NOT PRECEDENTIAL
UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT
_______________
Nos. 22-2739, 22-2813, 22-2889 and 23-3104
_______________
ELMAGIN CAPITAL, LLC,
Appellant in Nos. 22-2739 and 22-2889
v.
CHAO CHEN; KARL PETTY; ENTERGRID, LLC;
ENTERGRID FUND I, LLC
Appellants in Nos. 22-2813 and 23-3104
_______________
On Appeal from the United States District Court For the Eastern District of Pennsylvania (D.C. No. 2-20-cv-2576)
Circuit Judge: Honorable Stephanos Bibas*
_______________
Submitted Under Third Circuit L.A.R. 34.1(a)
November 3, 2023
Before: JORDAN, ROTH., and AMBRO, Circuit Judges
(Filed March 21, 2024 )
_______________
OPINION
_______________
*
The Honorable Stephanos Bibas, Circuit Judge sitting by designation pursuant to 28 U.S.C. § 291(b).
This disposition is not an opinion of the full court and, pursuant to I.O.P. 5.7, does not constitute binding precedent.
JORDAN, Circuit Judge.
Elmagin Capital, LLC appeals the District Court’s denial of motions for judgment as a matter of law and a new trial for breach of contract and trade secret misappropriation. Chao Chen cross-appeals the District Court’s denial of his request for attorneys’ fees. For the following reasons, we will affirm. I. BACKGROUND In 2014, brothers Richard and Kevin Gates joined with Chao Chen to found Elmagin Capital, LLC. It was established to trade in financial transmission rights (“FTRs”) in wholesale electricity markets.1 Chen developed trading strategies for Elmagin. During his tenure at Elmagin, he developed or participated in developing the two trading strategies still at issue in this case, one called “Breck,” and another called “Faber.” 2 While at Elmagin, Chen engaged defendant Karl Petty as a consultant to provide historical data analysis services. Before doing any work for Elmagin, Petty signed a
3
Consulting Agreement that prohibited him from using, exploiting, or disclosing Elmagin’s “confidential information.” In 2016, Chen informed the Gates brothers that he wished to leave Elmagin. The Gates brothers purchased Chen’s share in Elmagin, and Chen signed a Non-Disclosure and Non-Competition Agreement (“NDA”) that prohibited him from engaging in a competing business for one year and from ever using, exploiting, or disclosing Elmagin’s confidential information, including its trading strategies. Chen left Elmagin in January 2018.
In January 2019, he formed Entergrid, LLC and Entergrid Fund I, LLC for the purpose of trading in FTRs in wholesale electricity markets. He developed FTR trading strategies, including the allegedly misappropriating strategies, “Hydra” and “Gryphon.” Petty joined Entergrid in the spring of 2019. Chen taught Petty the basics of power markets and FTRs, and Petty kept notes.
Elmagin sued Chen, Petty and Entergrid in District Court, accusing them of disclosing and using the Breck and Faber strategies in violation of the NDA and the Defend Trade Secrets Act (“DTSA”), 18 U.S.C. § 1836, et seq., as well as Pennsylvania’s Uniform Trade Secrets Act (“PUTSA”), 12 Pa. Cons. Stat. § 5302, et seq. Chen responded that he only shared public information, that Breck and Faber combined well- known elements, and that his trading strategies were different from Elmagin’s.
The case went to trial in May of 2022. The jury found that the Breck and Faber strategies were trade secrets, but that Chen did not use them to develop Hydra and Gryphon. Chen accused Elmagin of bringing the suit in bad faith and asked the District Court for attorneys’ fees. The Court sent that question to the jury, and the jury found bad
4
faith, but the Court considered the jury’s answer advisory and declined to award attorneys’ fees. The Court also denied Elmagin’s motions for judgment as a matter of law and a new trial. The parties’ cross appeals followed. II. DISCUSSION3 A. The Misappropriation Claim To prevail on a claim for misappropriation of trade secrets under both the DTSA and the PUTSA, a plaintiff must establish (1) the existence of a trade secret, (2) that the trade secret was protectible, and (3) that it was misappropriated by the defendant. 18 U.S.C. §§ 1836(b)(1), 1839(3), (5); 12 Pa. Cons. Stat. § 5302.4 A trade secret may be a business method as long as it has “independent economic value” and “the owner … has taken reasonable measures to keep [it a] secret.” 18 U.S.C. § 1839(3); Oakwood Labs.
5
LLC v. Thanoo, 999 F.3d 892, 905 (3d Cir. 2021). The plaintiff must define his trade secret “with sufficient particularity … to permit the defendant to ascertain … the boundaries within which the secret lies.” Id. at 906. Trade secret law does not protect “general knowledge in the trade or … special knowledge of those persons who are skilled in the trade.” Id.
Misappropriation encompasses “disclosure or use of a trade secret” without consent, including “relying on the trade secret to assist or accelerate research or development[.]” 18 U.S.C. § 1839(5)(B); 12 Pa. Cons. Stat. § 5302; Oakwood, 999 F.3d at 909. Rarely can a plaintiff demonstrate misappropriation through direct evidence. Oakwood, 999 F.3d at 913. Instead, a plaintiff may rely on circumstantial evidence of access to the trade secrets and similarity between the secrets and the accused product. Id. at 909 n.18. But when public information and the defendant’s own knowledge, uninformed by the trade secret, effectively confine the scope of the trade secret, “more than … similarity” may be needed to prove misappropriation. Am. Can Co. v. Mansukhani, 742 F.2d 314, 331 (7th Cir. 1984); cf. Oakwood, 999 F.3d at 912 n.19 (suggesting a “plus factor” may be helpful when relying on circumstantial evidence).
At trial, Elmagin claimed that Hydra copied aspects of Breck, and Gryphon copied aspects of Faber. The jury was instructed to “assess whether the whole strategy, as opposed to its individual steps” warranted trade secret protection. (J.A. at 1526.) The jury found that both Faber and Breck constituted trade secrets. The jury was instructed to find misappropriation if the defendants, among other things, had access to the strategies
6
and if the strategies were “similar enough.” (J.A. at 1527-28.) They held that Chen, Petty, and Entergrid did not misappropriate the Breck and Faber strategies.
On appeal, Elmagin asks for a new trial or judgment as a matter of law on its claim that Hydra represents a misappropriation of Breck and a breach of the NDA. It also says that a new trial on all its claims is warranted because, in its view, the District Court erroneously permitted Chen’s prejudicial and irrelevant expert testimony.
1. The jury reasonably found that Hydra did not constitute a misappropriation of Breck.
Elmagin contends that the jury accepted its trade secret theory regarding Breck and that no evidence supported their finding regarding no misappropriation. Defining Breck, as it does, at a high level of abstraction and ignoring implementation details, Elmagin claims that Hydra is similar to Breck. But Elmagin cannot claim that the implementation details are irrelevant when it used them to defeat summary judgment and presented them to the jury at trial. Cf. Amazon.com, Inc. v. Barnesandnoble.com, Inc., 239 F.3d 1343, 1351 (Fed. Cir. 2001) (“A patent may not, like a ‘nose of wax,’ be twisted one way [for validity] and another to find infringement.”).
At summary judgment, the District Court only permitted Elmagin’s case to go forward because it pleaded its trade secret specifically using “voluminous exhibits … discussing the strategies and how they are constructed[.]” (J.A. at 3706 (emphasis added).) For example, Elmagin said that Breck filters the FTR universe “based upon overlays of other proprietary strategies, operational constraints on bid limits, correlation filters; or otherwise.” (D.I. 59 at 48 (cleaned up).) And at trial, Elmagin implored the
7
jury to ignore the difference between minimum spanning trees (“MSTs”)5 and shift factors6 and argued that Hydra used the same “correlation filtering” as Breck. (J.A. at 8021-22.) Elmagin was responsible for “identify[ing] the information it claims as a trade secret[,]” and it cannot fault the jury for relying on the implementation details Elmagin itself made relevant. Oakwood, 999 F.3d at 905; see id. at 906 (Whether a particular piece of information constitutes a trade secret is generally a question for the jury).
At trial, the jury heard evidence that Breck and Hydra were different. For example, Hydra uses shift factors to define the universe of possible FTRs and Breck uses MSTs]; Breck considers all biddable constraints before filtering, but Hydra begins with constraints that have a history of congestion; and Hydra uses a Python-based machine learning “hierarchical clustering” method7 to filter while Breck uses a “correlation
An MST is a “mathematical concept,” (Answering Br. at 15,) or “an algorithm”
5
(Opening Br. at 11,) that uses historical data to predict future congestion. (J.A. at 6939- 40.)
6 Shift factors measure the sensitivity of a path (and thus an FTR) to changes on the grid. That is, they determine how affected a given path on the power lines will be as the result of a given change to the power grid.
7 Python is a high-level computer programming language. Python, Encyclopedia Britannica, https://www.britannica.com/technology/Python-computer-language. Chen testified that clustering takes “a group of things and tell[s] me how similar or different … each thing is to every other[.]” (J.A. at 6998.) This is important because “if two FTR’s are very similar … I only want to trade one of them.” (J.A. at 6998.)
8
matrix”8 (J.A. at 6460-62, 6998-7000). The jury could have reasonably relied on those differences to support its finding of no misappropriation.9 Also, the jury could have reasonably relied upon the strategies’ different results in finding that Breck and Hydra were dissimilar. Elmagin’s expert testified that economists can find “functional identity” between strategies when they “perform[] the same sort of tasks in the same way[,] leading to … similar if not exactly identical results.” (J.A. at 7182-83.) And Chen’s expert testified that the strategies were “nowhere near identical” at the result level, with the parties’ bids having only a 1-6% overlap.10 (J.A. at 7497, 7515.)
8 The correlation matrix takes historical prices and removes highly correlated paths. Elmagin’s technical witness testified, “you kind of sum [up] the correlation of one path[] to all the other paths to see which one has the highest number” of correlations. (J.A. at 6461.)
Elmagin claims that Chen “developed the idea of using shift factors” to create the 9
FTR universe while at Elmagin, (Opening Br. at 46,) but the cited record only indicates that Chen was testing the theory for use in Breck, not that he invented the idea. (J.A. at 6559 (“Q. Did you do significant research into shift factor for use in Breck? A. I studied it for multiple days.”); see also J.A. at 6592 (“[Chen:] I would say most people that trade identify FTR’s using shift factors.”).)
Elmagin claims that Chen’s expert, Cristian Zarcu, should never have presented 10
his results-comparison testimony because his methodology was faulty. But the District Court found that Elmagin had forfeited that argument. We agree. While Elmagin tried to get the expert’s testimony thrown out before trial, it did so on the basis that the expert was not conversant in electricity trading, not on the ground that his results-comparison testimony was faulty. (J.A. at 138-47 (challenging qualifications and familiarity with record); 3979-81 (challenging purported legal conclusions).) And Elmagin cannot resurrect its lack-of-expertise argument on appeal since it did not raise it at the District Court in its motion for a new trial. See Unitherm Food Sys., Inc. v. Swift-Eckrich, Inc., 546 U.S. 394, 404 (2006) (“[A] party is not entitled to pursue a new trial on appeal unless that party makes an appropriate postverdict motion in the district court.”).
9
Finally, the jury found that Elmagin failed to prove that Chen did not independently develop Hydra. Elmagin challenges this finding as a matter of law. It claims “independent development” is only relevant to acquisition of trade secrets and not to use. (Opening Br. at 47-48.) It is mistaken. “If there is proof that the defendant independently developed a technique that resembles the trade secret, then the defendant did not ‘use’ the trade secret” under Pennsylvania law. Moore v. Kulicke & Soffa Indus., Inc., 318 F.3d 561, 567 (3d Cir. 2003); see also 4 Milgrim on Trade Secrets § 15.01[1] (proof of independent development enough to counter circumstantial evidence of similarity).11 Then, citing legal advice from a treatise and two out-of-circuit cases, Elmagin argues that independent development is impossible as a matter of law when the defendant had knowledge of the trade secret. Again, it is mistaken. The jury evidently determined that, despite Chen’s awareness, he developed Hydra around Elmagin’s proprietary information, using “general knowledge in the trade or … special knowledge.” Oakwood, 999 F.3d at 906. There was nothing unreasonable in that conclusion. For example, the jury heard evidence that using a 28-day pricing period, present in both Hydra and Breck, was “typical,” (J.A. at 7509.) and a “historical average is used throughout the investment
11 Elmagin claims Moore is inapplicable because it predates the DTSA and the PUTSA. (Reply Br. at 38.) But we have held that the PUTSA did not “effect[] a substantive shift” in defining shared terms. Bimbo Bakeries USA, Inc. v. Botticella, 613 F.3d 102, 109 n.7 (3d Cir. 2010). “Use” appears in the PUTSA and the DTSA. 12 Pa. Cons. Stat. § 5302; 18 U.S.C. § 1839(5)(B). That “independent derivation” is applicable to acquisition, 18 U.S.C. § 1839(6), does not make it irrelevant to use.
10
world, and ha[d] been for a very long time.” (J.A. at 7484.) The point is not whether we agree or disagree with the jury. It is that the jury was free to reach the conclusion it did based on the evidence presented. Kars 4 Kids Inc. v. Am. Can!, 8 F.4th 209, 218 n.8 (3d Cir. 2021).
2. The Court did not abuse its discretion in admitting Chen’s expert testimony.12
Elmagin claims that Chen’s expert, W. Anthony Mason, proffered irrelevant and prejudicial testimony by comparing the code for the strategies, while Elmagin’s theory of the case concerned the strategies themselves. Before trial, Elmagin made essentially the same arguments in a motion in limine, and the District Court denied the motion. The Court said that Elmagin was too late in seeking to exclude Mr. Mason entirely and that any jury confusion could be cured “with a jury instruction.” (D.I. 173 at 32; 35-36 (“If it’s fit, then you missed the Daubert deadline”).) At the pre-trial conference, the Court reiterated that cross-examination would be enough to “clarify in the jury’s mind that this is not about whether the code is the same, but whether the secrets and the algorithms are the same.” (J.A. at 1092.) The Court did not abuse its discretion in that ruling.
Indeed, Elmagin cross-examined Mr. Mason and had ample time to remind the jury of its trade secret theory. Moreover, the District Court noted that Elmagin did
12 We review decisions regarding the admissibility of expert testimony and “decision[s] to include or exclude evidence arising under the Federal Rules of Evidence 401, 402 and 403” for an abuse of discretion. Barker v. Deere & Co., 60 F.3d 158, 161 (3d Cir. 1995). A court abuses its discretion if its decision was “arbitrary, fanciful or clearly unreasonable.” Moyer v. United Dominion Indus., Inc., 473 F.3d 532, 542 (3d Cir. 2007).
11
intimate that code was relevant to its theory. During its closing arguments Elmagin’s counsel noted in describing trade secret access that “Mr. Mason said, well, I didn’t see any code checked in by Dr. Chen. … [Y]ou’ll have to decide for yourself, but it seems clear to me that there was access to all of this information.” (J.A. at 8018.) Elmagin cannot persuasively claim prejudice when it brought code to the jury’s attention.
B. The Breach of Contract Claim At summary judgment, the District Court held that Elmagin’s NDA with Chen protected “Elmagin’s trading strategies and any other confidential information related to the strategies.” (J.A. at 3718.) The jury was asked to determine whether Chen used or disclosed the same. Elmagin claimed that Chen used its strategies to develop Hydra and Gryphon and disclosed them to Petty in the spring of 2019. At trial, the jury heard evidence that Chen only used and disclosed public information. For example, Elmagin’s 28-day pricing period was “typical,” (J.A. at 7509,) and “a historical average is used throughout the investment world, and ha[d] been for a very long time” (J.A. at 7484); its price collar tactic was necessary in all buy/sell strategies; its method for finding the best FTR was common to all power market traders; its Faber strategy, “bid[ding] a price slightly above average,” (J.A. at 7337,) was available to the public because anyone could see Elmagin’s bids and because bidding a premium over an expected price was a common technique; its use of Dayzer, a third-party program common in the power trading world, was typical; its portfolio diversification technique was common; and Chen only told Petty things that Petty already knew about Elmagin. The jury was allowed to rely upon that evidence to find that Chen did not breach his NDA.
12
Elmagin faults the District Court for making a logical error. It argues that bidding with a price collar, bidding a premium, and using Dayzer are common tactics, but Chen told Petty that Elmagin did those things. Nonetheless, a jury could have reasonably found that those methods were so obvious that anyone in the power trading world would know Elmagin used them.
C. Attorneys’ Fees13 A district court “may” award “reasonable attorney’s fees to the prevailing party” if “a claim of the misappropriation is made in bad faith.” 18 U.S.C. § 1836(b)(3)(D); accord 12 Pa. Cons. Stat § 5305(1). A claim is brought in bad faith if the plaintiff “complete[ly] lack[ed] ... evidence” and “[knew] or [was] reckless in not knowing that its claim[s]” lacked merit. Krafft v. Downey, 68 A.3d 329, 336 & n.9 (Pa. Super. Ct. 2013); Sweet St. Desserts, Inc. v. Better Bakery, LLC, No. 12-cv-6115, 2017 WL 6311664, at *3 (E.D. Pa. Dec. 11, 2017).14 “[W]hen attorney’s fees are awarded pursuant to a statutory prevailing party provision, they are regarded as an ‘equitable’ remedy[,]” and the parties do not have a Seventh Amendment right to a jury in determining legally or factually whether attorneys’
13 For bad faith, we review “findings of fact for clear error, … conclusions of law de novo,” and “the District Court’s [discretionary] attorneys’ fees determination for abuse of discretion.” McCutcheon v. Am.’s Servicing Co., 560 F.3d 143, 147 (3d Cir. 2009).
14 Neither the DTSA or the PUTSA defines “bad faith” but district courts in our jurisdiction and others have used the test outlined above. See Krafft, 68 A.3d at 334 n.6 (collecting cases), 333 (citing Stilwell Dev. Inc. v. Chen, No. CV86-4487, 1989 WL 418783 (C.D. Cal. Apr. 25, 1989)). The jury was instructed to employ the Stilwell test in assessing whether Elmagin’s suit was brought in bad faith.
13
fees should be awarded. AIA Am., Inc. v. Avid Radiopharmaceuticals, 866 F.3d 1369, 1373 (Fed. Cir. 2017).
The jury found that Elmagin brought its claims in bad faith. But the District Court denied Chen’s attorneys’ fees claim, reasoning that Elmagin had some evidence for its misappropriation claims. We discern no clear error in that finding. While the jury decided the strategies were different enough to find no misappropriation, there was sufficient evidence for the trial judge to conclude Elmagin had not acted in bad faith. The Court had, in other words, an appropriate basis to rule, as a matter of law, that Chen did not carry his burden to demonstrate bad faith. (J.A. at 20 (citing Fed. R. Civ. P. 50(a)(1)).) Finally, the Court exercised its jurisdiction to deny attorneys’ fees, and there was no abuse of discretion in that.
On appeal, Chen argues that he is entitled to a jury decision on his attorneys’ fees claim, and that the Court was wrong as a matter of law. He says he is entitled to a jury decision because the DTSA and the PUTSA are modern statutes and not subject to the general rule that “determin[ing] whether a particular action” is entitled to a Seventh Amendment jury requires us to “compare the statutory action to 18th-century actions brought in the courts of England prior to the merger of the courts of law and equity.” Chauffeurs, Teamsters & Helpers, Loc. No. 391 v. Terry, 494 U.S. 558, 565 (1990) (internal quotation marks omitted). Regardless of the age of trade secret law, attorneys’ fees claims were available at law in English courts in the 1700s and were decided by the judge. AIA Am., 866 F.3d at 1373 (citing Arthur L. Goodhart, Costs, 38 Yale L. J. 849,
14
851-54 (1929)). That concludes the matter. Chen is not entitled to a jury decision on fees. III. CONCLUSION For the foregoing reasons, we will affirm the judgement of the District Court.