Fendi Adele S.R.L. v. Burlington Coat Factory Warehouse Corp.

689 F. Supp. 2d 585 (2010)

FENDI ADELE S.R.L., Fendi S.R.L., and Fendi North America, Plaintiffs,
v.
BURLINGTON COAT FACTORY WAREHOUSE CORP. et al., Defendants/Third Party Plaintiffs,
v.
546332 BC, Ltd., d/b/a/ Colton International, Summit Resources Imports LLC, Euro Moda, Inc., Moda Oggi, Inc., and Ashley Reed Trading, Inc., Third Party Defendants.

No. 06 Civ. 85(LBS).

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

February 8, 2010. Memorandum and Order Amended on Reconsideration March 23, 2010.

Richard L. Mattiaccio, Steven Skulnik, Victor Genecin, Squire, Sanders & Dempsey, L.L.P., New York, NY, for Plaintiffs, Fendi Adele S.R.L., Fendi S.R.L., Fendi North America, Inc.

J. Joseph Bainton, Smith, Gambrell & Russell, LLP, New York, NY, for Defendants/Third Party Plaintiffs.

Carmine Joseph Castellano, John Jongmin Lee, Smith, Gambrell & Russell, LLP, New York, NY, for Burlington Coat Factory Warehouse Corporation.

Robert Ian Goodman, Law Office of Louise M. Valiquette, Ossining, NY, for Colton International.

Bruce D. Katz, Bruce D. Katz, Esq., New York, NY, for 546332 BC Ltd.

Gerard Francis Dunne, Law Office of Gerard F. Dunne, P.C., New York, NY, for Ashley Reed Trading, Inc.

Kirk Gordon Downing, Law Offices of Kirk G. Downing, Beverly Hills, CA, for Moda Oggi, Inc.

John Smargiassi, Joseph & Smargiassi, LLC, New York, NY, for Euro Moda, Inc.

Lawrence Badash Fechner, Jaffe, Ross & Light LLP, New York, NY, for Summit Resource Imports LLC.

MEMORANDUM AND ORDER

SAND, District Judge.

In this Court's 2007 opinion, Fendi Adele S.R.L. v. Burlington Coat Factory, No. 06 Civ. 85, 2007 WL 2982295 (S.D.N.Y. Oct. 10, 2007), we granted partial summary judgment in favor of Plaintiffs Fendi Adele S.R.L., Fendi S.R.L., and Fendi North America (collectively known as "Fendi"), finding Defendants Burlington Coat Factory Warehouse Corporation and Cohoes Fashion, Inc., a wholly-owned subsidiary of Burlington Coat Factory, (collectively known as "Burlington") in contempt of this Court's 1987 Injunction that prohibited Burlington from selling any Fendi-branded merchandise without written permission from Fendi.^[1] Fendi now moves for summary judgment on its remaining causes of action.^[2] In addition, three other motions are pending before the Court: Burlington's Motion to Amend the Answer; Burlington's Motion for Summary Judgment against the Third Party Defendants; and Colton's Cross-Motion to Dismiss the Third-Party Complaint. We will also address Fendi and Burlington's objections to Magistrate Judge Dolinger's December 1, 2009 Report and Recommendation.

I. Background

Fendi owns the federally-registered Fendi trademarks and is the exclusive designer of handbags, shoulder bags, purses, wallets and key holders bearing this trademark. (Pl.'s 56.1 ¶ 1-2.) Five Fendi trademarks, each registered with the United States Patent Trademark Office, are used on their products. (Pl.'s 56.1 ¶ 68.) Each of its products is individually designed, as are the components used, in a complex design process. (Pl.'s 56.1 ¶¶ 73-75.) Fendi manufactures and sells its products in two annual seasons. Fendi's authorized customers visit Fendi's showrooms during the Fall and Spring Fashion Weeks in Milan, view the collection for the season, and place their orders for specific products in specific quantities. (Pl.'s 56.1 ¶¶ 77, 84.)

Fendi contracts with small companies in Florence, Italy to assemble the products. (Pl.'s 56.1 ¶ 97.) Each assembly order is for a stated quantity of a specific product in a specific combination of materials. (Pl.'s 56.1 ¶ 102.) Fendi gives the sub-contractors a sixteen digit code for each product. The code indicates the style number, combination of materials used, and the season for which the order is manufactured. (Pl.'s 56.1 ¶¶ 110-11.) The code is stamped on a piece of leather trim that is sewn into the inside lining of each product. (Pl.'s 56.1 ¶ 112.) Fendi monitors the coding of its products by its assembly subcontractors. (Pl.'s 56.1 ¶ 113.) As of the Fall-Winter season of 2003, Fendi began stitching a hologram label in the lining of every product. (Pl.'s 56.1 ¶¶ 117-18.) When the hologram is placed under a microscope, symbols can be seen within a genuine hologram, permitting it to be identified as authentic. (Pl.'s 56.1 ¶ 122.) A specialized vendor manufactures the hologram labels for Fendi; the holograms are delivered to Fendi's manufacturing facility and locked in a safe. (Pl.'s 56.1 ¶¶ 123-24.) One hologram is placed in each item, and Fendi tracks the number on each item. (Pl.'s 56.1 ¶¶ 125-26.)

Fendi visits its assembly contractors twice a week, monitors their work, and conducts inspections. (Pl.'s 56.1 ¶¶ 131-33.) Fendi also conducts inspections of its sub-subcontractors. (Pl.'s 56.1 ¶ 134.) Fendi inspects the finished products, and if a product has not been assembled correctly and cannot be repaired, Fendi destroys the product. (Pl.'s 56.1 ¶ 141.) The products are then shipped from Fendi's distribution center directly to Fendi's authorized customers. (Pl.'s 56.1 ¶ 141.) Each purchaser receives an invoice, which includes Fendi's name, the locations of the corporate offices in Italy, the date the goods were ready for shipment, the name and address of the authorized customer, and the order number assigned to the order. (Pl.'s 56.1 ¶¶ 147-48.) The invoice also details the contents of each box and the exact quantity of each style, which is described by Fendi's style number and materials code, and a narrative description of the article. (Pl.'s 56.1 ¶¶ 149-50.) Alberto Fabbri, the chief financial officer of Fendi Group, is able to access the computerized records of all the invoices going back to 2001. (Pl.'s 56.1 ¶ 152.)

Leonardo Minerva, Industrial Director of Leather Goods and Logistics Director for Fendi, was in charge of the manufacturing of the products from September 2002 to March 2008. (Pl.'s 56.1 ¶ 154.) As part of his duties, Minerva, along with his assistant, Massimo Lepri, examined hundreds of questioned Fendi-branded items each year for authenticity. (Pl.'s 56.1 ¶¶ 155-60.) Fendi has a multi-step process to determine whether an item is genuine, which involves: (1) visually inspecting the physical characteristics of the item, such as the quality of the materials, the purported Fendi trademark, the finishing, the metallic hardware, and the way the product was manufactured; (2) checking the item against Fendi's records to determine whether the particular style has been manufactured using the correct combination of materials; (3) inspecting each of the components for conformity with Fendi's specifications; (4) examining the sixteen digit code stamped on the leather trim inside the lining to determine if it matches Fendi's internal manufacturing and design records; and (5) examining the hologram for genuineness and checking the serial number on the hologram label to see if it matches Fendi's records. (Pl.'s 56.1 ¶¶ 157-65.)

At his deposition, Minerva testified that he never saw an instance in which a Fendi-branded item had correct codes but imperfect material. (Pl.'s 56.1 ¶ 117.) Minerva testified regarding forty three Fendi-branded items that were either sold by Burlington or that were selected from their inventory during discovery. These inspections were conducted by Minerva and his assistant Lepri. (Pl.'s 56.1 ¶ 166; Def.'s Response to 56.1 ¶ 166.) Minerva testified that thirty nine of the items were counterfeit, and three items were genuine Fendi products. (Pl.'s 56.1 ¶¶ 166-68.) Burlington's principal sources for Fendi-branded handbags and small leather goods were: 546332 BC, Ltd. d/b/a/ Colton International ("Colton"); Eura Moda, Inc. ("Euro Moda"); Moda Oggi, Inc. ("Moda Oggi"); Summit Resources LLC ("Summit"); and Ashley Reed Trading, Inc. ("Ashley Reed") (collectively known as the "Vendors"). (Pl.'s 56.1 ¶ 187.) Colton, Moda Oggi, Summit and Ashley Reed have never been customers of Fendi. (Pl.'s 56.1 ¶ 189.) In 2002 and 2003, Euro Moda made occasional purchases in small quantities of Fendi-branded items from an outlet store operated by Fendi. (Pl.'s 56.1 ¶ 189.) Minerva did not inspect any Fendi-branded goods supplied by Ashley Reed to Burlington but testified that he, along with Lepri, inspected twenty Fendi-branded items supplied by Ashley Reed to Filene's Basement, Big M, Inc., Nordstrom's Rack and Saks Off Fifth. (Pl.'s 56.1 ¶ 190.) Minerva testified that the items supplied by Ashley Reed were counterfeit. (Pl.'s 56.1 ¶ 190.)

On April 12, 2004, Stacy Haigney, Burlington's in-house counsel since 1990, met with counsel for Louis Vuitton Malletier, a company that has the same ultimate ownership as Fendi, to attempt to settle a trademark infringement litigation brought against Burlington. (Pl.'s 56.1 ¶ 204; Haigney Dec. ¶ 4, 13.) At the negotiations, counsel showed Haigney a Fendi-branded handbag that had been purchased from Burlington and informed Haigney that the handbag was counterfeit. (Pl.'s 56.1 ¶ 215.) Haigney requested that counsel send him a formal letter to that effect. (Def.'s Response to 56.1 ¶ 215.) Haigney confirmed that Colton had supplied the bag to Burlington and contacted counsel for Colton regarding the handbag. (Pl.'s 56.1 ¶¶ 216-17.) In late April 2004, Haigney received a letter on behalf of Fendi, which formally placed Burlington on notice that it was selling counterfeit branded goods and demanded that it cease and desist. (Pl.'s 56.1 ¶ 218; Haigney Dec. ¶ 18.) During the ensuing months, Fendi engaged in correspondence with Colton regarding the allegedly counterfeit bag. (Haigney Dec. ¶ 19.) On August 4, 2004, Colton sent a fax to Haigney containing a copy of an unsigned letter in Italian, purportedly from Minerva, together with an English translation stating that Fendi had examined the handbag and found that it had been made by Fendi and was within Fendi's standards of quality. (Pl.'s 56.1 ¶ 229.) Fendi challenged the authenticity of the letter. (Def.'s Response to 56.1 ¶ 232.) On December 23, 2005, Haigney received another cease and desist letter from Fendi, which alleged that all of Burlington's Fendi products were counterfeit and sold in violation of the 1987 Injunction. (Pl.'s 56.1 ¶ 275; Haigney Dec. ¶ 36.) Fend-branded items were still on Fendi's shelves as late as March 2008. (Pl.'s 56.1 ¶ 21.) This Court's 2007 Opinion found Burlington in contempt, granted Fendi's motion for an accounting and disgorgement of profits, and awarded Fendi costs and attorney's fees.

II. Fendi's Motion for Summary Judgment

A court may only grant a motion for summary judgment if there is no genuine issue as to any material fact and the moving party is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(c). The court looks to the substantive law to identify which facts are material; only disputes over facts that might affect the outcome of the suit under the governing law will preclude summary judgment. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S. Ct. 2505, 91 L. Ed. 2d 202 (1986); Kinsella v. Rumsfeld, 320 F.3d 309, 311 (2d Cir.2003). "[T]his standard provides that the mere existence of some alleged factual dispute between the parties will not defeat an otherwise properly supported motion for summary judgment; the requirement is that there be no genuine issue of material fact." Anderson, 477 U.S. at 247-48, 106 S. Ct. 2505 (emphasis in original). An issue is genuine if "there is sufficient evidence favoring the nonmoving party for a jury to return a verdict for that party." Id. at 249, 106 S. Ct. 2505. If the evidence is "merely colorable" or "not significantly probative," summary judgment may be granted. Id. at 149-50, 106 S. Ct. 2505.

Where the moving party would ultimately bear the burden of persuasion at trial, the moving party must make a prima facie showing with credible evidence that there is no genuine issue of material facts for trial. Celotex Corp. v. Catrett, 477 U.S. 317, 331, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986). If the moving party successfully makes such a showing, the burden of production shifts to the non-moving party to submit evidentiary materials demonstrating the existence of a genuine issue. Fed. R.Civ.P. 56(e); Celotex, 477 U.S. at 331, 106 S. Ct. 2548. "The nonmoving party may not rely on mere conclusory allegations nor speculation, but instead must offer some hard evidence showing that its version of the events is not wholly fanciful." D'Amico v. City of New York, 132 F.3d 145, 149 (2d Cir.1998). The court must draw all inferences in favor of the non-moving party. Id.

a. Admissibility of Minerva's Testimony

As a preliminary matter, we must determine whether or not Leonardo Minerva's deposition testimony is admissible. Burlington alleges that his testimony should be excluded because (1) Minerva lacked personal knowledge of the contents of the authenticity reports and (2) Minerva relied on the reports throughout the course of his deposition.^[3]

The district court has broad discretion to determine whether a witness is using a writing to refresh his or her memory or is offering the writing for the truth of something the witness can no longer recall. 20th Century Wear, Inc. v. Sanmark-Stardust Inc., 747 F.2d 81, 93 n. 17 (2d Cir.1984); Doty v. Elias, 733 F.2d 720, 725 (10th Cir.1984); United States v. Conley, 503 F.2d 520, 522 (8th Cir.1974). The court must ensure that the witness actually has a present recollection and that evidence, which would otherwise be inadmissible, does not inadvertently slip in for its truth. 20th Century Wear, 747 F.2d at 93 n. 17. In order to safeguard against this risk of prejudice, Federal Rule of Evidence 612 permits opposing counsel to inspect the writing used, cross-examine the witness on it, and introduce portions into evidence. Fed.R.Evid. 612; see also 20th Century Wear, 747 F.2d at 93 n. 17.

Burlington's allegation that Minerva had no recollection to refresh is without merit. Minerva testified that he had been conducting authenticity inspections for over four years. (Genecin Dec. Ex. 31 ("Minerva Dep.") at 12:21-24.) He described in detail the procedures Fendi has in place to protect the authenticity of its goods and identify counterfeit products. Minerva described the kits provided to sub-contractors, the codes that were placed on the products, the precise specifications of the distance between the "Fs" on Fendi fabric, the type of metal that was used for the hardware, and the manner of quality inspection that occurred when Fendi received the assembled products. (Minerva Dep. at 35:10-17.) Minerva testified that his assistant Massimo Lepri generally prepared the reports, and then Minerva reviewed the reports, discussed them with Lepri, and sometimes made corrections. (Minerva Dep. at 337:23-338:10). After Lepri signed the report, Minerva would sign the report. Nothing in Minerva's testimony suggests that he ever signed an authenticity report without conducting an inspection.^[4] Minerva's supervision and review of the reports is sufficient to establish personal knowledge, regardless of the fact that Lepri conducted the initial inspection and drafted the reports. See United States v. Muhammad, 120 F.3d 688, 699 (7th Cir.1997) ("[A]lthough Agent Gandolfo prepared the FBI 302 covering Muhammad's interview, [Agent Denny] reviewed that report for accuracy, made corrections to it, and signed it. In essence, the report was that of both Agent Gandolfo and Agent Denny.")

Furthermore, Minerva was entitled to consult the reports throughout his testimony. Courts have permitted witnesses to rely on notes or documents throughout the course of testimony if the witness demonstrates an independent recollection, and the testimony is of a detailed nature. See United States v. Rinke, 778 F.2d 581, 588 (10th Cir.1985) (finding that the witness was properly allowed to read from his notes where he had established an independent recollection by embellishing on the notes and testifying regarding subjects that were not explicitly contained in the notes regarding specific content of telephone conversations); United States v. Riccardi, 174 F.2d 883, 889 (3d Cir.1949) (trial judge did not err in permitting a witness to read a list of chattel because he "immediately recognized that the items of property involved were so numerous that in the ordinary course of events no one would be expected to recite them without having learned a list by rote memory").^[5]

Minerva demonstrated an independent recollection by going beyond the scope of the reports. Minerva identified a manufacturing number, which had once been a real Fendi number, but was now repeatedly found on counterfeit goods. (See e.g., Minerva Dep. at 73:3-18.) He also pointed to a hologram that was peeling off of a product, while a genuine Fendi hologram is not detachable. (Minerva Dep. at 72:18-73:2.) Throughout Minerva's testimony, he compared counterfeit bags with genuine bags, describing the differences. (See e.g., Minerva Dep. at 192:23-194:7.) Furthermore, the nature of Minerva's testimony was highly detailed. He provided testimony regarding approximately sixty counterfeit products. For each item he identified multiple deviations, which demonstrated that the products were counterfeit. In addition, each product involved deciphering the legitimacy of a sixteen digit code. Given the fact that Minerva did not rely exclusively on the reports and that the nature of the testimony was detailed, we find his use of the reports was proper.^[6]

b. Liability

i. Trademark Infringement and False Designation of Origin

Fendi's first two claims allege counterfeiting in violation of 15 U.S.C § 1114(a) and false designation of origin in violation of 15 U.S.C. § 1125(a). To prevail on these claims, Fendi must successfully establish that: (1) it had a valid mark entitled to protection under the Lanham Act; and (2) Burlington used a similar mark in commerce in a way that would likely cause confusion among the relevant consuming public. See Louis Vuitton Malletier v. Dooney & Bourke, Inc., 454 F.3d 108, 114-15 (2d Cir.2006); The Sports Authority, Inc. v. Prime Hospitality Corp., 89 F.3d 955, 960 (2d Cir.1996).

First, the Court must determine whether Fendi had a valid mark entitled to protection under the Lanham Act. Each of the five trademarks used on the products at issue is registered in the U.S. Patent and Trademark office. (Pl.'s 56.1 ¶ 68.) Under the Lanham Act a "certificate of registration of a mark upon the principal register provided by this chapter shall be prima facie evidence of the validity of the registered mark and of the registration of the mark, of the registrant's ownership of the mark, and of the registrant's exclusive right to use the registered mark...." 15 U.S.C. § 1057(b). In an action for infringement of a registered mark, "the defendant bears the burden to rebut the presumption of the mark's protectibility by a preponderance of the evidence." Lane Capital Management, Inc. v. Lane Capital Management, Inc., 192 F.3d 337, 345 (2d Cir.1999). As we previously noted, both the FENDI and FF monogram trademarks have been extensively advertised and are now regarded as prestigious symbols in fashion. Fendi S.a.s. Di Paola Fendi E Sorelle v. Cosmetic World, Ltd., 642 F. Supp. 1143, 1145 (S.D.N.Y.1986). Burlington has put forth no evidence to challenge the validity of the registered trademark; rather, Burlington brings a counterclaim to cancel Fendi's trademarks under a naked licensing theory. Burlington contends that Fendi has abandoned its rights to the marks. Burlington offers evidence that Fendi's assemblers were selling non-conforming bags directly from the manufacturers and that Fendi does not always act in conformity with its quality control measures. Specifically, Burlington alleges that a witness saw a bolt of fabric in an assembly warehouse, which should not exist in the one kit per bag assembly procedure.

A "naked license" is a license to use a trademark without sufficient quality control provided by the licensor. Under the naked licensing doctrine, a naked licensor may be estopped from challenging a breach of a licensing agreement, or even be deemed to have involuntarily abandoned the rights to the mark. See E.G.L. Gem Lab Ltd. v. Gem Quality Inst., Inc., 90 F. Supp. 2d 277, 300 (S.D.N.Y.2000), aff'd, 4 Fed.Appx. 81 (2d Cir.2001) ("A licensor must exercise some degree of supervision over the licensee on pain of abandonment of the mark."). The parties agree that the assembly subcontractors were not licensees and were contracted to do no more than assemble Fendi products. (Pl.'s 56.1 ¶¶ 97-98; Def.'s Response to 56.1 ¶¶ 97-98.) To succeed on a "naked licensing" abandonment claim, Burlington must meet a "high burden of proof." Patsy's Italian Restaurant v. Banas, 508 F. Supp. 2d 194, 212 (E.D.N.Y.2007) (citing Warner Bros., Inc. v. Gay Toys, Inc., 724 F.2d 327, 334 (2d Cir.1983)). Burlington has failed put forth actual evidence of back door dealings.^[7] Nor is one witness observing a bolt of fabric sufficient to demonstrate that Fendi has abandoned its mark. Based on the ample record showing Fendi's comprehensive control mechanisms, allegations of isolated incidents of non-compliance with those mechanisms is insufficient to support a naked licensing theory. Burlington's counterclaim is denied, and we find that Fendi has a valid trademark.

The second inquiry under the Lanham Act is whether there was a likelihood of confusion. To determine whether there is a likelihood of confusion, courts in the Second Circuit generally rely on the eight-factor test set forth in Polaroid Corp. v. Polarad Elecs. Corp., 287 F.2d 492, 495-96 (2d Cir.1961). However, where counterfeit marks are involved, the court need not engage in a Polaroid analysis because "counterfeit marks are inherently confusing." Chloe v. DesignersImports.com USA Inc., No. 07 Civ. 1791(CS), 2009 WL 1227927, at *6 (S.D.N.Y. Apr. 30, 2009); Gucci Am., Inc. v. Tyrrell-Miller, 678 F. Supp. 2d 117, 119-20 (S.D.N.Y.2008). To find a likelihood of confusion, a court need only determine that the items at issue are counterfeit and that the defendant distributed, offered for sale, or sold the items. Gucci Am., Inc. v. Duty Free Apparel, 286 F. Supp. 2d 284, 287 (S.D.N.Y.2003). Burlington does not dispute that the goods at issue were taken from its warehouse; therefore, the Court need only determine if the goods were counterfeit.

Minerva, the Industrial Director of Leather Goods and Logistics Director for Fendi from 2002 to 2008, testified that thirty nine of the Fendi-branded products taken from Burlington were counterfeit. He also identified twenty Fendi-branded products Ashley Reed supplied to other retailers as counterfeit. Minerva found numerous deviations for each counterfeit product, including falsified holograms, peeling holograms, improper coding, inferior materials, improper hardware, non-conforming fabric, and combinations of colors and hardware not used in genuine Fendi products. Minerva's testimony establishes a prima facie case that the goods at issue are counterfeit. See Motorola, Inc. v. Abeckaser, No. 07 Civ. 3963(CPS), 2009 WL 962809, at *5 (E.D.N.Y. Apr. 8, 2009) (finding that the plaintiff had made prima facie case based on the quality manager's findings of inferior quality, poor affixation of the marks on the products, inconsistency of the materials used with those used to make genuine goods, and the inconsistency in the packaging on the counterfeit goods); Duty Free Apparel, 286 F.Supp.2d at 288 (finding prima facie case established based on plaintiff's expert testimony, which identified several deviations in each of the items at issue).

Burlington has failed to provide any evidence to defeat Fendi's prima facie case and establish the existence of a genuine issue as to the authenticity of the goods in question. Burlington has not offered a countervailing expert, evidence that any of the goods in question were purchased from an authorized Fendi customer, or any other affirmative evidence that the bags are genuine. See Duty Free Apparel, 286 F.Supp.2d at 289 ("It is notable that Harvest Wrap failed to produce any affirmative evidence that the items in evidence are actually authentic, either through an expert or through an explanation of their origins."). Rather, Burlington offers evidence that Fendi goods are available from sources other than Fendi.^[8] Burlington also notes that it, along with Ashley Reed, has previously purchased authentic Fendi goods through the secondary market. (Def.'s Opp. Summ. J. 9-10.) Burlington's prior purchase of authentic goods has no bearing on Minerva's determination that the goods he inspected were counterfeit. See Motorola, Inc. v. Abeckaser, No. 07 Civ. 3963, 2009 WL 962809, at *5 (E.D.N.Y. Apr. 8, 2009) (finding that the fact that the defendants may have at one time purchased authentic goods was not inconsistent with the witness's testimony that the goods in question were counterfeit).

Burlington contends that genuine goods can be purchased from retailers other than authorized customers, and, therefore, the source of the goods is not dispositive as to authenticity. Thirty nine products were taken from Burlington; none of which were obtained through an authorized Fendi customer, and all of the products contained numerous deviations from authentic Fendi products. Even if authentic goods could be obtained through non-authorized customers, Burlington has put forth no evidence to counter the multiple deviations identified in each of the products. Burlington has failed to offer any "hard evidence showing that its version of the events is not wholly fanciful." D'Amico, 132 F.3d at 149 (2d Cir.1998). We find that the goods supplied by the Vendors were counterfeit, and Burlington is liable for counterfeiting and false designation of origin.^[9]

ii. Trademark Dilution

Fendi next moves for summary judgment for trademark dilution pursuant to 15 U.S.C. § 1125(c) and the New York anti-dilution statute, N.Y. Bus. Law. § 360-1. To prevail on a federal claim of trademark dilution, a plaintiff must establish that "(1) its mark is famous; (2) the defendant is making commercial use of the mark in commerce; (3) the defendant's use began after the mark became famous; and (4) the defendant's use of the mark dilutes the quality of the mark by diminishing the capacity of the mark to identify and distinguish goods and services." Savin Corp. v. Savin Group, 391 F.3d 439, 448-49 (2d Cir.2004). This Court previously found that the Fendi trademarks are famous and distinctive and that Burlington began selling counterfeit Fendi goods after Fendi had made its marks famous. With regard to the requirement of actual dilution, the Court of Appeals for the Second Circuit has found that if "a plaintiff who owns a famous senior mark can show the commercial use of an identical junior mark," there is a presumption of actual dilution. Savin Corp. v. Savin Group, 391 F.3d 439, 452-3 (2d Cir.2004). The marks at issue in this case are identical, and Burlington has put forth no evidence to overcome the presumption of actual dilution.

Under New York law, a plaintiff must demonstrate ownership of a distinctive mark and likelihood of dilution of that mark through blurring or tarnishment. The Sports Authority, Inc. v. Prime Hospitality Corp., 89 F.3d 955, 966 (2d Cir. 1996). Federal law's requirement of "actual dilution" is more stringent than New York's requirement of "likelihood of dilution." Savin, 391 F.3d at 456. Having already found trademark dilution under the more stringent federal standard, we also find trademark dilution under New York law.

iii. Unfair Competition Under New York Law

An unfair competition claim under common law is governed largely by the same standards as its Lanham Act counterparts, except common law requires a showing of bad faith or intent. Genesee Brewing Co., Inc. v. Stroh Brewing Co., 124 F.3d 137, 149 (2d Cir.1997). Use of a counterfeit mark creates a presumption of bad faith. Philip Morris U.S.A., Inc. v. Felizardo, No. 03 Civ. 5891(HB), 2004 WL 1375277, at *6 (S.D.N.Y. Jun. 18, 2004); see also Burberry Ltd. and Burberry USA v. Designers Imports, Inc., No. 07 Civ. 3997(PAC), 2010 WL 199906, at *8 (S.D.N.Y. Jan. 19, 2010) ("Burberry's evidence of Defendant's sale of counterfeit Burberry-branded merchandise creates a presumption of bad faith, satisfying the elements of Burberry's common law unfair competition claim."); Burberry Ltd. v. Euro Moda, Inc., No. 08 Civ. 5781(CM), 2009 WL 1675080, at *15 (S.D.N.Y. Jun. 10, 2009) (finding summary judgment appropriate because "evidence offered by Burberry concerning Moda Oggi's and Euro Moda's sale of merchandise with counterfeit Burberry Marks allows the Court to infer bad faith."). Burlington's sale of counterfeit Fendi-branded merchandise allows this Court to infer bad faith; therefore, summary judgment is appropriate as to Fendi's common law claims of unfair competition.

c. Damages

Fendi seeks disgorgement of profits, treble damages, reasonable attorney's fees, prejudgment interest and costs pursuant to 15 U.S.C. § 1117. Under Section 1117(a), a plaintiff is entitled to recover "(1) defendant's profits, (2) any damages sustained by the plaintiff, and (3) the costs of the action." 15 U.S.C. § 1117(a). In exceptional cases the court may award reasonable attorney's fees to the prevailing party. Id. In a case such as this one, which involves the use of a counterfeit mark, "the court shall, unless the court finds extenuating circumstances, enter judgment for three times such profits or damages, whichever amount is greater, together with a reasonable attorney's fee, if the violation consists of ... intentionally using a mark or designation, knowing such mark or designation is a counterfeit mark (as defined in section 1116(d) of this title), in connection with the sale, offering for sale, or distribution of goods or services." 15 U.S.C. § 1117(b). The question presented is whether there is a genuine issue of material fact as to Burlington's willfulness.^[10]

The standard for willfulness is whether the defendant had knowledge that his or her conduct represented infringement or recklessly disregarded the possibility. Kepner-Tregoe, Inc. v. Vroom, 186 F.3d 283, 288 (2d Cir.1999); see Koon Chun Hing Kee Soy & Sauce Factory, Ltd. v. Star Mark Mgmt., No. 04 Civ. 2293(JFB), 2007 WL 74304, at *11 (E.D.N.Y. Jan. 8, 2007) (applying the willfulness standard in Kepner-Tregoe to claims brought under the Lanham Act); Nike, Inc. v. Top Brand Co., No. 00 Civ. 8179(KMW), 2005 WL 1654859, at *6 (S.D.N.Y. July 13, 2005) (same). While caution must be exercised in granting summary judgment when state of mind is an issue, the summary judgment rule "would be rendered sterile" if the mere existence of an issue as to state of mind would automatically defeat an otherwise valid motion. Nora Beverages, Inc. v. Perrier Group of America, Inc., 269 F.3d 114, 125 (2d Cir.2001). Where the defendant offers no probative evidence raising a genuine issue of material fact regarding willfulness, summary judgment is appropriate. See Tanning Research Lab., Inc. v. Worldwide Import & Export Corp., 803 F. Supp. 606, 610 (E.D.N.Y.1992) ("It is true that questions of a defendant's state of mind are rarely suitable for determination on summary judgment.... But defendants have offered no probative evidence raising a genuine issue of material fact regarding their willfully blind violation.").

Fendi puts forth the following evidence to demonstrate that no reasonable trier of fact could find in favor of Burlington on the issue of willfulness: (1) Burlington never took any steps to comply with the 1987 Injunction; (2) Burlington failed to follow its own procedures for preventing the purchase and sale of counterfeits even though it understood the risk associated with buying Fendi-branded goods from parties other than the trademark owner; and (3) Burlington persisted in purchasing and selling counterfeit Fendi-branded goods even after Fendi notified it that the merchandise was counterfeit. (Pl.'s Mem. Summ. J. 28.) Burlington argues that the 1987 Injunction is not relevant. However, the 1987 Injunction was a form of prospective relief, which sought to prevent future counterfeit sales of Fendi-branded merchandise, by preventing Burlington from selling any Fendi-branded merchandise. Fendi entered into the 1987 Injunction to avoid litigation and because of its fundamental distrust of Burlington's ability or desire to distinguish counterfeit merchandise from genuine merchandise. As this Court previously found in our 2007 Opinion, Burlington did not implement any control mechanisms to ensure its compliance with the injunction. Burlington's in-house counsel, Stacy Haigney, admitted that he was not aware of a single piece of writing concerning any steps that the company took to comply with the injunction. (Pl.'s 56.1 ¶ 16.) By as early as September 23, 1993, Burlington began to violate the injunction. The 1987 Injunction put Burlington on notice, and Burlington's inaction in light of the injunction shows a complete indifference to the authenticity of the merchandise it purchased and sold.

Furthermore, Burlington's continued sale of counterfeit goods after receiving the April 2004 cease and desist letter establishes willfulness in and of itself. In April 2004, Fendi sent a cease and desist letter to Burlington. Burlington did not order the removal of Fendi-branded goods from its shelves until after Fendi commenced the instant litigation in 2006. Courts have repeatedly found willfulness where a defendant receives a cease and desist letter but continues the infringing conduct. See Motorola, Inc. v. Abeckaser, No. 07 Civ. 3963(CPS), 2009 WL 962809, at *8 (E.D.N.Y. Apr. 8, 2009) ("Even if defendants did not know that the goods they sold following receipt of the customs notices and cease-and-desist letter were counterfeit, which appears doubtful from the record, no reasonable juror could conclude that defendants did not at the very least recklessly disregard the possibility that the goods were counterfeit."); Microsoft Corp. v. Atek 3000 Computer Inc., No. 06 Civ. 6403(SLT), 2008 WL 2884761, at *3 (E.D.N.Y. July 23, 2008) (defendant's continued actions despite being apprised of its infringing conduct, were sufficient to find defendant's conduct willful); Merchant Media, LLC v. H.S.M. Int'l, No. 05 Civ. 2817(JES), 2006 WL 3479022, at *5 (S.D.N.Y. Nov. 30, 2006) ("[T]he fact that defendants continued to sell the infringing items in the face of two cease-and-desist letters from plaintiffs strongly suggests the willfulness of their conduct."); Microsoft Corp. v. Compusource Distributors, 115 F. Supp. 2d 800, 809 (E.D.Mich.2000) (finding willfulness where upon receipt of a cease and desist letter, the defendant called only two of its vendors and then continued to sell products from the same suppliers "without making any attempt to investigate their assurances that they were distributing genuine Microsoft software and hardware.").

Burlington asserts that it believed that the April 2004 letter was a bluff in an attempt to scare Burlington into taking Fendi merchandise off its shelves. (Def.'s Response to 56.1 ¶ 220.) Burlington points to the approximately seventeen month time lag between the first and second cease and desist letters and Fendi's failure to bring suit until 2006 as evidence that Fendi was bluffing. In Tanning Research Lab., Inc. v. Worldwide Import & Export Corp., 803 F. Supp. 606 (E.D.N.Y.1992), the plaintiffs sent letters to the defendants in 1988 and 1990, which informed the defendants of their Lanham Act violations. Tanning Research, 803 F.Supp. at 609. The defendants did not contact the plaintiffs for clarification of the letter but rather relied on "hearsay and self-interested representations" and the plaintiffs' failure to sue. Id. at 609-10. The court found that the defendants were nonetheless at least willfully blind from the date they received the letter in 1988 accusing them of infringement. Id. at 610.

Even accepting the evidence in the light most favorable to the defendant, we find that Burlington has failed to provide any evidence creating a genuine issue of material fact regarding willfulness. Burlington received a cease and desist letter and contacted Colton, who supplied the bag at issue. Four months later Colton sent Burlington an unsigned letter purportedly from Minerva, which stated that the bag was authentic. Burlington did nothing else to investigate the authenticity of the bag, despite the fact that the letter was unsigned and that Burlington knew that Fendi contested the authenticity of the letter. In addition, Haigney testified that buyers were to speak to him prior to purchasing any trademarked goods. (Pl.'s 56.1 ¶¶ 255-57.) Yet, Burlington's buyers did not always clear their purchases of Fendi-branded goods with the legal department. (Pl.'s 56.1 ¶ 257.) In October 2003, Haigney testified that a buyer approached him regarding the purchase of Fendi-branded goods from Colton. (Genecin Dec. Ex. 37 ("Haigney Dep.") at 113.) Haigney told the buyer not to buy the goods because Fendi previously had sued Burlington. (Haigney Dep. at 113-15.) This incident occurred just six months before Fendi's first cease and desist letter. Haigney knew that to prove authenticity he would need to provide "real documentation," (Haigney Dep. at 115), not an unsigned letter from an interested party. Burlington's failure to investigate the nature of the Fendi goods after the April 2004 cease and desist letter, in light of the 1987 Injunction and Burlington's failure to comply with its own internal procedures, demonstrates that Burlington was willfully blind.

Having determined that Burlington acted willfully, we find that Fendi is entitled to treble damages, along with reasonable attorney's fees,^[11] for violations on or after the date of the receipt of the April 2004 letter.^[12] In light of Magistrate Judge Dolinger's prior dealings on these issues with regard to the contempt action, we refer this matter to him for a report on the appropriate amount of damages, attorney's fees, prejudgment interest, and costs.

d. Supplemental Permanent Injunction

Fendi also moves to supplement the 1987 Injunction. Burlington began to violate the 1987 Injunction by receiving Fendi-branded goods into its inventory at least as early as 1993. (Pl.'s 56.1 ¶ 20.) Burlington did not stop selling Fendi-branded products until March 3, 2008, almost five months after this Court found Burlington in contempt.^[13] (Pl.'s 56.1 ¶ 20-21.) As the Magistrate Judge noted in his previous recommendation, "Burlington has not acted with due diligence, especially in the wake of the specific findings by the District Court as to the nature and extent of their obligations under the consent injunction." Fendi v. Burlington Coat Factory Warehouse Corp., 642 F. Supp. 2d 276, 303 (S.D.N.Y.2009). Thus, we adopted the recommendation of a $1,000 forward-looking sanction, which we understand to mean $1,000 per each item sold in violation of the 1987 Injunction.

"Injunctive relief should be narrowly tailored to fit specific legal violations. Accordingly, an injunction should not impose unnecessary burdens on lawful activity." Patsy's Brand, Inc. v. I.O.B. Realty, Inc., 317 F.3d 209, 220 (2d Cir. 2003). A district court is permitted to fashion an "injunction which will keep a proven infringer safely away from the perimeter of future infringement." Id. "[A] party who has once infringed a trademark may be required to suffer a position less advantageous than that of an innocent party." Oral-B Labs., Inc. v. Mi-Lor Corp., 810 F.2d 20, 24 (2d Cir.1987). The last twenty-three years supplies ample evidence of how little regard Burlington has for the 1987 Injunction; we cannot trust that Burlington will develop, let alone implement, compliance measures on its own.^[14] Therefore, within sixty days of this order, Burlington is directed to submit to this Court a written report, outlining in detail the manner and form in which Burlington intends to comply with the 1987 Injunction. The compliance plan must be specific to the requirements of the 1987 Injunction, which prohibits Burlington from purchasing or selling any Fendi-branded products without prior permission. The Court will not be satisfied with generic internal procedures intended to prevent trademark infringement. Six months after that date and every six months thereafter for the ensuing five years, Burlington is to submit to the Court a report of full compliance.

III. Burlington's Motion to Amend the Answer

Burlington seeks to amend its answer to include a statute of limitations defense to the contempt claim and limit Fendi's right to recovery to the three years prior to the instant lawsuit. Burlington first raised a statute of limitations defense on April 1, 2008 before Magistrate Judge Dolinger. Although the parties did not discover the sales dated back to 1993 until after our 2007 Opinion, Burlington was aware at the time of its Answer and throughout its briefing of the contempt claim that sales of Fendi-branded products occurred in 2002. Therefore, we adopted the Magistrate Judge's determination that Burlington waived the statute of limitations defense when it raised the affirmative defense of laches in its Answers and prior motion papers but did not raise a statute of limitations defense. Fendi, 642 F.Supp.2d at 285 n. 3. Furthermore, the Magistrate Judge found the statute of limitations defense was no longer available to Burlington because a final judgment on liability had already been rendered. Id.; see Petramale v. Local Union 17, Laborers' Int'l Union of North America, 625 F. Supp. 775, 782 (S.D.N.Y.1986), rev'd on other gds., 847 F.2d 1009 (2d Cir.1988) ("Since a final judgment has been rendered, the defense of statute of limitations can no longer be raised.... We must, therefore, deny [the] motion to amend its answer to include the defense of statute of limitations."). Burlington's motion to amend the answer to include a statute of limitations defense to the contempt claim is denied.

IV. Burlington's Motion for Partial Summary Judgment Against the Third-Party Defendants; Colton's Cross-Motion to Dismiss the Third Party Complaint

Burlington moves for partial summary judgment against the Third Party Defendants, the Vendors, as to liability only on its claim for indemnification. Colton cross-moves to dismiss the third party complaint. The Vendors—Colton, Summit, Euro Moda, Moda Oggi, and Ashley Reed—sold Fendi-branded merchandise to Burlington. A standard form purchase order (the "Purchase Order") was used for those sales. In the Purchase Order, the Vendors expressly warranted the authenticity of the merchandise they supplied. The Purchase Order also guaranteed that the Vendors would indemnify Burlington in trademark infringement actions.

a. Burlington's Motion for Partial Summary Judgment Against the Third-Party Defendants

The Vendors^[15] allege that Burlington could not have relied on the warranty in the Purchase Order because the 1987 Injunction prohibited Burlington from the purchase or sale of any Fendi-branded products. A prima facie claim for breach of express warranty requires the plaintiff to "show that there was an affirmation of fact or promise by the seller, the natural tendency of which [was] to induce the buyer to purchase and that the warranty was relied upon to the plaintiff's detriment." Nealy v. U.S. Surgical Corp., 587 F. Supp. 2d 579, 584 (S.D.N.Y.2008) (internal citations and quotations omitted). Burlington has demonstrated through the Purchase Order that the Vendors made a promise as to the authenticity of the goods. Furthermore, Burlington relied on that promise to its detriment. Regardless of the 1987 Injunction, Burlington would not be subject to the instant trademark infringement and counterfeit actions but for the Vendors breach of this warranty. The Vendors have pointed to no case law, nor even legal proposition, to support its argument that because of the 1987 Injunction, Burlington could not have relied on the warranty as to the authenticity of the goods. Burlington's motion for partial summary judgment as to liability is granted.^[16]

b. Colton's Cross-Motion to Dismiss the Third Party Complaint

Colton moves to dismiss the Third Party Complaint based on the forum selection clause in the Purchase Order. In order to determine whether to dismiss a complaint based on a forum selection clause the court conducts a four step analysis: (1) whether the clause was reasonably communicated to the party resisting enforcement; (2) whether the clause is mandatory or permissive; (3) whether the claims and parties involved in the suit are subject to the forum selection clause; and (4) whether the resisting party has rebutted the presumption of enforceability by making a sufficiently strong showing that "enforcement would be unreasonable or unjust, or that the clause was invalid for such reasons as fraud or overreaching." Phillips v. Audio Active Ltd., 494 F.3d 378, 383-84 (2d Cir.2007).

The forum selection clause states that "any dispute(s) with respect to the performance and/or interpretations of this contract ... will be determined under the laws of the State of New Jersey and that any lawsuit(s) arising from any such dispute(s) shall be venued exclusively in the New Jersey Superior Court for Burlington County to whose jurisdiction the Vendor hereby submits for all such lawsuit(s)." (Colton Opp. 3-4.) We can easily dispose of the first two inquires. First, no reasonable communication of the clause was necessary, as Burlington drafted the clause. Second, the clause confers exclusive jurisdiction and, therefore, is mandatory. See Phillips, 494 F.3d at 386 ("A forum selection clause is viewed as mandatory when it confers exclusive jurisdiction on the designated forum or incorporates obligatory venue language."). The third inquiry is whether the claims and parties involved in the suit are subject to the forum selection clause. Burlington alleges that it can proceed on its claims under the Uniform Commercial Code in this Court, even though indemnification based on the Purchase Order warranty may be venued in New Jersey. However, the clear language of the clause states that it applies to "any dispute(s) with respect to the performance and/or interpretations of this contract." (Colton Opp. 3-4.) Any action for indemnification based on Colton's performance of the contract falls within the scope of the clause.

"If the forum clause was communicated to the resisting party, has mandatory force and covers the claims and parties involved in the dispute, it is presumptively enforceable." Phillips, 494 F.3d at 383. At the fourth step of the inquiry, Burlington must rebut "the presumption of enforceability by making a sufficiently strong showing that enforcement would be unreasonable or unjust, or that the clause was invalid for such reasons as fraud or overreaching." Id. at 383-84. Burlington has failed to make any showing, let alone a showing sufficient to overcome the presumption of enforceability. We could postulate the prejudice that could result from a dismissal, such as increased expenses and duplicative proceedings. But given that Burlington has made no showing of prejudice, and we can foresee no such prejudice sufficient to overcome the presumption of enforceability. The forum selection clause is valid.

Burlington alleges that Colton has waived its right to enforce the forum selection clause. Burlington filed the Third Party Complaint in February 2006. For three years, Colton did not invoke the forum-selection clause.^[17] Colton answered the complaint with only a generic 12(b)(6) defense, and Colton did not object to this Court's jurisdiction. During the past three years, Colton has participated in discovery.

The waiver of a chosen forum "is not to be lightly inferred." Diesel Props S.r.L. v. Greystone Business Credit II LLC, No. 07 Civ. 9580(HB), 2008 WL 4833001, at * 15 (S.D.N.Y. Nov. 5, 2008) (citing Rush v. Oppenheimer & Co., 779 F.2d 885, 887 (2d Cir.1985)). A court may infer waiver where a long delay has caused prejudice to the opposing party. Mateco Inc. v. M/V Elli, 103 F. Supp. 2d 70, 73 (D.P.R.2000) (finding a waiver because if the forum-selection clause was after a year and a half delay "would inflict duplicative and unduly complex procedures upon [Plaintiffs]. They would incur severe prejudice through delay, expense, and uncertainty."). In Jockey Int'l v. M/V "LEVERKUSEN EXPRESS," Judge Haight distinguished Mateco, emphasizing that "the most significant factor was the delay in asserting the right which prejudiced the parties as well as the court." 217 F. Supp. 2d 447, 456 (S.D.N.Y.2002). Although Colton delayed three-years in raising the forum-selection clause, Colton's role in the lawsuit to date has been somewhat marginal. Dismissal of this case will lead to some increased expenses and duplicative proceedings, but Burlington has failed to allege any prejudice it will suffer from the enforcement of the forum-selection clause that it authored itself.

In LPR, SRL v. Challenger Overseas LLC, Judge Koeltl refused to find a waiver and found "no reason to depart from this usual rule in this case because the plaintiff had sufficient notice of the forum selection clause. It chose to bring the action in this Court in the face of a clear forum selection clause." LPR, SRL v. Challenger Overseas, LLC., No. 99 CIV. 8883(JGK), 2000 WL 973748, at *7 (S.D.N.Y. July 13, 2000). Burlington knew that the forum-selection clause venued this action in New Jersey but believed that it would be more efficient to have its third-party indemnification claims adjudicated in this Court where the underlying action giving rise to those claims was being adjudicated. (Burlington Reply Partial Summ. J. 2.) Burlington made the decision to bring this case in this Court with full knowledge of the forum-selection clause it drafted, and has alleged no prejudice that will result from the enforcement of that clause. We find that the forum selection clause is enforceable and Colton's motion to dismiss the third party complaint is granted.

V. Burlington's Objections to Magistrate Judge Dolinger's December 1, 2009 Report and Recommendation

Noting that the Court enjoys wide discretion in fashioning an appropriate judgment with regards to the amount and method of calculation of prejudgment interest, see, e.g., Jones v. UNUM Life Ins. Co., 223 F.3d 130, 139 (2d Cir.2000), Magistrate Judge Dolinger engaged in a carefully reasoned analysis in the December 1, 2009 Report and Recommendation ("R & R") that attempted to address the equities of the present case. See Jones, 223 F.3d at 139 (The goal of crafting an appropriate prejudgment interest calculation is to adequately address the "circumstances of the individual case" by considering the "remedial purpose" of the judgment, the "fairness and the relative equities of the award," and "the same considerations that inform the court's decision whether or not to award interest at all."). Both Fendi and Burlington agree with Magistrate Judge Dolinger's selection of the interest rate in 26 U.S.C. § 6621(a)(2), but Burlington objects to several elements of the proposed calculation. Fendi urges us to adopt the R & R.

We note at the outset that due to the paucity of Burlington's own records, Magistrate Judge Dolinger crafted his proposed calculation in the face of considerable uncertainty as to the timing and quantity of Burlington's contemptuous sales. (See, e.g., R & R at 620-21.) Burlington objects to Magistrate Judge Dolinger's recommendation that net profits from Burlington's sales of Fendi goods from 1993-1999 should be spread evenly over that time period due to Burlington's incomplete records. Burlington protests that profits for those years should be allocated based on season codes, which have been used by both parties' experts to estimate sales timing. (Def.'s Objections 8.)

However, Magistrate Judge Dolinger's recommendation was based in part on recognizing the very real possibility that the disgorgement amount did not capture all contemptuous sales, due to Burlington's lack of records for the years 1987-1993. (R & R at 620-21.) For this reason, we adopt Magistrate Judge Dolinger's recommendation as to the allocation of the profits over time, as well as his recommendation (1) to treat profits as if they were made on the first day of each fiscal year and (2) to spread shipping costs evenly over the relevant time period.^[18]

Burlington objects to using a fixed interest rate rather than a floating rate that is determined quarterly, as is used by the IRS in calculating interest on tax underpayments. See 26 U.S.C. § 6621(a)(2). However, as is made clear by Magistrate Judge Dolinger's decision not to compound interest daily, thoroughgoing adherence to the statutory scheme is not the goal in the instant matter. Rather, the goal is to craft a prejudgment interest calculation that adequately addresses the unique circumstances and equities of the case, which the R & R accomplishes admirably and persuasively. Burlington's objection is rejected.

Burlington next objects to the calculation of income taxes attributable to contemptuous sales, and claims it should be entitled to deduct $973,732 instead of $152,528 in income taxes. Because we adopted Magistrate Judge Dolinger's disallowance of the deduction of "store expenses" from Burlington's gross profit on the sale of Fendi goods, Burlington objects that it should be able to deduct more income tax than was recommended in the December 1, 2009 R & R. Mr. Berliner, Burlington's accounting expert, originally calculated income taxes based on the assumption that store expenses would be deductible, and applied a tax rate of 38% to pre-tax profits net of store and shipping expenses, resulting in an income tax provision of $152,528. This figure was adopted by Magistrate Judge Dolinger despite the disallowance of the store expenses deduction. Burlington claims it should be allowed to deduct a tax provision equaling 38% of a pre-tax profit figure unreduced by the store expenses deduction, resulting in an income tax provision of $973,732.

However, when income taxes are determined for the purpose of reducing a damage award, the goal is to ascertain as nearly as is practicable the actual amount of income taxes paid that were attributable to the contemptuous sales. See In Design v. K-Mart Apparel Corp., 13 F.3d 559, 566-67 (2d Cir.1994) ("[The district court] should have allowed K-Mart a deduction for the taxes it paid on its innocently-acquired unlawful profits.") (emphasis added); cf. Sheldon v. Metro-Goldwyn Pictures Corp., 106 F.2d 45, 54 (2d Cir.1939) (overhead that does not assist in production of infringing item "should not be credited to the infringer; that which does, should be"), aff'd, 309 U.S. 390, 60 S. Ct. 681, 84 L. Ed. 825 (1940). Furthermore, "the burden is on the contemnor `to prove any deductions for its costs from the gross revenues attributable to its contempt....'" Oral-B Laboratories, Inc. v. Mi-Lor Corp., 810 F.2d 20, 26 (2d Cir. 1987) (abrogated on other grounds as recognized in Paddington Corp. v. Attiki Importers & Distributors, Inc., 996 F.2d 577 (2d Cir.1993)). Regardless of the disallowance of the store expenses deduction from the damage award, the burden is on Burlington to show that the higher amount of taxes was actually paid and was distributed to the infringing sales. Because Burlington has not made this showing, the court finds Mr. Berliner's original estimation of income taxes paid and allocable to infringing sales more plausible. Accordingly, Burlington's objection is rejected.

For the foregoing reasons, we adopt Magistrate Judge Dolinger's December 1, 2009 Report and Recommendation in its entity.

VI. Conclusion

Fendi's motion for summary judgment is granted, and we refer the matter to Magistrate Judge Dolinger for a report on the appropriate amount of damages, attorney's fees, prejudgment interest and costs. Within sixty (60) days of this Order, Burlington is directed to submit a compliance plan to the Court. Six months after that date and every six months thereafter for the ensuing five years Burlington is to submit a compliance report to the Court. Cotton's cross-motion to dismiss the third-party complaint is granted. Burlington's motion for partial summary judgment against Ashley Reed, Euro Moda, Moda Oggi and Summit is granted. The December 1, 2009 Report and Recommendation is adopted in its entity.

SO ORDERED.

MEMORANDUM AND ORDER

On February 8, 2010, this Court granted summary judgment in favor of Plaintiffs Fendi Adele S.R.L., Fendi S.R.L., and Fendi North America (collectively known as "Fendi"), finding Defendants Burlington Coat Factory Warehouse Corporation and Cohoes Fashion, Inc., a wholly-owned subsidiary of Burlington Coat Factory, (collectively known as "Burlington") liable for trademark counterfeiting, trademark dilution, and common law unfair competition. Fendi Adele S.R.L. v. Burlington Coat Factory, No. 06 Civ. 85(LBS), 689 F. Supp. 2d 585, 600-01, 2010 WL 431509, at *10 (S.D.N.Y. Feb. 8, 2010) ("February 2010 order"). We awarded treble damages for the counterfeit goods sold after the April 2004 cease and desist letter and referred the matter to the Magistrate Judge for a calculation of damages.

Now before this Court is Fendi's Motion for Partial Reconsideration.^[1] Fendi seeks to clarify the damages awarded, if any, with regard to counterfeit goods sold prior to the April 2004 letter. If the sale of Fendi-branded goods prior to the cease and desist letter is not found to be willful, the issue remains as to whether the bad faith requirement still exists. The courts in this District are currently divided as to whether a finding of bad faith is still required in order to award treble damages in light of the Trademark Amendments Act of 1999. In our February 2010 order, we found the April 2004 letter established willfulness and did not reach the question of whether the bad faith requirement survives the 1999 Amendments. Fendi, 689 F.Supp.2d at 601-03, 2010 WL 431509, at *11. Fendi urges reconsideration of the issue of willfulness for the period prior to the April 2004 letter, or, if willfulness is not found, a determination as to whether the bad faith requirement survives the 1999 Amendments.

Upon reconsideration, we find that Burlington willfully infringed for the entire period. The Court's February 2010 order is modified to award Fendi treble damages for all of the counterfeit goods at issue.

I. Standard of Review

The standard for willfulness is whether the defendant had knowledge that his or her conduct represented infringement or recklessly disregarded the possibility. Kepner-Tregoe, Inc. v. Vroom, 186 F.3d 283, 288 (2d Cir.1999); see Koon Chun Hing Kee Soy & Sauce Factory, Ltd. v. Star Mark Mgmt., No. 04 Civ. 2293(JFB), 2007 WL 74304, at *11 (E.D.N.Y. Jan. 8, 2007) (applying the willfulness standard in Kepner-Tregoe to claims brought under the Lanham Act); Nike, Inc. v. Top Brand Co., No. 00 Civ. 8179(KMW), 2005 WL 1654859, at *6 (S.D.N.Y. July 13, 2005) (same). While caution must be exercised in granting summary judgment when state of mind is an issue, the summary judgment rule "would be rendered sterile" if the mere existence of an issue as to state of mind would automatically defeat an otherwise valid motion. Nora Beverages, Inc. v. Perrier Group of America, Inc., 269 F.3d 114, 125 (2d Cir.2001). Where the defendant offers no probative evidence raising a genuine issue of material fact regarding willfulness, summary judgment is appropriate. Tanning Research Lab., Inc. v. Worldwide Import & Export Corp., 803 F. Supp. 606, 610 (E.D.N.Y.1992).

II. Willfulness of Pre-2004 Sales

"Willful infringement may be attributed to the defendant's actions where he had knowledge that his conduct constituted infringement or where he showed a reckless disregard for the owner's rights." Johnson & Johnson Consumer Companies, Inc. v. Aini, 540 F. Supp. 2d 374, 396 (E.D.N.Y.2008); see also Cartier Int'l B.V. v. Ben-Menachem, No. 06 Civ. 3917(RWS), 2008 WL 64005, at *14 (S.D.N.Y. Jan. 3, 2008) (Conduct can be found to be willful where the defendant "recklessly disregarded the possibility" that his or her conduct constituted infringement.). Courts have repeatedly found willfulness where a defendant receives a cease and desist letter but continues the infringing conduct. See Fendi, 689 F.Supp.2d at 600-01, 2010 WL 431509, at *10 (collecting cases). Courts have also found willful infringement where the infringing party was found liable and then did not alter or cease the infringing activity. See Burberry Ltd. v. Euro Moda, No. 08 Civ. 5781(CM), 2009 WL 4432678, at *3 (S.D.N.Y. Dec. 4, 2009) (finding ample evidence of willfulness where a prior settlement put defendants on notice of the infringing activity, yet defendants persisted in the same conduct); Kepner-Tregoe, Inc., 186 F.3d at 288 (finding copyright infringement was willful where the defendant "chose to ignore the injunction [issued by a Texas district court and upheld by the Fifth Circuit] and continued to use the MPO program" that had been found to be infringing); Viacom Int'l Inc. v. Fanzine Int'l, Inc., No. 98 Civ. 7448(RCC), 2001 WL 930248, at *4 (S.D.N.Y. Aug. 16, 2001) (where the defendant had been sued at least two times in the previous year, with one suit resulting in a default judgment against the defendant, the court found that the defendant's subsequent "strikingly similar conduct" suggests that the defendant "knew or should have known that its actions constituted [copyright] infringement.").

When Burlington took no action to comply with the 1987 Injunction and began violating the injunction as early as 1993,^[2] Burlington showed a reckless disregard for Fendi's trademark rights. Johnson & Johnson, 540 F.Supp.2d at 396. The 1987 Injunction, like a cease and desist letter or a finding of liability, put Burlington on notice that its behavior was potentially infringing on Fendi's trademark. In fact, the injunction was even more specific than a cease and desist letter or a finding of liability because it mandated a specific course of action to protect against future infringement—Burlington was not to sell any Fendi-branded merchandise without first obtaining permission from Fendi. Burlington did not implement any control mechanisms to ensure compliance with the injunction. (Pl.'s 56.1 ¶ 16.) Burlington's in-house counsel, Stacy Haigney, who was present in court when the 1987 Injunction was entered, admitted that he was not aware of a single piece of writing concerning any steps that the company took to comply with the injunction. Id. Burlington did not even comply with its own trademark protection policies in the years following the injunction. Burlington's policies required that buyers (1) speak with Haigney prior to purchasing trademarked merchandise from third-party vendors, (Def. Response to 56.11254; Genecin Dec. Ex. 37 ("Haigney Dep.") at 156:20-157:23); and (2) obtain authentication documents for purchases from thirdparty vendors. (Pl.'s 56.1 ¶ 266.) Burlington purchased Fendi-branded goods from the Vendors as early as 2000. Haigney testified that the first time a buyer approached him regarding purchasing Fendibranded goods from a third-party vendor was not until 2003. (Haigney Dep. at 113:3-21.) No buyer ever sought approval from Haigney for any purchase of Fendibranded goods from Ashley Reed. (Pl.'s 56.1 ¶ 261.) Burlington did not obtain authentication documents for any of its purchases from the Vendors. (Pl.'s 56.1 ¶ 268.) Burlington was aware of the risk associated with buying trademarked goods from third-party vendors, yet it did not buy from authorized distributors, (Genecin Dec. Ex. 38 ("Goldstein Dep.") at 143:21-144:9; Pl.'s 56.1 11255), or verify the authenticity of the goods purchased from third-party vendors, (Goldstein Dep. at 143:3-20).

Burlington has failed to offer any evidence raising a genuine issue of material fact regarding its willfully blind infringement. See Tanning Research, 803 F.Supp. at 610 ("[D]efendants have offered no probative evidence raising a genuine issue of material fact regarding their willfully blind violation."). Burlington concedes that it took no action to implement procedures in order to ensure compliance with the 1987 Injunction, and it has offered no evidence to demonstrate that it complied with its own trademark protection policies. Burlington argues that it forgot about the injunction; however, whether Burlington forgot about the injunction is irrelevant. The crucial time for the purpose of finding willfulness is immediately after the 1987 Injunction was entered. At that moment, Burlington was on notice that its conduct potentially infringed on Fendi's trademark, and in doing nothing, Burlington demonstrated a reckless disregard for the injunction and for its purposeto prevent Burlington from infringing on Fendi's trademark in the future. See Burberry Ltd. v. Euro Moda, 2009 WL 4432678, at *3 ("In light of the prior June 30, 2005 settlement agreement where defendants admitted to selling scarves, hats and polo shirts with counterfeit Burberry marks, but stated it had ceased doing so in 2003 and promised not to do so in the future ..., defendants obviously knew that selling such items with counterfeit Burberry marks infringed Burberry's registered trademarks."). Burlington's failure to alter its conduct in the wake of the 1987 Injunction and lack of compliance with its own minimal trademark compliance procedures establishes willfulness for the entire period at issue.

III. Conclusion

This Court's February 8, 2010 Memorandum and Order is amended. Upon reconsideration, we find Burlington acted willfully for the entire relevant period, and Fendi is awarded treble damages with regard to all counterfeit goods at issue.

SO ORDERED.

REPORT & RECOMMENDATION

MICHAEL H. DOLINGER, United Stated Magistrate Judge.

TO THE HONORABLE LEONARD B. SAND, U.S.D.J.:

In this Report and Recommendation, we address whether the Court should use a state or federal benchmark in setting the rate at which prejudgment interest is to be assessed on an award of $2,528,768, representing disgorgement of defendants' profits from sales of Fendi goods for over fourteen years in violation of a court order. We then proceed to calculate accrued interest through December 14, 2009.

BACKGROUND

We set forth here the facts and procedural history of this case insofar as they are relevant to our recommendation as to the appropriate rate of prejudgment interest.^[1]

In 1987, Fendi agreed to discontinue a counterfeiting lawsuit against Burlington Coat Factory and Cohoes Fashion (hereinafter "defendants") in exchange for defendants' promise not to "purchase or sell any merchandise bearing the Fendi trademark unless permission in writing is received from Fendi." Fendi Adele S.R.L., 2007 WL 2982295, at *1. The settlement agreement was "so ordered" by this Court. Id.

At no point since then have defendants ever sought such permission from Fendi. Id. at *4. It is now clear that, with the passage of time, defendants—and plaintiffs—simply forgot about the 1987 consent decree. See id. at *1-2, 4. The lead-up to the instant litigation began with an exchange of letters between the parties between April 2004 and December 2005 in which plaintiffs alleged—and defendants denied—that Burlington was dealing in counterfeit Fendi products. Id. at *2. Presumably, Fendi's attorneys discovered the existence of the 1987 consent decree some time around December 2005, when they first mentioned it to Burlington in a letter. See id. Fendi filed this action on January 5, 2006. Id.; Compl., Docket Doc. 1.

In April 2006, plaintiffs moved for partial summary judgment, asking the district court to hold defendants in civil contempt for willful violations of the 1987 consent decree. They asked that, as a remedy, the Court order defendants to disgorge profits on their sales of goods bearing the Fendi trademark from 1987 to the present (with the disgorgement amount to be established by an accounting), and award plaintiffs attorneys' fees and costs. (Pls.' Mot. Summ. J., Doc. 51).

On October 10, 2007, the District Court found defendants in contempt of the 1987 court order. Fendi Adele S.R.L., 2007 WL 2982295, at *5. It found that Burlington had been selling Fendi goods without prior authorization since 2002, and that although defendant had taken steps to remove most Fendi products from its shelves after being reminded of the existence of the consent decree, it continued to sell Fendibrand perfume after that point. Id. The Court thus found Burlington Coat Factory's violation of the consent decree to be willful with respect to its sales of Fendi perfume after January 2006. Id.

The Court granted plaintiffs' request for disgorgement of defendants' profits, and ordered defendants to "open their financial books and statements to a consultant chosen by Plaintiffs for the purpose of determining profits from unauthorized sales of Fendi brand merchandise and the date of such sales." Id. at *6. The Court also granted plaintiffs' request for attorneys' fees and costs, noting the willfulness of Burlington's violation of the consent decree. Id. at *5.

Subsequently, Judge Sand referred the matter to me for a report and recommendation on the extent of the relief to be awarded to plaintiffs. (See Endorsed Order, Dec. 20, 2007, Doc. 160). Plaintiffs subsequent examination of defendants' records revealed that in fact they had begun selling Fendi goods as early as 1993. (Rep. & Recommendation, Apr. 28, 2009, Doc. 263 [hereinafter "R & R"] at 6). We also found that defendants had continued to sell some Fendi perfume even after Judge Sand's entry of judgment of contempt. (R & R at 49 (citing evidence of sales between October 10, 2007 and March 1, 2008)). The parties stipulated to the fact that gross revenues from sales in violation of the court order from 1993 through March 2008 totaled $2,772,945. (Id. at 7 (citing Mattiaccio Mar. 6, 2008 Letter at 1); see also Mem. & Order, Aug. 10, 2009, Doc. 293 at 3 (citing id.)).

Both sides submitted several rounds of briefing on the appropriate amount to be disgorged, arguing principally over the expenses that could be deducted from gross sales revenues to calculate net profits.^[2] Among other things, plaintiffs requested an award of prejudgment interest on the disgorgement amount, suggesting the use of the New York default statutory interest rate of 9 percent. (P's Mem. Law, Doc. 184, at 5-7). Plaintiffs calculated prejudgment interest through March 1, 2008. (Donohue Decl., Doc. 182, Ex. 13). Defendants did not explicitly oppose this application nor did they address prejudgment interest in any of their briefs.

I. Prior R & R on Prejudgment Interest

On April 28, 2009, I recommended that the Court award plaintiffs $2,528,768 as disgorgement of defendants' net profits from sales between 1993 and February 2008 in violation of the 1987 consent decree. (See R & R at 7, 51-52; Mem. & Order, Aug. 10, 2009, Doc. 293 at 2 (adopting recommendation as to calculation of net sales)). This included offsets of $91,649 in shipping expenses and $152,528 in income taxes based on rough estimates of these expenses provided by defendants. (R & R at 32). I also recommended that the Court award $541,913.65 in attorneys' fees and costs and prejudgment interest at the rate specified in 26 U.S.C. § 6621(a)(2) for the underpayment of federal taxes. (Id. at 52).

In recommending that the Court assess prejudgment interest, I noted that, although it is discretionary, prejudgment interest is appropriate in this case because it is consistent with the rationale for disgorgement: preventing unjust enrichment of defendants by the interest-free use of ill-gotten profits during the more than fourteen years during which they were in violation of the consent decree. (Id. at 47). I observed that the underpayment rate had been used to measure prejudgment interest in comparable cases arising under federal law. (Id. at 48). In a footnote, I opined that the nine-percent rate urged by plaintiffs was "divorced from the economic realities that should be the guide in circumstances of this sort." (Id. at 48 n. 30 (citing SEC v. First Jersey Sec., Inc., 101 F.3d 1450, 1476-77 (2d Cir.1996))).

In their objections to my report and recommendation, plaintiffs cited decisional law from within this circuit for the proposition that it is more appropriate to apply the New York statutory default rate of 9 percent interest than to apply the federal rate embodied in 26 U.S.C. § 6621(a)(2), which is adjusted quarterly but is generally lower than 9 percent. (Pls.' Partial Objection to Rep. & Recommendation, May 11, 2009, Doc. 273, at 9). Plaintiffs also argued that the particular facts of this case make a higher rate more appropriate than the one we recommended, pointing to defendants' continued violations of the consent decree even after being reminded of its existence, their prolonged resistance to providing an accounting, and their understatements and concealment of the extent of sales in violation of the consent decree. (Id. at 9-10). Finally, plaintiffs suggested that, even if this court were to borrow the federal statutory rate for nonpayment of taxes, it should borrow the rate from a different subdivision—section 6621(c), which sets a higher rate for underpayment of taxes in excess of $100,000.00—and should compound the interest. Defendants did not address prejudgment interest in their objection to my report and recommendation. (See Defs.' Objection to Rep. & Recommendation, May 12, 2009, Doc. 274).

II. The Instant Referral

In its subsequent memorandum and order, the Court adopted all of my recommendations except as to the interest rate to be applied to the disgorgement award. As to that issue, it remanded to me for a report and recommendation addressing in greater depth the appropriate benchmark to use in setting the rate of prejudgment interest to be awarded to plaintiffs.^[3] (See Mem. & Order, Aug. 10, 2009, Doc. 293 at 8).

In the wake of the remand, we invited further briefing on this question. (Order, Aug. 11, 2009, Doc. 294; see also Mem. & Order, Aug. 10, 2009, Doc. 293 at 6 (observing that defendants had not responded to plaintiff's objection to R & R's recommendation as to prejudgment interest)).

A. Plaintiffs' Arguments

Plaintiffs reiterate that this Court should assess 9 percent annual interest on the disgorgement amount, or in the alternative, interest calculated under 26 U.S.C. § 6621(c), compounded annually. (See Pls.' Mem. Law Concerning Calculation of Pre-Judgment Interest, Aug. 24, 2009, Doc. 304 [hereinafter "Pls.' Mem."] ). They acknowledge that the appropriate rate is a matter of the district court's discretionary judgment as to what seems reasonable given the context of the case. (See Pls.' Mem. at 3). They argue that the facts of this case militate toward the application of "the highest rate for which there is precedent in this [C]ourt," which plaintiffs identify as the New York statutory rate of 9 percent, citing two decisions imposing that rate for awards under the Lanham Act. Id. at 3 (citing Nat'l Ass'n for the Specialty Food Trade, Inc. v. Construct Data Verlag AG, 2007 WL 656274 (S.D.N.Y. Feb. 23, 2007); GTFM, Inc. et al. v. Solid Clothing, Inc., 2002 WL 31886349 (S.D.N.Y. Dec. 26, 2002)); see also Pls.' Reply Mem. Concerning Calculation of Pre-Judgment Interest, Doc. 309 [hereinafter "Pls.' Reply Mem."] at 4.

Plaintiffs identify several facts that they believe are relevant to the choice of interest rate. First, they note that defendants' poor record-keeping makes it impossible to know the full extent of their violations before May 31, 1999. (Pls.' Mem. at 3; Pls.' Reply Mem. at 2). In connection with this observation, they argue that, generally, doubts should be resolved against the contemnor. (Pls.' Reply Mem. at 2 (citing Goya Foods, Inc. v. Wallack Mgmt. Co., 344 F.3d 16, 21 (1st Cir.2003)). Second, they point out that the imprecision of defendants' records also makes unknowable the precise dates of contemptuous sales known to have occurred. (Pls.' Mem. at 3; Pls.' Reply Mem. at 3). The necessary result of this state of affairs, plaintiffs claim, is that, first, interest on the profits from all sales will be calculated as if they all took place on the last day of the fiscal year in which they occurred, and defendants will not be charged interest for any months between the actual sale date and the end of that fiscal year, and, second, all sales between 1993 and May 31, 1999 (the end of defendants' fiscal year 1999) will be treated as if they had occurred on May 31, 1999, and defendants will escape paying interest at all for almost six years. (Pls.' Mem. at 2; Pls.' Reply Mem. at 2-3). Third, plaintiffs opine that defendants' violation of the contempt order was particularly egregious, lasting over fourteen years and yielding large profits. (See Pls.' Mem. at 2, 5). Fourth, plaintiffs point to instances of defendants' conduct during this litigation that, according to plaintiffs, justify a higher interest rate. (Pls.' Mem. at 3-4 (citing R & R at 35-37)).

B. Defendants' Arguments

Defendants contend that the interest rate contained in 26 U.S.C. § 6621(a)(2), which I previously recommended, is appropriate in this case. (Defs.' Mem. Law Concerning Rate of Pre-Judgment Interest, Doc. 305 [hereinafter "Defs.' Mem."]). They argue that this is so because this rate of interest fully compensates plaintiffs by depriving defendants of their contemptuous gains, (id. at 4), because a higher rate would be a "windfall" for plaintiffs (id. at 3), and because the section 6621(a)(2) rate "lacks any unfairness" to defendants (id. at 4).

ANALYSIS

Both sides acknowledge that this Court has wide discretion in selecting an appropriate rate of prejudgment interest to be applied to the award in this case. (See Defs.' Mem. at 2; Pls.' Mem. at 3). Although there is a federal statute that sets the interest rate to be applied post-judgment to "any money judgment in a civil case recovered in a district court," 28 U.S.C. § 1961(a), there is no analogous statute governing prejudgment interest generally. Jones v. UNUM Life Ins. Co., 223 F.3d 130, 139 (2d Cir.2000). In exercising their discretion, courts in this circuit have adopted a wide variety of prejudgment interest rates.^[4]

I. Available Statutory Interest Rates

Although courts in this circuit do not consistently apply any single prejudgment interest rate, they consistently borrow one of the many interest rates set out in statute to govern assessment of interest in other circumstances.^[5] We describe here the various statutory rates contemplated by the parties as possible candidates for application in this case.

New York law prescribes that, in the absence of another applicable statute prescribing the interest rate, judgments shall be assessed interest at a constant rate of 9 percent per annum from the earliest date on which the cause of action existed through to payment of the judgment. N.Y. C.P.L.R. §§ 5001(b), 5004 (McKinney 2009). The federal statutes setting the postjudgment interest rate (see 28 U.S.C. § 1961) and the rates of interest applicable to underpayment of taxes generally (see 26 U.S.C. § 6621(a)(2)) and to large underpayments of corporate taxes (see 26 U.S.C. § 6621(c)) all cross-reference rates determined by a federal agency and changed periodically.

Section 1961 sets postjudgment interest on federal judgments at the "weekly average 1-year constant maturity Treasury yield, as published by the Board of Governors of the Federal Reserve System, for the calendar week proceeding [the date of judgment]." 28 U.S.C. § 1961(a). During the prejudgment interest period in this case,^[6] this weekly average yield fluctuated between a low of 0.29 percent per annum (for the week ending November 20, 2009) and a high of 7.24 percent per annum (week ending Jan. 6, 1995). See Federal Reserve Statistical Release Historical Data, http://www.federalreserve.gov/ Releases/H15/ (follow "Historical Data" link and then click "Weekly (Friday)" on the table under "Treasury bills constant maturity—1-year"). This weekly yield averaged approximately 3.92 percent per annum during this period. See id. (average of weekly rates for one-year constant-maturity nominal Treasury yield for the week ending June 4, 1993 through the week ending November 20, 2009, inclusive).

The statutory provision governing most underpayments of federal taxes sets interest on underpayments at the federal short-term rate established quarterly by the Secretary plus 3 percentage points, rounded to the nearest percent. 26 U.S.C. § 6621(a)(2), (b)(1), (b)(3). However, "large corporate underpayment[s]" accrue interest at the federal short-term rate plus 5 percentage points, rounded to the nearest percent. 26 U.S.C. § 6621(c). From June 1993 through November 2009, the general underpayment rate embodied in section 6621(a)(2) fluctuated from a low of 4 percent per annum^[7] to a high of 10 percent,^[8] averaging approximately 7.13 percent per annum. See Rev. Rul. 2009-37, Nov. 23, 2009, available at 2009 WL 4023061 (average of quarterly non-corporate underpayment rates for second quarter 1993 through fourth quarter 2009, inclusive). The large corporate underpayment rate fluctuated correspondingly from 6 percent to 12 percent, and averaged approximately 9.13 percent per annum. See id. (average of quarterly large corporate underpayment rates for second quarter 1993 through fourth quarter 2009, inclusive).

Although the New York statute was designed to govern the assessment of prejudgment interest, these federal statutory schemes were not. Thus, if this Court decides to borrow the interest rate embodied in one of the federal statutes, it would likely have to adapt the method of calculation to make sense in the context of prejudgment interest. For example, 28 U.S.C. § 1961 directs courts to calculate post-judgment interest by applying the relevant Treasury yield rate for the week preceding the entry of judgment to the entire period between judgment and the date of repayment. See 28 U.S.C. § 1961(a), (b). Using the rate at one point in time to govern the entire period makes a good deal of sense in the context of postjudgment interest; it simplifies the calculation of what defendants owe, avoiding any further delay in payment of a judgment that might potentially be created by a complex calculation scheme involving a fluctuating rate. However, the need for simplification of calculations is less compelling in the context of prejudgment interest, where the relevant time period over which interest must be calculated is predetermined. Moreover, in this case, application of the relevant Treasury yield rate on the week preceding the entry of judgment would be particularly unjust because it is likely that judgment will be entered at a point in time when the relevant Treasury yield rate is particularly low,^[9] whereas defendants had use of the money for over sixteen years during which the Treasury yield rate was frequently much higher.

Similarly, the statutory scheme governing the underpayment of federal taxes contemplates daily compounding of interest. However, taxes come due every year and the I.R.S. has a strong enforcement mechanism; the statutory scheme is not designed to accommodate the facts of this case, in which we must assess interest over a period of sixteen years. See 26 U.S.C. § 6621.

Not surprisingly, then, as another court in this circuit has observed, in cases in which a statutory benchmark is borrowed to calculate interest over time periods encompassing various short-term rates, courts in this Circuit use a wide variety of formulas to compute interest. Del Turco v. Taylor Tile Co., 2007 WL 2581882, at *4 (E.D.N.Y. Aug. 6, 2007); see also In re Crazy Eddie Sec. Litig., 948 F. Supp. 1154, 1167 (E.D.N.Y.1996) (observing that courts faced with practical difficulties regarding calculation of prejudgment interest may choose any reasonable method of calculation (citing Conway v. Icahn & Co., 16 F.3d 504, 512 (2d Cir.1994)).

Thus, our assessment of prejudgment interest is not necessarily complete once we have pointed to a statute setting an interest rate or defining a formula for a rate that we think should be applied in this case. Assessing prejudgment interest encompasses not only the identification of a particular rate-determining mechanism to apply, but also determinations as to other aspects of how to calculate prejudgment interest, such as whether and how often to vary the interest rate and whether and how often to compound the interest. As we will explain below, these considerations may in some circumstances affect the overall award of prejudgment interest even more than the court's choice of rate.

With this in mind, we turn to the question of how this Court should exercise its discretion as to the appropriate prejudgment interest rate and method of calculation in this case.

II. Factors Guiding Our Exercise of Discretion

In our choice of a rate of prejudgment interest, it is clear that the specific facts of the case at hand are paramount. See, e.g., Jones, 223 F.3d at 139 (stating that the appropriate rate of prejudgment interest "will depend on the circumstances of the individual case"); id. at 140 (remanding to district court for factual findings to support its choice of prejudgment interest rate).

Plaintiffs point to two decisions in which courts assessed prejudgment interest on awards for willful violations of the Lanham Act at the New York statutory rate of 9 percent per annum. (See Pls.' Mem. at 3 (citing Construct Data Verlag AG, 2007 WL 656274; GTFM, Inc., 2002 WL 31886349)). In both cases, the court invoked its broad discretion to set the prejudgment interest rate, but did not otherwise explain or seek to justify the use of the New York statutory rate. See Construct Data Verlag AG, 2007 WL 656274, at *6-7; GTFM, Inc., 2002 WL 31886349, at *2. Given the fact-intensive nature of the inquiry before us, these decisions, which are of course not controlling, are also not particularly instructive.

In assessing what prejudgment interest rate the facts support, a district court should draw guidance from the same factors that are relevant to its decision whether or not to award prejudgment interest at all: (1) "the need to fully compensate the wronged party for actual damages suffered," (2) "considerations of fairness and the relative equities of the award," (3) "the remedial purpose of the statute involved," and (4) "such other general principles as are deemed relevant by the court." Jones, 223 F.3d at 139 (quoting First Jersey Sec., Inc., 101 F.3d at 1476).

The first factor, compensation for actual damages, is not applicable to the instant context. As Judge Sand has previously noted, a plaintiff need not have actually suffered any damages in order to receive an award of disgorgement in a contempt action. See Fendi Adele S.R.L., 2007 WL 2982295, at *5 (citing Manhattan Indus., Inc. v. Sweater Bee By Banff, Ltd., et al., 885 F.2d 1, 6-7 (2d Cir.1989); Stephen King v. Allied Vision Ltd., 155 F.R.D. 440, 452 (S.D.N.Y.1994)). Rather than directly compensating the plaintiff for an injury suffered, disgorgement in this context is aimed at preventing defendant from being enriched by the violation of the court order. See id. Because the third factor commands us to consider an appropriate rate of prejudgment interest in light of the purpose of the legal regime at issue, prejudgment interest on a contempt award should be aimed in large part at stripping defendants of any benefit it may have obtained unjustly.

Defendants argue that any rate higher than that incorporated into 26 U.S.C. § 6621(a)(2) would be inappropriate because it would "give Fendi a windfall". (Defs' Mem. at 3). To the extent that they mean that an award of 9 percent interest would exceed plaintiffs' actual damages, this is beside the point. If, however, defendants' complaint is understood to mean that 9 percent interest would go further than necessary to strip defendants of any benefit from the use of the funds, this argument may have more merit. However, defendants do not explain why they believe 9 percent interest would go too far. Likewise, although defendants assert that the interest rate embodied in 26 U.S.C. § 6621(a)(2) will be sufficient to "deprive BCF of its contemptuous gains," they do not offer the reasoning by which they arrived at this conclusion. (Defs.' Mem. at 4).

A. Measurement of Unjust Enrichment

1. Return on investment

In measuring the benefit to defendants of the free use of the money, a court may consider the likely rate of return that defendants could have achieved by investing the funds. Jones, 223 F.3d at 139 (citing Diduck v. Kaszycki & Sons Contractors, Inc., 974 F.2d 270, 286 (2d Cir. 1992)).

The Second Circuit has suggested on several occasions that in estimating the rate of return on a hypothetical investment, the court should presume conservative investments. E.g., N.Y. Marine & Gen'l Ins. Co. v. Tradeline, LLC, 266 F.3d 112, 131 (2d Cir.2001) (rate of return on investment "should be measured by interest on short-term, risk-free obligations" such as treasury bills (citing Indep. Bulk Transp., Inc. v. Vessel "Morania Abaco," 676 F.2d 23, 27 (2d Cir.1982))). The rationale for this presumption seems to be that the risk inherent in high-yield investments could work either to the investor's advantage (yielding a high rate of return) or to his disadvantage (causing the loss of his investment), and that the court should estimate the likely returns ex ante, without the benefit of hindsight. See, e.g., Sec. Ins. Co. of Hartford v. Old Dominion Freight Line, Inc., 314 F. Supp. 2d 201, 204-05 (S.D.N.Y.2003) (declining to consider plaintiff's usual return on investment and explaining presumption of conservative investments by noting that plaintiff could have invested "to its advantage or disadvantage"). Thus, although we know now that defendants would have realized an average annual return of approximately 8.8 percent between 1993 and 2009 had they invested their ill-gotten profits in an S & P 500 index fund (a broad measure of stock market returns),^[10] for our purposes it is more appropriate to measure defendants' likely rate of return on investments during that period by looking to the rate of return on government-backed instruments such as treasury bills or treasury notes.^[11]

Treasury bill maturities range from one month to one year.^[12] During the prejudgment-interest period, the yield on three-month treasury bills ranged from as low as 0.0 percent per annum (on December 10, 18, and 24, 2008) to as high as 6.42 percent per annum (on November 6-7, 2000), and averaged approximately 3.61 percent per annum. See Federal Reserve Statistical Release, www.federalreserve.gov/ RELEASES/H15/ (average of daily nominal rates for Treasury 3-month constant maturities from June 1, 1993 through November 20, 2009, inclusive). Treasury notes are long-term government-backed debt obligations with maturities of greater than one year. 31 C.F.R. § 365.5(b)(5). During the prejudgment-interest period, the yield on ten-year treasury notes ranged from as low as 2.08 percent per annum (on December 18, 2008) to as high as 8.05 percent (on November 7, 1994), and averaged approximately 5.15 percent per annum. See Federal Reserve Statistical Release, www.federalreserve.gov/ RELEASES/H15/ (average of daily nominal rates for Treasury 10-year constant maturities from June 1, 1993 through November 20, 2009, inclusive).

Thus, an appropriately conservative estimate of what defendants might have earned had they invested the funds they obtained through their unlawful sales would be somewhere between approximately 3.6 and 5.2 percent per annum on average.

2. Cost of borrowing funds

The Second Circuit has indicated that, in measuring the extent of a defendant's unjust enrichment, it may also be appropriate to look to the rate that the defendant would have had to pay to borrow funds equivalent to its ill-gotten gains.^[13]First Jersey Sec., Inc., 101 F.3d at 1474-75. At issue in First Jersey Securities was a violation of the securities laws and civil RICO that the Court described as "a massive and continuing fraud" and a "wilfull[ ] and deliberate[ ]" violation of the law. Id. at 1459. The Court acknowledged that the purpose of disgorgement in that context was "to deprive violators of their ill-gotten gains, thereby effectuating the deterrence objectives of those laws." Id. at 1474 (internal citations omitted). The Court then approved the district court's use of the interest rate embodied in the federal statute governing the underpayment of taxes (26 U.S.C. § 6621) because this rate "reflects what it would have cost to borrow the money from the government and therefore reasonably approximates one of the benefits the defendant derived from its fraud." Id. at 1476; see also Aimsi Tech., Inc., 650 F.Supp.2d at 304-05 (applying 26 U.S.C. § 6621(a)(2) rate, citing id.).

Defendants seem to concede that it would be inappropriate not to consider what they would have had to pay to borrow the funds they enjoyed and to apply a lower rate that only reflected a conservative estimate of the likely rate of return. (Defs.' Mem. at 4 (the underpayment rate "reflects what it would have cost the defendant to borrow money from the government and, therefore, reasonably approximates one of the benefits the defendant derived from its fraud" (citing First Jersey Sec., Inc., 101 F.3d at 1476))). Indeed, defendants cite First Jersey Securities extensively, and argue that the rate that the Second Circuit upheld in that case (the general rate for underpayment of federal taxes embodied in 26 U.S.C. § 6621) is appropriate here. (Defs.' Mem. at 2-4).

Here, the average underpayment rate from 1993 to 2008 (approximately 7.13 percent per annum, see supra, at 15) is slightly higher than the likely annual rate of return on conservative government-backed investments (approximately 3.6 to 5.2 percent per annum, see supra, at 24). The Second Circuit's guidance in First Jersey Securities and defendants' implicit concession that it is appropriate for us to consider the cost of borrowing funds provide ample reason for us to conclude that prejudgment interest should be set at the federal underpayment rate, at a minimum. However, we do not believe that our analysis should end here for two reasons. First, plaintiff argues that the disgorgement principal actually falls short of encompassing all of defendants' ill-gotten gains because of the vast shortcomings of defendants' own bookkeeping, and thus, that it is appropriate in this case to use a higher interest rate than would normally be used to measure defendants' benefit from the free use of the funds in order to fully divest defendants of their spoils.^[14] Second, the Second Circuit has also advised district courts to consider the general balance of equities in setting the rate of prejudgment interest in a particular case. E.g., Jones, 223 F.3d at 139. Plaintiff argues that, here, this balance militates toward a higher rate of interest. We consider these arguments in turn.

B. Deficiencies in the Disgorgement Amount

As plaintiffs remind us, it is unlikely that the gross sales figure to which the parties stipulated captures all sales of Fendi goods in violation of the court order due to the incomplete nature of defendants' records. (E.g., Pls.' Mem. at 2). For one thing, defendants made no records of any kind available for any year before 1993. (Donohue Decl., Doc. 182, ¶ 15 n. 10). Thus, while plaintiffs know a sale was made as early as 1993,^[15] it is possible that sales could have occurred even earlier, but impossible to confirm or refute that hypothesis. Furthermore, defendants offered no records of sales before October 4, 2003, although plaintiffs were able to estimate sales for 1993 through 2003 based on information about the Fendi products stocked by defendants. (Id. ¶ 55 & nn. 43, 44). Plaintiffs also note that the exact date of sales cannot be known for any year because of quirks in defendants' recordkeeping. (E.g., Pls.' Mem. at 2; see also R & R at 30 & n. 20 (discussing the imprecision as to the timing of sales resulting from quirks in Burlington's recordkeeping)).^[16]

In this case, in addition to obscurities in determining the details of defendants' past practices because of their accounting failures, defendants have also demonstrated themselves to be highly reluctant to lay their accounts bare. (R & R at 37 (noting defendants'"prolonged resistance to being called to account for [their] violation of the injunction.")). We observed in our previous report and recommendation that it was far from clear that defendants were making their best efforts to obtain the historical data that was available to them. (Id. at 30 n. 19). Because of the limitations of defendants' records, we are faced here with a choice between charging defendants interest for extra days they may not actually have enjoyed the use of their ill-gotten profits and allowing them to escape paying interest for extra days on which they actually did enjoy the free use of these funds.

Plainly, the determination of the amount to be disgorged in order to prevent unjust enrichment and thus deter future violations of the law is not an exact science. First Jersey Sec., Inc., 101 F.3d at 1475 (disgorgement amount need only be a "reasonable approximation of profits causally connected to the violation" (quoting SEC v. Patel, 61 F.3d 137, 139 (2d Cir. 1995))). During this process, doubts should generally be resolved against the wrongdoer. Id. ("any risk of uncertainty [in calculating disgorgement] should fall on the wrongdoer whose illegal conduct created that uncertainty" (alteration in original) (quoting Patel, 61 F.3d at 139)); Aimsi Tech., Inc., 650 F.Supp.2d at 304-05 (citing id.); Acticon Tech., 2008 WL 356872, at *2 (where damages for willful patent infringement are uncertain because of inadequacy of evidence available from infringer, doubts must be resolved against the infringer) (internal citations omitted); see also Pls.' Reply Mem. at 2 (doubts regarding the precise timing of contemptuous sales should be resolved against the contemnor (citing Goya Foods, Inc., 344 F.3d at 21)).

Accordingly, courts have previously deemed it appropriate to increase the pre-judgment interest rate in order to compensate for deficiencies in the disgorgement calculation in a contempt action. See Warner Bros., Inc. v. Gay Toys, Inc., 598 F. Supp. 424, 433 (S.D.N.Y.1984) (explaining that prejudgment interest award would not create a "windfall" for plaintiff where plaintiff has received less than what it calculated total profits to be due to court's acceptance of defendant's accounting method). Our calculations of a principal amount to be disgorged and of prejudgment interest on that award are aimed at the same goal; they are complementary components of the contempt remedy. See West Virginia v. United States, 479 U.S. 305, 306, 107 S. Ct. 702, 93 L. Ed. 2d 639 (1987) (observing that "prejudgment interest is an element of complete compensation"). We believe that in this case, the limitations on our ability to make a comprehensive determination of defendants' profits do require some upward adjustment of the interest rate to compensate for the real possibility that defendants made sales of Fendi goods between 1987 and 1993 for which they cannot be held accountable.

However, rather than increasing the interest rate above the general underpayment rate contained in section 6621(a)(2), as plaintiffs urge, we believe that we can compensate for any deficiencies in the disgorgement amount due to defendants' poor recordkeeping through a sensible methodology for choosing the dates on which to assess prejudgment interest, which we will lay out below. The general underpayment rate, which averaged approximately 7.13 percent per annum during the prejudgment interest period, modestly exceeds the return that plaintiffs would have enjoyed from most reasonable investments, excluding stocks and other speculative financial instruments, and reflects the cost of borrowing from the government the funds whose use defendants enjoyed. It is thus an appropriate rate under the facts of this case.

C. Other Equitable Considerations

Finally, Fendi argues that the underlying facts of the case and defendants' questionable litigation tactics warrant application of the New York statutory interest rate of 9 percent per annum, which plaintiffs describe as "the highest rate for which there is precedent in this Court." (Pls.' Mem. at 3). We disagree.

It is true that we previously observed that defendants' violations were "extreme and extended" and persisted even after they were reminded of the consent decree, and indeed even after the Court held them in contempt.^[17] (R & R at 43, 51). But the violations of the law at issue in First Jersey Securities, where the Second Circuit upheld the application of the rate embodied in 26 U.S.C. § 6621(a)(2), were equally deliberate and egregious. 101 F.3d at 1459. We also do not believe that defendants' behavior during litigation, including their initial failure to disclose profits and concealment of some profits,^[18] demands a 9 percent interest rate.

However, while we do not agree with plaintiffs' choice of interest rate, these equitable considerations are certainly relevant to the appropriate exercise of our discretion. We believe that they militate towards a choice of methodology for calculating prejudgment interest at the rate set forth in 26 U.S.C. § 6621(a)(2) that favors plaintiff.

III. Methodology for Calculating Prejudgment Interest

A. Dividing the Disgorgement Principal Over Time

Prejudgment interest is calculated "from the earliest ascertainable date on which the cause of action existed" through to the date of judgment. Briarpatch Ltd., L.P. v. Geisler Roberdeau, Inc., 513 F. Supp. 2d 1, 5 (S.D.N.Y.2007); A I Marine Adjusters, Inc. v. M/V Siri Bhum, 2007 WL 760415, at *5 (S.D.N.Y. Feb. 8, 2007). Thus, prejudgment interest in this case should be assessed from the date on which defendants realized the profits they are now being ordered to disgorge through to the date of entry of judgment for plaintiffs in an amount representing a disgorgement award including prejudgment interest. (See supra, at 13-14 n. 6).

Here, where the disgorgement principal encompasses profits made over more than fourteen years, it would be unfair to defendants to award interest on the entire disgorgement amount from the date of the earliest known sale (which plaintiffs estimate occurred some time during the early part of the fiscal year which began on June 1, 1993).^[19] However, there is no basis in the record for a finding as to how defendants' profits were distributed over the more-than-fourteen-year period in which they were profiting from sales of Fendi goods. The parties stipulated to a figure representing total sales between June 1, 1993 and March 1, 2008, and plaintiffs' accountant initially calculated prejudgment interest based on a rough estimate of when these sales occurred, treating all sales as if they had occurred on the last day of the fiscal year (and all sales known to have occurred between 1993 and 1999 as if they occurred on the last day of the 1998-99 fiscal year). (Donahue Decl., Doc. 182, Ex. 13).^[20] This Court then ruled that the appropriate means of calculating net profits yielded a total disgorgeable amount of $2,528,768. (R & R at 51; Mem. & Order, Doc. 293 at 2 (adopting recommendation as to calculation of net profits)). However, we did not make findings as to how the gross revenues and offsetting costs were spread out over the more-than-fourteen-year period of sales. Indeed, this would have been difficult to do. As noted, there is no evidence as to year-by-year revenues from sales of Fendi goods. As for costs, defendants were allowed to subtract from the total gross profits shipping expenses of $91,649 and income taxes of $152,528, but both of those figures were very rough estimates based on data from a few years rather than actual costs incurred each year between 1993 and 2008. With respect to shipping costs, defendants' accountant relied on data from only three years, and conservatively estimated shipping expenses for all the other years based on net sales. (R & R at 29 (citing Berliner Decl., Doc. 196, at 3 & n. 8)). He calculated income taxes on sales of Fendi goods by averaging the income taxes paid by defendants from 1998 to 2007, estimating taxes for 1993 through 1997 based on this average, and attributing a percentage of those estimated taxes to Fendi sales based on the assumption that Fendi sales represented 38 percent of net profits for the entire period. (R & R at 31 (citing Berliner Decl., Doc. 196, at 4)). We thus can hardly be said to have precise data on when defendants acquired the $2,528,768 in disgorgeable net profits.

In approximating how the total disgorgement award of $2,528,768 was distributed between June 1, 1993 (the first day of the fiscal year in which the first sale is known to have occurred) and March 1, 2008 (the first month in which no sale is known to have occurred, as plaintiffs stopped investigating at this point), we can use as a starting point plaintiffs' own figures as to roughly when the gross sales occurred. (See Donohue Decl., Doc. 182, Ex. 13).^[21] Since we lack an analogous time frame for when defendants incurred the costs that this Court allowed them to deduct, the fairest and most accurate method available to us for accounting for when defendants' profits were made is to spread this offset of $244,177.00 for shipping costs and income taxes evenly over the entire disgorgement amount, discounting the gross sales for each period proportionately to its percentage of total gross sales for 1993 to 2008.^[22] We offer here plaintiffs' determinations of gross sales revenues for each period alongside our own calculation of net profits for that period by this methodology:

----------------------------------------------------------------
                             Findings as to      Calculation of
  Time Period                 Gross Sales          Net Profits
----------------------------------------------------------------
6/1/1993-5/31/1999            $  775,091.00        $  701,439.48
----------------------------------------------------------------
6/1/1999-5/31/2000            $  229,929.00        $  208,080.44
----------------------------------------------------------------
6/1/2000-5/31/2001            $   52,348.00        $   47,373.73
----------------------------------------------------------------
6/1/2001-5/31/2002            $   92,433.00        $   83,649.73
----------------------------------------------------------------
6/1/2002-5/31/2003            $  326,617.00        $  295,580.85
----------------------------------------------------------------
6/1/2003-5/31/2004            $  322,681.00        $  292,018.86
----------------------------------------------------------------
6/1/2004-5/31/2005            $  507,499.00        $  459,274.89
----------------------------------------------------------------
6/1/2005-5/31/2006            $  407,611.00        $  368,878.55
----------------------------------------------------------------
6/1/2006-5/31/2007            $   75,619.00        $   68,433.45
----------------------------------------------------------------
6/1/2007-3/1/2008             $    4,462.00        $    4,038.01
----------------------------------------------------------------
     TOTAL                    $2,794,290.00        $2,528,768.00
----------------------------------------------------------------

(See Donohue Decl., Doc. 182, Ex. 13).

B. Calculating Simple Interest on Each Period

1. Date from which to calculate interest

In terms of the relevant day in each time period on which to begin calculating interest, for each period between June 1, 1999 and March 1, 2008, profits should be treated as if they were made on the first day of the fiscal year rather than the last day of that year.^[23] This is an appropriate methodology in light of the high degree of uncertainty as to when the illegal profits were earned (apart from the broad outlines determined by plaintiffs), the responsibility that defendants bear for that uncertainty, the unfairness that would result from assuming that all sales occurred on the last day of the fiscal year when the precise sales dates are unknown, and defendants' failure either to argue in support of the originally-advanced method or to offer an alternative method.

Plaintiffs were able to pinpoint sales after 1999 to a particular fiscal year. For the prior period, however, although defendants made $775,091.00 in gross sales of Fendi profits between June 1, 1993 and May 31, 1999, plaintiffs were unable to break these sales down to specific years with as much certainty as sales in subsequent years. (See Donohue Decl., Doc. 182, Ex. 13). The figure for sales during the 1993 to 1999 fiscal years constitutes 28 percent of the total gross sales the parties stipulated to ($2,794,290), and yet, under plaintiffs' original calculation, these sales were treated for interest purposes as if they had all occurred on the last day of this six-year period. This approach seems to reward defendants for the lack of clarity in their records, a result we would prefer to avoid. One alternative would be to treat this six-year period as we treated each subsequent one-year period, and to calculate interest on these sales as if they had all occurred on the first day of the period (June 1, 1993, the beginning of the fiscal year). However, we believe a fairer approach is to divide the gross sales and net profits for this period evenly among the six fiscal years it encompasses. See Appendix A. The resulting figures, while an hypothesized account of defendants' actual sales and profits, seems to be as accurate and fair an account as is possible on the information available to us. We recommend that the Court then assess prejudgment interest on defendants' net profits (as distributed in Appendix A) from the first day of each period on which it is reasonable to say they were earned.

2. Rate to apply to each period

It is within our discretion to apply the average non-corporate underpayment rate for the entire period from June 1, 1993 through December 14, 2009, rounded to the nearest hundredth of a percent, to the entire disgorgement amount.^[24] Here, this approach would result in an award of simple interest totaling $1,556,122.38.^[25] Indeed, it would likely be well within our discretion to pick the prevailing underpayment rate at any one point during this period and apply it to the entire period. See, e.g., Trs. of Local 813 I.B.T. Ins. Trust Fund v. Bay Area Portables, Inc., 2006 WL 1313819, at *4 (E.D.N.Y. May 9, 2006) (noting that the governing interest rate had fluctuated between 3 and 6 percent during the relevant period, and applying a rate of 5 percent). However, we can be, and believe that in this case we should be, more precise, and should account for shifting rates over time. We thus recommend applying the average non-corporate underpayment rate for each period to the net profits earned during that period, as laid out in Appendix A. Greater precision in our choice of interest rate comports with the principle that the prejudgment interest should ultimately reflect economic realities, and, in this case, works slightly to plaintiffs' advantage, which is more consistent with the balance of equities in this case.^[26]

C. Compounding of Interest

Interest assessed on underpayment of federal taxes is compounded daily. 26 U.S.C. § 6622(a) ("In computing the amount of any interest required to be paid under this title ... such interest and such amount shall be compounded daily."); see also Rev. Rul. 2009-37 at 1 (stating that interest under section 66212 is subject to daily compounding). Post-judgment interest on federal judgments is compounded annually. 28 U.S.C. § 1961(b) ("Interest... shall be compounded annually."). However, as with all aspects of calculation of prejudgment interest, we have discretion as to whether and how often to compound.^[27]

Daily compounding, as prescribed by 26 U.S.C. § 6621, would result in an award of prejudgment interest that dwarfs the principal that defendants were ordered to disgorge. Indeed, even compounding interest annually would more than quadruple the amount of prejudgment interest that defendants would otherwise owe.^[28]

Plaintiffs argue for compounding of interest only as an alternative to our application of the New York statutory rate; they seem not to appreciate the financial impact that compounding would have in this case. Indeed, interest assessed at the large corporate underpayment rate embodied in section 6621(c) and compounded annually according to our methodology would total $5,793,996.20, while simple interest calculated by the same methodology at 9 percent per annum, which plaintiffs say they would prefer, would total $1,964,249.85.

Although we rejected defendants' arguments that the disgorgement amount should be discounted because plaintiffs "slept" on their rights under the 1987 contempt order,^[29] it is clear here that compounding interest for well over a decade during which plaintiffs failed to bring suit would be inconsistent with the balance of equities in this case. We therefore recommend that this court decline to order compounding of prejudgment interest.

CONCLUSION

We recommend that the Court assess prejudgment interest on defendants' net profits at the underpayment rates embodied in 26 U.S.C. § 6621(a)(2), as we recommended in our previous report and recommendation, and as urged by defendants. We further recommend that the Court calculate interest based on the net profits and time periods laid out in Appendices A and B, using the average underpayment rate for each period and assessing simple interest for each period from the first day of the period through December 14, 2009. By our calculations, this yields a total of $1,556,953.55. See Appendix B.

Pursuant to Rule 72 of the Federal Rules of Civil Procedure, the parties shall have ten (10) days from this date to file written objections to this Report and Recommendation. Such objections shall be filed with the Clerk of the Court and served on all adversaries, with extra copies to be delivered to the chambers of the Honorable Leonard B. Sand, Room 1650, and to the chambers of the undersigned, Room 1670, 500 Pearl Street, New York, New York 10007. Failure to file timely objections may constitute a waiver of those objections both in the District Court and on later appeal to the United States Court of Appeals. Fed.R.Civ.P. 72, 6(a), 6(d); 28 U.S.C. § 636(b)(1); see also Thomas v. Arn, 474 U.S. 140, 150, 106 S. Ct. 466, 88 L. Ed. 2d 435 (1985); Small v. Secretary of Health and Human Services, 892 F.2d 15, 16 (2d Cir.1989).

DATED: New York, New York.

December 1, 2009.

                                       APPENDIX A
----------------------------------------------------------------------------------
                   Days from beginning of      Gross sales for    Net profits for
Time Period        period until 12/14/09         the period         the period
----------------------------------------------------------------------------------
6/1/93-5/31/94            6040                 $  129,181.84      $  116,906.58
----------------------------------------------------------------------------------
6/1/94-5/31/95            5675                 $  129,181.84      $  116,906.58
----------------------------------------------------------------------------------
6/1/95-5/31/96            5310                 $  129,181.83      $  116,906.58
----------------------------------------------------------------------------------
6/1/96-5/31/97            4944                 $  129,181.83      $  116,906.58
----------------------------------------------------------------------------------
6/1/96-5/31/98            4579                 $  129,181.83      $  116,906.58
----------------------------------------------------------------------------------
6/1/98-5/31/99            4214                 $  129,181.83      $  116,906.58
----------------------------------------------------------------------------------
6/1/99-5/31/00            3849                 $  229,929.00      $  208,080.44
----------------------------------------------------------------------------------
6/1/00-5/31/01            3483                 $   52,348.00      $   47,373.73
----------------------------------------------------------------------------------
6/1/01-5/31/02            3118                 $   92,433.00      $   83,649.73
----------------------------------------------------------------------------------
6/1/02-5/31/03            2753                 $  326,617.00      $  295,580.85
----------------------------------------------------------------------------------
6/1/03-5/31/04            2388                 $  322,681.00      $  292,018.86
----------------------------------------------------------------------------------
6/1/04-5/31/05            2022                 $  507,499.00      $  459,274.89
----------------------------------------------------------------------------------
6/1/05-5/31/06            1657                 $  407,611.00      $  368,878.55
----------------------------------------------------------------------------------
6/1/06-5/31/07            1292                 $   75,619.00      $   68,433.45
----------------------------------------------------------------------------------
6/1/07-3/1/08              927                 $    4,462.00      $    4,038.01
----------------------------------------------------------------------------------
   TOTAL:                                      $2,794,290.00      $2,528,768.00
----------------------------------------------------------------------------------

                                           APPENDIX B
--------------------------------------------------------------------------------------
                   Days from                         Average
                  beginning of                      6621(a)(2)         Simple interest
                  period until      Net profits     rate (per           under average
Time Period         12/14/09       for the period    annum)            5621(a)(2) rate
--------------------------------------------------------------------------------------
6/1/93-5/31/94        6040          $  116,906.58      0.07              $  135,419.46
--------------------------------------------------------------------------------------
6/1/94-5/31/95        5675          $  116,906.58      0.086             $  156,318.51
--------------------------------------------------------------------------------------
6/1/95-5/31/96        5310          $  116,906.58      0.09              $  153,067.55
--------------------------------------------------------------------------------------
6/1/96-5/31/97        4944          $  116,906.58      0.088             $  139,350.08
--------------------------------------------------------------------------------------
6/1/96-5/31/98        4579          $  116,906.58      0.088             $  129,062.30
--------------------------------------------------------------------------------------
6/1/98-5/31/99        4214          $  116,906.58      0.078             $  105,277.42
--------------------------------------------------------------------------------------
6/1/99-5/31/00        3849          $  208,080.44      0.082             $  179,928.58
--------------------------------------------------------------------------------------
6/1/00-5/31/01        3483          $   47,373.73      0.088             $   39,781.48
--------------------------------------------------------------------------------------
6/1/01-5/31/02        3118          $   83,649.73      0.068             $   48,591.10
--------------------------------------------------------------------------------------
6/1/02-5/31/03        2753          $  295,580.85      0.056             $  124,846.87
--------------------------------------------------------------------------------------
6/1/03-5/31/04        2388          $  292,018.86      0.046             $   87,884.08
--------------------------------------------------------------------------------------
6/1/04-5/31/05        2022          $  459,274.89      0.05              $  127,212.85
--------------------------------------------------------------------------------------
6/1/05-5/31/06        1657          $  368,878.55      0.066             $  110,524.10
--------------------------------------------------------------------------------------
6/1/06-5/31/07        1292          $   68,433.45      0.078             $   18,894.38
--------------------------------------------------------------------------------------
6/1/07-3/1/08          927          $    4,038.01      0.0775            $      794.80
--------------------------------------------------------------------------------------
     TOTAL:                         $2,528,768.00                        $1,556,953.55
--------------------------------------------------------------------------------------

NOTES

[1] The Court's 2007 Opinion provides a full discussion of the lengthy history of this case. See Fendi, 2007 WL 2982295, at *1-3 (S.D.N.Y. Oct. 10, 2007).

[2] Burlington withdrew its cross-motion for summary judgment at oral argument. (T. at 27.)

[3] Burlington argues for the first time in its Surreply that Minerva's testimony is inadmissible because Fendi failed to produce a written expert report as required by Federal Rule of Civil Procedure 26. Rule 26(a)(2)(B) requires written reports "if the witness is one retained or specially employed to provide expert testimony in the case or one whose duties as the party's employee regularly involve giving expert testimony." Fed.R.Civ.P. 26(a)(2)(B). The "structure of Rule 26(a)(2)(B) provides a clear distinction between the retained class of experts and the unretained class of experts." Bank of China, New York Branch v. NBM LLC, 359 F.3d 171, 182 n. 13 (2d Cir.2004) (quoting Kent v. Katz, No. 99 Civ. 189, 2000 WL 33711516, at *1 (D.Vt. Aug. 9, 2000)). Burlington does not acknowledge the limited applicability of Rule 26(a)(2)(B), and even if Minerva qualifies as an employee who regularly gives expert testimony, the testimony of an expert need not be excluded if the error is harmless. Fed. R.Civ.P. 37(c); see Patterson v. Balsamico, 440 F.3d 104, 117 (2d Cir.2006). Prior to Minerva's testimony, Fendi produced the authentication reports that Minerva relied on in his testimony. These reports provided Burlington with the opinions he would express, the data or other information he considered, the exhibits that he would use, and Minerva's title of employment at Fendi. See Fed. R.Civ.P. 26(a)(2)(B)(i)-(iv). Any failure to produce a formal written expert report was harmless error.

[4] Burlington's allegation that Minerva only reviewed fifty percent of Lepri's authentication reports misstates Minerva's testimony. (Def.'s Opp. Summ. J. 7.) As one part of the authenticity inspection, the distance between the two F's on the fabric is measured to see if it is in compliance with Fendi's standards. Lepri would conduct this measurement and include it in his report; Minerva, upon reviewing the report, would check this measurement at least fifty percent of the time, sometimes more, depending on how many reports he had to review. (Minerva Dep. at 332:18-333:13.) This specific measurement is the only aspect of the inspection that Minerva testified he sometimes checked only fifty percent of the time.

[5] See also S. Broun, McCormick on Evidence § 9 (6th ed.2009) ("[E]ven if a witness recognizes from present memory the correctness of a set of facts recorded in a memorandum, the witness may be unable to detail those facts from memory without consulting the writing. Accordingly, the generalization that, once refreshed, a witness must speak independently of the writing is too inflexible. Again, the matter ought to be entrusted to the trial judge's discretion. The judge may permit the witness to consult the memorandum as she speaks, especially when it is so lengthy and detailed that even a witness with a fresh memory would realistically be unable to recite all the items unaided.").

[6] Any prejudice to Burlington, as a result of Minerva's reliance on the reports, was mitigated by Burlington's ability to cross-examine Minerva and attack his credibility. See 20th Century Wear, 747 F.2d at 93 n. 17 (noting that the courts and Congress have devised two safeguards to protect against the admission of inadmissible evidence for its truth: (1) the broad discretion of the trial judge; and (2) the right to inspect whatever is used to refresh the witness's recollection, to cross-examine the witness, and to introduce into evidence relevant portions); Rinke, 778 F.2d at 588 ("[The witness's independent recollection], coupled with the opportunity defense counsel had to cross-examine Pieper and/or to introduce the notes into evidence and attack Pieper's credibility vitiates the prejudice, if any, to the defendants when the court allowed Pieper to rely on his notes while testifying.").

[7] The testimony Burlington puts forth does not support its allegations. (Def.'s Opp. Summ. J. 25-26.) At Scott Ressler's deposition, he did not testify that he purchased goods directly from Fendi manufacturers. He testified that he purchased St. John's merchandise directly from the manufacturers but did not remember ever purchasing Fendi merchandise directly from the manufacturer. (Castellano Dec. Ex. 1 ("Ressler Dep.") at 433-34). Colton testified that he purchased goods on two occasions from a Fendi outlet, and on one of those occasions all he purchased was a cashmere Fendi blanket for his home. (Castellano Dec. Ex. 2 ("Colton Dep.") at 55-57.)

[8] The bolts of fabric in the warehouse, which Burlington claims challenges the credibility of Fendi's procedures, does not raise a genuine issue as to the authenticity of the goods. In Gucci v. Duty Free Apparel, 286 F. Supp. 2d 284 (S.D.N.Y.2003), the court rejected a similar argument. Even though the court recognized "that Gucci's quality control measures are likely not foolproof and that authentically-manufactured Gucci items containing deviations may wind up in the retail market," the court concluded that "mere speculation that this unlikely possibility came about in this particular case" was insufficient to defeat summary judgment. Duty Free Apparel, 286 F.Supp.2d at 289. The bolt of fabric provides nothing more than speculation and is insufficient to defeat summary judgment.

[9] Burlington raises an affirmative defense of laches. A party that seeks equity's assistance must stand before the Court with clean hands. Hermes Int'l v. Lederer de Paris Fifth Ave., Inc., 219 F.3d 104, 107 (2d Cir.2000). The defense of laches is not available where the defendant has been found to have intended the infringement. Id. As the Court subsequently concludes that Burlington willfully infringed, infra Part II.C, the laches defense is unavailable.

[10] Prior to the Trademark Amendments Act of 1999 (the "1999 Amendment"), the Court of Appeals for the Second Circuit required a showing of bad faith or intentional misconduct in order to recover profits for infringement and counterfeiting claims. See Int'l Star Class Yacht Racing Assoc. v. Tommy Hilfiger, U.S.A., Inc., 80 F.3d 749, 753 (2d Cir. 1996). However, the 1999 Amendment specifically added "willful" in reference to a violation under Section 1125(a) but not in reference to a violation under Section 1125(c). The courts in this Circuit are currently split as to whether this amendment eliminated the requirement of willfulness or "bad faith" for violations of 1125(c). See Mr. Water Heater Enterprises v. 1-800-Hot Water Heater, 648 F. Supp. 2d 576, 589-90 (S.D.N.Y.2009) (collecting cases); Malletier v. Dooney & Bourke, Inc., 500 F. Supp. 2d 276, 280 (S.D.N.Y.2007) (finding the willfulness requirement unaffected because the 1999 Amendment "did not alter or even address the relevant subsections of the federal trademark infringement statute"); Nike, Inc. v. Top Brand Co. Ltd., No. 00 Civ. 8179(KMW), 2005 WL 1654859, at *10 (S.D.N.Y. July 13, 2005) ("Without reason to believe that Congress intended to preserve the Second Circuit rule that preceded the 1999 amendment, the plain language of the statute should govern."). For the reasons set forth below, we find that Burlington's infringement was willful, and, therefore, need not reach the question of whether the "bad faith" requirement survives the 1999 Amendment.

[11] Where a plaintiff establishes a case of trademark infringement or false designation, the plaintiff is entitled to recover attorney's fees in "exceptional cases." 15 U.S.C. § 1117(a). The district courts are given considerable discretion in awarding reasonable attorney's fees. Quaker State Oil Refining Corp. v. Kooltone, Inc., 649 F.2d 94, 95 (2d Cir. 1981). Willful infringement generally renders a case "exceptional." Id.; Motorola, 2009 WL 962809, at *10. Attorney's fees are appropriate in this case based on Burlington's willful infringement.

[12] Burlington is also ordered to deliver the counterfeit goods to Fendi for their destruction. Johnson & Johnson Consumer Co. Inc. v. Aini, 540 F. Supp. 2d 374, 396 (E.D.N.Y. 2008); Fendi S.A.S. Di Paola Fendi E Sorelle v. Cosmetic World, Ltd., 642 F. Supp. 1143, 1146-47 (S.D.N.Y.1986). Those goods that are still relevant to pending litigation will not be destroyed until the end of that litigation.

[13] Burlington argues that the perfume sold during this period was not a violation of the injunction because it was sold pursuant to a "lease agreement" arrangement with Scents of Worth; therefore, though the sales were made through the Burlington retail infrastructure, Burlington never owned or gained title to the merchandise at any time. (Def.'s Response to 56.1 ¶ 21.) The Court rejected this very argument in our 2007 Opinion, noting that "Fendi-branded perfume was stocked and handled by Burlington employees, the perfume was sold at the same registers as other Burlington products, and Burlington received 26 percent of the net sales obtained from those items." Fendi, 2007 WL 2982295, at *5.

[14] In arguing that the injunctive relief sought by Fendi is draconian, Burlington argues that "Fendi's proposed injunction would require[] BCF to obtain written permission before dealing in any Fendi-branded merchandise (not merely counterfeit merchandise)...." (Def.'s Opp. Summ. J. 21.) Burlington's continued failure to grasp the requirements of the 1987 Injunction even after this Court found it in contempt is troubling.

[15] For the purpose of the discussion of this motion, the Vendors include Euro Moda, Moda Oggi and Summit. Ashley Reed does not dispute its liability to the extent that the goods it supplied to Burlington are determined to be counterfeit. For the reasons set forth in Part IV.b, Colton's motion to dismiss the third-party complaint is granted and, therefore, Colton is no longer a party in this action.

[16] Euro Moda and Moda Oggi also allege that a settlement was entered into in June 2008, where Euro Moda and Moda Oggi agreed to give up their independent claims against Burlington, if Burlington agreed to dismiss its claims against Euro Moda and Moda Oggi. Euro Moda and Moda Oggi have put forth no evidence of a settlement; this contention is without merit.

[17] We do not accept Colton's argument that it brought the forum-selection clause objection at the first opportunity. Colton cannot justify the three year delay based on Burlington's failure to specify the basis of the indemnification claim, and at the same time argue that any claim for indemnification based on trademark is governed by the forum-selection clause in the Purchase Order. The Third-Party Complaint, while it does not explicitly reference the Purchase Order, asserts that in selling counterfeit goods the Vendors "breached one or more contracts of sale pursuant to which such counterfeit goods were purchased by retailers." (Third Party Compl. ¶ 16.) The clear language of the Purchase Order states that all disputes arising out of the performance of the contract are venued in New Jersey. However, a waiver is not inferred simply because a party does not invoke a forum-selection clause at the first opportunity. Thus, the question remains whether Colton's three year delay is sufficient to infer a waiver.

[18] Although spreading shipping costs evenly over this time period may mean some shipping expenses are allocated to perfume sales, a development favoring Burlington, Fendi has not objected to the R & R. (Pl.'s. Resp. to Def.'s. Objections 1.)

[1] For the purposes of this motion, we assume familiarity with the facts.

[2] Burlington's records have been a persistent obstacle in this litigation. At the time of our 2007 opinion, we believed Burlington's first sale of Fendi goods was in 2002. Subsequent discovery revealed that Burlington began selling Fendi-branded goods no later than 1993. Burlington has made no records available for the six years immediately after the injunction was entered. Fendi, 689 F.Supp.2d at 620-21, 2010 WL 431509, at *24. As Magistrate Judge Dolinger noted in his Report and Recommendation, "while plaintiffs know a sale was made as early as 1993, it is possible that sales could have occurred even earlier, but impossible to confirm or refute that hypothesis." Id.

[1] For a fuller overview of the facts of this case, see Mem. & Order, Oct. 10, 2007, Docket Doc. 147, available on Westlaw at Fendi Adele S.R.L. et al. v. Burlington Coat Factory Warehouse Corp. et al., 2007 WL 2982295 (S.D.N.Y. Oct. 10, 2007).

[2] See, e.g., Pls.' Mem. Law Supp. Additional Contempt Remedies, Feb. 22, 2008, Doc. 176; Defs.' Mem. Law re: Proposed Contempt Judgment, Mar. 5, 2008, Doc. 180; Pls.' Mem. Law Supp. Proposed Judgment for Contempt, Mar. 5, 2008, Doc. 184; Defs.' Mem. Law Opp. Pls.' Mot. for Additional Contempt Remedies, Mar. 14, 2008, Doc. 188; Pls.' Reply Mem. Law Supp. Additional Contempt Remedies, Mar. 21, 2008, Doc. 193; Defs.' Suppl. Mem. Law re: Profits, Apr. 1, 2008, Doc. 195. See also R & R at 3-5 (describing submissions by both sides).

[3] Judge Sand articulated the precise question for our analysis first as "whether the appropriate interest rate should be based on the federal statutory benchmark for non-payment of federal taxes, 26 U.S.C. § 6621(a)(2), or for analogous causes of action under state law" and subsequently, in broader language, as "whether the calculation of pre-judgment interest should be based on federal or state statutory benchmarks in the instant action." Compare Mem. & Order, Doc. 293 at 8 with Order, Doc. 294.

[4] E.g., U.S. Philips Corp. v. Iwasaki Elec. Co., 607 F. Supp. 2d 470 (S.D.N.Y.2009) (setting prejudgment interest in patent case at prime rate, calculated and compounded quarterly, and rejecting 1-year treasury-bill rate embodied in 28 U.S.C. § 1961); S.E.C. v. Aimsi Tech., Inc., 650 F. Supp. 2d 296 (S.D.N.Y.2009) (setting prejudgment interest rate in securities case at rate specified in 26 U.S.C. § 6621(a)(2)); Nat'l Ass'n for the Specialty Food Trade, Inc. v. Construct Data Verlag A.G., 2007 WL 656274, at *2 (S.D.N.Y. Feb. 23, 2007) (setting prejudgment interest in case involving willful infringement of Lanham Act at New York statutory rate of 9 percent); Sec. Ins. Co. of Hartford v. Old Dominion Freight Line, Inc., 314 F. Supp. 2d 201 (S.D.N.Y.2003) (setting prejudgment interest in suit for actual damages under Interstate Commerce Act at 52-week maturity treasury-bill rate embodied in 28 U.S.C. § 1961); GTFM, Inc. et al. v. Solid Clothing Inc., 2002 WL 31886349, at *4 (S.D.N.Y. Dec. 26, 2002) (setting prejudgment interest in case involving willful infringement of Lanham Act at New York statutory rate of 9 percent); AccuScan, Inc. v. Xerox Corp., 2000 WL 280005, at *2 (S.D.N.Y. Mar. 15, 2000) (setting prejudgment interest in patent infringement case at 52-week-maturity treasury-bill rate, compounded annually, and rejecting use of prime rate).

[5] Judge Sand's framing of the question presented for our analysis ("whether the calculation of pre-judgment interest should be based on federal or state statutory benchmarks") reflects this practice. (See Order, Doc. 294).

[6] The prejudgment interest period is based on the period during which defendants in this case enjoyed the use of profits from the unlawful sale of Fendi goods, which runs approximately from June 1, 1993—the beginning of the fiscal year during which defendants made the first known sales of Fendi goods in violation of the consent decree— through the date on which this Court enters a judgment against defendants in an amount that includes the disgorgement principal and prejudgment interest. This Court's August 10, 2009 order approved a disgorgement principal of $2,528,768, see Mem. & Order, Aug. 10, 2009, Doc. 293, but the Court has not yet ordered the entry of judgment against defendants. We assume it will do so following its decision as to the appropriate amount of prejudgment interest. For purposes of simplifying the calculation of prejudgment interest, we assume that the Clerk of the Court will enter judgment approximately two weeks after this report and recommendation, and we accordingly base our calculation of prejudgment interest on a judgment date of December 14, 2009.

[7] The rate was 4 percent during the fourth quarter 2003, first quarter 2004, third quarter 2004, and third and fourth quarters of 2009. Rev. Rul.2009-37, Nov. 23, 2009, available at 2009 WL 4023061.

[8] The rate was 10 percent during the second quarter 1995. Id.

[9] The relevant Treasury yield rate has been dropping steadily for years. See Federal Reserve Statistical Release Historical Data, http://www.federalreserve.gov/Releases/H15 (follow "Historical Data" link and then click "Weekly (Friday)" on the table under "Treasury bills constant maturity—1-year"). For the week ending November 20, 2009 (the most recent period for which weekly average data is available at the time of this writing), the rate was 0.29 percent, the lowest it has been during the prejudgment interest period. Id. (weekly average rate for week ending Nov. 20, 2009).

[10] Callan Periodic Table of Investment Returns, http://www.callan.com/research/institute/download/?file=periodic/free/311.pdf (last visited Nov. 24, 2009) (S & P 500 returns for 1993 through 2008, inclusive). Figures for 2009 are not yet available. The S & P 500 annual return rate for 2008 (negative 37 percent) demonstrates well the risk inherent in such investments.

[11] E.g., N.Y. Marine & Gen. Ins. Co., 266 F.3d at 131 (affirming district court's application of treasury-bill rate provided in 28 U.S.C. § 1961 rather than 17 percent interest rate reflecting what plaintiff actually could have earned on investment); Sec. Ins. Co. of Hartford, 314 F.Supp.2d at 205 (stating that 52week-maturity treasury bill rate prescribed by 28 U.S.C. § 1961 reflects a "conservative yet valid investment option"); AccuScan, Inc., 2000 WL 280005, at *2 (applying 52-week-maturity treasury-bill rate and observing that this rate, embodied in 28 U.S.C. § 1961, is frequently applied in awarding prejudgment interest in patent cases); Barcia v. Sitkin, 1997 WL 66785, at *4 (S.D.N.Y. Feb. 19, 1997) (awarding prejudgment interest at treasury bill rate embodied in 28 U.S.C. § 1961 in civil contempt action).

[12] Prior to 2001, however, the shortest maturity was three months. See, e.g., Federal Reserve Statistical Release, www.federalreserve. gov/RELEASES/H15/.

[13] Somewhat analogously, the Court has suggested that the cost of borrowing funds may serve as a measure of prejudgment interest when the defendant has injuriously withheld from the plaintiff money to which he was entitled. See, e.g., Jones, 223 F.3d at 139-40; Saulpaugh v. Monroe Cmty. Hosp., 4 F.3d 134, 145 (2d Cir.1993).

[14] Cf. Acticon Tech. v. Heisei Electronics Co., 2008 WL 356872, at *4 (S.D.N.Y. Feb. 5, 2008) (holding no abuse of discretion in magistrate judge's use of § 1961 rate, compounded annually, absent evidence that this rate would not fully compensate plaintiff) (emphasis added).

[15] R & R at 6 (citing Donohue Decl., Doc. 182, ¶ 14 & n. 9); see also Donohue Decl., Doc. 182, Ex. 2 at 19 (reflecting Fendi products received as early as September 23, 1993).

[16] We also note that plaintiffs took issue with our recommendation that defendants be allowed to deduct shipping expenses and income taxes in spite of their thin evidentiary proffer on these issues, although the District Court rejected that argument and plaintiffs do not raise it here. (See Pls.' Partial Objection to Rep. & Recommendation, May 11, 2009, Doc. 273, at 3-8).

[17] However, we also found that the sales after the entry of judgment of contempt "were scattered and not part of any evident pattern of active resistance by defendants to compliance with their obligations under the Injunction." (R & Rat 50-51).

[18] See, e.g., R & R at 36-37 (stating that defendants "drastically understated the scale of [their] sales of Fendi-branded items," "concealed at least some of [their] sales activities in merchandising Fendi perfume," and "engaged in prolonged resistance to being called to account for [their] violation of the injunction").

[19] But see Del Turco, 2007 WL 2581882, at *4. In Del Turco, Magistrate Judge Azrack found that defendant owed unpaid benefits from 1999 through 2003 totaling $42,845.51. Id. at *3. (The report and recommendation did not specify how much of this total amount was owed for each of the years 1999 through 2003.) She then assessed interest on the entire amount of unpaid benefits for each month from January 1999 (the earliest date on which any unpaid benefits were due) through February 2007 (the month in which judgment was rendered). Id. at *4.

[20] In our previous report and recommendation, we approved of this methodology (albeit applied to net profits rather than gross sales). (R & R at 48). However, we also rejected plaintiff's choice of interest rate, and plaintiff has subsequently drawn our attention to the injustice that would result from this method coupled with a less generous interest rate. (E.g., Pls.' Mem. at 2; Pls.' Reply Mem. at 2).

[21] We note that defendant has not objected to these figures.

[22] For example, plaintiffs determined that $775,091.00 of the total $2,794,290.00 in gross sales was made between June 1, 1993 and May 31, 1999. (See Donohue Decl., Doc. 182, Ex. 13). $775,091.00 represents approximately 27.74 percent of the total gross sales; net profits for this period should represent the same percentage of the total net profits ($2,528,768). This approach yields net profits for this period of $701,439.48. (775,091/2,794,290 x 2,528,768 = 701,439.48).

[23] For example, plaintiffs determined that defendants made $229,929 in gross sales between June 1, 1999 and May 31, 2000. (Donohue Decl., Doc. 182, Ex. 13). They then calculated prejudgment interest on that amount for 2,831 days (from May 31, 2000 until March 1, 2008). We believe it is more appropriate here to calculate prejudgment interest on the net profits corresponding to that amount for 3,196 days (from June 1, 1999 until March 1, 2008).

[24] E.g., Del Turco, 2007 WL 2581882, at *4 (noting that court has discretion as to how to calculate interest under 26 U.S.C. § 6621 and applying the average prescribed rate over the relevant five-year period to the nearest tenth of a percent); Hollie v. Korean Air Lines Co., 834 F. Supp. 65, 70 (S.D.N.Y.1993) (applying the monthly average of 52-week-maturity treasury-bill rates for the relevant period).

[25] The average non-corporate underpayment rate for the disgorgement period, rounded to the nearest hundredth of a percent, is 7.13 percent per annum. See supra, at 15. Using this rate, we calculate simple interest for each period by dividing .0713 by 365 days, multiplied by the number of days between the first day of that period and December 14, 2009, multiplied by the net profits for that period. See Appendix A for a table of net profits for each period and the number of days of interest to be assessed on each period's profits.

[26] This approach was taken in Finkel v. Tech Man, Inc., 2007 WL 433399, at *3 (E.D.N.Y. Feb. 6, 2007). In that ERISA case, the court found that defendant owed $26,325.75 for unpaid benefits for the period from December 29, 2005 through July 26, 2006, and that defendant only owed $1,335.60 in unpaid benefits for the period from August 10 through August 30, 2006 because it had paid some of the required contributions for those weeks. The court then applied the interest rate for the first and second quarters of 2006 to the portion of the principal due for the period from December 29, 2005 through July 26, 2006, and the interest rate for the third quarter of 2006 to the portion of the principal accrued during that period. Id. at *3. The underpayment rate was 7 percent for the first and second quarters of 2006 and 8 percent for the third quarter of 2006. Rev. Rul. 2009-37, 2009 WL 4023061 (January 1, 2006—September 30, 2006). We note that this method of calculation was somewhat imprecise; the first period during which unpaid benefits accrued actually extended from the fourth quarter 2005 into the third quarter of 2006, but the court applied the average interest rate for the first and second quarters of 2006, presumably for ease of calculation.

[27] See, e.g., U.S. Philips Corp., 607 F.Supp.2d at 483 (setting prejudgment interest in patent case at prime rate, calculated and compounded quarterly); AccuScan, Inc., 2000 WL 280005, at *2 (applying 52-week-maturity treasury-bill rate and compounding interest annually, rather than quarterly as plaintiff urged); Hollie, 834 F. Supp. at 70-71 (applying average of monthly rates for 52-weekmaturity treasury bills and compounding interest annually).

[28] As laid out in Appendix B, simple interest at the average quarterly non-corporate underpayment rate for each period calculated from the first day of each period to December 14, 2009 yields a total of $1,556,953.55. Annually compounded interest is calculated by multiplying net profits for that year by (1 + the annual interest rate for that period) to the power of (the number of whole years between the first day of the period and December 14, 2009), plus simple interest assessed on the remainder of days. Thus, interest calculated at the average § 6621(a)(2) rate for each period and compounded annually and assessed from the first day of each period through December 14, 2009 would total $4,836,494.08. Daily compounded interest is calculated by multiplying net profits for that year by (1 + the annual interest rate for that period divided by 365) to the power of (the number of days between the first day of the period and December 14, 2009). Daily compounded interest at the average § 6621(a)(2) rate for each period would total $5,072,885.24.

[29] See R & R at 34-36 (recommending denial of defendants' request to exclude profits from the period before plaintiffs invoked the consent decree); Mem. & Order, Aug. 10, 2009, Doc. 293, at 2 (adopting that recommendation).