Reebok International Ltd. v. K-Mart Corp.

849 F. Supp. 252 (1994)

REEBOK INTERNATIONAL LTD. and Reebok International Limited, Plaintiffs,
v.
K-MART CORPORATION; Melville Corporation; and Leif J. Ostberg, Inc., Defendants.

No. 92 Civ. 8871 (CHT).

United States District Court, S.D. New York.

April 15, 1994.

Lewin & Laytin, New York City (Harley Lewin, William Reid, Eric Weinstein, of counsel), for plaintiffs Reebok Intern. Ltd. and Reebok Intern. Ltd.

Brumbaugh, Graves, Donohue & Raymond, New York City (Joseph Garon, Richard Clark, Parker Bagley, of counsel), for defendants K-Mart Corp. and Melville Corp.

Kensington & Ressler, P.C., New York City (Michael Venditto, Helen Williamson, of counsel), for defendant Leif J. Ostberg, Inc.

OPINION AND ORDER

TENNEY, District Judge.

Plaintiffs Reebok International Ltd. ("Reebok USA") and Reebok International Limited ("Reebok UK") (collectively "Reebok"), bring this action against defendants K-Mart Corporation ("K-Mart"), Melville Corporation ("Melville") and Leif J. Ostberg Corporation ("LJO"), alleging that K-Mart stores have sold a sneaker that infringes on Reebok trademarks and trade dress. Plaintiffs' complaint alleges: (1) trademark infringement under Lanham Act § 32(1), 15 U.S.C. § 1114(1), (2) false designation of origin and unfair competition under Lanham Act § 43(a), 15 U.S.C. § 1125(a), (3) trademark counterfeiting under Lanham Act § 32(1), 15 U.S.C. § 1114(1), and (4) unfair competition and dilution of a trademark under N.Y.Gen. Bus.Law § 368-d and New York common law. Reebok seeks treble damages, injunctive relief, destruction of any shoes remaining in inventory, and attorneys' fees and costs.

Defendants deny the allegations of the complaint and seek attorneys' fees. K-Mart and Melville also bring a counterclaim, seeking cancellation of Reebok's federally registered trademark in the "Stripecheck" side design. K-Mart and Melville base their counterclaim on allegedly false affidavit statements made by Reebok to the Patent and Trademark Office (the "PTO") in 1981 in connection with its initial trademark application and in 1988 in connection with its application for incontestible trademark status.

Judge Patterson presided over discovery and other pre-trial matters before transferring the case to these chambers on November 16, 1993. I ruled on certain motions in limine on December 3. A four-day bench trial lasted from December 6 to December 9, 1993. The parties submitted substantial pre-trial and post-trial briefs. After reviewing the evidence presented at trial, I find for defendants on all of plaintiffs' claims. I also find that K-Mart and Melville have failed to satisfy the burden of proving their counterclaim.

BACKGROUND

A. The Parties and Their Products

Reebok USA, a Massachusetts corporation, designs, manufactures and sells shoes and sportswear. Reebok UK, organized as a limited company under the laws of the United Kingdom and a wholly-owned subsidiary of Reebok USA, also designs, manufactures and sells shoes and sportswear. Among Reebok's all-time best selling shoe models are the "Ex-O-Fit," a men's sneaker first sold in the United States in 1983, and the "Freestyle," a women's sneaker first sold in the United States in 1982. Trial Transcript ("Tr.") 65. Both of these sneakers are sold in high-top and low-top versions, and are popular among aerobics enthusiasts, gym goers and the general consumer. The Ex-O-Fit and Freestyle benefitted greatly from the aerobics boom of the 1980's, selling 3.03 million and 9.56 million pairs in 1986, respectively. Pl. exh. 169. These sales decreased drastically in the next several years. By 1990, Ex-O-Fit sales had slipped to 1.50 million pairs, and Freestyle sales fell to 2.49 million pairs. Id.^[1]

The external design of the Ex-O-Fit differs from that of the Freestyle. Both shoes incorporate a side design based on Reebok's Stripecheck trademark. Reebok UK owns a registered federal trademark for miscellaneous design for the Stripecheck, Reg. No. 1,196,293. Reebok USA retains the exclusive license for use of the Stripecheck in the United States. The bottom part of the two-part Stripecheck design consists of a curve starting from the mid-sole of the shoe, extending upwards and rearward to the upper part of the heel, narrowing as it extends backwards. This "stripe" is then overlaid with a diagonal "check" extending from the eyelets of the shoe towards the heel of the shoe, narrowing as it continues rearward and downward. The shape of the horizontal and vertical elements of the side designs of the Ex-O-Fit and Freestyle, and their proportions, differ from the registered Stripecheck design.

Reebok claims that the Titan model 17112 shoe ("the Titan") infringes the Stripecheck trademark and the trade dress of the Ex-O-Fit and Freestyle.^[2] The Titan is an inexpensive all-black high-top sneaker, sold exclusively at K-Mart stores through the Meldisco subsidiary of the Melville Corporation. An arrangement between K-Mart and Melville grants Meldisco the exclusive right to operate the footwear business in K-Mart stores. Pl. exh. 173 at 21. Melville, a New York corporation, owns 51% of Meldisco, while K-Mart, a Michigan corporation, owns the remaining 49%. Meldisco leases floorspace in K-Mart's stores, which is then used as the sales space for the footwear operations. Id. Meldisco had roughly $1.2 billion in sales in 1992, $140 million of which came from men's athletic shoes. Tr. 344.

By 1990, all-black high-top sneakers had become increasingly popular with consumers. Nearly all of the leading athletic shoe companies had introduced, or had plans to introduce, an all-black high-top. Recognizing the popularity of all-black high-top sneakers, Meldisco followed the trend and commissioned its own all-black high top. After considering working with other companies, Meldisco decided to work with defendant LJO, a footwear buying agent and New Jersey corporation.

In late 1990, Meldisco collaborated with LJO and designed the all-black Titan high-top. Def. exh. BQ at 19-22; Pl. exh. 173 at 117. The Meldisco and LJO personnel responsible for the Titan design knew of the existence of the Freestyle, the Ex-O-Fit and the Stripecheck design. Def. exh. BQ at 25-26. Meldisco used LJO as the buying agent for an initial order for 29,700 pairs of Titans in 1991. Tr. 413; Def. exh. BQ at 47. Meldisco later decided to cut out the middlemen at LJO and worked directly with a Taiwanese factory to order an additional 65,700 pairs. Pl. exh. 173 at 83.

Meldisco began selling the Titan in 1991. It sold almost all, if not all, of the shoes by the time of trial. Through November 10, 1993, Meldisco had sold over 90,500 pairs of the Titan, generating revenues of approximately $1.3 million. Tr. 346. Meldisco sold the Titan at $16.99 a pair through September 1992, after which it marked down the shoes to $14.99. Pl. exh. 173 at 109. In August 1993, Meldisco marked down the shoes remaining in inventory to $9.99. Id. Meldisco no longer imports, manufactures or sells the Titan 17112 or any shoes under the Titan name.

Defendants acknowledged that the Titan incorporates lower quality materials and construction methods than the Ex-O-Fit or Freestyle. Joint Pre-Trial Order ("JPTO") Undisputed Facts ¶ bb. For example, Reebok constructed the Ex-O-Fit and Freestyle with a garment leather upper, a more durable, comfortable and breathable material than the vinyl used in the Titan. The respective prices of the shoes reflect the difference in quality. While the Titan retailed at between $9.99 and $16.99, the retail prices of the Ex-O-Fit and Freestyle ranged from $49.00 to $65.00.

The sales outlets for the shoes also reflect the disparity in the quality and price of the sneakers. Reebok retailed the Ex-O-Fit and Freestyle through sporting goods stores, footwear specialty stores and department stores. The Titan, in contrast, was sold exclusively at 800 of K-Mart's 2400 discount stores.

The footwear selection at K-Mart stores consists almost entirely of inexpensive brands and "private labels." Private label shoes, such as the Titan, are designed exclusively for sale in K-Mart stores. Meldisco has used the Titan name as a private label brand for several years, generating $93 million in sales from 1990 through 1993.^[3] Def. exh. F.

K-Mart stores have carried premium shoe brands such as Nike, Adidas and Reebok only on rare occasions. Reebok does not sell its sneakers to warehouse clubs and discount chains such as K-Mart. Tr. 92. Meldisco occasionally made an "opportunistic buy" of "A-brand" shoes such as Nike, Adidas, or Reebok from third parties. Less than 1% of Meldisco sales of athletic shoes, and 1/10th of 1% of total Meldisco sales, came from these "A-brands," and Meldisco has deemphasized the practice since 1990. Pl. exh. 173 at 127.

The evidence established that K-Mart stores have sold only very small numbers of Reebok shoes, and that Reebok shoes were never sold at any of the same K-Mart stores as the Titan. From 1989 through 1993, K-Mart stores only carried one Reebok model, a white court shoe that was sold in 1991. Tr. 381, 387. During the rare occasions that Meldisco made an opportunistic buy of Reeboks, it sold the shoes in the 75 to 200 K-Mart stores with the highest economic demographics. Meldisco sold the Titan in 800 K-Mart stores with lower economic demographics. Tr. 414-15. In addition, Reebok presented no evidence that Meldisco purchased any Freestyles or Ex-O-Fits since 1984.

B. Procedural History

Reebok first became aware of the Titan in November 1992. It is unclear whether Reebok discovered the Titan through its own policing efforts or from an inquiry by the United States Customs Service ("Customs"). In November and December of 1992, Customs examined whether importation of the Titan violated U.S. law. Reebok filed a complaint in this court on December 8, 1992, later amended on April 23, 1993.

The Customs examination initially determined that the Titan side design and the Stripecheck were confusingly similar, and that future import shipments of the Titan should be detained. In February of 1993, Customs rescinded the initial determination of infringement after meeting with counsel for Meldisco. Reebok's counsel met with Customs officials in May of 1993 to request reconsideration of the decision to rescind the finding of infringement.^[4]

In August of 1993, Customs made another finding, this time ruling that the Titan infringed on the Stripecheck design. Defendants moved that the district court stay further Customs proceedings related to this matter. Judge Patterson denied the motion to stay. His order noted that since defendants no longer imported the Titan, it was "not necessary for them to participate in the Customs proceeding, eliminating any potential duplication of their efforts."

At trial and in its briefs, Reebok claimed that the Customs proceeding eased its evidentiary burden in this civil suit. See Plaintiff's Post-Trial Brief at 41-42 & n. 171. The court has carefully read the various Customs decisions in this matter and has considered their findings. However, the Customs proceedings do not alter the evidentiary burdens of the parties in the district court.

C. The Stripecheck Trademark

Reebok UK filed an initial trademark application for the Stripecheck in November of 1979, claiming first use of the mark "in commerce with the United States at least as early as November 1975." The PTO refused registration in an Office Action dated December 10, 1980, ruling that the design was ornamental and not indicative of the origin of the goods. Reebok later filed a § 2(f) declaration that the mark had become distinctive through exclusive and continuous use for at least five years from the November 1979 date of application. The PTO approved the application and registered the mark under 15 U.S.C. § 1052(f) in 1982. Reebok applied for incontestible status for the Stripecheck in 1988. The PTO granted the application on August 29, 1988.

At trial, Reebok failed to introduce persuasive evidence that the Stripecheck trademark had been in "continuous use" since November 1974. Reebok was virtually unknown in the United States until 1979. Def. exh. P. Reebok had some sales in the United States in the early 1970's, but the evidence failed to demonstrate conclusively that any Reebok shoes sold in the United States during this period bore the Stripecheck mark. Tr. 80, 103. Reebok produced a witness who testified that he purchased a Reebok shoe bearing the Stripecheck mark in 1971 or 1972. However, this witness did not testify to continuous use between 1972 and 1975,^[5] nor did Reebok produce any documentary evidence indicating use of the Stripecheck mark during this period. In addition, defendants introduced documentary evidence that the shoe purchased by Reebok's witness was "reintroduced" in 1976, indicating that there was some gap in continuous use between 1972 to 1976. Def. exh. BB4.

DISCUSSION

I. Validity of the Marks

A. The Stripecheck Trademark

Reebok's Stripecheck trademark, although having been granted incontestible status by the PTO, is not immune from attack. An incontestible trademark may still be challenged on the basis of fraud. 15 U.S.C. § 1115(b)(1). A trademark applicant assumes the obligation to make any statements to the PTO with "uncompromising candor." Orient Express Trading Co. v. Federated Dep't Stores, Inc., 842 F.2d 650, 653 (2d Cir.1988). Defendants claim that Reebok's § 2(f) declaration, filed on November 9, 1981, and Reebok's § 15 affidavit, filed on May 19, 1988, included fraudulent statements. Fraud requires a "deliberate attempt to mislead the PTO," and is not established merely through a showing of "mere error or inadvertence." Id. A party seeking cancellation of a registered trademark through an allegation of fraud must demonstrate fraud by clear and convincing evidence. Id.

Defendants failed to introduce any evidence of a deliberate attempt to mislead the PTO with the 1981 § 2(f) affidavit. Although Reebok's evidence at trial failed to substantiate the statements made in its 1981 affidavit, this was not Reebok's obligation. Defendants bore the burden of proving that Reebok deliberately attempted to mislead the PTO.

The 1988 § 15 affidavit presents a closer question. This affidavit stated that the validity or ownership of the Stripecheck was not at issue in any pending legal proceeding. It appears that at the time of the § 15 affidavit, Reebok was a plaintiff in Reebok v. Alon, et al., No. CV 87-06466, 1987 WL 123992 (CBM) (C.D.Cal.), a trademark infringement action then pending in federal court, in which the defendants' answer denied Reebok's ownership interest in the Stripecheck. Def. exhs. BC1, BC2, BD. However, the Alon defendants raised neither a counterclaim nor an affirmative defense disputing the validity of the mark. The Alon suit was not resolved until March 31, 1989.

The court does not believe that the fact that the Alon suit was unresolved at the time of the § 15 declaration necessarily made the declaration fraudulent. Reebok may well have failed to fulfill its duty of uncompromising candor by not calling the Alon suit to the attention of the PTO. However, the court cannot find that Reebok engaged in a deliberate attempt to mislead the PTO. Any error was more likely the result of mere error or inadvertence, or possibly the result of a good faith belief that Reebok did not have to call the Alon matter to the PTO's attention. See also, § 1604.03 of the Trademark Manual of Examining Procedures (2d ed. 1993) (§ 15 affidavit can be accepted by PTO even if registrant is plaintiff in suit that does not raise a counterclaim challenging validity of plaintiffs mark). The court concludes that the Stripecheck is a valid federally registered trademark.

B. Differences Between the Registered Stripecheck Mark and the Mark Used On the Freestyle and Ex-O-Fit

The side design of the shoes introduced at trial by Reebok differed from the side design registered by the PTO. "Slight and inconsequential" deviations in a registered trademark are to be expected when a two-dimensional registered mark is incorporated into the design of a three-dimensional item of merchandise. Montres Rolex, S.A. v. Snyder, 718 F.2d 524, 532 (2d Cir.1983), cert. denied, 465 U.S. 1100, 104 S. Ct. 1594, 80 L. Ed. 2d 126 (1984). Variations in the form of the mark are protected "as long as the new and old forms create the same, continuing commercial impression." Cullman Ventures, Inc. v. Columbian Art Works, Inc., 717 F. Supp. 96, 128 (S.D.N.Y.1989).

The differences between the Stripecheck as applied to the Ex-O-Fit and Freestyle exceed the "slight and inconsequential." However, they still convey a single commercial impression, and do not alter the distinctive characteristics of the mark. See Montres Rolex, 718 F.2d at 532; Ilco Corp. v. Ideal Security Hardware Corp., 527 F.2d 1221, 1224 (C.C.P.A.1976). As such, the differences between the marks represent a continuity of trademark rights, and merit protection under Reebok's trademark registration.

C. The Ex-O-Fit and Freestyle Trade Dress

Protectible trade dress may be created through the combination of geometric shapes, coloring, lettering or descriptive elements. Paddington Corp. v. Attiki Importers & Distrib., Inc., 996 F.2d 577, 584 (2d Cir.1993). The use of commonly used elements does not bar trade dress protection of the entirety of the trade dress. Id. The court's determination of distinctiveness focuses on the combination of elements and the total impression that the trade dress gives the observer. Id.

The most important elements of the Ex-O-Fit and Freestyle high-top^[6] trade dress are (in no particular order) the velcro straps around its ankles, the soft concentric circles of padding around the ankle, the Stripecheck mark, the soft leather upper, the monochromatic color, the Reebok logo/Union Jack labels on the side of the shoe, and the Starcrest label on the tongue of the shoe. Several of these features appear to be functional. However, the combination and arrangement of otherwise unprotectible functional features can create protectible trade dress. LeSportsac, Inc. v. K-Mart Corp., 754 F.2d 71, 76 (2d Cir.1985).

The evidence indicated that the respective trade dresses of the Freestyle and Ex-O-Fit have acquired secondary meaning and are therefore protectible. Among the factors examined in determining secondary meaning are sales success, advertising expenditures, unsolicited media coverage, attempts by others to copy the trade dress, and previous successful enforcement actions. Despite a substantial decline in sales over the last five years, the Freestyle and Ex-O-Fit continue to enjoy significant sales success. Reebok has succeeded in previous enforcement actions against several third parties who copied the Ex-O-Fit and Freestyle trade dress. Reebok failed to present credible evidence of advertising expenditures for the Ex-O-Fit and Freestyle. Reebok conceded that the Ex-O-Fit and Freestyle were not part of Reebok's major paid-for advertising campaigns between 1988 and 1993, Tr. 205-06, although the shoes appeared in at least two department store catalogs during this period. Reebok submitted no documentary evidence of unsolicited media coverage in which the Ex-O-Fit or Freestyle are visible.

Analyzing these factors demonstrates that the Freestyle and Ex-O-Fit continue to be popular sneakers whose respective trade dress have acquired secondary meaning among consumers.

II. Factors Relevant to Analyzing the Likelihood of Confusion

Ownership of an incontestible mark does not relieve a trademark owner from the requirement of proving likelihood of confusion. Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d 1072, 1077-78 (2d Cir.1993). Determination of the likelihood of confusion revolves around analysis of the factors set out in Polaroid Corp. v. Polarad Elec. Corp., 287 F.2d 492, 495 (2d Cir.), cert. denied 368 U.S. 820, 82 S. Ct. 36, 7 L. Ed. 2d 25 (1961) and developed in later caselaw. These factors are "evaluated in the context of how it bears on the ultimate question of likelihood of confusion as to the source of the product." Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 872 (2d Cir.1986).

A. Trademark Strength

The starting point of the Polaroid analysis is the strength of the trademark. The strength of a mark is the "distinctiveness of the mark, or more precisely, its tendency to identify the goods sold under the mark as emanating from a particular, although possibly anonymous, source." McGregor-Doniger, Inc. v. Drizzle, Inc., 599 F.2d 1126, 1131 (2d Cir.1979). The strength of the mark defines the scope of protection it is accorded. Id. In balancing the Polaroid factors, trademark strength is best conceptualized as a sliding scale, not as a binary categorization as either strong or weak. Stronger marks enjoy protection over a wider range of products and services, and over a wider range of variations in visual format, than weaker marks. McCarthy on Trademarks § 11.24[1]. Analysis of the strength of a mark requires examination of the degree to which the mark is inherently distinctive, and the degree to which the mark is distinctive in the marketplace. W.W.W. Pharmaceutical Co. v. Gillette Co., 984 F.2d 567, 572 (2d Cir.1993). The fact that a mark is registered establishes that a mark is strong, but does not relieve a court of the obligation of examining the precise strength of the mark in a specific context.

The evidence presented at trial established that the Stripecheck is a strong mark entitled to a relatively high level of protection. The design combines a horizontal and a vertical design element. As discussed supra, these two elements are each closely related to designs used by other shoe manufacturers to identify their sneakers. But by combining these two elements, Reebok has created a distinct, recognizable side design.

Accepting the Stripecheck as inherently distinctive and therefore strong, the court turns to the distinctiveness of the Stripecheck in the marketplace to determine its relative strength. See Paddington, 996 F.2d at 585 (courts may consider secondary meaning when analyzing the strength of an inherently distinctive mark). Secondary meaning is analyzed under "rigorous evidentiary requirements." 20th Century Wear, Inc. v. Sanmark-Stardust, Inc., 815 F.2d 8, 10 (2d Cir.1987).

Several factors cause the court to recognize limitations to the strength of the Stripecheck. First, Reebok has used the Stripecheck inconsistently, and has focused primarily on other marks in establishing its brand identity among consumers. Second, the Stripecheck design closely resembles other registered marks used by Reebok's competitors. And third, Reebok presented very little evidence regarding either the strength of the Stripecheck in the marketplace or the use of the Stripecheck in advertising.

The field of sneaker side designs is crowded with designs similar to the Stripecheck. Several other side designs, used by other companies to identify their sneakers, closely resemble the two elements of the Stripecheck design. For example, the "stripe" part of the Stripecheck is nearly identical to the Puma "Formstrip," and is very similar to the horizontal part of the side design used on Asics' shoes. The "check" part closely resembles trademarks used on Brooks, Bata and Mitre sneakers. Many of these third party uses preceded the introduction of the Stripecheck. The similarity of these designs is understandable. Side stripes that flow or that are angled to the rear of a shoe create a visual impression of movement, action and speed — all attributes associated with athletic activities. There are a limited number of shapes that designers can use to create this feeling of motion.

Third party use of similar designs does not by itself provide a ceiling on the strength of the mark. However, it can be a factor weighing against the strength of a mark, see W.W.W., 984 F.2d at 573 (extensive third party use of words "sport" and "stick" weighs against strength of "Sportstick" trademark), especially when the third party uses occur on competing products.

Defendants' expert on shoe marketing and design, Mr. Philip Nutt, testified that the Stripecheck design was not one of the stronger trademarks in the sneaker market.^[7] He attributed the Stripecheck's relative lack of strength to many factors, including the mark's similarity to other marks in a crowded field, Reebok's lack of brand discipline in its shoe designs and advertisements, and its decision to rely primarily on the printed Reebok logo to establish brand identity.

The evidence overwhelmingly supported Mr. Nutt's observation that Reebok lacks a uniform external visual identity among consumers. Reebok uses the Stripecheck on only some of its sneakers. Reebok lacks guidelines regarding which shoes do or do not carry the Stripecheck. Reebok's witnesses conceded that Reebok does not incorporate the Stripecheck into all of its shoes, and that the Stripecheck is used in a "variety of forms." Tr. 68-69. Reebok's vice-president for design felt that the Stripecheck restricted design options, and stated that Reebok designers responded to trends and changing fashions by freely departing from the registered Stripecheck design or leaving the mark off of shoes. Def. exh. BO at 51-53; Pl. exh. 176 at 112-13. As a result, many of the "versions" of the Stripecheck bear almost no resemblance to the design used on the Ex-O-Fit or Freestyle. Reebok's competitors, in contrast, use their trademark side designs on a much higher proportion of their sneakers, and with only minor variations in appearance.^[8] Reebok witnesses conceded that it did not have any standards for uniformity of application for the Stripecheck mark,^[9] Def. exh. BN at 34; Def. exh. BO at 43, even though its competitors applied their marks consistently. Def. exh. BR at 30-31.

Reebok's 1992 sneaker catalog demonstrates Reebok's lack of a consistent external visual identity. The shoes depicted in the catalog display a dizzying array of different side designs. Some of the shoes depicted in the catalog bore a close variation on the Stripecheck design. A larger number of shoes bore a side design that utilized only a portion of the Stripecheck, often only the "stripe" portion broken in some manner. Another large portion of the Reebok product line bore side designs with no relation whatsoever to the Stripecheck design. Reviewing these various designs shows that a very large majority of the sneakers in Reebok's 1992 product line utilize side designs that do not represent a continuity of trademark rights, or utilize side designs entirely unrelated to the Stripecheck. There did not appear to be any pattern or system governing which models carried close variants of the Stripecheck and which shoes carried distant variations or designs completely unrelated to the Stripecheck.

Reebok argued that the court should not look to Reebok's overall sneaker line in assessing its external brand identity. Instead, Reebok argued that the court should limit its analysis to the shoes categorized as "Classics," a product category of ten to fifteen shoe models established in 1991. According to Reebok, the Classics line had the brand discipline and reliance on the Stripecheck that the larger Reebok product line lacked.

The "Classics" argument was unpersuasive. Reebok presented no convincing evidence that the Classics designation had any significance to retail consumers.^[10] Reebok also failed to establish that it expended any appreciable advertising or consumer marketing resources on Classics before 1993. In addition, the Classics, like the rest of Reebok's product line, did not rely on the Stripecheck for brand identity. Despite Reebok's claims that it used the Stripecheck consistently across the Classic line of shoes, Tr. 69, Reebok's product catalog demonstrates that this was untrue. Several Classics incorporate side designs completely unrelated to the Stripecheck.^[11]

Reebok's failure to consistently focus on the Stripecheck is explained by the fact that Reebok has relied primarily on the Reebok logo, consisting of the Reebok name printed in its distinctive typeface and usually accompanied by the Union Jack flag, to establish an identity among consumers. Reebok's brand identity guideline manual for 1990 to early 1993^[12] demonstrates that the Stripecheck was only one of several secondary elements used to establish Reebok's identity among consumers. The "Reebok Brand Identity System" brochure is a glossy, pull-out series of detailed guidelines signed by the President of Reebok International Ltd. These guidelines repeatedly stress to Reebok employees that the "cornerstone of the [brand identity] system is the Reebok logo" and that this printed logo is the "central component" of Reebok's identity in the "minds and hearts of people worldwide." Although the Brand Identity Manual includes substantial specific guidelines for use of the Reebok logo and Union Jack design, there are no guidelines for use of the Stripecheck, nor any statements regarding its importance in identifying the Reebok brand to consumers. The brand identity manual mentions the Stripecheck only briefly. The manual classes the Stripecheck along with the Union Jack flag design and the Starcrest design as "among the most familiar" of "several symbols with which Reebok has long been associated."

The 1992 product catalog demonstrates the reliance on the Reebok logo to establish brand identity. Although only a portion of Reebok's shoes bore the Stripecheck, almost all of its shoes bore the Reebok logo on the side of the shoe, usually applied in a contrasting color. See Tr. 96 (The Reebok name "always" appears on its shoes). This contrasts with Nike, Asics and Puma, who include their respective names on the sides of only a small portion of their shoes. And the limited number of advertisements Reebok submitted as evidence all featured the Reebok logo and Union Jack design as their designation of origin and most prominent feature. One ad, for example, displays the Reebok logo in no fewer than eight places. Pl. exh. 36.

Reebok has apparently realized the fact that it has not presented a consistent external visual identity to the public. It has very recently changed its marketing strategy to focus upon a stylized variation of the Stripecheck mark called the "Performance logo." See Tr. 99 (Reebok has evolved its trademark in "the last several months to highlight the [Stripecheck] in a more consistent manner"); see Pl. exh. 176 at 45-46. This variation of the Stripecheck will be incorporated into the Reebok logo. A new 1993 version of Reebok's brand identity manual replaces the previous brand identity system, Tr. 99, and states that "The Reebok and Performance Logo replaces the Union Jack in all applications with the exception of the Reebok Classic Collection." Def. exh. H3.

The move to more standardized trademark usage occurred sometime in 1993, Def. exh. BO at 44-45, 63 (designers were implementing changes in August of 1993), and may lead to greater standardization and brand identification for the Stripecheck. However, it has little, if any, effect on the time period relevant to this case. Since the design of the Titan occurred in 1990, and almost all of the Titan sales occurred before the change in Reebok's marketing focus, the court believes that the new marketing efforts had only a negligible effect on the strength of the Stripecheck and the likelihood that consumers were confused regarding the origin of the Titan during the relevant time period.

Reebok introduced surprisingly little evidence regarding the strength of its mark. Reebok relied on an unconvincing expert, a Reebok advertising executive, and a handful of print ads and press clippings in order to establish the strength of its mark. Given the "rigorous evidentiary requirements" used to establish secondary meaning, see 20th Century Wear, 815 F.2d at 10, this evidence did little to establish the strength of the Stripecheck.

The evidence failed to convince the court that Reebok focused on the Stripecheck in its advertisements to establish brand identity. Reebok failed to provide any examples of ads used in its major advertising campaigns since 1987: the "Performance," "U.B.U.," "Pump," and "Life's Short, Play Hard" campaigns. Tr. 204-06. The Ex-O-Fit and Freestyle were not part of these advertising campaigns,^[13] and Reebok did not provide any examples of the Stripecheck being featured in these campaigns. Reebok produced no persuasive evidence of television ads that might have featured shoes with the Stripecheck. Reebok also contended that "promotion by athlete endorsements ... is essential to success in the athletic shoe industry," but failed to introduce any evidence regarding the extent to which Reebok athlete endorsers wore shoes bearing the Stripecheck. See JPTO, Pl.Contentions of Fact ¶ 23.

The only reliable advertising evidence came from a handful of print ads, published over several years, for shoes bearing the Stripecheck. These scattered print ads provided little guidance in establishing what portion of Reebok's $60-100 million annual advertising budget went to products bearing the Stripecheck. A Reebok advertising executive testified, but failed to provide any reliable testimony regarding the extent of Reebok's advertising of shoes bearing the Stripecheck during the relevant time period.^[14] Reebok also failed to introduce objective evidence regarding consumer perceptions of Reebok's brand identity in general, or of the Stripecheck in particular.^[15]

The court therefore concludes that the Reebok Stripecheck was a strong and inherently distinctive mark from 1990 through 1993. Many consumers would recognize the Stripecheck as an indication of origin on sneakers. However, consumers would by no means universally recognize the Stripecheck as a designation of origin. Only consumers with some familiarity with premium sneaker brands would be able to distinguish the Stripecheck from competing side designs in the marketplace. And a significant number of consumers, particularly those most familiar with premium sneaker brands, would expect to see the Reebok logo or Union Jack design on a pair of Reeboks. Given Reebok's lack of discipline in side designs, inconsistent use of the Stripecheck, and reliance on other logos, many consumers might even view the Stripecheck as merely a decorative flourish, and not as an indicator of origin.

Of course, a manufacturer need not use a mark throughout its product line to maintain its strength. Nor does a manufacturer's use of several different trademarks create any inherent limitation on the strength of any one of those marks. But in weighing the evidence produced by defendants and the very limited evidence produced by Reebok at trial regarding the strength of the Stripecheck in the marketplace, the court recognizes certain specific limitations on the strength of the mark.

B. Similarity of Marks

The Titan's side design generally resembles the side designs of the Ex-O-Fit and the Freestyle, but differs in some material respects. The Titan design is also somewhat similar to the registered Stripecheck design, although it bears less resemblance to the registered design than it does to the designs used on the Ex-O-Fit and Freestyle.

Each shoe has a horizontal stripe element overlaid by a vertical, angular element. The horizontal element of the Titan design utilizes parallel, virtually straight lines, rather than the curved and narrowing lines of the Stripecheck, Ex-O-Fit and Freestyle. The vertical element of the Titan is a thin bar, falling at a steep angle. This contrasts with the curved "V" shape of the vertical elements of the Stripecheck and the Ex-O-Fit, which fall at a much shallower angle than the vertical element of the Titan. The vertical element of the Titan is much closer to the thin vertical element of the Freestyle. The Titan bears the closest resemblance to the Freestyle side design. It bears less of a resemblance to the Ex-O-Fit design, and the least resemblance to the Stripecheck as depicted on the trademark registration.

Given the crowded field of sneaker side designs, the court pays closer attention to the differences between the Titan and Reebok designs than it might in a case involving different products. The similarity of the side designs used by shoe manufacturers indicate that there are a limited number of angled shapes that designers can use to create a feeling of motion. The court is therefore careful to limit the range of variations protected by a valid trademark.

The Titan and Reebok side designs are far from identical. The material differences in their respective side designs, when viewed in their entirety and in context, reduce the likelihood of confusion by a small amount. See Gruner + Jahr, 991 F.2d at 1078 (dissimilarity of logos can lessen the likelihood of confusion). Nonetheless, the marks are still generally similar, and are closely enough related to create at least the potential for confusion.

C. Similarity of Trade Dress

Similarity of trade dress is related to some of the other Polaroid factors. For the sake of clarity, the court will discuss this factor separately.

The Titan, Ex-O-Fit and Freestyle are each monochromatic high-top sneakers. There are several material differences between the shoes. The Titan lacks the velcro ankle straps of the Reebok models, a distinctive design feature incorporated into many of Reebok's sneakers. The Reebok shoes also incorporate distinctive concentric circles of padding around their ankles. Although the Titan has ankle padding, it is not arranged in concentric circles. The Titan is also entirely black, in contrast with the Ex-O-Fit and Freestyle, which have contrasting white labels with the Reebok, Union Jack and Starcrest designs applied throughout the shoes. The shape and proportions of the shoes also differ. Other, less important differences include differences in the position of airholes, stitching, and other cosmetic features.

At the point of sale, the packaging of the products differed. See W.W.W., 984 F.2d at 573 (package design relevant to the similarity of the products). Reebok sold the vast majority of its shoe models in a distinctive blue box emblazoned with the Union Jack flag design covering the entire lid of the shoe box. Def. exh. BV at 28. Meldisco sold the Titan in a plain white box, although it sold some Titans in open boxes. The difference in packaging would only be relevant to consumers who had previously purchased Reeboks or who had seen Reeboks sold in shoe stores.

The differences between the shoes, particularly the Titan's lack of Reebok's distinctive velcro ankle straps and concentric-circle ankle padding, reduces the likelihood of confusion of trade dress by a significant amount. However, these differences are not sufficient to completely dismiss the possibility of consumer confusion.

D. Similarity of Use of the Products

Each of the shoes enjoy nearly identical uses. Consumers use all three sneakers primarily for casual use. Consumers might also put the shoes to athletic uses.

E. Quality of Products

The construction and materials of the Titan are inferior to the materials and construction of the Reebok models. Rigorous athletic use of the Titan would exaggerate these differences. Reebok established the inferiority of the materials and construction of the Titan, but did not conduct any specific tests on the Titan. The court concludes that the Titan would wear out more quickly than the Reebok models, and would be hotter, less comfortable and provide less support to the wearer. Without specific tests, the court cannot conclude with any reasonable certainty what the effect of these differences would be on the exterior appearance of the shoe or how quickly these differences would become noticeable.

F. Marketing Channels

Meldisco sold the Titan through marketing channels completely dissimilar from those used by Reebok in selling the Ex-O-Fit, Freestyle or any of its shoes. Meldisco sold the Titan exclusively at K-Mart stores, with a retail price from $9.99 to $16.99. Reebok sells its shoes in sporting goods stores, footwear specialty stores and department stores. The Ex-O-Fit and Freestyle sold at a retail price of between $45.00 and $65.00. Other Reebok models vary in retail price from $29.90 to over $100.

The substantial price difference between the Titan and Reebok's shoes weighs strongly against a finding of confusion. See McGregor-Doniger, 599 F.2d at 1135 (court looks to the manner of display or sale, place where goods are sold, and class of customers for whom product is designed and sold); Speedry Prods. Inc. v. Dri Mark Prods., Inc., 271 F.2d 646, 651 (2d Cir.1959). Most consumers who are sufficiently familiar with shoe side designs to recognize the Stripecheck as a Reebok mark would also realize that Reebok sneakers sell for much more than $9.99 to $16.99.

The fact that the shoes were sold in different types of stores also weighs heavily against a finding of confusion. See L.A. Gear, Inc. v. Thom McAn Shoe Co., 988 F.2d 1117, 1134 (Fed.Cir.), cert. denied, ___ U.S. ___, 114 S. Ct. 291, 126 L. Ed. 2d 240 (1993); Vitarroz Corp. v. Borden, Inc., 644 F.2d 960, 967 (2d Cir.1981). Reeboks were not sold at any of the same K-Mart stores as the Titan,^[16] and K-Mart has not carried any Freestyles or Ex-O-Fits since 1984. K-Mart stores last carried Reebok shoes in 1991. Even prior to 1991, K-Mart stores sold only a small number of "A-brand" shoes, and a very small number of Reebok shoes. These limited sales did not create any appreciable overlap in the K-Mart chain of stores, and do not alter the conclusion that the Titan was sold in a completely different channel of trade.

G. Likelihood that Reebok will Bridge the Gap Between the Products

Reebok introduced no persuasive evidence that it intends to enter the low-priced end of the sneaker market, and conceded that it has no plans to sell its shoes through discounters.

Reebok theorized that some consumers might believe that the Titan was part of a low-priced line of Reeboks, but did not establish that this had occurred or was likely to occur.^[17]

H. Consumer Care

The parties disputed whether persons intending to purchase shoes at a K-Mart exercise care in evaluating the shoes. Reebok alleged that the low price of the shoes at a K-Mart store attracts an unsophisticated consumer, and that the Titan's low price leads to impulse purchases. The court does not believe that the price of the Titan necessarily indicates a lack of care by the purchaser, nor did the evidence show that the K-Mart customer is necessarily unsophisticated.

The parties have not cited any case holding that the price of an item alone establishes a lack of care by consumers. To the contrary, this circuit has previously found that newsstand purchasers of a $2.95 magazine are reasonably sophisticated. Gruner + Jahr, 991 F.2d at 1079.

In addition, the court cannot find that K-Mart customers, by definition, are necessarily unsophisticated. It strikes the court as more than a little elitist to believe that K-Mart shoppers fail to exercise care merely because they shop at a discount store. The fact that a consumer may prefer to purchase low-priced footwear (or have no other option) does not transform that consumer into an unsophisticated, careless shopper.^[18] In fact, an argument can be made that shoppers with limited budgets use more care in spending their more limited resources than shoppers at non-discount stores.

Shoes intended for casual everyday street use, such as the Titan, are likely to receive a relatively high amount of consumer care. Shoes are an item worn daily, and are much less likely to be the subject of impulse purchases than other low priced consumer goods or consumables. Compare with Stern's Miracle-Gro Prods., Inc. v. Shark Prods., Inc., 823 F. Supp. 1077, 1089 (S.D.N.Y.1993) (hair care products and houseplant fertilizer examples of products likely to be subjects of impulse purchases). Unlike other inexpensive consumer products, a careless decision about footwear can lead a customer to experience ongoing discomfort and physical pain. Consumer care is further encouraged by the mostly self-service nature of K-Mart footwear departments. By laying the shoes out in their boxes, Meldisco allows K-Mart customers to closely inspect the shoes.

The court therefore believes that the Titan is an example of a low-priced item that consumers exercise more than a reasonable amount of care in purchasing.

I. Actual Confusion

Evidence of actual confusion can take two forms — scientific surveys testing the statistical likelihood of confusion among consumers,^[19] and incidents of actual confusion. Reebok did not offer a survey into evidence. Nor did Reebok show any incidents of actual confusion between the Titan and Reebok shoes.^[20] Reebok also failed to produce any evidence that the Titan was "passed off" as a Reebok product at any of the 800 K-Mart stores that carried the Titan.

Although evidence of actual confusion is not required in order to show a likelihood of confusion, the absence of actual confusion may be used against a plaintiff. Nikon, Inc. v. Ikon, Corp., 987 F.2d 91, 95 (2d Cir.1993). "[I]t is certainly proper for the trial judge to infer from the absence of actual confusion that there was also no likelihood of confusion." McGregor-Doniger, 599 F.2d at 1136; see E.S. Originals Inc. v. Stride Rite Corp., 656 F. Supp. 484, 490 (S.D.N.Y. 1987) (even where plaintiff presented unpersuasive, anecdotal evidence of actual confusion, the failure to undertake a survey may "strongly suggest that a likelihood of confusion cannot be shown").

The court believes that such an inference is warranted in this case. Reebok had ample time to prepare a survey or other evidence of actual confusion. The court rejects Reebok's claim that it did not conduct a survey or seek evidence of actual confusion since it believed that its declarations and mall interviews would be sufficient. In another action, a district court had specifically informed Reebok and its counsel that it "cannot place any value" on the declarations and individual mall interviews similar to those offered by Reebok in this case. See Payless Shoesource, Inc. v. Reebok Int'l Ltd., 804 F. Supp. 206, 212 n. 4 (D.Kan.1992), order vacated, 998 F.2d 985 (Fed.Cir.1993). And in evaluating Reebok's claim against the Titan, Customs specifically informed Reebok that consumer declarations would be insufficient evidence in the district court. Pl. exh. 76 at 6. Having been given these explicit warnings, Reebok's decision not to conduct a survey strongly suggests that a likelihood of confusion could not be shown.

J. Bad Faith and Intent to Confuse

Bad faith involves an attempt by a junior user to adopt a mark with either the intention of capitalizing on confusion between his and the senior user's product, or an intention to capitalize on the senior user's reputation and good will. Lang v. Retirement Living Publishing Co., 949 F.2d 576, 583 (2d Cir.1991). Knowledge of an existing registered trademark does not give rise to an inference of bad faith, Id. at 583-84, although use of an identical mark permits an inference of bad faith.

Here, the Titan's side design was far from identical to the registered Stripecheck mark, and differed materially from the side designs of the Ex-O-Fit and Freestyle. The court therefore looks to the other evidence to determine whether defendants intended to promote confusion between the products and appropriate the goodwill of another.

The fact that the Titan designers placed the Titan name on the side of the shoe helps demonstrate their good faith. Had they intended to confuse customers, the court does not believe the designers would have placed the Titan name on the sides and tongue of the shoe.^[21] The similarities in the sneaker side designs of other manufacturers also supports defendant's good faith. Shoe designers appear to be working from the same limited palette of angled geometric shapes. Even the leading manufacturers use similar angled geometric designs without any apparent intent to capitalize on another's goodwill.

The Titan's designers were aware of the Stripecheck mark. Defendants did not seek the advice of counsel at the design stage, but did seek this advice once Reebok voiced its objections to the Titan, Pl. exh. 174 at 87-88. The combination of the fact that defendants had knowledge of the other marks and failed to seek advice of counsel at the design stage raised some doubts in the court's mind. However, the trial and deposition testimony of Mr. Capozzoli and Mr. DeStefano, the parties responsible for the design of the Titan, removed these doubts and satisfied the court that defendants did not design the Titan with an intent either to promote confusion or to appropriate Reebok's goodwill.

Reebok also attempted to show that prior suits involving Melville and K-Mart justified a lower burden of proof in establishing bad faith. After reviewing these cases, the court does not find that defendants' prior Lanham Act experiences justifies a finding of bad faith in relation to the Titan.

K. Additional Factors

The Polaroid factors are not exclusive. An additional factor relevant to this case is the fact that the name "Titan" appears on the side panel and tongue of the shoe. A counterdesignation of origin certainly does not prevent a finding of infringement, Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1047 (2d Cir.1992) (goods carrying name of the manufacturer can still be confusingly similar), and can be an insignificant factor in defendant's favor. Reebok Int'l Ltd. v. Daytona Indus., Inc., No. 89 Civ. 5759 (KTD), 1989 WL 111049 at *5 (S.D.N.Y. Sept. 18, 1989) (label unlikely to dispel likelihood of confusion). However, the factor may still play some role in lessening the likelihood of confusion. See W.W.W., 984 F.2d at 573 (identifying name may lessen likelihood of confusion); Stormy Clime Ltd. v. ProGroup, Inc., 809 F.2d 971, 977 (2d Cir.1987) (label in traditional place was a factor differentiating products).

Aware of the limitations of this factor, the court believes that the presence of the Titan label helped some prospective purchasers realize that the product is not affiliated with Reebok. The self-service nature of K-Mart's shoe departments encourages consumers to physically handle and closely inspect the shoes, increasing the likelihood that they will notice the Titan name and realize its significance.

The presence of the Titan name, while relevant to the issue of purchaser confusion, is not very relevant to the issue of post-sale confusion. The Titan name is difficult to see from a distance of more than a few feet, especially in low light conditions, or when the shoe is worn with long pants. It therefore has little or no value as a counterdesignation of origin in the post-sale context. See Keds Corp. v. Renee Int'l Trading Corp., 888 F.2d 215, 222 (1st Cir.1989).

III. Purchaser Confusion

A. Trademark Claim

Likelihood of confusion requires a finding that numerous ordinarily prudent purchasers were likely to be confused as to the source of the product in question. Gruner + Jahr, 991 F.2d at 1077. Consumer confusion can include a belief that the trademark holder sponsored, endorsed, or was otherwise affiliated with the junior product. Stop the Olympic Prison v. United States Olympic Committee, 489 F. Supp. 1112, 1121-22 (S.D.N.Y.1980). The court's analysis looks beyond similarity in overall appearance, and balances all of the Polaroid factors to determine whether the similarity is likely to provoke confusion. McGregor-Doniger, 599 F.2d at 1133. After weighing the Polaroid factors in context and in their entirety, see Paddington, 996 F.2d at 584, the court does not find that an appreciable number of consumers were likely to believe that Reebok was associated with the Titan.

The court begins with the fact that Reebok Stripecheck is a strong mark, and that the side design of the Titan is generally similar to the Stripecheck. Also in favor of a finding of likelihood of confusion are the facts that the Titan and Reebok models are intended for similar use as casual footwear, and that the Titan's quality is inferior to Reebok's sneakers.

However, the other Polaroid factors all strongly indicate that there is little chance that a consumer would mistake the Titan for a Reebok shoe. K-Mart shoppers exercise a reasonable degree of care in purchasing sneakers intended for casual wear. The vast price difference between the Titan and any Reebok model would help inform a purchaser that the Titan is not part of the Reebok line. Consumers who knew enough about sneakers to recognize the Stripecheck as a Reebok mark would be especially likely to realize that Reeboks do not sell at $9.99 or $16.99. The presence of the "Titan" name on the shoe also helps prevent consumer confusion, particularly in light of the self-serve nature of K-Mart's shoe departments.

Another important factor in this case is the fact that the Titan and Reebok shoes are sold in completely different channels of trade. K-Mart did not carry Reeboks, with occasional exceptions for small numbers of shoes. K-Mart did not sell any Titans alongside Reebok shoes, and has not sold any Reeboks since 1991. Reebok had no plans to enter the low-priced end of the sneaker market and "bridge the gap" in the channels of trade. Reebok surmised that a consumer might believe they are buying a second-tier Reebok product, but presented no persuasive evidence that this occurred or was likely to occur.

Given these facts,^[22] the court concludes that there was no likelihood that consumers would be confused at the point of sale that the Titan was affiliated with Reebok. The differences in price and channels of trade, taken by themselves, removed most of the likelihood of confusion. Combined with the obvious presence of the Titan name on the shoe and Reebok's absence from the low-priced end of the sneaker market, the court is satisfied by the great preponderance of the evidence that there was no likelihood of confusion. Reebok's lack of survey evidence or other evidence of consumer confusion dispelled any remaining doubts.

B. Trade Dress Claim

Analysis of trade dress infringement also focuses on balancing the Polaroid factors in order to assess the ultimate question of the likelihood of confusion. Paddington, 996 F.2d at 584. The Titan, Ex-O-Fit and Freestyle are all monochromatic high-top sneakers intended for use as casual footwear, and the quality of the Reebok shoes exceeds the quality of the Titan. However, the same balance of factors that indicates no likelihood of trademark confusion also indicates that there was no likelihood of trade dress confusion.

In addition, the trade dress of the Titan differs substantially from the trade dress of the Ex-O-Fit and Freestyle. The lack of Reebok's distinctive velcro straps, concentric circle ankle cushioning, Starcrest, Union Jack and Reebok labels, and other differences in trade dress remove any remaining likelihood that consumers would be confused that the trade dress of the Titan was somehow associated with the Ex-O-Fit or Freestyle.

Differences in trade dress do not prevent a finding of infringement, and may in some cases be immaterial to the question of the likelihood of confusion. Here, however, the differences are substantial, and would be obvious to a consumer familiar with Reebok's trade dress. These differences, when viewed in conjunction with the other Polaroid factors, eliminate the likelihood that an appreciable number of consumers would be confused by the trade dress of the Titan.

IV. Post-Sale Confusion

Post-sale confusion can establish a claim of trademark infringement. Lois v. Levi Strauss, 799 F.2d at 872. Post-sale confusion occurs when a consumer sees a product worn by another individual outside of a retail store, wrongly associates the product with the trademark holder, and then allows that association to influence a later purchasing decision. Id. at 872-73.

In this case, post-sale confusion presents a closer question than point-of-sale confusion. Several Polaroid factors are immaterial in the context of post-sale confusion. Those factors include the price differences in the shoes and their different marketing channels. In addition, any physical differences between the products become less apparent when viewed outside of the point of sale, and any identifying labels are more likely to be removed by the purchaser. Adjusting the Polaroid analysis to account for these aspects of the post-sale context, there is at least the possibility of post-sale confusion whenever the market includes two products similar in appearance. However, the circuit has specifically noted that harm to a plaintiff is not inevitable merely because a competing product incorporates a generally similar mark. Vitarroz, 644 F.2d at 967.

In the context of point-of-sale confusion, courts have developed a large body of case law balancing the Polaroid factors, helping define the point at which the mere possibility of confusion becomes the likelihood of confusion. Post-sale confusion, on the other hand, is a relatively new cause of action whose contours are not yet thoroughly defined. The court relies primarily on Lois v. Levi Strauss, 799 F.2d 867 (2d Cir.1986), the circuit's leading post-sale confusion case, for guidance.

Lois involved a case in which it was undisputed that the trademark stitching pattern was "intimately associated" with the manufacturer, and had acquired "profound" secondary meaning. Id. at 876-77. The Stripecheck, although a strong mark, appears significantly weaker than the back pocket stitching of Levi jeans presented to the court in Lois. In that case, Levi established the strength of the stitching pattern by showing that this pattern was used on "all" of its jeans continuously since 1873, "undisputed examples of the intimate association between the stitching pattern and [Levi]'s products in the buying public's mind," and numerous examples of the "phenomenal reaction" to the stitching pattern, all supporting the court's conclusion that the back pocket stitching has become the "embodiment of Levi Jeans in the minds of the jeans buyers." Id. at 869. Here, in contrast, the Stripecheck was a strong but secondary mark, used on only a portion of Reebok's sneakers, and for which Reebok produced little reliable evidence of trademark strength, advertising expenditures, or consumer recognition. And unlike the present case, Levi had plans to "bridge the gap" in the channels of trade between the products.

The two marks in Lois were "essentially identical." Id. at 873. In this case, the Reebok and Titan side designs are generally similar, with some material differences. However, these differences become less discernable when the Titan is viewed at a distance, an important consideration in assessing post-sale confusion.

The Titan's monochromatic black color also contrasts with the mark in Lois. Lois involved highly visible back pocket stitching applied in a contrasting color. Here, the marks are applied in the same monochromatic tone as the sneakers, making the marks much less prominent to the observer. Black absorbs many decorative features of a shoe. As a result, all side stripe designs become less visible and identifiable when observed on all-black shoes in the post-sale context.

And unlike the present case, the Lois plaintiff produced at least some survey evidence of consumer confusion. Id. at 875-77. Thus, although Lois provides some guidance, numerous material distinctions prevent Lois from controlling the outcome of this case.

As discussed previously, post-sale confusion narrows the Polaroid factors, since differences in price and channels of trade are immaterial to post-sale confusion, and differences in the appearance of the products are minimized in the post-sale context. As a result, courts must be even more cognizant of the fact that balancing the remaining Polaroid factors is not a matter of keeping score, but is instead a helpful tool used to analyze the ultimate question — whether an appreciable number of consumers are likely to be confused as to the source of the products.

In Reebok's favor, the Stripecheck is a strong mark, the Titan is an inferior product put to the same use as the Ex-O-Fit and Freestyle, and the products are generally similar in appearance. In favor of defendants are the lack of any evidence of actual confusion, the lack of any research or reliable third-party testimony showing a likelihood of confusion, differences in the overall appearance of the shoes, and defendants' good faith. In addition, the general similarities in appearance between the Titan and the examples of the Ex-O-Fit and Freestyle presented at trial are due to some extent to the fact that they are all monochromatic all-black high-tops, rather than to the similarity of their side stripe designs.

The evidence presented at trial simply failed to establish that an appreciable number of consumers were likely to have been confused by the Titan in the post-sale context. The court cannot rule out the possibility that some consumers, viewing the Titan in the post-sale context, might believe that the Titan is a Reebok product. However, a court can only find a Lanham Act violation if there is a likelihood of confusion, not merely the possibility of confusion. See Eldon Indus. Inc. v. Rubbermaid, Inc., 735 F. Supp. 786, 796, 828 (N.D.Ill.1990) (likelihood of confusion minimal despite possibility that some consumers might confuse products in post-sale setting); Vitarroz, 644 F.2d at 967 (similarity of marks on competing products does not make harm to plaintiff inevitable).

Even had Reebok demonstrated a likelihood that an appreciable number of consumers would mistake the Titan for a Reebok shoe, a claim of post-sale confusion requires the additional steps of showing that the consumers will perceive something about the product, and that what they observe will affect a later purchasing decision.^[23] These steps can often only be inferred. Of course, courts analyzing the likelihood of confusion must necessarily often infer the existence of relevant factors in order to answer the ultimate question. In appropriate cases, these inferences are warranted, and will lead to a finding of post-sale confusion. However, at some point unfounded inferences become too remote from the evidence to support a conclusion that confusion is likely. Courts must be particularly wary of wholly speculative claims in the context of post-sale confusion. As one of the bench's leading experts on the Lanham Act has stated:

Point-of-sale confusion is by far the more important, because it directly affects individuals who are in the market for the particular product. Post-sale confusion may affect future purchasers of the product, but in a more indirect and diffuse manner; post-sale confusion is far less likely to cause erroneous purchases than point-of-sale confusion.

Plasticolor Molded Prods. v. Ford Motor Co., 713 F. Supp. 1329, 1336 n. 8 (C.D.Cal.1989) (Kozinski, J.), vacated by consent judgment, 767 F. Supp. 1036 (1991).

Regardless of whether it occurs at the point of sale or post-sale, a likelihood of confusion still requires the same thing — a showing that an appreciable number of consumers are likely to be confused. Here, the claim of post-sale confusion is wholly speculative^[24] and is too far removed from the evidence to permit an inference that confusion was likely. Reebok has simply failed to demonstrate by a preponderance of the evidence that an appreciable number of consumers were likely to be confused as to the source of the Titan in the post-sale context.

V. Trademark Counterfeiting

A counterfeit trademark is a falsely applied mark, identical to or substantially indistinguishable from the registered trademark. 15 U.S.C. § 1127; Montres Rolex, 718 F.2d at 532. The court compares the Titan not only to the trademark as formally registered with the PTO, but also to the trademark as actually used on Reebok's merchandise. Montres Rolex, 718 F.2d at 532. Here, the side design of the Titan is materially distinguishable from the Stripecheck as depicted in the trademark registration and as used on the Freestyle and Ex-O-Fit. The horizontal element of the Titan side design is completely different from the horizontal element of Reebok's designs. In addition, the shape, proportion and angle of the vertical element of the Titan differs materially from Reebok's vertical elements. In fact, the vertical element of the Titan bears little resemblance to the vertical element of Stripecheck either as it appears on the trademark registration or as it appears on the Ex-O-Fit. These substantial material differences preclude a finding of trademark counterfeiting.

VI. Exclusion Of Evidence At Trial

At trial, Reebok sought to introduce several consumer declarations obtained after mall-intercept interviews in order to demonstrate actual confusion. Reebok also asked that two of the interviewees be permitted to testify. The court refused introduction of any of this documentary or testimonial evidence.

Reebok's offer of proof and pre-trial memoranda indicated that this evidence originated in mall-intercept interviews conducted with passersby at several sites throughout the country, gathered solely for use in connection with this litigation. Counsel for Reebok prepared a series of questions, and hired two market research companies to conduct the mall-intercepts. Once intercepted, passersby were asked a few cursory screening questions, and then asked a series of questions after being shown a Titan shoe.

This type of methodology can be very helpful in assessing the likelihood of confusion when it is part of a scientifically based, statistical survey of consumer reaction. Here, however, Reebok claimed that the results of these interviews constituted specific examples of "actual confusion," and repeatedly denied that this evidence constituted survey evidence.

The court refused to accept the results of these interviews as examples of actual confusion. Reebok obtained these interviews through a staged, artificial situation. Reebok did not provide an offer of proof establishing that the interview conditions realistically simulated the point-of-sale or post-sale context. In fact, Reebok did not even attempt to introduce testimony from any of the interviewers or their supervisors concerning the testing procedures or conditions. Reebok also avoided mentioning how many persons participated in the mall-intercept.

Even had Reebok made an offer of proof that the mall intercept realistically simulated the point-of-sale or post-sale context, examples of "actual" confusion could not result from the controlled mall-intercept setting used by Reebok. The Second Circuit has frequently "commented upon the difficulty of establishing actual confusion on the part of retail customers." Mushroom Makers, Inc. v. R.G. Barry Corp., 580 F.2d 44, 48 (2d Cir.1978), cert. denied, 439 U.S. 1116, 99 S. Ct. 1022, 59 L. Ed. 2d 75 (1979). Introduction of the type of evidence offered by Reebok would transform this difficult burden into a laughably simple task by allowing plaintiffs to manufacture "incidents" of actual confusion in mall intercepts. For example, a plaintiff might commission a consumer survey of 1000 consumers in order to assess the statistical likelihood of confusion of two products. If the survey results failed to indicate a statistically significant likelihood of confusion, perhaps by indicating that only 1% of the participants confused the products, the plaintiff could still attempt to introduce the 1%, or ten participants, as incidents of "actual confusion." As one commentator noted:

Most surveys do not measure actual confusion. Surveys only give us information about a controlled and artificial world.... One might be able to draw helpful inferences from a survey of randomly selected pedestrians in a shopping mall who are interrogated about pictures of two products, but to claim that their responses are direct proof of the responses of actual consumers as they make their purchasing decisions is going too far.

Perlman, "The Restatement of the Law of Unfair Competition: A Work in Progress", 80 Trademark Rep. 461, 472 (1990). There is no question that Reebok's staged and artificial consumer interviews were not evidence of actual confusion.

In its post-trial brief, Reebok claims that it suffered prejudice as a result of the court's denial of defendant's earlier motion in limine to exclude this evidence.^[25] Plaintiff's posttrial memorandum of law at 41 n. 170. The court fails to see how the court's in limine ruling prejudiced Reebok. A ruling on a motion in limine does not constitute a final ruling on admissibility. Luce v. United States, 469 U.S. 38, 41-42, 105 S. Ct. 460, 463-64, 83 L. Ed. 2d 443 (1984). Reebok has not pointed to any evidence that it would have submitted had the court excluded the consumer interviews before trial.

Trial counsel assumes the risks of its decisions regarding trial tactics. Counsel was well aware of the fact that a court "cannot place any value" on this type of evidence when offered to show actual confusion. See Payless, 804 F.Supp. at 212 n. 4; Pl. Exh. 76 at 6 (Reebok presented Customs with several consumer declarations as evidence of actual confusion, but Customs unambiguously notified counsel that "a court would require substantially more than [these declarations] ..."). Reebok's decisions regarding trial tactics do not excuse the resulting evidentiary failures.

The court also looks with great skepticism on Reebok's claim that it declined to conduct a survey due to a desire to avoid "a waste of resources." Pl. Post-Trial Br. at 41. Reebok, using experienced trademark counsel at trial, was well aware of the pivotal role that consumer surveys and other evidence of confusion play in Lanham Act claims. Furthermore, counsel admitted that Reebok made "valiant efforts," Tr. 13, and incurred significant expenses in preparing a questionnaire, hiring outside firms to interview a large number of individuals in mall-intercepts at several locations throughout the nation, and paying airfare and hotel bills to have these interviewees appear in court. Any waste of expenditures resulted from counsel's decision to expend significant time and financial resources on manufacturing unpersuasive, unreliable and inadmissible "evidence" instead of expending the same resources to gather competent, reliable and admissible evidence.

VII. State Law Claims

Reebok also brings state law claims based on violation of New York's antidilution statute, N.Y.Gen.Bus.L. § 368-d, and state common law. In Escada AG v. The Limited, Inc., 810 F. Supp. 571, 573-74 (S.D.N.Y.1993), this court held that federal patent law preempted § 368-d as applied to potentially patentable designs. Reebok has stipulated that the Stripecheck design "is of the type on which a patent may have issued, assuming the requirements of novelty, non-obviousness and lack of functionality were met." JPTO Undisputed Facts ¶ ii. The PTO has issued Reebok Design Patent No. 294,654 on Mar. 15, 1988 for an "ornamental design for an athletic shoe upper" bearing the Stripecheck design. Def. exh. BA1. The court concludes that federal design patent law preempts Reebok's antidilution claim under § 386-d.

Reebok's claim of common law unfair competition fails on the merits. A common law unfair competition claim is based on the same issue determinative in a Lanham Act claim — the likelihood of confusion. Andy Warhol Enters. v. Time Inc., 700 F. Supp. 760, 763 (S.D.N.Y.1988). As discussed in the analysis of Reebok's Lanham Act claims, Reebok has failed to show any likelihood of confusion. Accordingly, its claim of common law unfair competition fails.

CONCLUSIONS

The court finds for defendants on all of Reebok's claims of trademark infringement and unfair competition. The court finds for Reebok on defendants' counterclaim. The parties will bear their own attorneys' fees.

So Ordered.

NOTES

[1] Approximately 50-60% of 1991-93 Ex-O-Fit sales came from sales of high-top versions, while 60-70% of Freestyle sales came from high-tops. Pl. exh. 4. All-black versions constituted a large portion of the sales of both the Ex-O-Fit and the Freestyle.

[2] Line drawings of the Titan and Stripecheck appear as Appendix A to this opinion.

[3] Meldisco attempted to register the Titan name as a trademark. It abandoned the application in 1993 after discovering that an industrial footwear manufacturer had previously registered the Titan name.

[4] In its presentation to Customs, Reebok misrepresented the fact that K-Mart stores sold the Titan alongside Reebok shoes. Pl. exh. 76 at 5.

[5] The court notes that this witness has a strong professional interest in Reebok's ownership of the Stripecheck. He has been a Reebok employee since 1982, is now president of Reebok's fitness division.

[6] Reebok did not submit low-top versions of these shoes into evidence, and did not base its claim of trade dress infringement on the low-top versions.

[7] The court can be skeptical of the credibility of paid experts. Mr. Nutt, however, proved a highly credible witness. Documentary materials produced by Reebok, statements by Reebok's witnesses, and the other evidence produced at trial supported Mr. Nutt's conclusions. The court found Mr. Nutt to have great expertise and his statements to have had a solid factual basis. Reebok's expert on the sneaker market, in contrast, testified in the abstract and without documentary support, and was thoroughly unconvincing.

[8] For example, Nike uses its "Swoosh" design, with only minor variations, on all but three of the several hundred sneaker models depicted in Nike's 1990 catalog. Def. exh. AO. Asics and Puma each use their respective side designs on nearly all of their shoes, and generally use these designs with only small variations in appearance. See Exh. Y3, AB1.

[9] According to Reebok's vice-president for design, Reebok designers labor under only one trademark directive — that the Reebok name always appear in an identical typeface on the shoe. Pl. exh. 176 at 39. He conceded that Reebok did not instruct its designers to use Reebok's other trademarks uniformly or consistently. Pl. exh. 176 at 111.

[10] Reebok did not introduce any persuasive testimonial or documentary evidence that consumers recognized Reebok Classics as a separate product category, or that the designation had any other significance to consumers.

In addition, Reebok failed to provide evidence of what advertising expenditures, if any, it applied to the Classics line in 1991, 1992 or 1993. Ex. 169. A summary of advertising expenditures stated that Reebok expended $0 on advertising the Classics line in 1991 and 1992. Id. The court could not accept the explanation of another witness (who did not prepare this exhibit) that the summary included Classics expenditures under another heading, even though Classics were given a heading of their own in the advertising summary. The only objective evidence of advertising for Reebok Classics was a single print advertisement for a running sneaker, with a "Reebok Classic" tag line and logo. This ad bore a 1993 copyright date, and could not even suggest possible consumer awareness of the Reebok Classics line in the 1990-92 period.

[11] Many of the Classics do not bear the Stripecheck. The "Princess" aerobic shoe, the "Phase II," "NPC Insignia" and "NPC II" tennis shoes, and the "Pro Cheer II" cheerleading shoe, all categorized as "Classics" in Reebok's 1992 catalog, do not use the Stripecheck. Def.exh. H4; see Amended PTO Undisputed Facts ¶ m (shoes were in Classic line in Spring of 1993). The Princess is one of the best selling shoes in the Classic line. Def.exh. BN at 16. The Phase II has since been dropped from the Classics line. Id.

[12] Exhibit H4 bore a date of 1990, and a Reebok advertising executive testified that the brochure was circulated to the advertising department in 1993. When questioned about the 1990 copyright date on the brochure, the witness remembered that the brochure was circulated in late 1990. Tr. 200-01. Reebok circulated a new set of brand identity guidelines sometime in 1993, which replaced the earlier 1990-93 brand identity manual.

[13] The Ex-O-Fit and Freestyle received almost no advertising attention from 1988 until 1993. From 1988 through the first half of 1991, Reebok did not use any paid-for advertising for the Ex-O-Fit. Tr. 205. Following that, the Ex-O-Fit and Freestyle were not part of Reebok's main paid-for advertising campaigns. Tr. 206.

[14] This witness failed to produce any evidence regarding what portion of Reebok's advertisements depicted products bearing the Stripecheck mark, Tr. 160, 166-67, initially misrepresented the fact that the Classics line of shoes all bore the Stripecheck mark, Tr. 192, dubiously claimed that an ad bearing a copyright date of 1993 had been running since 1991, Tr. 172, and conceded that she did not work personally on even the limited number of ads introduced at trial. Tr. 208.

The court could not give much weight to her conclusory testimony regarding the frequency with which Reebok advertises shoes bearing the Stripecheck. In her deposition testimony, this witness specifically disclaimed knowing which shoes incorporated the Stripecheck design, and denied that she was qualified to determine whether specific shoes carried the Stripecheck mark. Def.exh. BU at 30-31, 67. This witness also testified that "All our shoes — most of our shoes feature the Stripecheck, ...," a statement completely at odds with even a cursory examination of Reebok's product line. Def.exh. BU at 67.

[15] A Reebok marketing executive testified that Reebok conducted a study of consumer recognition of the Stripecheck. Tr. 97-98. Reebok failed to produce this evidence to defendants in discovery and objected to defendants inquiring into the study at trial. Id.

[16] The court believed that this issue had been well-settled at trial, only to be surprised by a claim in Reebok's post-trial brief that "Meldisco sold the Titan alongside Reebok shoes in K-Mart stores." Pl. Post-Trial Br. at 15 n. 84. Reebok supports this misstatement with several phantom citations to the record, which unambiguously do not support Reebok's proposition.

[17] A professional investigator employed by Reebok for several years stated that he had overheard some anonymous customers in unnamed stores on unspecified occasions remark that Reebok and other shoe companies made cheap shoes and sold them under different names. Tr. 321. This meager, vague testimony did not convince the court that any K-Mart shoppers believed this.

[18] A recent opinion in this district stated that: "Even if some of the prospective purchasers of Dom Perignon are from low income groups, and are therefore less sophisticated shoppers than wealthier purchasers,...." Schieffelin & Co. v. The Jack Co., 1994 WL 144884 (BN) 1994 U.S.Dist.Lexis 3785 at *55 (S.D.N.Y.1994). The court can accept this statement for the proposition that low income purchasers might have less familiarity with purchase of a super-luxury item such as a $100 bottle of champagne than would wealthy consumers. However, the court expressly disagrees with this statement's implication that there is a direct relationship between income and consumer intelligence. Careless shopping habits are not a necessary by-product of a low income.

[19] Courts generally categorize consumer surveys as evidence of "actual confusion." Although this phrase is faithful to the language of Judge Friendly's list of eight Polaroid factors, "actual confusion" does not accurately describe the nature of consumer surveys. Consumer surveys are probably better described as a statistical means of predicting the likelihood that actual consumers will confuse the products.

[20] As discussed supra, Reebok claimed that it had "evidence" of "actual confusion." This proved to be neither admissible evidence nor examples of actual confusion.

[21] The decision to apply the name in black was consistent with the design objective of an all-black shoe, and does not demonstrate an attempt to deceive consumers. A name in a contrasting color would detract from the monochromatic appeal of the shoe.

[22] Defendants' good faith in designing the Titan plays little role in the court's analysis.

[23] Reebok's witnesses and trial memoranda failed to articulate what post-sale observers would notice about the Titan, or how those observations would affect later purchasing decisions.

No Reebok witnesses testified that the Titan was aesthetically inferior to Reebok's shoes. Reebok established that the materials and construction of the Titan were inferior to those of Reebok's shoes. The Titan's inferiority would affect both the durability of the shoe and the comfort and support afforded the wearer. The court infers from this testimony that some of the differences in quality might be apparent to an observer after a Titan had been in use for a significant period of time. Lacking any specific laboratory tests on the Titan, the court cannot conclude what form these differences would take, how quickly they would occur, or how visible they would be to an observer. Nonetheless, the differences in quality create at least the possibility that a Titan might convey a negative impression to some consumers.

[24] In fact, none of Reebok's witnesses specifically addressed the question of post-sale confusion. Reebok's trial and post-trial memoranda of law do not specifically discuss post-sale confusion, mentioning it only in passing.

[25] Defendants had previously sought to exclude this evidence through a motion in limine dated November 11, 1993. Defendants argued that this evidence constituted a consumer survey, conducted in violation of Judge Patterson's previous instruction that Reebok complete any survey and submit its results to defendants by August 4, 1993. Reebok's memorandum in opposition to the motion in limine denied that the evidence constituted a survey, but argued in the alternative that the questionnaires would still be admissible survey evidence. The parties completed briefing the motion in limine on December 2, 1993.

From the papers filed in connection with the motion, it was unclear at the time whether the evidence constituted a consumer survey or evidence of actual confusion. In addition, Reebok's memoranda did not concede the existence of Judge Patterson's scheduling instruction, nor was such an instruction formally docketed. Given these facts, and lacking a formally docketed order barring a survey conducted after August 4, I denied the motion in limine on December 3 in order to see how the case unfolded and better assess the nature and relevancy of the evidence and its relation to the issues developed at trial.