Silverthorne v. Sterling Seismic

• Court: Court of Appeals for the Fifth Circuit
• Decided: January 3, 2025
• Docket: 24-20006
• Status: Published

Opinion

Case: 24-20006 Document: 85-1 Page: 1 Date Filed: 01/03/2025

United States Court of Appeals for the Fifth Circuit United States Court of Appeals Fifth Circuit ____________ FILED January 3, 2025 No. 24-20006 ____________ Lyle W. Cayce Clerk Silverthorne Seismic, L.L.C.,

Plaintiff—Appellant,

versus

Sterling Seismic Services, Limited, doing business as Sterling Seismic & Reservoir Services,

Defendant—Appellee. ______________________________

Appeal from the United States District Court for the Southern District of Texas USDC No. 4:20-CV-2543 ______________________________

Before Smith, Clement, and Higginson, Circuit Judges. Jerry E. Smith, Circuit Judge: In advance of a trial for trade-secret misappropriation, the district court instructed the parties on how the plaintiff could prove reasonable- royalty damages under the Defend Trade Secrets Act of 2016, 18 U.S.C. § 1836. It certified that order for interlocutory appeal under 28 U.S.C. § 1292(b), and an administrative, or motions, panel of this court granted the plaintiff leave to appeal. That was error. The parties have not yet gone to trial, and the plaintiff hasn’t yet proven liability. Damages may never come up. The parties do not Case: 24-20006 Document: 85-1 Page: 2 Date Filed: 01/03/2025

No. 24-20006

need our input to proceed, and we would not speed up the litigation by weigh- ing in. The time spent on this appeal would only slow the proceedings. There is no reason to buck the hallmark rule that a party may appeal once and only after final judgment. We vacate the order granting leave to appeal, and we dismiss the appeal for want of jurisdiction and remand for further proceedings. 1 I. Casillas Petroleum Resource Partners II, L.L.C. (“Casillas”), li- censed seismic data from Silverthorne Seismic, L.L.C. (“Silverthorne”). Under their arrangement, Silverthorne would send the data to Sterling Seis- mic Services, Ltd. (“Sterling”), which would then process the data and pass it on to Casillas. Sterling’s process required more data than Casillas had paid for. Silverthorne thus required Sterling to agree to forward only the data that Casillas had licensed. But according to Silverthorne, Sterling sent to Casillas some unlicensed data, too, which Casillas then showed to potential investors. Silverthorne sued Sterling for misappropriating its trade secrets under the Defend Trade Secrets Act. Relevant to this appeal, Silverthorne seeks a remedy of a “reasonable royalty” under § 1836(b)(3)(B)(ii). In an order issued five days before trial, the district court set out the standard for calculating a “reasonable royalty,” adopting this court’s defini- tion of the same term under state law in University Computing Co. v. Lykes-

_____________________ 1 “A panel hearing the merits of an appeal may review a motions panel ruling, and overturn it where necessary.” X Corp. v. Media Matters for Am., 120 F.4th 190, 198 n.6 (5th Cir. 2024) (per curiam) (quoting Mattern v. Eastman Kodak Co., 104 F.3d 702, 704 (5th Cir. 1997) (citation omitted), abrogated in part on other grounds by Burlington N. & Santa Fe Ry. Co. v. White, 548 U.S. 53 (2006)).

2 Case: 24-20006 Document: 85-1 Page: 3 Date Filed: 01/03/2025

Youngstown Corp., 504 F.2d 518 (5th Cir. 1974). The order did not dispose of Silverthorne’s claims or otherwise foreclose it from recovering. On Silverthorne’s motion, the district court certified the order for interlocutory appeal under § 1292(b), concluding that (1) the reasonable- royalty standard was a controlling question of law, (2) there was substantial ground for difference of opinion regarding the standard, and (3) immediate appeal would materially advance ultimate termination of the litigation. The district court stayed the proceedings pending this court’s resolution of the certified question. An administrative panel of this court granted leave to appeal. II. We normally have jurisdiction to review only “final decisions of the district courts.” Firestone Tire & Rubber Co. v. Risjord, 449 U.S. 368, 373 (1981) (quoting 28 U.S.C. § 1291). A final decision is generally one “that ends the litigation on the merits and leaves nothing for the court to do but execute the judgment.” Id. (cleaned up). A party unhappy with an interlocu- tory ruling must wait until that final judgment, when it can appeal all claims of error at once. Id. at 374. The final-judgment rule “conserves judicial energy and eliminates the delays, harassment, and costs that would be occasioned by a succession of separate interlocutory appeals.” 2 “Even in [a] favorable state of our docket,” appeals “require [several] months.” 3 Clark-Dietz & Assocs.-Eng’rs, _____________________ 2 United States v. Bear Marine Servs., 696 F.2d 1117, 1119 (5th Cir. 1983), abrogated in part on other grounds by Burlington, 548 U.S. 53. 3 Take this case. The district court certified its order for interlocutory appeal five days before trial, then froze the trial proceedings pending this appeal. That was over a year ago. Trial would long have been finished. At that point, Silverthorne could have appealed, and, if necessary, this court could have remanded for a short postlude in the district court.

3 Case: 24-20006 Document: 85-1 Page: 4 Date Filed: 01/03/2025

Inc. v. Basic Constr. Co., 702 F.2d 67, 69 (5th Cir. 1983) (cleaned up). Section 1292(b) provides one “narrow exception” to the final- judgment rule. Garner v. Wolfinbarger (Garner II), 433 F.2d 117, 120 (5th Cir. 1970) (citing § 1292(b)). It allows a district court to certify an order for inter- locutory appeal where (1) the “order involves a controlling question of law,” (2) “there is substantial ground for difference of opinion” on that question, and (3) an immediate appeal “may materially advance the ultimate termina- tion of the litigation.” § 1292(b); see Garner v. Wolfinbarger (Garner I), 430 F.2d 1093, 1096–97 (5th Cir. 1970). “Interlocutory appeals are generally disfavored,” and § 1292(b) “must be strictly construed.” Allen v. Okam Holdings, Inc., 116 F.3d 153, 154 (5th Cir. 1997). After the district court certifies an order under § 1292(b), we may, but need not, grant leave to appeal the order. Bear Marine, 696 F.2d at 1119. The appellant must “persuad[e] the court of appeals that exceptional circumstances justify a departure from the basic policy of postponing appel- late review until . . . final judgment.” 4 And if we later conclude that we have improvidently accepted an appeal under § 1292(b), we “must vacate the earlier order granting leave to appeal and must remand the case to the district court.” 5

_____________________ Even if we decided the issue presented, Silverthorne’s interlocutory appeal would be a roadblock, not a shortcut, to final resolution of its action. On remand, this case will go to trial more than a year late. The loser (maybe also the winner) will bring the case back to this court on another monthslong journey and then, possibly, back to the district court. 4 ICTSI Or., Inc. v. Int’l Longshore & Warehouse Union, 22 F.4th 1125, 1131 (9th Cir. 2022) (O’Scannlain, J.) (cleaned up). 5 Bear Marine, 696 F.2d at 1119; accord Parcel Tankers, Inc. v. Formosa Plastics Corp., 764 F.2d 1153, 1156 (5th Cir. 1985); see also ICTSI, 22 F.4th at 1131 (9th Cir. 2022) (“The merits panel . . . must agree that the requirements of § 1292(b) are met”).

4 Case: 24-20006 Document: 85-1 Page: 5 Date Filed: 01/03/2025

We imprudently granted leave to appeal. The district court’s order involved no controlling question of law, and this appeal would not materially advance the ultimate termination of the litigation.

A. We have no appellate jurisdiction where the strict requirements of § 1292(b) are not met. “This court’s appellate jurisdiction under § 1292(b) extends only to interlocutory orders that involve a controlling question of law.” 6 “A controlling question of law must be one of law—not fact—and its resolution must materially affect the outcome of litigation in the district court.” ICTSI, 22 F.4th at 1130 (cleaned up). That effect must be “imme- diate” 7 and cannot depend on a party’s ability to prove additional facts. “[T]he purpose of § 1292(b) is not to offer advisory opinions rendered on hypotheses which evaporate in the light of full factual development.” Benoit v. Saint-Gobain Performance Plastics Corp., 959 F.3d 491, 508 (2d Cir. 2020) (cleaned up). Our answer to a question can immediately and materially impact an action by “terminat[ing] [it] in the district court” or by restoring it from an incorrect termination. 8 Controlling questions thus “include . . . whether a claim exists as a matter of law, whether a defense that will defeat the claim is available, and questions as to subject-matter jurisdiction, proper venue,

_____________________ 6 Malbrough v. Crown Equip. Corp., 392 F.3d 135, 136 (5th Cir. 2004) (cleaned up); accord McMillan v. Amazon.com, Inc., 983 F.3d 194, 198 (5th Cir. 2020). Contra post, at 5 (Higginson, J., dissenting) (proposing that once the district court has certified an order under § 1292(b), “there is nothing jurisdictional standing in [our] way” of deciding the appeal). 7 See Fujitsu Ltd. v. Tellabs, Inc., 539 F. App’x 1005, 1007 (Fed. Cir. 2013). 8 19 James W. Moore et al., Moore’s Federal Practice § 203.31[2] (3d ed. 2024).

5 Case: 24-20006 Document: 85-1 Page: 6 Date Filed: 01/03/2025

personal jurisdiction, and standing to maintain the action.” 9 That is why so many of our § 1292(b) appeals have come from dismis- sal or summary judgment. For example, in Overdam v. Texas A&M Univer- sity, we decided a § 1292(b) appeal from dismissal, identifying controlling questions in (1) whether a student was constitutionally entitled to cross- examine his accuser in Title IX proceedings; and (2) what the pleading stan- dard was for a Title IX claim. 43 F.4th 522, 526 (5th Cir. 2022) (per curiam). By reversing, we would have restored the case from dismissal, immediately impacting it. See Fujitsu, 539 F. App’x at 1007. In “rare” circumstances, even discovery orders can contain control- ling questions of law. Hyde Constr. Co. v. Koehring Co., 455 F.2d 337, 338–39 (5th Cir. 1972). For example, in Garner I, we accepted an interlocutory appeal to decide whether a core privilege shields evidence that is “essential” to a “complex case.” 10 “[P]ending motions to dismiss” depended in part on that evidence. Garner I, 430 F.2d at 1097. Our resolution of the privilege could have obviated a trial and the intervening years of pretrial litigation, so that issue was controlling in that context. Though a controlling question’s resolution “need not . . . terminate an action,” 11 a question is not controlling just because its answer would complicate a litigant’s ability to make its case. Nor is a question controlling _____________________ 9 Id. (cleaned up). 10 Note, Interlocutory Appeals in the Federal Courts under 28 U.S.C. § 1292(b), 88 Harv. L. Rev. 607, 623 (1975) [hereinafter Note, Interlocutory Appeals] (citing Gar- ner I, 430 F.2d 1093); Garner I, 430 F.2d at 1096–97 (controlling question in attorney-client privilege); Hyde, 455 F.2d at 338–39 (same). 11 Klinghoffer v. S.N.C. Achille Lauro Ed Altri-Gestione Motonave Achille Lauro in Amminstrazione Straordinaria, 921 F.2d 21, 24 (2d Cir. 1990); Garner I, 430 F.2d at 1097 (accepting a § 1292(b) appeal even where the decision would not “requir[e] complete dismissal”).

6 Case: 24-20006 Document: 85-1 Page: 7 Date Filed: 01/03/2025

because our answer could save the parties from a post-judgment appeal. 12 Otherwise, almost any disputed legal question could qualify for interlocutory review, and “the exception [provided by § 1292(b)] might well swallow the rule.” 13 With those principles in mind, damages issues generally do not control a case until the plaintiff establishes liability. 14 Only then would a plaintiff be entitled to damages. If this court reversed an incorrect damages order before then, its impact on the litigation would not be “immediate” but, instead, would be contingent on the rest of the plaintiff’s case. Fujitsu, 539 F. App’x at 1007. And if the plaintiff fails to show that the defendant is liable, our reversal will have had no effect on the litigation. ICTSI, 22 F.4th at 1130. One exception, among others, is where the damages issue would be dispositive. For example, in Vaughan v. Anderson Regional Medical Center, this court asserted § 1292(b) jurisdiction to decide whether a plaintiff could seek “pain and suffering and punitive damages” under the Age Discrimina- tion in Employment Act. 15 The district court had dismissed the plaintiff’s claims based on its answer to that question. By reversing, we would have restored the proceedings, immediately and materially impacting the case. Here, the district court’s pretrial order set out the standard for

_____________________ 12 ICTSI, 22 F.4th at 1130 (stating that the controlling question must affect the liti- gation “in the district court” (emphasis added)). 13 See Baldridge v. SBC Commc’ns, Inc., 404 F.3d 930, 932 (5th Cir. 2005) (cleaned up). 14 Fujitsu, 539 F. App’x at 1007; see also Note, Interlocutory Appeals, supra note 10, at 619 (“[A]n order involving computation of damages in complex litigation would not be controlling until a finding of liability, since it is only at that stage that reversal of trial court error could result in time savings.”). 15 849 F.3d 588, 590 (5th Cir. 2017).

7 Case: 24-20006 Document: 85-1 Page: 8 Date Filed: 01/03/2025

reasonable-royalty damages under the Defend Trade Secrets Act. Relying on University Computing’s definition of the same term under state law, the court required Silverthorne to prove “what Silverthorne and Sterling would have agreed to for Sterling to use the alleged trade secret.” The court then certi- fied that the damages issue was a controlling question of law. But that issue is not yet controlling. If we reversed now, we would have no “immediate impact on the course of the litigation” because Silver- thorne has not yet proven liability. Fujitsu, 539 F. App’x at 1007. The parties will proceed to trial regardless of whether we weigh in, and “nothing that we can do will prevent [the] trial.” Bear Marine, 696 F.2d at 1120. If Silver- thorne fails to establish liability, our premature answer to the question will not have affected the litigation at all. ICTSI, 22 F.4th at 1130. Any dispute about damages will have “evaporate[d] in the light of full factual develop- ment.” Benoit, 959 F.3d at 508 (cleaned up). Even assuming arguendo that Silverthorne could establish liability, the reasonable-royalty standard may still not be controlling. Silverthorne claims that the district court’s order prevents it from proving damages. If that were true, the question could have controlled the case. 16 As the district court noted, though, its order did “not automatically bar [Silverthorne] from prov- ing damages,” as long as it does so according to the standard defined in Uni- versity Computing. Any effect of this court’s reversal would be speculative and dependent

_____________________ 16 For example, the district court might have granted summary judgment against Silverthorne after concluding that it could not prove damages under University Computing. See, e.g., Vaughan, 849 F.3d at 590 (asserting § 1292(b) jurisdiction to answer dispositive damages question). If we reversed such an order, we would have “immediate[ly] im- pact[ed]” the litigation by resuming proceedings that had been incorrectly stopped. Fujitsu, 539 F. App’x at 1007.

8 Case: 24-20006 Document: 85-1 Page: 9 Date Filed: 01/03/2025

on Silverthorne’s evidence. If we adopted Silverthorne’s standard, Silver- thorne may nonetheless fail to establish a reasonable royalty, and if we af- firmed the district court’s interpretation, Silverthorne could nonetheless find a way to prove damages. The dissent makes hay out of Garner I, where we asserted § 1292(b) jurisdiction over a discovery order denying the attorney-client privilege. Resolution of that privilege in that “complex case” could have abated years of pretrial litigation and a long trial. See 430 F.2d at 1097. Conversely, the inevitable trial here “will be short, and nothing we might do is likely to abbre- viate it significantly.” Bear Marine, 696 F.2d at 1120. This appeal would not “avoid[] the ‘waste of precious judicial time while the case grinds through to a final judgment.’” Post, at 4 (quoting Hadjipateras v. Pacific, S.A., 290 F.2d 697, 702–03 & n.13 (5th Cir. 1961)). Also misplaced is the dissent’s reliance on Yamaha Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199 (1996). Post, at 8. Even if Yamaha sheds light on when a question is “controlling,” 17 the district court’s interpretation of “reasonable royalty” here would not be. The Yamaha district court had certified for appeal its summary-judgment order, which had foreclosed most theories of damages. 516 U.S. at 203–04. But the district court here ex- plained that its definition of a reasonable royalty “does not automatically bar [Silverthorne] from proving damages.” See post, at 2. Because our answer to the certified question would not “immedi- ate[ly]” or “materially affect the outcome of litigation in the district court,” the question is not yet controlling. Fujitsu, 539 F. App’x at 1007; ICTSI, _____________________ 17 Yamaha decided two issues: (1) whether a court of appeals could “exercise jur- isdiction over any question that is included within the order that contains the controlling question . . . identified by the district court,” and (2) whether federal law “suppl[ied] the exclusive remedy” in certain maritime cases. 516 U.S. at 204–05 (cleaned up).

9 Case: 24-20006 Document: 85-1 Page: 10 Date Filed: 01/03/2025

22 F.4th at 1130 (cleaned up).

B. Section 1292(b) also requires that an “immediate appeal [may] mate- rially advance the end of the litigation,” which means that this court’s decision on the interlocutory appeal could “abbreviate [the district court pro- ceedings] significantly.” Bear Marine, 696 F.2d at 1119–20. This interlocutory appeal would not do that. See id. at 1120. The par- ties will continue to trial whether this court weighs in. The trial, including any trial on damages, will probably “be short, and nothing we might do is likely to abbreviate it significantly.” Id. Indeed, the parties may not even reach the damages issue because Silverthorne has not yet proven liability. Silverthorne and the dissent aver that we could save judicial resources by shortening post-judgment appeals and proceedings. But § 1292(b) requires that an interlocutory appeal shorten the proceedings in the district court, not on post-judgment appeal. 18 Otherwise, almost any disputed legal question could qualify under § 1292(b), and the “narrow exception” allow- ing interlocutory review would “swallow the rule.” See Garner II, 433 F.2d at 120; Baldridge, 404 F.3d at 932. This interlocutory appeal would not materially advance the ultimate termination of the litigation.

C. The dissent bemoans that “the parties and this court have already _____________________ 18 Bear Marine, 696 F.2d at 1120 (concluding that the appeal would not “materially advance” litigation where “nothing we might do is likely to abbreviate [the trial] signifi- cantly”); ICTSI, 22 F.4th at 1131 (explaining that an appeal “materially advance[s]” liti- gation where “resolution of the question may appreciably shorten . . . the district court pro- ceedings” (emphasis added) (cleaned up)).

10 Case: 24-20006 Document: 85-1 Page: 11 Date Filed: 01/03/2025

expended time and resources on briefing and argument, all while the case is abated in the district court.” Post, at 9–10. But “[t]hat is no reason for th[is] court to dissipate further energies on the appeal or to decide questions that may prove to be hypothetical.” Bear Marine, 696 F.2d at 1120. By declining to decide this appeal, the dissent complains, we leave the law of reasonable royalties “a mess.” Post, at 10. Maybe so. But itching, though we may be, to reach the merits of an interesting issue, “[w]e do not sit to decide moot questions[] or to issue advisory opinions.” Bear Marine, 696 F.2d at 1120. We unwisely accepted this interlocutory appeal under § 1292(b). Because the certified question is not controlling, and the appeal would not materially advance the ultimate termination of the litigation, Silverthorne has not shown “that exceptional circumstances justify a departure from the basic policy of postponing appellate review until after the entry of a final judg- ment.” Coopers & Lybrand v. Livesay, 437 U.S. 463, 475 (1978) (cleaned up). * * * * * The order granting leave to appeal is VACATED, the appeal is DISMISSED for want of jurisdiction, and the case is REMANDED for further proceedings as appropriate. We express no opinion on the ultimate merits of any question that the district court may address.

11 Case: 24-20006 Document: 85-1 Page: 12 Date Filed: 01/03/2025

Stephen A. Higginson, Circuit Judge, dissenting: The question in this appeal was certified for our review by a veteran district judge who felt compelled to adhere to case law that is out of step with more recent statutory text, as confirmed by two other circuit courts. A unanimous panel of our colleagues agreed to revisit our law, granting permission to appeal. Today, nearly a year later, we vacate and dismiss that grant as “imprudent[],” ante at 5, and “unwise[],” ante at 11, and we return the case to the inquiring district court with no answer. This is a mistake. Because the certified issue is a question of law; because it is decisive for this litigation; because decisions from the other courts of appeals convincingly explain that our half-century-old decision in University Computing Co. v. Lykes-Youngstown Corp., 504 F.2d 518 (5th Cir. 1974), is inconsistent with statutorily expanded liability in the Defend Trade Secrets Act of 2016, Pub. L. No. 114-153, § 2(a), 130 Stat. 376, 376–80 (codified at 18 U.S.C. § 1836) (DTSA)—we can, and should, answer the district court’s question. I. The history of the proceedings shows why the case pivots on the certified question. Silverthorne Seismic, L.L.C. (Silverthorne) alleged misappropriation of its trade secrets under the DTSA and sought a reasonable royalty from Sterling Seismic Services, Ltd. (Sterling). This was the sole claim in issue at all points relevant. During motions practice, the district court found multiple triable issues of fact regarding liability under the DTSA. But the district court excluded the opinion of Silverthorne’s damages expert. Silverthorne, applying the reasonable royalty benchmark of a hypothetical negotiation between a willing licensor and licensee, then proposed to “present evidence at trial of the fair market value of a license to the . . . data in a fictitious transaction between willing seller and willing buyer,” with lay witness Barton Wilson, Silverthorne’s owner, “testify[ing]

12 Case: 24-20006 Document: 85-1 Page: 13 Date Filed: 01/03/2025

based on his knowledge of actual transactions . . . in the time period in question.” Relying on University Computing, Sterling moved to exclude the testimony, contending that it was expert testimony and irrelevant because Wilson was “opining on . . . what an oil and gas operating company . . . would be willing to pay,” rather than what Sterling, a data processing company, would have paid in a hypothetical negotiation to license the trade secret for its own use. See Univ. Computing, 504 F.2d at 539. At a pretrial conference, the parties and the district court extensively discussed settling the statutory measure for a reasonable royalty by interlocutory appeal. Counsel for Silverthorne stated that “if the Court is ruling that the measure of damages would be what Silverthorne would sell and what Sterling would pay, . . . there’s not going to be any evidence of that adduced with plaintiff’s case. . . . I think the Court’s going to be left with no alternative but to grant a directed verdict on the DTSA.” The court responded, “I’m afraid you’re right. I’m afraid there is no means of showing damages . . . .” On October 25, 2023, having already excluded testimony from Silverthorne’s damages expert, the district court denied the motion to exclude Wilson’s lay testimony. But with respect to the reasonable royalty standard, the court decided to adhere to University Computing, stating that “the proper measure of a reasonable royalty analysis is ‘what the parties would have agreed to as a fair price for licensing the defendant to put the trade secret to the use the defendant intended at the time the misappropriation took place.’” See Univ. Computing, 504 F.2d at 539. Thus, “the proper inquiry is what Silverthorne and Sterling would have agreed to for Sterling to use the alleged trade secret. It is not what Silverthorne would have agreed to for an operating company to use the alleged trade secret.” The court stated that “this holding does not automatically bar Plaintiff from proving damages.” With respect to the propriety of putting on Wilson as a

13 Case: 24-20006 Document: 85-1 Page: 14 Date Filed: 01/03/2025

lay witness, the district court stated that Wilson could testify from personal knowledge as company president but could not testify on a foundation of “general industry or business knowledge.” See Metro Hosp. Partners, Ltd. v. Lexington Ins. Co., 84 F. Supp. 3d 553, 564 (S.D. Tex. 2015). On November 15, 2023, the court certified its order for interlocutory appeal and framed the following question: “whether a plaintiff is entitled to prove reasonable royalty damages under the DTSA using willing buyer(s) detached from the parties to the litigation when willing buyers (here, oil and gas exploration companies) exist for plaintiff’s alleged trade secret (here, seismic data), but the defendant and comparable entities (here, seismic processors) do not buy or license that trade secret.” We unanimously granted leave to appeal on January 4, 2024. II. Certification of interlocutory orders under 28 U.S.C. § 1292(b) requires the district court to find that the “order involves a controlling question of law,” that “there is substantial ground for difference of opinion” upon that question, “and that an immediate appeal from the order may materially advance the ultimate termination of the litigation.” 1 The appeal proceeds only if the court of appeals receives a timely application for and allows review. The statute otherwise affords “unfettered discretion” to the court of appeals. See Microsoft Corp. v. Baker, 582 U.S. 23, 31 (2017). The majority opinion’s constricted application of § 1292(b) is out of step with that discretion and with our previous decisions. “The statute was

_____________________ 1 Like the majority, I focus on the first and third factors, but there is substantial ground for difference of opinion as well. Cf. David O. Taylor, Using Reasonable Royalties to Value Patented Technology, 49 Ga. L. Rev. 79, 134 (2014) (referring to a similar “unresolved question” concerning reasonable royalties for patent infringement).

14 Case: 24-20006 Document: 85-1 Page: 15 Date Filed: 01/03/2025

framed in broad language” and we interpret it with the recognition that its “considerable flexibility” avoids the “waste of precious judicial time while the case grinds through to a final judgment as the sole medium through which to test the correctness of some isolated identifiable point of fact, of law, of substance or procedure, upon which in a realistic way the whole case or defense will turn”—rather than according to a set of “handy modifiers” such as “‘strictly construed.’” Hadjipateras v. Pacifica, S.A., 290 F.2d 697, 702– 03 & n.13 (5th Cir. 1961) (quoting Milbert v. Bison Lab’ys, Inc., 260 F.2d 431, 435 (3d Cir. 1958)). Thus, while we duly assess the propriety of an appeal, Coopers & Lybrand v. Livesay, 437 U.S. 463, 475 (1978); Pub. Emp. Union, Local No. 1279 v. Alabama, 453 F.2d 922, 924 (5th Cir. 1972), “[w]e do not believe it does any good to echo epithets uttered by others that § 1292(b) is to be ‘sparingly applied.’” Ex parte Tokio Marine & Fire Ins. Co., 322 F.2d 113, 115 (5th Cir. 1963) (quoting Milbert, 260 F.2d at 433). 2 Instead, we take a pragmatic approach in assessing whether a district court order involves a controlling question of law the review of which would materially advance the litigation. As Judge Posner put it, a “‘question of law’ means an abstract legal issue” that “the court of appeals c[an] decide quickly and cleanly” without “hunting through the record.” Ahrenholz v. Bd. of Trs. of Univ. of Illinois, 219 F.3d 674, 677 (7th Cir. 2000). And while “mere[] fact- review questions” like sufficiency of the evidence to sustain summary judgment may often be “inappropriate for § 1292(b) review,” Clark-Dietz & Assocs.-Engineers, Inc. v. Basic Const. Co., 702 F.2d 67, 69 (5th Cir. 1983), “the meaning of a statutory or constitutional provision” is a question of law.

_____________________ 2 We do strictly enforce certain procedural requirements, see Local No. 1279, 453 F.2d at 924, where the clear congressional instructions stand in contrast to the open-ended discretion afforded to the district court judge to determine whether a question is “controlling” or whether an appeal “advances” the litigation.

15 Case: 24-20006 Document: 85-1 Page: 16 Date Filed: 01/03/2025

Ahrenholz, 219 F.3d at 676. The interpretation of a statute like the DTSA does not metamorphose into an evidence-weighing question simply because its import to the case—like all questions of law—“depend[s] on a party’s ability to prove additional facts,” ante at 5. While we might decline to hear such an appeal for prudential reasons, there is nothing jurisdictional standing in the way. See, e.g., Spong v. Fid. Nat. Prop. & Cas. Ins. Co., 787 F.3d 296, 304 (5th Cir. 2015) (stating that a question of preemption “certainly falls within the ambit of 28 U.S.C. § 1292(b)”). The design of § 1292(b) supports this usual approach. Appeals under the provision, although they are taken from certified orders, involve propositions of law, which are prospective in nature, in an interlocutory posture. The whole point is that legal principles are generally applicable, so the court of appeals does not need to “hunt[] through the record”—particularly not in order to second-guess the district court judge by scrutinizing the record to determine whether a question is really dispositive. See Coopers, 437 U.S. at 475 n.25 (quoting H.R. Rep. No. 85-1667, at 5–6 (1958), as stating that the district court’s certification provides “the best informed opinion that immediate review is of value”); United States v. Garner, 749 F.2d 281, 289 (5th Cir. 1985) (“[E]ven a cursory review of the provision’s legislative history would reveal that the drafters of § 1292(b) felt the participation of the trial judge in procedurally screening interlocutory orders to be ‘essential’ in fulfilling the provision’s dual purpose of ensuring that only appropriate cases are subject to interlocutory review and avoiding wasteful and fruitless jurisdictional determinations in the courts of appeals.”), supplemented, 752 F.2d 116 (5th Cir. 1985); In re Trump, 874 F.3d 948, 952 (6th Cir. 2017) (“In most cases, certified orders already stand out as ‘exceptional’ by virtue of another Article III judge’s opinion. District courts do not make a habit of certifying their own orders for interlocutory appeal.”).

16 Case: 24-20006 Document: 85-1 Page: 17 Date Filed: 01/03/2025

Because § 1292(b) appeals are not taken from final judgments, courts including ours have emphasized that imposing a simulacrum of the final- judgment rule on the provision makes little sense. “[I]t is never one hundred percent certain in advance that the resolution of a particular question will determine the outcome or even the future course of the litigation.” Johnson v. Burken, 930 F.2d 1202, 1205–06 (7th Cir. 1991). That is why “the issue need not . . . be dispositive to be a ‘controlling question.’” Hadjipateras, 290 F.2d at 702 n.10; accord Sokaogon Gaming Enter. Corp. v. Tushie-Montgomery Assocs., Inc., 86 F.3d 656, 659 (7th Cir. 1996); Kuehner v. Dickinson & Co., 84 F.3d 316, 318–19 (9th Cir. 1996); Katz v. Carte Blanche Corp., 496 F.2d 747, 755 (3d Cir. 1974). The contingency that worries the majority can be spotted in even our most garden-variety § 1292(b) decisions. We have repeatedly entertained appeals from summary judgment orders involving certified questions such as “whether a plaintiff must prove an adverse effect on competition in order to prevail under 7 U.S.C. §§ 192(a)–(b),” Wheeler v. Pilgrim’s Pride Corp., 591 F.3d 355, 357 (5th Cir. 2009) (en banc), or whether workers were “entitled to be paid the minimum wage,” Castellanos-Contreras v. Decatur Hotels, LLC, 622 F.3d 393, 398 (5th Cir. 2010) (en banc). If the court of appeals disagrees with the district court on a legal question of this variety—which, after all, must be susceptible of a difference in opinion to be eligible for § 1292(b) review—the district court may not yet have determined how the new rule applies to the evidence in the case. It makes no difference whether such rules go to liability or to damages. From a reviewing court’s perspective they are equally “hypothe[tical].” See ante at 5. Another way to view this, of course, is that as soon as a legal proposition has been relied upon in the order from which appeal is taken, its application to the case is no longer hypothetical at all.

17 Case: 24-20006 Document: 85-1 Page: 18 Date Filed: 01/03/2025

I am accordingly doubtful that the principles the majority extracts from a law review note and an unpublished decision from another circuit 3 can be reconciled with the full sweep of our case law. We have on multiple occasions resolved discovery privilege disputes under § 1292(b). See Hyde Constr. Co. v. Koehring Co., 455 F.2d 337, 338–39 (5th Cir. 1972); Garner v. Wolfinbarger, 430 F.2d 1093, 1096–97 (5th Cir. 1970); see also Mohawk Indus., Inc. v. Carpenter, 558 U.S. 100, 110–11 (2009) (stating that “district courts should not hesitate to certify an interlocutory appeal” of “a privilege ruling involv[ing] a new legal question”). Perhaps it speaks for itself that the majority opinion distinguishes these cases by stating that the issues were “controlling” for discovery. See ante at 6. At any rate, that is not how we have viewed the situation. We have flatly stated that “[r]eview under § 1292(b) is available where decision on an issue would affect the scope of the evidence in a complex case, even short of requiring complete dismissal.” Wolfinbarger, 430 F.2d at 1097. In line with our rule (which bears repeating) that “the issue need not . . . be dispositive to be a ‘controlling question,’” Hadjipateras, 290 F.2d at 702 n.10, we have previously considered it sufficient that “[t]he availability or unavailability of the testimony and documents sought may affect the disposition of the pending motions to dismiss.” Wolfinbarger, 430 F.2d at 1097 (emphasis added). The question here not only “affect[s] the scope of the evidence” but has a much more direct and immediate effect on the litigation than potential limits on the exchange of documents among the litigants’ lawyers years before trial if it ever occurs. It is undisputed that the district court’s standard _____________________ 3 A decision, moreover, which is not even clearly the law of that circuit. The Federal Circuit considered this issue at length in the context of a Rule 54(b) appeal from the grant of a motion in limine to exclude damages evidence and concluded that § 1292(b) would have been the proper avenue for an appeal. See Ultra-Precision Mfg. Ltd. v. Ford Motor Co., 338 F.3d 1353, 1359 (Fed. Cir. 2003).

18 Case: 24-20006 Document: 85-1 Page: 19 Date Filed: 01/03/2025

for reasonable royalty damages forecloses the principal theory upon which Silverthorne’s witness has foundation to opine. And even if our discovery privilege cases approach the outer bounds of § 1292(b), we are nowhere near those jurisdictional hinterlands. The certified question here is on all fours with the keystone Supreme Court precedent on § 1292(b) review. Yamaha Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199 (1996), involved a summary judgment order deciding that federal maritime law provided the cause of action in the case, so that federal law preempted certain avenues of recovery under state law, id. at 203. The certified questions asked whether the plaintiffs might recover damages for loss of society, damages for future earnings, or punitive damages. Id. at 204. The Supreme Court held that damages were “properly governed by state law,” id. at 216—and did not address which state’s law this might be, or “the standards governing liability,” leaving such questions for the district court on remand, id. at n.14. So liability was far less established than in this case, and the Supreme Court’s decision, which left open the choice of law governing both liability and remedies, was far less dispositive. Yamaha is not an exception either. As the majority notes, ante at 7, this court reviewed the dismissal of pain and suffering and punitive damages for claims under an antidiscrimination statute. Vaughan v. Anderson Reg’l Med. Ctr., 849 F.3d 588, 590 (5th Cir. 2017). This court similarly considered the scope of various damages measures under the antitrust laws even though “on both the issue of liability and damages, there are disputed issues of material fact which make summary judgment improper.” Pollock & Riley, Inc. v. Pearl Brewing Co., 498 F.2d 1240, 1244–46 (5th Cir. 1974) (footnote omitted). In a consolidated case, the Pollock & Riley court further considered whether jury instructions had to mention treble damages in civil antitrust cases. Id. at 1242. And this court sitting en banc reviewed the certified question of “whether the [Fair Labor Standards Act, 29 U.S.C. §§ 201–219

19 Case: 24-20006 Document: 85-1 Page: 20 Date Filed: 01/03/2025

(FLSA)] generally applies to the Workers” and reached “the question of whether . . . travel, visa, and recruitment expenses” were subject to reimbursement, Castellanos-Contreras, 622 F.3d at 398–99, even though violation of the FLSA had not yet been established, id. at 423 (Dennis, J., dissenting). Although the damages measure was not certified, the court explained that it was “a legal question that can properly be the subject of interlocutory review.” Id. at 399 (majority opinion). As for our discretion, it weighs in favor of finishing what we started. The Castellanos-Contreras court noted that although “whether we should address these questions at this stage . . . is a question about which reasonable jurists can—and, in the case of this court, do—debate,” id.: A motions panel of this court permitted Decatur to pursue this appeal, and the original panel exercised its discretion to hear the appeal. Others on our court might have had a different take had they been on either panel. But we are no longer at the beginning of this case; instead, we are very far along. Considerable time has passed, two panel opinions have issued, and the parties have briefed the merits three times: to the original panel, in connection with the rehearing petitions, and in merits briefing to the en banc court. Additionally, this case has been the subject of two oral arguments. After so much time and effort has been expended by both the parties and the court as a whole, the discretionary decision now becomes much different, and the majority of the court agrees it should be resolved in favor of hearing the merits. Id. at 399–400. As in Castellanos-Contreras, our discretion has already been exercised to hear this appeal. I think those points of the majority opinion’s which are well taken sound at most in the further exercise of our discretion to finish what we started and render a decision. But when a motions panel has already granted the appeal, when the parties and this court have already expended

20 Case: 24-20006 Document: 85-1 Page: 21 Date Filed: 01/03/2025

time and resources on briefing and argument, all while the case is abated in the district court, “the discretionary decision now becomes much different.” Id. at 399. Declining to decide this appeal now amounts only to “acquiescing in a useless trial and later appeal” of the same issue with the same arguments. Cf. Ex parte Tokio Marine, 322 F.2d at 115. In the meantime, the “uncertain legal environment” will ensure that “[d]amage awards, rationales, and percentages are widely disparate”—in other words, that “[r]easonable royalty damage awards are a mess.” Daralyn J. Durie & Mark A. Lemley, A Structured Approach to Calculating Reasonable Royalties, 14 Lewis & Clark L. Rev. 627, 644 (2010). Rather than assenting to that state of affairs, in my view, we should answer the certified question. III. The question certified by the district court, and previously accepted by us, involves a provision in the DTSA 4 which permits royalty awards in compensation for mere disclosure of trade secrets, regardless of the traditional rule in University Computing that the basis for a royalty award must be the defendant’s commercial use. The district court, however, followed University Computing and formulated the standard for a reasonable royalty award as “what the parties would have agreed to as a fair price for licensing the defendant to put the trade secret to the use the defendant intended at the time the misappropriation took place.” See Univ. Computing, 504 F.2d at 539. The district court thus certified the question of whether the legal test for a hypothetical negotiation as between a “willing buyer” and a “willing seller” was “case-specific,” or “whether a plaintiff is entitled to prove

_____________________ 4 The DTSA does not preempt state-law remedies, which are not at issue. See 18 U.S.C. § 1838.

21 Case: 24-20006 Document: 85-1 Page: 22 Date Filed: 01/03/2025

reasonable royalty damages under the DTSA using” third party “willing buyer(s).” 5 We would greatly assist the district court and speed the resolution of this litigation if we were to answer the legal question as other circuit courts have, and must, by “start[ing] with the statutory text.” Rest. L. Ctr. v. U.S. Dep’t of Labor, 120 F.4th 163, 171 (5th Cir. 2024) (quoting Tanzin v. Tanvir, 592 U.S. 43, 46 (2020)). When we interpret a statute like the DTSA, plain language is enforced “according to its terms.” King v. Burwell, 576 U.S. 473, 486 (2015). “[W]hether the language is plain” depends on context and statutory scheme. Id. If a statute requires interpretation, this court’s “task is to resolve that ambiguity.” Life Techs. Corp. v. Promega Corp., 580 U.S. 140, 141 (2017). “Congress ‘legislate[s] against a background of common- law adjudicatory principles,’ and it ‘expect[s]’ those principles to ‘apply except when a statutory purpose to the contrary is evident.’” Minerva Surgical, Inc. v. Hologic, Inc., 594 U.S. 559, 572 (2021) (quoting Astoria Fed. Sav. & Loan Assn. v. Solimino, 501 U.S. 104, 108 (1991)). In conducting this analysis, the court may “address any issue fairly included within the certified order.” Yamaha, 516 U.S. at 205. I would answer the district court’s question in two parts. First, I would explain why the district court’s failure to allow for misappropriation by disclosure in its reasonable royalty standard was out of step with the plain text of the DTSA. Second, with the benefit of this analysis, I would explain why third parties may provide useful points of reference to arrive at a royalty remedy for such disclosure.

_____________________ 5 The district court noted that “[t]he parties . . . disagree over who the appropriate ‘willing buyer’ is in this inquiry: Defendant argues for a case-specific buyer (here, a seismic processing company such as Defendant), while Plaintiff argues for a willing buyer desiring to use the trade secret (here, an oil and gas exploration company).”

22 Case: 24-20006 Document: 85-1 Page: 23 Date Filed: 01/03/2025

A. The district court assumed based on University Computing that only a defendant’s use was compensable by a reasonable royalty. However, by its plain text, the DTSA authorizes reasonable royalty damages not just for commercial use of misappropriated trade secrets but also for disclosure of them, as is alleged here by Silverthorne against Sterling. The statute allows courts to award “the damages caused by the misappropriation measured by imposition of liability for a reasonable royalty for the misappropriator’s unauthorized disclosure or use of the trade secret.” 18 U.S.C. § 1836(b)(3)(B)(ii). The district court felt compelled to start not with the DTSA but with the test in University Computing, which applied Georgia’s common law of misappropriation at the time: In calculating what a fair licensing price would have been had the parties agreed, the trier of fact should consider such factors as the resulting and foreseeable changes in the parties’ competitive posture; the prices past purchasers or licensees may have paid; the total value of the secret to the plaintiff, including the plaintiff’s development costs and the importance of the secret to the plaintiff’s business; the nature and extent of the use the defendant intended for the secret; and finally whatever other unique factors in the particular case which might have affected the parties’ agreement, such as the ready availability of alternative processes. 504 F.2d at 539. But this common law has since been succeeded by codification, specifically first in the Uniform Trade Secrets Act (UTSA), as confirmed by decisions interpreting states’ versions of that statute. A notable decision recognizing this change was the Tenth Circuit’s in StorageCraft Tech. Corp. v. Kirby, 744 F.3d 1183 (10th Cir. 2014). StorageCraft’s former director

23 Case: 24-20006 Document: 85-1 Page: 24 Date Filed: 01/03/2025

shared its source code with a competitor, but claimed that he never used it for profit and that there was no evidence the competitor did either. Id. at 1185. That, said then-Judge Gorsuch, was of little help under Utah’s UTSA, which made reasonable royalties available “for a misappropriator’s unauthorized disclosure or use of a trade secret.” Id. at 1186 (quoting Utah Code Ann. § 13-24-4(1)) (emphasis added in original). This clear statutory language led the Tenth Circuit to reject the common-law rule from University Computing that “a misappropriation defendant generally had to ‘put the trade secret to some commercial use’ before a reasonable royalty award was allowed,” id. at 1187 (quoting Univ. Computing, 504 F.2d at 539), although the court noted that the defendant’s uses might be informative about “the scope of the license the defendant assumed for himself,” id. at 1189. The court explicitly declined to limit consideration to either the University Computing factors or the patent law factors from Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116, 1120–21 (S.D.N.Y. 1970), modified sub nom. Georgia-Pac. Corp. v. U.S. Plywood-Champion Papers, Inc., 446 F.2d 295 (2d Cir. 1971). Id. at 1189. Similarly, the Fourth Circuit instructively interpreted parallel language in Maryland’s UTSA in AirFacts, Inc. v. Amezaga, 30 F.4th 359 (4th Cir. 2022). A former AirFacts employee had sent internal company documents to another company in the same general field (but not a competitor) as part of a job application after resigning. Id. at 362. The court, applying the statutory text, again rejected the idea of “condition[ing] such awards on a defendant putting a trade secret to commercial use,” stating that “University Computing only considered a common law claim for misappropriating a trade secret. It didn’t interpret a statute, as we do here.” Id. at 368. In 2016, after StorageCraft and before AirFacts, and over forty years after University Computing, the DTSA was enacted. As noted, the plain

24 Case: 24-20006 Document: 85-1 Page: 25 Date Filed: 01/03/2025

language of the statute permits courts in civil actions for misappropriation of trade secrets to award “the damages caused by the misappropriation measured by imposition of liability for a reasonable royalty for the misappropriator’s unauthorized disclosure or use of the trade secret.” 18 U.S.C. § 1836(b)(3)(B)(ii) (emphasis added). The House report noted that “Subparagraph (B)” is “drawn directly from § 3 of the UTSA.” H.R. Rep. No. 114-529, at 13 (2016), reprinted in 2016 U.S.C.C.A.N. 195, 207. 6 The report referred to decisions of “courts interpreting the UTSA’s analogous provision.” Id. at n.13. Accordingly, as explicated by our two sister courts in nearly identical contexts, the DTSA by its plain text authorizes reasonable royalty damages not just for use of misappropriated trade secrets but also for disclosure of them, as alleged here. Indeed, other provisions of the statute confirm this expansion of liability. Section 1836(b)(3) applies to “civil action[s] brought under this subsection with respect to the misappropriation of a trade secret,” thereby incorporating 18 U.S.C. § 1839(5)(B), which defines “misappropriation” to include some disclosures. Moreover, as a matter of statutory scheme § 1836(b)(3)(B)(ii) applies “in lieu of” other remedies— which are relatively likely to be inadequate in instances of disclosure, since lost profits and unjust enrichment may reflect the defendant’s commercial use. See StorageCraft, 744 F.3d at 1186. The DTSA provides this remedy at the district court’s discretion. See 18 U.S.C. § 1836(b)(3). In recognizing it as a potential option, then-Judge Gorsuch cautioned that he “d[id]n’t mean to suggest that the reasonable

_____________________ 6 See, e.g., Utah Code Ann. § 13-24-4(1) (“In lieu of damages measured by any other methods, the damages caused by misappropriation may be measured by imposition of liability for a reasonable royalty for a misappropriator’s unauthorized disclosure or use of a trade secret.”); Md. Code Ann., Com. Law § 11-1203(c) (same).

25 Case: 24-20006 Document: 85-1 Page: 26 Date Filed: 01/03/2025

royalty measure of damages is always the most sensible remedy.” StorageCraft, 744 F.3d at 1187. And the legislative history of the DTSA disavows “intent to encourage the use of reasonable royalties.” H.R. Rep. No. 114-529, at 13. Congress would have expected courts to award a reasonable royalty for disclosure in cases where, as with any type of damages award, the court has determined that the remedy would provide fair compensation for the wrong. The logic of the reasonable royalty is that when someone puts the intellectual property of another to work, a royalty can be assessed as if on the fruits of the endeavor. See Tex. Advanced Optoelectronic Sols., Inc. v. Renesas Elecs. Am., Inc., 895 F.3d 1304, 1324 (Fed. Cir. 2018). But as discussed, in a suit for misappropriation by disclosure, the right to use is not necessarily taken at all. Instead, what is taken is confidentiality—for unlike patents, which are public, a trade secret holder’s interest generally depends on keeping the information secret. See Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 480–84 (1974). Sufficiently pressing, such incongruities may require a different measure of damages or may allow the district court to award none at all. See 18 U.S.C. § 1836(b)(3). On remand in AirFacts, for example, the district court found that AirFacts had failed to prove that “there [wa]s any fair licensing price” for the trade secrets disclosed in the defendant employee’s job application. AirFacts, Inc. v. De Amezaga, No. CV DKC 15- 1489, 2022 WL 17584258, at *7 (D. Md. Dec. 12, 2022) (emphasis added). The district court’s application of University Computing foreclosed this option from the outset. Rather than inquiring into the value of a license for “the use the defendant intended,” the district court should have framed the hypothetical negotiation as directed to a license that would have covered the alleged disclosures. At the threshold, the court should have focused not only on whether Sterling intended to use the trade secret in commerce, but on the more general question of whether the purpose could be attributed to

26 Case: 24-20006 Document: 85-1 Page: 27 Date Filed: 01/03/2025

Sterling of attempting to convert either the content or the confidentiality of Silverthorne’s trade secrets into its own profit upon which a prospective royalty could be reasonable to assess. B. With this statutory framework in hand, I would turn to the specifics of the hypothetical negotiation. The district court has asked for our advice on how to apply the construct of a hypothetical negotiation between willing buyer and seller in order to arrive at a reasonable royalty. The district court was correct that the reasonable royalty is “case-specific” insofar as any damages remedy is specific to the case at hand—the scope of the misappropriation helps define the license over which the parties are hypothesized to be negotiating. But the district court applied this principle too rigidly. A hypothetical negotiation is only one way to measure a reasonable royalty. And when it is used, it is hypothetical. The question is how the parties would have acted, had they negotiated reasonably toward a license—not how they did act. The district court thus erred in concluding that a comparison to third-party benchmarks could never inform that assessment. A reasonable royalty award is compensatory. The DTSA states that reasonable royalty awards should “measure[]” “the damages caused by the misappropriation.” 18 U.S.C. § 1836(b)(3)(B)(ii). Thus “[t]he ‘value of what was taken’ . . . measures the royalty.” Aqua Shield v. Inter Pool Cover Team, 774 F.3d 766, 770 (Fed. Cir. 2014) (quoting Dowagiac Mfg. Co. v. Minn. Moline Plow Co., 235 U.S. 641, 648 (1915)). And what was taken by the defendant “undoubtedly depends on what he did with the secret.” StorageCraft, 744 F.3d at 1189. As explained, under the DTSA, a defendant may take an interest in keeping a trade secret confidential by disclosing it. The idea of a hypothetical negotiation in a disclosure case is to value this

27 Case: 24-20006 Document: 85-1 Page: 28 Date Filed: 01/03/2025

misappropriation by evaluating what price the parties would have agreed to for a license allowing the defendant to make the actionable disclosures. See Univ. Computing, 504 F.2d at 537. Thus, the remedy is case-specific in the same way that any compensatory remedy is. But because the reasonable royalty reflects “the actual value of what has been appropriated,” Univ. Computing, 504 F.2d at 537 (quoting Vitro Corp. of Am. v. Hall Chem. Co., 292 F.2d 678, 683 (6th Cir. 1961)); cf. Aqua Shield, 774 F.3d at 770, “every case requires a flexible and imaginative approach to the problem of damages,” Univ. Computing, 504 F.2d at 538; see also StorageCraft, 744 F.3d at 1189; cf. Georgia-Pacific, 318 F. Supp. at 1120– 21. That is particularly so in a case like this, with a cause of action established in 2016. A meticulous focus on the details of the hypothetical negotiation should not be allowed to curtail the requisite flexibility or to draw attention from the primary inquiry of what price, if any, makes for a fair remedy. See, e.g., StorageCraft, 744 F.3d at 1189. As an initial matter, this means that the district court erred in positioning the hypothetical negotiation as the sole measure of a reasonable royalty that it was permitted to consider. A court’s discretion to select a “methodology for arriving at a reasonable royalty” in a particular case is not so confined. See SmithKline Diagnostics, Inc. v. Helena Lab’ys Corp., 926 F.2d 1161, 1164 (Fed. Cir. 1991). Besides analytical methods, see, e.g., Univ. Computing, 504 F.2d at 538; cf., e.g., TWM Mfg. Co., Inc. v. Dura Corp., 789 F.2d 895, 899 (Fed. Cir. 1986); Hanson v. Alpine Valley Ski Area, Inc., 718 F.2d 1075, 1080–81 (Fed. Cir. 1983), “in some cases the damages will be subject to exact measurement, either because the parties had previously agreed on a licensing price . . . or because some industry standard provides a clear measure,” Univ. Computing, 504 F.2d at 538–39; cf. Hanson, 718 F.2d at 1078. Such an established royalty almost definitionally involves the prices paid by third parties.

28 Case: 24-20006 Document: 85-1 Page: 29 Date Filed: 01/03/2025

And when the district court does select a hypothetical negotiation to value the royalty, the method still cannot be too specific to the parties, because it requires assuming that the parties behaved precisely as they did not. Instead of keeping its trade secret close, the plaintiff licenses the secret out; instead of misappropriating, the defendant obtains a license. Of necessity, therefore, the exercise abstracts from the actual parties in order to posit a sequence of events that did not occur. As Chief Judge Markey (then of the Court of Customs and Patent Appeals) explained in the patent context: Determination of a “reasonable royalty” after infringement, like many devices in the law, rests on a legal fiction. Created in an effort to “compensate” when profits are not provable, the “reasonable royalty” device conjures a “willing” licensor and licensee, who like Ghosts of Christmas Past, are dimly seen as “negotiating” a “license.” There is, of course, no actual willingness on either side, and no license to do anything . . . . Panduit Corp. v. Stahlin Bros. Fibre Works, 575 F.2d 1152, 1159 (6th Cir. 1978). It may thus often be “inaccurate, and even absurd” to assume reasonably conducted “negotiations between the plaintiff and defendant.” Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538, 1554 & n.13 (Fed. Cir. 1995) (en banc). And that assumption seems particularly tenuous when the defendant is alleged to have committed the misappropriation knowingly, see 18 U.S.C. § 1839(5), and the supposed license threatens the integrity of the trade secret. Again I would emphasize that the district court’s choice of remedy (and if a royalty is selected, methodology) provides a safety valve that can relieve some of the pressure from these tensions, but they are otherwise inherent in a counterfactual analysis and cannot be assumed away by saying that the remedy is “case-specific.” Because the reasonable royalty relates to the parties in this counterfactual sense, the district court should not have framed the

29 Case: 24-20006 Document: 85-1 Page: 30 Date Filed: 01/03/2025

hypothetical negotiation in the “case-specific” way that it did. The willing negotiators are counterfactual constructions of the parties negotiating reasonably “in the marketplace” toward a reasonable price. See LaserDynamics, Inc. v. Quanta Comput., Inc., 694 F.3d 51, 76 (Fed. Cir. 2012). It may be crucial to extrapolate from third parties what such reasonable people might do, and what a reasonable price would be. See Univ. Computing, 504 F.2d at 539 (referring to “prices past purchasers or licensees may have paid”). 7 The whole notion of a suit for misappropriation by disclosure under the DTSA is that the defendant has taken from the plaintiff an interest in keeping a trade secret confidential as against certain third parties. When the parties are presumed to negotiate over a fair price for a license to such an interest, a district court may well be justified in considering those third parties to be natural points of reference. Of course, comparable licenses must be actually comparable. District courts must take care when assessing evidence and methodology on the reasonable royalty question, lest confusion amid “the individual and collective flexibility of the factors . . . impede[] the hypothetical negotiation from leading to an objectively fair and reasonable outcome.” See John C. Jarosz & Michael J. Chapman, The Hypothetical Negotiation and Reasonable

_____________________ 7 To be sure, these fictitious negotiators are “the parties in the marketplace” at the relevant time. LaserDynamics, 694 F.3d at 76 (emphasis added); see also Nathaniel C. Love, Comment, Nominal Reasonable Royalties for Patent Infringement, 75 U. Chi. L. Rev. 1749, 1756 & n.42 (2008) (noting Georgia-Pacific factors keyed to the parties). Thus the parties’ individual circumstances may provide constraints on the negotiation. For example, patent cases have considered it “implicit” “that a reasonable royalty would leave an infringer with a reasonable profit,” Hanson, 718 F.2d at 1081 (quoting Square Liner 360°, Inc. v. Chisum, 691 F.2d 362, 377 (8th Cir. 1982)), as assessed from the standpoint of the hypothetical negotiation, see Powell v. Home Depot U.S.A., Inc., 663 F.3d 1221, 1238–39 (Fed. Cir. 2011). But see, e.g., Monsanto Co. v. Ralph, 382 F.3d 1374, 1384 (Fed. Cir. 2004). But this case- specific reference point does not change the counterfactual nature of the hypothetical negotiation.

30 Case: 24-20006 Document: 85-1 Page: 31 Date Filed: 01/03/2025

Royalty Damages: The Tail Wagging the Dog, 16 Stan. Tech. L. Rev. 769, 807 (2013). “[T]here must be a basis in fact to associate the royalty rates used in prior licenses to the particular hypothetical negotiation at issue in the case.” Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292, 1317 (Fed. Cir. 2011); see also ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 869 (Fed. Cir. 2010). But the mere fact that licenses were granted to third parties does not by itself make those licenses unacceptably “different from the hypothetical agreement under consideration.” Uniloc, 632 F.3d at 1316 (quoting Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1327 (Fed. Cir. 2009)). Because a reasonable royalty may compensate for mere disclosure under the DTSA, as similarly explicated by the Tenth and Fourth Circuits, and the standard for such a royalty does not categorically rule out proof by means of comparable licenses, I respectfully dissent and would vacate the district court’s order and remand for reconsideration.