BDT Products, Inc. v. Lexmark International, Inc.

OPINION McKEAGUE, Circuit Judge. In 2005, we affirmed a grant of partial summary judgment for Lexmark International (“Lexmark”) in a suit brought by Lexmark’s one-time partners, BDT Products and Buro-Datentechnik GMBH & Company KG (hereinafter collectively “BDT”), arising from the contention that Lexmark had misappropriated trade secrets in developing a printer tray that substantially resembled a tray developed by BDT. Among the firms appearing for BDT were Higgs, Fletcher & Mack (“Higgs”) and Meisenheimer Herron & Steele (“Meisenheimer”). Following our ruling, the district court granted Lexmark’s motion for attorney fees and imposed sanctions to the extent of those fees (more than five million dollars) on BDT, Higgs, and Meisenheimer under Kentucky Revised Statute § 365.886, 1 28 U.S.C. *744 § 1927, and its inherent powers. Meisenheimer now appeals the imposition of these sanctions, arguing that courts may not impose sanctions under § 1927 on law firms (as opposed to individual attorneys), and that, regardless, BDT’s suit was not necessarily meritless and Lexmark has not demonstrated that Meisenheimer (as opposed to BDT or Higgs) acted in bad faith or with improper purpose. 2 We agree that courts may not impose sanctions on law firms under § 1927; the critical question, then, is whether the court abused its discretion in imposing sanctions under its inherent powers. We conclude that, as BDT’s tray was commercialized and sold before BDT even transmitted some of its information to Lexmark, BDT and its attorneys pursued a suit based at its heart on misappropriation of “trade secrets” that were, in fact, not secret at all. In determining that Meisenheimer acted in bad faith or with improper purpose in pursuing this meritless lawsuit, however, the district court relied in part upon a misstatement of Sixth Circuit law and, without sufficient proof, ascribed the improper purpose sought by BDT and Higgs to Meisenheimer as well. Accordingly, we vacate the district court’s order insofar as it imposes sanctions on Meisenheimer and remand to the district court for entry of an order denying Lexmark’s motion for sanctions as to Meisenheimer and for any other necessary actions consistent with this opinion. I. Beginning in 1990, Lexmark and BDT, which both develop paper handling technology for printers, worked together on various projects. As part of this relationship, between 1990 and 1996 the two companies entered into seven confidentiality agreements, including four “one-way agreements” (“Confidential Disclosure Agreements”) under which only Lexmark was to disclose confidential information to BDT and three “two-way agreements” (“Confidential Exchange Agreements”) under which Lexmark and BDT disclosed confidential information to each other. Each of these agreements expressly provided that Lexmark was free to use (and disclose through sales of its products) any information BDT disclosed to Lexmark. Language of the Confidentiality Agreements Each of the four one-way Confidential Disclosure Agreements, which were signed by BDT respectively in February of 1990, August of 1991, June of 1992, and March of 1995, included the following Paragraph 9: In connection with this agreement, Lexmark does not wish to receive any information which may be considered confidential or proprietary by Contractor. Accordingly, except with respect to the rights of Contractor under valid patents and copyrights, no obligation of any kind is assumed by or is to be implied against Lexmark by virtue of Lexmark’s discussions with Contractor or with respect to any information received (in whatever form) from Contractor and Lexmark will be free to reproduce, use and disclose such information to others without limitation. Moreover, discussions and/or correspondence, or other activities under this agreement shall not in any respect, impair the right of Lexmark to make, procure, or market products or services now or in the future which may be com *745 petitive with those offered by Contractor. (Appellee App’x VL1674, 1676-77, 1680, 1683.) The first of the two-way Confidential Exchange Agreements, which was signed by BDT in May of 1993, stated as part of Paragraph 4: Except with respect to rights under valid patents, the receiving party shall be free to use any such Confidential Information provided by the disclosing party, any reports and written documentation prepared by the receiving party, and any ideas, concepts and/or techniques contained in any such Confidential Information for any purpose including the use of such Information in the development, manufacture, marketing and maintenance of its products and services, subject only to the obligation not to disclose, publish, or disseminate such Confidential Information during such foregoing specific period of confidentiality- (Appellee App’x VI: 1686.) Each of the other two Confidential Exchange Agreements, which were signed by BDT in March of 1995 and March of 1996 respectively, contained the same language in Paragraph 3: Except with respect to rights under valid patents and copyrights, Recipient shall be free to use any such Confidential Information provided by Discloser subject only to the obligation not to disclose, publish or disseminate such Confidential Information during the foregoing specified period of confidentiality. (Appellee App’x VI: 1690.) Section 5(c) of the 1995 and 1996 Confidential Exchange Agreements (and Section 6(c) of the 1993 Confidential Exchange Agreement) stated that any obligations under the agreements to maintain confidentiality would not apply to information that was or that became publically available without breach. Similarly, Section 5(f) of the 1995 and 1996 Confidential Exchange Agreements (and Section 6(f) of the 1993 Confidential Exchange Agreement) provided an exemption from disclosure obligation for information that “is inherently disclosed in the use, lease, sale or other distribution of, or publicly available supporting documentation for, any present or future product or service by or for Recipient____” (Appellee App’x VI: 1691,1687,1695.) Lexmark’s Sale of Technology Resembling BDT’s LF 2000 Printer Tray In the confines of a closed and “confidential” prototype room at the 1993 Hannover Fair, an annual printer industry trade show in Germany, BDT demonstrated to Lexmark both a diverter (output device) and an advanced high-capacity input tray, known for the purposes of this suit as the LF 2000. Following the Fair, BDT (at Lexmark’s request) provided prototypes of the LF 2000, information on the timing sequences and algorithms for the printer tray, and the costs of a license for the technology and of the product itself if manufactured by BDT. Between 1993 and 1995, Lexmark repeatedly asked for, and BDT repeatedly provided, advice and information on the workings of the LF 2000. Ultimately, however, Lexmark informed BDT that it was not interested in licensing the LF 2000. During this period, BDT also entered into discussions with other clients, including Hewlett Packard Guadalajara (“Hewlett Packard”) and Tektronix, regarding paper-tray technology. As part of those discussions, BDT executed one-way Confidential Disclosure Agreements with both clients, under which BDT promised to safeguard the confidential information of its clients, but its clients were not required to safeguard confidential information provided by BDT. Late in 1993, BDT began to develop what ultimately became the auxil *746 iary tray for Hewlett Packard’s “HP4v” printer, which Hewlett Packard released to the public in 1994. As the district court noted in granting Lexmark’s motion for partial summary judgment, “The HP4v tray was the first commercial tray to embody every component of the BDT System that is at issue in this case.” BDT Prods., Inc. v. Lexmark Int’l, Inc., 274 F.Supp.2d 880, 884 (E.D.Ky.2003). Given Lexmark’s decision not to license the LP 2000 technology, BDT representatives were taken aback at the end of 1996, when Lexmark demonstrated its new “Optra S” printer — a printer that, to BDT’s eyes, utilized that technology. BDT’s Preparation for Filing a Lawsuit Having concluded that Lexmark had somehow acted inappropriately by developing the Optra S, BDT contacted its intellectual property attorney, Bernard L. Kleinke, who was then with the Higgs firm. BDT did not, however, immediately file suit, because it did not “know what [it] could file about.” Instead, BDT asked Kleinke to research the issue and determine whether he thought that Lexmark “took [BDT’s] look and feel” or “trade.” At the end of 1997, Kleinke suggested to BDT that there might be a good case against Lexmark, but that BDT should wait before filing suit to see whether Lexmark would also release the diverter Lexmark had seen at the Hannover Fair, on which BDT had a patent. Lexmark never released the diverter, and early in 1998 Higgs agreed to take the LF 2000 suit on a contingency fee basis. In explaining why BDT chose to file its lawsuit, Freidhelm Steinhilber, the head of the management board of Buro-Datentechnik GMBH & Company KG, stated that “the main reason was that we don’t like if somebody takes information from us and uses it freely” and “to protect our IP, number one.... ” (District Court Mem. & Order, June 27, 2008 (DCM) at 8.) Glenn Klein, the president of BDT Products, stated that “the biggest reason” was that “we had offered them for a license. We offered them a license three times, and three times they turned us down on the license.” (DCM at 8-9.) “We wanted to get a license from those guys, yeah,” Klein concluded. (DCM at 9.) BDT was also worried that BDT’s existing customers would believe that the technology they had purchased from BDT actually belonged to another company. In preparation for filing a suit, BDT provided Higgs with correspondence and communications between BDT and Lexmark from 1990-1997. From these materials, and from correspondence between Klein and Kleinke in 1997, Higgs created the “BDT v. Lexmark Chronology of Facts.” As a result of a misstatement or falsehood in a December 1997 letter from Klein to Kleinke, the Chronology of Facts contained the claim — which was, quite simply, false- — that BDT and Lexmark had executed a Confidential Disclosure Agreement in 1993 explicitly covering the LF 2000 technology. Also as part of its pre-filing investigation, Higgs researched whether commercial sale of the HP4v tray released in 1994 revealed BDT’s trade secrets. Higgs, strongly influenced by BDT representatives, ultimately concluded that the trade secrets encompassed much more than the physical embodiment of the tray. “In the minds of [Higgs] attorneys,” the district court observed, “this included BDT’s know-how ‘regarding how each part functioned and worked together, BDT’s reliability and testing data, and BDT’s algorithms, none of which could be revealed through a physical examination of the tray.’ ” (DCM at 19.) The district court concluded that Higgs’ intended purpose as it prepared to file a *747 complaint was quite clear: “[Higgs] intended to force a settlement on behalf of its client, BDT.” (DCM at 21.) A November 1997 Higgs memo stated that Lexmark would settle the case quickly, as otherwise Lexmark would face $5,000,000 in re-tooling expenses. A May 1998 Higgs memo stated that it was the Higgs litigation team’s “collective judgment that the target audience for this Complaint is less the court than the corporate managers and Lexmark’s lawyers, who we hope will see from the Complaint the wisdom of an early, negotiated settlement.” (DCM at 21.) The Verified Complaint (August 13, 1998) On August 13, 1998, Higgs on behalf of BDT filed the initial Verified Complaint in California. The Complaint, which was sworn by Klein, contained the false claim that in 1993 BDT and Lexmark had executed a Confidential Disclosure Agreement covering the LF 2000. (The district court noted that “[t]he Verified Complaint contains a remarkable amount of detail regarding a document which BDT now admits never existed.” (DCM at 17 n. 8.)) Higgs attorney John Morris, the draftsman of the Verified Complaint, later stated that he wished “to convey a ... theme ... concerning the evolving relationship of trust and confidence” and was not “trying to convey the idea that the particular information — that the particular trade secrets that were the essence of this lawsuit were governed by and controlled by the particular [Confidential Disclosure Agreement] mentioned.” (DCM at 20.) The Association of the Meisenheimer firm and the Amending of the Complaint Following the departure in 1999 of a Higgs litigator for other employment, BDT and Higgs invited Matthew Herron of Meisenheimer to join the litigation team. Before agreeing to do so, Herron conducted an extensive review of the litigation materials, met with Klein, and attended Klein’s deposition. The court received notice of the association of Herron as counsel for BDT on August 10,1999; as of November 5, 1999, Kleinke was no longer assigned to the case by Higgs. Upon joining the litigation, Herron sought and received permission to file an amended complaint in which he removed references to any confidentiality agreements covering the LF 2000 technology and focused on five of the eight claims stated in the Verified Complaint. (The district court noted that Herron would have preferred to focus only on the complaint’s third claim, misappropriation of trade secrets, but that the Higgs firm insisted on inclusion of the other four claims.) Developments in the Suit: Lexmark’s Letter and Paul Vickrey as a Consultant On June 6, 2001, at Lexmark’s insistence, Lexmark’s counsel sent Herron a letter in part identifying the existence of the confidentiality agreements that Lexmark believed permitted Lexmark to use BDT’s confidential information. The letter also pointed out that “the final alleged trade secrets were evident from ... sale of devices that plainly exhibit the alleged trade secrets.” (DCM at 23.) “It can only be concluded,” the letter added, “that you and your clients know that the claims are unsupportable and lack probable cause and that plaintiffs are pursuing this litigation for nefarious purposes.” (DCM at 23.) As the district court noted, “[Meisenheimer], then charged with conducting the matter on behalf of BDT, continued with the prosecution of this matter, nonetheless. The rest, as they say, is history.” (DCM at 23.) Late in 2002, as the parties engaged in extensive discovery, Meisenheimer sought assistance from Kleinke, who had in the meantime left Higgs and joined the law firm of Foley & Lardner in San Diego. *748 (Foley & Lardner entered its appearance on behalf of BDT on November 4, 2002.) According to Meisenheimer, Foley & Lardner’s work ultimately directed Meisenheimer to Paul Vickrey, a Chicago attorney with extensive experience in trade secret matters, and Meisenheimer received BDT’s permission to hire Vickrey as a consultant. Vickrey, having reviewed Lexmark’s motion for summary judgment, informed Herron that he believed that BDT had a strong case, and that his firm would be willing to join the litigation on a contingent fee basis if BDT survived summary judgment. (Vickrey Deposition, Appellant App’x at 907-928.) Procedural History Higgs first filed a Verified Complaint in the California state court on August 13, 1998. Following removal to federal court and a protracted venue fight that made its way to the Ninth Circuit, venue was transferred to the Eastern District of Kentucky on February 22, 1999. Following his association with the suit, Herron sought and received permission to amend the complaint, which he did, filing the Amended Complaint on August 4, 1999. Higgs, and not Meisenheimer, was counsel of record at the time the Amended Complaint was filed. The Amended Complaint asserted five causes of action: (1) breach of an implied-in-fact contract; (2) breach of an implied-in-law contract; (3) misappropriation of trade secrets pursuant to California Civil Code §§ 3426, et seq.; (4) unfair competition pursuant to California Business & Professional Code §§ 17200, 1702-OS; and (5) breach of confidence. Lexmark moved to dismiss counts 2, 3, 4, and 5, but the court denied the motion without prejudice. After extensive discovery, Lexmark filed motions for partial summary judgment and summary judgment. On March 11, 2003, the district court granted Lexmark’s motion for partial summary judgment, determining that Kentucky substantive law governed the matter, and dismissed BDT’s claims for breach of implied-in-law contract, unfair competition, and breach of confidence. (DCM at 25.) On July 31, 2003, the court dismissed BDT’s remaining claims for breach of implied-in-fact contract and trade secret misappropriation under KUTSA. The district court concluded that: (1) by BDT’s own admission there was never any agreement under which Lexmark promised not to use the LF 2000 technology, and Lexmark could not have misappropriated BDT’s information because the Confidential Disclosure Agreements and Confidential Exchange Agreements between BDT and Lexark specifically permitted Lexmark to use that information, BDT Prods., 274 F.Supp.2d at 893-98; and (2) the information at issue in the suit was not a legally cognizable trade secret, as, prior to Lexmark’s supposed misappropriation, BDT had disclosed its technology to both Hewlett Packard and Tektronix under one-way Confidential Disclosure Agreements that did not protect BDT’s confidential information, id. at 891. In its decision, the district court did not discuss the importance of the commercial sale in 1994 of Hewlett Packard’s HP4v paper tray — although it did lay out the facts behind the Hewlett Packard release. The district court entered a judgment in favor of Lexmark on September 2, 2003. BDT immediately appealed the court’s grant of summary judgment, and in 2005 we affirmed the district court judgment in toto. See BDT Prods., Inc. v. Lexmark Int’l, Inc., 124 Fed.Appx. 329, 330 (6th Cir.2005). 3 The three judges on the panel did not agree completely as to the grounds on which their ruling should be based. *749 The majority concluded that: (1) through the Confidential Disclosure Agreements and Confidential Exchange Agreements between Lexmark and BDT, BDT had contracted away any rights it had against Lexmark’s disclosure of BDT’s information, id. at 334; (2) through its disclosure of the LF 2000 to Hewlett Packard and Tektronix subject only to one-way Confidential Disclosure Agreements, BDT had extinguished its proprietary interest, such that no trade secrets existed in the case, id. at 333; and (3) through commercial sales of the HP4v paper tray in 1994, BDT’s trade secrets had similarly been freely disclosed, such that they no longer constituted trade secrets, more than two years before Lexmark introduced the Optra S, id. at 334. In contrast, Judge Merritt based his concurrence “on the single ground” that by the time Lexmark introduced its tray in 1997, Hewlett Packard had been selling the same tray on the open commercial market for over two years. Id. at 335 (Merritt, J., concurring). Judge Merritt added: Had the trade secret not already been revealed to the public in the “HP4V” tray long before the Lexmark tray came on the market, I would have some doubts about the proper outcome in this case, and I would be reluctant to dispose of the case on summary judgment.... The fact that here there were written contracts expressly giving Lexmark confidentiality protection, but not BDT, certainly weakens BDT’s claims, as the Court concludes. But a reasonable jury might still find that BDT’s claimed trade secrets were protected by the inherently confidential relationships between the designer, BDT, and its clients, Lexmark, Hewlett Packard and Tektronix. Officials from Hewlett Packard and Tektronix gave depositions in which they testified that officials of those clients considered similar BDT design information confidential in the absence of a written agreement. Id. at 335-36 (Merritt, J., concurring). To summarize: two of the panel judges agreed with the district court’s two grounds for approving Lexmark’s motion for summary judgment, while one disagreed, concluding that a reasonable jury might agree with BDT; all three agreed that Lexmark was entitled to summary judgment because the 1994 release of the HP4v had disclosed BDT’s confidential information. Following our ruling, the case returned to district court on Lexmark’s motion for attorney fees. After receiving memoranda and proposed findings from the parties and holding a three-day evidentiary hearing, the district court found that BDT, Higgs, and Meisenheimer had pursued “a lawsuit that should never have been brought, and in which no attorney should have persisted.” (DCM at 54.) The district court granted Lexmark’s motion and — acting under Kentucky Revised Statute § 365.886, 28 U.S.C. § 1927, and its inherent power — leveled sanctions against BDT, Higgs, and Meisenheimer in the amount of Lexmark’s attorney fees. “Lexmark,” the district court noted, has set forth clear and convincing evidence that BDT’s counsel asserted and pursued frivolous claims on behalf of its client from day one — seeking damage for breach of a contract which never existed and could never be implied as well as the alleged misappropriation of a putative trade secret (at the heart of all of BDT’s claims) which had been publically disclosed well in advance of any prefiling investigation by [Higgs], which had been disclosed to others pursuant to agreements which offered no protection to BDT’s LF 2000 technology, and when the only agreement between the parties *750 to the suit never afforded protection to any of BDT’s technology. (DCM at 54.) After considering what part BDT and its attorneys had respectively played in bringing, pursuing, and prolonging the suit, the court allocated 50% of the responsibility to BDT, 30% to Higgs, and 20% to Meisenheimer. As to Meisenheimer’s individual responsibility, the court stated: Unlike [Higgs], [Meisenheimer’s] Herron made efforts to streamline the suit by paying careful attention to the evidence available and by amending the complaint accordingly. [See, e.g., at Record No. 547 at 576:11-19.] Indeed, Herron would have dismissed all claims, with the exception of the KUTSA claim, but for the insistence of [Higgs] that the other claims which appeared in the amended complaint remain. [Record No. 547 at 578:7-579:15.] Herron crafted an amended complaint which no longer averred that the disclosure of the LF 2000 technology was made specifically “pursuant to” the one-way [Confidential Disclosure Agreements] and removed any reference to any confidentiality agreements specific to the LF 2000— agreements which did not exist. That said, Herron looked specifically at testimony elicited from BDT representatives and documents during the suit, but he remained blind to the importance of the information that was always was [sic ] in BDT’s — and counsel’s — possession prior to and during the course of this lawsuit. His actions may have been laudable in many respects but they were not enough to end this unfortunate lawsuit. (DCM at 61-62.) This appeal followed the actual imposition of sanctions on August 6, 2008. II. Meisenheimer’s first argument is that the district court erred to the extent that it imposed sanctions under 28 U.S.C. § 1927, as § 1927 permits the imposition of sanctions against individual lawyers, but not against law firms. According to § 1927: Any attorney or other person admitted to conduct cases in any court of the United States or any Territory thereof who so multiplies the proceedings in any case unreasonably and vexatiously may be required by the court to satisfy personally the excess costs, expenses, and attorneys’ fees reasonably incurred because of such conduct. 28 U.S.C. § 1927 (2006). Whether this language prohibits the sanctioning of law firms is an issue of law that the court reviews de novo. Claiborne v. Wisdom, 414 F.3d 715, 722 (7th Cir.2005). While we have never directly ruled on this question, after the district court issued its order requiring Meisenheimer to pay sanctions in this case, a Sixth Circuit panel stated in dicta that “ § 1927 does not authorize the imposition of sanctions on a represented party, nor does it authorize the imposition of sanctions on a law firm.” Rentz v. Dynasty Apparel Indus., Inc., 556 F.3d 389, 396 n. 6 (6th Cir.2009) (imposing sanctions against attorneys in their individual capacities solely under Fed.R.Civ.P. 11 rather than under § 1927) (citing Claiborne, 414 F.3d at 722-24). “[0]ne panel of [the Sixth Circuit] is not bound by dicta in a previously published panel opinion.” United States v. Burroughs, 5 F.3d 192, 194 (6th Cir.1993). Nonetheless, “[although dictum is unnecessary to the decision, it may nevertheless be followed if ‘sufficiently persuasive.’ ” PDV Midwest Ref., L.L.C. v. Armada Oil & Gas Co., 305 F.3d 498, 510 (6th Cir. 2002). The Rentz court, while offering no analysis itself, cited to Claiborne, 414 F.3d 715, in which the Seventh Circuit collected cases and issued a well-reasoned explanation of why, under § 1927, judges may not *751 appropriately sanction law firms. As the Claiborne court observed: Our sister circuits have come to differing conclusions without focusing on the precise legal question at stake.... [Tjhese decisions are inconclusive. We therefore consider for ourselves whether a law firm is subject to sanctions under § 1927. The statute itself refers to “[a]ny attorney or other person admitted to conduct cases in any court of the United States.” ... Individual lawyers, not firms, are admitted to practice before both the state courts and the federal courts. [Citations omitted.] The fact that § 1927 refers to “other person[s]” admitted to conduct cases is of no help to the defendants. This language reflects the fact that in limited circumstances non-attorneys may appear in judicial proceedings, such as in patent proceedings or where law students receive special permission to conduct cases before they are admitted to the bar. [Citations omitted.] It is too much of a stretch to say that a law firm could also be characterized as such a person. Our conclusion has the virtue of being consistent with the rationale the Supreme Court used in Pavelic & LeFlore v. Marvel Entertainment Group, 493 U.S. 120, 110 S.Ct. 456, 107 L.Ed.2d 438 (1989), when it considered the question whether sanctions were possible against a law firm under an earlier version of Fed.R.Civ.P. 11. (The rule was amended as of December 1, 1993, to ensure that law firms could be subject to sanctions under its authority.) In Pavelic & LeFlore, however, the Court had to construe language that permitted sanctions only against “the person who signed” the offending document. 493 U.S. at 121, 110 S.Ct. 456, 107 L.Ed.2d 438. The district court, affirmed by the court of appeals, had found that this language permitted it to impose sanctions not only against the lawyer who signed the papers, but also against his law firm. The Supreme Court reversed, finding that in context the phrase “the person who signed” could only mean the individual signer, not his partnership, either in addition to him or in the alternative. The language of § 1927 raises exactly the same problem as the earlier version of Rule 11. Even if Pavelic & LeFlore does not strictly dictate the outcome here, it points strongly in the direction we have taken. Claiborne, 414 F.3d at 722-23. We find the analysis and reasoning of the Claiborne court persuasive. Even if firms can admittedly be personified in a literary sense through briefs, there is no reason to consider a law firm a “person” under the statute. More importantly, law firms are not “admitted” to “conduct cases” in court. We therefore confirm what the Rentz court stated in dicta, that 28 U.S.C. § 1927 does not authorize the imposition of sanctions on law firms. We thus conclude that the district court erred insofar as it imposed sanctions on Meisenheimer under § 1927. III. In addition to sanctioning Meisenheimer under 28 U.S.C. § 1927, the district court imposed sanctions to the extent of attorney fees under its inherent power to sanction for conduct which abuses the judicial process. See, e.g., Chambers v. NASCO, Inc., 501 U.S. 32, 44-45, 111 S.Ct. 2123, 115 L.Ed.2d 27 (1991). We review a court’s imposition of sanctions under its inherent powers for abuse of discretion. Id. at 55, 111 S.Ct. 2123. Before considering whether the district court abused its discretion in imposing such sanctions, we take the opportunity to clarify our case law on when a district court may assess attorney fees under its inherent powers, and we affirm the validity of the three-part test we *752 laid out in Big Yank Corp. v. Liberty Mut. Fire Ins. Co., 125 F.3d 308, 313 (6th Cir.1997). Generally, fee shifting is prohibited under the “American Rule,” which is “deeply rooted in our history and in congressional policy.” See Alyeska Pipeline Serv. Co. v. Wilderness Soc’y, 421 U.S. 240, 247, 271, 95 S.Ct. 1612, 44 L.Ed.2d 141 (1975). Congress, however, “while fully recognizing and accepting the general rule,” has made “specific and explicit provisions for the allowance of attorneys’ fees under selected statutes granting or protecting various federal rights.” Id. at 260-61, 95 S.Ct. 1612. In addition to awarding fees under one of these statutory provisions, moreover, a court may assess attorney fees against a party under the court’s inherent powers “when a party has ‘acted in bad faith, vexatiously, wantonly, or for oppressive reasons.’ ” Chambers, 501 U.S. at 45, 111 S.Ct. 2123 (quoting Alyeska Pipeline Serv. Co., 421 U.S. at 258-59, 95 S.Ct. 1612). This “bad faith exception” to the American Rule — the imposition of sanctions under a court’s inherent powers — thus requires a finding of bad faith or of conduct “tantamount to bad faith.” Youn v. Track, Inc., 324 F.3d 409, 420 (6th Cir.2003) (quoting First Bank of Marietta v. Hartford Underwriters Ins. Co., 307 F.3d 501, 517 (6th Cir.2002); Roadway Exp., Inc. v. Piper, 447 U.S. 752, 767, 100 S.Ct. 2455, 65 L.Ed.2d 488 (1980)). In the Sixth Circuit, we apply a three-prong test to determine whether a district court may properly impose sanctions to the extent of attorney fees under this bad faith exception. “In order to award attorney fees under this bad faith exception, a district court must find [1] that ‘the claims advanced were meritless, [2] that counsel knew or should have known this, and [3] that the motive for filing the suit was for an improper purpose such as harassment.’ ” Big Yank, 125 F.3d at 313 (quoting Smith v. Detroit Fed’n of Teachers, Local 231, 829 F.2d 1370, 1375 (6th Cir.1987)). The Big Yank requirement that a court find that a party have an improper purpose in filing a suit resembles the Supreme Court’s general requirement that a court find bad faith or conduct tantamount to bad faith. Put another way, satisfying the overarching bad faith requirement mandated by the Supreme Court before imposition of sanctions under a court’s inherent powers often automatically satisfies the improper purpose prong of the Big Yank test. See First Bank of Marietta, 307 F.3d at 519 n. 15 (“This Court has likewise used ‘improper purpose’ and ‘bad faith’ interchangeably.”). Cf. Chambers, 501 U.S. at 46 n. 10, 111 S.Ct. 2123 (“[T]he bad-faith exception resembles the third prong of Rule ll’s certification requirement, which mandates that a signer of a paper filed with the court warrant that the paper ‘is not interposed for any improper purpose, such as to harass or to cause unnecessary delay or needless increase in the cost of litigation.’ ”). As this case demonstrates, there is some confusion regarding what behavior constitutes improper purpose, bad faith or conduct tantamount to bad faith. In particular, it is not entirely clear from our case law whether the simple fact that a party pursued a clearly frivolous and meritless lawsuit constitutes conduct that is “tantamount to bad faith” sufficient for the imposition of sanctions. 4 Such a rule clearly *753 risks collapsing the third Big Yank prong (that the motive for filing was improper purpose) into the first two prongs (that the claims were meritless, and that counsel knew or should have known that the claims were meritless). See First Bank of Marietta, 307 F.3d at 531 (Gilman, J., concurring in part and dissenting in part) (“Nothing in Big Yank Corp. or its progeny, however, indicates that this third requirement is surplusage.”). This exact concern was the subject of a debate in First Bank of Marietta, 307 F.3d at 524. Having found that the evidence in that case demonstrated that the claim was clearly meritless and that counsel knew or should have known so, the majority observed that the district court had expressly concluded that the sanctioned party acted in bad faith in bringing the action. Id. “Implicit in this finding,” the court explained, “is that [the sanctioned party] improperly used the court system to try to force a result that it could not obtain under the applicable law, which is separate and district [sic ] from the issue of whether [the sanctioned party] was attempting to prevail on the merits of its lawsuit and collect payment on a meritless claim.” Id.; see also id. at 525 n. 19 (“Our inference is in accord with this Court’s ruling ... that ‘[i]f, from the facts found, other facts may be inferred which will support the judgment, such inferences shall be deem [sic ] to have been drawn by the District Court.’ ”) (quoting Zack v. Comm’r of Internal Revenue, 291 F.3d 407, 412 (6th Cir.2002)). 5 The dissent, however, challenged the suggestion that it is possible to infer “that a claim is filed for an ‘improper purpose’ if the claim is ‘invalid’ and is put forth by a litigant who knows that the claim is invalid.” Id. at 531 (Gilman, J., concurring in part and dissenting in part). The majority’s discussion of this inference, moreover, was dicta: the majority ultimately affirmed the district court’s finding of bad faith after concluding that the plaintiffs had made “improper use of the courts.” To demonstrate this improper use, the majority pointed both to the plaintiffs pursuit of a meritless suit and to the plaintiffs withholding of material evidence in support of its claim. See id. at 523 n. 18. In other words, the First Bank of Marietta court did not find that the plaintiffs knowing pursuit of a meritless claim by itself demonstrated bad faith or improper purpose within the meaning of the third Big Yank prong. In order to clarify any remaining confusion, we hereby hold that the “mere fact that an action is without merit does not amount to bad faith.” Miracle Mile Assocs. v. City of Rochester, 617 F.2d 18, 21 (2d Cir.1980) (citing Runyon v. McCrary, 427 U.S. 160, 183-84, 96 S.Ct. 2586, 49 L.Ed.2d 415 (1976)). By this, we mean that in order for a court to find bad faith sufficient for imposing sanctions under its inherent powers, the court must find something more than that a party knowingly pursued a meritless claim or action at any stage of the proceedings. 6 In *754 Big Yank, the court indicated that the “something more” could be a finding that the party had filed the suit “for purposes of harassment or delay, or for other improper reasons.” 125 F.3d at 313-14 (quoting Colombrito v. Kelly, 764 F.2d 122, 133 (2d Cir.1985)). In First Bank of Marietta, the “something more” was a party’s improper use of the courts, which was demonstrated by the fact that the plaintiff “file[d] a meritless lawsuit and [withheld] material evidence in support of a claim.” 307 F.3d at 523 n. 18 (emphasis in original) (citing Mansmann v. Tuman, 970 F.Supp. 389, 396 (E.D.Pa.1997)). In Chambers, the Supreme Court suggested that a court could find that “fraud has been practiced upon it, or that the very temple of justice has been defiled” by actions such as a party “delaying or disrupting the litigation or by hampering the enforcement of a court order.” Chambers, 501 U.S. at 46, 111 S.Ct. 2123 (citations removed). Harassing the opposing party, delaying or disrupting litigation, hampering the enforcement of a court order, or making improper use of the courts are all examples of the sorts of conduct that will support a finding of bad faith or improper purpose; given the requirements of the Big Yank test, however, these sorts of conduct cannot be demonstrated solely by the fact that a party knowingly pursued a meritless claim or action. IV. Having clarified the law as to when a district court may properly assess sanctions under its inherent power, we turn to the question of whether, in this case, the district court abused its discretion in imposing a sanction of more than one million dollars on Meisenheimer. Meisenheimer admits to knowing the facts the district court identified as rendering BDT’s claims meritless; the questions remaining are thus whether in light of those facts: (1) BDT’s suit was meritless; (2) Meisenheimer knew or should have known that its suit was meritless; and (3) Meisenheimer acted in bad faith or with an improper purpose. See Big Yank, 125 F.3d at 313. A. Whether the suit was meritless, as Meisenheimer knew or should have known. The district court identified three facts known to BDT, Higgs, and Meisenheimer that, in its view, rendered BDT’s suit meritless: (1) that BDT had disclosed its technology to Hewlett Packard and Tektronix under one-way agreements offering no protection to BDT; (2) that BDT’s Confidential Disclosure Agreements with Lexmark were in place at all relevant times and permitted Lexmark to disclose BDT’s technology; and (3) that there had been a public sale of BDT’s supposedly secret technology before Lexmark sold its printer tray. In another context, we have concluded that legal theories advanced by a plaintiff must be considered “indisputably meritless” when a plaintiff cannot, even through the amendment of his pleadings, truthfully establish the existence of any injury. Wilson v. Yaklich, 148 F.3d 596, 602 (6th Cir.1998). While Meisenheimer makes several arguments as to why the first and second of these facts were not necessarily and obviously fatal to BDT’s suit, we need not reach any conclusions regarding these facts, as we find that the third fact — -that Hewlett Packard sold the *755 HP4v in 1994, two years before Lexmark released the Optra S — alone meant that BDT, aided by Higgs and Meisenheimer, was bringing a suit that was at its heart about the misappropriation of trade secrets that were no longer secret. In other words, this third fact meant that BDT was never injured, and so its suit was necessarily meritless. In response, Meisenheimer makes three arguments: first, it points out that while all three judges on the Sixth Circuit panel found this fact dispositive on appeal, the district court did not rely on this fact in ruling on Lexmark’s motion for summary judgment. Nonetheless, there is no requirement for a district court to include every possible basis for its determination to grant summary judgment; here, the district court presented two justifications, both of which were affirmed by the Sixth Circuit. One of those justifications, moreover, was that the information Lexmark allegedly misused did not constitute a cognizable trade secret, as, prior to Lexmark’s supposed misappropriation, BDT had disclosed its technology to both Hewlett Packard and Tektronix under one-way Confidential Disclosure Agreements that did not protect BDT’s confidential information. BDT Products, 274 F.Supp.2d at 891. Given that the district court had thus already concluded that the “trade secrets” were not secret at all at the time Lexmark allegedly used BDT’s information, the absence of what the district court now sees as an obvious point, even if telling, is not dispositive. Meisenheimer next argues that the fact that alternative means could have been used to obtain trade secrets is no defense to misappropriation of trade secrets. As Lexmark points out, however, the fact that the HP4v was sold in 1994 meant (provided that the HP4v revealed all of BDT’s trade secrets) that by 1996 (when Lexmark released the Optra S printer) there were no more trade secrets for Lexmark to misappropriate. In other words, Lexmark did not need a defense to trademark misappropriation, because it never actually appropriated any trade secrets. See 4-15 Milgrim On Trade Secrets § 15.01[1] (2009) (“Plaintiff, by the preponderance of the evidence, must prove ... Plaintiff is the owner of a trade secret.”); § 15.01[l][d][iv] (2009) (“[I]f plaintiff has engaged in acts, such as selling a marketed product that reveals the trade secret ... then the plaintiff may have difficulty in meeting its burden of showing that the subject matter sued upon is a trade secret.”). See also id. at § 1.05[2] (“Thus, it is an almost undisputed proposition that when an article, the ‘secret’ nature of which is fathomable upon scrutiny and inspection, is marketed, the ‘secret’ is lost.... A similar loss of trade secret status may occur when possessors of a trade secret elect to disclose it either without adequate safeguards or subject to an insufficiently protective contract.”). Finally, Meisenheimer argues that as part of its suit BDT identified eight trade secrets, only one of which was the product design, and so the sale of the HP4v could not have revealed seven of the secrets. It is true that, in addition to the product design, BDT specifically claimed as trade secrets: (1) research demonstrating the feasibility, reliability, and expense of the product design; (2) data from testing the product design; (3) prototypes of the final product design; (4) line drawings of the LF 2000; (5) three 1993 proof-of-concept drawings related to the product design; (6) two 1994 proof-of-concept drawings related to the product design; and (7) an engineering feasibility analysis related to how the product design might work. (PI. Resp. to Mem. Op. and Order and Req. for Status Conf., Dec. 13, 2000, R.E. 103.) All of these “secrets,” however, were super-ceded by the finished construction of the *756 HP4v, Hewlett Packard’s successes or difficulties with commercialization of the HP4v, and the reception in the market to the HP4v. 7 That the HP4v tray was commercialized years before Lexmark produced the Optra S — and even years before BDT provided Lexmark with some of the information BDT later claimed Lexmark misappropriated — necessarily meant that, as of 1994, BDT’s supposed “trade secrets” were in fact not secret at all. As there were no trade secrets for Lexmark to misappropriate, any suit relying at its heart on the misappropriation of trade secrets must necessarily have been meritless — as Meisenheimer either knew or certainly should have known. B. Whether Meisenheimer acted in bad faith or with improper purpose. The most difficult question raised by this appeal is whether the district court abused its discretion by finding that Meisenheimer acted in bad faith or with improper purpose, or that Meisenheimer’s conduct was tantamount to bad faith. The district court appears to have based its finding of bad faith or improper purpose upon two determinations, the first of which was that BDT, Higgs, and Meisenheimer all worked together to force Lexmark into a settlement. As the district court wrote: Nonetheless, BDT enlisted the services of [Higgs], then [Meisenheimer], to pursue claims — any claims really — that would enable BDT to achieve what it wanted: an award that would make up for BDT’s failure to obtain a negotiated royalty or license fee from Lexmark and that would provide assurances to its other clients of BDT’s ability to maintain its own “secrets,” for which those clients had paid, even though no secrets remained. Each and every step taken in the litigation of this matter — complaint, amended complaint, the various pleadings, discovery, and appeals — was a maneuver to obtain these ends in the face of facts which would defeat the claims and to, effectively, harass BDT’s former client and new-found competitor, Lexmark, until a settlement was obtained. BDT and its counsel gambled, hoping that Lexmark would simply determine that it was too expensive and too arduous to litigate the matter and decide to settle for a tidy sum before the merits of the claim could be decided. (DCM at 46-47.) Certainly, evidence that a party or its attorneys attempted to force the other side into settlement in what they knew was a meritless suit constitutes improper purpose. The district court, moreover, pointed to evidence demonstrating convincingly that this is what BDT and Higgs had in mind. That evidence, however, relates only to BDT and Higgs, and dates from November of 1997 and May of 1998 — long before Meisenheimer became involved with the suit. The district court did not identify any evidence in the record before us specifically indicating that Meisenheimer too was focused on forcing a settlement by Lexmark. Nor, for that matter, did Lex *757 mark, either at oral argument or in supplemental briefing directed to the issue of bad faith or improper purpose. The second determination on which the district court based its bad faith finding was that Meisenheimer, BDT, and Higgs were pursuing a meritless suit, and that bad faith can be demonstrated solely by “the filing of a clearly meritless claim ... that resulted in extensive discovery and an appeal.” (DCM at 42 (quoting First Bank of Marietta, 307 F.3d at 525).) In Part III of this opinion, however, we concluded that this determination relies on a misreading of our case law. As we indicated there, a determination that a plaintiff filed and pursued a meritless claim, by itself, cannot satisfy the bad faith/improper purpose prong of the Big Yank test. Given that the court did not identify sufficient evidence in the record to support its conclusion that Meisenheimer was attempting to force Lexmark into an unnecessary settlement, and given that the court based part of its bad faith finding on a misreading of Sixth Circuit law, we conclude that the court failed to support the finding of bad faith or improper purpose required by the third Big Yank prong, 125 F.3d at 313. It is true that even where a district court “made no express finding of willfulness, bad faith or recklessness, we may nonetheless affirm if ‘the record sets forth sufficient evidence to support [the district court’s] decision.’ ” Red Carpet Studios Div. of Source Advantage, Ltd. v. Sater, 465 F.3d 642, 647 n. 2 (6th Cir.2006) (citing Toombs v. Leone, 777 F.2d 465, 471 (9th Cir.1985)). It is a logical extension of that rule that where a district court made an express finding of bad faith, but perhaps made it on the wrong grounds, we may nonetheless affirm provided that there is sufficient evidence in the record to support the finding. Here, however, despite having several opportunities to do so, Lexmark has simply been unable to point to any such evidence demonstrating that Meisenheimer acted in bad faith or with improper purpose. Accordingly, we find that the district court abused its discretion when it imposed sanctions under its inherent powers on Meisenheimer, as there was not sufficient evidence in the record demonstrating that Meisenheimer acted in bad faith or with improper purpose. V. We conclude first that sanctions under 28 U.S.C. § 1927 may be imposed only on individual attorneys, and not law firms. The continued validity of sanctions imposed on Meisenheimer therefore depends on whether the district court abused its discretion in imposing sanctions under its inherent power. We conclude that the district court did not abuse its discretion by determining that, at the very least, Meisenheimer should have known that the commercialization of the HP4v in 1994 meant that, well before 1996, BDT no longer had any trade secrets for Lexmark to misappropriate. As a result, the district court did not abuse its discretion in finding that Meisenheimer pursued a meritless lawsuit, and that Meisenheimer knew or should have known that it was pursuing a meritless suit. In concluding that Meisenheimer had acted in bad faith in doing so, however, the district court relied in part on a misstatement of Sixth Circuit law and without sufficient evidence imputed Higgs and BDT’s improper purpose to Meisenheimer. Accordingly, we VACATE the imposition of sanctions on Meisenheimer and REMAND to the district court for entry of an order denying Lexmark’s motion for sanctions as to Meisenheimer and for any other necessary actions consistent with this opinion. Footnotes [1] KRS § 365.886, which is a Kentucky statute permitting the award of prevailing party at *744 torney fees under the Kentucky Uniform Trade Secrets Act (KUTSA), is not at issue in this appeal. [2] BDT and Higgs initially appealed the sanctions decision, but then reached settlement agreements with Lexmark. On October 8, 2008, we granted BDT’s and Higgs’ motions to dismiss their appeals with prejudice. [3] We similarly upheld the district court's grant of Lexmark’s motion for costs. See BDT Prods., Inc. v. Lexmark Int’l, Inc., 405 F.3d 415, 416 (6th Cir.2005). [4] Some of this confusion might arise from a misreading or misapplication of the decision in First Bank of Marietta. In that case, the court observed that "[o]f course, the filing of a clearly meritless claim ... that resulted in extensive discovery and an appeal, is evidence of a bad faith and abuse of the courts....” 307 F.3d at 525. The district court in this case seized upon this observation as support for the conclusion that the filing of a clearly *753 meritless claim by itself demonstrates bad faith or improper purpose and abuse of the courts. Saying that the filing of a meritless claim is evidence of bad faith, however, is clearly not the same thing as saying that the filing of a meritless claim proves bad faith. [5] The court slightly misquoted Zack, which in fact read “such inferences should be deemed to have been drawn by the District Court.” Zack, 291 F.3d at 412 (emphasis added) (quoting Grover Hill Grain Co. v. Baughman-Oster, Inc., 728 F.2d 784, 793 (6th Cir.1984)). [6] For a court to impose sanctions under its inherent powers, it is not necessary that the court find that an action was meritless as of filing, or even shortly thereafter. It can become apparent part-way through a suit that an action that initially appeared to have merit is in fact meritless; parties and attorneys have a responsibility to halt litigation whenever they realize that they are pursuing a merit-less suit. As in this case, moreover, a party or firm might enter an action long after the filing of the initial complaint, but may still be sane *754 tionable under a court's inherent powers if it acts in bad faith. The "something more” that a court must find to meet the third prong of the Big Yank test may similarly occur at any stage of the proceedings. A court imposing sanctions under its inherent powers may consider the nature and timing of the actions that led to a finding of bad faith in determining whether to impose sanctions on conduct from that point forward, or instead to infer that the party’s bad faith extended back in time, perhaps even prior to the filing of the action. [7] Meisenheimer also argues that the district court “impliedly found in favor of BDT” by observing in its decision granting Lexmark summary judgment that Lexmark continued to ask for and receive assistance from BDT even after commercial release of the HP4v. See BDT Prods., 274 F.Supp.2d at 882. Meisenheimer here misreads the district court’s decision. Desiring assistance from experts and being unable to develop a product without such assistance are two different things. The fact that Lexmark continued to ask for BDT’s help does not mean that the HP4v did not reveal all of BDT’s trade secrets, or that Lexmark would have been unable to develop the Optra S without BDT's assistance and solely with the guide of the HP4v.