Summary
noting that the erroneous admission of expert testimony is subject to harmless error review
Summary of this case from Gallardo v. United StatesOpinion
No. 12–4182.
2014-03-11
STORAGECRAFT TECHNOLOGY CORPORATION, Plaintiff–Appellee, v. James KIRBY, an individual, Defendant–Appellant.
Richard F. Ensor of Vantus Law Group, P.C., Salt Lake City, UT, for Defendant–Appellant. Thomas R. Karrenberg (Heather M. Sneddon with him on the brief), of Anderson & Karrenberg, P.C., Salt Lake City, UT, for Plaintiff–Appellee.
Richard F. Ensor of Vantus Law Group, P.C., Salt Lake City, UT, for Defendant–Appellant. Thomas R. Karrenberg (Heather M. Sneddon with him on the brief), of Anderson & Karrenberg, P.C., Salt Lake City, UT, for Plaintiff–Appellee.
Before TYMKOVICH, SEYMOUR, and GORSUCH, Circuit Judges.
GORSUCH, Circuit Judge.
James Kirby says the jury's award against him is too much. True, he helped start and served as a director of StorageCraft, a computer software company. True, after a falling out with his colleagues he stole the computer source code on which the company's products depend. True, he shared the source code with NetJapan, a rival company that quickly produced a competing software product much like StorageCraft's. But the jury's $2.92 million trade secret misappropriation award is still too much. Too much, Mr. Kirby says, because he never used the secret for his own personal profit. And too much because StorageCraft never sought to prove at trial that NetJapan made commercial use of its trade secret either. Maybe he was angry about how his former colleagues had treated him, maybe he disclosed the trade secret to a rival out of vengeance. But without firmer proof that someone profited from his misdeed Mr. Kirby insists the jury's verdict should be overturned.
The trouble is Utah law doesn't distinguish between a misappropriator's venial motives. When someone steals a trade secret and discloses it to a competitor he effectively assumes for himself an unrestricted license in the trade secret. And that bears its costs. After all, what value does a trade secret hold when it's no longer a secret from the trade? The misappropriator may act with a wish to line his pockets or satisfy a vendetta or for some other purpose still. All the same Utah's trade secret statute holds him to account for the full value of the license he arrogated to himself. Just as the district court held.
I
Mr. Kirby's argument otherwise takes various but related forms. In its first and most ambitious guise Mr. Kirby insists a trade secret plaintiff cannot seek and obtain a “reasonable royalty” measure of damages—as StorageCraft did in this case—without proving the misappropriating defendant made commercial use of the secret. He presses this same essential point in at least three different ways—suggesting he is entitled to judgment as a matter of law because the record contains no evidence of commercial use, claiming that a new trial is warranted for the same reason, and arguing the district court erred by failing to give a jury instruction requiring proof of commercial use. But however the argument is dressed, underneath lies the same problem: Utah's trade secret statute, the law governing Mr. Kirby's case, expressly allows a reasonable royalty measure of damages when the misappropriator uses or discloses the trade secret. And no one disputes that Mr. Kirby did at least that— disclosing the secret to NetJapan.
Utah's Uniform Trade Secrets Act provides three possible measures of damages for misappropriation—the defendant's unjust enrichment, the plaintiff's “actual loss,” or “a reasonable royalty.” Utah Code Ann. § 13–24–4(1). This last option is sometimes described as “the price that would be set by a willing buyer and a willing seller” for a license in the trade secret, a measure of damages that seeks to recreate “an actual market transaction ... [in] which both parties gain from the transaction.” Restatement (Third) of Unfair Competition § 45 cmt. g (1995). At trial, the district court allowed StorageCraft to present evidence premised on a reasonable royalty damages theory. The company argued that its source code was its lawful trade secret; that Mr. Kirby stole it; that he disclosed it to a rival; that in doing so he effectively assumed for himself a license to reveal the trade secret to StorageCraft's competitor; that this diminished the value of its intellectual property and the products depending on it; and that Mr. Kirby should pay a royalty reflecting that much, whether or not he or NetJapan have to date made commercial use of that intellectual property in products of their own.
Utah law allows a plaintiff to proceed just as StorageCraft did. Contrary to Mr. Kirby's supposition, nothing in the state's trade secret statute categorically restricts the availability of “reasonable royalty” damages to cases in which the misappropriator used a trade secret commercially rather than disclosed it to others. To the contrary, the statute expressly provides that “[i]n lieu of damages measured by any other methods,” the reasonable royalty measure of damages is available “for a misappropriator's unauthorized disclosure or use of a trade secret.” Utah Code Ann. § 13–24–4(1) (emphasis added). In this respect, Utah's statute tracks the Uniform Trade Secrets Act, which itself provides that reasonable royalty damages are a “general option” for cases involving disclosure as well as use. SeeUnif. Trade Secrets Act § 3 & cmt. (amended 1985); cf. Hertz v. Luzenac Grp., 576 F.3d 1103, 1115 (10th Cir.2009) (noting that under Colorado's materially identical trade secret statute reasonable royalty damages are allowed for a misappropriator's disclosure or use of a trade secret); Sonoco Prods. Co. v. Johnson, 23 P.3d 1287, 1290 (Colo.App.2001) (same).
Neither does Utah's policy choice on this score come without its reasons. In the first place, the line between use and disclosure is hardly as crisp as Mr. Kirby suggests. Can't disclosing a trade secret for a particular end or purpose (be it retribution or profit or otherwise) be a way of putting it to use, at least in a broad sense of the word? What happens when the disclosure is made to a third party (like NetJapan) with the intent the third party itself put the trade secret to commercial use in ways harmful to the secret's owner? Isn't at least that disclosure a use of the secret, whether or not the third party takes up the invitation?
Beyond these definitional difficulties, where (as here) a defendant discloses a trade secret to a rival company in a fit of retaliatory pique without any desire for personal riches, the other two measures of damages may not always be entirely fit for the task. An award based on unjust enrichment risks undercompensating the plaintiff when the defendant has no gains of his own to disgorge. SeeRestatement (Third) of Unfair Competition § 45 cmt. g (commending use of reasonable royalty measure of damages “when the plaintiff's loss ... is ... greater than any gain acquired by the defendant”). Though what the Utah statute calls the “actual loss” measure of damages doesn't suffer from this particular problem, it may invite practical difficulties of its own. In cases like ours the best evidence about the extent of the plaintiff's lost sales isn't readily available from the defendant before the court but resides instead in the hands of far-flung third parties like NetJapan. Proving a causal connection between the plaintiff's claimed lost profits and the defendant's conduct might be difficult, too, in these circumstances. Complexities like these may be surmountable, but the cost of doing so may be enough to explain why a state would wish to make reasonable royalty awards generally available to misappropriation plaintiffs. After all, it is hardly unknown for the law to resolve ambiguities about the appropriate quantity of damages against the proven wrongdoer rather than his victim. See, e.g., Russo v. Ballard Med. Prods., 550 F.3d 1004, 1021 (10th Cir.2008).
In recognizing this much we don't mean to suggest that the reasonable royalty measure of damages is always the most sensible remedy. Or that every state must be as solicitous to it as Utah. Rather than follow the Uniform Trade Secrets Act and deem the reasonable royalty remedy a “general option” for disclosure cases, some states allow the remedy only when the plaintiff is unable to prove the amount of its actual losses or the misappropriator's unjust gains. See Cacique, Inc. v. Robert Reiser & Co., 169 F.3d 619, 623 (9th Cir.1999). We can imagine arguments that might lead reasonable lawmakers in that direction as well, including worries that hypothetical royalty negotiation exercises themselves might be difficult to administer in certain circumstances or might yield damages in excess of the plaintiff's actual losses. Cf. Rite–Hite Corp. v. Kelley Co., 56 F.3d 1538, 1554 n. 13 (Fed.Cir.1995) (en banc) (finding in a patent case that the “willing licensor/willing licensee” description is “an inaccurate, and even absurd, characterization when ... the patentee does not wish to grant a license”); Mark A. Lemley, Distinguishing Lost Profits from Reasonable Royalties, 51 Wm. & Mary L.Rev. 655, 667 (2009) (worrying that “some patentees who can prove lost profits elect instead to seek a ‘reasonable’ royalty that is far in excess both of what the parties would have negotiated and of the actual losses the patentee suffered”).
In the end, though, arguments like these are more appropriately directed to those charged with writing Utah's trade secret statute than those charged with applying it. To decide this case it's enough for this court to recognize and respect Utah's policy choice to permit “reasonable royalty” awards as a “general option” in “disclosure” cases. We are in no position to override that legislative choice simply because we might prefer another. Neither does Mr. Kirby suggest Utah's legislative choice offends any higher principle of law that we do have the power to enforce. He does not, for example, suggest that the use of the reasonable royalty measure of damages in his case or more generally is so speculative that it offends the Federal Constitution's due process guarantee.
Instead, Mr. Kirby replies a good deal more modestly by directing our attention to University Computing Co. v. Lykes–Youngstown Corp. and citing it for the proposition that at common law a misappropriation defendant generally had to “put the trade secret to some commercial use” before a reasonable royalty award was allowed. 504 F.2d 518, 539 (5th Cir.1974). The problem is, University Computing offered its view of the common law's requirements well before Utah passed the statute we are called upon to apply and well before the adoption of the Uniform Trade Secrets Act on which the Utah statute is modeled. What's more, University Computing proceeded to hold that the defendant there engaged in “a clear commercial use” of the trade secret merely by displaying it to a third party who could use the secret to the owner's disadvantage. See id. at 542;see also Metallurgical Indus. Inc. v. Fourtek, Inc., 790 F.2d 1195, 1205 (5th Cir.1986) (noting the breadth of University Computing 's definition of commercial use). As we've seen, the same description could easily be applied to our case—so even taking University Computing and its description of the common law at full value does nothing to help Mr. Kirby's cause.
Beyond even these problems lurks still another. To the extent University Computing or similar cases cited by Mr. Kirby happen to emphasize commercial use apart from disclosure, it's noteworthy that they tend to borrow generously from patent law's approach to reasonable royalty damages. See, e.g., Univ. Computing, 504 F.2d at 535, 536–37;Olson v. Nieman's, Ltd., 579 N.W.2d 299, 310 (Iowa 1998). Under the terms of the federal patent statute (unlike those of the Uniform Trade Secrets Act), a reasonable royalty award does depend on “the use made of the invention by the infringer.” 35 U.S.C. § 284. But the difference between the Patent Act and Uniform Trade Secrets Act on this score is hardly surprising given basic differences between patents and trade secrets. Sharing information about another's patented invention is generally not the injury disclosing a trade secret is. After all, patentees aren't allowed to keep their innovations entirely to themselves; telling us how to practice their invention is the price they must pay for their patent. Universal Oil Prods. Co. v. Globe Oil & Ref. Co., 322 U.S. 471, 484, 64 S.Ct. 1110, 88 L.Ed. 1399 (1944); see also35 U.S.C. § 112(a). Because the crucial information about patented inventions is already a matter of public record, patent cases on reasonable royalties have had no cause to address unauthorized disclosure. Their focus has been and must be on authorized uses. To be sure, patent cases still provide useful instruction for courts considering damages for trade secret misappropriation, as we have noted before. See, e.g., Telex Corp. v. Int'l Bus. Machs. Corp., 510 F.2d 894, 930 (10th Cir.1975). But in light of the differences between the patent and trade secret statutes—and between patents and trade secrets themselves—patent law's treatment of reasonable royalties can't be read to suggest that a reasonable royalty award in trade secret cases must always and as a matter of logical necessity require commercial use.
Mr. Kirby relies on various Utah cases which say a precondition for trade secret misappropriation liability is the “use of the secret that injures” its owner. See, e.g., USA Power, LLC v. PacifiCorp, 235 P.3d 749, 758 (Utah 2010); Water & Energy Sys. Tech., Inc. v. Keil, 974 P.2d 821, 822 (Utah 1999). But these decisions also rely on a standard that predates the enactment of Utah's Uniform Trade Secrets Act. Microbiological Research Corp. v. Muna, 625 P.2d 690, 697–98 (Utah 1981). Because we are told elsewhere to make the Utah Uniform Trade Secrets Act's “plain language” our first source of guidance, Water & Energy Sys. Tech., Inc. v. Keil, 48 P.3d 888, 894 (Utah 2002), and because that statute clearly finds misappropriation in improper “ disclosure or use of a trade secret,” Utah Code Ann. § 13–24–2(2)(b) (emphasis added), we're confident that “use” in these cases should be understood at least today in the broad sense we've identified. Neither does Mr. Kirby contest his liability for misappropriation (only the scope of the damages he must pay) and formally at least liability is the only subject of these cases.
II
Even if reasonable royalty damages are permissible in principle for disclosure cases like his, Mr. Kirby insists the damages assigned by the jury in his case are still too much. If the animating idea behind the reasonable royalty exercise is to fix a price the parties would have agreed upon for a license, Mr. Kirby insists a court must ask, a license for what purpose? And because the trial record shows he did nothing much with the trade secret he stole, Mr. Kirby insists that the jury's $2.92 million award represents anything but a reasonable royalty. For this reason and yet again he insists he is entitled to judgment as a matter of law or a new trial.
Our earlier disagreement with Mr. Kirby was about the law. This time it is about the facts. As we've seen, a reasonable royalty award seeks to estimate the fee the defendant would have had to pay, after negotiation, to secure a lawful license for the trade secret he unlawfully assumed for himself. And in any licensing negotiation it's surely true that many considerations can come into play when setting a price, considerations ranging from the duration and scope of the license to market prices paid for similar licenses. One court has claimed to identify no fewer than fifteen factors that may enter into the mix when attempting to fix a reasonable royalty award. Another has generated a different but in some ways overlapping list. Compare Ga.-Pac. Corp. v. U.S. Plywood Corp., 318 F.Supp. 1116, 1120 (S.D.N.Y.1970), modified in part sub nom. Ga.-Pac. Corp. v. U.S. Plywood–Champion Papers, Inc., 446 F.2d 295 (2d Cir.1971), with Univ. Computing, 504 F.2d at 539. Because some of these factors aren't relevant to every case, those tasked with estimating a reasonable royalty bear the considerable burden of identifying the ones that are most pertinent in the particular circumstances at hand before attempting to assign each an appropriate weight. In this way, reasonable royalty calculations demand no small measure of judgment—perhaps like damages calculations in nearly every complex business dispute where courts are called on to imagine a “but for” world that should have been but isn't.
Having said all that, we don't doubt the breadth of a trade secret license can have a significant effect on price, just as Mr. Kirby suggests. No one doubts that a trade secret holder and licensee might agree to one price for a highly restrictive license and an entirely different price for an unrestricted license in the very same secret. For this reason, it is surely important when setting a reasonable royalty award to account for the scope of the license the defendant assumed for himself, to aim at a price that reflects the particular “use of the trade secret made by the defendant.” Restatement (Third) of Unfair Competition § 45 cmt. g; see also Ga.-Pac. Corp., 318 F.Supp. at 1120;Univ. Computing, 504 F.2d at 539. At least in this very broad sense of the word, then, the nature of the defendant's “use” of the trade secret matters a great deal in the reasonable royalty analysis. While a defendant in Utah can't avoid reasonable royalty damages because he disclosed the trade secret to others without anyone making commercial use of it, the amount of reasonable royalty damages he must pay undoubtedly depends on what he did with the secret—what uses he made of it.
Even acknowledging all this, however, still does nothing to help Mr. Kirby. And here his problem lies in the record. He would have us equate him with a hypothetical misappropriator who stole a trade secret and then disclosed it only to his long since retired, seriously aged, and tight-lipped great-grandfather, someone who didn't and couldn't make use of the secret and who therefore couldn't do serious damage to the trade secret or its rightful holder. Were these the facts of our case, we might agree with Mr. Kirby that the jury's $2.92 million award was unreasonable as a matter of law. But those aren't the facts of this case. Instead, the evidence at trial showed that Mr. Kirby took StorageCraft's trade secret and intentionally disclosed it to NetJapan, aware that NetJapan was an able competitor, and aware that NetJapan could well use the secret to compete with StorageCraft. Mr. Kirby may have acted for no personal gain save the satisfaction of revenge. StorageCraft may not have sought to prove to the jury that NetJapan's coincidentally timed product actually depended on its trade secret. But the facts all the same permitted the jury to issue an award premised on an understanding that the license Mr. Kirby assumed allowed him the right to share the trade secret with a rival and permit (if not compel) its commercial exploitation by that rival, leaving the remaining financial value of the (former) secret seriously compromised. That set of facts—that particular kind of use—is precisely the basis on which StorageCraft's expert reached his damages estimate and the jury reached its award. Mr. Kirby does not suggest before us that an award of $2.92 million is unreasonable as a matter of law when premised on facts like these, nor does he dispute that the value of a trade secret is seriously diminished when it is no longer a secret from a trading rival. And we are hardly permitted to accept his invitation to substitute his alternative and self-serving version of the facts for those the jury reasonably found at trial. See, e.g., Escue v. N. Okla. Coll., 450 F.3d 1146, 1156–57 (10th Cir.2006).
III
That takes us finally to Mr. Kirby's challenge to StorageCraft's expert. Before admitting expert testimony at trial, a district court must assure that it is both reliable and relevant. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993); Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 147, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). When a party objects to proposed expert testimony, the court “must adequately demonstrate by specific findings on the record” that it has taken these gate-keeping responsibilities seriously. Goebel v. Denver & Rio Grande W. R.R. Co., 215 F.3d 1083, 1088 (10th Cir.2000). This, Mr. Kirby says, the district court failed to do.
We have yet to identify some unifying theory or principle for discerning the precise point at which a district court's gate-keeping findings prove sufficient. But several lessons emerge from a review of our existing decisions.
First, it is not sufficient for a district court simply to say on the record that it has decided to admit the expert testimony after due consideration. Instead, the district court must furnish enough of a record to permit a reviewing court to say with confidence that it “properly applied the relevant law.” United States v. Avitia–Guillen, 680 F.3d 1253, 1259 (10th Cir.2012); see also United States v. Roach, 582 F.3d 1192, 1207 (10th Cir.2009).
Second, the district court must reply in some meaningful way to the Daubert concerns the objector has raised. So, for example, if the reliability of an expert's methodology is at issue, it's not good enough for the district court to stress the expert's qualifications. See, e.g., Dodge v. Cotter Corp., 328 F.3d 1212, 1227 (10th Cir.2003); Goebel, 215 F.3d at 1088. At the same time, a district court doesn't have to discuss in every case all of the reliability factors that the Supreme Court identified in Daubert and Kumho. A district court's gate-keeping function is more flexible than that, requiring the court to focus its attention on the specific factors implicated by the circumstances at hand. See Kumho, 526 U.S. at 141–42, 119 S.Ct. 1167;Goebel, 215 F.3d at 1088. And, other things equal, more complicated challenges demand lengthier discussions while less complicated challenges require less discussion. See, e.g., Avitia–Guillen, 680 F.3d at 1259 (distinguishing between “a challenge to an expert's methodology in a complicated area of medical science” and a challenge to a fingerprint “expert's qualifications to testify to a commonly used method of identification”); Dodge, 328 F.3d at 1228 (stressing “the novelty of the [experts'] medical causation theory”); cf. Charles Alan Wright & Victor James Gold, Federal Practice and Procedure § 6266 (1st ed. 1997 & Supp.2013).
Third, a district court's insufficient gate-keeping findings may not warrantreversal if the appellee can persuade us the error was harmless. If, for example, it is readily apparent from the record that the expert testimony was admissible, it would be pointless to require a new trial at which the very same evidence can and will be presented again. See, e.g., Kinser v. Gehl Co., 184 F.3d 1259, 1271–72 (10th Cir.1999), abrogated in part on other grounds by Weisgram v. Marley Co., 528 U.S. 440, 120 S.Ct. 1011, 145 L.Ed.2d 958 (2000). Even if this court concludes the expert's testimony was wrongly admitted, the presentation of that evidence might still qualify as harmless error “if other competent evidence is ‘sufficiently strong’ to permit the conclusion that the improper evidence had no effect on the decision.” Goebel, 215 F.3d at 1089 (internal quotation mark omitted). Either way, we will not demand a new trial when the existing one reached the right result.
United States v. Roach denied that an appellate court could rely on the record to make its own determination that an expert witness was qualified to testify. 582 F.3d at 1207 n. 7. “[T]hat determination,” we said, “is strictly reserved for the district court.” Id. at 1208 n. 7. But Roach did not seek to explain how this statement could be squared with our earlier decision in Kinser that did exactly what Roach suggested an appellate court may not do. Because one panel of our court cannot overrule prior panel decisions and earlier panel decisions control over later ones, Muskrat v. Deer Creek Pub. Schs., 715 F.3d 775, 792 (10th Cir.2013), we must continue to treat Kinser as binding. In fact, it seems Kinser simply wasn't brought to the court's attention in Roach; neither, for that matter, was Roach's statement even a part of the court's holding in that case.
Before the district court Mr. Kirby presented several objections to the admission of StorageCraft's damages expert, but before this court he continues to press just two. The first pursues a familiar theme. He faults StorageCraft's expert for failing to adopt his legal theory that trade secret royalty damages require proof of commercial use. Plainly, though, this objection has a good deal less to do with the expert and his methodology than it does with Mr. Kirby's view of the law, amounting less to a Daubert objection than a sort of renewal of his motion for judgment as a matter of law. Given that, it's equally plain to us the district court adequately discharged its duty to address this argument when it analyzed the language of Utah's trade secret statute and interpreted that language as authorizing royalty damages for either commercial use or disclosure.
Mr. Kirby's remaining objection focuses on various assumptions the expert employed when seeking to quantify the costs StorageCraft incurred in developing its computer code. Mr. Kirby doesn't dispute the principle that StorageCraft could have sought to take into account its development costs when negotiating a license for its intellectual property. Instead, he attacks the particular assumptions the expert used to calculate those costs. For example, the expert assumed developers worked 40–hour weeks over 15 to 20 months to develop the source code and Mr. Kirby says this assumption is too high, unsupported by the evidence. Mr. Kirby complains as well that the district court failed to discharge its Daubert duties when it dismissed his complaints summarily, stating only that they “raised great points for cross-examination.”
As we've seen, by their nature some objections to expert testimony don't require from the district court more than a brief reply, and Mr. Kirby's objections to the expert's development cost assumptions may well fall into this category. But even if we assume for argument's sake that the district court's exegesis wasn't detailed enough to discharge its gate-keeping duties, we are confident any error here was harmless. Returning by way of example to the expert's assumption about the time required to develop the computer source code, Mr. Kirby's own deposition testimony confirms that he and three other engineers devoted a very large number of hours to its creation over a 15 to 20 month period. The expert's assumption, then, was grounded in record evidence, Mr. Kirby's own testimony belies his methodological complaint, and we can rest assured that more words from the district court would not have altered the admissibility of the expert's evidence.
Two housekeeping items. Besides attacking StorageCraft's trade secret verdict, Mr. Kirby also attacks its breach of contract verdict. But at this point there's nothing left for him to complain about. It's a fact that StorageCraft sued Mr. Kirby for breach of contract as well as trade secret misappropriation. The company alleged that Mr. Kirby breached a promise to return the company's intellectual property when he quit. The jury accepted this theory in its verdict. But because the parties stipulated that StorageCraft's misappropriation and breach of contract damages awards were duplicative, the district court entered judgment only on the misappropriation claim. Because of that, and because we affirm the judgment against Mr. Kirby for misappropriating StorageCraft's trade secret, there's no need to address the contract claim here. See, e.g., Celeritas Techs., Ltd. v. Rockwell Int'l Corp., 150 F.3d 1354, 1361–62 (Fed.Cir.1998). Mr. Kirby's challenge to the jury's award of exemplary damages and the district court's award of attorneys' fees likewise require no extended discussion. He acknowledges that his arguments on these issues depend for their success on the success of the arguments we've already addressed and rejected above.
The judgment is affirmed.