A H Sportswear Co. v. Victoria's Secret Stores

Full title: A H SPORTSWEAR CO., INC. and Mainstream Swimsuits, Inc., Plaintiffs, v…

Court: United States District Court, E.D. Pennsylvania

Date published: Aug 17, 2001

167 F. Supp. 2d 770 (E.D. Pa. 2001)

Opinion

No. 94-7408.

August 17, 2001.

Gus Milides, Easton, PA, Norman Seidel, Laub, Seidel, Cohen Hof, Easton, PA, Stephen J. Meyers, Seidel, Gonda, Lavorgna Monaco, P.C., Philadelphia, PA, Ronald N. Weikers, Seidel, Gonda, Lavorgna and Monaco, Philadelphia, PA, Arthur H. Seidel, Drinker Biddle Reath LLP, Philadelphia, PA, for A H Sportswear, Inc., Mainstream Swimsuits, Inc.

Frank J. Colucci, Colucci Umans, New York, NY, H. Robert Fiebach, Cozen and O'Connor, Philadelphia, PA, Richard P. Jacobson, Colucci and Unams, New York, NY, for Victoria's Secret Stores, Inc., Victoria's Secret Catalogue, Inc.

DECISION AND ORDER

VAN ANTWERPEN, District Judge.

For nearly seven years this trademark dispute has moved back and forth between this Court and the Third Circuit Court of Appeals. As a result of the most recent partial remand from the Third Circuit Court of Appeals, we again find ourselves weighing in on the issues dividing these parties.

I. BACKGROUND

We begin with a brief review of the tortured history of this case. In the initial trial, we found that the use of THE MIRACLE BRA as applied to lingerie did not create a likelihood of confusion, but there was a possibility of confusion when the mark was applied to swimwear. A H Sportswear, Inc. v. Victoria's Secret Stores, Inc., 926 F. Supp. 1233, 1269 (E.D.Pa. 1996) ( "A H I"). Since much work would be required to set damages, we certified our entire opinion to the Third Circuit for review, but they refused. Following a damages trial, we ruled that Defendants could not use THE MIRACLE BRA trademark with respect to swimwear unless they used a disclaimer and paid Plaintiffs a reasonable royalty. A H Sportswear, Inc. v. Victoria's Secret Stores, Inc., 967 F. Supp. 1457, 1482-83 (E.D.Pa. 1997) ( "A H II")

Both parties appealed. The Third Circuit, en banc, affirmed our finding of no likelihood of confusion with regard to lingerie but determined that the proper standard in this Circuit was now "likelihood of confusion." A H Sportswear, Inc. v. Victoria's Secret Stores, Inc., 166 F.3d 197, 206 (3d Cir. 1999) ( "A H III").¹ They remanded the case to us to consider whether there was a likelihood of confusion with regard to THE MIRACLE BRA swimwear. Id. The Third Circuit also directed that we examine whether the doctrine of reverse confusion was implicated in this case. Id. at 207.

1 The Court also vacated the remedy we awarded. A H III, 166 F.3d at 208-9.

On remand, we found that there was no likelihood of direct confusion. A H Sportswear Co. v. Victoria's Secret Stores, 57 F. Supp.2d 155, 169 (E.D.Pa. 1999) ( "A H IV"). That is, it was unlikely that a consumer would regard THE MIRACLE BRA swimwear as a product of A H. Id. We then determined, under existing case law, that there were no grounds for a claim of reverse confusion. Id. at 176-178. Accordingly, we denied injunctive relief. Id.

The parties appealed again. The Court of Appeals, this time in a panel decision, affirmed the decision as to our conclusion that there was no likelihood of direct confusion. A H Sportswear Inc. v. Victoria's Secret Stores, Inc., 237 F.3d 198, 227 (3d Cir. 2000) ( "A H V"). With regard to the claim of reverse confusion, the Court of Appeals held that we applied an incorrect test. Id. at 236. The Court remanded the case to us, directing that we apply a new ten factor test to evaluate the merits of Plaintiffs' reverse confusion claim. Id. at 229. The Court also noted that we need not hear new evidence. Id. at 238.

On January 11, 2001, Plaintiffs filed a "Motion For Hearing on Contempt and Additional Hearing." We denied this motion in its entirety. A H Sportswear, Inc. v. Victoria's Secret Stores, Inc., 134 F. Supp.2d 668, 669 (E.D.Pa. 2001) ( "A H VI"). In particular, we refused to open up the voluminous record and receive additional evidence. Id. at 681. Previously, the parties were provided with extensive opportunities to submit evidence and make arguments.² From October 25 to November 3, 1995, we held a non-jury civil trial to determine the issue of liability. A H I, 926 F. Supp. at 1235. We received into evidence over three-hundred exhibits. On May 24, 1996, we issued a nearly forty page opinion which made 75 findings of fact and 6 conclusions of law. Id. To settle the issues related to damages, we made additional findings of fact and further evaluated the evidence. A H II, 967 F. Supp. at 1460-1483. On appeal, the case was the subject of two substantive opinions by the Third Circuit. See A H Sportswear v. Victoria's Secret Stores, 166 F.3d 191, 195 (3d Cir. 1999) (panel decision); A H III, 166 F.3d at 197 (en banc decision). After the case was remanded to us, we received multiple memorandums from the parties concerning the issues on remand. We then issued another substantive opinion addressing the issues of likelihood of direct and reverse confusion. A H IV, 57 F. Supp.2d at 155. The case was appealed again. The Court of Appeals issued yet another substantive opinion which remanded the case to us again. A H III, 166 F.3d 197. Plaintiffs then filed a motion for hearing on contempt and additional hearing on issues remanded to us. A H VI, 134 F. Supp.2d at 669. We received extensive briefs on these issues. Although we refused to hold a hearing, allow further discovery, or receive additional evidence, we did permit the parties to file lengthy briefs and submit up to twenty five additional proposed findings of fact. Id. at 681. This Court has expended an extraordinary amount of time and devoted tremendous resources to this case, reacting to matters both properly before us and the extraneous issues the parties choose to bring up. We do not think that the parties should be able to re-open the record just because the litigation has been protracted. In addition, for the reasons set forth in our opinion of March 12, 2001, we believe that the existing record is sufficient to decide the issues before us.

2 The District Court Docket for this case presently contains 229 entries. In addition to the papers submitted in connection with our four previous substantive opinions for this matter, there have also been numerous other motions and requests.

In A H VI, we directed the parties to brief the issue of reverse confusion only, since our adjudication of Plaintiffs' direct confusion claim was affirmed on appeal. 134 F. Supp.2d at 681. Thus, presently before us are:

1. Plaintiffs' Memorandum Concerning Reverse Confusion,³ filed on April 11, 2001;

2. Defendants' Memorandum Concerning the Likelihood of Reverse Confusion, filed on May 14, 2001;

3. Plaintiffs' Reply Memorandum, filed on May 31, 2001; and

4. The parties proposed additional findings of fact and responses thereto.

3 As filed, Plaintiffs memorandum also addressed the likelihood of confusion and what they labeled as "Six Additional Issues." In our July 12, 2001 Order we struck these portions of the brief because they violated our March 12, 2001 opinion. Accordingly, they are not before us and we have not addressed them in this opinion.

Before turning to these issues, we wish to make plain the fact that the sole issue before us is the likelihood of reverse confusion between the MIRACLESUIT trademark and THE MIRACLE BRA trademark as applied to swimwear. The issue of likelihood of confusion between MIRACLESUIT and THE MIRACLE BRA lingerie has already been settled. A H I, 926 F. Supp. at 1263-4. There, we found that Plaintiffs had failed to show a likelihood of confusion with respect to Defendants' use of its mark on lingerie. Id. The Third Circuit affirmed this determination. A H III, 166 F.3d at 201-10. On remand, the Third Circuit only asked us to consider whether the doctrine of reverse confusion was implicated with respect to the swimwear. Id. at 201, 207. We limited our discussion of reverse confusion to the swimwear in A H IV and neither the parties nor the Third Circuit disagreed with this decision. In the briefs presently before us, the parties only discuss the likelihood of reverse confusion with respect to the swimwear.⁴

4 "[T]he sole issue before the Court is whether Victoria's Secret's use of THE MIRACLE BRA on swimwear causes reverse confusion with A H's MIRACLESUIT." Defs. Mem. at 17. Pages nineteen to forty-two of Plaintiffs' brief only discusses swimwear. The remainder of the brief dealt with issues beyond the scope of the remand and was therefore struck by our July 12, 2001 Order.

In any event, we do not find that that there is a likelihood of reverse confusion with respect to Defendants' use of THE MIRACLE BRA trademark on lingerie. As discussed more fully below, the factors looked at are the same, regardless of whether the claim is for direct or reverse confusion. Considering the relative minor changes to the Lapp factors, and in light of the extensive analysis in A H I, we find that Plaintiffs have not shown a likelihood of reverse confusion with respect to the use of THE MIRACLE BRA trademark on lingerie.

II. FACTS

We will provide only a brief overview of the facts in the record since they were previously presented in multiple decisions. See, e.g., A H I, 926 F. Supp. at 1235-54. A H Sportswear, Co. Inc. ("A H")⁵ is a manufacturer of about 10 percent of the swimsuits made in the United States. Its bathing suits are made from a patented fabric designed to make the wearer look slimmer. On October 27, 1992, it received a federal trademark registration for use of the MIRACLESUIT mark for swimwear.

5 Mainstream Swimsuits, Inc. is a separate corporation which serves as the exclusive distributor of MIRACLESUIT products manufactured by A H. Because the nature of the corporate relationship between these two entities does not effect the issue of Defendants' liability for reverse confusion, we will refer to both Plaintiffs collectively as A H.

Victoria's Secret Stores, Inc., and Victoria's Secret Catalogue, Inc. (collectively "Victoria's Secret"), have sold women's lingerie under the name THE MIRACLE BRA since 1993, holding a federal trademark registration for use of the mark with lingerie since 1994. In 1994, they introduced THE MIRACLE BRA swimwear. Later that same year, A H commenced this litigation for trademark infringement and related claims. After suit was filed, Defendants began to use a disclaimer which stated that it was not associated with the MIRACLESUIT. Eventually, Defendants were denied a trademark registration for the use of THE MIRACLE BRA mark on swimwear by the Patent and Trademark Office (PTO).

In connection with this motion the parties have also agreed to several additional findings of fact.

1. Victoria's Secret has a well established trade name and the economic power to advertise THE MIRACLE BRA swimwear.

2. Victoria's Secret is a wealthier, more powerful company A H Sportswear, Inc. or Mainstream Swimsuits, Inc.

3. Victoria's Secret was aware of the Miraclesuit mark for swimwear when it adopted THE MIRACLE BRA for swimwear.

4. Victoria's Secret considered whether the adoption of the mark might result in confusion.

5. Victoria's Secret was aware, at the time it was the junior user of the mark that the Patent and Trademark Office had "refused registration" because of "confusingly similarity between the registrant's mark MIRACLESUIT and the applicant's mark would cause consumer confusion as to the source" of the goods.

6. "Similarity of marks" is an important factor in determining whether there is reverse confusion.

III. DISCUSSION

A. Trademark Infringement

Section 32(1) of the Lanham Act states:

Any person who shall, without the consent of the registrant —

(a) use in commerce any reproduction, counterfeit, copy, or colorable imitation of a registered mark in connection with the sale, offering for sale, distribution, or advertising of any goods or services on or in connection with such use is likely to cause confusion or cause mistake, or to deceive; . . . shall be liable in a civil action by the registrant. . . .

15 U.S.C. § 1114(1). To establish infringement under this act, the trademark owner must prove that: "(1) the mark is valid and legally protectable; (2) the mark is owned by the plaintiff; and (3) the defendant's use of the mark to identify goods or services is likely to create confusion concerning the origins of the goods or services." Fisons Horticulture, Inc. v. Vigoro Indus. Inc., 30 F.3d 466, 472 (3d Cir. 1994); Express Servs. Inc. v. Careers Express Staffing Servs., 176 F.3d 183, 185 (3d Cir. 1999) (quoting same).

With regard to the first two elements, there is no real dispute that Plaintiffs' own a valid and protectable mark. Plaintiffs' mark is registered with the U.S. Patent and Trademark Office on the Principal Register, which is reserved for distinctive marks. Id. Pursuant to Section 7(b) of the Lanham Act, 15 U.S.C. § 1057(b), A H's Principal Register registration is: "prima facie evidence of the validity of the registered mark and of its registration of the mark, of the registrant's ownership of the mark, and of the registrant's exclusive right to use the mark in commerce or in connection with the goods or services specified in the certificate." Such registration also entitles Plaintiffs to: "a strong prima facie presumption that its registered mark is either not merely descriptive or, if descriptive, that secondary meaning is presumed, which amounts to the same thing." McCarthy on Trademarks and Unfair Competition, 4th ed. § 11:43 (hereafter "McCarthy"). See Playboy Enter., Inc. v. Chuckleberry Publ'g, Inc., 687 F.2d 563, 567 (2d Cir. 1982) (Principal Register registration "indicates that the mark is not merely descriptive and gives to it a strong presumption of validity"). A H has used its MIRACLESUIT mark for swimsuits since at least 1991. It has extensively advertised and promoted its goods under the mark. Further, there have been substantial sales of goods under the MIRACLESUIT mark throughout the United States, including Pennsylvania. Thus, we find that Plaintiffs have established that they own a valid and protectable mark.

Having addressed the first two elements of trademark infringement, we turn to an examination of whether the Defendants' mark is likely to create consumer confusion.

B. Reverse Confusion Generally

The traditional pattern of direct confusion occurs when customers mistakenly think that the junior user's goods are from the same source as, or are connected with, the senior user's goods. McCarthy § 23:10. Conversely, with reverse confusion, customers purchase the senior user's goods under the mistaken impression that they are getting the junior user's goods. Id. This is likely to occur when a person familiar with the well-known junior user first comes into contact with the lesser-known senior user and, because of the similarity of the marks, mistakenly thinks that the senior user is the same as, or is affiliated with, the junior user. Id. Rather than try to profit from the senior user's mark, the junior user attempts to saturate the market and overwhelm the senior user. Id. The result of such action is that the senior user loses the value of the trademark, its product identity, corporate identity, control over its goodwill and reputation, and ability to move into new markets. See Commerce Nat. Ins. v. Commerce Ins. Agency, 214 F.3d 432, 444 (3d Cir. 2000) quoting Fisons, 30 F.3d at 474. See also Ameritech, Inc. v. American Info. Tech. Corp., 811 F.2d 960 (6th Cir. 1987); Sands, Taylor Wood Co. v. Quaker Oats Co., 978 F.2d 947 (7th Cir. 1992).

Reverse confusion is not completely different from the more typical direct confusion claim. With both types of trademark claims, the presence, or absence, of likelihood of confusion remains the key. The law simply permits no remedy unless it can be proved in some manner that the public was likely to be confused by the defendant's conduct. Thad G. Long Alfred M. Marks, Reverse Confusion: Fundamentals and Limits, 84 Trademark Rep. 1, 2-3 (1994) (hereafter "Long Marks"); Pignons v. Polaroid Corp., 657 F.2d 482, 492 n. 4 (1st Cir. 1981). It is only incidental whether the confusion reflects a belief that defendant's products or services are those of the plaintiff (likelihood of direct confusion) or a belief that plaintiff's products or services are those of the defendant (likelihood of reverse confusion). Id.

C. The Lapp Test for Reverse Confusion

To evaluate whether there is a likelihood of confusion, the Third Circuit has now developed a ten factor test which tracks the direct confusion test for noncompeting goods articulated in Interpace Corp. v. Lapp, Inc., 721 F.2d 460, 463 (3d Cir. 1983) (hereafter "Lapp"). A H V, 237 F.3d at 234. The factors we are to evaluate are as follows:

1. The degree of similarity between the owner's mark and the allegedly infringing mark;

2. The strength of the owner's mark;

3. The price of the goods and other factors indicative of the care and attention expected of consumers when making a purchase;

4. The length of time the defendant has used the mark without evidence of actual confusion arising;

5. The intent of the defendant in adopting the mark;

6. The evidence of actual confusion;

7. Whether the goods, competing or not competing, are marketed through the same channels of trade and advertised through the same media;

8. The extent to which the targets of the parties sales efforts are the same;

9. The relationship of the goods in the minds of consumers; and

10. Other facts suggesting that the consuming public would expect the prior owner to manufacture both products or to manufacture a product in the defendant's market, or would expect that the prior owner is likely to expand into the defendant's market.

Id. As with the test for direct confusion, we must weigh all ten factors; no single one being dispositive. Id.

In connection with our review of Plaintiffs' direct confusion claim, we considered all of these Lapp factors, albeit under a slightly different framework. See A H IV, 57 F. Supp.2d at 163; A H V, 237 F.3d at 215-216. Four of the Lapp factors are analyzed the same for both direct and reverse confusion claims: attention expected of consumers when making a purchase ( Lapp factor no. 3); the channels of trade ( Lapp factor no. 7); the targets of sales efforts ( Lapp factor no. 8); and the relationship of goods in consumer's minds ( Lapp factor no. 9). A H V, 237 F.3d at 237. Our consideration of these factors is unlikely to differ from what we found when we looked at the identical factors in conjunction with the direct confusion claim. Id. The Third Circuit also noted that under the circumstances of this case, our evaluation of factors four and six, pertaining to actual confusion, was unlikely to result in us drawing new conclusions as to whom the factors favored. Id. As to the remaining Lapp factors, the Third Circuit determined that the application of these factors can differ when utilized to analyze reverse confusion claims. Id. at 236-237. Therefore, we must reconsider our prior determinations concerning: the similarity of the marks ( Lapp factor no. 1); the strength of the marks ( Lapp factor no. 2); the intent of Victoria Secret in adopting the mark ( Lapp factor no. 5); and other relevant factors pertaining to reverse confusion ( Lapp factor no. 10). Id. At the Third Circuit's suggestion, we will discuss each factor in the same order as they appear in the Lapp test. Id. at 216.

Lapp Factor 1: Similarity of The Marks

Although not dispositive, the degree of similarity between marks is one of the most important of the Lapp factors. Fisons, 30 F.3d at 476; A H V, 237 F.3d at 229; FF # 6. Where a trademark owner and alleged infringer deal in competing goods, the court rarely needs to look beyond the mark itself. Marks are confusingly similar if ordinary consumers would likely conclude that the products: "share a common source, affiliation, connection or sponsorship." Fisons, 30 F.3d at 477. To see if this is the case, we look at the sights, sounds and meanings of the marks, and then evaluate whether these elements combine to create the same overall commercial impression. See A H V, 237 F.3d at 229; Fisons, 30 F.3d at 478-79.

When we examined the similarity of the marks in conjunction with Plaintiffs' direct confusion claim we concluded that the distinctions between the marks weighed against a likelihood of confusion. A H IV, 57 F. Supp.2d at 169-70. In reaching this determination, we discussed the use of housemarks by the parties, and we afforded particular weight to Defendants' use of a disclaimer which disavowed any association with Plaintiffs' product. Id. at 168-69. The effectiveness of the housemark and disclaimer changes in the reverse confusion context, requiring us to examine this factor anew. A H V, 237 F.3d at 236.

1. Sight, Sound and Meaning

Consistent with our evaluation of the marks in connection with the direct confusion claim, we find that the MIRACLESUIT mark and THE MIRACLE BRA mark are somewhat distinct in sight, sound and meaning. See A H IV, 57 F. Supp.2d at 168. From a visual perspective, the two marks are distinct because THE MIRACLE BRA mark is presented either in "all capital block letters or in small capital letters with the letters, T, M, B alone in capital letters, while the MIRACLESUIT has sometimes been advertised with the initial letter M capitalized and the rest of the mark in lower-case letters in italicized script or as one word with both the M and S capitalized." A H IV, 57 F. Supp.2d at 167, quoted in A H V, 237 F.3d at 217. In addition, the MIRACLESUIT mark is frequently accompanied by various incarnations of the slogan "Look ten pounds lighter in ten seconds!" A H V, 237 F.3d at 217. The two marks also sound different. Id.; A H IV, 57 F. Supp.2d at 167. They have different numbers of syllables, and the last syllable is different. A H V, 237 F.3d at 217. "MIRACLESUIT bleeds two words together while THE MIRACLE BRA consists of three discrete words." Id. Finally, the meanings of the marks differs slightly. A H IV, 57 F. Supp.2d at 168. The descriptive word "bra" focuses attention on the brassiere portion of the Victoria's Secret swimsuit, while the "suit" portion of Plaintiffs' mark focuses attention on the entire swimsuit. A H V, 237 F.3d at 217, quoting A H IV, 57 F. Supp.2d at 166-7.

The fact that there are differences in what two names suggest does not, by itself, result in a finding that the names are not confusingly similar. Fisons, 30 F.3d at 477. "A subsequent user may not avoid likely confusion by appropriating another's entire mark and adding descriptive or non-descriptive matter to it." Id. quoting 2 McCarthy § 23:15[8]. Fisons concerned the marks Fairway and Fairway Green used on lawn care products. Id. Here, the Defendants did not appropriate Plaintiffs' entire mark. Each mark is composed of different words. Still, the marks share the same dominant portions. Considered in isolation, the distinctions present in the sight, sound, and meaning, are not enough to weigh this factor against finding a likelihood of confusion.

Beyond appearance, sound, and meaning, other things affect how a mark is likely to be perceived in the marketplace. For instance, a party's use of a disclaimer and/or housemark in connection with the mark may affect its general commercial impression. A H IV, 57 F. Supp.2d at 167. Although, the presence of the disclaimer and housemark are weighted less in the reverse confusion paradigm, they remain relevant to the ultimate consideration as to whether the marks can be deemed similar. A H V, 237 F.3d at 236. We will evaluate the effect of Defendants' disclaimer and the use of housemarks by both parties in turn.

2. Disclaimer

In November 1996, Defendants voluntarily began to utilize the following disclaimer: "The Miracle Bra Swimwear Collection is exclusive to Victoria's Secret and not associated with MIRACLESUIT by Swimshaper." In determining that there was no likelihood of direct confusion, we found that this disclaimer created a meaningful difference between the marks. A H IV, 57 F. Supp.2d at 169; A H V, 237 F.3d at 230.

As mentioned above, the Defendants' disclaimer is less effective at combating reverse confusion. A H V, 237 F.3d at 229-30. This is not to say that it is of no utility. Id. Disclaimers can be effective at dispelling consumer confusion even in the reverse confusion context. By definition, for there to be reverse confusion, the consumer would have had to come into contact with the junior user's goods. Typically, in becoming familiar with the Defendants' product, consumers will be exposed to the disclaimer, and this may mitigate some consumer confusion. Id. at 230 ("if consumers are faced with the disclaimer every time they flip through the Victoria's Secret catalogue they are less likely to forget that MIRACLESUIT is unrelated to THE MIRACLE BRA swimwear").

Nonetheless, many consumers will become familiar with Defendants' product without being exposed to the disclaimer. A number of consumers will learn about THE MIRACLE BRA swimwear from sources beyond the control of the Defendants. In addition, no disclaimer policy is flawless. See Defs. Opp'n to A H's Mot. for Recons. Filed 4/6/01. Human and mechanical error will result in occasional gaps in the use of a disclaimer. Id. Even among those who view and understand the disclaimer, a number won't recall it when they view Plaintiffs' mark. We find ourselves unable to conclude that the Defendants' disclaimer creates a meaningful distinction between the marks.

Defendants argue that our finding that the marks were somewhat distinct in terms of sight, sound, and meaning, should lead us to conclude that the marks are not confusingly similar despite the limited utility of the disclaimer in the reverse confusion context. Defs. Br. at 14-15. Plaintiffs counter this argument by pointing to A H IV where we stated: "Thus, solely based on the presumption that Defendants will continue to use the disclaimer when marketing THE MIRACLE BRA swimwear, we find that the two marks are not similar." 57 F. Supp.2d at 169; A H V, 237 F.3d at 237; Reply Br. at 4. These statements are not as contradictory as the parties suggest. In evaluating Plaintiffs' direct confusion claim, we concluded that because there were some distinctions between the marks a disclaimer could effectively combat consumer confusion. When marks are identical and used on competing goods, a disclaimer is rarely going to be enough to weigh this factor in favor of a junior user. On the other hand, when two marks are not identical, a disclaimer can create a meaningful distinction between marks. While there are important distinctions between MIRACLESUIT and THE MIRACLE BRA, these alone are not enough for us to weigh this factor in favor of Defendants. A H IV, 57 F. Supp.2d at 169. The Third Circuit noted our reliance on the use of a disclaimer to draw distinctions between the marks. A H V, 237 F.3d at 229. With reverse confusion, the disclaimer, which is used only on Defendants' goods, does not effectively address the likelihood that consumers will associate Plaintiffs' goods with the Defendants'.

3. Housemarks

Conflicting marks must be compared in their entirety, including any housemark which either party appends to its mark. McCarthy § 23:43. A junior user's addition of a housemark to a possibly infringing mark has the potential to reduce or eliminate the likelihood of confusion. Id. With reverse confusion claims, several courts have taken a hostile stance towards a defendant's use of a housemark; refusing to regard them as mitigating factors. See A H V, 237 F.3d at 230 (discussing this). These courts have found that a defendant's housemark aggravates consumer confusion because it can create a situation in which consumers begin to associate the possibly infringing mark exclusively with the junior user. See, e.g., Sands, Taylor Wood, 978 F.2d at 960, Americana Trading Inc. v. Russ Berrie Co., 966 F.2d 1284, 1288 (9th Cir. 1992).

We accept the established rule that a junior user cannot justify its confusing use of another's mark simply by tacking on its own housemark or tradename. McCarthy § 23:10. Still, a housemark should not, automatically, be considered an aggravating factor. See A H V, 237 F.3d at 230. The better approach is to look at whether the housemark increases or decreases consumer confusion in the particular case. Id. (refusing to adopt a per se rule that housemarks are aggravating factors in the reverse confusion context).

In A H IV, we concluded that the use of the famous Victoria's Secret housemark created "a distinct and separate impression" which diminished the likelihood of confusion. 57 F. Supp.2d at 168-69. We also found that Plaintiffs' use of the SWIM SHAPER housemark helped to create an impression distinct from THE MIRACLE BRA. Id. at 168 n. 17. We will first evaluate the effect of Plaintiffs' housemark, before turning to the more complicated issue of Defendants' housemark.

(a) Plaintiffs' Housemark

From its inception, Plaintiffs' MIRACLESUIT trademark has often appeared in the marketplace followed by the words "by SWIM SHAPER." A H I, 926 F. Supp. at 1248. Hangtags affixed to the MIRACLESUIT swimsuit when it is sold in commerce variously read "MIRACLESUIT® by SWIM SHAPER® — Look 10 pounds lighter in 10 seconds. The ten seconds it takes to slip it on."; or "MIRACLESUIT™ by SWIM SHAPER™ — Look 10 pounds lighter in 10 seconds!" with the MIRACLESUIT silhouette design and the repeated phrase "Look 10 pounds lighter in 10 seconds." Id. Mainstream's press kits promoting MIRACLESUIT, refer to it as "MIRACLESUIT by SWIM SHAPER" or "MIRACLESUIT from SWIM SHAPER." Id. Nevertheless, many advertisements for MIRACLESUIT swimsuits never mention the SWIMSHAPER mark. Id.

We find that Plaintiffs' housemark somewhat mitigates consumer confusion. Some potential purchasers will look at the MIRACLESUIT, notice the distinctive tagging and housemark, and neither Victoria's Secret nor THE MIRACLE BRA will enter their minds. We do not think that the mitigating effect is significant though. MIRACLESUIT is the first word on the hangtag, occasionally used alone in advertisement, and the only word on the sewn in label.⁶ Id. These facts reduce the likely impact of the housemark. See TV Land, L.P. et al. v. Viacom Int'l, Inc. et al., 908 F. Supp. 543 (N.D.Ill. 1995). In TV Land, the Defendants planned to officially refer to their television station as "Nick at Nite TV Land." Id. In evaluating the effect of the "Nick at Nite" housemark, the court took into consideration the fact that it was going to primarily be referred to as "TV Land." Id. We also find that even among those consumers who are exposed to the SWIMSHAPER housemark, there will still be a likelihood of confusion. Some will not recall that Victoria's Secret puts out THE MIRACLE BRA swimwear. Others will not realize that THE MIRACLE BRA products are available only at Victoria's Secret. Thus, we find that although the SWIMSHAPER housemark creates some distinctions between the marks, its impact is not significant enough to weigh Lapp factor one in favor of the Defendants.

6 When this litigation commenced, the labels sewn into the swimsuits read "Swim Shaper." A H I, 926 F. Supp. at 1248. Plaintiffs decided to use a sewn-in tag bearing the MIRACLESUIT name in August or September 1994. Id. Plaintiffs actually began using the new sewn-in label in June 1995, to avoid a change in the middle of the swimsuit season. Id.

(b) Defendants' Housemark

The effect of Defendants' housemark in the reverse confusion context is less clear. Several courts have found that housemarks aggravate the likelihood of reverse confusion. See Americana Trading, 966 F.2d at 1288 (use of a housemark did not, as a matter of law, render marks dissimilar); Sands, Taylor Wood, 978 F.2d at 960 (finding defendant's use of a housemark to be an "aggravation not a justification"); Robert Bruce, Inc. v. Sears, Roebuck Co., 343 F. Supp. 1333, 1347 (E.D.Pa. 1972). Other courts have held that housemarks created useful distinctions between marks, even in the reverse confusion context. See Universal Money Ctr., Inc. v. American Tel. Tel. Co., 22 F.3d 1527, 1531 (10th Cir. 1994) (finding that the AT T housemark created a meaningful distinguish between the marks); Harlem Wizards Entm't Basketball, Inc. v. NBA Prop., 952 F. Supp. 1084 (D.N.J. 1997) (NBA trade name helped to distinguish Harlem Wizards from Washington Wizards); Worthington Foods, Inc. v. Kellogg Co., 732 F. Supp. 1417, 1440 (S.D.Ohio 1990); Packerware Corp. v. Corning Consumer Prod. Co., 895 F. Supp. 1438, 1449 (D.Kan. 1995).⁷

7 In Packerware, the marks were identical but the court weighed the similarity factor in favor of the defendant, in large part due to the use of the well known Corning housemark. 895 F. Supp. at 1449. The court considered both the likelihood of direct and reverse confusion. Id. at 1446.

We do not think that the presence of the housemark on the Defendants product will mitigate consumer confusion. First, there is the possibility that consumers will fail to remember the mark when encountering A H's swimwear. A H V, 237 F.3d at 230. In addition, the housemark reinforces the association of the word "miracle" exclusively with Victoria's Secret. Id. (discussing this possibility). This later concern is of particular importance because it increases the possibility that Plaintiffs' swimsuit could be seen as an imitator of the Defendants' swimsuit. Such a result is unacceptable because Plaintiffs were the first to use "miracle" as part of a trademark for swimsuits.

In reaching this conclusion, we have not come to view the Defendants' housemark as having a singularly negative effect. Its presence does reduce the likelihood of direct confusion. A H IV, 57 F. Supp.2d at 168. It is also relevant that Defendants have not adopted Plaintiffs' entire mark. By this we mean that we are not comparing "Victoria's Secret Miraclesuit" with "Miraclesuit." Cf. Robert Bruce, 343 F. Supp. at 1341. Thus, the negative effect of the Defendants' housemark is not as great as was present in Robert Bruce and Sands Taylor Wood. Despite this, we cannot conclude that the extensive use of the Victoria's Secret housemark creates a meaningful difference between the marks.⁸

8 Even if the Defendants' housemark did not aggravate consumer confusion, we would not weigh this factor in favor of the Defendants. When we concluded in A H IV that the similarity of the marks should not be weighed in favor of finding a likelihood of confusion, we relied on the distinctions in the sight, sound and meaning of the marks, as well as the presence of the disclaimer and the parties' use of housemarks. 57 F. Supp.2d at 166-69. All of these things were integral to our finding. The same is true here. Even if the Defendants' housemark had a mitigating effect on confusion it would not be enough because there is still no effective disclaimer in place, and the distinctions in sight, sound and meaning are not significant.

4. PTO Determination of Mark Similarity

The PTO rejected THE MIRACLE BRA registration for swimwear. While the PTO's adjudication may assist us in deciding whether trademark infringement has occurred, the decision is not binding on this court. A H V, 237 F.3d at 221. In A H IV, we afforded no weight to this adjudication and the Third Circuit held that it was not error to do so. Id. The PTO did not have disclaimer or the voluminous trial record available to it when it made its determination. Nor did the PTO consider the role of housemarks. Plaintiffs argue that since the disclaimer and housemarks play a lesser role in the reverse confusion context, we should now defer to the PTO's decision regarding mark similarity. We are mindful of the general obligation of courts to avoid unnecessary conflict with the PTO. Although we find ourselves reaching the same conclusion as the PTO in connection with our evaluation of Plaintiffs' reverse confusion claim, we did not rely on their opinion to find that the similarity of the marks weighs in favor of a likelihood of confusion. Given the fact that we are in possession of much more information than the PTO had, we continue to rely on our own evaluation of the marks.

5. Conclusion: Overall Commercial Impression

We turn now to the overall commercial impression of the marks. The overall commercial impression considers the sight, sound, and meaning of the marks as well as other indicia demonstrative of the manner in which the marks are presented to the consuming public. If the overall impression created by the marks is "essentially the same, it is very probable that the marks are confusingly similar." Jews for Jesus v. Brodsky, 993 F. Supp. 282, 302 (D.N.J. 1998) quoting Opticians Ass'n of America v. Indep. Opticians of America, 920 F.2d 187, 195 (3d Cir. 1990).

The overall commercial impression of the marks is confusingly similar. The marks are not identical, but they do share the same dominant portion. Analytic Recruiting, Inc. v. Analytic Res., LLC, 156 F. Supp.2d 499, 513-514 (E.D.Pa. 2001). It is proper to give greater force and effect to the dominant feature of a mark. See A H V, 237 F.3d at 216; Country Floors Inc. v. P'ship Composed of Gepner and Ford, 930 F.2d 1056, 1065 (3d Cir. 1991); Checkpoint Sys. v. Check Point Software, 104 F. Supp.2d 427, 457 (D.N.J. 2000); Birthright v. Birthright, Inc., 827 F. Supp. 1114, 1135 (D.N.J. 1993). Consumers are not going to be viewing the marks side by side, and from this perspective, they are more likely to recall the `miracle' portion of the marks. Id.

In the direct confusion context, consumers have the disclaimer and the well known Victoria's Secret housemark readily available to draw distinctions between the marks. These things do not carry the same weight in the reverse confusion context. The Third Circuit noted with approval, our finding that the disclaimer tipped the balance in favor of the Defendants on this factor. A H V, 237 F.3d at 219, 229. When evaluating the similarity of the marks to determine if there was a likelihood of direct confusion, we asked ourselves whether a consumer viewing THE MIRACLE BRA bikini in the pages of the Victoria's Secret catalogue and with a message stating that it was not affiliated with A H or the MIRACLESUIT would be confused as to the source of the swimwear. We concluded that such confusion was unlikely to occur. Now, we are being asked whether a consumer viewing the MIRACLESUIT might conclude that the bathing suit was either a product of Victoria's Secret or an imitation of THE MIRACLE BRA swimwear. In this situation, consumers will not have the benefit of a disclaimer stating that it is not affiliated with THE MIRACLE BRA or Victoria's Secret.⁹ We are mindful of the fact that disclaimers are often ignored by consumers. See Int'l Kennel Club of Chicago, Inc. v. Mighty Star, Inc., 846 F.2d 1079, 1092-93 (7th Cir. 1988). For this reason, a disclaimer can rarely cure otherwise clear cases of likely confusion. Id., Brockum Co. v. Blaylock, 729 F. Supp. 438, 445 (E.D.Pa. 1990); McCarthy § 23:51. The party seeking to avoid a finding of likely confusion on the basis of a disclaimer bears the burden of demonstrating that the disclaimer was or is likely to be effective. E.G.L. Gem Lab Ltd. v. Gem Quality Inst., Inc., 90 F. Supp.2d 277, 297 (S.D.N.Y. 2000), citing Charles of the Ritz Group Ltd. v. Quality King Distrib., Inc., 832 F.2d 1317, 1324 (2d Cir. 1987); Home Box Office, Inc. v. Showtime/The Movie Channel, Inc., 832 F.2d 1311, 1315 (2d Cir. 1987). We find that the Defendants have not met this burden with respect to reverse confusion.¹⁰ Consumers are unlikely to recall Defendants' disclaimer when they view the MIRACLESUIT.

9 Defendants point out that Plaintiffs are free to utilize a disclaimer, hinting that if Plaintiffs really were concerned about being mistakenly affiliated with the Defendants, they could address the problem themselves. As the senior user of a registered mark, Plaintiffs do not have an affirmative responsibility to address confusion caused by a junior user. Boston Athletic Ass'n v. Sullivan, 867 F.2d 22, 29 (1st Cir. 1989) ("When one uses a mark similar to one already in use, there is generally an affirmative duty to avoid the likelihood of confusion"). Moreover, the placement of a disclaimer on Plaintiffs' goods may only exacerbate the likelihood that consumers would view the MIRACLESUIT as an imitator of THE MIRACLE BRA swimwear.

10 We recognize that Plaintiffs bear the ultimate burden of proving the likelihood of confusion. A H V, 237 F.3d at 210. Once a likelihood of confusion has been shown, a defendant can sometimes avoid injunctive relief by showing that a disclaimer will remove the likelihood of confusion.

Nor does the use of housemarks eliminate the likelihood of confusion. Plaintiffs' SWIMSHAPER housemark has not been as ubiquitously or consistently used as the Victoria's Secret housemark. A H I, 926 F. Supp. at 1248. This reduces the housemark's effectiveness at dispelling confusion. TV Land, 908 F. Supp. at 551. Even when the housemark is present, "miracle" remains the dominant portion of the marks making it the focus of consumer attention. See A H V, 237 F.3d at 216 (more forceful and distinctive aspects of marks should be given more weight); Analytic Recruiting, 156 F. Supp.2d at 513-514; Checkpoint, 104 F. Supp.2d at 458 (noting that if the parties consistently used their additional descriptive terms and logos, the similarity between the marks would be more pronounced). As to Defendants' housemark, it actually aggravates consumer confusion because it perpetuates the perception that "miracle" is associated exclusively with Victoria's Secret. See A H V, 237 F.3d at 230.

Given the diminished effectiveness of the disclaimer and housemarks, the distinctions in the sight, sound, and meaning of the marks are insufficient to weigh this factor against a likelihood of confusion.¹¹ Therefore, we will weigh Lapp factor number one in favor of the Plaintiffs.

11 In connection with their contempt motion, Plaintiffs had argued that any lapse in Defendants' adherence to the disclaimer practice it had voluntarily implemented should result in us automatically concluding that a likelihood of confusion existed. A H VI, 134 F. Supp.2d at 672-674. As we stated in response to this motion, an occasional unintentional lapse in the disclaimer practice does not vitiate our conclusion that there is no likelihood of direct confusion. Id. Plaintiffs should not read this statement as having any bearing on that conclusion.

Lapp Factor 2: Strength of the Marks

Trademark strength is a function of a mark's distinctiveness and commercial presence developed through use and promotion. Checkpoint Sys., 104 F. Supp.2d at 458, citing ChiChi's, Inc. v. Chi-Mex, Inc., 568 F. Supp. 731, 736 (W.D.Pa. 1983). The stronger the mark, the greater the amount of protection it is entitled. McCarthy § 11:73. We will first evaluate the parties relative commercial strength. We will then examine the conceptual strength of Plaintiffs' mark.

In A H IV, we considered the parties relative commercial strength. Drawing on then existing precedent and traditional ideas about reverse confusion, we held that because A H had not shown that Victoria's Secret used its strength to overwhelm the marketplace, the doctrine of reverse confusion was not implicated. 57 F. Supp.2d at 176-78. The Third Circuit reversed, concluding that it was error to use an economic disparity test as a necessary prerequisite to finding a likelihood of reverse confusion. A H V, 237 F.3d at 235. Nonetheless, the strength of the marks, in terms of both their commercial and conceptual strength remains a key aspect of the reverse confusion test. Id.

1. Commercial Strength

With reverse confusion, the evidence of commercial strength differs from that expected with direct confusion claims. Fisons, 30 F.3d at 479. With direct confusion, the junior user is getting a free ride on the recognition value of a strong senior mark. Therefore, courts examine the relative strength of the marks to determine whether the senior user's mark is strong enough to cause confusion. In the reverse confusion context, there is no reason to apply such a test. McCarthy § 23:10. Rather, we must evaluate the strength of the junior user's mark so as to gauge its ability to overpower the senior user's mark. Id.; Dreamwerks Prod. Group, Inc. v. SKG Studio, 142 F.3d 1127 (9th Cir. 1998). We do this by looking at: (1) the commercial strength of the junior user vis a vie the senior user; and (2) any advertising or marketing campaign by the junior user that resulted in the public awareness of the junior user's mark. A H V, 237 F.3d at 231; Fisons, 30 F.3d at 474, 479.

(a) Relative Economic Power of the Parties

The parties agree that Victoria's Secret is wealthier than the A H Sportswear, Inc., or Mainstream Swimsuits, Inc. FF # 2. This is not to say that Plaintiffs are without commercial strength. They are not a small corporation operating in a geographically limited area. They manufacture 4 million swimsuits annually, which accounts for approximately 10% of all swimsuits sold in the United States. A H IV, 57 F. Supp.2d at 162. About 200,000 of the swimsuits they manufacture are the MIRACLESUIT. Id. at 175.

(b) Promotion of the Marks

Previously, we concluded that although the Defendants spent considerably more in advertising dollars, both parties had successfully increased public awareness for their products. A H IV, 57 F. Supp.2d at 177. The Third Circuit held that it was error for us not to specifically articulate the amount of free advertising received by the Defendants since we did note the amount of free publicity the Plaintiffs received. A H V, 237 F.3d at 237. The Third Circuit concluded that this omission resulted in us comparing the parties on unequal terms. Id.

Initially, we note that the Defendants did not receive any free publicity for THE MIRACLE BRA swimwear. They did receive free publicity in conjunction with their promotion of THE MIRACLE BRA bra. It has already been established, at trial and on appeal, that the use of THE MIRACLE BRA mark on lingerie did not infringe on Plaintiffs' MIRACLESUIT trademark. A H IV, 57 F. Supp.2d at 179, aff'd, A H V, 237 F.3d at 227. Based on this, it would appear logical to only consider the advertising and promotion for THE MIRACLE BRA swimsuit. Def. Mem. at 17-18. The problem with this approach is that the swimwear leverages off the success of lingerie. THE MIRACLE BRA bra was heavily promoted by Victoria's Secret and, by all accounts, the marketing campaign was a tremendous success. A H IV, 57 F. Supp.2d at 177. THE MIRACLE BRA became "a familiar trademark in a relatively short time." Id. Seeking to capitalize on their popular product, Defendants introduced THE MIRACLE BRA swimwear collection. Id. at 160. Unlike the lingerie, the introduction of the swimwear was not accompanied by extensive advertising or promotion. Instead, since its inception, it has been promoted almost exclusively through its appearance in the Victoria's Secret Catalogue. Defendants neither sought, nor received, any significant free advertising for the swimwear.

Defendants argue that we should ignore the free publicity received for the lingerie. Defs. Mem. at 17-19. They argue that we should only look at the meager amount of expenditures solely attributable to the swimwear. Id. Previously, we rejected such an approach when we examined the paid promotional expenses. A H IV, 57 F. Supp.2d at 177. Instead, we concluded that a "meaningful" portion of the amount spent for lingerie should be considered as benefitting the swimwear. Id. We did not attempt to come up with some formula to try to decipher to what degree the swimwear benefitted from the lingerie promotion; nor have the parties suggested a principled approach to such an undertaking. We find that we should treat the free publicity valuations for THE MIRACLE BRA the same way that we treated the paid expenses in A H IV: we will assume that a meaningful portion benefitted the swimwear.

(1) Paid Promotional Expenses

Plaintiffs requested a hearing and additional discovery on the amount the Defendants spent on advertising since 1996, the last year in which there are figures in the record. Pls. Mem. at 22. We opted to deny this request. A H VI, 134 F. Supp.2d at 681. The Court of Appeals has not required us to revisit the issues presented here as of the present day. A H V, 237 F.3d at 238. Rather, we have been authorized to review the evidence already presented in a new way. Id. Accordingly, we will use the expenditure figures in the record presently before us.

From the launch of THE MIRACLE BRA through 1996, Victoria's Secret Stores spent approximately $9,295,000, and Victoria's Secret Catalogue spent $4,000,000 on advertising. A H IV, 57 F. Supp.2d at 177. These amounts represent the total amount spent on promoting both the swimwear and the lingerie. Id. Since the swimsuit has never been extensively available from Victoria's Secret Stores, we presume that the vast majority of advertising expenditures by the stores went to promote the lingerie.¹² Id.

12 Defendants allege that Victoria's Secret Stores spent only $13,000 on marketing swimwear under THE MIRACLE BRA name. A H IV, 57 F. Supp.2d at 177 n. 31.

Plaintiffs have also conducted extensive marketing campaigns for their product. Id. They have spent approximately $300,000 annually to advertise MIRACLESUIT in various forums. Id. From August 1991 to August 1995, these expenditures totaled $1,244,599. Id.

(2) Comparison of Free Publicity Valuations

Placing dollar values on publicity is an inexact science. Pls. Trial Ex. 315. Typically, experts employ a two step process to determine what it would cost a company to obtain the publicity it has received for free. In step one, the cost of the advertising space or broadcast time is determined. Id. This is known as the "Ad Rate." Id. This number does not properly reflect the value of such publicity because, unlike advertising, publicity is viewed by consumers to be a more accurate and credible source of product information. Id. Logically, this positioning provides greater value to companies than simple advertising. Id. To measure the enhanced value of publicity, experts attach a multiple of between three and ten times to the Ad Rate to come up with the public relations value of the advertising. Id.; A H I, 926 F. Supp. at 1250.

Plaintiffs have received a substantial amount of free publicity. MIRACLESUIT swimsuits have been "featured" or mentioned in hundreds of newspaper and magazine articles throughout the country. A H I, 926 F. Supp. at 1250. A video news release, created by A H, was placed with more than twenty local and five national TV programs. See Pl.'s Trial Ex. 10. Such publicity has an approximate three year residual effect. A H I, 926 F. Supp. at 1250. The publicity campaign for MIRACLESUIT swimsuits was apparently successful because it increased the number of "consumer impressions" from twenty-three million in 1992 to more than 142 million by 1995. Id. at 1250. The public relations value of this publicity is as follows:

---------------------------------------------------------------------- Media Number Ad Rate PR Rate ^† ---------------------------------------------------------------------- Print 83 (out of a total of $476,009.91 $1,430,458.00 335 articles) ---------------------------------------------------------------------- Television 17 (out of 59 spots) $ 83,622.99 $ 250,868.97 ---------------------------------------------------------------------- Radio not available not available not available ---------------------------------------------------------------------- Total $559,632.90 $1,681,326.97 ----------------------------------------------------------------------

† Calculated by multiplying the Advertising Rate by 3. Source: Pls. Trial Ex. 333.

Defendants also received free publicity but comparing it to that received by the Plaintiffs is difficult. As explained above, the publicity focused on the lingerie, not the swimwear. The successful public relations campaign orchestrated by Victoria's Secret Stores, preceded the expansion of the mark into swimwear. As to Victoria's Secret Catalogue, it does not produce press kits and its only advertising is the catalogue itself. A H II, 967 F. Supp. at 1466. To complicate matters further, the parties used different approaches to come up with a monetary value for the free publicity they respectively received. Plaintiffs' media expert only utilized 83 of 335 print articles and 17 of 59 television spots to come up with the advertising value for the MIRACLESUIT's free publicity. The expert also used 3 as the multiplier to translate the "Ad Rate," to the "PR Rate." Meanwhile, the Defendants' valuation for its free publicity it received, took all of the television and radio spots into consideration and used 5 as the multiplier to translate the "Ad Rate" to the "PR Rate."

To address these difficulties, the Defendants suggested that we make some calculations of our own. The Defendants proposed that we extrapolate from the findings of Plaintiffs' expert to determine the advertising value for all the television spots and articles, not just the ones considered by the expert. After coming up with this value, Defendants argue that we should use five as the multiplier to translate the Ad Rate to the PR rate.

While we understand Defendants critique of the valuation approach utilized by Plaintiffs' expert, we are reluctant to step beyond what are properly matters for judicial interpretation and notice. As to the different multipliers used by the parties, neither side's expert provides a justification for the multiplier selected. The report prepared by Plaintiffs' expert simply states that it chose "the most conservative multiple of three times." Pls. Trial Ex. 315 at 6.¹³ Defendants do not state why it utilized five as the multiplier. In light of the lack of evidence supporting either choice, we will not criticize the approach taken by either side. Instead, we will take the discrepancy into account when we compare the amounts of free publicity received by both sides. As to Defendants other concern, we are reluctant to treat all of the articles and television spots the same. Plaintiffs' expert stated that it focused on the "top" instances of free publicity. Pls. Trial Ex. 315 at 6. It makes sense that certain articles and television spots are going to be of greater value than others. This being said, we recognize that the total value of the free publicity received by Plaintiffs was higher than the amount stated in the expert report. We note that the same is true for Defendants. Its evidence of the value of free publicity it received did not include an amount for print media. Defs. Mem. at 22. Despite our inability to compare the parties free publicity in strict quantitative terms, we are confident in our ability to make a qualitative comparison.

13 Attached to Defendants' Memorandum at Tab 7.

The public relations valuation totals for THE MIRACLE BRA lingerie are as follows:

-------------------------------------------------------------- Media Number^_ Ad Rate PR Rate^† -------------------------------------------------------------- Print 224 articles not available not available -------------------------------------------------------------- Radio 78 spots $ 228,040.80 $1,140,204.00 -------------------------------------------------------------- Television 54 spots $ 493,975.80 $2,469,879.00 -------------------------------------------------------------- Total $1,140,204 $3,610,083.00 --------------------------------------------------------------

_ Every spot used to calculate the Ad Rate

† Calculated by multiplying the Ad Rate by 5. Source: Defendants Exhibit 23; Defendants Mem. at 22.

There are two gaps in the free publicity evidence: (1) there is no valuation for the free publicity Defendants received as a result of print articles mentioning the product; (2) there is no information on whether Plaintiffs benefitted from free publicity on the radio. Despite the lack of precise symmetry in the evidence produced by both sides, it is apparent that both parties received substantial amounts in free advertising. Plaintiffs were not overwhelmed by Defendants advertising. They clearly held their own on the public relations front. MIRACLESUIT was mentioned in more articles and seen in more television spots. Although we may not be able to say for certain who received more in terms of free publicity, we are confident in concluding that both parties benefitted greatly from good PR.

(3) Overall Evaluation of the Parties Promotional Campaigns

Consistent with the Third Circuit's instructions, we re-evaluated the promotional campaign undertaken by the Defendants. We find that, in light of the advertising campaign engaged in by Plaintiffs, Defendants did not saturate the marketplace with its advertising to promote THE MIRACLE BRA swimwear.

In a strict numerical sense, the Defendants spent more on promoting THE MIRACLE BRA than the Plaintiffs spent promoting MIRACLESUIT. Defendants spent $13.4 million promoting THE MIRACLE BRA mark on both lingerie and swimwear and received in excess of $3.6 million in free publicity. Plaintiffs spent $1,244,599 to promote the MIRACLESUIT mark and received at least $1,681,326 in free publicity. A cursory glance at such numbers suggests a significant disparity. The gap is not nearly so great once several factors are taken into consideration. First, only a small portion of the amount spent by Defendants directly benefitted the swimwear. Second, Plaintiffs took a more conservative approach in valuing the free publicity it received.

We continue to believe that the disparity in economic power and advertising expenditures in the case at bar "does not even begin to approach the disparity that exits in other reverse confusion cases." A H IV, 57 F. Supp.2d at 178, relying on, Big O Tire Dealers Inc. v. Goodyear Tire Rubber Co., 561 F.2d 1365, 1367 (10th Cir. 1977) (plaintiff has net worth of $200,000 versus defendant who spent $10 million on advertising); Sands, Taylor Wood v. Quaker Oats Co., 18 U.S.P.Q.2d 1457, 1458, 1474 (N.D.Ill. 1990) (plaintiff had $200,000 net worth while defendant spent $365 million to advertise over a six-year period), aff'd in part and rev'd in part, 978 F.2d 947 (7th Cir. 1992); Fisons, 30 F.3d at 470-71 (plaintiff spends nominal amounts on advertising while in its first few months defendant had spent over $500,000 on advertising); Americana Trading, 966 F.2d at 1287 (not discussing the difference in numerical terms, but there were obvious and apparent differences in the economic power of the senior user vis a vie the junior user); Banff, 841 F.2d at 487-490 (same).

Including the free publicity accorded to Victoria's Secret THE MIRACLE BRA does not lead us to the conclusion that the Defendants overwhelmed or wiped out the market strength of the MIRACLESUIT mark. A H spent considerable sums on advertising and received the benefit of free publicity. Its swimwear sales have steadily increased over time and the company continues to enjoy a significant share of the swimsuit market.

2. Conceptual Strength

In terms of conceptual strength, the stronger the mark is, the greater the amount of protection it receives. Ameritech, 811 F.2d at 966; McCarthy § 11:73. A senior user of a conceptually weak mark will receive less protection than the senior user of a conceptually distinct mark. See A H V, 237 F.3d at 231. Courts use a four part sliding scale scheme to classify marks as to their distinctiveness: (1) generic; (2) descriptive; (3) suggestive; or (4) arbitrary or fanciful. Marks that are suggestive, arbitrary, or fanciful are deemed inherently distinctive and are entitled to protection upon adoption. Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 768, 112 S.Ct. 2753, 120 L.Ed.2d 615 (1992). This classification process, in addition to being used by courts to determine whether a mark is entitled to trademark protection, is also a "useful guide in assessing the strength or weakness of a mark." A H V, 237 F.3d at 222, citing Express Servs., 176 F.3d at 186. A mark's classification is not dispositive of its conceptual strength or in whose favor Lapp factor two should be weighted. Id.

The Third Circuit has determined that the MIRACLESUIT mark is "at best, merely suggestive." A H V, 237 F.3d at 223.¹⁴ The record contains several examples of trademarks in actual use for apparel, fashion and cosmetic items that incorporate the word "miracle" as a part of their trademark or tradename. A H I, 926 F. Supp. at 1258. The term is not widely used within the swimsuit market; Plaintiffs and Defendants being the only ones to do so. Id. The mark has been used in other fields, which can affect its conceptual strength. See A H V, 237 F.3d at 223 relying on, Sun Banks of Florida, Inc. v. Sun Federal Savings Loan Ass'n, 651 F.2d 311, 316-17 (5th Cir. 1981); Triumph Hosiery Mills, Inc. v. Triumph Int'l Corp., 308 F.2d 196, 199 (2d Cir. 1962). A H V suggests that we could conclude that the MIRACLESUIT mark was descriptive. 237 F.3d at 223. We think that suggestive is the proper categorization.¹⁵ That Plaintiffs' mark is not descriptive is shown inferentially by the fact that it is registered with the U.S. Patent and Trademark Office on the Principal Register, which is reserved for distinctive marks. Id. Section 7(b) of the Lanham Act is tracked by Section 33(a) 15 U.S.C. § 1115(a) which provides that registration on the Principal Register "shall be prima facie evidence" of the validity of the registered mark, of its registration, of the registrant's ownership, and of the registrant's exclusive right to use the mark on the goods specified in the registration. Courts have interpreted this section to mean that a party: "is entitled to a strong prima facie presumption that its registered mark is either not merely descriptive or if descriptive, that secondary meaning is presumed, which amounts to the same thing." McCarthy § 11:43. See e.g., Playboy Enter., Inc. v. Chuckleberry Publ'g, Inc., 687 F.2d 563, 567 (2d Cir. 1982) (PTO registration "indicates that the mark is not merely descriptive and gives to it a strong presumption of validity").

14 In A H I we determined that the MIRACLESUIT mark was in the arbitrary/fanciful category and entitled to the highest level of protection. Neither the original three judge panel nor the en ban panel over-ruled this determination. See A H Sportswear v. Victoria's Secret Stores, 166 F.3d 191, 195 (3d Cir. 1999) (panel decision discussing our classification of the mark as fanciful); A H Sportswear v. Victoria's Secret Stores, 166 F.3d 197 (3d Cir. 1999) (en banc decision). When we reviewed the matter in connection with the first remand, we specifically considered each portion the MIRACLESUIT mark. We noted our previous determination that the use of the word "miracle" on swimwear was fanciful because it revealed nothing about the product. The second part of the mark, "suit", was generic. When these two aspects of the mark were combined, we concluded that the descriptiveness of the mark ranged from suggestive to arbitrary. The Third Circuit held that this determination was clearly erroneous. A H V, 237 F.3d at 237.

15 The Third Circuit left the proper categorization to us: "the court must gauge the strength of the MIRACLESUIT mark." A H V, 237 F.3d at 237. Defendants have not argued that MIRACLESUIT is a generic mark and there is no basis in the record for such a conclusion.

In a literal sense, MIRACLESUIT cannot be considered as descriptive of the swimwear. Making a person appear slightly thinner is not a miracle. See Checkpoint Sys., 104 F. Supp.2d at 427, 459 ("Checkpoint" has no descriptive significance when used in connection with the products at issue). The Third Circuit suggested that "miracle" may fall into a class of terms which, although technically fitting into the arbitrary or fanciful category, should actually be considered to be descriptive or generic. A H V, 237 F.3d at 231-32. This can occur when a suggestive or arbitrary mark is used in connection with a number of different product. Id. at 222. The example given H. Lubovsky, Inc. v. Esprit de Corp., 627 F. Supp. 483 (S.D.N Y 1986) is particularly instructive. The court explained: "if a customer saw a doll in a toy store bearing a strong familiar mark like `Exxon,' he might well assume that the oil company had gone into the toy business; if, on the other hand, he saw a doll bearing a familiar but weak laudatory trademark like Merit, hie would be unlikely to assume that it is connected with the similarly named gasoline or cigarettes." Id. at 487; see also A H V, 237 F.3d at 231 (discussing same).

We have repeatedly evaluated the distinctiveness of Plaintiffs' mark and find that "miracle," as applied to swimwear, should not be considered such a term. "Miracle" is not among the 5,000 most commonly used English words. A H I, 926 F. Supp. at 1259. Its use in the apparel field is limited. Id. Even if "miracle" was considered a common word, "the significant factor is not whether the word itself is common, but whether the way the word is used in a particular context is unique enough to warrant trademark protection." Id. quoting Fisons, 30 F.3d at 478. In Henri's Food Products Co., Inc. v. Kraft, Inc., 717 F.2d 352 (7th Cir. 1983), the seventh circuit found that the word "miracle" is fanciful, because it reveals nothing about the product. Id. at 355. We find the same to be true here, "miracle" does not reveal anything about the swimwear. A H IV, 57 F. Supp.2d at 164. In addition, when the Defendants registered their mark, the PTO did not require them to file a disclaimer¹⁶ of the word "miracle" in the trademark registration. A H I, 926 F. Supp. at 1248. This amounts to their recognition that "miracle" as a part of the phrase THE MIRACLE BRA is a protectable word. Id.; McCarthy § 11:52.

16 Under § 6 of the federal Lanham Act, the Commissioner has the power to require an applicant to disclaim rights in descriptive portions of a composite mark. McCarthy § 11:52. Because of this rule, trademark protection can be afforded to phrases composed otherwise generic words. Id. ("Even if a descriptive portion of a composite mark has been disclaimed, the total composite will be considered for purposes of determining infringement)"; McCarthy § 11:26.

This case is distinguishable from Sun Banks. There, the Fifth Circuit gave special weight to fact that 25 competing financial institutions used the word sun in their titles. 651 F.2d at 316-17. In a footnote, the court also noted the widespread use of the word "sun" in the names of other Florida businesses. 651 F.2d at 317 n. 8. The number of other products utilizing the word "miracle" as all or part of their trademarks, is a far cry from the 4,400 Florida businesses which were found to use the word sun in their titles. See id.

Nor do we find cases involving common given names or surnames persuasive. See A H V, 237 F.3d at 223-24 discussing Steve's Ice Cream v. Steve's Famous Hot Dogs, 3 U.S.P.Q.2d 1477, 1479, 1987 WL 124289 (T.T.A.B. 1987), and S.C. Johnson Son, Inc. v. Johnson, 116 F.2d 427, 430 (2d Cir. 1940). When people choose to use their name for their business or product, they run the risk of another person with the same name deciding to do the same thing. Id. Accordingly, when surnames or given names are used, courts should be more cautious before finding that a mark is entitled to a high level of protection. Id.

In the case at bar, there was "only limited evidence concerning the use of `miracle' marks in the marketplace." A H I, 926 F. Supp. at 160. The products which were pointed to as having "miracle" as a component of their trademarks were "virtually unrelated to either lingerie or swimwear." Id. We stated that these marks and the goods covered by them "in no way approached the degree of similarity shared by the MIRACLESUIT and THE MIRACLE BRA." Id. Therefore, we concluded that the use of the word "miracle" when applied to swimwear was a strong mark. Id.

In A H IV, we re-iterated our conclusions concerning the strength of Plaintiffs' mark. 57 F. Supp.2d at 164. We discussed the seventh circuit case which deemed the word "miracle" to be fanciful. Id. We evaluated third party use of similar marks, concluding that the evidence was not significant enough to warrant further downgrading in the level of distinctiveness we attributed to Plaintiffs' mark. Id. at 165 n. 16. Ultimately, we concluded that the mark ranges from being suggestive to arbitrary and afforded it "a high level of trademark protection."¹⁷ A H IV, 57 F. Supp.2d at 165. The Third Circuit stated that we erred because we did not consider evidence of the use of the miracle mark in other markets. A H V, 237 F.3d at 223-24. We have now considered this evidence and we respectfully do not believe that this precludes us from reasserting the finding we made in A H I that Defendants' evidence of the use of the word "miracle" as part of the trademark for other products was unconvincing. Accordingly, we find that the MIRACLESUIT mark is suggestive, which places it at the mid point on the range of descriptiveness.

17 At times, the Third Circuit stated that we concluded the mark should receive the "highest" level of protection. A H V, 237 F.3d at 223.

3. Conclusion

Lapp factor two favors the senior user when there is a commercially strong junior user who adopts a conceptually strong mark. This is not such a case because both parties have significant commercial strength and the mark is at the mid-point on the range of descriptiveness. We recognize that it is a close call, but ultimately we must conclude that this factor weighs against finding a likelihood of confusion.¹⁸

18 If we had concluded that the MIRACLESUIT mark was entitled to a low level of protection, this factor would have weighed more heavily in Defendants favor. Such a change in degree, however, would not have effected our overall conclusion in this case.

Plaintiffs argued that simply because it was the weaker economic unit, we should weigh this factor in their favor. A H V establishes that courts should examine the relative strength of the marks by a two part balancing test. 237 F.3d at 230-32. The first part of the test evaluates relative commercial strength. Id. at 230-31. The concern is about whether a senior user has been overwhelmed by an aggressive junior user. Id. We find that A H has consistently promoted its mark and enjoys considerable commercial strength in the relevant swimsuit market. In stating this, we do not mean to suggest that the onus is on the trademark owner to undertake extraordinary efforts to defend against aggressive junior users. Boston Athletic Ass'n, 867 F.2d at 29. Generally, there is a higher burden on the second user to accommodate a smaller, prior use. Id. Nonetheless, in this case, we have concluded that the commercial disparity between the parties, coupled with the conceptually weak nature of the mark, reduces the likelihood of confusion. The Defendants did not use their economic power to overwhelm the market with advertising of THE MIRACLE BRA swimwear. A H IV, 57 F. Supp.2d at 178.

Accordingly, we will weigh Lapp factor two in favor of the Defendants.

Lapp Factor 3: Attention Expected

In the reverse confusion context, courts must examine whether the product is the type that consumers will care enough about to notice the differences between the products or is it the type of purchase hastily made with only a limited impression. See Fisons, 30 F.3d at 476 n. 12. Consistent with our previous determination, we find that this factor weighs against finding a likelihood of confusion. A H IV, 57 F. Supp.2d at 171-172; A H V, 237 F.3d at 225. The record contains undisputed evidence showing that buyers of women's apparel are sophisticated purchasers. Id. Such purchasers are likely to take care before they make a purchase. The premium price of both swimsuits also suggests that consumers are likely to pay attention before they buy either Plaintiffs or Defendants swimsuits. Id. Because the product in issue is not the type that unsophisticated consumers are likely to hastily purchase, Lapp factor three favors the Defendants.

Lapp Factors 4 and 6: Actual Confusion

The fourth factor asks us to look at the length of time the mark was used without actual confusion arising. The sixth factor concerns whether a plaintiff can show that consumers have actually been confused by the marks. Because these factors are intrinsically related, we will discuss them together. See Ziff Davis Media, 145 F. Supp.2d at 422.

The Third Circuit stated that we "may, but need not" consider these factors on remand. A H V, 237 F.3d at 236. We find that there is no need to re-examine the actual confusion evidence in this case. With the exception of one article, A H's evidence of actual confusion supported its claim for direct confusion.¹⁹ We previously concluded that this evidence was too weak to support even that claim. A H IV, 57 F. Supp.2d at 178. The Third Circuit agreed with our conclusion that A H put forth insufficient credible evidence of instances of actual confusion. A H V, 237 F.3d at 227. Accordingly, we will weigh Lapp factors four and six in favor of the Defendants.

19 Plaintiffs claim that they have evidence of four people being actually confused. As we stated in A H VI, we refuse to open up the record to allow for the submission of additional evidence. 134 F. Supp.2d at 681. We re-iterated this point on Plaintiffs' Motion for Reconsideration. Then, on July 12, 2001, we struck the documents which Plaintiffs submitted that were outside of the record and constituted an attempt to introduce new evidence.
Even if we were to consider this evidence, it would not have persuaded us to weigh this factor in Plaintiffs' favor. The evidence consisted of four print-outs of responses to a question posted on the Cyberswim website. Cyberswim exclusively sells Miraclesuit swimwear. The question asked shoppers "`Miraclesuit' is made by what company?" Plaintiffs submit four responses, all dated as occurring in 2001. One person responded by writing: "miracle bra?" The second response read: "The company that pattened (sic) the `Miracle Bra'?" The third response said: "Victoria's Secret?" And the fourth response was: "not sure, but Victoria's Secret?" All of the responses ended with a question mark. Only one of these could really be considered showing an instance of actual reverse confusion.
It is well established that trademark ownership does not guarantee freedom from all confusion. The test is likelihood of confusion, not the possibility that some stray consumer might be confused.
Again, we wish to make plain our decision to reject these materials. Defendants, who abided by this Court's directions and orders, have not had the opportunity to respond to this information. We write here merely to make plain to Plaintiffs that this extraneous information would not have been useful or persuasive had we wished to open up the record.

Lapp Factor 5: Intent

Ordinarily, the relevant inquiry in a trademark infringement case involves determining: "whether the defendant adopted a mark with the intent of promoting confusion and appropriating the prior user's good will." Fisons, 30 F.3d at 479. Intent remains relevant in the reverse confusion context, albeit in a different format. In reverse confusion cases, the junior user typically does not attempt to take advantage of the senior user's goodwill and reputation; instead the junior user seeks to overwhelm it. Id. Therefore, the appropriate inquiry is an examination of whether the junior user intended to usurp the goodwill of the senior user by deliberately seeking to cause consumer confusion and thereby destroy senior user's business identity. A H V, 237 F.3d at 237. Mere carelessness in adopting the mark is not enough: "in light of the policy concerns implicated by the reverse confusion doctrine, it would be troubling to hold that a lesser degree of culpability would weigh in the plaintiff's favor for a reverse confusion claim than it would for a direct confusion claim." A H V, 237 F.3d at 233. The Third Circuit directed us to consider whether our previous finding of good faith on the part of the Defendants is dispositive of the reverse confusion intent analysis, or whether further examination of this issue is needed. Id. at 237.

Under the circumstances here, we find that no further review of the Defendants' intent is necessary. Defendants expanded the use of their mark for legitimate reasons. A H I, 926 F. Supp. at 1260-61; A H IV, 57 F. Supp.2d at 172; A H V, 237 F.3d at 226. It was not their intention to usurp Plaintiffs' goodwill; rather, they sought to "leverage off" the success of THE MIRACLE BRA lingerie line. See A H I, 926 F. Supp. at 1243; A H II, 967 F. Supp. at 1237; A H V, 237 F.3d at 226 (affirming the determination that that Victoria's Secret brought the mark into swimwear because of its success in lingerie). Plaintiffs' arguments to the contrary remain unconvincing.

Plaintiffs argue that they should be allowed to present additional evidence of bad faith which occurred after our determination in 1996. Pls. Mem. at 27. They have not shown how such evidence is probative of Defendants' intent when it adopted the mark. More importantly, none of the facts cited by Plaintiffs indicate bad faith or an intent to destroy their business reputation.

First, Plaintiffs attempt to use our decision in A H VI as evidence of Defendants' alleged bad faith. Id. In connection with that motion, Defendants argued, in the alternative, either that they did not have to utilize a disclaimer on the website, or that the practice they did follow was consistent with the orders of this court. A H VI, 134 F. Supp.2d at 673. Although we rejected the argument that disclaimers did not have to be used on the website, Defendants putting forth such a suggestion is not evidence of bad faith.

Plaintiffs next cite the PTO's determination of the marks' similarity as evidence of bad faith or improper intent. Pls. Mem. at 27. We note that this decision occurred after the pendency of this action. As discussed above, the PTO's determination of the similarity of the marks is not controlling and, under the circumstances of this case, we do not consider it persuasive. Since we ourselves disagreed with the PTO's decision, we are hard pressed to say that Defendants' failure to stop using the mark as a result of this decision is a sign of bad faith. See A H IV, 57 F. Supp.2d at 173.

Plaintiffs then point to the fact that Defendants expanded the use of the mark after the commencement of this lawsuit. Pls. Mem. at 27. Again, we find no reason to draw an inference of bad faith from this. See SecuraComm Consulting Inc. v. Securacom Incorporated, 166 F.3d 182, 189 (3d Cir. 1999) ("A defendant's refusal to cease using a mark upon demand is not necessarily indicative of bad faith"). This case is far from being clear cut. Prior to this case, trademark law in this Circuit was in a state of flux. See A H III, 166 F.3d at 203-7. In A H I, we found that the Defendants' mark when used on lingerie did not constitute trademark infringement. 926 F. Supp. at 1269. As to its application to swimwear, we concluded that there was no likelihood of confusion, finding only a possibility of confusion. Id. at 1265-68. After the Third Circuit eliminated the possibility of confusion standard, the case was remanded to us. A H III, 166 F.3d at 209-10. On remand, under a different test, we did not believe that Plaintiffs had proven their claim of reverse confusion. See A H IV, 57 F. Supp.2d at 178. In light of the state of the case law at the institution of this lawsuit and the closeness of the issues dividing the parties, Defendants' continued use of the mark on swimwear after the commencement of this litigation does not show bad faith or an intent to destroy Plaintiffs' business reputation. See SecuraComm, 166 F.3d at 188, quoting, Blockbuster Videos, Inc. v. City of Tempe, 141 F.3d 1295, 1300 (9th Cir. 1998) ("[i]nfringement is not willful if the defendant might have reasonably thought that its proposed usage was not barred by the statute"). We recognize that continued use of a mark in the face of litigation will not always be seen as innocence or good faith. For instance, in Big O Tire Dealers, Goodyear clearly expressed an intent to drag out litigation while it utilized the plaintiff's trademark. 408 F. Supp. at 1232. Goodyear all but admitted that it knew it was wrongly using the mark and that it intended to use litigation as a weapon so it could prolong the amount of time it had to do so. Id.

Plaintiffs also argue that we should make a finding of bad faith because Defendants continued to use the mark despite instances of actual confusion. Pls. Mem. at 27 As stated above, there is no strong evidence of actual confusion. We were not convinced by the meager evidence Plaintiffs put forth. Defendants apparently viewed the evidence in much the same way as we did: not convincing enough to warrant the cessation of selling swimwear with THE MIRACLE BRA mark.

Finally, Plaintiffs argue that we should weigh the intent factor in favor of finding a likelihood of confusion because the Defendants had knowledge of the MIRACLESUIT mark prior to expanding THE MIRACLE BRA line into swimwear. Pls. Mem. at 27-30, 33; FF 4. Plaintiffs misquote a passage from A H I in which we cite Syntex v. Norwich, 315 F. Supp. 45 (S.D.N.Y. 1970), as support for their proposition.²⁰ Pls. Mem. at 33. In A H I, we concluded that Defendants' actions were not willfully tortious. A H I, 926 F. Supp. at 1255. Instead, we surmised that the actions, "at most," amounted to a decision to run the risk of "coming very close to infringement" in order to utilize THE MIRACLE BRA mark on swimwear. A H I, 926 F. Supp. at 1255 (emphasis added). It is important to point out that this opinion discusses the Defendants' liability for damages under the possibility of confusion standard. It does not address the likelihood of reverse confusion presently before us. As to Defendants' prior knowledge of the MIRACLESUIT mark, Plaintiffs improperly analogize this case to the situation in Syntex. See Pls. Mem. at 33. The marks in Syntex were virtually identical, Vagitrol versus Vagestrol, but the products were for the treatment of different medical conditions. 315 F. Supp. at 52. In evaluating the defendant's intent in adopting the mark, the court considered the fact that the defendant had knowledge of both the existence of plaintiff's mark and the PTO's determination that the marks were confusingly similar. Id. The court was displeased with the defendant's actions and refused to find that it was an innocent junior user. Id. Nonetheless, the court specifically stated that it could not characterize the actions as "wilfully tortious." Id. The court did not explicitly weigh the intent factor in favor of the granting the preliminary injunction. Id. We find that the factual issues in Syntex are distinct from the issues before us. In particular, only two letters differed between the marks and they sounded virtually the same. Id. at 50-53. Furthermore, there was a public health risk and the court concluded that such a threat warranted more caution in the selection of a product name. Id. at 52-53. Finally, the defendant in Syntex knew of the PTO's determination of the marks prior to its adoption of the confusingly similar mark. Despite these facts, which show more egregious conduct than Victoria's Secret is guilty of, the court was still unwilling to condemn the defendant's actions. Id. at 52. Syntex does not persuade us to weigh the intent factor in favor of the Plaintiffs. We quoted the case in A H I, not because the facts were identical, but because, "at most", Victoria's Secret's actions could be considered risky, not willfully tortious. See Ziff Davis Media, 145 F. Supp.2d at 437 (knowledge of competitor's similarly named product prior to adoption of the mark was not enough to weigh the intent factor in favor of finding a likelihood of confusion).

20 Plaintiffs misquote and take a statement completely out of context in their Memorandum. Pls. Mem. at 33. A H I states: "In our consideration of this case, we have not come to believe Defendants' actions with respect to Plaintiffs' MIRACLESUIT were willfully tortious. Rather, we surmise that these actions at most amounted to a `decision to take the risk of coming very close to infringement, for the sake of a trade name defendant evidently believed would be more valuable to it than any of the available alternatives.' Syntex Laboratories, Inc. v. Norwich Pharmacal Co., 315 F. Supp. 45, 52 (S.D.N.Y. 1970), aff'd, 437 F.2d 566 (2nd Cir. 1971) (citations omitted)." A H I, 926 F. Supp. at 1255. In their Memorandum, Plaintiffs quoted this passage as reading "The Defendants made the `conscious decision to take the risk of coming very close to infringement for the sake of a trade name defendant evidently believed would be more valuable than any of the available alternatives.'" Pls. Mem. at 33. Plaintiffs leave out the words "at most" and insert the word "conscious." They then go on to state that we incorrectly "inferred innocense (sic) from this scenario." Id. According to Plaintiffs, the court in Syntex found something "more closely resembling `bad faith' or `intent'" when it evaluated similar facts. Id. We disagree with both their reading of A H I and Syntex.

Despite their awareness of Plaintiffs' mark, the Defendants' intent is distinguishable from that found in Commerce National and Big O Tire Dealers, Inc. v. Goodyear Tire Rubber Co., 408 F. Supp. 1219, 1232 (D.Colo. 1976). In Commerce National, the larger company adopted the mark with the deliberate intent of pushing its rival out of the market. 214 F.3d at 444. Similarly, in Big O Tire Dealers the defendant was aware of plaintiff's mark and the potential conflict because it unsuccessfully sought a letter of consent. 408 F. Supp. at 1232. Nonetheless, Goodyear launched an advertising campaign for the identical mark due to significant amount of money invested and Goodyear's belief that could out litigate the plaintiff. Id. Here, it is apparent that Defendants did not deliberately undertake to cause consumer confusion or destroy Plaintiffs' business when they adopted THE MIRACLE BRA mark and later expanded it to swimwear. See A H V, 237 F.3d at 232 (distinguishing Commerce National from this matter); Ziff Davis Media, 145 F. Supp.2d at 433 (clear knowledge of other mark was not enough to demonstrate an intent to confuse).

Having found no evidence of an intent to confuse or push Plaintiffs out of the swimsuit market, Lapp factor five weighs against finding a likelihood of confusion and favors the Defendants.

Lapp Factor 7: Channels of Trade

In A H IV, in connection with the direct confusion claim, we concluded that the MIRACLESUIT swimwear and THE MIRACLE BRA swimwear were sold through overlapping channels of trade. 57 F. Supp.2d at 170. The products of both companies are available nationwide. MIRACLESUIT is sold to department stores and catalogues that compete with Victoria's Secret stores and the Victoria's Secret Catalogue. Id. at 171. A woman looking for a swimsuit might pursue a variety of retailers and could conceivably view both swimsuits before making a purchasing decision. Id.

Despite these similarities, the channels of trade are not identical. With the exception of three catalogues issued in 1992, the MIRACLESUIT has not been available in the same store or catalogue as THE MIRACLE BRA swimwear.²¹ A H IV, 57 F. Supp.2d at 171 n. 22. Further, the MIRACLESUIT is available from a variety of retailers, while THE MIRACLE BRA swimsuit is available exclusively through Victoria's Secret. Id. Defendants offered evidence about the length it goes to cultivate a distinctive exclusive image. All stores have a distinctive trade dress and aspects of this design are carried over into the catalogue layouts and overall appearance. Defendants do not sell products with their housebrand to other retailers or to discount stores. These steps reduce the likelihood that a consumer would think that a department store now carried Victoria's Secret products.

21 The third catalogue issued was identical to the one prior to it with the exception of the cover. A H I, 926 F. Supp. at 1242.

On balance, we find that the channels of trade are similar enough to weigh this factor in favor of the Plaintiffs. This is a close call; closer than in the direct confusion context because the differences in the channels of trade reduce the harm that could result if consumers wrongly concluded that the MIRACLESUIT is a product of Defendants. Were this our only concern we would weigh this factor in favor of the Defendants. However, reverse confusion is aimed at redressing a second potential wrong: the harm that could result if consumers view the senior user as being an imitator of the junior user. A consumer familiar with THE MIRACLE BRA swimwear could think that the MIRACLESUIT was an imitator. The distinctions in the channels of trade are not so great as to address this aspect of reverse confusion. Accordingly, for Lapp factor seven, we think that the balance tips in favor of finding that there is a likelihood of confusion, which favors Plaintiffs.

Lapp Factor 8: Targets of Sales Efforts

Under the eighth Lapp factor, a court must consider whether the targets for the parties' sales efforts are the same. Both the MIRACLESUIT and THE MIRACLE BRA target women nineteen and older who are willing to pay a premium price for a swimsuit. A H IV, 57 F. Supp.2d at 170. The targets differ slightly in that the primary function of THE MIRACLE BRA swimwear is cleavage enhancement while the primary function of the MIRACLESUIT is to make the wearer look slimmer. Id. Thus, consumers who are more concerned about their bustline may prefer THE MIRACLE BRA swimwear while those concerned with their waistline may prefer the MIRACLESUIT. This is still a small distinction. Both companies focus on improving physical appearance. No reasonable profit minded swimsuit maker is going to try to produce swimwear that makes the wearer look worse than they really are. Furthermore, a woman may decide not to purchase a MIRACLESUIT if she confuses it with the cleavage enhancing features of the Defendants' swimsuit. Overall, we find that the similarity in the targets of the sales efforts weighs in favor of finding a likelihood of confusion. Thus, Lapp factor eight favors Plaintiffs.

Lapp Factor 9: Relationship of the Goods in Consumer's Minds

This factor requires us to consider whether the goods offered by the parties are similar enough in their functions that a consumer could assume they were offered by the same source. Fisons, 30 F.3d at 481. We find that consumers are likely to view the products similarly. Both are swimsuits which come in two styles: bikini and one-piece. A H IV, 57 F. Supp.2d at 170-171. Both products are directed at a similar segment of the public and both market themselves as being figure enhancing and addressing perceived body flows. Id.

Granted, there are some noticeable differences in the underlying properties of the swimsuits. Id. See Richards v. Cable News Network, Inc., 15 F. Supp.2d 683, 690 (E.D.Pa. 1998) (no similarity of goods where plaintiff used the mark to sell reggae music and defendant used it as the title of a news program); Harlem Wizards, 952 F. Supp. 1084 (services offered by show basketball team and professional basketball team distinctive). The MIRACLESUIT offers body control and claims to make the wearer look slimmer through horizontal control and vertical stretch. A H IV, 57 F. Supp.2d at 170. THE MIRACLE BRA swimwear provides cleavage enhancement. Id. A single retailer has never offered both products for sale at the same time.²² Id. at 171. Finally, the use of housemarks also makes it less likely that consumers would conflate the goods. See Ziff Davis Median, 145 F. Supp.2d at 439.

22 Prior to the launch of THE MIRACLE BRA swimwear, the MIRACLESUIT was offered for sale in three issues of the Victoria's Secret Catalogue. Victoria's Secret is the only retailer who could conceivably offer both products for sale. While the company does not sell its products to other retailers or discounters, it does sell products from other manufacturers. Thus, a department store, the typical retailer of Plaintiffs' products, would not be able to offer both the MIRACLESUIT bathing suit and THE MIRACLE BRA swimwear.

These distinctions, while important, ultimately are not enough to remove the likelihood that consumers are likely to view the products similarly. See id. ("the MIRACLESUIT and THE MIRACLE BRA swimwear products could be considered somewhat interchangeable"). Therefore, we find that Lapp factor nine weighs in favor of finding a likelihood of confusion, which favors Plaintiffs.

Lapp Factor 10: Other Relevant Facts

In the reverse confusion context, this catch all factor requires: "an examination of other facts suggesting that the consuming public might expect the larger, more powerful company to manufacture both products, or expect the larger company to manufacture a product in the plaintiff's market, or expect that the larger company is likely to expand into the plaintiff's market." A H V, 237 F.3d at 234. Plaintiffs have not suggested that there are any other relevant facts for us to consider. Defendants suggest three other facts which we should consider: (1) commercial impression; (2) survey evidence; (3) continued success of the Plaintiffs' MIRACLESUIT line. Def. Mem. at 45-47. We will address each of these in turn.

1. Commercial Impression

Defendants argue that consumers overall impressions of the goods are different on the grounds that: (1) MIRACLESUIT is sold through department stores and specialty stores not associated with A H, while THE MIRACLE BRA swimwear is exclusively sold through entities related to Victoria's Secret; (2) the respective swimsuits focus on different areas of the body; (3) the distinctive labeling used by the parties. Id. at 44-45. We find that these grounds are adequately taken into consideration by the other Lapp factors and do not warrant separate consideration under this catch-all factor. Defendants first point is taken into consideration in Lapp factor seven where we looked at the different channels of trade. Defendants' second point was addressed in our discussion of Lapp factor nine which concerns the similarities of the products. The third point is addressed by Lapp factor one, which focuses on the similarity of the marks as they are presented in the marketplace.

2. Survey Evidence

Defendants next cite to the survey evidence submitted at trial. Id. at 45-46. We have reviewed this data and find that it does not sway our opinion in one way or another.²³ See A H I, 926 F. Supp. at 1257. As we explained in our first opinion on this matter, the survey evidence, from both sides, focused on comparing THE MIRACLE BRA bra with the MIRACLESUIT swimsuit. Id. at 1256. Since the issue before us is the likelihood of confusion among swimsuit purchasers, the data is not very persuasive. Id. Moreover, none of surveys conducted by the parties' respective experts examined whether the study participants were familiar with THE MIRACLE BRA. 10/30/95 Trial Tr. at 56; Defendants 1/12/96 Proposed Findings of Fact # 40.²⁴ Reverse confusion occurs when a consumer, familiar with a junior user's goods wrongly associates the senior user's goods with the junior user. Absent at least some familiarity with the junior user's goods (or mark) there cannot be reverse confusion.

23 This survey evidence was a part of the trial record. We invited the parties to utilize our prior findings of fact and previously submitted evidence. A H VI, 134 F. Supp.2d at 681-82.

24 In the study conducted by Defendants' expert, James Fouss, respondents were shown a black MIRACLESUIT swimsuit. They were then asked a number of questions to try and gauge whether any one associated the swimsuit with any particular company. Defendants 1/12/96 Proposed Findings of Fact # 47(a). The studies conducted by another defense expert, Ivan Ross, was designed to test for direct confusion, i.e. whether potential purchasers of THE MIRACLE BRA swimwear associated it with MIRACLESUIT. Id. at # 48-50. Plaintiffs' expert, Daniel Callahan, did not test whether respondents had prior familiarity with THE MIRACLE BRA or MIRACLESUIT. 10/30/95 Tr. at 56.

3. Plaintiffs' Profitability

As to the third point, Defendants criticize A H's previous failure to offer evidence of lost profits or other pecuniary harm. Id. at 46, citing A H II, 967 F. Supp. at 1465. When we conducted the initial damages trial in this case, we made a finding that Plaintiffs had not offered any evidence of lost profits or pecuniary harm as a result of Defendants' use of THE MIRACLE BRA mark on swimwear. A H II, 967 F. Supp. at 1465. We further found that sales of the MIRACLESUIT swimwear increased after the Defendants began marketing their swimwear. Id.

We specifically informed the parties not to address the damages issue in their briefs submitted in connection with the Third Circuit's remand. A H VI, 134 F. Supp.2d at 681. We preferred to reach issues related to damages, if at all, after the liability issue had been answered. Facts concerning profitability are not sufficiently probative of whether consumers will be confused into thinking that Defendants manufacture the MIRACLESUIT. Accordingly, we will not consider Plaintiffs' previous failure to offer evidence of lost profits or pecuniary harm in deciding whether a likelihood of confusion exists.

4. Conclusion

We find that since there are no other relevant facts for us to consider under this catch-all factor, Lapp factor ten favors neither party.

D. Balancing the Lapp Factors

In the typical case of reverse confusion, the senior user, who is a lesser player in the commercial world, adopts and uses a mark. Long Marks at 2. Then, a junior user with greater economic power adopts the same mark and promotes it extensively. Id. Through this promotion, the public recognizes the mark as belonging to the junior user. Id. As a result, the consumer mistakenly comes to believe that the junior user is the source of the senior user's product. Id. citing Banff, 841 F.2d at 490; Ameritech, 811 F.2d at 966; Big O Tire Dealers, 408 F. Supp. 1219.

Reverse confusion claims involve a number of different variables. The ideal case for finding reverse confusion occurs where: a senior user uses a mark intensively in a large and growing geographic area so that it acquires a secondary meaning which the public associates exclusively with the senior user and its products. Long Marks at 1. When this occurs,

the senior user can enjoin a junior user in that geographic area, and in other areas within a reasonable zone of expansion, from using the same mark or any other mark likely to cause confusion or to deceive, especially where the junior user has not adopted the mark in good faith and especially where the senior user has obtained a federal registration.

Id. When all of these circumstances are present, trademark infringement is clearly established and a senior user should receive some relief. Id. Unsurprisingly, such clear cut cases where all the vectors point toward a single conclusion are rare. More common are situations like the one we are faced with here where some factors point in favor of finding a likelihood of confusion and others do not.

In Plaintiffs' column there is: the similarity of the marks ( Lapp factor no. 1), the overlapping nature of the channels of trade ( Lapp factor no. 7), the sales targets for the goods ( Lapp factor no. 8); and the relationship of the goods in the minds of consumers ( Lapp factor no. 9). It is also important to point out that Plaintiffs' valid, protectable mark was in widespread use prior to the Defendants expanding their mark into swimwear.

On Defendants' side of the ledger there is: the strength of mark ( Lapp factor no. 2); the care and attention expected of consumers making a purchase ( Lapp factor no. 3); the meager evidence of actual confusion ( Lapp factors 4 and 6); and its intent in adopting the mark ( Lapp factor no. 5).

In a numerical sense, more factors favor the Defendants. Still, we find that Plaintiffs have established a likelihood of reverse confusion stemming from Defendants' use of THE MIRACLE BRA mark on swimwear. Mark similarity is the most important aspect of the Lapp test. Fisons, 30 F.3d at 476; A H V, 237 F.3d at 229; FF # 6. When highly similar marks are used on directly competing goods, a court may evaluate the likelihood of confusion based on the similarity of the marks alone. See A H V, 237 F.3d at 214; Analytic Recruiting, 156 F. Supp.2d at 513-514. Both marks share important visual and auditory similarities. In the direct confusion context, the Defendants' use of a disclaimer results in there being a meaningful and significant distinction between the two marks. When consumers cannot make quick reference to the disclaimer, the similarities between the marks is readily apparent. The problem with the similarity of the marks is exacerbated by the fact that the they are used on the same type of product, marketed to the same group of people through similar channels of trade.

The evidence which favors the Defendants does not eliminate the likelihood of confusion created by the marks. Although Lapp factor two favors the Defendants, it was a close call. The Defendants are the commercially stronger party and they did engage in extensive advertising and promotion for their mark. In addition, the mark is on the mid-point on the range of descriptiveness. As for the limited evidence of actual confusion, many courts have recognized that a Plaintiff need not show such evidence to win a Lanham Act claim. Such evidence is difficult to gather and its absence does not mean that harm has not occurred. Finally, while we have never found that the Defendants adopted the mark with bad faith or malicious intent, they were aware of Plaintiffs' mark prior to expanding THE MIRACLE BRA line into swimwear. FF # 3.

Overall, we find that the Lapp factors decisively favor A H. Thus, we conclude that there is a likelihood of reverse confusion between MIRACLESUIT and THE MIRACLE BRA as applied to swimwear. We find that Plaintiffs have satisfied all three elements required for trademark infringement. Specifically, Defendants' use of the mark, THE MIRACLE BRA on swimwear, is likely to create confusion concerning the origin of Plaintiffs' swimsuit.

IV. REMEDIES

The Lanham Act provides for injunctive relief, noting in relevant part, that the courts "shall have power to grant injunctions, according to the principles of equity and upon such terms as the court may deem reasonable to prevent the violation of any right of the registrant of a mark registered in the Patent and Trademark Office . . ." 15 U.S.C. § 1116(a). The Third Circuit has explained, "once a trademark owner demonstrates likelihood of confusion, it is entitled to injunctive relief." Lapp, 721 F.2d at 462. "It is difficult to imagine an unfair competition case where damages are adequate to remedy the problem of defendant's continued acts." McCarthy § 30:2. We recognize that injunctive relief is appropriate in this case and will allow the parties an opportunity to comment on the scope of the injunctive relief necessary.

The statutory and case law also permit the award of monetary damages, but courts do not automatically award profits, granting them only in light of equitable considerations. See SecuraComm, 166 F.3d at 189-90; Acxiom Corp. v. Axiom, Inc., 27 F. Supp.2d 478, 505-506 (D.Del. 1998). The Third Circuit has held that profits will be denied where an injunction forbidding future infringing acts satisfies the equities of the case. See National Dryer Mfg. Corp. v. National Drying Mach. Co., 228 F.2d 349, 350 (3d Cir. 1955); Microsoft Corp. v. CMOS Tech., Inc., 872 F. Supp. 1329, 1336-37 (D.N.J. 1994) (citations omitted). The Supreme Court has held that profits and damages may be awarded in cases where the infringement is "willfully calculated to exploit the advantage of an established mark." Champion Spark Plug Co. v. Sanders, 331 U.S. 125, 131, 67 S.Ct. 1136, 91 L.Ed. 1386 (1947). The Third Circuit has explained that "the propriety of an accounting depends upon whether [the infringer's] use was in good faith and whether it was palming off." Natural Footwear Ltd. v. Hart, Schaffner Marx, 760 F.2d 1383, 1407 (3d Cir. 1985). Without any showing of fraudulent conduct, palming off, or deliberate intent to confuse, the court should not award profits. See Acxiom, 27 F. Supp.2d at 506; Microsoft, 872 F. Supp. at 1337. Courts may also consider whether there is evidence of pecuniary harm suffered by the owner of the marks. Acxiom, 27 F. Supp.2d at 506; Alfred Dunhill of London, Inc. v. Kasser Distillers Prod. Corp., 350 F. Supp. 1341, 1369 (E.D.Pa. 1972), aff'd. 480 F.2d 917 (3d Cir. 1973) (table text). We have serious reservations about awarding monetary damages in this case. See A H III, 166 F.3d at 209 (expressing reservation about the award of monetary damages in this case). We note that Plaintiffs have not shown fraudulent conduct or a deliberate intent to confuse. Moreover, when we initially set damages in A H II, we made a finding that Plaintiffs had failed to offer any evidence of lost profits or other direct pecuniary harm proximately caused by Defendants' use of THE MIRACLE BRA on swimwear. 967 F. Supp. at 1465.

As part of their monetary damages, Plaintiffs requested recovery for corrective advertising. Pls. Mem. at 42. Following the trial on the merits, we evaluated Plaintiffs' claim for a monetary award of prospective corrective advertising. A H II, 967 F. Supp. at 1478. Plaintiffs did not claim or prove that they were entitled to damages for corrective advertising expenditures already made. Id. at 1478 n. 12. We determined that such relief was not appropriate. Id. discussing Big O Tire Dealers, 561 F.2d at 1374-75. See also Adray v. Adry-Mart, Inc., 76 F.3d 984, 988-89 (9th Cir. 1995). Plaintiffs had not shown that the Defendants have acted in "wanton and reckless disregard for plaintiff's rights." Id. Nor were they able to prove any sizable damages. Id. at 1478-79. For the same reasons cited in A H II, we continue to believe that an award for corrective damages is inappropriate in the case. Nonetheless, we will withhold final judgement on whether such relief will be granted until after the parties have been given an opportunity to brief the issue of compensatory damages.

Finally, Plaintiffs have requested various more far reaching forms of relief in the form of treble damages and attorneys' fees. Pls. Mem. at 42. Under 15 U.S.C. § 1117(a), we have discretion to increase the award of damages three times the amount of actual damages proven. Damages may be enhanced on the rational of the difficulties in proving the amount of loss in a market context and in order to deter cases of egregious infringement. This power to enhance damages does not permit us to award damages when none have been proven. Caesars World, Inc. v. Venus Lounge, Inc., 520 F.2d 269, 186 U.S.P.Q. 497 (3d Cir. 1975) ("Three times zero is zero."); Donsco, Inc. v. Casper Corp., 587 F.2d 602, 199 U.S.P.Q. 705 (3d Cir. 1978). Since the evidence at trial did not support a finding of willful infringement, an enhancement of damages is inappropriate in this case. See A H II, 967 F. Supp.2d at 1482 n. 16; SecuraComm, 166 F.3d at 190 (evidence did not support a finding of willful infringement making the trebling of damages inappropriate). Plaintiffs have never proven bad faith, deliberate intent, or egregious infringement. Id. Furthermore, no other equities weigh significantly against the Defendants or in favor of Plaintiffs. Id.

As to attorneys' fees, 15 U.S.C. § 1117(a) provides that a court "in exceptional cases may award reasonable attorney fees to the prevailing party." The Third Circuit has explained that a court must find "culpable conduct on the part of the losing party, such as bad faith, fraud, malice, or knowing infringement," for a case to qualify as "exceptional." Ferrero U.S.A., Inc. v. Ozak Trading, Inc., 952 F.2d 44, 47 (3d Cir. 1991); Acxiom, 27 F. Supp.2d at 505-6. The Third Circuit has construed "exceptional" cases narrowly, noting that: "typically, attorneys' fee cases involve deliberate attempts by the defendant to pass off its goods as those of the plaintiff by applying plaintiff's trademark to defendant's goods." Id. There needs to be an intentional showing of bad faith. Id. For the same reasons we explained above in our analysis of Lapp factor five, we find that the Defendants did not act in bad faith when they expanded the use of THE MIRACLE BRA mark. Plaintiffs have not shown evidence that the Defendants adopted the mark with an improper intent or otherwise sought to willfully infringe on Plaintiffs. This not an exceptional case and attorney's fees will not be awarded.

V. CONCLUSIONS OF LAW

Consistent with the foregoing discussion and the findings of fact made by this court, see A H I, 926 F. Supp. at 1235-54; A H II, 967 F. Supp. at 1462-67, we state the following conclusions of law:

1. Plaintiffs have met their burden of establishing a likelihood of reverse confusion between the MIRACLESUIT and THE MIRACLE BRA swimwear trademarks under the relevant precedent for directly competing goods.

2. Plaintiffs have met their burden in establishing their entitlement to injunctive relief.

3. Plaintiffs have failed to establish willful infringement or other facts which would entitle them to treble damages under 15 U.S.C. § 1117(a).

4. Plaintiffs have failed to establish that this case is an exceptional one as required for award of attorneys' fees under 15 U.S.C. § 1117(a).

An appropriate order follows.

ORDER

AND NOW this 17th day of August 2001, upon consideration of Plaintiffs' Brief Sur Remand, filed April 11, 2001, and Defendants' Brief Upon Remand, filed on May 14, 2001, and Plaintiffs Reply Brief, filed on May 31, 2001, it is hereby ORDERED as follows:

1. We find in favor of Plaintiffs with regard to claims of reverse confusion with respect to THE MIRACLE BRA trademark and the MIRACLESUIT trademark as applied to the swimwear market;

2. We find that Plaintiffs have not met their burden as to the wilfulness of Defendants' conduct to entitle them to relief beyond an injunction and compensatory damages

3. The parties are to report for an immediate settlement conference before Judge Weiner.

4. Plaintiffs must submit within sixty days of this order a brief outlining their entitlement to compensatory damages and an injunction, consistent with that discussed in this opinion and upon this record.

5. Within thirty (30) days of the filing of Plaintiffs' Brief, Defendants shall submit a response brief.

6. All briefs, responses, and replies shall be reasonable in length, not to exceed thirty (30) pages. They shall be fully annotated to the opinions in this case, to this Court's prior findings of fact and to the established record.