Case Information

United States Court of Appeals for the Fifth Circuit United States Court of Appeals Fifth Circuit FILED August 31, 2021 No. 19-40566 Lyle W. Cayce Consolidated with Clerk HTC Corporation; HTC America, Incorporated,

Plaintiffs—Appellants , versus

Telefonaktiebolaget LM Ericsson; Ericsson, Incorporated,

Defendants—Appellees . Appeal from the United States District Court for the Eastern District of Texas USDC No. 6:18-CV-243 Before Jones, Elrod, and Higginson, Circuit Judges .

Jennifer Walker Elrod, Circuit Judge :

After negotiations for a patent licensing agreement soured, the plaintiffs sued the defendants and alleged that they had breached their contractual obligation to offer a license on fair, reasonable, and non- discriminatory terms. The jury found in the defendants’ favor, and the district court entered a separate declaratory judgment holding that the defendants had affirmatively complied with their contractual obligations. Now, the plaintiffs appeal, challenging the district court’s declaratory judgment in the defendants’ favor (which relies on the facts found by the jury) and certain of the district court’s jury instructions and evidentiary rulings. Finding no error in any of the district court’s decisions, we AFFIRM .

I. With the small cell phones we carry in our pockets, instant access to communication, information, and entertainment are never more than a few simple taps away. Devices communicate seamlessly with one another to bring us the data we need in seconds, regardless of which company developed those devices or what network they are on.

To make all of this possible, industry standard setting organizations have settled on technical standards that permit interoperability between devices. These standards dictate how cell phones and cellular towers communicate with one another, and they promote fast, reliable, and secure connections between devices. This is why a person with an iPhone on AT&T’s network can call or text someone with a Samsung phone on Verizon’s network without any trouble. Mobile device manufacturers and network providers all use the same set of standards when designing their devices and networks to ensure compatibility between various devices and networks.

Standards may be generally beneficial for both consumers and industry participants. See generally Mark A. Lemley, Intellectual Property Rights and Standard-Setting Organizations , 90 Calif. L. Rev. 1889, 1896–98 (2002). However, industry standards also pose a risk of anticompetitive behavior. See, e.g. , Ericsson, Inc. v. D-Link Sys., Inc. , 773 F.3d 1201, 1209 (Fed. Cir. 2014) (explaining two “potential concerns” with standardization). Standards frequently incorporate patented technology, known as standard- essential patents. See id. When a patent becomes essential to an industry standard, other companies face a choice if they want to have a standard- compliant device: They can either attempt to license the patented technology in order to develop their own product, or they can risk a patent infringement action against them. This puts the holder of a standard-essential patent in a powerful position to extract as much money as they want from potential licensees.

To combat the potential for anticompetitive behavior, standard setting organizations require standard-essential patent holders to commit to licensing their patents on fair, reasonable, and non-discriminatory terms (FRAND terms). See id. Companies seeking to license under these terms become third-party beneficiaries of the contract between the standard- essential patent holder and the standard setting organization. They are thus enabled to enforce the terms of that contract. See Microsoft Corp. v. Motorola, Inc. , 696 F.3d 872, 885 (9th Cir. 2012).

This case involves such an attempt to license a standard-essential patent. The European Telecommunications Standards Institute (ETSI) is a preeminent standard setting organization in the mobile telecommunications field. The France-based organization established many of the global standards for 3G, 4G, and 5G cellular communications technology. ETSI asks its members that own standard-essential patents to provide “an irrevocable undertaking in writing that [they are] prepared to grant irrevocable licences on fair, reasonable and non-discriminatory” terms. If a patent owner is not prepared to grant such licenses, ETSI may exclude the patented technology from the standard.

Telefonaktiebolaget LM Ericsson and Ericsson, Inc. (collectively Ericsson) hold numerous patents that are considered essential to the standards for 2G, 3G, and 4G technology. Ericsson has agreed with ETSI to grant licenses to other companies to use its standard-essential patents on FRAND terms. HTC Corp. and HTC America, Inc. (collectively HTC) produce mobile devices that implement the 2G, 3G, and 4G standards. HTC also owns several standard-essential patents related to 2G, 3G, and 4G technology. But to manufacture standard-compliant mobile devices, HTC has to obtain a license to use Ericsson’s patents.

Ericsson and HTC have previously entered into three cross-license agreements for their respective patents—one in 2003, one in 2008, and one in 2014. In the 2014 licensing agreement, HTC agreed to pay Ericsson a lump sum of $75 million to license Ericsson’s patents regardless of HTC’s phone sales. Based on HTC’s sales, the company paid Ericsson approximately $2.50 per mobile device.

In 2016, a short time before the 2014 license was about to expire, HTC and Ericsson began negotiating to renew the license. Ericsson offered a rate of $2.50 per 4G device. HTC did not accept this offer. Instead, it independently examined the value of Ericsson’s patents and reasoned that the royalty rate should be based on the smallest salable patent-practicing unit rather than the net sales price of the entire end-user device. HTC concluded that the relevant unit was the baseband processor, the hardware component of a cellular device that allows it to connect to cellular networks. Based on this conclusion, HTC met with Ericsson in late March 2017 and counteroffered a rate of $0.10 per 4G device.

Ericsson considered HTC’s offer to be “so far off of the norm” that it “chilled” the negotiations. Ericsson left the meeting intending to justify its initial $2.50 per 4G device offer. A few days after that meeting, on April 6, 2017, HTC filed the present lawsuit, alleging that Ericsson had breached its commitment to provide a license on FRAND terms. In June, while the lawsuit was pending, Ericsson proposed a new licensing agreement whereby HTC would pay 1% of each 4G device sold, subject to a $1 floor and a $4 cap. HTC did not accept this second offer either. Instead, it filed an amended complaint seeking declaratory relief that Ericsson had breached its contractual obligations to ETSI by insisting on non-FRAND terms and had failed to negotiate in good faith. ^[1] Ericsson counterclaimed for a declaration that its offers were FRAND and alleged that HTC had failed to negotiate in good faith.

At trial, the parties debated how to determine the value of Ericsson’s patents, which would affect whether Ericsson had offered FRAND terms to HTC. For its part, HTC sought to prove that Ericsson had not offered FRAND terms. It advocated apportioning the value of Ericsson’s patents using the profit margin on the baseband processor, rather than the net sales prices of the finished mobile device, as the royalty base. HTC also pointed to several larger mobile device manufacturers, like Apple and Huawei, who had received significantly more favorable licensing agreements than the one offered to HTC.

Ericsson argued that it had offered FRAND terms to HTC and negotiated in good faith. It contended that the best way to determine whether proposed terms were fair and reasonable was to look at licenses that Ericsson had negotiated with other, similarly situated companies. All of those licenses were based on the value that Ericsson’s product provided to the end-user product, not just the smallest salable unit. Ericsson pointed to its licenses with several of HTC’s competitors, each of which contained terms similar to those offered to HTC.

Ericsson and HTC both submitted proposed jury instructions. HTC proposed highly detailed instructions to help the jury interpret FRAND, but Ericsson objected to most of these instructions and proposed a more general FRAND instruction. The district court considered the two proposals, but ultimately gave the following instruction to the jury: “Whether or not a license is FRAND will depend upon the totality of the particular facts and circumstances existing during the negotiations and leading up to the license. Ladies and gentlemen, there is no fixed or required methodology for setting or calculating the terms of a FRAND license rate.”

Before issuing this charge to the jury, the district court held a formal charge conference with both HTC and Ericsson where it invited “any possible objections” to its proposed language. HTC made no objection to this instruction. Instead, at the end of the charge conference, HTC requested that eleven of its proposed jury instructions also be included in the final jury instructions. The district court overruled “[t]hose offerings and any objection relating for their non-inclusion in the final jury instructions.”

The jury was tasked with answering three questions: (1) “whether HTC prove[d] by a preponderance of the evidence that Ericsson breached its contractual obligation to offer HTC a license, on FRAND terms, to Ericsson’s cellular standard-essential patents”; (2) whether Ericsson had breached its duty of good faith negotiations; and (3) whether HTC had breached its duty to negotiate in good faith.

After deliberation, the jury returned a verdict finding that HTC had not proven that Ericsson had breached its FRAND duties and that both parties had breached their obligations to negotiate in good faith. HTC did not make a post-verdict Rule 50(b) motion. Ericsson moved for a declaratory judgment on its counterclaim. The district court granted Ericsson’s request and declared “that, in its dealings with HTC as presented in this case, Ericsson complied with its FRAND assurance to HTC, as set forth in” its licensing declarations with ETSI. HTC now appeals: (1) the district court’s exclusion of three of HTC’s requested jury instructions; (2) the district court’s declaratory judgment that Ericsson had complied with its obligation to provide HTC a license on FRAND terms; and (3) the district court’s exclusion of certain expert testimonial evidence as hearsay.

II. We first consider whether the district court erred in not giving HTC’s proposed jury instructions. Although HTC initially asked the district court to include eleven instructions at the end of the charge conference, HTC has appealed the exclusion of only three of those eleven instructions. HTC challenges the district court’s refusal to give two instructions relating to apportioning the value of Ericsson’s patents and one instruction regarding the non-discrimination requirement of FRAND.

We typically review a challenge to the district court’s jury instructions “under an abuse of discretion standard, affording the trial court substantial latitude in describing the law to the jurors.” Jimenez v. Wood Cnty. , 660 F.3d 841, 845 (5th Cir. 2011) ( en banc ) (quoting United States v. Santos , 589 F.3d 759, 764 (5th Cir. 2009)). Ericsson contends that HTC failed to properly object in the district court, and thus our review should be limited to plain error. See id. ; Fed. R. Civ. P. 51(d)(2) (“A court may consider a plain error in the instructions that has not been preserved as required by Rule 51(d)(1) if the error affects substantial rights.”). However, we do not decide this issue because we conclude that, even under the abuse-of-discretion standard, the district court did not commit reversible error by excluding HTC’s proposed instructions. ^[2] When reviewing for abuse of discretion, we will reverse a district court’s refusal to give a requested jury instruction “only if the instruction (1) was a substantially correct statement of law, (2) was not substantially covered in the charge as a whole, and (3) concerned an important point in the trial such that the failure to instruct the jury on the issue seriously impaired the [party’s] ability to present a given [claim].” Kanida v. Gulf Coast Med. Pers. LP , 363 F.3d 568, 578 (5th Cir. 2004) (quoting United States v. McClatchy , 249 F.3d 348, 356 (5th Cir. 2001)).

HTC contends that, rather than being instructed that “[w]hether or not a license is FRAND will depend upon the totality of the particular facts and circumstances” and that “there is no fixed or required methodology for setting or calculating the terms of a FRAND license,” the jury should have been told to apportion the value of the patents from the non-patented features of HTC’s phones. Moreover, according to HTC, the jury should have been told that the non-discrimination requirement of FRAND “serves to level the playing field among competitors” by requiring a patent holder to provide similar terms to similarly situated licensees.

HTC has failed to demonstrate that the district court committed reversible error in refusing to give HTC’s proposed instructions. To begin with, HTC’s requested instructions were not “substantially correct” statements of the law. Id. Proposed jury instructions that rely on inapplicable law cannot be “substantially correct.” See United States v. Spalding , 894 F.3d 173, 188 (5th Cir. 2018) (holding that instructions that relied on a case that “was inapposite” did not need to be given); United States v. Webster , 162 F.3d 308, 322–23 (5th Cir. 1998) (“Because the factor has nothing to do with Webster’s conduct or intent, his proffered general instruction would have incorrectly stated the law.”).

Here, all of HTC’s proposed instructions miss the mark because they rely on inapplicable law. Ericsson’s FRAND commitment is part of an agreement that it made with ETSI. Accordingly, the district court held, and HTC has not challenged on appeal, that Ericsson’s FRAND commitment “is ‘governed by the laws of France,’ and is ‘solely [] contractual [in] nature.’” Whether Ericsson’s offer complied with the terms of its agreement with ETSI is thus a matter of French contract law. See, e.g. , Microsoft Corp. v. Motorola, Inc. , 795 F.3d 1024, 1040 (9th Cir. 2015) (explaining that a similar case was “not a patent law action” but a breach-of-contract case); Apple, Inc. v. Motorola Mobility, Inc. , 886 F. Supp. 2d 1061, 1081 (W.D. Wis. 2012) (applying “French law to Apple’s claims involving ETSI” in a breach-of- FRAND case); King Fung Tsang & Jyh-An Lee, Unfriendly Choice of Law in FRAND , 59 Va. J. Int’l L. 220, 263 (2019) (“Generally, U.S. cases have characterized all three FRAND issues as contractual.”).

In spite of this, HTC’s proposed jury instructions are based on United States patent law. HTC did not even attempt to justify its proposed instructions under French contract law or to argue that French contract law and United States patent law are equivalent. Nor did it try to support its position by pointing to the specific, governing provisions of Ericsson’s agreement with ETSI. Because HTC relied on inapplicable law to support its proposed instructions, the district court was well within its discretion in refusing to give those instructions.

Even if United States law were to govern the case, none of HTC’s instructions would have been required. For the proposed apportionment instructions, HTC relied on patent-damages cases for support, not other breach-of-contract cases. In patent-damages cases involving a FRAND commitment, the Federal Circuit has held that “district courts must make clear to the jury that any royalty award must be based on the incremental value of the invention, not the value of the standard as a whole or any increased value the patented feature gains from its inclusion in the standard.” D-Link Sys., Inc. , 773 F.3d at 1235. In such cases, then, a general apportionment instruction may be necessary. See id. HTC urges us to adopt a similar requirement in this case by holding that the district court abused its discretion in refusing to provide HTC’s requested instructions. The United States as an amicus supports the use of a general apportionment instruction where, as here, the agreement at issue does not specifically define what constitutes FRAND.

However, as other courts have recognized, a breach-of-contract case involving a breach-of-FRAND claim “is not a patent law action.” Microsoft Corp. , 795 F.3d at 1040. The act of determining the value of a standard- essential patent for purposes of calculating damages is distinct from assessing whether a particular offer made during unique licensing negotiations was fair and reasonable. So, while “the Federal Circuit’s patent law methodology can serve as guidance in contract cases on questions of patent valuation,” id. , it does not explicitly govern the interpretation of contractual terms, even terms that are intertwined with patent law. Cf. Bonzel v. Pfizer, Inc. , 439 F.3d 1358, 1363 (Fed. Cir. 2006) (“The nature of this contract action does not change because the contract is a patent license . . . .”). Thus, although the district court may have been permitted to use United States patent-law principles to guide its instructions (assuming that United States law were to apply), it was not obligated to do so.

As for the non-discrimination instruction that HTC requested, it is erroneous because it flatly contradicts the agreement between Ericsson and ETSI. No federal court of appeals has directly considered a FRAND non- discrimination claim in detail. But as with the other elements of FRAND, “[t]rial courts should . . . consider the patentee’s actual [F]RAND commitment in crafting the jury instruction.” D-Link Sys., Inc. , 773 F.3d at 1231 (discussing jury instructions in a patent-damages case). HTC’s proposed non-discrimination instruction would do away with any “difference in terms” offered to potential licensees if the difference “creates a competitive disadvantage for a prospective licensee.” The proposed instruction described the non-discrimination prong of FRAND as something designed to “level the playing field among competitors . . . by prohibiting preferential treatment that imposes different costs to different competitors.”

HTC’s proposed instruction would transform the non-discrimination element of FRAND into a most-favored-licensee approach, which would require Ericsson to provide identical licensing terms to all prospective licensees. But ETSI has already rejected a most-favored-licensee approach and chosen to give patent holders some flexibility in coming to reasonable agreements with different potential licensees. See J. Gregory Sidak, Fair and Unfair Discrimination in Royalties for Standard-Essential Patents Encumbered by a FRAND or RAND Commitment , 2 Criterion J. on Innovation 301, 313–14 (2017). Giving HTC’s proposed instruction to the jury would have caused confusion regarding the scope of the non-discrimination element of the actual FRAND agreement between Ericsson and ETSI. The district court was well within its discretion to refuse to give such an instruction. ^[3]

Even if HTC’s proposed instructions were substantially correct statements of the law, we still would not reverse the district court’s decision to exclude them. Such an exclusion would constitute reversible error only if the proposed instructions “concerned an important point in the trial such that the failure to instruct the jury on the issue seriously impaired [HTC’s] ability to present a given [claim].” Kanida , 363 F.3d at 578 (quoting McClatchy , 249 F.3d at 356). HTC has not demonstrated that this is the case.

On appeal, HTC contends that the district court should have instructed the jury to apportion the value of Ericsson’s patents from the end product and base the reasonable royalty rate “on the incremental value that the patented invention adds to the end product.” But whether apportionment was necessary was not an issue at trial. Both parties agreed that apportionment was required and that the patent’s value needed to be based on the value that the patent added to the end product and not on any value derived from being a part of the standard.

Only the methodology for determining the patent’s value was disputed. HTC fought to convince the jury that the patent’s apportioned value should be based on the smallest salable patent-practicing unit rather than the net value of the device. Ericsson agreed that apportionment was necessary; it simply thought that the best way to determine the value that its patents added to the end product was by looking to the negotiated-for prices that others had paid to license those patents. Thus, HTC’s proposed apportionment instructions did not concern an important point at trial because they addressed an undisputed issue.

Moreover, the failure to give HTC’s three proposed instructions in no way impaired HTC’s ability to present its claims. In other cases, we have declined to correct the erroneous exclusion of jury instructions when the jury instructions that were given were substantively correct “and trial counsel was able to present the jury with the inferences they were permitted to make from the evidence.” Kanida , 363 F.3d at 579; see also United States v. Cain , 440 F.3d 672, 674–75 (5th Cir. 2006) (holding that the exclusion of a requested instruction was not reversible error because the defendant “was in no way precluded from producing evidence that” supported his position, “nor was he precluded from arguing that the Government had failed to present sufficient evidence” to convict). Under those circumstances, we reasoned that the party “was free to tailor her argument to fit the court’s instructions and to argue based upon those instructions that the jury” should make certain inferences in her favor. Kanida , 363 F.3d at 579. This meant that her ability to present her claim was unimpaired, which in turn meant that the error was not reversible. See id.

In this case, we likewise conclude that there is no reversible error. Even assuming arguendo that the district court had incorrectly excluded the apportionment instructions, its other FRAND instructions were legally accurate. And, as in Kanida , HTC presented its evidence and argued to the jury that apportionment was required and that HTC’s apportionment method was the superior one. HTC then explicitly covered the substance of its proposed instructions during its closing arguments. HTC was fully able to present its case to the jury. The absence of its requested instructions did not affect its ability to present its case to the jury. HTC has therefore failed to demonstrate both that the district court was wrong to exclude HTC’s proposed jury instructions and that any alleged error is reversible by this court. ^[4]

III. HTC next challenges the district court’s declaratory judgment— which ruled that Ericsson’s licensing offers to HTC were FRAND—as legally erroneous. HTC contends that “[g]iven the acknowledged, substantial disparities between license terms offered to HTC and several of its competitors, the district court could not hold those terms nondiscriminatory as a matter of law.”

HTC’s contention is little more than a sufficiency-of-the-evidence challenge to the jury verdict masquerading as a challenge to the declaratory judgment. It is true that the jury verdict and the declaratory judgment vary slightly—the first finding that Ericsson’s offers were not a breach of FRAND and the second holding that Ericsson’s offers actually were FRAND. Therefore, the evidentiary threshold for each is also slightly different. Cf. United States v. Espinosa-Cerpa , 630 F.2d 328, 330–33 (5th Cir. 1980) (explaining that “an acquittal is not to be taken as the equivalent of a finding of the fact of innocence”); Stevenson v. Sears, Roebuck & Co. , 713 F.2d 705, 710–11 (Fed. Cir. 1983) (explaining that failure to prove a patent invalid does not automatically establish validity).

Nevertheless, in this case, the evidence used to establish that Ericsson’s offers did not breach its FRAND obligations is also the same evidence used to establish that Ericsson’s offers affirmatively were FRAND. Although the declaratory judgment was a conclusion of law, it expressly relied on the facts as found by the jury. Challenging the evidentiary sufficiency of the declaratory judgment thus also challenges the sufficiency of the evidence supporting the jury’s verdict.

To properly preserve such a challenge, HTC needed to file a Rule 50(b) motion. See Roberts v. Wal-mart Stores, Inc. , 7 F.3d 1256, 1259 (5th Cir. 1993). It failed to do so. “Absent such a motion, . . . an appellate court is ‘powerless’ to review the sufficiency of the evidence after trial.” Ortiz v. Jordan , 562 U.S. 180, 189 (2011) (quoting Unitherm Food Sys., Inc. v. Swift- Eckrich, Inc. , 564 U.S. 394, 405 (2006)). We are therefore powerless to review the sufficiency of the evidence supporting the jury’s verdict, and by extension the sufficiency of the evidence supporting the district court’s declaratory judgment. ^[5]

Even if had HTC adequately preserved its challenge, we would not reverse the district court’s declaratory judgment. A court should grant a Rule 50(b) motion “ only if ‘the facts and inferences point so strongly and overwhelmingly in favor of one party that the Court believes that reasonable men could not arrive at a contrary verdict.’” Brown v. Bryan Cnty. , 219 F.3d 450, 456 (5th Cir. 2000) (quoting Boeing v. Shipman , 411 F.2d 365, 374 (5th Cir. 1969) ( en banc ), overruled on other grounds by Gautreaux v. Scurlock Marine, Inc. , 107 F.3d 331 (5th Cir. 1997) ( en banc )). Moreover, on appeal we would “consider all of the evidence, drawing all reasonable inferences and resolving all credibility determinations in the light most favorable to the non- moving party”—in this case, Ericsson. Id. The same standards would apply to HTC’s challenge to the declaratory judgment because the declaratory judgment relies so heavily on the facts as found by the jury.

Ericsson presented substantial evidence to support its position that it had offered FRAND terms to HTC. Ericsson pointed to licenses with similarly situated companies to HTC that had terms that were remarkably similar to those offered to HTC. This supported both its valuation arguments (suggesting that the value of Ericsson’s patents was best determined by looking to comparable licenses) and its non-discrimination arguments (suggesting that other similarly situated competitors had received similar licenses).

HTC countered with evidence regarding the value of Ericsson’s patents using its own valuation method based on the smallest salable patent- practicing unit. HTC further argued that Ericsson’s licenses with companies like Apple, Samsung, and Huawei were much more favorable, but Ericsson presented additional evidence indicating that those companies were not similarly situated to HTC due to a variety of factors. The jury was not required to accept HTC’s arguments or evidence regarding valuation or non- discrimination. Indeed, as the district court concluded, the jury must have rejected HTC’s evidence and arguments in order to conclude that HTC had not carried its burden of proving that Ericsson’s offers to HTC were not FRAND. And the evidence that HTC points to in its briefing before this court is not so overwhelmingly in HTC’s favor that, even drawing all inferences in Ericsson’s favor, no reasonable jury could have found for Ericsson. Accordingly, even if we had power to review HTC’s challenge to the district court’s declaratory judgment, we would still reject it.

IV. Finally, HTC challenges the district court’s exclusion at trial of testimony and a report proffered by an expert used by Ericsson in prior litigation but not in this litigation. HTC wanted to demonstrate that Ericsson’s experts in a previous case had calculated royalty rates in comparable licensing agreements at a lower rate than the one offered to HTC. The district court granted Ericsson’s motion in limine to exclude this evidence, yet HTC tried to present it to the jury via cross-examination of another expert. The district court excluded the evidence as hearsay and HTC now appeals the exclusion of the report and testimony of one expert in particular.

We review evidentiary rulings for abuse of discretion. EEOC v. Manville Sales Corp. , 27 F.3d 1089, 1092–93 (5th Cir. 1994). “A trial court abuses its discretion when its ruling is based on an erroneous view of the law or a clearly erroneous assessment of the evidence.” Hinojosa v. Butler , 547 F.3d 285, 292 (5th Cir. 2008) (quoting United States v. Yanez Sosa , 513 F.3d 194, 200 (5th Cir. 2008)). Even where error is present, we will not vacate a judgment “if the error does ‘not affect any party’s substantial rights.’” Id. (quoting Fed. R. Civ. P. 61).

Here, HTC sought to cross-examine Ericsson’s expert, Mr. Mills, using testimony from an expert that Ericsson had used in a prior litigation, Mr. Kennedy. The district court excluded this testimony as hearsay. HTC insists that Kennedy’s prior testimonial evidence is not hearsay but is instead a statement of an opposing party. Under Federal Rule of Evidence 801(d)(2)(C), a statement that “is offered against an opposing party” and “was made by a person whom the party authorized to make a statement on the subject” is not hearsay. According to HTC, Ericsson’s previous expert was a person given authority to speak on Ericsson’s behalf in court on the topic of patent valuation.

We disagree. Although “[t]here are circumstances where the testimony of an expert witness can serve as an adverse admission against the presenting party,” we have explained that “[s]uch circumstances must be narrowly construed.” Fox v. Taylor Diving & Salvage Co. , 694 F.2d 1349, 1355 (5th Cir. 1983). Those cases where we have allowed expert testimony to serve as a party admission have been limited to circumstances where the expert was an agent of the opposing party or where the expert’s statements or reports were prepared for the very case in which admission was sought. See, e.g. , id. at 1354–55 (“The expert testified at a very early stage in the plaintiff’s case.”); Collins v. Wayne Corp. , 621 F.2d 777, 781–82 (5th Cir. 1980), superseded on other grounds by amendment to Fed. R. Evid. 103(a), as recognized in Mathis v. Exxon Corp. , 302 F.3d 448, 459 n.16 (5th Cir. 2002); Theriot v. J. Ray McDermott & Co. , 742 F.2d 877, 882 (5th Cir. 1984) (explaining that Collins stands for the proposition that “an agent’s statement, made within the scope of his authority . . . is admissible against the principal as an admission”).

That is not the case here. Kennedy was not an agent of Ericsson simply by virtue of acting as an expert witness in a prior case. “In theory, despite the fact that one party retained and paid for the services of an expert witness, expert witnesses are supposed to testify impartially in the sphere of their expertise.” Kirk v. Raymark Indus., Inc. , 61 F.3d 147, 164 (3d Cir. 1995). This is different from an agent who is hired to investigate and prepare a report on an incident for the company’s own purposes. See id. at 163–64. “Since an expert witness is not subject to the control of the party opponent with respect to consultation and testimony he or she is hired to give, the expert witness cannot be deemed an agent.” Id. at 164; see also Vicki Franks, An Expert’s Expertise: Is Expert Deposition Testimony Impermissible Hearsay or an Admission by a Party-Opponent at Trial? , 36 Am. J. Trial Advoc. 263, 299–300 (2012 ) (“‘[A]uthority’ includes an underlying premise that a person is ‘authorized’ to speak on behalf of a company or organization; there is a presumed supervisory role by the party giving the authorization and underlying commonality in purpose. That does not exist with an expert. An expert acts as an independent source for objective evidence . . . .” (footnotes omitted)).

Kennedy’s testimony was prepared and given at a different time for a different case. Although Kennedy’s testimony may touch on the same general topic as Mills’s, Kennedy’s valuation was necessarily given for a different purpose and dealt with a distinct licensing situation. It had nothing to do with the present dispute over the licensing negotiations between Ericsson and HTC. The district court was well within its discretion in excluding Kennedy’s prior opinions.

* * * Finding no error in the district court’s jury instructions, declaratory judgment, or evidentiary rulings, we AFFIRM the district court’s judgment. Stephen A. Higginson, Circuit Judge , concurring in the judgment :

I write separately because, in my view, the district court erred when it refused to incorporate a mutually-requested apportionment instruction. However, because the district court’s error did not seriously impair HTC’s ability to present its case, I concur in the judgment.

Following the district court’s instruction-by-instruction review of its proposed final jury instructions, HTC objected to the exclusion of certain instructions, stating: “we’d like to request that certain instructions relating to various patent issues, the determination of FRAND and the like, be included.” ^[1] HTC then specifically identified each of the requested instructions, referring to them by the names and numbers used in HTC’s proposed jury instructions. HTC also directed the district court to supporting authorities for each of its instructions, explaining that the instructions were “based upon the authorities” cited in its proposed jury instructions. HTC’s apportionment instructions, in particular, were supported by a citation to Ericsson, Inc. v. D-Link Systems, Inc. , 773 F.3d 1201, 1232 (Fed. Cir. 2014). As detailed below, I conclude that D-Link presents significant support for the apportionment instructions.

I. Apportionment To understand whether HTC’s apportionment instructions were substantially correct statements of law, I take guidance from the Federal Circuit, which regularly confronts questions of patent valuation. ^[2] When a jury is asked to value an infringed SEP, the Federal Circuit has held that the “jury must be instructed” to “apportion damages for SEPs that cover only a small part of a standard.” D-Link , 773 F.3d at 1232–33; see also VirnetX, Inc. v. Cisco Sys., Inc. , 767 F.3d 1308, 1326–27 (Fed. Cir. 2014). In D-Link , a jury was asked to determine whether D-Link infringed Ericsson’s patents and, once it found infringement, to approximate damages for that infringement. D-Link , 773 F.3d at 1214. The judge then fixed an ongoing royalty award for Ericsson’s infringed patents. Id. The Federal Circuit vacated the jury’s damages award and the ongoing royalty award because the district court did not properly “caution the jury about the importance of apportionment” in the jury instructions. Id. at 1228, 1232–33, 1236.

The Federal Circuit held that “a royalty award for a SEP must be apportioned to the value of the patented invention (or at least to the approximate value thereof), not the value of the standard as a whole.” Id. at 1233; see also Commonwealth Sci. & Indus. Research Org. v. Cisco Sys., Inc. , 809 F.3d 1295, 1305 (Fed. Cir. 2015) (“[R]easonable royalties for SEPs generally . . . must not include any value flowing to the patent from the standard’s adoption.”). Instructing a jury on apportionment ensures that SEP holders are “only . . . compensated for the added benefit of their inventions,” not for the value of “unpatented features reflected in the standard” or “value added by the standard’s adoption of the patented technology.” D-Link , 773 F.3d at 1232–33.

The Federal Circuit’s unmistakable command that a jury assessing patent value must be instructed on apportionment comports with settled tenets of patent law. Indeed, the Supreme Court recognized long ago that a patentee “must in every case be given evidence tending to separate or apportion the defendant’s profits and the patentee’s damages between the patented feature and the unpatented features, and such evidence must be reliable and tangible.” Garretson v. Clark , 111 U.S. 120, 121 (1884). These legal requirements are rooted in practical fairness considerations—not all of the value of a particular standard can be attributed to its constituent parts. Instead, “[o]nce incorporated and widely adopted, . . . technology is not always used because it is the best or the only option; it is used because its use is necessary to comply with the standard.” D-Link , 773 F.3d at 1233. A widespread cellular network or patent contributing to that network, even if of average quality, is extraordinarily valuable merely because of its universality. Apportionment allocates that value appropriately and avoids any artificial windfall to the SEP holder. Thus, D-Link directs that the failure to instruct a jury on proper apportionment is error when the jury is asked to assess patent value.

D-Link , though decided in the patent infringement context, applies instructively to breach-of-FRAND cases. While a breach-of-FRAND case “is not a patent law action,” the “Federal Circuit’s patent law methodology can serve as guidance in contract cases on questions of patent valuation.” ^[3] Microsoft Corp. v. Motorola, Inc. , 795 F.3d 1024, 1040 (9th Cir. 2015); Bonzel v. Pfizer, Inc. , 439 F.3d 1358, 1363 (Fed. Cir. 2006) (noting that a federal circuit court outside of the Federal Circuit may “deem[] it appropriate to apply the law of patent infringement” to a breach of contract case in which the contract is a patent license). That is because a jury assessing patent infringement damages undertakes the same task of assessing whether an offered rate is FRAND. See id. ; Realtek Semiconductor, Corp. v. LSI Corp. , No. C-12-3451-RMW, 2014 WL 2738226, at *6 (N.D. Cal. June 16, 2014) (“[D]etermining damages for patent infringement is equivalent to declaring the parties’ rights under the RAND contract.”). After finding patent infringement, a jury is tasked with “‘award[ing] the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer.’” VirnetX , 767 F.3d at 1326 (quoting 35 U.S.C. § 284). Similarly, a jury assessing whether a SEP holder has breached its FRAND obligation must determine the value of the SEP holder’s portfolio and compare that value to the offered rates. Valuing the portfolio requires apportionment, just as valuing a patent requires apportionment. See id.

By refusing to instruct the jury on apportionment, the district court did not just put aside the Federal Circuit’s D-Link decision, but also contravened the wishes of the parties. Indeed, HTC and Ericsson both requested an apportionment instruction. Ericsson’s experts and corporate representative agreed at trial that the value of Ericsson’s SEP portfolio must be isolated from the value of standardization. And even the United States, as amici, agrees that a jury in a breach of FRAND matter should be instructed on apportionment. ^[4] The fundamental importance of an apportionment instruction was evidence to each party at trial, as well as the United States as an amici participant before us.

II. Whether the District Court Abused its Discretion in Refusing the

Requested Jury Instructions Given that HTC preserved its challenge to the district court’s exclusion of its apportionment instructions, I would reach the merits as follows.

“We review a district court’s jury instructions for abuse of discretion.” Jowers v. Lincoln Elec. Co. , 617 F.3d 346, 352 (5th Cir. 2010). Although trial judges are afforded “wide latitude in fashioning jury instructions,” the “trial court must ‘instruct the jurors, fully a nd correctly, on the applicable law of the case, and . . . guide, direct, and assist them toward an intelligent understanding of the legal and factual issues involved in their search for truth.’” Bender v. Brumley , 1 F.3d 271, 276 (5th Cir. 1993) (quoting Charles A. Wright & Arthur R. Miller, 9 Federal Practice and Procedure § 2556 (1971)).

Similarly, we “review[] a district court’s refusal to provide a requested jury instruction for abuse of discretion.” Kanida v. Gulf Coast Med. Pers. LP , 363 F.3d 568, 578 (5th Cir. 2004) (emphasis added) (quoting United States v. McClatchy , 249 F.3d 348, 356 (5th Cir. 2001)). “[T]he district court’s refusal to give a requested jury instruction constitutes reversible error ‘only if the instruction 1) was a substantially correct statement of law, 2) was not substantially covered in the charge as a whole, and 3) concerned an important point in the trial such that the failure to instruct the jury on the issue seriously impaired the party’s ability to present a given claim.’” Id. (quoting McClatchy , 249 F.3d at 356) (alterations omitted); see also Travelers Cas. & Sur. Co. of Am. v. Ernst & Young LLP , 542 F.3d 475, 489 (5th Cir. 2008); Russell v. Univ. of Tex. of Permian Basin , 234 F. App’x 195, 207 (5th Cir. 2007). Finally, “a party is entitled to reversal for a district court’s failure to give a particularly requested instruction only if the jury was misled by the given instructions.” ^[5] Baker v. Canadian Nat’l/Ill. Cent. R.R. , 536 F.3d 357, 364 (5th Cir. 2008); Folks v. Kirby Forest Indus. Inc. , 10 F.3d 1173, 1175 (5th Cir. 1994); Pierce v. Ramsey Winch Co. , 753 F.2d 416, 425 (5th Cir. 1985).

HTC’s apportionment instructions were substantially correct statements of law and were not covered in any other portion of the jury charge. Nevertheless, I do not conclude that the district court’s refusal to include the apportionment instruction “seriously impaired [HTC]’s ability to present [its] claim.” See Kanida , 363 F.3d at 578. Therefore, while I perceive that the district court erred by refusing to instruct the jury on apportionment, the error is not reversible.

A. HTC’s Proposed Instructions Were “Substantially” Correct

HTC’s proposed instructions on apportionment consisted of the following:

Reasonable Royalty – Apportionment Generally When the product(s) that are the subject to a claim of royalties by an SEP holder have both patented and unpatented features, a “reasonable royalty” requires that the value of the patented features is apportioned from the non-patented features using reliable and tangible evidence.

Reasonable Royalty – Apportionment of Royalties in FRAND context When dealing with SEPs subject to FRAND obligations, apportionment requires that the patentee’s royalty be premised on the value of the SEP holder’s patented invention(s), not any value added by the standard’s adoption of the patented invention(s). That is, a FRAND royalty must reflect only the value of the patented invention(s) and not the additional value that resulted from the patent(s)’ inclusion in the standard or the value resulting from the standard’s success in the marketplace.

Additionally, in this context, to ensure that a FRAND royalty is based on the incremental value that the [SEP holder’s] patented invention(s) add to the product(s), the SEP holder’s patented invention(s) must be apportioned from all of the other technology reflected in the standard that is not patented by the SEP holder.

The essential requirement is that the ultimate reasonable royalty award must be based on the incremental value that the patented invention adds to the end product.

These are “substantially” correct statements of law. Kanida , 363 F.3d at 578. The instructions are largely consistent with pattern jury instructions from a variety of sources. Model Patent Jury Instructions § 6.7 ( Fed. Circuit Bar Ass’n July ); Model Patent Jury Instructions § 5.10 ( United States Dist. Court for the N. Dist. of Cal. Dec. 2018 ); Model Patent Jury Instructions §§ 10.2.5.1, 10.2.5.4 ( Am. Intellectual Prop. Law Ass’n 2018 ).

Further, these instructions mimic language from leading cases on apportionment. See Garretson , 111 U.S. at 121; D-Link , 773 F.3d at 1232–33. For instance, although Ericsson now contends that requiring “reliable and tangible evidence” that apportionment has been accounted for is too high of a bar, Garretson uses that standard verbatim. Garretson , 111 U.S. at 121. Moreover, HTC’s proposed instructions incorporate the fundamental requirement of apportionment, asking that factfinders carefully isolate the value of SEPs from the value of standardization more broadly. See D-Link , 773 F.3d at 1232; Commonwealth Sci. , 809 F.3d at 1304–05; Broadcom Corp. v. Qualcomm Inc. , 501 F.3d 297, 314 (3d Cir. 2007).

At the charge conference, Ericsson also appears to have agreed that the instructions were legally correct. Instead of objecting to the instructions because they were legally erroneous, Ericsson merely objected that the instructions were “unnecessary” and “potentially confusing to the jury.” Ericsson now targets the legal accuracy of HTC’s proposed apportionment instructions. HTC’s proposed apportionment instructions state that “the value of the patented features” should be apportioned “from the non- patented features.” Ericsson contends that this is the wrong standard. Instead, Ericsson argues, a reasonable royalty award that accounts for apportionment should be “based on the incremental value that the patented invention adds to the end product.” Id. at 1226. But HTC’s broader instructions contained exactly the language Ericsson agrees should have been included: “The essential requirement is that the ultimate reasonable royalty award must be based on the incremental value that the patented invention adds to the end product.” ^[6]

B. HTC’s Proposed Instructions Were Not Covered Elsewhere

HTC’s proposed instructions “w[ere] not substantially covered in the charge as a whole.” Kanida , 363 F.3d at 578. The district court gave the jury no instructions regarding patent valuation or apportionment. The district court only instructed the jury regarding a breach-of-contract claim and then told the jury that there is no set methodology for determining a FRAND rate. C. HTC’s Ability to Present its Case Was Not Substantially Impaired

While the district court erred by refusing to include an apportionment instruction, I concur in the judgment because the absence of such an instruction in this particular case did not “seriously impair[]” HTC’s very “ability to present” its position. Id.

The district court did nothing to prevent HTC from presenting the jury with evidence about the importance of apportionment. And that is exactly what HTC did: repeatedly clarifying that the jury needed to engage in apportionment. At times, Ericsson’s witnesses even agreed. And, to the extent Ericsson presented testimony tying the perceived market value of standardized cellular connectivity to Ericsson’s SEPs, HTC could have—but did not—object to this testimony on apportionment grounds. HTC was not seriously impaired from presenting the evidence necessary to convince the jurors that apportionment was appropriate.

Moreover, the jury instructions actually administered left the jury with discretion. Specifically, with respect to FRAND, the district court instructed the jury as follows:

Whether or not a license is FRAND will depend upon the totality of the particular facts and circumstances existing during the negotiations and leading up to the license.

Ladies and gentlemen, there is no fixed or required methodology for setting or calculating the terms of a FRAND license rate.

Far from impairing HTC’s case or prohibiting application of any given theory, the district court gave the jury latitude to apportion value as the consensus of expert testimony told it to. ^[7]

III. Conclusion In my view, the district court erred when it did not instruct the jury to isolate the value of Ericsson’s SEPs from the value of standardization. The district court denied proposed apportionment instructions from both parties and diverged from the fundamental apportionment-instruction requirement currently required by the Federal Circuit. I recognize that this case presents what apparently is the first breach-of-FRAND jury trial in this country, making the district court’s omission understandable. But, although it may be the first such trial, it is unlikely to be the last. Indeed, recently, the Supreme Court was asked to consider whether a FRAND-bound SEP holder has a jury trial right in a proceeding seeking specific performance. TCL Comm. Tech. Holdings Ltd. v. Telefonaktiebolaget LM Ericsson , 943 F.3d 1360 (Fed. Cir. 2019), cert. denied , 141 S. Ct. 239 (2020). If more FRAND disputes are submitted to juries, proper instructions about apportionment and the FRAND obligation generally will be significant.

In short, because I conclude that the absence of apportionment instructions did not “substantially impair” HTC’s ability to present its case, I would hold that “the district judge’s failure to give the requested [apportionment] instruction . . . does not rise to the level of reversible error in this case.” Kanida , 363 F.3d at 579 (emphasis added).

HTC a license, on FRAND terms, to Ericsson’s cellular standard-essential patents.” Importantly, when the district court consulted with the parties about how to respond to the jury’s note, HTC articulated no linkage between the jury’s question and apportionment. Likewise, HTC did not object during Ericsson’s closing arguments when Ericsson urged valuation theories that do not align with apportionment.

[1] HTC also sought back royalties and brought antitrust claims against Ericsson under the Sherman Act. The district court compelled arbitration on these claims, however, and they are not at issue in this appeal.

[2] We note, however, that “[t]o preserve a jury instruction error, a party must make ‘a specific, formal, on -the-record objection.’” Eastman Chem. Co. v. Plastipure, Inc. , 775 F.3d 230, 241 (5th Cir. 2014) (quoting Jimenez , 660 F.3d at 845). Our precedent requires parties to meticulously comply with the instructions of Federal Rule of Civil Procedure 51 in order to properly preserve an alleged error for appeal. See id. ; Jimenez , 660 F.3d at 845; Kelly v. Boeing Petroleum Servs., Inc. , 61 F.3d 350, 361 (5th Cir. 1995); Russell v. Plano Bank & Tr. , 130 F.3d 715, 719 (5th Cir. 1997) (“We have repeatedly held that a general objection to the district court’s jury instructions is insufficient to satisfy Rule 51.”); Taita Chem. Co. v. Westlake Styrene, LP , 351 F.3d 663, 668 (5th Cir. 2003) (holding that an objection was inadequate where the party only generally objected “to every one of the numbered charges proposed and offered by [the party], one through 103, to the extent they were not incorporated in the final charges”).

[3] HTC’s proposed instructions also depart from the leading model instructions provided by the Federal Circuit Bar Association for patent-damages cases involving FRAND terms. Those instructions state that the jury “must consider the following two factors in setting a royalty: (1) Any royalty for the patented technology must be apportioned from the value of the standard as a whole; and (2) The royalty must be based on the value of the invention, not any value added by the standardization of that invention. Generally, the non-discrimination requirement of FRAND means that similarly situated firms should receive similar terms and conditions (or have access to the same range of terms and conditions).” Model Pat. Jury Instructions § 5.11 (Federal Circuit Bar Association 2020). HTC’s proposed instructions go well beyond these straightforward directives.

[4] We hold that HTC fails on both the first and third prongs of abuse-of-discretion review. Our analyses of each prong constitute independent bases for rejecting HTC’s position, and as such they are not dicta. “In this circuit, ‘alternative holdings are bi nding precedent and not obiter dicta .’” Ramos-Portillo v. Barr , 919 F.3d 955, 962 n.5 (5th Cir. 2019) (quoting Whitaker v. Collier , 862 F.3d 490, 496 n.14 (5th Cir. 2017)).

[5] HTC also argues that if we were to vacate or reverse the jury verdict and declaratory judgment, then we should also vacate the prevailing-party determination and costs awarded by the district court. Because we affirm the district court’s judgment in all respects, we need not address this issue.

[1] Prior to the formal, on-the-record, charge conference, the district court held an informal charge conference—in chambers and off-the-record—during which the parties apparently engaged in a “fulsome discussion” about the proposed jury instructions. Unfortunately, there is no record of what transpired during this informal charge conference, even though the parties repeatedly referred back to it during the formal charge conference. I would have benefitted from the complete record. Fed. R. Civ. P. 51(b)(2); see also King v. Ford Motor Co. , 597 F.2d 436, 440 n.3 (5th Cir. 1979); cf. Johnson v. Helmerich & Payne, Inc. , 892 F.2d 422, 424 n.3 (5th Cir. 1990).

[2] Ericsson takes the position that American caselaw is inapplicable because the ETSI FRAND commitment is governed by French law. I am not convinced. First , neither party has presented evidence that foreign law differs from American law on this issue. Indeed, Ericsson states in its brief “under French law and American law alike, the factfinder construes the terms [of an ambiguous contract] based on evidence of ‘the parties’ common intentions’ and the ‘prevailing industry standard or approach.’” As Ericsson has argued elsewhere, “[w]here neither party presents evidence that foreign law differs from American law – as here – a U.S. court must ‘apply [i ts] own law to the case.’” Corrected Non-Confidential Brief for Appellants Ericsson Inc. and Telefonaktiebolaget LM Ericsson at 39, TCL Comm. Tech. Holdings Ltd. v. Telefonaktiebolaget LM Ericsson , 943 F.3d 1360 (5th Cir. 2019), 2018 WL 3440544, at *39–40 (quoting The Scotland , 105 U.S. 24, 29 (1882)); Second , Ericsson itself cited to American caselaw to support its proposed jury instructions. This means that both parties and the district court relied on American law when assessing the propriety of the proposed jury instructions. Third , Ericsson does not dispute that the jury should engage in an apportionment analysis of some type when assessing a FRAND rate. Indeed, Ericsson emphasizes on appeal that its experts told the jury just that: “Nobody disputed at trial that the FRAND royalty must reflect patent value, apart from the value of other features or standardization itself.” Fourth , as Ericsson told the district court, “no other country today employs juries in adjudicating patent matters.” This leaves a dearth of caselaw from any jurisdiction beyond American borders about how to instruct a jury valuing a patent portfolio.

[3] See generally J. Gregory Sidak, Apportionment, FRAND Royalties, and Comparable Licenses After Ericsson v. D-Link, 2016 U. Ill. L. Rev. 1809, 1864–68 (2016); Richard H. Stern, Who Should Own the Benefits of Standardization and the Value it Creates? , 19 Minn. J.L. Sci. & Tech. 107, 170–71 & n.271 (2018).

[4] The United States was similarly emphatic during oral argument. Oral Argument Tr. at 28:00-28:05 (“In a case where there is a more general FRAND commitment as there is here, yes, we think an apportionment instruction must be given.”).

[5] I note for future clarification that numerous decisions articulating this court’s standard of review for omitted jury instructions do not include this fourth element. See, e.g. , Kanida , 363 F.3d at 578; Travelers Cas. & Sur. Co. of Am. , 542 F.3d at 489. Because I conclude HTC’s ability to present its case was not “substantially impaired,” I do not consider whether the instructions as given misled the jury—and therefore do not address whether this fourth element is necessary to the analysis.

[6] I note that, contrary to concerns expressed by Ericsson and amici, HTC’s proposed apportionment instructions do not dictate any particular patent valuation methodology. HTC’s apportionment instructions do not require use of the SSPPU as the royalty base. HTC’s apportionment instructions merely seek to alert the jury to the risk of awarding unearned, inflated, and unapportioned value to a given SEP portfolio. Ericsson’s point—that SSPPU may not be the appropriate royalty base and should not be mandated to the jury—is well taken. See Commonwealth Sci. , 809 F.3d at 1302–03; D-Link , 773 F.3d at 1226–28. But HTC’s proposed instructions do not collide with that concern, nor do HTC’s instructions even preclude the jury from relying on comparable license evidence.

[7] Although the jury expressed some confusion during deliberations, the jury’s confusion was generalized, asking the district court to “help us frame question #1 in different terms” and stating that the jury had “great difficulty getting jurors to agree on what the question is asking.” Question one asked “whether HTC prove[d] by a preponderance of the evidence that Ericsson breached its contractual obligation to offer