PARKS LLC, Appellant v. TYSON FOODS, INC; HILLSHIRE BRANDS COMPANY
OPINION OF THE COURT
This case concerns a trademark that once enjoyed widespread recognition but has since grown considerably weaker. Since the 1950s, Parks Sausage Company has manufactured or licensed sausage under the brand name “PARKS.”1 At one point, PARKS was placed on the Principal Register of trademarks at the United States Patent and Trademark Office (“USPTO”), but, sometime in the early 2000s, Parks failed to renew the registration. In 2014, Tyson Foods, Inc. and Hillshire Brands Company (collectively, “Tyson”),2 the owners of the frankfurter brand BALL PARK, launched a premium frankfurter product called PARK'S FINEST. Parks sued, arguing that Tyson was engaged in false advertising and was infringing Parks's trademark.
The District Court determined that Parks's claim for false advertising was really a repetition of its trademark claim, and that the PARKS mark was too weak to merit protection against Tyson's use of the PARK'S FINEST name. We agree with the District Court and will affirm in all respects.
A. Parks and PARKS
Parks was founded in the 1950s by Henry G. Parks, Jr., a pioneering African-American businessman. The company had the distinction at one point of being the first African-American-owned company to be publicly traded on the New York Stock Exchange. Parks engaged in radio and television advertising directed to consumers and developed a well-known slogan, “More Parks Sausages, Mom, Please [.]” (Opening Br. at 29.) Though the PARKS brand had likely developed prominence sufficient for common law trademark protection earlier than 1970, the name was not registered at the USPTO until that year.
Following the death of Mr. Parks in 1989, the company he built fell on hard times. It eventually went bankrupt and was purchased by its current owners. Parks stopped making and selling PARKS products and instead entered into a licensing agreement in 2000 with Dietz & Watson, a Philadelphia-based producer of delicatessen meats, to make and sell PARKS-branded products. Around that time, the USPTO registration of the mark lapsed. In 2002, Parks also granted a license to Super Bakery, Inc., a supplier of baked goods that is related to Parks by common ownership, to sell PARKS-branded products in military commissaries. At least since the licensing agreement with Dietz & Watson 17 years ago, PARKS-branded products have been advertised primarily through grocery store handbills and circulars rather than through television and radio advertising.
From 2008 through 2014, Dietz & Watson sold over $38 million worth of PARKS-branded products. In 2014, PARKS sales through Dietz & Watson increased 40% from the previous year. In addition, from 2003 through August 2013, Super Bakery sold some $31 million in PARKS products.
B. Tyson, BALL PARK, and PARK'S FINEST
BALL PARK brand frankfurters are well known, accounting for 23% of the revenue of all franks sold in the United States.3 Parks, LLC v. Tyson Foods, Inc., Civ. No. 5:15-00946, 2015 WL 4545408, at *4 (E.D. Pa. July 28, 2015). Tyson owns the BALL PARK mark and claims that the brand is recognized by 90% of American adults over the age of eighteen. Id. In 2014, Tyson introduced a line of “super-premium” frankfurters that it decided to call “[PARK'S FINEST].” Parks, LLC v. Tyson Foods, Inc., 186 F. Supp. 3d. 405, 412 (E.D. Pa. 2016). It says it chose that name after conducting extensive consumer research. According to Tyson, surveys showed that the name “conveys premium quality in a clever, memorable way that should be ownable for the [BALL PARK] brand.” (App. at 375.)
Packaging for the frankfurters includes the BALL PARK logo superimposed over the words PARK'S FINEST.
Tabular or graphical material not displayable at this time.
(App. at 11.) Tyson says that it designed the packaging to “strongly convey [ ]” the BALL PARK mark so that consumers would make the connection between BALL PARK and the PARK'S FINEST product. (App. at 390.) In advertisements, the product would be referred to as “[PARK'S FINEST] from [BALL PARK],” Parks, 2015 WL 4545408 at *2, or sometimes “BALL PARK'S FINEST.” (Opening Br. at 17.) Before the product was launched, Tyson's attorneys undertook a trademark search, discovered the lapsed PARKS mark, and confirmed the cancellation of the mark with the USPTO.
C. Procedural Background
Parks filed suit against Tyson in 2015, asserting false advertising, false association, and trademark dilution claims under the Lanham Act, as well as Pennsylvania common law and statutory claims.4 It requested a nationwide injunction and an accounting of the sales of PARK'S FINEST franks.
The District Court denied Parks's motion for a preliminary injunction because it concluded that Parks was unlikely to succeed on the merits of its false advertising claim. Parks, 2015 WL 4545408 at *16. Once discovery ended, Tyson moved for summary judgment on all of Parks's claims. Parks conceded that its trademark dilution claim and its state law claims should be dismissed, and they were. The Court then granted summary judgment on the remaining Lanham Act claims of false association and false advertising, concluding that “no reasonable factfinder could find in Parks's favor[.]” Parks, 186 F. Supp. 3d. at 413 (footnote omitted). Parks timely appealed.
II. DISCUSSION 5
Section 43(a) of the Lanham Act, codified at 15 U.S.C. § 1125(a), creates “two distinct bases of liability: false association ․ and false advertising.” Lexmark Int'l, Inc. v. Static Control Components, Inc., __ U.S. __, 134 S. Ct. 1377, 1384 (2014); see also J. Thomas McCarthy, 5 McCarthy on Trademarks and Unfair Competition § 27:9 (4th ed. 2017) [hereinafter “McCarthy on Trademarks”] (describing “two major and distinct types” of claims under § 1125(a)). Section 1125(a)(1)(A) prohibits “false or misleading” claims that are “likely to cause confusion, or to cause mistake, or to deceive as to ․ the origin, sponsorship, or approval of his or her goods, services, or commercial activities by another person[.]” That provision is “the foremost federal vehicle for the assertion of ․ infringement of ․ unregistered marks, names and trade dress[.]”6 5 McCarthy on Trademarks § 27:9. Claims made under it are often called “false designation of origin” or “false association” claims. Id. at n. 1.05. We will use the latter term here. To establish a false association claim, the owner of an unregistered mark “has the burden ․ of proving the existence of a protectable mark.” E.T. Browne Drug Co. v. Cococare Prod., Inc., 538 F.3d 185, 191 (3d Cir. 2008). When, as in this case, the mark is a surname, a necessary step in showing that it is eligible for protection as a trademark is demonstrating that it has acquired secondary meaning. Id. at 191-92. “Secondary meaning” is a term of art in trademark law that refers to “a mental association in buyers' minds between the alleged mark and a single source of the product.” 2 McCarthy on Trademarks § 15:5.
Another portion of the statute, subsection (a)(1)(B), forbids “commercial advertising or promotion” that “misrepresents the nature, characteristics, qualities, or geographic origin of his or her or another person's goods, services, or commercial activities[.]” 15 U.S.C. § 1125(a)(1)(B). Claims under that provision are called “false advertising” claims. 5 McCarthy on Trademarks § 27:9. False advertising claims do not require proof of secondary meaning, so litigants may be tempted to frame a false association claim as a false advertising claim, to ease their evidentiary burden. Cf. Pernod Ricard USA, LLC v. Bacardi U.S.A., Inc., 653 F.3d 241, 247 & n.8 (3d Cir. 2011) (noting that, in that case, the “false advertising dispute [was] a proxy for the real fight the parties want[ed] to have, which [was] over the right to the ․use of ․ a trademark” and observing that “[t]his [was] not the first time the false advertising provision of the Lanham Act has been asked to stand in for a trademark action”). That is what seems to have happened here, and we take this opportunity to clarify the distinction between claims brought under § 1125(a)(1)(A) and § 1125(a)(1)(B). As the District Court recognized, Parks's false advertising claim fails because it is essentially a false association claim in disguise. The false association claim is also infirm, for the reasons described by the District Court.
A. False Advertising
As noted above, the statement at issue in a false advertising claim must “misrepresent[ ] the nature, characteristics, qualities, or geographic origin” of a product.7 15 U.S.C. § 1125(a)(1)(B); see Kehoe Component Sales Inc. v. Best Lighting Prods., Inc., 796 F.3d 576, 590 (6th Cir. 2015) (“Absent a false statement about geographic origin, a misrepresentation is actionable under § 1125(a)(1)(B) only if it misrepresents the characteristics of the good itself– such as its properties or capabilities. The statute does not encompass misrepresentations about the source of the ideas embodied in the object (such as a false designation of authorship) [.]” (internal quotation marks and citation omitted)). Parks's false advertising claim fails because it depends upon the purported false association between Tyson's PARK'S FINEST brand and the PARKS mark. So the false advertising claim rises and falls with the false association claim, which we will subsequently address. For now, we note simply that the primary argument Parks advances is that the name PARK'S FINEST falsely implies that Tyson's product is one of Parks's products. As Parks puts it, Tyson marketed PARK'S FINEST as “Parks' sausages.” (Opening Br. at 21.) In other words, PARK'S FINEST is only misleading in the way that Parks suggests if a consumer makes the connection between PARK'S FINEST and PARKS and has in mind a pre-existing association between PARKS and high quality products. This is a false association claim and nothing more.
Parks also argues that PARK'S FINEST is misleading with respect to the “nature, characteristics, [or] qualities,” 15 U.S.C. § 1125(a)(1)(B), of the product. Parks alleges that the name PARK'S FINEST will imply to consumers that it is a sausage when in reality it is a frankfurter – an item consumers may see as inferior. Again, that contention largely duplicates the one that we have already found wanting. Unless a consumer knows that PARKS is a mark for sausages, the name PARK'S FINEST does not carry any such implications. At bottom, then, this too is a false association claim.8
To the extent that Parks is advancing the related argument that the name “PARK'S FINEST” is misleading because it blurs the distinction between frankfurters and sausages and is therefore confusing to consumers, that argument falters on two grounds.9 First of all, because the packaging for PARK'S FINEST displays “a factually accurate, unambiguous statement” that the product is a frankfurter, “[n]o reasonable consumer could be misled by those statements, and the rest of the label does not put those statements in doubt.” See Pernod Ricard, 653 F.3d at 252 (reaching the same conclusion with regards to a rum called Havana Club that was actually made in Puerto Rico but informed consumers on the packaging that it was made in Puerto Rico).10 Second, Parks has undercut its own argument by repeatedly saying both at the District Court and before us on appeal that hot dogs and sausages are actually not distinctive.11 If a frankfurter is a kind of a sausage, as Parks suggests, then there is nothing false or misleading if the advertising for PARK'S FINEST suggests that to consumers.
Parks's final argument is that the name PARK'S FINEST is misleading with regard to “origin.” (Opening Br. at 15-16.) To a large extent that argument falters because it repeats the same mistakes we have already discussed. But it also fails for another reason. Section 1125(a)(1)(B) focuses specifically on statements that are false with regard to “geographic origin” and not other types of “false designation[s] of origin,” § 1125(a)(1).12 So, § 1125(a)(1)(B) can be the proper vehicle for bringing a challenge to, for example, a Swiss army knife not made in Switzerland, Black Hills gold jewelry that is not made in the Black Hills, or Scotch whiskey not manufactured in Scotland. Cf. Forschner Grp., Inc. v. Arrow Trading Co. Inc., 30 F.3d 348, 355 (2d Cir. 1994) (concluding that the term “swiss” in swiss army knife was not a geographic term of origin but adjudicating the claim under § 1125(a)(1)(B)); Black Hills Jewelry Mfg. Co. v. Gold Rush, Inc., 633 F.2d 746, 750 (8th Cir. 1980) (noting that the Lanham Act's prohibition on false advertising “continue[s] [a] tradition of providing protection against outsiders who use the same geographical designation”); Scotch Whiskey Ass'n v. Barton Distilling Co., 489 F.2d 809, 811 (7th Cir. 1973) (applying an earlier version of the Lanham Act to resolve a geographic origin claim). But that subsection is not the right vehicle for addressing claims of false designation of origin that are not concerned specifically with geographic origin. See Pernod Ricard, 653 F.3d at 250 n.11 (noting that the most natural meaning of “geographic origin” would extend only to “the place of a product's manufacture, not a broad inquiry into the product's background,” but affirming dismissal on alternative grounds). We have suggested as much before in dicta, but never in a holding. Id. The question is now squarely before us and we conclude that the term “geographic origin” refers solely to the place of origin and not to the creator, manufacturer, or any broader conception of the term “origin.”13
That conclusion is consistent with precedent from other circuit courts. See Forschner Grp., 30 F.3d at 355 (“The question is whether [the mark] can be construed to mean that the product is made in a certain locale.”); cf. Black Hills Jewelry Mfg. Co., 633 F.2d at 750 (noting that the Lanham Act follows the common law “tradition of providing protection against outsiders who use ․ [a] geographical designation”). It is also consistent with Supreme Court precedent interpreting the term “origin of goods” in the context of a false association claim under the Lanham Act. In Dastar Corp. v. Twentieth Century Fox, 539 U.S. 23, 37 (2003), the Court emphasized that “origin” could refer to either “geographic origin” or “to origin of source or manufacture” but rejected the argument that the term origin could “be stretched” to include broader concepts of origin such as “the creator of the underlying work,” id. at 29-33, or “the author of any idea, concept, or communication embodied in those goods,” id. at 37. So the term “origin” has already been cabined.
Moreover, as we suggested in Pernod Ricard, the term is further narrowed in § 1125(a)(1)(B) by the addition of the modifier “geographic.” Pernod Ricard, 653 F.3d at 250 n.11 (questioning the district court's effort to “use the modifier ‘geographic’ to expand the meaning of ‘origin’ into the realm of history, heritage, and culture”). Dictionary definitions of the term “geographic” are consistent with a focus on the place of origin or manufacture. See Geographic, Webster's Third New International Dictionary 948 (1986) (defining “geographic” as “belonging to or characteristic of a particular region”). Likewise, under the common law of trademarks, a “geographically descriptive trademark” is one that uses “a geographic name to indicate where the goods are grown or manufactured.” Geographically Descriptive Trademark, BLACK'S LAW DICTIONARY 1631 (9th Ed. 2009) (emphasis added); see also 2 McCarthy on Trademarks § 14:3 (providing examples of “geographically descriptive term[s]”). When Congress used the term “geographic origin,” it was not writing on a blank slate, but instead appears to have intended to link the protections of the false advertising provision to a well-defined and readily understood concept: the place where the goods come from.14 Cf. Forschner Grp., 30 F.3d at 354 (concluding that “[u]nder the false advertising provision of the Lanham Act, a phrase is eligible for protection as a representation of geographic origin only if the phrase is geographically descriptive”).
The name PARK'S FINEST says nothing about the product's “geographic origin.” In the end, Parks has not made a valid claim for false advertising because none of its grievances concern the “nature, characteristics, qualities, or geographic origin” of PARK'S FINEST.15
B. False Association
The elements of a false association trademark claim under the Lanham Act track the elements of a common law trademark infringement claim: a plaintiff must prove that “(1) the marks are valid and legally protectable; (2) the marks are owned by the plaintiff; and (3) the defendant's use of the marks to identify goods or services is likely to create confusion concerning the origin of the goods or services.” Ford Motor Co. v. Summit Motor Prod., Inc., 930 F.2d 277, 291 (3d Cir. 1991) (internal quotation marks omitted). Even when those elements are satisfied, relief is limited in scope to where “market penetration is significant enough to pose the real likelihood of confusion among the consumers in that area.” Charles Jacquin Et Cie, Inc. v. Destileria Serralles, Inc., 921 F.2d 467, 472 (3d Cir. 1990) (internal quotation marks omitted).
A valid and legally protectable mark must be “distinctive,” which may be shown in two ways. Some marks are, by their very nature, considered distinctive. 2 McCarthy on Trademarks § 11:2. Such inherently distinctive marks include ones that are arbitrary or fanciful, such as APPLE for computers or SHELL for gasoline, id. at § 11:11, as well as ones that are suggestive of a product's function but not descriptive such as PENGUIN for freezers or SAMSON for weight training machines, id. at § 11:67. On the other hand, marks that are merely descriptive of the product are not inherently distinctive and secondary meaning must be proven before such a name will be protectable.16 Id. at 11:2; see Commerce Nat'l. Ins. Servs., Inc. v. Commerce Ins. Agency, Inc., 214 F.3d 432, 438 (3d Cir. 2000) (“If the mark has not been federally registered ․ then validity depends on proof of secondary meaning, unless the unregistered or contestable mark is inherently distinctive.” (internal quotation marks omitted)). The District Court concluded that PARKS was not inherently distinctive and had not achieved secondary meaning. We agree.
1. Lack of Secondary Meaning
Trademarks based on the surname of a founder are not inherently distinctive. Doeblers' Pennsylvania Hybrids, Inc. v. Doebler, 442 F.3d 812, 827 n.17 (3d Cir. 2006), as amended (May 5, 2006). In a painful stretch, Parks argues inherent distinctiveness should be a jury question because, even though its mark is, in fact, the surname of the company's founder, the word “parks” is also the plural of “park,” as in recreational land, and therefore could be seen as an “arbitrary” mark. But it is undisputed that Parks was named after its founder, someone who Parks describes with justifiable pride as “an important figure in the history of American Business,” (App. at 60) and Parks's reputation is closely linked to its founder.17 Based on that record, no reasonable juror could conclude that the name PARKS was anything other than a reference to the founder. It is obvious that the mark is not inherently distinctive.
Parks was therefore required to demonstrate that the mark had secondary meaning at the time that Tyson began to use the name PARK'S FINEST. As noted earlier, “[s]econdary meaning exists when the mark ‘is interpreted by the consuming public to be not only an identification of the product or services, but also a representation of the origin of those products or services.’ ” Commerce Nat'l., 214 F.3d at 438 (quoting Scott Paper Co. v. Scott's Liquid Gold, Inc., 589 F.2d 1225, 1228 (3d Cir. 1978), superseded on other grounds by statute, Fed. R. Civ. P. 52(a), as recognized in Shire US Inc. v. Barr Labs., Inc., 329 F.3d 348, 352 n.10 (3d Cir. 2003)). In assessing secondary meaning, we have relied on the following factors, to the extent relevant:
(1) the extent of sales and advertising leading to buyer association; (2) length of use; (3) exclusivity of use; (4) the fact of copying; (5) customer surveys; (6) customer testimony; (7) the use of the mark in trade journals; (8) the size of the company; (9) the number of sales; (10) the number of customers; and, (11) actual confusion.
Id. (citing Ford Motor Co., 930 F.2d at 292).
As did the District Court, we consider each of the factors relevant to this case and conclude that no reasonable juror could decide that PARKS enjoyed secondary meaning at the time of the alleged infringement.18
i. Extent of Advertising
Secondary meaning is generally “established through extensive advertising which creates in the minds of consumers an association between the mark and the provider of the [products or] services advertised under the mark.” Id. Use of a mark “for a long period of time in a prevalent advertising campaign” can “create a reasonable inference” of secondary meaning. E.T. Browne Drug Co., 538 F.3d at 200. Since approximately 2001, PARKS has not been advertised directly to consumers aside from around $14,000 a year that Dietz & Watson has spent on circular advertisements and in-store product demonstrations. Super Bakery has primarily engaged in direct marketing to “institutions and military facilities.” Parks, 186 F. Supp. 3d at 422. Decades ago, in the 1960s and ‘70s, PARKS had employed a ubiquitous and long-running ad campaign to reach consumers, and while there may still be some faint echoes of the campaign in the minds of some people, there is no evidence of recent “extensive advertising” such as would create the necessary mental association between the mark and the product. Commerce Nat'l., 214 F.3d at 438. This factor thus cuts against a finding of secondary meaning.
ii. Length and Exclusivity of Use
The “length of use” factor favors Parks, as its mark has been in continuous use for more than 50 years. Additionally, Parks has used or licensed the mark exclusively throughout the entirety of the company's existence. And, of course, the fact that PARKS was once on the Principal Register at the USPTO indicates that PARKS had acquired secondary meaning once upon a time. Cf. 2 McCarthy on Trademarks § 15:32 (describing how registration creates a presumption of secondary meaning). But consumer perceptions have a half-life, and “once a mark, always a mark” has never been a principle of trademark law. Merely proving length and exclusivity of use does not prove widespread familiarity. As the District Court noted, “Parks has not cited to any evidence to attempt to quantify how widespread the name was known over those years before the present owners purchased the company out of bankruptcy in the late 1990s[.]” Parks, 186 F. Supp. 3d at 426. So while the length and exclusivity of use unquestionably favor Parks, those factors alone cannot carry the day.
iii. Evidence of Copying
Parks contends that Tyson copied the PARKS mark when it chose the name PARK'S FINEST. But Parks's only evidence of copying is a bare inference from the fact that Parks and Tyson compete in a similar space and that Tyson, upon a trademark search, found the lapsed PARKS mark. On the other hand, Tyson submitted extensive focus group and survey data that showed how the name PARK'S FINEST was selected without any reference to PARKS. Tyson emphasized that the name PARK'S FINEST was a finalist for the new brand name even before the trademark search was conducted. While on summary judgment the nonmoving party is entitled to reasonable inferences, it would be unreasonable on this record to conclude that Tyson copied PARKS. Even the most generous weighing of this factor for Parks leaves it neutral.
iv. Customer Surveys
Parks conducted a survey that is used primarily to test for consumer confusion, but it then sought to use that same survey to also prove secondary meaning. The attempt to make the survey do double duty was unwise.
There are two predominant formats of consumer surveys used in trademark litigation to show a likelihood of confusion. The first, the Ever-Ready survey, named after the case of Union Carbide Corp. v. Ever-Ready, Inc., 531 F.2d 366, 385-88 (7th Cir. 1976), superseded on other grounds by statute, Fed. R. Civ. P. 52(a), as recognized in Scandia Down Corp. v. Euroquilt, Inc., 772 F.2d 1423, 1429 (7th Cir. 1985), involves showing consumers only the potentially-infringing product and asking open-ended questions to determine whether they believe the product is associated with the senior mark. See Jerre B. Swann, Likelihood of Confusion Studies and the Straitened Scope of Squirt, 98 Trademark Rep. 739, 746 (2008) (describing the Ever-Ready format as the “gold standard” for likelihood of confusion surveys). Even though that survey design is most helpful for illustrating a likelihood of confusion, it can also indicate secondary meaning by showing a high degree of familiarity with the senior mark. Id. at 745 (“The [Ever-Ready] format ․ addresses ․ brand strength.” (emphasis added)).
An Ever-Ready survey is usually employed by owners of commercially strong marks. Id. at 739. Holders of weaker marks more frequently employ a Squirt survey, named after the type used in SquirtCo v. Seven-Up Co., 628 F.2d 1086, 1089 n.4 (8th Cir. 1980). Parks's expert, Mark Lang, chose to use a Squirt survey because of what he described as “the relatively weak commercial strength of the [PARKS] brand[.]” (App. at 658-59.) He also said that a Squirt survey was appropriate because PARKS and PARK'S FINEST “have a high degree of proximity in the marketplace[.]” (App. at 659.) In a Squirt survey, two products are placed side by side, often with other products that serve as controls, and participants are asked questions to determine if confusion exists as to the source of the products.19 Swann, 98 Trademark Rep. at 749-50. Courts have sometimes criticized Squirt surveys for utilizing closed-ended questions that can lead participants to the desired answer. See Scott Fetzer Co. v. House of Vacuums Inc., 381 F.3d 477, 487-88 (5th Cir. 2004) (critiquing a Squirt survey for pushing “survey participants to search for any connection, no matter how attenuated ․ instead of permitting participants to make their own associations”); Riviana Foods Inc. v. Societe Des Produits Nestle S.A., Civ. A. No. H-93-2176, 1994 WL 761242, at *4 (S.D. Tex. Dec. 20, 1994) (rejecting a Squirt survey because it “used a leading question on the likelihood of confusion issue”); see also Swann, 98 Trademark Rep. at 752-53 (compiling cases). Nevertheless, a well-designed Squirt survey may show a likelihood of confusion. What it does not do or even purport to do, however, is prove secondary meaning.
As the District Court perceived, there was a “fundamental[ ] flaw in the survey's methodology,” if the point was to show secondary meaning. Parks, 186 F. Supp. 3d at 418. Because the survey presented an image of both PARKS and PARK'S FINEST, a consumer who had never heard of PARKS could still conclude that the two products were affiliated. Specifically, participants were shown two products with the words “Parks” or “Park's” in the title and several other hot dog or sausage products with names bearing no obvious linguistic connection to Parks, and then asked whether any of the products were affiliated. Given the products shown, PARKS and PARK'S FINEST were the obvious choices. Cf. THOIP v. Walt Disney Co., 788 F. Supp. 2d 168, 183 (S.D.N.Y. 2011) (rejecting a Squirt survey in part because the allegedly infringing product “stood out like a bearded man in a lineup with four clean-shaven men” and therefore participants were pushed to reach a particular outcome (internal quotation omitted)). Participants were therefore primed to reach that conclusion, even if they were not familiar with either PARKS or PARK'S FINEST as trademarks.20 Cf. Nat'l Distillers Prods. Co., LLC v. Refreshment Brands, Inc., 198 F. Supp. 2d 474, 484 (S.D.N.Y. 2002) (observing that a similar survey was flawed because “every respondent was exposed to the [allegedly infringed] product ․ thus acquainting them with a product that they would almost certainly have been unfamiliar with otherwise, due to [the product's] very limited distribution network and weak sales”).
As a result, while Lang's survey may or may not have been useful for illustrating a likelihood of confusion,21 it was certainly not probative of whether PARKS had secondary meaning. Rather, there were at least three equally plausible conclusions that a participant could have reached when responding that PARKS and PARK'S FINEST were associated: 1) that PARK'S FINEST came from Parks (the inference Parks obviously preferred); 2) that PARKS was made by Tyson as an extension of the BALL PARK mark; or 3) that PARK'S FINEST and PARKS were both made by some unknown third party. None of those inferences is more likely than the other, so the survey tells us nothing about whether the PARKS mark had achieved sufficient consumer recognition to qualify as having secondary meaning. Cf. Itamar Simonson, The Effect of Survey Method on Likelihood of Confusion Estimates: Conceptual Analysis and Empirical Test, 83 Trademark Rep. 364, 387 (1993) (noting that the “[Squirt] [f]ormat[ ] tend [s] to lead to relatively high confusion estimates when the senior and junior marks appear as logical extensions”).
Parks nevertheless argues that “proof of one – likelihood of confusion – is proof of the other – secondary meaning.” (Opening Br. at 35) (relying on Interpace Corp. v. Lapp, Inc., 721 F.2d 460, 465 (3d Cir. 1983).) That is simply wrong. It is true that evidence proving secondary meaning and evidence proving likelihood of confusion may sometimes overlap. But not always. See 2 McCarthy on Trademarks § 15:11 n.1 (“Not every response rate that shows likely confusion establishes secondary meaning and not every survey that fails to show likely confusion establishes an absence of secondary meaning.” (quoting Vincent N. Palladino, Secondary Meaning Surveys, in TRADEMARK AND DECEPTIVE ADVERTISING SURVEYS: LAW, LAW SCIENCE, AND DESIGN 98 (2012))). Two marks can be confusingly similar even if neither has secondary meaning. See Scott Paper, 589 F.2d at 1229 (noting that “[l]ikelihood of confusion is an analytically distinct” concept from secondary meaning). Consumers may find an association even if both marks were previously unknown to them. Establishing that two marks are similar does not necessarily lead to any valid conclusion about whether either of the two has secondary meaning. See Spraying Sys. Co. v. Delavan, Inc., 762 F. Supp. 772, 779 (N.D. Ill. 1991) (“However proper the survey question may have been to prove likelihood of confusion between the marks, it was improper to prove secondary meaning.”), aff'd, 975 F.2d 387 (7th Cir. 1992); Spraying Sys. Co., v. Delavan, Inc., 975 F.2d 387, 394 (7th Cir. 1992) (noting that a leading survey design “created a bias in favor of identifying a single company” as the source of the products and therefore the survey results could not prove secondary meaning). The “customer survey” factor, Commerce Nat'l., 214 F.3d at 438, thus favors neither Parks nor Tyson.
v. Size of the Company and Number of Sales and Customers
The size of a company, its total sales, and the size of its customer base can also be probative of secondary meaning because the jury is entitled to draw the logical inference that “[t]he larger a company and the greater its sales, the greater the number of people who have been exposed to [the] symbol used as a trademark, and the greater the number of people who may associate [that] symbol with a company or source with which they should be familiarized.” 2 McCarthy on Trademarks § 15:49. But “[r]aw sales figures need to be put into context to have any meaning.” Id. When put into context, the sales figures for PARKS are not probative of secondary meaning. Since 2011, sales of PARKS sausages accounted for no more than 1.3% of the breakfast sausage market in the northeast and 0.01% of the market in the mid-south in any given year.22 Dinner sausage sales made up less than 1% of the sales in the northeast and less than 0.5% of the sales in the mid-south. Sales figures outside of those regions were even less significant. None of the sales numbers are large enough to indicate secondary meaning.
Sales in Pennsylvania and New Jersey were considerably greater than sales in other states, so “[i]t is possible that sales in ․ th[o]se states ․ could be large enough, relative to the market in those states, to be probative of secondary meaning in those markets.” Parks, 186 F. Supp. 3d at 424 n.12. Parks, however, made the expansive and ultimately unfounded claim that its PARKS mark had secondary meaning throughout the whole eastern United States,23 and it failed to break down data on market share by state. So, at most, Parks's sales data might weakly support a finding that the PARKS mark has secondary meaning in one portion of the northeast, but it does not support the broader claim that Parks made in its complaint.
vi. Actual Confusion
The evidence that Parks put forth of actual confusion was similarly unimpressive After extensive discovery, the company could only produce two declarations from employees of Dietz & Watson and Super Bakery. The first declaration came from an employee at Dietz & Watson with some responsibility for PARKS-branded products. He recounted three instances of consumer confusion: one consumer contacted Parks to complain about “[PARKS] from BALL PARK,” and two consumers called Parks to complain about the nitrate content of its sausages – presumably because PARK'S FINEST prominently advertises that it is free of nitrates. Parks, 2015 WL 4545408 at *6. The second declaration came from a manager at Super Bakery who recalled his own personal confusion when he first encountered the PARK'S FINEST product. Id.
Such declarations from friendly sources are potentially “self-serving and of little probative value.” 815 Tonawanda St. Corp. v. Fay's Drug Co., Inc., 842 F.2d 643, 648 (2d Cir. 1988). More importantly, though, PARK'S FINEST has sold “many millions of units,” Parks, 186 F. Supp. 3d at 426 (internal quotation marks omitted), and one would expect to see more than a handful of vague complaints of confusion. If anything, the paucity of proof of actual confusion suggests that the PARKS mark lacks secondary meaning. As we have said before, “harmonious coexistence in the same geographical area ․ cuts against [a] claim to secondary meaning.” Commerce Nat'l., 214 F.3d at 440.
viii. Conclusion with Regard to Secondary Meaning
At the end of the day, the fact that the PARKS mark has existed for a long time and that it enjoyed secondary meaning half a century ago cannot overcome the weight of the factors against Parks. The record shows that there is almost no direct-to-consumer advertising, that Parks had a miniscule market share, and that there was practically no record of actual confusion.24 To find secondary meaning, jurors would have to make an impermissible “leap of faith[.]” Parks, 186 F. Supp. 3d at 426 (quoting E.T. Browne Drug Co., 538 F.3d at 199). Accordingly, the District Court correctly concluded that no reasonable jury could find that the PARKS mark had secondary meaning. 25
For the foregoing reasons, we will affirm the District Court's order granting summary judgment to Tyson on Parks's claims.
1. We use “Parks” in lowercase letters to refer to the company and “PARKS” in capital letters to refer to the mark.
2. Hillshire Farms was acquired by Tyson in 2014 and remains a wholly owned subsidiary. Parks, LLC v. Tyson Foods, Inc., Civ. No. 5:15-00946, 2015 WL 4545408, at *1 (E.D. Pa. July 28, 2015).
3. Though it may distress the cognoscenti, we use the terms “frankfurters,” “franks,” and “hot dogs,” as synonyms. Not so with the term “sausage,” which we use to denote something akin to but arguably different from hot dogs.
4. Specifically, Parks's claims were: (1) false advertising under the Lanham Act, 15 U.S.C. § 1125(a)(1)(B); (2) trademark infringement in the eastern United States (all states east of the Mississippi river) under the Lanham Act, 15 U.S.C. § 1125(a)(1)(A); (3) trademark dilution under the Lanham Act, 15 U.S.C. § 1125(c); (4) trademark infringement under Pennsylvania's Unfair Trade Practices and Consumer Protection Act, 54 Pa. Const. Stat. § 1123, although Parks confusingly says in the body of Count 4 that it is “under common law”; (5) trademark dilution under Pennsylvania statutory law, 54 Pa. Const. Stat. § 1124; and (6) unfair competition under Pennsylvania common law.
5. The District Court had jurisdiction over the Lanham Act claims under 28 U.S.C. § 1331 and the state law claims under 28 U.S.C. § 1367. We have jurisdiction pursuant to 28 U.S.C. § 1291. Summary judgment is appropriate only if “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). A dispute is genuine when “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986).
6. For trademarks that are on the Principal Register at the USPTO, as the PARKS mark once was, section 32 of the Lanham Act, codified at 15 U.S.C. § 1114, also provides a cause of action for infringement.
7. Broadly stated, the elements a false advertising claim are:1) that the defendant has made false or misleading statements as to his own product [or another's]; 2) that there is actual deception or at least a tendency to deceive a substantial portion of the intended audience; 3) that the deception is material in that it is likely to influence purchasing decisions; 4) that the advertised goods traveled in interstate commerce; and 5) that there is a likelihood of injury to the plaintiff in terms of declining sales, loss of good will, etc.Warner-Lambert Co. v. Breathasure, Inc., 204 F.3d 87, 91-92 (3d Cir. 2000). The specific point at issue in this case bears on the first element.
8. The name PARK'S FINEST does touch upon the “nature, characteristics, [or] qualities,” 15 U.S.C. § 1125(a)(1)(B), of the product in at least one respect. It implies high quality. But Parks does not allege that PARK'S FINEST is of poor quality or does not in fact deserve the positive appellation of “finest.” Indeed, such a claim would likely fail since calling a product the finest is “common marketplace puffery,” Castrol Inc. v. Pennzoil Co., 987 F.2d 939, 945 (3d Cir. 1993) (internal quotation marks omitted).
9. It isn't entirely clear whether Parks is actually making such an argument, but we address it for the sake of completeness.
10. Parks points to evidence in the record suggesting that Tyson may have hoped to create a cross-over product that would appeal to both sausage and frankfurter consumers. For instance, focus groups sessions conducted before launch suggested that PARK'S FINEST was seen “as a satisfying sausage in a convenient, fully-cooked, smaller link.” (App. at 788 (emphasis omitted).) However, as the District Court noted “[a] consumer who encounters the [PARK'S FINEST] product would not be privy to how [Tyson] intended for the product to be seen” and would be exposed to the unambiguously true statement that the product contains “uncured beef frankfurters.” Parks, 186 F. Supp. 3d at 416 n.2.
11. For instance, Parks notes that some dictionaries define a “hot dog” as “a small cooked sausage,” (Opening Br. at 22 (quoting hot dog, Merriam-Webtster's Online Dictionary, https://www.merriam-webster.com/dictionary/hotd̈og (last visited on May 1, 2017))), and frankfurter as “a cured cooked sausage[.]” (Opening Br. at 22 (quoting frankfurter, Merriam-Webtster's Online Dictionary, https://www.merriam-webster.com/dictionary/frankfurter (last visited on May 1, 2017)).) Parks also points out that hot dogs and sausages are grouped in the same category under the Nice Classification system, an international classification system for the registration of trademarks, and that Parks's food marketing expert testified that “the food industry considers and treats hot dogs and sausages as one category.” (Opening Br. at 22; App. at 633-34.)
12. Prior to 1989, § 43(a) of the Lanham Act did not clearly distinguish between false association and false advertising claims and merely prohibited “false designation[s] of origin” generally. Lanham Act, Pub L. No. 79-489, § 43, 60 Stat. 427, 441 (1946); see also 5 McCarthy on Trademarks §§ 27:6, 27:7 (describing the history of § 43(a)). In 1989, the statute was amended by creating separate subsections covering false association and false advertising, and by adding the modifier “geographic” before “origin” in the false advertising subsection. 15 U.S.C.A. § 1125(a)(1)(B); 5 McCarthy on Trademarks §§ 27:6, 27:7.
13. We leave for another day the question of how precisely the claimed geographic origin must match the actual place of origin. We do not need to decide, for instance, whether a company headquartered in a particular place can truthfully label a product as coming from that location even though the product was manufactured elsewhere. In Re Nantucket Allserve Inc., 28 U.S.P.Q.2d 1144 (T.T.A.B. Aug. 11, 1993) (refusing to register the mark NANTUCKET NECTARS when the company was headquartered in Nantucket but the product was manufactured in Worcester, MA); but see In Re Joint-Stock Co. Baik, 80 U.S.P.Q.2d 1305 (T.T.A.B. June 8, 2006) (registering a trademark for BAIKALSKAYA VODKA manufactured in a town near Lake Baikal and using some water from the Lake in the manufacturing process). Nor do we need to resolve a case like Pernod Ricard, in which the product in question had once been manufactured in Havana but was no longer. Pernod Ricard, 653 F.3d at 244.
14. To be clear, we are not opining that a mark must be a “geographically distinctive trademark” before a manufacturer may be found liable for falsely advertising the geographic origin of its goods. The Second Circuit appears to have reached that conclusion, see Forschner Grp., 30 F.3d at 353 (“Questions of false designation of geographic origin are properly considered within the analytical framework used to gauge the distinctiveness of trademarks.”), but whether “geographic origin” in § 1125(a)(1)(B) is a term of art incorporated in full from the common law of trademark protection was not briefed by the parties and we need not resolve that question today. We only hold that the term “geographic origin” must refer, at the very least, to the place of the origin of goods and that PARK'S FINEST does not meet that requirement.
15. The District Court also concluded that PARK'S FINEST did not have a “tendency to deceive a substantial portion of the intended audience.” Parks, 186 F. Supp. 3d at 417 (relying on Pernod Ricard, 653 F.3d at 248). Because we conclude that Parks's claims are not false advertising claims, there is no necessity to address that conclusion.
16. Wholly generic names cannot have trademark significance, 2 McCarthy on Trademarks § 12:1, but genericity is not at issue in this case.
17. Indeed, before its bankruptcy and acquisition, the name of the Parks Sausage Company was H.G. Parks, Inc.
18. There is no evidence in the record concerning customer testimony or the use of the mark in trade journals, so factors (6) and (7) of the Commerce National list are not relevant.
19. Lang surveyed 893 individuals for Parks. Participants were assigned randomly to one of four groups: two test groups and two control groups. The test groups were shown an array of five hot dog or sausage products including PARKS, PARK'S FINEST and three other brands. The control groups were shown PARKS sausages and an imaginary mark called BALL PARK OUR FINEST. All four groups were asked an identical series of questions to gauge product confusion. First, they were asked if “two or more” of the products were “from the same company or are affiliated or connected [.]” (App. at 664.) Then, those who said yes were asked to identify which two brands were affiliated and to explain why they felt the products were affiliated.Of those in the test groups, 49.1% thought that two or more of the products were affiliated, while only 19.3% thought that in the control group. Based on the answers to the follow-up questions, Lang concluded that around one in five hot dog or sausage consumers were likely to confuse PARKS and PARK'S FINEST.
20. As the District Court noted, the survey might have shed light on secondary meaning if participants had been asked to identify the source of PARKS or PARK'S FINEST, but no such follow-up questions were asked. Parks, 186 F. Supp. 3d at 425; see also Ideal Toy Corp. v. Plawner Toy Mfg. Corp., 685 F.2d 78, 82 (3d Cir. 1982) (relying on a consumer survey showing participants an imitation product and asking them to identify the manufacturer); E. I. DuPontde Nemours & Co. v. Yoshida Int'l, Inc., 393 F. Supp. 502, 520 (E.D.N.Y. 1975) (involving a trademark dispute between the makers of TEFLON products and a zipper named EFLON, and highlighting a survey that established that TEFLON was “fairly well known” among prospective EFLON purchasers).
21. Tyson raised a variety of methodological concerns with the survey that Parks employed. Given our conclusion that the survey design is incapable of proving secondary meaning, we need not discuss those other flaws.
22. The District Court relied on Tyson's data with regard to the market share of PARKS products in various regions of the country, because Parks did not provide a detailed breakdown of sales. Tyson's expert relied on data provided by IRI, a data analytics company. In the IRI dataset, the “northeast” was defined as including Pennsylvania, New Jersey, New York, Connecticut, Rhode Island, Massachusetts, New Hampshire, Vermont, and Maine, and the “mid-south” as Maryland, West Virginia, Virginia, Kentucky, Tennessee, and North Carolina. On appeal, Parks does not question the District Court's reliance on the IRI data or the categorization of various regions of the country, and so we likewise rely on Tyson's data.
23. The District Court noted that Parks failed to define “[e]astern United States” but concluded, based on Parks's pleadings, that the term encompassed “all states east of the Mississippi River.” Parks, 186 F. Supp. 3d at 421 n.8 (quoting Parks's brief). Parks has not contested that conclusion.
24. Although the lack of secondary meaning would itself be a sufficient basis for affirming the District Court's grant of summary judgment, Tyson's alternative argument that there is insufficient evidence of “market penetration” is persuasive. (Ans. Br. at 41.) “[T]he trademark of a prior user should be protected from infringement by a subsequent user of the same mark only in areas where the prior user has established a market for its goods[.]” Nat. Footwear Ltd. v. Hart, Schaffner & Marx, 760 F.2d 1383, 1394 (3d Cir. 1985); see also Charles Jacquin Et Cie, Inc. v. Destileria Serralles, Inc., 921 F.2d 467, 473-74 (3d Cir. 1990) (applying Nat. Footwear to the secondary meaning context). To determine market penetration, we consider(1) the volume of sales of the trademarked product; (2) the growth trends (both positive and negative) in the area; (3) the number of persons actually purchasing the product in relation to the potential number of customers; and (4) the amount of product advertising in the area.Nat. Footwear Ltd., 760 F.2d at 1398-99 (footnotes omitted).For many of the same reasons that we conclude that the PARKS mark does not enjoy secondary meaning throughout the eastern United States, we also conclude that it lacked adequate market penetration. In particular, Parks sought to enjoin Tyson from selling in at least five states where there is no evidence of sales at all. And sales in many states outside of the northeast are extremely limited if not “de minimis,” id. at 1400.
25. Because we conclude that PARKS does not have secondary meaning, we do not need to consider whether Parks offered sufficient proof of a likelihood of confusion. Likewise, we do not address Tyson's affirmative defense of abandonment through uncontrolled licensing.
JORDAN, Circuit Judge.