Alpo Petfoods, Inc. v. Ralston Purina Company

Opinion for the court filed by Circuit Judge THOMAS.

In this case, Ralston Purina Co. and ALPO Petfoods, Inc., two of the leading dog food producers in the United States, have sued each other under section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a) (1982) (amended 1988), 1 alleging false advertising. ALPO asserts that Ralston has violated section 43(a) by claiming that its Puppy Chow products can lessen the severity of canine hip dysplasia (CHD), a crippling joint condition. Ralston, for its part, attacks ALPO’s claims that ALPO Puppy Food contains “the formula preferred by responding vets two to one over the leading puppy food.”

*961 After a sixty-one-day bench trial, the district court decided that Ralston’s CHD-related advertising and ALPO’s veterinarian preference advertising both violated section 43(a). ALPO Petfoods, Inc. v. Ralston Purina Co., 720 F.Supp. 194, 209-11 (D.D.C.1989). The court permanently enjoined both companies from making “advertising or other related claims” similar to those held false, and ordered both parties to disseminate corrective statements. Id. at 216-17. Applying section 35(a) of the Lan-ham Act, 15 U.S.C. § 1117(a) (1982 & Supp. V 1987) (amended 1988), the court also awarded ALPO $10.4 million (plus costs and attorneys’ fees). The court reached this figure by determining the amount that Ralston spent on its CHD-related advertising, using that amount as a measure of Ralston’s benefit from the advertising, and then doubling the amount to capture the full harm that the advertising caused ALPO. ALPO, 720 F.Supp. at 215 (citing U-Haul Int’l, Inc. v. Jartran, Inc., 793 F.2d 1034, 1037 (9th Cir.1986)). Ralston, in contrast, was awarded only its costs and attorneys’ fees. Id. at 215, 216.

Ralston appeals the district court’s judgment, focusing on the court’s determination that the CHD-related advertising claims were false, as well as the court’s monetary award to ALPO, its refusal to award similar relief to Ralston, and its broad and expansively implemented injunction. 2 Convinced that the court properly applied section 43(a) and found facts that are not clearly erroneous, we affirm the court’s conclusion that both Ralston and ALPO have violated section 43(a). On several issues concerning remedies, however, we vacate and remand. The monetary award to ALPO is an award of Ralston’s profits, rather than ALPO’s actual damages, yet we do not see in Ralston’s conduct willful, targeted wrongdoing, which is what an award of profits requires. See Foxtrap, Inc. v. Foxtrap, Inc., 671 F.2d 636, 641-42 (D.C.Cir.1982) (per curiam). ALPO is entitled to its actual damages, however, and we try below to clarify what an award of actual damages under section 35(a) can include. Because ALPO, too, has violated section 43(a), Ralston is entitled to any actual damages that it can prove; in a case involving a meritorious claim and a meritorious counterclaim, section 35(a) does not authorize a court to compensate only the party considered less blameworthy. Attorneys’ fees, in contrast, are available under section 35(a) only “in exceptional cases,” which this court has defined as cases involving willful or bad-faith conduct. Lanham Act § 35(a), 15 U.S.C. § 1117(a); see Reader’s Digest Ass’n v. Conservative Digest, Inc., 821 F.2d 800, 808 (D.C.Cir.1987). This case does not fall into that category, so we reverse the district court’s decision to grant attorneys’ fees. Finally, we instruct the district court to enter an injunction more closely tailored to the harm posed by any repeat of Ralston’s false advertising.

I. Background 3

A. Ralston’s CHD-related Advertising

Ralston manufactures and sells Puppy Chow, which, in its two versions (dry and Chewy Morsels), is the leading puppy food sold in the United States. In September 1985, Ralston changed the formula for Puppy Chow. At the same time, it began running print ads stating that the new Puppy Chow formula could reduce the laxity of (i.e., extra space in) dogs’ hip joints and improve the fit of those hip joints, thereby lessening the severity of CHD. See ALPO, 720 F.Supp. at 196 n. 2, 198, 202 (describing CHD and explaining relationship between hip joint laxity and CHD). For over a year, Ralston directed this print advertising at veterinarians, breeders, dog enthusiasts, and others interested in dog nutrition. In addition, Ralston ran a thirty-second Puppy Chow commercial on national television *962 networks from June 1986 through October 1986, claiming that Puppy Chow “help[s] critical bone development.” Id. at 200; see id. at 199-200. To at least some extent, all of the ads claimed that feeding Puppy Chow to puppies could ameliorate or prevent CHD in those puppies. See id. at 198-200 (quoting Ralston’s direct and indirect claims regarding a Puppy Chow diet’s effects on CHD).

Ralston’s claims had a weak empirical basis. The hypothesis behind Ralston’s CHD-related product change and advertising was the “anion gap theory” of Dr. Richard Kealy, a Ralston nutritionist. This theory holds that the smaller the difference between the chlorine content and the combined sodium and potassium content of a dog’s diet, the more snugly the dog’s hip joints will tend to fit. Beginning in 1980, Dr. Kealy conducted a series of studies exploring the effect of a low-anion-gap diet on dogs’ hip joint fit. In 1984, results of Dr. Kealy’s first four studies (Trials I through IV) 4 led Dr. Kealy to prepare a monograph that reported a connection between a low-anion-gap diet and reduced hip joint laxity. At about the same time, Dr. Kealy briefed Ralston’s marketing executives on his findings, eventually leading Ralston to reformulate Puppy Chow and make its CHD-related advertising claims. In mid-1985, however, Dr. Kealy began Trial V, his first long-term study of the effects of a low-anion-gap diet. Although the parties now dispute whether the results of Trial V were statistically significant, these results undermined Ralston’s CHD-related claims so much that Dr. Kealy ended the study, which he had projected would last for almost three years, after only thirty-three weeks. After reviewing Ralston’s research findings and the conflicting expert opinions on those findings’ statistical significance, the district court found that the anion gap theory lacked empirical support. Id. at 205 & n. 12, 208-09. It therefore held that the CHD-related advertising claims were false and deceptive. Id. at 213.

At trial, the district court also heard conflicting evidence on the market effect of Ralston’s CHD-related advertising. This evidence included surveys of veterinarians and consumers, Ralston officials’ own assessments of the advertising campaign, and several expert witnesses’ interpretations of data on puppy food sales and market shares. The court concluded that the challenged ads materially increased Ralston’s sales and profits, and that, given the vigorous interbrand competition during the period at issue, the increases came at the expense of ALPO and other competitors of Ralston. Id. at 209; see also id. at 214 (supplementing “the court’s earlier finding that Ralston’s claims were material in fact” with a presumption, drawn from PPX Enterprises v. Audiofidelity Enterprises, 818 F.2d 266, 272 (2d Cir.1987), that actually false claims are material).

B. ALPO’s Veterinarian Preference Advertising

Over roughly the same period covered by Ralston’s CHD-related advertising, ALPO claimed in several media that veterinarians preferred “the formula” of ALPO Puppy Food over that of “the leading puppy food,” Puppy Chow. The district court found no basis for this assertion, or for the additional claim, appearing on ALPO Puppy Food bags and in other media, that the veterinarian preference was “2 to 1.” See id. at 209-11. The court therefore held that ALPO’s advertising was false, material, and aimed at Ralston. ALPO has not appealed. 5

*963 C. District Court’s Findings and Conclusions

Relying on its findings of fact concerning Ralston’s and ALPO’s advertising and the effects of that advertising, the court held that both companies had made material, false claims about goods in interstate commerce, in violation of section 43(a). The court found that the likelihood of deception and injury from each company’s advertising justified a permanent injunction against each company under section 34(a) of the Lanham Act, 15 U.S.C. § 1116(a) (1982 & Supp. V 1987) (amended 1988). See ALPO, 720 F.Supp. at 214-15, 215-16. On the issue of monetary relief, the court found “that ALPO Puppy Food lost sales to Puppy Chow on account of the CHD claims” and that “ALPO spent a substantial amount of money in advertising expenditures in order to counter the CHD campaign.” Id. at 212. In contrast, the court found Ralston undeserving of monetary relief because it saw Ralston as the greater wrongdoer, id. at 216, because it believed that Ralston’s counterclaim was “instituted only as an afterthought,” id., and because it concluded that an injunction against ALPO would suffice to vindicate the public interest, id. at 212. On the issue of attorneys’ fees, the opinion includes no explicit findings supporting the partially offsetting fee awards. At a postjudgment hearing, however, the court explained that it had awarded fees in an effort “to encourage private attorneys general.” Transcript of Motions Hearing at 7 (Sept. 18, 1989).

II. Discussion

A. Holding That Ralston’s CHD-related Advertising Violated Section 43(a)

Since ALPO has not appealed, we review only half of the district court’s conclusion that Ralston and ALPO both violated section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a) (1982). 6 As the district *964 court correctly stated, to prevail in a false advertising suit under section 43(a), a plaintiff must prove that the defendant’s ads were false or misleading, 7 actually or likely deceptive, material in their effects on buying decisions, connected with interstate commerce, and actually or likely injurious to the plaintiff. ALPO, 720 F.Supp. at 213 (citing, inter alia, Skil Corp. v. Rockwell Int’l Corp., 375 F.Supp. 777, 783 (N.D.Ill.1974)); accord Harper House, Inc. v. Thomas Nelson, Inc., 889 F.2d 197, 208 (9th Cir.1989). Ralston challenges the court’s holding that ALPO successfully proved all of these elements with respect to the CHD ads. Its key claim is that the court’s legal conclusions rest on erroneous findings of fact.

In reviewing the district court’s findings on the elements of ALPO’s section 43(a) claim, we have no authority to weigh the evidence anew. See Fed.R.Civ.P. 52(a) (“Findings of fact, whether based on oral or documentary evidence, shall not be set *965 aside unless clearly erroneous_ ). In Anderson v. City of Bessemer City, 470 U.S. 564, 573-76, 105 S.Ct. 1504, 1511-13, 84 L.Ed.2d 518 (1985), the Supreme Court described in expansive terms the deference that the “clearly erroneous” standard affords trial courts. Going beyond the familiar “ ‘definite and firm conviction that a mistake has been committed’ ” standard, id. at 573, 105 S.Ct. at 1511 (quoting United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948)), the Court said this:

Documents or objective evidence may contradict the witness’ story; or the story itself may be so internally inconsistent or implausible on its face that a reasonable factfinder would not credit it.... But when a trial judge’s finding is based on his decision to credit the testimony of one of two or more witnesses, each of whom has told a coherent and facially plausible story that is not contradicted by extrinsic evidence, that finding, if not internally inconsistent, can virtually never be clear error.

Id., 470 U.S. at 575, 105 S.Ct. at 1512.

The above passage from Bessemer City squarely applies. The district court heard weeks of conflicting expert testimony on the basis for and effectiveness of Ralston’s CHD-related advertising. In finding that advertising false, deceptive, material, and injurious, the court cited specific experts’ testimony, sometimes crediting that testimony over other evidence. See, e.g., ALPO, 720 F.Supp. at 204 (crediting Dr. E.A. Corley’s criticism of Ralston’s research monograph on anion gap theory); id. at 208 (“The results of Trial V should not have been excluded from Ralston’s analysis. The testimony of Dr. Corley and Dr. Robson are credited on this point. The justification provided by Drs. Kealy and Helms as to why Trial V was excluded is unconvincing and contrary to the weight of the credible evidence.”); id. at 209 (finding Dr. Bruce Owen’s regression studies sufficient to show that CHD-related advertising had a material effect). We have reviewed the record, and we cannot say that this evidence and other evidence supporting the court’s view of Ralston’s CHD-related claims were incoherent, facially implausible, or contradicted by extrinsic proof. See Bessemer City, 470 U.S. at 575, 105 S.Ct. at 1512. Accordingly, we affirm the district court’s determination that ALPO satisfactorily carried its burdens of proof and persuasion on each element of its false advertising case against Ralston.

In only one respect do we disturb the district court’s analysis of ALPO’s section 43(a) claim. As we discuss in sections (B) and (D) below, the court granted ALPO certain remedies that would be proper only in a case involving actions that evince willfulness or bad faith, such as passing off a product as another seller’s product. See, e.g., W.E. Bassett Co. v. Revlon, Inc., 435 F.2d 656, 662 (2d Cir.1970) (affirming award of profits in case involving willful passing off). The district court did not explicitly state that Ralston acted willfully or in bad faith, but its choice of remedies strongly suggests such a finding. See, e.g., ALPO, 720 F.Supp. at 215, 216 (awarding attorneys’ fees to ALPO); see also Conservative Digest, 821 F.2d at 808 (award of attorneys’ fees to plaintiff requires a finding of willfulness or bad faith). Moreover, several of the court’s findings on Ralston’s conduct, taken together, reflect the court’s conclusion that Ralston ran advertisements that it knew lacked support. See, e.g., ALPO, 720 F.Supp. at 213 (Ralston’s claims have “perpetrated a cruel hoax on dog owners”); id. at 216 (“Ralston persists in its position that its thoroughly inadequate and distorted research permits it to continue to claim its dog food can ameliorate CHD.”). In sum, the court found, without stating explicitly, that Ralston had acted willfully or in bad faith. For the reasons that follow, we set that finding aside as clearly erroneous.

None of the district court’s observations that amount to a finding of willfulness or bad faith relate to Ralston’s intentions at the time that it violated section 43(a). For example, the opinion stresses that Ralston has not recanted its CHD-related claims or shown remorse for having made them. Id. at 214-15, 216. Although *966 these considerations can affect the propriety of a permanent injunction by showing whether a defendant is likely to cause future harm, see, e.g., SEC v. First City Fin. Corp., 890 F.2d 1215, 1233 (D.C.Cir.1989) (Ruth B. Ginsburg, J., joined by Edwards, J., concurring); ALPO, 720 F.Supp. at 214-15, they cannot show that an unrepentant party has willfully done its commercial misdeeds. Particularly in a case involving disputes over scientific support for advertising claims, a party’s protestations of innocence can reflect an honest difference of scientific opinion, rather than a specific intent to mislead consumers and attack business rivals. Indeed, insisting that a losing party show contrition, especially when an appeal lies ahead, overlooks the nonpenal nature of section 35(a) remedies. See infra p. 970; cf. United States v. Williams, 891 F.2d 921, 923 (D.C.Cir.1989) (noting criminal defendant’s “acceptance of responsibility,” which has mitigating effect under federal sentencing guidelines).

The district court also condemns Ralston for its efforts to control regulators’ and potential litigants’ access to its CHD research — efforts that reportedly included destroying some documents. ALPO, 720 F.Supp. at 208, 216. If Ralston manipulated the evidence, the court had at its disposal tools for resolving the problem directly. See, e.g., Fed.R.Civ.P. 37 (establishing sanctions for discovery abuse). Litigation misconduct alone, however, does not demonstrate indifference to competitors’ or consumers’ rights, the state of mind that some Lanham Act remedies require. See Foxtrap, 671 F.2d at 641-42; see also Gold Seal Co. v. Weeks, 129 F.Supp. 928, 940 (D.D.C.1955) (identifying competitors’ and consumers’ rights as the rights that section 43(a) primarily protects), aff'd sub nom. S.C. Johnson & Son v. Gold Seal Co., 230 F.2d 832 (D.C.Cir.) (per curiam), cert. denied, 352 U.S. 829, 77 S.Ct. 41, 1 L.Ed.2d 50 (1956).

Lastly, the court’s finding that Ralston carried out its CHD-related research, product change, and advertising with its competitors in mind, see ALPO, 720 F.Supp. at 201, added to its finding that the research did not bear out the advertising claims, id. at 209, 213, does not necessarily support a conclusion that Ralston acted willfully or in bad faith. The decisions that require willfulness or bad faith for certain Lanham Act remedies are trademark infringement cases under section 43(a). E.g., Conservative Digest, 821 F.2d at 808 (limiting awards of attorneys’ fees); Foxtrap, 671 F.2d at 641-42 (limiting awards of profits). In these cases, willful or bad-faith infringement usually means passing off a product or service as another seller’s better-established one, or some other deliberate theft of a mark holder’s good will — in sum, conduct aimed at a victim targeted by the defendant. See, e.g., id. at 642 n. 10 (stating, as evidence of willfulness, that infringing night club’s manager commented to newspaper that “he hoped the [infringed] club’s glitter would ‘rub off’ on his club”); W.E. Bassett, 435 F.2d at 662 (noting, in affirming finding of willful infringement, that infringer first tried to buy out its victim, then infringed victim’s mark); see also J.T. McCarthy, Trademarks and Unfair Competition § 30:25(A), at 498 (2d ed. 1984) (before awarding an infringer’s profits, courts demand proof of “ ‘intent,’ or a knowing act denoting an intent, to infringe or reap the harvest of another’s mark and advertising”). This “targeting” consideration is at least as appropriate in a false advertising case. See, e.g., Harper House, 889 F.2d at 209 n. 8 (indirectly equating direct comparative advertising with passing off). In the broader false-advertising context, as in the trademark infringement context, “willfulness” and “bad faith” require a connection between a defendant’s awareness of its competitors and its actions at those competitors’ expense. Cf. General Mills, Inc. v. Kellogg Co., 824 F.2d 622, 627 (8th Cir.1987) (trademark case) (“Knowledge of another’s product and an intent to compete with that product is not, however, equivalent to an intent ... to mislead and to cause customer confusion.”).

To summarize, we affirm the district court’s holding that Ralston violated section 43(a), but reverse its finding that the violation was willful or in bad faith.

*967 B. Monetary Award in Favor of ALPO

Besides challenging the district court’s conclusion that its CHD-related advertising violated section 43(a), Ralston attacks the monetary remedy for that violation: a $10.4 million judgment in favor of ALPO under section 35(a) of the Lanham Act, 15 U.S.C. § 1117(a) (1982 & Supp. V 1987). 8 Ralston concentrates its attack on the court’s decision to use Ralston’s advertising costs as a measure of monetary relief, a method derived from U-Haul Int’l, Inc. v. Jartran, Inc., 793 F.2d 1034, 1042 (9th Cir.1986). Ralston argues that the court made an error of law in adopting U-Haul, and that the court incorporated clearly erroneous findings of fact into its U-Haul analysis. Reviewing the court’s decision on monetary relief for abuse of discretion, see Conservative Digest, 821 F.2d at 807, we agree that the award against Ralston must be vacated.

The district court’s opinion states that Ralston’s false advertising caused ALPO financial harm, and describes the $10.4 million award to ALPO as damages. See, e.g., ALPO, 720 F.Supp. at 212 (“The court finds that ALPO was damaged by Ral-ston’s misconduct.... A particularly effective way to measure the damages sustained would be to look to the sums Ral-ston expended in its CHD advertising campaign.”). But cf. id. at 215 (“The measure of damages and profits may be assessed on a number of different bases where necessary to effectuate the purposes of the Act_”). The two-part analysis supporting the amount of the monetary relief, however, shows that the court actually awarded Ralston’s profits to ALPO. In deciding the amount of relief, the district court first adopted the reasoning in U-Haul, 793 F.2d at 1042. See ALPO, 720 F.Supp. at 215. In that case, the Ninth Circuit affirmed an unprecedented $40 million award to a competitor injured by a section 43(a) violation. See U-Haul, 793 F.2d at 1037, 1041-42 (explaining district court’s alternative justifications for monetary award and deciding to affirm award based on one of those justifications). Of the award as affirmed, $12 million constituted the defendant’s profits, calculated on the assumption that “the financial benefit [to the defendant] was at least equal to [its] advertising expenditures” and adjusted upward 100% under section 35(a). Id. at 1042; see also id. (repeatedly calling this part of award an award of profits). The district court here, after using the U-Haul approach to calculate Ralston’s profits at $10.4 million, confirmed that figure by comparing it with “the 11 million dollar adjusted net profits Ralston earned from the sales of its Puppy Chow products during the period of its CHD advertising program.” ALPO, 720 F.Supp. at 215; see id. at 212, 215 (“adjusted net profits” figure *968 equals Ralston’s nationwide pre-tax profits, multiplied by ALPO’s percentage share of non-Ralston puppy food market).

Leaving aside whether the U-Haul standard 9 or the district court’s alternative calculation 10 accurately measures the profits that Ralston derived from its false advertising, we hold that this case does not justify an award of profits. Section 35(a) authorizes courts to award to an aggrieved plaintiff both plaintiff’s damages and defendant’s profits, but, as this court noted in Foxtrap, 671 F.2d at 641, courts’ discretion to award these remedies has limits. Just as “any award based on plaintiff’s damages requires some showing of actual loss,” id. at 642; see also infra p. 969 (discussing actual damages under section 35(a)), an award based on a defendant’s profits requires proof that the defendant acted willfully or in bad faith, see Foxtrap, 671 F.2d at 641; Frisch’s Restaurants, Inc. v. Elby’s Big Boy, 849 F.2d 1012, 1015 (6th Cir.1988). Proof of this sort is lacking. Ralston’s decision to run CHD-related advertising that lacked solid empirical support does not, without more, reflect willfulness or bad faith. See supra pp. 965-66; cf. U-Haul Int’l, Inc. v. Jartran, Inc., 601 F.Supp. 1140, 1147-48 (D.Ariz.1984) (describing U-Haul defendant’s targeted comparative advertising, which misrepresented plaintiff’s and defendant’s prices), aff'd in part and rev’d in part, 793 F.2d 1034 (9th Cir.1986).

In Conservative Digest we “left open the possibility that a court could properly award damages to a plaintiff when the defendant has been unjustly enriched.” 821 F.2d at 807-08 (citing Foxtrap, 671 F.2d at 641 & n. 9). The unjust-enrichment theory, which emerged in trademark cases in which the infringer and the infringed were not competitors, holds that courts should divest an infringer of his profits, regardless of whether the infringer’s actions have harmed the owner of the infringed trademark. Awards of profits are justified under the theory because they deter infringement in general and thereby vindicate consumers’ interests. See, e.g., Monsanto Chemical Co. v. Perfect Fit Prods. Mfg. Co., 349 F.2d 389, 392, 395-97 (2d Cir.1965), cert. denied, 383 U.S. 942, 86 S.Ct. 1195, 16 L.Ed.2d 206 (1966); see also 2 J.T. McCarthy, supra p. 14, § 30:25(B) (citing cases). As we state below, however, we doubt the wisdom of an approach to damages that permits courts to award profits for their sheer deterrent effect.

Relying on W.E. Bassett, 435 F.2d at 664, and Monsanto Chemical, 349 F.2d at 396, ALPO proposes such an approach. See Brief of Appellee at 40-43. Indeed, at oral argument, counsel for ALPO claimed that because Foxtrap, 671 F.2d at 641, cites W.E. Bassett, this circuit has “adopted the Bassett theory for [profits awards based solely on] deterrence in an egregious case.” Transcript of Oral Argument at 25 (Apr. 13, 1990). Foxtrap, how *969 ever, cites W.E. Bassett to suggest that courts can award profits only when they find “a relatively egregious display of bad faith.” Foxtrap, 671 F.2d at 641. Indeed, this court in Foxtrap advised a district court to make an award that would “deter the defendant, yet not be a windfall to plaintiff nor amount to punitive damages.” Id. at 642 n. 11 (emphasis added). Based on Foxtrap, as well as our concern that deterrence is too weak and too easily invoked a justification for the severe and often cumbersome remedy of a profits award, see Koelemay, Monetary Relief for Trademark Infringement Under the Lanham Act, 72 Trademark Rep. 458, 493-94, 536-37 (1982), we hold that deterrence alone cannot justify such an award.

Since this case lacks the elements required to support the court’s award of Ral-ston’s profits, we vacate the $10.4 million judgment in favor of ALPO. We do not mean, however, to deny ALPO all monetary relief for Ralston’s false advertising. Because the district court has so far focused on awarding Ralston’s profits, it has not yet decided what actual damages ALPO has proved. On remand, the court should award ALPO its actual damages, bearing in mind the requirement that any amount awarded have support in the record, see Foxtrap, 671 F.2d at 642; Gold Seal, 129 F.Supp. at 940, as well as the following points about the governing law.

In a false-advertising case such as this one, actual damages under section 35(a) can include:

—profits lost by the plaintiff on sales actually diverted to the false advertiser, see, e.g., Foxtrap, 671 F.2d at 642 (trademark case);

—profits lost by the plaintiff on sales made at prices reduced as a demonstrated result of the false advertising, see, e.g., Burndy Corp. v. Teledyne Indus.,

Additional Information