A few months ago, we analyzed the Ninth Circuit’s decision in Webb v. Trader Joe’s Company, No. 19-56389 (June 4, 2021), which held that a private plaintiff’s challenge to poultry labeling claims were preempted by federal law, warranting dismissal at the pleadings stage.  However, the Ninth Circuit’s recent decision in Cohen v. ConAgra Brands, Inc., No. 20-55969 (Oct. 26, 2021), declined to apply preemption in a similar challenge to labeling claims on poultry products.  In this post, we examine the two decisions and conclude that, notwithstanding Cohen, Webb remains the benchmark for future litigation in this area.

In Webb, a consumer objected to label statements on Trader Joe’s All Natural Boneless Chicken Breasts, All Natural Chicken Thighs, and All Natural Chicken Wings products, which said the products contained “[u]p to 5% retained water.”  According to Webb, that statement was misleading because, based on independent testing she had commissioned, the percentage of retained water was greater than claimed on the labels.  Trader Joe’s moved for judgment on the pleadings, arguing that Webb’s claims were preempted by the federal Poultry Products Inspection Act (PPIA), which prohibits states from imposing requirements “in addition to, or different than those” described under federal law. 

The district court agreed that Webb’s claims were preempted and dismissed the case, and the Ninth Circuit unanimously affirmed.  The panel reasoned that federal regulations require the Food Safety and Inspection Service (FSIS) to preapprove the labels of all poultry products that include certain “special statements.”  The Trader Joe’s product labels had three special statements—“no antibiotics ever,” “no added hormones,” and “all vegetarian fed”—which by law triggered FSIS review of the entire product label.  The panel agreed with the district court that it was proper to take judicial notice of the label, since its authenticity was undisputed.  And although there was no evidence that FSIS had endorsed those specific labeling statements, the panel held that FSIS’s approval could be inferred from the fact that FSIS, by regulation, necessarily reviews the “entire label” before product labeling can be used in commerce, and it did not object to Trader Joe’s’ labels.  It followed that “[a]ny additional label requirements Webb seeks to place on Trader Joe’s through application of her retained water data would necessarily be ‘different than’ those required by the PPIA, and her claims are thus preempted.”

Four months later, the Cohen court ruled on a consumer challenge regarding the labeling of ConAgra chicken products.  Cohen claimed the labeling statements “Made with 100% Natural White Meat Chicken,” “No Preservatives,” “No Artificial Colors,” “No Added Hormones,” “No Artificial Flavors,” and “0g Trans Fat per Serving” were false and misleading due to the alleged presence of three “synthetic” ingredients in the products.  ConAgra moved to dismiss the complaint, arguing that Cohen’s claims were preempted because the challenged labels included multiple special statements (including the same “no added hormones” statement from Webb), and thus FSIS had necessarily reviewed the entire label and concluded that it complied with federal law.

The district court agreed and granted ConAgra’s motion to dismiss.  Consistent with the approach in Webb, it took judicial notice of an image of ConAgra’s front label found on the internet, and concluded that FSIS had “approved the labeling of the Chicken Products, including the specific representation challenged by [Cohen].”  As such, the district court deemed Cohen’s challenge preempted.

The Ninth Circuit, however, reversed.  The panel began by reaffirming many of the principles stated in Webb:  (1) all poultry labels with special statements “must be submitted to FSIS in the form of a final label for approval”; (2) “when the agency reviews and approves a label, the agency is deciding that it is not false or misleading under the PPIA”; (3) “allowing private consumers to second-guess the agency’s decisions through state law claims against producers would both circumvent that pre-approval process and conflict with the PPIA’s goal of national uniformity”; and, (4) as a result, “if ConAgra’s labels were reviewed and approved by FSIS, then Cohen’s claims challenging the labels would be preempted.”

However, the panel ultimately declined to dismiss Cohen’s claims because “there [were] no affidavits or other documentary evidence showing that the label was submitted to and approved by FSIS.”  The panel explained that preemption is an affirmative defense, and “the mere existence of the label is insufficient to establish that it was reviewed and approved by FSIS.”  Although the panel recognized that in Webb, “label evidence alone was enough to conclude that a retained water claim was federally approved,” it found Webb distinguishable because the plaintiff there “did not challenge whether the label was reviewed by FSIS.”  Cohen, by contrast, “contend[ed] that ConAgra used the generic approval process for its labels, improperly bypassing FSIS review.”  In other words, unlike Webb, Cohen had disputed that the use of the label in commerce was itself evidence that FSIS had approved it. 

Based on this reasoning, the panel concluded that ConAgra was required to “produce evidence that the label was reviewed and approved by FSIS,” and it remanded for further fact-finding on that particular point.  The panel stressed the “limited nature of [its] remand,” directing the parties to engage in narrow discovery on whether ConAgra’s labels had, in fact, been approved by FSIS.  As the panel explained, “this is hardly a significant burden”:  “If the evidence shows that ConAgra’s label was approved by FSIS, then Cohen’s claims are preempted.  Cohen may not try to argue or show that FSIS’s approval decision was wrong.” 

In two separate footnotes, the panel noted that after oral argument, ConAgra had filed, pursuant to Appellate Rule 28(j), a product application with an FSIS approval stamp—seemingly clear evidence that the product had in fact received FSIS approval.  The court explained that this belated submission was procedurally improper, but strongly hinted that the document would otherwise be sufficient for ConAgra to establish the preemption defense.

In our view, Cohen’s basis for distinguishing Webb does not hold (or in this case, retain) water:  the panel relied entirely on the supposed fact that Cohen (unlike Webb) asserted that ConAgra improperly bypassed FSIS review.  However, Cohen’s complaint made no mention of the FSIS review process, let alone included any plausible allegation that the process was not followed in this instance.  Indeed, the panel acknowledged that Cohen did not “challenge[] the facts underlying the agency approval process.”  In our view, without such an allegation, the court should have reached the same conclusion as in Webb, i.e., FSIS’s approval of ConAgra’s labels can be inferred from the fact that FSIS, by regulation, necessarily reviews the labeling before they can be used in commerce.  That determination would have required dismissal on preemption grounds, as the panel itself acknowledged.

Regardless, Cohen presents a unique factual and procedural history that is unlikely to recur.  It was unusual for the panel to have credited Cohen’s claim of an improper FSIS review process without any such allegation having appeared in the complaint.  Moreover, in future cases, it is unlikely that a plaintiff will be able to plausibly allege, consistent with Rule 11, that the FSIS review protocol required by regulation was not followed for a given poultry label.  And even if a plaintiff could so allege, Cohen offers a roadmap for quick resolution by authorizing manufacturers to submit documentary evidence of FSIS pre-approval when seeking dismissal on preemption grounds, prior to engaging in costly and protracted discovery.  Technically speaking, the submission of such extra-pleadings evidence might result in conversion of a defendant’s motion to dismiss into one for summary judgment—but the Cohen panel’s decision plainly supports this abbreviated procedure, so plaintiffs should not be able to complain that summary judgment is premature in such cases.

Thus, even after Cohen, Webb remains good law, and consistent with the long line of decisions that have dismissed claims due to federal-law preemption at the pleadings stage, without requiring development of a factual record.[1] For these reasons, we expect courts to continue to follow Webb and to dismiss poultry labeling challenges at the pleadings stage.