Bread and lawsuits: consumer beliefs about what "local" means don't matter

Bimbo Bakeries USA, Inc. v. Sycamore, Nos. 18-4062, 19-4031,19-4040 (10th Cir. Mar. 18, 2022)

Previous discussion (one of four opinions in the case that I blogged).

Bimbo sells Grandma Sycamore’s Home-Maid Bread; it sued defendants, who include the baker who developed the Grandma Sycamore’s recipe, for trade secret misappropriation, trade dress infringement, and false advertising when it sold a comparable bread product, Grandma Emilie’s, with the tagline “Fresh. Local. Quality.” The district court kicked out the trade dress claim on summary judgment and sent the other claims to the jury, which returned a verdict favoring Bimbo on both and awarded over $2 million in damages for trade secret misappropriation, which the court increased by nearly $800,000 for willfulness. The district court remitted the false advertising damage award from over $8,000,000 to under $85,000 to reflect that it wasn’t sure whether Utah residents (the ones surveyed) had the same definition of “local” as others. The court of appeals affirmed on trade dress and reversed on the other claims, leading to total defeat for Bimbo.

Grandma Sycamore

Grandma Emilie

Grandma Sycamore “seeks to emulate bread made at home—for example, the bread is baked in a special pan so that two loaves can be pulled apart, giving the bread a ‘breakaway side’”—and has been “immensely popular” in Utah. Hostess sold a homemade bread product in Utah known as Grandma Emilie’s, then went bankrupt in 2012; defendant USB acquired a Salt Lake City bakery, several warehouses throughout Utah, and the Grandma Emilie’s line. USB soon decided to make Grandma Emilie’s in-house and closed its Salt Lake City bakery. When it chose packaging, it “took into consideration” the Grandma Sycamore’s packaging. It used the “Fresh. Local. Quality” tagline at points of sale and placed the tagline on its trucks. Along with Salt Lake City, U.S. Bakery had bakeries in Alaska, Idaho, Montana, Oregon, and Washington. “It sold bread products under the ‘Fresh. Local. Quality.’ tagline throughout those states regardless of where the products were baked, as well as in California and Wyoming, before moving away from the tagline in mid-2015.” It briefly used “Freshly Baked in Utah” until it shut down the Salt Lake City bakery.

Trade dress: Bimbo claimed the following elements: “(1) a horizontally-oriented label; (2) a design placed at the top center of the end; (3) the word ‘White’ in red letters; (4) the use of a red, yellow, and white color scheme; and (5) stylized font below the design outlined in white.” Based on the packaging for Grandma Sycamore’s and its competitors in the homemade bread products space, the district court found “[t]he combination of [Bimbo Bakeries’] purported trade dress is clearly generic” and concluded it was “not subject to protection” under the Lanham Act because combining those elements was “the custom in the industry.” The court of appeals agreed, rejecting Bimbo’s argument that the trade dress was protectable because consumers associate it with Grandma Sycamore’s; Bimbo Bakeries spent millions of dollars advertising its product; and USB intentionally copied it.

USB submitted evidence that the homemade bread products that compete with Grandma Sycamore’s “all tend to combine the purported trade dress elements.” “The purported trade dress claimed by Bimbo Bakeries is thus defined at a broad enough level of generality to sweep in its competitors. Bimbo Bakeries may well have a protectable Grandma Sycamore’s trade dress, but its claim in this litigation extends far beyond its product’s more specific attributes.” Bimbo’s arguments didn’t counter the evidence of what was customary in the category. Though it offered a survey, the court doubted its value and reliability: “

The expert asked consumers whether they thought a picture of Grandma Sycamore’s exact package with just the “Grandma Sycamore’s” name removed (but several other features more specific than the purported trade dress, including the “Home-Maid Bread” marker, still on it) came from a single source, rather than asking about the far more general purported trade dress alleged by Bimbo Bakeries. Recognizing the Grandma Sycamore’s packaging with two words omitted is not the same as associating every iteration of the highly general purported trade dress claimed by Bimbo Bakeries with solely Grandma Sycamore’s.

[Discussion of trade secret claim omitted; the court’s opinion is heavily redacted even though the court of appeals agreed that “[n]o reasonable juror could find that the compilation is not generally known or readily ascertainable.” Then what justifies the redaction?]

Following Mead Johnson and similar cases, the court found that “local” was not a falsifiable factual claim, no matter what Bimbo’s survey said. Local has too many possible meanings: “that a company hires local workers, that it uses local materials, that it is locally based, that it participates in outreach efforts with local organizations, or that it donates money to local causes.”  But even assuming that it meant manufacture,

the word lacks any specific objective meaning beyond the general concept it conveys. Definitions of ‘local’ and views about whether something is “local” vary wildly, so the word’s usage in marketing can only communicate U.S. Bakery’s position that its products are local. … Without more, then, the veracity of a locality claim cannot be judged in an empirically verifiable way. Locality is fundamentally subjective. Without any consensus definition of what “local” objectively means, we are unable to conclude that a claim of locality admits of being proven true or false.

Also, it wasn’t clear how large the area defined by “local” could be, or its borders. Something could be local but just across the state line, or from the other side of the same state and not local.

[Here’s the question, though: What if a substantial number of reasonable consumers agree on one definition, while substantial numbers of reasonable consumers don’t have a definition or have other definitions? That is, the survey might find that, to a subset of consumers, the word does have a specific objective meaning. Usually, deceiving a substantial number of reasonable consumers suffices for liability. The survey is the “more,” or could be if courts were willing to consider reasonable consumers to be actual consumers—which they sometimes do!

Anyway, “local” was just opinion, not specific or detailed enough to be falsifiable. “Drawing a particular and verifiable factual meaning from as vague a word as ‘local’ … requires ‘a more specific assurance,’” and the court found none here, because the survey didn’t help. “All it can do is juxtapose U.S. Bakery’s own opinion about when something qualifies as ‘local’ with the opinions of its customers. And while they may often differ, it is the very essence of opinions that they differ.”

Comment: It’s not so much that this is obviously wrong, but it gives the court the power to disregard any empirical evidence that ordinary consumers think differently than the court does. What are the boundaries of this argument?  Would the court find that products imported from France could be “local” (because we might be speaking in galactic terms)? But even setting that aside, what kinds of statements are falsifiable by evidence of consumer reception? The thing that annoys me about this line of cases is that they never tell you what’s on the other side. If the answer is “things where the consensus definition is shared by essentially 100% of the population,” then we are no longer allowing misleading claims to be governed by the Lanham Act—and our survey thresholds are all wrong. For example, in the McNeil v. Pfizer “Listerine is as effective as floss” case, the court rejected Pfizer’s argument that “effective” was not falsifiable and credited the survey showing that about a third of consumers took away the message that Listerine could replace floss—not for nothing, roughly the deception rate in the survey here. But why doesn’t the floss survey just indicate that opinions differ about what “as effective” means, as it is the essence of opinions to do?

The court addresses people like me by worrying that Bimbo’s theory would allow “a bakery selling ‘local’ bread in a supermarket one block away from its production line [to] be liable for false advertising whenever customers interpret ‘local’ as baked in-store, as several survey respondents did here.” [Why doesn’t the “substantial number of reasonable consumers” standard guard against this risk?] “That is a difference of opinion, not commercial deception, and we share our sister circuits’ concerns with using consumer surveys to transform non-actionable statements into Lanham Act claims of fact.” “If a statement is not one of fact, it is legally irrelevant whether consumers agree with it.” We don’t care what consumers think about the matter, because we don’t trust them to give good answers to surveys, because surveys are unpredictable and could produce liability “for a wholly unanticipated claim the advertisement’s plain language would not support,” and the resulting “unintended” liability would chill commercial speech.

Again, this concern is not unfounded, but I don’t see what the limiting principle is. If there are three reasonable meanings, one of which is false, is a survey ok? What about five? I also think this logic makes nonsense of the black-letter requirement that “misleadingness” be proven by extrinsic evidence. We don’t trust surveys, except that we require them (unless we agree there’s falsity by necessary implication). We should get rid of these epicycles, which have zero basis in the statute, and let the factfinder consider all the relevant evidence. But, of course, courts don’t trust juries.  [Oh, and also, about that “unintended” liability: The Lanham Act is supposedly strict liability, at least if you are a trademark defendant. False advertising, courts aren’t so sure!]

Anyway, putting “local” within a tagline using two other words that didn’t provide any more specific meaning to “local” (as “baked in Utah” briefly did) didn’t change anything. “All three words in the tagline are inchoate buzzwords, ‘unquantifiable and subject to an individual’s fancy.’”

OK, so read the following passage and tell me: which claims are subject to proof of implicit falsity by survey, if “some segment” means “28%”?

It is possible that, as Bimbo Bakeries argues, some segment of consumers (1) interpreted the “Fresh. Local. Quality.” tagline’s use of the word “local” as an assertion that U.S. Bakery’s products were locally baked, (2) determined that locally baked meant baked within the state of sale, and (3) found that false fact material to the decision to purchase the product. But not every subjective interpretation of ambiguous language is actionable false advertising.  When the language in question is incapable of objective verification as to truth or falsity, it is not a statement of fact, and no amount of misunderstanding will give rise to an action under the Lanham Act.

Remember, this is about language, not about whether the underlying claim received by consumers—this bread was baked in Utah—is “[]capable of objective verification.”

‘When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean—neither more nor less.’

‘The question is,’ said Alice, ‘whether you can make words mean so many different things.’

‘The question is,’ said Humpty Dumpty, ‘which is to be master—that’s all.’

I suspect the court thinks that this Alice in Wonderland quote supports its position—but only because the court is the master, not consumers.

Anyway, all Bimbo’s claims were too broad: it had a protectable trade secret and a protectable trade dress, but claimed them too broadly for protectability, and “local” was also “too broad to convey anything verifiable …. Bimbo Bakeries’ claims fail as a matter of law because of their breadth.”