buy a product causes an injury, and it becomes “ascertainable” when the consumer can quantify the difference in value between what was promised and what was received. “The New Jersey Supreme Court has repeatedly and explicitly endorsed a benefit-of-the-bargain theory under the Consumer Fraud Act that requires nothing more than that the consumer was misled into buying a product that was ultimately worth less to the consumer than the product he was promised…. [T]here is no requirement that the product actually received be defective or deficient in any way other than that it is not what was promised.” Thus, if a seller promises a racecar and delivers a well-running and spacious minivan, the consumer has a fraud claim.

A benefit-of-the-bargain claim requires (1) a reasonable belief about the product induced by a misrepresentation; and (2) that the difference in value between the product promised and the one received can be reasonably quantified. There must be a direct causal connection between the misrepresentation and the plaintiff’s expectations. That was properly pled here.

Ascertainable loss must be supported by some evidence that is not hypothetical or illusory and must be capable of calculation. Plaintiffs pled such loss by stating what they paid for the soup, which they thought had less sodium than the cheaper alternative, and subtracting the price of the regular tomato soup. The difference in value between what was received and what was promised is the measure of damages under the NJCFA as well as an ordinary warranty claim, and it’s often calculated by reference to the market price of buying or making the product received like the one warranted.

Here, where neither replacement cost nor cost of modification to improve the product until it performed as represented would be useful, the plaintiffs’ alternative calculation was perfectly acceptable. Just as the standard approach doesn’t require the plaintiff to actually pay for replacement or modification to make out a NJCFA claim, plaintiffs didn’t have to plead they would actually have purchased the reference product. These are approaches to valuing injury, not the injury itself. And the valuations don’t have to be perfect, as long as they provide a reasonable basis that is not speculative or unquantified. “Plaintiffs' approach to valuation need not assume that the prices of these products are the result of every consumer being concerned about the same product characteristics as Plaintiffs, because the prices are not being used to show how much the market values certain amounts of sodium, but are instead being used to objectively measure the values to Plaintiffs of the product promised and product received.” (The inverse of fraud on the market, I guess.) According to plaintiffs, “the soup they paid for is identical, for their purposes, to soup that is 20 to 80 cents cheaper, and they bought it based on a misrepresentation.” That was sufficient ascertainable loss, and indeed it was difficult to imagine a more precise or objective way to value food products at the pleading stage.

The same arguments preserved plaintiffs’ express warranty claims against Cambpell’s “meritless” objections.

Defendants argued that plaintiffs failed to specify whether they saw the label of the 480-mg regular tomato soup, so they couldn’t have been misled. But the statements could be misleading regardless. “If a manufacturer represents that a higher-priced television has a 25% higher resolution than the manufacturer's regular 50" television, when in fact the two have the same resolution, it is not necessary for the misled consumer to know that the regular 50" television also had 1920 x 1080 resolution in order for the consumer to have been misled by the claim.” Consumers are not required to examine the comparable product in order to determine whether to accept the seller’s representations about the comparison.

With the exception of some too-vague statements about “marketing materials,” the complaint survived.