Souza v. Exotic Island Enterprises, Inc.

68 F.4th 99

• Court: Court of Appeals for the Second Circuit
• Decided: May 19, 2023
• Docket: 21-2149
• Status: Published

Opinion

21-2149-cv Souza v. Exotic Island Enterprises, Inc.

UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT

August Term, 2022

Argued: December 2, 2022 Decided: May 19, 2023

Docket No. 21-2149-cv

ALANA SOUZA, AKA ALANA CAMPOS, BROOKE BANX, BROOKE TAYLOR-JOHNSON, JACLYN SWEDBERG, JAIME EDMONDSON-LONGORIA, JESSICA HINTON, TIFFANY TOTH-GRAY, URSULA SANCHEZ, AKA URSULA MAYES,

Plaintiffs-Appellants,

— v. —

EXOTIC ISLAND ENTERPRISES, INC., DBA MANSION GENTLEMEN’S CLUB & STEAKHOUSE, KEITH SLIFSTEIN,

Defendants-Appellees,

EXCLUSIVE EVENTS & PROMOTIONS INC., DBA THINK SOCIAL FIRST,

Third-Party-Defendant.*

* The Clerk of Court is directed to amend the caption as displayed above. B e f o r e:

LYNCH, NARDINI, and MENASHI, Circuit Judges.

Plaintiffs-Appellants, a group of current and former professional models, appeal from a judgment of the United States District Court for the Southern District of New York (Karas, J.) granting summary judgment against them on a variety of claims arising from the use of their images in social media posts promoting a “gentlemen’s club” operated by Defendants-Appellees. On appeal, Plaintiffs argue, among other things, that the district court misapplied this Court’s framework for evaluating the likelihood of consumer confusion in the context of a Lanham Act false endorsement claim, misconstrued Supreme Court guidance constraining the Lanham Act’s reach in the false advertising context, and applied the wrong statute of limitations to Plaintiffs’ state law right of publicity claims. We disagree. We conclude that the district court properly granted summary judgment on Plaintiffs’ federal claims and the majority of their state law claims, and permissibly declined to exercise supplemental jurisdiction over their remaining claims. We therefore AFFIRM the judgment of the district court.

JOHN V. GOLASZEWSKI, Casas Law Firm, P.C., New York, NY, for Plaintiffs-Appellants.

MICHAEL KOLB, O’Connor & Partners, PLLC, Kingston, NY, for Defendants-Appellees.

2 GERARD E. LYNCH, Circuit Judge:

This appeal concerns several claims brought by Plaintiffs-Appellants Alana

Souza (a/k/a Alana Campos), Brooke Banx, Brooke Taylor-Johnson, Jaclyn

Swedberg, Jaime Edmondson-Longoria, Jessica (a/k/a Jessa) Hinton, Ursula

Sanchez (a/k/a Ursula Mayes), and Tiffany Toth-Gray (together, “Plaintiffs”) – all

current or former professional models – against Defendants-Appellees Exotic

Island (“Exotic”) and Keith Slifstein (together, “Defendants”). Those claims arise

from the basic undisputed allegation that Defendants, through a third-party

vendor, used images of Plaintiffs without their permission in social media posts

promoting a “gentlemen’s club” operated by Defendants.

After the parties cross-moved for summary judgment, the United States

District Court for the Southern District of New York (Kenneth M. Karas, J.)

granted summary judgment in Defendants’ favor. Specifically, it concluded that

(1) Plaintiffs’ false endorsement claims, as supported by the evidentiary record

on summary judgment, were foreclosed by our decision in Electra v. 59 Murray

Enters., Inc., 987 F.3d 233 (2d Cir.), cert. denied, 142 S. Ct. 563 (2021); (2) their false

advertising claims were founded upon injury that either fell outside the zone of

interests protected by the Lanham Act, or that was unsubstantiated by the record;

3 and (3) the bulk of their state-law right of publicity claims were barred by New

York’s one-year statute of limitations for such claims. The district court then

declined to exercise supplemental jurisdiction over the few state-law claims that

were not time-barred.

We agree with the district court on all counts, and therefore AFFIRM its

judgment in full.

BACKGROUND

Although the parties cross-moved for summary judgment below, because

this appeal concerns the district court’s grant of Defendants’ motion, we construe

the record in the light most favorable to Plaintiffs. See Heublein, Inc. v. United

States, 996 F.2d 1455, 1461 (2d Cir. 1993). The factual backdrop of this case,

however, is simple and largely undisputed.

I. Factual Background

A. The Parties

Exotic and its president, Slifstein, operate Mansion Gentlemen’s Club &

Steakhouse (“Mansion”) in Newburgh, New York. Plaintiffs are or were

professional models whose pictures appeared without their consent, and without

compensation, on social media sites associated with Mansion. The Instagram and

4 Facebook posts at issue were actually created and published to Defendants’

accounts by Third-Party Defendant Exclusive Events & Promotions d/b/a Think

Social First, a third-party vendor authorized by Exotic to operate those accounts

on its behalf.

Each Plaintiff works or has worked as a professional model, promoting her

“image, likeness and/or identity . . . for the benefit of various clients, commercial

brands, media and entertainment outlets.” E.g., Joint Appendix (J.A.) 63. In

substantially identical declarations, Plaintiffs have testified that because they

“rely on [their] professional reputation[s] to book modeling and advertising

jobs,” their reputations are “critical” to the opportunities they are offered, and

they therefore “have spent considerable time and energy” protecting and

policing their images and reputations, and carefully negotiating their modeling

fees based on “informed assessment[s]” of any given job’s effect on their brands.

E.g., id. 63-64.

Plaintiffs have enjoyed varying levels of success and visibility in their

modeling careers. Several have appeared in magazines, advertising campaigns,

television episodes, and films. Some are former Playboy Playmates, including

five (Swedberg, Campos, Hinton, Edmondson-Longoria, and Toth-Gray) who

5 were named Playmate of the Month between 2010 and 2012 and one (Swedberg)

who was named 2012 Playmate of the Year. Their highest single-year modeling

earnings range from around $18,300 to around $107,000. Their social media

footprints range from several thousand to a few million followers.

Plaintiffs’ links to New York State are fleeting at best. None have lived in

New York, many have never even worked in New York, and several others have

made just a single modeling appearance in the state. Only Mayes recalled making

multiple promotional appearances in New York, between 2005 and 2009, though

she never lived in the state.

Most of the Plaintiffs no longer work as full-time models. Banx and Taylor-

Johnson both stopped modeling around 2014, followed soon thereafter by

Edmondson-Longoria and Swedberg in, respectively, 2015 and 2017. Mayes has

worked as a model only sporadically since 2014. Campos began working

primarily as a real estate agent in 2015. Hinton and Toth-Gray continue to do

modeling work, alongside other professional activities.

B. The Social Media Posts

Published between 2014 and 2018, each of the posts at issue set revealing

photographs of Plaintiffs against advertising copy linked thematically to each

6 visual in some way. For example, one of Mansion’s September 2014 Facebook

posts featured a picture of Taylor-Johnson in an apparent school uniform that

included a short plaid skirt, captioned: “Friday Oct 17th SEXY SCHOOL GIRL

PARTY! No Cover For Ladies That Wear A Sexy School Girl Skirt[.] All Our

Dancers Will Be Wearing Short Plaid Skirts!” J.A. 37. A May 2015 Instagram post

featured Swedberg, astride a motorcycle in lingerie, captioned: “EVERY

TURSDAY [sic] is BIKE NIGHT AT #THEMANSION #bikenight

#adultentertainment #stripclub #strippers #gentlemensclub #newburgh

#hudsonvalley #steakhouse.” Id. 38. A December 2015 Facebook post featured

Hinton, in a Santa Claus cap and topless save for a bra, captioned: “The

Naughtiest party of the year is happening tonight! Are you naughty enough to

party with our girls?” Id. 41. Matching Instagram and Facebook posts from

January 2016 featured Banx posing suggestively in a cropped Pittsburgh Steelers

top, captioned, respectively, “No sexier place to watch the #NFL” and “No sexier

place to watch the Steelers vs. Broncos!!” Id. 35-36. A Facebook post from the

same month featured Edmondson-Longoria posed in a similarly scant Seattle

Seahawks top, captioned: “It’s FOOTBALL Sunday at #Mansion! Watch all the

BIG games today with us!” Id. 40. In each of those pictures, Plaintiffs’ faces were

7 visible. Several other posts included images of Plaintiffs whose full faces were

not visible.

Defendants claim to have taken all possible steps to remove the posts once

they were made aware of Plaintiffs’ grievances. However, as of July 2020, at least

one (depicting Campos) remained live on Mansion’s Instagram page. Defendants

attribute their failure to remove the image to password difficulties that hampered

their access to that Instagram account.

During discovery, each Plaintiff was asked in an interrogatory to identify

“any and all jobs and/or work lost as a result of the allegations asserted in the

Complaint.” E.g., J.A. 329. None identified any specific lost opportunities.

Instead, each gave the following response, verbatim:

[I]t is a well-known fact that prospective clients have no duty to disclose to the model their reasoning for why the model was denied an endorsement opportunity. It is also a widely known fact that on the outset of creating a highly coveted endorsement deal, clients and/or their advertising agencies will conduct due diligence of models in advance of contacting a model to discuss an endorsement opportunity. As a result of these common industry practices, Plaintiff has not been contacted directly by a third party with notice of a refusal to do business or the rescission of an offer to hire due to Defendant’s use of Plaintiff’s image.

E.g., id.

8 II. Procedural Background

A. This Litigation

Plaintiffs filed their Complaint in October 2018. The Complaint asserted

several causes of action: (1) false advertising and (2) false endorsement under

Section 43 of the Lanham Act; (3) violation of their privacy/publicity rights under

New York Civil Rights Law §§ 50-51; (4) violation of New York’s Deceptive Trade

Practices Act; and (5) defamation. Plaintiffs later withdrew their deceptive trade

practices and defamation claims. In January 2019, Defendants answered,

asserting a statute of limitations affirmative defense, and the parties began

several years of discovery. In November 2019, Defendants sought and were

granted leave to file a third-party complaint against Exclusive Events. Eventually,

in February 2021, the parties filed dueling summary judgment motions.

Defendants also moved to strike the expert report and survey of Plaintiffs’ expert,

Martin Buncher.

In August 2021, the district court denied Plaintiffs’ motion and granted

Defendants’ motion for summary judgment (and along with it, Defendants’

motion to exclude portions of Buncher’s report), dismissing the bulk of Plaintiffs’

9 claims as a matter of law and declining to exercise supplemental jurisdiction over

a few leftover state law claims. This appeal followed.

B. Similar Lawsuits Brought by Plaintiffs’ Counsel

Meanwhile, less than a week after the parties had filed their summary

judgment motions in February 2021, another panel of this Court decided Electra v.

59 Murray Enters., Inc., 987 F.3d 233 (2d Cir. 2021). In Electra, a group of

professional models – including several of the Plaintiffs in this case, represented

by the same counsel – sued a group of defendants who, also through a

third-party vendor, had allegedly used those models’ photographs to promote

the defendants’ strip clubs. Id. at 240-41. The Electra plaintiffs asserted several of

the same causes of action as those asserted here. Id. at 242. For reasons discussed

below, we affirmed in relevant part the district court’s grant of summary

judgment for the defendants on most plaintiffs’ false endorsement and right of

publicity claims. Id. at 239, 251, 257-58 (also overturning portions of the district

court’s summary judgment order for reasons not pertinent here).

Electra was just one of a group of such cases brought in the Southern

District of New York by the same counsel on behalf of different groups of

models, including many of the same plaintiffs. See Edmondson v. RCI Hosp.

10 Holdings, Inc., No. 16-CV-2242-VEC, 2021 WL 4499031 (S.D.N.Y. Oct. 1, 2021);

Gibson v. SCE Grp., Inc., 391 F. Supp. 3d 228 (S.D.N.Y. 2019).1 Plaintiffs’ counsel

and his firm have also filed many similar lawsuits in other jurisdictions, again

often on behalf of some of the same plaintiffs.2

1 One of those cases has subsequently settled. Edmondson v. RCI Hosp. Holdings, Inc., No. 16-CV-2242-VEC (S.D.N.Y. Aug. 10, 2022), Dkt. No. 225. The other is currently on appeal before this Court. Gibson v. SCE Grp., Inc., No. 22-916 (2d Cir.). 2 See, e.g., Pepaj v. Paris Ultra Club LLC, No. CV-19-01438-PHX-MTL, 2021 WL 632623 (D. Ariz. Feb. 18, 2021); Takeguma v. Freedom of Expression LLC, No. CV-18-02552-PHX-MTL, 2021 WL 487884 (D. Ariz. Feb. 10, 2021); Pinder v. 4716 Inc., 494 F. Supp. 3d 618 (D. Ariz. 2020); Gray v. LG&M Holdings LLC, No. CV-18-02543-PHX-SRB, 2020 WL 6200165 (D. Ariz. Sept. 23, 2020), reconsideration denied, 2020 WL 9074801 (D. Ariz. Oct. 14, 2020); Longoria v. Million Dollar Corp., No. 18-CV-02266-PAB-NYW, 2021 WL 1210314 (D. Colo. Mar. 31, 2021); Moreland v. Beso Lounge & Rest. LLC, No. 3:19-CV-00958 (VLB), 2020 WL 5302312 (D. Conn. Sept. 4, 2020); Burciaga v. Gold Club Tampa, Inc., No. 8:16-CV-790-T-27JSS, 2016 WL 9526567 (M.D. Fla. Dec. 28, 2016); Moreland v. Club 390 Corp., No. 18 C 7441, 2019 WL 13076638 (N.D. Ill. Aug. 26, 2019); Lundberg v. One Three Five, Inc., No. 2:19-CV-00692-RJC, 2022 WL 2669136 (W.D. Pa. Jan. 14, 2022), report and recommendation adopted, 2022 WL 2668557 (W.D. Pa. Mar. 15, 2022); Geiger v. Abarca Fam. Inc., No. 3:21CV771 (DJN-EWH), 2022 WL 4242838 (E.D. Va. July 29, 2022), report and recommendation adopted, No. 3:21CV771(RCY), 2022 WL 4241649 (E.D. Va. Sept. 14, 2022).

11 DISCUSSION

Plaintiffs present five basic arguments on appeal, spread across their three

remaining causes of action. First, they contend that the district court erred in

evaluating their false endorsement claims by (1) oversimplifying its inquiry into

the strength of Plaintiffs’ marks by focusing on recognizability alone;

(2) wrongfully excluding certain expert evidence; and (3) bungling its overall

balancing of the Polaroid likelihood of confusion factors. Next, Plaintiffs argue

that the district court (4) adopted too restrictive a take on the zone of interests

protected by the Lanham Act. Finally, they challenge the district court’s decision

(5) to apply a one-year statute of limitations to Plaintiffs’ right of publicity claims.

We address, and reject, Plaintiffs’ arguments below.

I. Standards of Review

A. Summary Judgment, Generally

“We review a district court’s grant of summary judgment de novo,

resolving all ambiguities and drawing all permissible inferences in favor of the

nonmoving party.” Tiffany & Co. v. Costco Wholesale Corp., 971 F.3d 74, 83 (2d Cir.

2020). Although “[t]he party seeking summary judgment bears the burden of

establishing that no genuine issue of material fact exists,” Vivenzio v. City of

12 Syracuse, 611 F.3d 98, 106 (2d Cir. 2010) (internal quotation marks omitted),

“[w]hen the burden of proof at trial would fall on the nonmoving party, it

ordinarily is sufficient for the movant to point to a lack of evidence to go to the

trier of fact on an essential element of the nonmovant’s claim,” CILP Assocs., L.P.

v. PriceWaterhouse Coopers LLP, 735 F.3d 114, 123 (2d Cir. 2013) (internal quotation

marks omitted). In that scenario, “once such a showing is made, the non-movant

must set forth specific facts showing that there is a genuine issue for trial.”

Weinstock v. Columbia Univ., 224 F.3d 33, 41 (2d Cir. 2000) (internal quotation

marks omitted).

B. Likelihood of Confusion

Our standard of review has historically been tougher to pin down in

Lanham Act cases involving our eight familiar Polaroid factors – discussed in

further detail below – which measure the likelihood of consumer confusion in a

given case. The source of that tension is that, although we have always held that

the district court’s balancing of those factors should be reviewed de novo, some of

our past cases have “purported to afford ‘considerable deference’ to district

courts’ findings ‘with respect to predicate facts underlying each Polaroid factor,’”

Tiffany, 971 F.3d at 85, quoting Playtex Prods., Inc. v. Ga.-Pac. Corp., 390 F.3d 158,

13 162 (2d Cir. 2004), or even to the district court’s “‘finding on each factor’

generally,” id., quoting Natural Organics, Inc. v. Nutraceutical Corp., 426 F.3d 576,

578 (2d Cir. 2005).

But although our stance may have wobbled over the years, recent cases

have solidified our view that, “[i]nsofar as the determination of whether one of

the Polaroid factors favors one party or another involves a legal judgment – which

it often does – we must review that determination de novo.” Car-Freshner Corp. v.

Am. Covers, LLC, 980 F.3d 314, 327-28 (2d Cir. 2020) (internal quotation marks

omitted). To that end, we have cautioned that past cases hinting at “deference” to

the district court “should not be read to suggest that a district court deciding a

motion for summary judgment in a trademark infringement case has greater

discretion than it would have in a non-trademark case.” Tiffany, 971 F.3d at 85

(internal quotation marks and alteration omitted) (adding that “we have never

purported to expand a district court’s license to make factual findings at summary

judgment beyond those very limited circumstances in which the uncontroverted

evidence and the reasonable inferences to be drawn in the nonmoving party’s

favor support only a single conclusion” (emphasis in original) (internal quotation

marks omitted)).

14 And so, to reiterate, “we review de novo a ruling on whether the plaintiff

has shown a likelihood of confusion because we consider the issue to be a

question of law.” Car-Freshner, 980 F.3d at 326.

c. Other Matters

Finally, we review for abuse of discretion a district court’s decision to

admit or exclude expert evidence. Restivo v. Hessemann, 846 F.3d 547, 575 (2d Cir.

2017) (adding that the district court’s decision “is to be sustained unless

manifestly erroneous” (internal quotation marks omitted)). The same is true of a

district court’s decision to decline to exercise supplemental jurisdiction over state

law claims. Motorola Credit Corp. v. Uzan, 388 F.3d 39, 56 (2d Cir. 2004).

II. False Endorsement

Section 43(a) of the Lanham Act prohibits the

use[] in commerce [of] any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact which . . . is likely to cause confusion, or to cause mistake, or to deceive as to the affiliation, connection, or association of such person with another person, or as to the origin, sponsorship, or approval of his or her goods, services, or commercial activities by another person[.]

15 15 U.S.C § 1125(a)(1)(A). This provision is intended to “prevent consumer

confusion regarding a product’s source,” to “enable those that fashion a product

to differentiate it from others on the market,” EMI Catalogue P’ship v. Hill,

Holliday, Connors, Cosmopulos Inc., 228 F.3d 56, 61 (2d Cir. 2000) (internal

quotation marks omitted), and to protect against the risk that consumers will

mistakenly “believe that the trademark owner sponsors or endorses the use of the

challenged mark,” Kelly-Brown v. Winfrey, 717 F.3d 295, 304 (2d Cir. 2013)

(internal quotation marks omitted).

To prevail on a so-called false endorsement claim under Section 43 of the

Lanham Act, a plaintiff must prove, among other uncontested requirements,

“that there is the likelihood of confusion between the plaintiff’s good or service

and that of the defendant.” Electra, 987 F.3d at 257 (internal quotation marks

omitted). To determine whether there is a likelihood of consumer confusion, we

look to our eight familiar Polaroid factors:

(1) strength of the trademark; (2) similarity of the marks; (3) proximity of the products and their competitiveness with one another; (4) evidence that the senior user may bridge the gap by developing a product for sale in the market of the alleged infringer’s product; (5) evidence of actual consumer confusion; (6) evidence that the imitative mark was adopted in bad faith; (7) respective

16 quality of the products; and (8) sophistication of consumers in the relevant market.

Kelly-Brown, 717 F.3d at 307 (internal quotation marks omitted); see Polaroid Corp.

v. Polarad Elecs. Corp., 287 F.2d 492, 495 (2d Cir. 1961). Those factors are neither

exhaustive nor applied mechanically. See Kelly-Brown, 717 F.3d at 307; Brennan’s,

Inc. v. Brennan’s Rest., L.L.C., 360 F.3d 125, 130 (2d Cir. 2004). No single factor is

dispositive; rather, each is evaluated “in the context of how it bears on the

ultimate question of likelihood of confusion as to the source of the product.”

Brennan’s, 360 F.3d at 130 (internal quotation marks omitted).

A. Strength of Mark

The district court held that the first Polaroid factor – strength of mark –

favored Defendants. We agree.

1. Recognizability as Strength

Plaintiffs challenge that conclusion on a few fronts. First, they argue that

the district court misstepped by treating recognizability as the “bottom line”

barometer for strength of mark in false endorsement claims of this sort.

Appellant’s Br. 21-28.

17 Plaintiffs are mistaken. We recently endorsed that precise approach in

Electra. In that case, the district court had declared without qualification that in

“celebrity” false endorsement cases “the ‘strength of the mark’ refers to the level

of recognition that the plaintiff has among the consumers to whom the

advertisements are directed.” Toth v. 59 Murray Enters., Inc., No. 15-CV-8028-

NRB, 2019 WL 95564, at *6 (S.D.N.Y. Jan. 3, 2019) (internal quotation marks

omitted). On appeal, the Electra panel held that the district court in that case had

“properly analyzed the record of each [plaintiff’s] public prominence to

determine the strength of their marks.” Electra, 987 F.3d at 258. Then, amplifying

a prior district court’s holding that the “misappropriation of a completely

anonymous face could not form the basis for a false endorsement claim,” the

panel reiterated that “because the ultimate question under Polaroid . . . is the

likelihood of consumer confusion, the district court properly analyzed

[plaintiffs’] recognizability.” Id., quoting Bondar v. LASplash Cosms., No. 12-CV-

1417-SAS, 2012 WL 6150859, at *7 (S.D.N.Y. Dec. 11, 2012). We are bound by

Electra, which presented effectively identical issues in an effectively identical

factual context.

18 Perhaps recognizing that we are so constrained, Plaintiffs rely upon several

critiques of the Electra panel’s approach. Most significantly, they cite a number of

cases for the proposition that the strength inquiry is intended to be broader than

a simple, one-dimensional “fame” test. See, e.g., Brennan’s, 360 F.3d at 130

(explaining that the strength of mark factor also includes a mark’s “inherent

distinctiveness”); Coach Servs., Inc. v. Triumph Learning LLC, 668 F.3d 1356, 1367

(Fed. Cir. 2012) (remarking that “fame cannot overwhelm the other” factors

relevant to the likelihood of confusion calculus); see also 15 U.S.C. § 1125(a)(1)

(providing for civil remedies for “any person who believes that he or she is or is

likely to be damaged” (emphasis added)).

As a general matter, Plaintiffs are correct that we have long recognized that

the strength of a mark typically “encompasses two different concepts, both of

which relate significantly to likelihood of consumer confusion.” Virgin Enters. Ltd.

v. Nawab, 335 F.3d 141, 147 (2d Cir. 2003) (emphases added). The first is “inherent

distinctiveness,” id., which looks to the mark itself, divorced from consumers’

actual knowledge of it, and classifies marks on a spectrum ranging from stronger

“fanciful” or “arbitrary” marks down to progressively weaker “suggestive,”

“descriptive,” or “generic” marks, see Gruner + Jahr USA Pub., a Div. of Gruner +

19 Jahr Printing & Pub. Co. v. Meredith Corp., 991 F.2d 1072, 1075 (2d Cir. 1993). The

second concept is “acquired distinctiveness,” which we have defined as “fame, or

the extent to which prominent use of the mark in commerce has resulted in a

high degree of consumer recognition.” Virgin, 335 F.3d at 147; see Guthrie

Healthcare Sys. v. ContextMedia, Inc., 826 F.3d 27, 41 n.4 (2d Cir. 2016) (“The theory

is that a mark similar to a famous mark is more likely to cause confusion, or at

least more likely to cause a more widespread confusion, than a mark similar to a

relatively unknown one.”).

Plaintiffs run aground, however, when they suggest that we may not focus

on recognizability in this context. They insist that such an approach cannot be

reconciled with the Supreme Court’s rejection, in Two Pesos, Inc. v. Taco Cabana,

Inc., of this Court’s old rule declaring “protection for trade dress unavailable

absent proof of secondary meaning.” 505 U.S. 763, 772 (1992). But the Supreme

Court’s problem with our old rule was not that it permitted secondary meaning

(i.e., acquired distinctiveness) to supplant inherent distinctiveness. To the

contrary, the Supreme Court expressly reiterated that distinctiveness can arise,

independently, in either form: “An identifying mark is distinctive and capable of

being protected if it either (1) is inherently distinctive or (2) has acquired

20 distinctiveness through secondary meaning.” Id. at 769 (emphasis in original).

Rather, the Supreme Court’s concern was that this Court’s rule did not permit

trade dress to be protected even where it had been found (in that case, by a jury) to

be inherently distinctive. Id. at 772. We have applied no such rule in the false

endorsement context. Nor are we saddled with any such inherent distinctiveness

finding – or even an argument from Plaintiffs that their marks are inherently

distinctive – in this case.

And therein lies the wisdom underpinning the Electra panel’s approach.

The concept of inherent distinctiveness is simple enough to apply where, say, one

restaurant sues another for coopting its “festive” dining setup, “decorated with

artifacts, bright colors, paintings and murals,” as in Two Pesos, 505 U.S. at 765, or

even when the subject matter is human names, as in, e.g., 815 Tonawanda Street

Corp. v. Fay’s Drug Co., 842 F.2d 643, 648 (2d Cir. 1988); Paco Sport, Ltd. v Paco

Rabanne Perfumes, 234 F.3d 1262 (2d Cir. 2000) (summary order). It is more

awkward to apply when it effectively interrogates how much one human being

does, or does not, physically resemble another. And that includes, as this case

vividly illustrates, inquiries concerning the extent to which one unnamed model,

21 whose face may or may not be shown, and who may appear to be of a certain

race, ethnicity, body type, physical stature, etc., resembles another.

The usual criteria for inherent distinctiveness, in any event, have little

application here. In a false endorsement case like this one, the “mark” in question

is the identity of the purported endorser herself. Bondar v. LASplash Cosms., No.

12-CV-1417-SAS, 2012 WL 6150859, at *5 (S.D.N.Y. Dec. 11, 2012) (“Courts in this

Circuit have recognized that celebrities have a trademark-like interest in their

name, likeness, and persona that may be vindicated through a false endorsement

claim under the Lanham Act.”). But unlike a conventional adopted mark, an

endorser’s face and body fall nowhere on the familiar spectrum from “arbitrary”

to “generic”; their identity inherently is their mark. And where any face or figure

regarded as “attractive,” J.A. 1547, will do, notwithstanding the anonymity of the

actual person whose face or figure is depicted (and the negligible endorsement

value derived from that actual person’s connection to the product being sold), the

unauthorized use of that person’s image may invade rights granted by other

statutes or common law sources, see, e.g., the discussion of N.Y. Civ. Rights Law

22 §§ 50-51 below,3 but creates no risk of consumer confusion as conceived under

the Lanham Act.

The Electra panel’s focus on recognizability thus serves the purposes of

trademark law in the false endorsement context.4 It properly calibrates strength

as a function of the extent to which the purported endorser’s identity and

goodwill can be linked to the product being sold. It is also consistent with our

precedent recognizing that “even a common name mark may warrant protection

as a strong mark if it has achieved distinctiveness in the marketplace,” but

emphasizing that “if the mark is not recognized by the relevant consumer group,

3 Defendants in fact conceded liability under §§ 50-51 with respect to the one Plaintiff’s claim whose timeliness they did not contest below. 4 Although Electra was the first time we endorsed that approach, district courts in this Circuit have long applied it to false endorsement cases. See, e.g., Pelton v. Rexall Sundown, Inc., No. 99-CV-4342-JSM, 2001 WL 327164, at *3 (S.D.N.Y. Apr. 4, 2001) (remarking that the “strength of [plaintiff’s] mark or name is a crucial factor in determining likelihood of consumer confusion,” and granting summary judgment to defendants because “there [was] no evidence that [plaintiff] is a recognizable celebrity”); Jackson v. Odenat, 9 F. Supp. 3d 342, 357 (S.D.N.Y. 2014) (“In a celebrity endorsement case, the ‘mark’ is the plaintiff’s persona and the ‘strength of the mark’ refers to the level of recognition that the plaintiff has among the consumers to whom the advertisements are directed.”); Gibson, 391 F. Supp. 3d at 245-46 (“In [a false endorsement case] the Court, like other courts in this District, interprets the strength of the mark to mean the level of recognition the celebrity has among the segment of the public to whom the goods are advertised.” (internal quotation marks omitted)).

23 a similar mark will not deceive . . . consumers.” Brennan's, 360 F.3d at 132.

Finally, the adoption of that approach in Electra is, of course, binding on us, as it

was binding on the district court. And the district court faithfully and correctly

applied it.

2. Lack of Evidence of Recognizability

The district court also correctly evaluated the evidence relevant within that

framework. First, it permissibly excluded Plaintiffs’ putative expert testimony as

unreliable; it then correctly concluded that because Plaintiffs were left with next

to no evidence of recognizability, the strength-of-mark Polaroid factor weighed in

Defendants’ favor.

As to the expert evidence, Plaintiffs argue that the district court abused its

discretion in excluding as unreliable testimony from Plaintiffs’ expert, Martin

Buncher. We disagree. “In deciding whether a step in an expert’s analysis is

unreliable, the district court should undertake a rigorous examination of”

multiple factors, including “the method by which the expert draws an opinion,”

and “should only exclude the evidence if the flaw is large enough that the expert

lacks good grounds for his or her conclusions.” Amorgianos v. Nat’l R.R. Passenger

Corp., 303 F.3d 256, 267 (2d Cir. 2002) (internal quotation marks omitted). Here,

24 the court permissibly determined that Buncher’s survey of 812 respondents “who

had patronized . . . a ‘Bikini Bar/Gentlemen’s Club/Strip Club’ in the past two

years,” J.A. 160, suffered from methodological flaws and therefore did not lay a

reliable foundation for his recognizability analysis.5 Specifically, the court flagged

the study’s absence of a control group, its overly inclusive approach to actual

recognition,6 and its failure to show respondents several Plaintiffs’ full faces

(which, the court noted, somehow did not appear to have meaningfully affected

respondents’ professed ability to recognize the Plaintiffs who were pictured).

Exclusion on those grounds was comfortably within the court’s discretion. See

Amorgianos, 303 F.3d at 266 (where the court finds that an expert’s opinion is

“based on data, a methodology, or studies that are simply inadequate to support

the conclusions reached, Daubert and Rule 702 mandate the exclusion of that

5 Buncher’s conclusions included his calculation that “almost half the respondents felt they recognized the Plaintiff[s’] images in the ads in some manner having seen them prior to this research.” J.A. 174. 6 Recall that the question of recognizability goes to the strength of the purported endorser’s “mark” – that is, the mark’s capacity to cause a consumer to infer “sponsorship or approval of the Clubs’ goods and services” by the particular person whose mark is at issue. Toth, 2019 WL 95564, at *6. That a survey respondent found a Plaintiff’s image vaguely familiar, or “in some manner [had] seen [that Plaintiff] prior to” taking the survey, J.A. 174, sheds little light on that question.

25 unreliable opinion testimony”); see also Edmondson v. RCI Hosp. Holdings, Inc., No.

16-CV-2242-VEC, 2020 WL 1503452, at *7-8 (S.D.N.Y. Mar. 30, 2020) (excluding a

similar report by Buncher for similar reasons, including the lack of a control

group and the use of a survey design that left “no way to verify whether

respondents truly recognized any of the Plaintiffs”).

Deprived of that evidence, Plaintiffs were left with precious few indicia of

recognizability. The district court correctly recognized that what remained7 fell

well short of establishing that any Plaintiff was sufficiently recognizable to

establish a strong mark.

7 This consisted of (1) vague and conclusory (and identical) written testimony from Plaintiffs that each had “achieved celebrity status and fame” and that “[o]n any given day, regardless of where I’m at, I am recognized by complete strangers and my fans who follow me on social media,” e.g., J.A. 62; and (2) evidence of Plaintiffs’ relatively modest modeling income, professional prominence, and social media footprints, which the district court determined (in a finding Plaintiffs do not contest on appeal) to be, at best, comparable to evidence the Electra panel had deemed insufficient to establish a strong mark. See Toth, 2019 WL 95564, at *7; see also Electra, 987 F.3d at 257-58 (affirming that portion of the Toth ruling).

26 B. Actual Confusion

The district court next concluded, after excluding more of Buncher’s

testimony,8 that the actual confusion Polaroid factor likewise favored Defendants.

Plaintiffs once again challenge the court’s exclusion analysis. But once

again, the district court was well within its discretion to determine that

methodological shortcomings counseled against admitting Buncher’s testimony

as to actual confusion. First, the court observed that the structure of the excluded

portion of the survey only allowed respondents to give their impressions of all of

the images of all of the Plaintiffs, and therefore did not permit respondents to

differentiate between specific images and/or specific Plaintiffs. Second, the court

underscored that the survey neither provided respondents with a “don’t know”

option nor instructed them “not to guess,” and therefore did not allow

respondents any recourse or guidance if they were unsure about the correct

answer. S.A. 27-29.

8 This portion of Buncher’s expert submission reported, among other things, that: (1) 62% of respondents agreed with the statement that “[a]ll the women shown in these ads have some affiliation, connection or association with those clubs in whose ad they appear”; (2) 75% agreed that “[a]ll of the women in these ads have agreed to sponsor, endorse or promote the club represented in these ads”; and (3) 76% agreed that “[a]ll of the women in the ads approve of the use of their image in those Club advertisements in which they appear.” J.A. 179.

27 Although Plaintiffs muster plausible defenses for their expert’s survey

design, and we recognize that reasonable district judges might differ as to

whether the proper outlet for such structural concerns is exclusion or cross-

examination, we endorsed exclusion on substantially similar grounds in Electra.

See 987 F.3d at 258 (concluding “for substantially the same reasons as those set

out in the district court’s opinion,” that the district court “did not abuse its

discretion in striking . . . the ‘Buncher Report’”), aff’g Toth, 2019 WL 95564, at *8

(“Buncher failed to provide survey takers with an opportunity to indicate lack of

knowledge or an instruction for participants not to guess . . . .”); see also

Edmondson, 2020 WL 1503452, at *7 (same, as to a similar report by the same

expert). Electra is indistinguishable in this respect, and remains binding on us.

Because that permissibly excluded evidence represented the only

meaningful evidence9 of actual confusion in Plaintiffs’ arsenal, the district court

9 Plaintiffs also brandish deposition testimony from Slifstein, who was asked if the fact that one of the posts at issue included the phrase “our girls” would cause a “reasonable customer” to believe “well, they’re saying our girls; this is one of the girls who is going to be there.” J.A. 149-50. Slifstein responded: “That’s up to them. I’m not – some may, some may not. It’s the individual’s discretion if they choose to believe what they see.” Id. at 150. But despite Plaintiffs’ dogged claims to the contrary, Slifstein’s testimony is not evidence of actual confusion. That is, Slifstein’s own subjective perception of the post’s effect on consumers may be probative of his state of mind, but nothing about Slifstein’s speculative answer to

28 was correct to conclude that the actual confusion Polaroid factor weighed in

a speculative question establishes that any actual consumer was actually confused. See W.W.W. Pharm. Co. v. Gillette Co., 984 F.2d 567, 574-75 (2d Cir. 1993) (observing that “speculative” testimony that did not purport to suggest that an actual consumer had been confused by the alleged infringement was not evidence of actual confusion). Moreover, we note that even if consumers were hoodwinked into believing that the “girls” in the posts were, in fact, the “girls” working at Mansion, thus giving rise to a plausible deceptive trade practices claim, see Electra, 987 F.3d at 259, that is not at all equivalent to a trademark claim founded upon the premise that one party has “exploit[ed] the goodwill [another party] has created for its trademark.” Streetwise Maps, Inc. v. VanDam, Inc., 159 F.3d 739, 745 (2d Cir. 1998). A generic misconception that the anonymous, unrecognized models in the posts are in fact the same models who work at Mansion inflicts the same injury on a consumer irrespective of any goodwill those models have cultivated in their own “marks,” because that misconception does not rely upon such goodwill in the first place. In such a scenario, the consumer is harmed because the anonymous “attractive” models from the advertisements that the consumer expects to see upon arrival at Mansion are not actually there, not because the consumer’s expectations were at all shaped by the models’ own identity or reputation or esteem or any of the other elements embedded in the concept of “goodwill” protected under the Lanham Act. See Robert G. Bone, Hunting Goodwill: A History of the Concept of Goodwill in Trademark Law, 86 B.U. L. REV. 547, 583 (2006). In other words, the misconception goes to the nature of the product itself, not the mistaken belief that someone whose imprimatur the consumer values has vouched for that product.

29 C. Bad Faith

Finally, the district court correctly held that the bad faith Polaroid factor

also weighed in Defendants’ favor. Once again, Electra is directly and

indistinguishably on point. In that case, as in this one, the defendant companies

“used third-party contractors to create the advertisements and publish them on

the Clubs’ websites and social media.” Electra, 987 F.3d at 241. And in that case,

as in this one, there was no evidence that the defendant clubs ever “asked [the

third-parties] to use a photograph of a specific person, instead requesting

photographs that would complement the advertised event or the purpose of a

particular webpage.” Id. That was enough at the summary judgment stage for the

Electra panel to award the bad faith Polaroid factor to the defendants. The same is

true here.

D. Balancing the Polaroid Factors

Plaintiffs’ final false endorsement argument is that the district court erred

in how it balanced the Polaroid factors in the aggregate. We disagree.

We start with one aspect of the district court’s Polaroid balancing that

Plaintiffs do not specifically target, but that is nonetheless woven into the

portions they do challenge. In this case, the district court focused its analysis on

30 just three of the eight Polaroid factors: strength of Plaintiffs’ marks, actual

confusion, and bad faith. It then “assume[d] without deciding” that the

remaining factors favored Plaintiffs, S.A. 31, briefly elaborating in a footnote as to

just one of those other factors and devoting no analysis to the rest.

That approach is discouraged – if not necessarily proscribed – in this

Circuit. Although it was once our position that district courts need not “slavishly

recite the litany of all eight Polaroid factors in each and every case,” Orient Exp.

Trading Co. v. Federated Dep’t Stores, Inc., 842 F.2d 650, 654 (2d Cir. 1988), we have

since charted a different course. In Natural Organics, for example, we

acknowledged that legacy, but observed that “our most recent cases on this issue

confirm” the opposite. Natural Organics, 426 F.3d at 579-80 (vacating a bench trial

judgment in part because the district court “did not discuss” several Polaroid

factors despite “considerable evidence” having been presented at trial on those

factors). That more recent perspective dates back to Judge Cabranes’s opinion in

Arrow Fastener, which explained that “it is incumbent upon the district judge to

engage in a deliberate review of each factor, and, if a factor is inapplicable to a

case, to explain why.” Arrow Fastener Co. v. Stanley Works, 59 F.3d 384, 400 (2d Cir.

1995) (reversing a bench trial judgment for other reasons).

31 Although many of these admonitions have arisen in our review of bench

trial proceedings, we have also noted in the summary judgment context that

“[t]his Court has repeatedly urged district courts to apply the Polaroid factors

even where a factor is irrelevant to the facts at hand.” Int’l Info. Sys. Sec.

Certification Consortium, Inc. v. Sec. Univ., LLC, 823 F.3d 153, 156 (2d Cir. 2016)

(internal quotation marks omitted) (vacating summary judgment for other

reasons); see also New Kayak Pool Corp. v. R & P Pools, Inc., 246 F.3d 183, 185 (2d

Cir. 2001) (vacating a preliminary injunction denial where the district court “did

not apply the Polaroid test” at all).

There is good reason for that. On appeal, we weigh the Polaroid factors de

novo based in large part upon the district court’s presentation and culling of the

record, albeit not in actual deference to its conclusions at the summary judgment

stage. See Tiffany, 971 F.3d at 85. Accordingly, where a district court has punted

on factors that it deems irrelevant for reasons that we cannot discern – or where it

has erroneously analyzed one factor and neglected to address others (relegating

those other factors, in all likelihood, to appellate afterthoughts) – the appeal “will

generally result in a costly and avoidable remand in order to elicit findings on the

other Polaroid factors” in the first instance. Natural Organics, 426 F.3d at 579-80; see

32 also Thompson Med. Co. v. Pfizer Inc., 753 F.2d 208, 214 (2d Cir. 1985) (“[T]he

complexities attendant to an accurate assessment of likelihood of confusion

require that the entire panoply of elements constituting the relevant factual

landscape be comprehensively examined.”). Courts and litigants have better

things to do with their time and resources. We thus reiterate that, as a general

matter, district courts should typically address all the Polaroid factors and, if it

deems one of the factors irrelevant, “explain why.” Arrow Fastener, 59 F.3d at 400.

That does not mean, however, that a district court’s judgment must

inevitably be vacated wherever it neglects to account for all eight Polaroid factors.

To be sure, such a failure is risky, and may well undermine our ability to conduct

adequate appellate review, thus necessitating a remand. In rare cases, though, the

weight of binding precedent may obviate the need for a complete Polaroid

analysis. See Natural Organics, 426 F.3d at 580 (“In an appropriate case, this Court

may undertake a full Polaroid balancing without the benefit of findings on each

Polaroid factor.”).

This is such a case. Here, the district court directly modeled its approach

after Electra. In that case, faced with effectively identical issues and facts, the

panel held that the “relevant” Polaroid factors “include[d], inter alia, the strength

33 of the mark, evidence of actual consumer confusion, and evidence that the mark

was adopted in bad faith.” 987 F.3d at 257. It then swiftly concluded that the

district court had “correctly dismissed Appellants’ Lanham Act claim,”

incorporating by reference the district court’s reasoning and holding that

“because the ultimate question under [Polaroid] is the likelihood of consumer

confusion, the district court properly analyzed Appellants’ recognizability.” Id. at

258. And although the district court in that case had itself addressed additional

factors, see Toth, 2019 WL 95564, at *6 (addressing six of the Polaroid factors), the

Electra panel made no mention of any Polaroid factor aside from strength of mark,

actual confusion, and bad faith. 987 F.3d at 258.

The district court in this case followed suit. Applying what it gleaned to be

Electra’s implicit lesson that, in this context, those three factors alone may be

dispositive, it granted summary judgment to Defendants based on its

determination that each of those factors weighed in Defendants’ favor. We cannot

fault a district court for its reasonable adherence to recent, directly-on-point,

binding precedent constructed upon substantially indistinguishable facts. On

appeal, Plaintiffs do not present, and we do not discern, any reason to suggest

that the other Polaroid factors not explored by the district court weighed more

34 strongly in their favor in this case than those same factors weighed in favor of the

Electra plaintiffs. The district court thus did not err in concluding that the same

balancing that led our colleagues to affirm summary judgment in Electra required

the same conclusion here.10

In any event, Plaintiffs do not urge us to vacate the district court’s grant of

summary judgment and to remand for fuller consideration on these (or any)

grounds; rather, they seek outright reversal of that judgment (or, in the

alternative, certification of a state law question to the New York Court of

Appeals, as discussed below). They have therefore waived any argument for

such a remand. See Norton v. Sam's Club, 145 F.3d 114, 117 (2d Cir. 1998) (“Issues

not sufficiently argued in the briefs are considered waived and normally will not

be addressed on appeal.”).

10 We underscore just how closely this case resembles Electra. We do not hold, for instance, that all false endorsement claims, or even all false endorsement claims involving roughly similar facts, may rise or fall on these three factors alone, nor that district courts should make a habit of assuming without deciding that seemingly peripheral Polaroid factors favor whichever party does not prevail on a given motion. This case simply presents exceptional circumstances that permit us to conclude that the factors deemed dispositive in Electra operate in an identical fashion in this effectively identical context.

35 Rather than argue that the district court erred procedurally, Plaintiffs insist

that the district court substantively “got the Polaroid balancing wrong.”

Appellants’ Br. 37. Even as they acknowledge that we have rejected a

“mechanical application of the Polaroid factors,” their argument amounts to little

more than highlighting the district court’s assumption (without deciding) that a

greater number of factors favored Plaintiffs than Defendants, and urging that this

should have been enough for their false endorsement claim to survive summary

judgment. Id. at 38.

But Plaintiffs’ initial concession swallows the argument it precedes: we

have indeed long instructed that “the evaluation of the Polaroid factors is not a

mechanical process where the party with the greatest number of factors weighing

in its favor wins.” RiseandShine Corp. v. PepsiCo, Inc., 41 F.4th 112, 124 (2d Cir.

2022), quoting Paddington Corp. v. Attiki Importers & Distributors, Inc., 996 F.2d 577,

584 (2d Cir. 1993). And even setting that point aside, Plaintiff’s argument is

foreclosed in this specific context by Electra, which held under effectively

identical circumstances that the same three factors were sufficient to definitively

tilt the Polaroid balance at the summary judgment stage. Electra, 987 F.3d at 257-

58; see also RiseandShine, 41 F.4th at 124 (“Weak marks are entitled to only an

36 extremely narrow scope of protection, unless a convincing combination of other

Polaroid factors militates strongly in favor of likelihood of confusion.” (internal

quotation marks omitted)).

We therefore affirm the district court’s grant of summary judgment in

Defendants’ favor as to Plaintiffs’ false endorsement claims.

III. False Advertising

The district court also correctly granted summary judgment to Defendants

on Plaintiffs’ false advertising claims.

use[] in commerce [of] any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact which . . . in commercial advertising or promotion, 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).

To prevail on a false advertising claim, a plaintiff must establish that the

message at issue is “(1) either literally or impliedly false, (2) material, (3) placed

in interstate commerce, and (4) the cause of actual or likely injury to the

37 plaintiff.” Church & Dwight Co. v. SPD Swiss Precision Diagnostics, GmBH, 843 F.3d

48, 65 (2d Cir. 2016). Only the injury prong – which was the sole basis for the

district court’s disposition of this claim – is at issue on appeal.

Looking to the Supreme Court’s decision in Lexmark International, Inc. v.

Static Control Components, Inc., 572 U.S. 118 (2014), the district court held that

Plaintiffs had established no evidence of any injury falling within the zone of

interests protected by the Lanham Act. In Lexmark, the Supreme Court recognized

that the “broad language” of 15 U.S.C. § 1125(a)(1) authorizing “any person who

believes that he or she is likely to be damaged” to sue “might suggest that an

action is available to anyone who can satisfy the minimum requirements of

Article III.” 572 U.S. at 129. Nonetheless, it rejected that expansive reading. Id.

Instead, it conditioned statutory protection upon two requirements: (1) that

a plaintiff’s injury fall within the “zone of interests” protected by the Lanham

Act, and (2) proximate causation. Id. To satisfy the first requirement, a plaintiff

must demonstrate injury specifically to a “commercial interest in reputation or

sales.” Id at 131-32. By contrast, plaintiffs injured in other ways – for example, a

“consumer who is hoodwinked into purchasing a disappointing product” or a

“business misled by a supplier into purchasing an inferior product” – may not

38 invoke the Lanham Act even as they “may well have an injury-in-fact cognizable

under Article III.” Id. at 132.

Turning to the proximate cause requirement, the Lexmark Court explained

that although in some sense “all commercial injuries from false advertising are

derivative of those suffered by consumers who are deceived by the advertising,”

liability under the Lanham Act is ordinarily limited to those who can “show

economic or reputational injury flowing directly from the deception wrought by

the defendant’s advertising.” Id. at 133 (emphasis added). “[T]hat occurs,” it

continued, “when deception of consumers causes them to withhold trade from

the plaintiff.” Id.

We have had little occasion to apply that guidance. Although we have

construed Lexmark’s broader teachings in other contexts, see, e.g., Am. Psychiatric

Ass’n v. Anthem Health Plans, Inc., 821 F.3d 352, 359 (2d Cir. 2016); Moya v. United

States Dep’t of Homeland Sec., 975 F.3d 120, 130 (2d Cir. 2020), we have applied

Lexmark in just a few cases actually involving the Lanham Act, and never in a

way that is particularly instructive here, cf. SM Kids, LLC v. Google LLC, 963 F.3d

206, 213-14 (2d Cir. 2020). As a result, we have never examined whether Lexmark

39 abrogates this Circuit’s relevant authority regarding actionable injury in a

Lanham Act false advertising claim.

Addressing that question now, we conclude that far from undermining our

precedent, Lexmark reinforces it. Lexmark is entirely consistent, for instance, with

our holding in Merck Eprova AG v. Gnosis S.p.A., 760 F.3d 247 (2d Cir. 2014),11 that

(1) a viable false advertising claim requires the plaintiff to have been “injured as a

result of the misrepresentation, either by direct diversion of sales or by a

lessening of goodwill associated with its products,” id. at 255 (internal quotation

marks omitted); and (2) although such injury may be “presumed” from a direct

competitor’s “false comparative advertising claim,” in all other cases, a plaintiff

must present some affirmative “indication of actual injury and causation,” id. at

259 (internal quotation marks omitted). All of that jibes with Lexmark, which

similarly requires plaintiffs to demonstrate “economic or reputational injury”

proximately caused by the alleged false advertisement, 572 U.S. at 133, and

similarly does not limit protection to direct competitors only, id. at 136. Nor,

11 Merck was argued well before Lexmark but decided a few months after, and without reference to, Lexmark. No Rule 28(j) letters were filed in Merck after Lexmark was decided. See generally Merck Eprova AG v. Gnosis S.p.A., Nos. 12-4218-cv, 13-513-cv (2d Cir.). We thus think it prudent to consider Merck as we would a case decided prior to Lexmark.

40 importantly, does it foreclose courts from granting a presumption of injury to

direct competitors while requiring others to present evidence of injury and

causation. See id. at 138-39; Merck, 760 F.3d at 259.12

We are thus bound by Lexmark and Merck alike. And the upshot of that

binding authority as applied to this case is that if Plaintiffs are in direct

competition with Defendants, and if Defendants’ false advertising implicated

Plaintiffs in some way, then injury and proximate cause are presumed. If not,

both must be affirmatively shown. Here, there is no evidence that Plaintiffs –

professional models who have brought this lawsuit precisely because they object

to the suggestion that they are even associated with Defendants’ marketplace –

directly compete with Defendants. Unsurprisingly, then, Plaintiffs concede that

“perhaps” they do not directly compete with Defendants. Appellants’ Br. 41.

Instead, they insist that they have affirmatively established injury cognizable

under the Lanham Act and proximate causation.

12 See also ThermoLife Int’l, LLC v. BPI Sports, LLC, No. 21-15339, 2022 WL 612669, at *2 (9th Cir. Mar. 2, 2022) (unpublished disposition) (“[W]e have generally held that when a plaintiff competes directly with a defendant, a misrepresentation will give rise to a presumed commercial injury that is sufficient to establish standing.” (internal quotation marks and alterations omitted)); cf. 3B Med., Inc. v. SoClean, Inc., 857 F. App’x. 28, 29 (2d Cir. 2021) (summary order) (applying both Lexmark and Merck to a Lanham Act false advertising claim).

41 They are mistaken. Plaintiffs claim two injuries: (1) that they may have lost

out on work opportunities due to the reputational hit from being linked with a

“gentlemen’s club”; and (2) that they were deprived of the revenue they would

typically expect to have received directly from Defendants for an authorized use

of their images. Both theories miss the mark.

The first purported injury type would likely satisfy Lexmark’s

requirements, if only there were any evidence that such an injury actually

occurred in this case. In Plaintiffs’ own words, their “uncontradicted testimony is

that their association with a strip club was potentially devastating to their

careers.” Appellants’ Br. 42 (emphasis added). That may well be possible, but

there is no evidence that anything of the sort actually happened. Plaintiffs

concede that, as far as they know, no third party has ever “refus[ed] to do

business or [rescinded] an offer to hire due to Defendant’s use of Plaintiff’s

image.” E,g., J.A. 329. And even if it is true, as Plaintiffs aver, that this ignorance

is to some degree attributable to the customary industry practice not to tell a

model why they did not receive a job offer, Plaintiffs have made no attempt to

present other evidence conceivably available to people in their position. For

example, they admit that there is nothing in the record to suggest that anyone

42 who might have been expected to hire Plaintiffs ever saw the posts in question, or

was likely to see the posts, or ever mentioned the posts. There is no temporal

evidence correlating downturns in Plaintiffs’ careers with the appearance of the

posts. There is no expert opinion testimony, let alone expert empirical analysis,

illustrating the effect of this kind of R-rated association on a typical model’s

career – much less on these particular models’ careers. There is, in short, nothing

that could permit a reasonable juror to find that the posts proximately caused

actual or likely “economic or reputational” injury here. Lexmark, 572 U.S. at 133.

To the contrary, Plaintiffs conceded at oral argument that their only

evidence of actual injury is tied to their claim that each Plaintiff “lost the income

she should have been paid had Defendants operated through legal channels and

paid her for her appearance in their advertisements.” J.A. 1922; see, e.g., id. 68-69.

Unfortunately for Plaintiffs, this second injury type – which sounds in trademark

infringement and in theft of services – fails to check any of Lexmark’s boxes. It

does not constitute “reputational” injury, nor does it flow “from the deception

wrought by the defendant’s advertising,” nor is there any reason to believe that it

would cause consumers (even assuming that term encompasses purchasers of

43 modeling services) “to withhold trade from the plaintiff.”13 Lexmark, 572 U.S. at

133 (emphasis added); see Merck, 760 F.3d at 255 (plaintiffs must show they were

“injured as a result of the misrepresentation,” and that the misrepresentation

concerned “an inherent quality or characteristic of the product” (internal

quotation marks omitted)). It is, in short, not the kind of injury that can sustain a

false advertising claim under the Lanham Act.

The district court was therefore correct to grant summary judgment to

Defendants on Plaintiffs’ false advertising claims.

13 Although the resulting authority is not binding on us, we note that Plaintiffs’ counsel has unsuccessfully made these same arguments in some of the other similar lawsuits it has filed across the country on behalf of similar groups of plaintiffs. See, e.g., Gray, 2020 WL 6200165, at *9 (granting summary judgment to defendants on plaintiffs’ false advertising claim because, among other reasons, the Lanham Act affords plaintiffs no right “to receive fair market value for use of their Images”); Pepaj, 2021 WL 632623, at *11 (same). Some courts, however, have adopted the argument. See Order at 21-22, Edmondson v. Velvet Lifestyles, LLC, No. 15-CV-24442 (S.D. Fla. July 28, 2017), Dkt. No. 174 (granting summary judgment on false advertising claim to professional models whose images were used by a swingers club because it was “undisputed that Plaintiffs are entitled be compensated for at least the fair market value for the use of their photos”), rev’d on other grounds, 43 F.4th 1153 (11th Cir. 2022); Order at 4, Underwood v. KAB Business Holdings, Inc., No. 5:20-CV-00881 (E.D.P.A. Mar. 31, 2021), Dkt. No. 41 (granting summary judgment to professional models on false advertising claim in relevant part because the plaintiffs “have lost the royalties or endorsement income they would have received if [the defendant] had paid them for the use of their images”).

44 IV. Right of Publicity

Finally, the district court correctly determined the majority of Plaintiffs’

right of publicity claims to be time-barred, and permissibly declined to exercise

supplemental jurisdiction over the remaining timely claims.

A. Statute of Limitations

Sections 50 and 51 of the New York Civil Rights Law prohibit the “(i) usage

of plaintiff’s name, portrait, picture, or voice, (ii) within the state of New York,

(iii) for purposes of advertising or trade, (iv) without plaintiff’s written consent.”

Electra, 987 F.3d at 249, quoting Molina v. Phx. Sound Inc., 747 N.Y.S.2d 227, 230

(1st Dep’t 2002); see N.Y. Civ. Rights Law §§ 50-51.

The district court held that these claims are subject to New York’s one-year

statute of limitations for any “violation of the right of privacy under section

fifty-one of the civil rights law,” N.Y. C.P.L.R. § 215(3). And because New York

courts apply a “single publication rule” for claims subject to § 215(3), whereby a

cause of action “accrues on the date the offending material is first published,”

Nussenzweig v. diCorcia, 9 N.Y.3d 184, 188 (2007); see Richardson v. Proctor &

Gamble Co., 176 N.Y.S.3d 605, 607 (1st Dep’t 2022), and because all but two of the

offending posts were published more than a year before the Complaint was filed,

45 the district court granted summary judgment with respect to the lion’s share of

Plaintiffs’ §§ 50-51 claims.

The only portion of this holding that Plaintiffs challenge on appeal is the

applicability of the one-year statute of limitations itself. Urging us either to

reverse or, in the alternative, to certify the question to the New York Court of

Appeals, they argue that § 215(3) does not control here because it applies only to

privacy claims, not to publicity claims. We disagree.

Certainly, Plaintiffs are correct that those two rights are in some ways

conceptually distinct. In general, privacy rights protect “individuals who have

not placed themselves in the public eye . . . from the embarrassment of having

their faces plastered on billboards and cereal boxes without their permission.” Jim

Henson Prods., Inc. v. John T. Brady & Assocs., Inc., 867 F. Supp. 175, 188 (S.D.N.Y.

1994). The interests protected are personal: individuals’ “dignity and peace of

mind,” id., with damages “designed primarily to compensate for injury to

feelings,” Electra, 987 F.3d at 255 (internal quotation marks omitted). By contrast,

publicity rights generally protect the “commercial value that attaches to [the]

identities” of persons who do place themselves “in the public eye.” Jim Henson,

867 F. Supp. at 188. Publicity rights are, in that sense, property-like in nature. See

46 Lerman v. Flynt Distrib. Co., 745 F.2d 123, 134 (2d Cir. 1984) (”Because the plaintiff

must generally have developed a property interest with financial value in order

to prove that he suffered damages, the right [of publicity] is most frequently

invoked by public figures or celebrities.”). Accordingly, “its infringement is a

commercial, rather than a personal tort,” with damages calibrated in terms of

commercial harm. Jim Henson, 867 F. Supp. at 188.

But none of that has any bearing on the applicable statute of limitations in

this case. To the extent that New York law recognizes a right of publicity, that

right is “encompassed” under the state’s statutory right of privacy; it has no other

source. See Stephano v. News Grp. Publications, Inc., 64 N.Y.2d 174, 183 (1984)

(“[T]he ‘right of publicity’ is encompassed under the Civil Rights Law as an

aspect of the right of privacy, which, as noted, is exclusively statutory in this

State . . . “); Darden v. OneUnited Bank, 128 N.Y.S.3d 640, 642 (2d Dep’t 2020)

(“[T]here is no common-law right of publicity [under New York law].”).

Consequently, there is no basis for Plaintiffs’ attempt to construe the language of

§ 215(3), which sets forth a one-year limitations period for any “violation of the

right of privacy under section fifty-one of the civil rights law,” as somehow

applicable only to the encompassing right of privacy and not the encompassed

47 right of publicity. However one may conceptualize different aspects of the right

protected by §§ 50-51, New York law provides a cause of action only for acts that

fall within the singular statutory definition set forth in those provisions, and

supplies a specific statute of limitations keyed to that cause of action. See N.Y.

C.P.L.R. § 215(3). Unsurprisingly, New York courts have applied the one-year

statute of limitations to §§ 50-51 claims that, like the claims in this case, could be

classified as right of publicity claims. See, e.g., Richardson, 176 N.Y.S.3d at 607;

Sirico v. F.G.G. Prods., Inc., 896 N.Y.S.2d 61, 66-67 (1st Dep’t 2010); see also Brooks

ex rel. Est. of Bell v. The Topps Co., No. 06-CV-2359-DLC, 2007 WL 4547585, at *3

(S.D.N.Y. Dec. 21, 2007); Fischer v. Forrest, No. 14-CV-1304-PAE-AJP, 2017 WL

1063464, at *5 (S.D.N.Y. Mar. 21, 2017).

For their part, Plaintiffs cite no New York authority applying, in the statute

of limitations context, the privacy/publicity distinction at the heart of their

argument. Nor do they identify any New York authority actually applying to any

claim brought under any aspect of §§ 50-51 some limitations period other than

the one-year window applicable to such claims under § 215(3). Because § 215(3) is

unambiguous – and the case law applying § 215(3) unanimous – on this point,

there is no need to certify the question to the New York Court of Appeals. And

48 indeed we have already endorsed this very conclusion, albeit in what is arguably

dictum.14 We reiterate that conclusion today, and for that reason, we hold that the

14 In its account of the procedural history of the case, the Electra panel remarked that the district court had “correctly” applied the one-year limitations period. 987 F.3d at 240. However, the panel provided no explanation for that conclusion and never returned to the subject in its analysis of the issues presented in the case. That is not surprising: the plaintiffs in Electra did not dispute that the one-year limitations period applied, so the matter was not actually at issue there. See Appellees’ Br. 31 n.14, Electra, 987 F.3d 233 (No. 19-235) (noting plaintiffs’ failure to contest the issue and arguing that the plaintiffs “thus concede the point”); Appellants’ Br. 41-46, Electra, 987 F.3d 233 (No. 19-235) (not addressing the issue); Appellants’ Reply Br. 28-31, Electra, 987 F.3d 233 (No. 19-235) (same). Plaintiffs here do not directly attack the Electra panel’s treatment of the issue, but instead seize upon a post-Electra change in New York law. See N.Y. Civ. Rights Law § 50-f(3) (effective May 29, 2021) (providing that “[t]he rights recognized under this section are property rights, freely transferable or descendible . . . .”). We are not convinced. Aside from having no direct bearing on any statutory limitations period, the new provision codified at §50-f in relevant part simply echoes the longstanding principle that the limited publicity right already encompassed in §§ 50-51 implicates property interests (while also expanding the scope of that right in other ways). See Lerman, 745 F.2d at 134; Brinkley v. Casablancas, 438 N.Y.S.2d 1004, 1012 (1st Dep’t 1981) (“The damages that flow from [a violation of §§ 50-51] should be compensable whether the injury is to one’s feelings or to his ‘property’ interest. Both injuries are caused by the same wrong and should be redressed by the same cause of action.”). Since § 50-f was enacted, New York courts have continued to apply a one-year statute of limitations to §§ 50-51 claims that could be classified as right of publicity claims, see, e.g., Richardson, 176 N.Y.S.3d at 607 (affirming summary judgment on timeliness grounds of §§ 50-51 claims brought by a model who claimed her image had been used to market hair products without her authorization), as has at least one federal court applying New York law, see Edmondson, 2021 WL 4499031, at *4 (denying motion to reconsider dismissal as untimely of §§ 50-51 claims brought by a group of models that included many of the Plaintiffs in this case).

49 district court correctly concluded that all but a few of Plaintiffs’ right of publicity

claims were time-barred.

B. Supplemental Jurisdiction

After dismissing most of Plaintiffs’ state law claims as untimely, the

district court declined to exercise supplemental jurisdiction over the remaining,

timely, §§ 50-51 claims. It is undisputed that, as a general matter, “after properly

granting summary judgment on the [federal] claims, the District Court had

discretion not to exercise supplemental jurisdiction” over any remaining state

law claims. Boyd v. J.E. Robert Co., 765 F.3d 123, 126 (2d Cir. 2014). We note,

however, that it is somewhat unusual for a district court to exercise supplemental

jurisdiction over some state law claims to one party’s benefit (e.g., by disposing of

them on timeliness grounds) while declining to exercise supplemental

jurisdiction over other similar (but timely) state law claims to the other party’s

detriment. But Plaintiffs do not argue that the district court abused its discretion

in doing so, and indeed expressly disavowed any such contention at oral

argument. Accordingly, any challenge they might have raised with respect to the

district court’s decision not to exercise supplemental jurisdiction over their

remaining timely claims is waived. See Norton, 145 F.3d at 117.

50 CONCLUSION

We have considered Plaintiffs’ other arguments and conclude that they are

without merit. Thus, for the foregoing reasons, we AFFIRM the judgment of the

district court.