Mike's Train House, Inc. v. Lionel, L.L.C., Korea Brass and Yoo Chan Yang

*403 OPINION

Defendant-Appellant Lionel, L.L.C., appeals a jury verdict finding it liable for misappropriation of trade secrets and unjust enrichment, and awarding Plaintiff-Appellee Mike’s Train House, Inc. (“MTH”) damages exceeding $40 million. After unsuccessfully moving the district court for a new trial and judgment as a matter of law, Lionel appeals the jury verdict, the district court’s evidentiary decisions, the specificity with which MTH identified its “trade secrets,” the imposition of joint and several liability, and the amount of the damage award. Lionel also appeals the district court’s order granting MTH’s request for an injunction. Because Lionel correctly argues that the district court erred in admitting expert testimony and in imposing joint-and-several liability, and because the jury award improperly “double counts” MTH’s damages, we REVERSE the district court’s order denying Lionel’s motion for a new trial and REMAND this case for further proceedings consistent with this opinion. Because the district court’s injunction is based entirely on the jury’s verdict, we REVERSE that order as well.

I. BACKGROUND

This case involves MTH’s claim that Lionel misappropriated MTH’s trade secrets and was unjustly enriched by that misappropriation. Before MTH brought this claim, the parties were involved in both criminal and civil litigation in South Korea, based on the same underlying events. After several individuals associated with Lionel were convicted of criminal charges, and Lionel and MTH settled the civil suit, MTH filed the instant complaint against Lionel in federal district court.

Lionel and MTH are both distributors of O-gauge model trains. The model trains they distribute are scale replicas of actual trains. Both Lionel and MTH contract with third parties for the manufacture of their trains and then release the trains for distribution under their own names. “Gauge” refers to the width of the track; collectors tend to collect trains of only one gauge (in this case “0 gauge”), so that all their trains can operate on the same track. The trains are manufactured by a process known as zinc die-casting, which is a precise method of construction in which molten metal is injected into pre-shaped molds. This method requires extensive design work for the construction of a single train; accordingly, designers produce “design drawings,” which contain all the information necessary to build a die-cast train. About 200 to 300 design drawings are required to produce each scale engine.

Lionel has been producing O-gauge model trains since the 1930s. In 1987, Lionel hired MTH, founded by Michael Wolf, to work with Samhongsa, a Korean supplier, to design and manufacture die-cast trains, for release under Lionel’s name. MTH and Samhongsa designed and produced several different die-cast trains for Lionel between 1987 and 1993. MTH’s relationship with Lionel ended in 1993. Between 1993 and 1998, Lionel used a number of die-cast manufacturers, but none produced high-quality trains at competitive prices. In 1994, Samhongsa began making O-gauge die-cast trains for resale under MTH’s name, a relationship that continued until 2002.

Soon-Gap Ahn, one of the leading designers at Samhongsa from 1997-98, assisted in designing trains for both Lionel and MTH during each company’s relationship with Samhongsa. Ahn testified that Samhongsa designers, subcontract designers, members of various departments at *404 Samhongsa, and other subcontractors had access to all Samhongsa designs and were allowed to make and retain copies of those designs. Ahn also testified that Korean designers use a common numbering system in their design drawings to refer to the same parts of the model train. For example, engine parts are numbered in the “1000 range” by Korean design companies.

During the time that Samhongsa worked for both Lionel and MTH, Samhongsa subcontracted with a company named Korea Brass (“KB”), a supplier of wax castings for model trains. KB had been formed in 1984 by Sung-Won Cho, a former Sam-hongsa employee. In 1998, KB expanded and became a direct supplier and manufacturer of model trains. After Cho and Yoo Chan Yang, KB’s sales agent in the United States, approached Lionel in April 1998, Lionel placed an order with KB to manufacture two die-east-model-train lines.

KB then hired Ahn to design trains that it intended to manufacture for Lionel. Before departing Samhongsa, Ahn copied design drawings for MTH trains onto computer disks; Ahn testified that he took the drawings to work on them at home, and that the other designers acted similarly. While working for KB, Ahn hired Bok-Dong Chung, a Samhongsa subcontractor, to assist him with the drawings for an 0-gauge die-cast model train. Ahn testified that he instructed Chung to draw the designs differently from those produced by Samhongsa.

In 1999, Lionel distributed the two die-cast trains designed and manufactured by KB. After purchasing samples of each model in July of that year, Wolf immediately traveled to Korea because he was concerned that the trains had been designed and manufactured based on MTH’s design drawings. Wolf asked Samhongsa to conduct an internal investigation, during which Chung admitted to copying Sam-hongsa designs of MTH trains and giving them to Ahn. Korean law-enforcement authorities launched an investigation into whether crimes had been committed in connection with the use of these designs. As part of that investigation, the police seized Ahn’s computer and the computers belonging to other former and current Samhongsa employees to determine whether they contained evidence of criminal activity.

The Korean authorities prosecuted Ahn, along with three other Samhongsa employees, for misappropriation of trade secrets, a criminal offense in Korea. As part of that case, Professor Kun-Woo Lee, a Korean engineering expert, compared 3307 pairs of Samhongsa and KB drawings and found substantial evidence of copying. A Korean court found Ahn and three Sam-hongsa employees guilty of the unlawful misappropriation of MTH’s trade secrets. Samhongsa then brought a civil suit in Korea against Ahn, Cho, KB, and the three Samhongsa employees. The Korean court found all of the defendants liable for misappropriation of MTH’s trade secrets. During the defendants’ preparation for an appeal of that judgment, the parties settled the civil litigation and the case was dismissed.

In 2000, MTH filed a complaint against Lionel, KB, and Yang in the United States District Court for the Eastern District of Michigan, alleging (1) misappropriation of trade secrets; (2) unfair competition by the misappropriation of products; (3) unfair competition in violation of the Lanham Act, 15 U.S.C. § 1125(a); (4) violations of Korea’s Prevention of Unfair Competition and Trade Secret Protection Law; and (5) unjust enrichment. The district court granted the defendants’ motion for summary judgment as to the count alleging a violation of Korea’s trade-secrets act, and *405 MTH withdrew its claims of unfair competition and violations of the Lanham Act. On May 12, 2004, a jury trial began on the remaining claims, i.e., misappropriation of trade secrets and unjust enrichment.

The jury returned a verdict in favor of MTH against all defendants on both claims. It awarded MTH $11,978,887 for lost profits and $13,794,518 for future lost profits. The jury also awarded an additional $12,834,820 against Lionel and $2,167,440 against KB and Yang for unjust enrichment. Further, the court granted MTH’s request for a permanent injunction, prohibiting Lionel from using the design drawings at issue. The court denied Lionel’s motion for a new trial and for judgment as a matter of law. Lionel now appeals, requesting that we reverse the district court’s order denying its motion to vacate the jury’s verdicts as to both liability and damages, and grant either its motion for a new trial or for judgment as a matter of law. Lionel also asks that we reverse the district court’s injunction.

II. STANDARD OF REVIEW

We review de novo a district court’s refusal to grant a party’s motion for judgment as a matter of law. Greene v. B.F. Goodrich Avionics Sys., Inc., 409 F.3d 784, 788 (6th Cir.2005). Under Rule 50, a court may grant judgment as a matter of law on a particular issue if “there is no legally sufficient evidentiary basis for a reasonable jury to find for that party on that issue....” Fed.R.CivJP. 50(a)(1). In cases based on diversity jurisdiction, we apply the forum state’s substantive standard for determining when judgment as a matter of law is appropriate. Greene, 409 F.3d at 788 (citing Morales v. Am. Honda Motor Co., 151 F.3d 500, 506 (6th Cir.1998)). For a party to be entitled to judgment as a matter of law after the jury enters an adverse verdict in Michigan, the evidence viewed in the light most favorable to the non-moving party must fail to establish a claim as a matter of law. Forge v. Smith, 458 Mich. 198, 580 N.W.2d 876, 880 (1998). Specifically, judgment as a matter of law is appropriate when “viewing the evidence in the light most favorable to the non-moving party, there is no genuine issue of material fact for the jury, and reasonable minds could come to but one conclusion, in favor of the moving party.” Noble v. Brinker Int’l, Inc., 391 F.3d 715, 720 (6th Cir.2004).

We review a district court’s refusal to grant a motion for a new trial under an abuse-of-discretion standard. In re Brown, 342 F.3d 620, 627 (6th Cir.2003). An abuse of discretion occurs when the district court relies on clearly erroneous findings of fact, improperly applies the law, or uses an erroneous legal standard. Id. at 633 (citing Southward v. S. Cent. Ready Mix Supply Corp., 7 F.3d 487, 492 (6th Cir.1993)). We “will find an abuse of discretion only when the Court has ‘a definite and firm conviction that the trial court committed a clear error of judgment.’ ” Engebretsen v. Fairchild Aircraft Corp., 21 F.3d 721, 728 (6th Cir.1994) (quoting Logan v. Dayton Hudson Corp., 865 F.2d 789, 790 (6th Cir.1989)). Under Rule 59, after a jury trial a court may grant a new trial “for any of the reasons for which new trials have heretofore been granted in actions at law in the courts of the United States.” Fed.R.Civ.P. 59(a). We have interpreted this Rule to require a new trial only “when a jury has reached a ‘seriously erroneous result’ as evidenced by[ ] (1) the verdict being against the weight of the evidence; (2) the damages being excessive; or (3) the trial being unfair to the moving party in some fashion, i.e., the proceedings being influenced by prejudice or bias.” Holmes v. City of Massillon, 78 F.3d 1041, 1045-46 (6th Cir.1996).

*406 III. LIABILITY ISSUES

On appeal, Lionel argues that the district court committed three errors related to Lionel’s liability: (1) allowing MTH to offer improper expert testimony; (2) determining that MTH had sufficiently identified what actual trade secrets Lionel had allegedly misappropriated; and (3) admitting into evidence various documents from the Korean court proceedings under exceptions to the rule proscribing hearsay evidence. We address these issues in turn.

A. Expert Testimony

MTH offered the testimony of Dr. Jeffery L. Stein, a professor of mechanical engineering at the University of Michigan, to establish that Lionel’s design drawings had been copied from drawings Samhong-sa completed for MTH. Stein testified that he examined two sets of design drawings, one from Lionel/KB and one from MTH/Samhongsa, concerning ten different train models, for a total of 162 pairs of drawings. Stein testified that his evaluation was based on twenty-one self-selected criteria, including factors such as the drawing’s title and the part number assigned to the drawing. Stein calculated a score from zero to one based on these criteria; zero indicated no association between the drawings, and one indicated an extremely high association. Stein testified that approximately fifty-five percent of the KB drawings were copied from Samhongsa drawings. Stein further testified that designers working independently to develop similar design drawings of the same train would not develop identical or substantially similar drawings, because detailed dimensions cannot be gleaned from examining pictures of full-scale trains, and thus it was likely that designers would assign slightly different dimensions and measurements to their drawings of the same train part. Furthermore, he testified that “even if all of the numbers were exactly the same, the way that the information is represented in the drawing, there is a certain amount of creative input that the designer has in terms of how to represent this in the drawing.” Stein also testified that designers would calculate different tolerances, i.e., the acceptable degree to which the completed part could deviate from the design dimensions:

If you make something with a very high tolerance then it’s very expensive and time consuming to produce, so it’s always a big engineering decision about when you make a tolerance as to how much it’s going to be, what is the tradeoff going to be between cost and time and the quality of the final product.

Given the delicate decisions that must be made between expense and quality, Stein testified that a company’s engineering team generally decides what the tolerance values for a given product will be.

Stein testified that he reviewed the report Professor Lee prepared, which compared 3307 drawing pairs. Stein explained that he independently corroborated Lee’s work using a regression analysis to compare it to the 162 drawing pairs Stein reviewed. Stein testified that he found copying in sixty-nine percent of the drawings, whereas Lee found copying in sixty-one percent. Stem’s regression analysis indicated that his criteria would produce conclusions about copying that were consistent with Lee’s conclusions in ninety-one percent of the drawings.

Lionel argues that (1) the district court should have excluded Stein’s testimony under Federal Rule of Evidence 702 as unreliable, and (2) even if reliable, Stein’s testimony about Lee’s conclusions was improper under Rule 703. As discussed below, we agree on both points and conclude that the errors were not harmless.

*407 1. Rule 702

A district court must determine whether expert testimony is reliable before it can be admitted under Rule 702. 1 See Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) (“Federal Rule of Evidence 702 imposes a special obligation upon a trial judge to ‘ensure that any and all scientific testimony ... is not only relevant, but reliable.’ ”) (citing Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)). In Daubert, the Supreme Court identified factors that a court may consider in determining whether an expert’s testimony is sufficiently reliable, including (1) whether a theory or technique can be or has been tested; (2) whether the technique has been subjected to peer review and publication; (3) whether the technique has a known or potential rate of error and the existence of standards controlling its operation; and (4) whether the theory or technique enjoys general acceptance in a relevant scientific community. Daubert, 509 U.S. at 593-94, 113 S.Ct. 2786; see also Nelson v. Tenn. Gas Pipeline Co., 243 F.3d 244, 251 n. 5 (6th Cir.2001). Because the Daubert factors may not be pertinent in evaluating the reliability of expert testimony in every case, they are neither definitive nor exhaustive. A court must consider whether those factors are reasonable measures of reliability in a given case, however. Nelson, 243 F.3d at 251. Moreover, even if we determine that Dr. Stein’s testimony was improperly admitted, we are not to disturb the jury’s verdict if the improper admission of that testimony was harmless. See United States v. Smithers, 212 F.3d 306, 315-17 (6th Cir.2000) (conducting harmless-error analysis after applying Daubert factors to determine whether district court erred in excluding expert testimony).

We conclude that the district court abandoned its gate-keeping function by failing to make any findings regarding the reliability of Stein’s testimony. The district court allowed Stein to testify as to his opinion of Lee’s report, but failed to make specific findings regarding the reliability of Stein’s own technique for comparing the design drawings as opposed to Stein’s technique of comparing his conclusions to those reached by Lee. Although the district court’s failure to make specific findings does not itself require that we vacate the jury’s verdict, a proper application of the Daubert factors in this case reveals that the district court abused its discretion in admitting Stein’s testimony.

First, Stein created the criteria with which he compared the design drawings; however, there is no evidence that his methodology had ever been tested, subjected to peer review, possessed a known or potential rate of error, or enjoyed general acceptance. Although it is true that “in some instances well-grounded but innovative theories will not have been published,” and that “[s]ome propositions ... are too particular, too new, or of too limited interest to be published,” Daubert, 509 U.S. at 593, 113 S.Ct. 2786, the novelty of a theory does not shield an expert’s testimony from judicial scrutiny. In Nelson, although we acknowledged that peer *408 review may not always be available in the case of novel theories or methodology, we emphasized that scrutiny by the scientific community is one of several indicators of “ ‘good science,’ in part because it increases the likelihood that substantive flaws in methodology will be detected.” 243 F.3d at 251 (quoting Daubert, 509 U.S. at 594, 113 S.Ct. 2786).

The novelty of Stein’s theory is not the only reason that his testimony is unreliable. Our conclusion is also compelled by the clear evidence that Stein lacked a rudimentary understanding of the Korean model-train design industry, and was thus unable to identify those aspects of the design drawings that might be indicative of copying. For example, one factor that Stein relied upon in evaluating the similarity between two drawings was whether the part was assigned the same number. The record clearly establishes, however, that Korean manufacturers share a common numbering system for train parts. For example, all Korean design companies assign the number “2600” to the train’s boiler. Based on his criteria, however, Stein would consider the fact that both MTH’s and Lionel’s design drawings of boilers were labeled “2600” as evidence of copying. Similarly, Stein gave weight to the name each drawing bore. Thus, if MTH’s design drawing of a boiler and Lionel’s design drawing of a boiler bore the name “boiler,” Stein interpreted that naming as a sign of copying. In addition to the obvious flaw inherent in a methodology that identifies copying by looking at the names of discrete component parts, Stein’s methodology reveals a lack of insight into this industry by considering the numbers assigned to each train part as evidence of copying.

As indicative of the lack of reliability in his methodology, Stein testified that he “arbitrarily” determined how much significance each of the twenty-one criteria would have on his assessment of whether there was evidence of copying. Among the twenty-one criteria Stein evaluated to determine whether copying was evident in a pair of drawings were part number, part name, the drawings’ dimensions, when the drawing was made, the distance from the edges to the image, and the color index. Stein testified that whether drawings had a similar part name and part number bore more heavily on his assessment than any other criterion. Even if Stein purposefully placed more significance on part name and number, as opposed to doing so “arbitrarily,” by discounting the significance of more relevant criteria, Stein’s methodology does not reflect the realities of the Korean design industry, and thus cannot be relied upon as an adequate test by which the similarities between MTH’s and Lionel’s design drawings can be evaluated.

Finally, the record indicates, and MTH concedes, that Stein not only created his report for the purposes of litigation, but that he created the precise methodology at issue for that purpose as well. We have been suspicious of methodologies created for the purpose of litigation, because “expert witnesses are not necessarily always unbiased scientists.” Turpin v. Merrell Dow Pharms., Inc. 959 F.2d 1349, 1352 (6th Cir.1992). Other courts, too, have considered whether an expert’s testimony relates to matters growing naturally and directly out of their own research, or whether the expert’s testimony has been developed for the purposes of litigation. See, e.g., Daubert v. Merrell Dow Pharms., Inc., 43 F.3d 1311, 1317 (9th Cir.1995) (noting that this is a “very significant fact to be considered”). That this methodology was created for the purposes of litigation further supports our conclusion that Stein’s testimony was not reliable under Daubert. See Nelson, 243 F.3d at 252.

*409 2. Rule 703

Lionel argues that even if Stein’s methodology were adequately reliable under Rule 702, the district court erred in allowing Stein to testify as to the conclusions drawn by Professor Lee. Rule 703 of the Federal Rules of Evidence allows an expert to rely on facts not admitted into evidence if those facts are “of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject.” Fed.R.Evid. 703. Although an expert may testify as to his or her opinion formed from facts not admitted into evidence, “[f]acts or data that are otherwise inadmissible shall not be disclosed to the jury by the proponent of the opinion or inference unless the court determines that their probative value in assisting the jury to evaluate the expert’s opinion substantially outweighs their prejudicial effect.” Id.

Although Stein testified about the conclusions he drew based on his examination of the 162 drawing pairs, he also testified extensively about the conclusions reached by Professor Lee and the degree to which Stein’s and Lee’s conclusions overlapped. MTH argues that Stein could properly testify about the similarity of his and Lee’s conclusions regarding copying within the drawing pairs, and that the methodology by which Stein compared his conclusions with Lee’s (a regression analysis) meets DaubeH’s standards of reliability. Lionel does not challenge the general use of regression analysis. Rather, Lionel correctly argues that, regardless of the methodology Stein used to compare his conclusions to those of Lee, Rule 703 simply does not allow Stein to testify as to the degree of similarity between Lee’s conclusions and his own.

Stein’s testimony as to Lee’s report and conclusions, which were not admitted into evidence, was based on hearsay. See Fed. R.Evid. 801(c) (“ ‘Hearsay’ is a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted.”). Other circuits have squarely rejected any argument that Rule 703 extends so far as to allow an expert to testify about the conclusions of other experts. See United States v. Tran Trong Cuong, 18 F.3d 1132, 1143 (4th Cir.1994) (holding that an expert witness could not bolster his opinion testimony by testifying that a non-testifying expert’s conclusions were essentially the same as the expert witness’s); United States v. Grey Bear, 883 F.2d 1382, 1392-93 (8th Cir.1989) (“We are persuaded that Fed. R.Evid. 703 does not permit an expert witness to circumvent the rules of hearsay by testifying that other experts, not present in the courtroom, corroborate his views.”). We have also held that a district court erred by admitting expert testimony that was based “upon the opinion of others who were not even qualified as experts, nor present at the trial.” Taylor v. B. Heller & Co., 9 Ohio Misc. 104, 364 F.2d 608, 613 (6th Cir.1966). Stein’s testimony regarding Lee’s conclusions was therefore improperly admitted.

3. Harmless Error

Under Rules 702 and 703, the district court abused its discretion in admitting Stein’s testimony regarding (1) his conclusions about the degree of copying between the design drawing pairs, and (2) the similarity between his conclusions and Lee’s. We may, however, overturn the jury’s verdict based on these evidentiary errors only if they were not harmless. Smithers, 212 F.3d at 317. The harmless-error standard provides that if “one cannot say, with fair assurance, ... that the judgment was not substantially swayed by the error, it is impossible to conclude that *410 substantial rights were not affected.” DePew v. Anderson, 311 F.3d 742, 751 (6th Cir.2002) (quoting Kotteakos v. United States, 328 U.S. 750, 765, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946)).

Stein’s testimony likely had a substantial effect on the verdict. Stein was the only expert to testify about the degree of copying between the drawing pairs. Without his testimony, the jury would not have learned of Lee’s conclusions, nor would they have heard testimony from any other expert witness regarding the degree of copying. Furthermore, as Lionel points out, of the 3307 drawing pairs, only eleven were presented to the jury. Because MTH relied on Stein’s testimony as its source of expertise and analysis regarding the degree of copying, it is impossible to conclude with any certainty that Stein’s testimony did not sway the jury’s verdict. We therefore remand for a new trial. And because there remains a genuine issue of material fact as to whether Lionel copied MTH’s design drawings, we also affirm the district court’s order denying Lionel’s motion for judgment as a matter of law.

B. Specificity of Trade Secrets

The first element a plaintiff alleging misappropriation of trade secrets must prove is that the information at issue actually constitutes a “trade secret.” Stromback v. New Line Cinema, 384 F.3d 283, 302 (6th Cir.2004); Rothschild v. Ford Motor Co., 2 F.Supp.2d 941, 950 (E.D.Mich.1998). For information to constitute a trade secret under Michigan law, the information must (1) derive economic value from the fact that it is not known to others who could make use of it, and (2) be the subject of efforts that are reasonable under the circumstances to maintain its secrecy. M.C.L. § 445.1902(d). Lionel does not argue on appeal that the secrecy of the information at issue was not adequately maintained; rather, it argues that MTH did not introduce evidence indicating what, precisely, constituted the trade secret.

Citing A.H. Emery Co. v. Marcan Products, 389 F.2d 11 (2d Cir.1968), MTH argues that design drawings and engineering plans are prima facie trade secrets. Id. at 16 (“[I]t is well settled that detailed manufacturing drawings and tolerance data are prima facie trade secrets.”); see also Rockwell Graphic Sys. v. DEV Indus., 925 F.2d 174, 175, 180 (7th Cir.1991). Although it concedes that the design drawings contain some non-secret information, MTH argues that the drawings contain several kinds of secret information — including tolerances (the acceptable degree to which the completed part could deviate from the design dimensions), data and reference points, clearances, pivot points, spring tensions, and specific alloys — that cannot be determined by simply measuring the parts on a manufactured engine or looking at a picture. MTH also argues that the design drawings are trade secrets even though certain elements of them may have been publicly known.

In 3M v. Pribyl, 259 F.3d 587 (7th Cir.2001), the Seventh Circuit held that a plaintiff may prevail in a trade-secrets case without identifying a specific item of information that is not publicly known or readily accessible. There, the plaintiff had a valid trade secret in its operations-and-procedures manual, despite that most of the information it contained was publicly available:

In order to be considered a trade secret, a pattern, technique, or process need not reach the level of invention necessary to warrant patent protection. A trade secret can exist in a combination of characteristics and components, each of which, by itself, is in the public domain, but the unified process, design and operation of which, in unique combination, affords a *411 competitive advantage and is a protecta-ble secret.

Id. at 595-96. The Seventh Circuit explicitly rejected the argument that Lionel advances before this Court: “[Defendants seem to suggest that if SM cannot point to specific items within its manuals that are not known by the industry, then SM cannot claim a trade secret in the combined product. We disagree.” Id. at 595. Several courts have followed the Seventh Circuit’s decision in SM. See Catalyst & Chem. Servs., Inc. v. Global Ground Support, 350 F.Supp.2d 1, 9 (D.D.C.2004), aff'd 173 Fed.Appx. 825 (Fed Cir.2006) (“[I]t is widely accepted that a trade secret can exist in a combination of characteristics each of which, by itself, is in the public domain.”) (citation omitted); Norbrook Lab. Ltd. v. G.C. Hanford Mfg. Co., 297 F.Supp.2d 463, 483 (N.D.N.Y.2003) (citing 3M, 259 F.3d at 595-96), aff'd 126 Fed.Appx. 507 (2d Cir.2005); PRG-Schultz Intern., Inc. v. Kirix Corp., No. 03-C-1867, 2003 WL 22232771, at *6 (N.D.Ill. Sept.22, 2003) (same). This line of cases is consistent with Arco Industries Corp. v. Chemcast Corp., 633 F.2d 435, 442 (6th Cir.1980), in which this Court held that under Michigan law a new combination of known steps or processes can be entitled to trade-secret protection.

As in SM, the design drawings here are properly considered trade secrets even though they contain a mixture of secret information (e.g., dimensions, tolerances, and data-reference points) and non-secret information. The Seventh Circuit’s reasoning is persuasive: When material such as design drawings or manuals are trade secrets based on a unique combination of both protected and unprotected material, a plaintiff should not be obligated to identify which components of the protected material is secret. Thus, the district court did not abuse its discretion in denying Lionel’s motion for a new trial on this ground.

C. Documentary Evidence

The district court admitted forty-five exhibits that are English translations of documents generated during the Korean investigation and judicial proceedings. These documents include judgments of the criminal convictions of Ahn and the three Samhongsa employees, a final civil judgment against KB, statements by various witnesses to Korean authorities, indictments to which Ahn and his colleagues pleaded guilty, seizure reports of allegedly misappropriated trade secrets, investigative reports, arrest notices, seizure orders for trains at issue in this suit, documents involving Cho’s travels to the United States, the civil complaint against Cho, the Korean trade-secrets statute, and an appraisal by the Korean special master. Although the district court acknowledged that the documents were hearsay, it determined that they were admissible under specific hearsay exceptions. Lionel appeals the district court’s decision to admit all forty-five exhibits. It divides the documents into two categories: the actual Korean judgments and those documents relating to the Korean proceedings.

Following United States v. Garland, 991 F.2d 328, 333 (6th Cir.1993), in which this Court held that foreign criminal judgments are admissible to prove that those judgments were in fact rendered, the district court correctly held that the judgments were admissible under Federal Rule of Evidence 201, which allows courts to take judicial notice of adjudicative facts. 2 Lion *412 el’s reading of Garland, as applying to only foreign criminal judgments, is too narrow. In that opinion, we relied on Supreme Court precedent in which the Court took judicial notice of non-criminal foreign judgments and noted that it is common for courts to take judicial notice of prior judgments and to use them as prima facie evidence of the facts stated in them. Garland, 991 F.2d at 333-34 (citing Ennis v. Smith, 55 U.S. 400, 430, 14 How. 400, 14 L.Ed. 472 (1852) (holding that foreign judgments in rem are admissible as evidence of the underlying facts)); see also Moses v. United States, 166 U.S. 571, 600, 17 S.Ct. 682, 41 L.Ed. 1119 (1897) (holding that a judgment recovered against a non-party principal by the government was properly admitted in evidence against the surety of a bond as evidence of that bond); Drummond v. Prestman, 25 U.S. (12 Wheat.) 515, 519, 6 L.Ed. 712 (1827) (holding that the record of a prior judgment in an action by plaintiff against defendant’s son “was certainly competent to prove a fact [that] every judgment is competent to prove between any parties, to wit, that such a judgment was obtained between certain parties, on a certain cause of action”); Merck’s Mortgage Co. v. Bogan, 434 F.2d 490, 493 (D.C.Cir.1970) (holding that a prior foreclosure judgment was properly admitted as prima facie evidence in the mortgage company’s suit against the mortgage guarantor); Southern Ry. Co. v. Bouknight, 70 F. 442, 448 (4th Cir.1895) (holding, in a mortgage case, that a prior judgment was admissible, “not simply as establishing the fact of its rendition, but as proof of when the action was brought, what for, and the amount”). Admitting these documents was not error.

Furthermore, the district court did not abuse its discretion in admitting the other documents. The notices of arrest, Korean complaint, and the investigative reports are generally admissible under the public-records exception, as set forth in Rule 803(8). See Fed.R.Evid. 803(8); Blue Tree Hotels Inv. (Canada), Ltd. v. Starwood Hotels & Resorts Worldwide, Inc., 369 F.3d 212, 218 (2d Cir.2004) (referring to complaints filed in state court as public records, and noting that courts may consider them in deciding motions to dismiss); United States v. Loyola-Dominguez, 125 F.3d 1315 (9th Cir.1997) (noting that warrants of deportation are public records, and leaving undisturbed the district court’s conclusion that arrest warrants are public records). 3 The Korean court records, including the indictments, are admissible under Rule 803(22), which excepts judgments of previous convictions from the general ban against hearsay. Fed.R.Evid. 803(22). Several courts have held that an indictment from a previous conviction is properly included within the scope of Rule 803(22) and is thus admissible despite being hearsay. See Maynard v. Dixon, 943 F.2d 407, 414 (4th Cir.1991) (holding that an indictment relating to a previous conviction is admissible). Most of the Korean witness statements and interrogation transcripts are admissible under Rule 804(b)(3), which allows the admission of statements made against the declarant’s interests, so long as the witness is unavailable. See Fed.R.Evid. 804(b)(3); McClung v. Wal-Mart Stores, Inc., 270 F.3d 1007, 1014-15 (6th Cir.2001) (holding that self-inculpatory and corroborated statements were admissible under Rule 804(b)(3)). Moreover, the statements made by, and the interrogation of, Ahn, Yang, and Cho *413 can be admitted under Rule 801(d)(2), which excepts from the rule against hearsay those statements made by and offered against a party. Fed.R.Evid. 801(d)(2). Finally, although we have never considered whether a search warrant is an admissible document, even if the district court erred in admitting several documents that are analogous to search warrants, that error is certainly harmless because those documents were duplicative of the Korean-litigation documents, which were properly admitted.

IV. DAMAGES

The jury awarded MTH total damages of $40,775,665, allocated as follows: (1) $11,978,887 for past lost profits; (2) $13,794,518 for future lost profits; (3) $12,834,820 for Lionel’s unjust enrichment; and (4) $2,167,440 for KB and Yang’s unjust enrichment. The first two parts of the award (for past and future lost profits, totaling $25,773,405) were imposed jointly and severally against all three defendants. Lionel argues that this violates Michigan’s ban on joint-and-several liability in tort actions. Lionel also argues that the part of the total damages based on its unjust enrichment ($12,834,820) is improperly excessive because at least part of it duplicates the award for past lost profits. Our review of a jury’s damage award is extremely deferential, and we will not order a remittitur or new trial unless the award “is contrary to all reason.” In re Lewis, 845 F.2d 624, 635 (6th Cir.1988). Nonetheless, we agree with both of these arguments and therefore reverse the district court’s denial of Lionel’s motion for a new trial on these grounds as well.

A. Joint-and-Several Liability

Based on the jury’s verdict, the district court awarded MTH $25,773,405 for past and future lost profits, to be imposed jointly and severally against Lionel, KB and Yang. The district court held that because this is a trade-secrets case, the Michigan Tort Reform Act (MTRA), M.C.L § 600.2956, which prohibits joint liability in tort actions, does not apply. 4 Because the Supreme Court of Michigan has not conclusively ruled whether the MTRA applies to the misappropriation of trade secrets, we must determine how that court would respond if confronted with the issue. See Stratienko v. Cordis Corp., 429 F.3d 592, 600 (6th Cir.2005). “In order to determine how the state supreme court would rule, we look to the decisions of the state’s intermediate courts unless we are convinced that the state supreme court would decide the issue differently.” Melson v. Prime Ins. Syndicate, Inc., 429 F.3d 633, 636 (6th Cir.2005). We review de novo a district court’s interpretation of state law. Salve Regina College v. Russell, 499 U.S. 225, 231, 111 S.Ct. 1217, 113 L.Ed.2d 190 (1991).

Michigan courts have repeatedly held that the misappropriation of trade secrets is a tort. See Bliss Clearing Niagara, Inc. v. Midwest Brake Bond Co., 270 F.Supp.2d 943, 948-49 (W.D.Mich.2003); Vic’s Quality Fruit Market, III, Inc. v. Busch’s, Inc., No. 231176, 2002 WL 1374174, at *4 (Mich.Ct.App. Jun.25, 2002). Thus, the only question before us is whether the Supreme Court of Michigan would extend the MTRA to torts beyond those involving *414 personal injury, property damage, and wrongful death.

Several Michigan Court of Appeals holdings espouse this broader reading. See Holton v. A + Ins. Assoc., Inc., 255 Mich.App. 318, 661 N.W.2d 248, 252 (2003) (“[T]he rule, by its language, appears to limit its application to three types of actions, arguably excluding plaintiffs’ action for lost insurance proceeds. However, we conclude that the rule’s applicability is not strictly limited to those three actions.”); AAA Mortgage Corp. v. Legghio, No. 239016, 2003 WL 22439665, at *14 (Mich.Ct.App. Oct.28, 2003) (noting that “M.C.L. § 600.2956 applies to all torts, and not only to those seeking damages for personal injury, property damage, or wrongful death”); K-Mart Corp. v. Logan, No. 232393, 2003 WL 21583385, at *2-3 (Mich.Ct.App. July 10, 2003) (“the phrase, ‘personal injury, property damage, or wrongful death’ modifies only the phrase ‘or another legal theory seeking damages’ ”) (citing Holton, 661 N.W.2d at 252). Although MTH points to a series of Michigan Court of Appeals opinions that read M.C.L. § 600.2956 narrowly, these cases are not persuasive. In each such case, the Michigan Court of Appeals was asked to apply M.C.L. § 600.2956 to cases involving personal injury, property damage, or wrongful death, and was thus not squarely confronted with whether the statute applies outside those particular contexts. See also Gerling Konzern Allgemeine Versicherungs AG v. Lawson, 472 Mich. 44, 693 N.W.2d 149, 151 (2005) (noting that M.C.L. § 600.2925a-d eliminated joint-and-several liability in “certain tort actions” in the context of a personal-injury action); Lamp v. Reynolds, 249 Mich.App. 591, 645 N.W.2d 311, 315 (2002) (applying M.C.L § 600.2956 to a claim for damages arising out of personal injury).

Alternatively, MTH argues that the Michigan Uniform Trade Secrets Act (“MUTSA”) sets forth its own method of allocating damages among defendants, and thus M.C.L. § 600.2956 should not apply in this case. MTH cites M.C.L. § 445.1904, which sets forth the recovery of damages for misappropriation. 5 Citing John Hancock Financial Services v. Old Kent Bank, 346 F.3d 727, 731-33 (6th Cir.2003), in which this Court held that the MTRA did not apply when an applicable provision of the UCC explicitly allocated damage liability, MTH argues that the MUTSA should similarly preclude the application of the MTRA to claims governed by the MUTSA. Old Kent Bank does not govern the outcome of this claim, however; the provision of the MUTSA that generally allows the recovery of damages in misappropriation eases is not in tension with M.C.L. § 600.2956, because that statute concerns the allocation of liability, as opposed to the kinds of damages a misappropriation plaintiff can recover. The MTRA and the MUTSA do not conflict, and thus both can and should be applied here. It was therefore error to impose the $25,773,405 award for lost profits jointly and severally against all the defendants. On remand, any award *415 for lost profits should be assessed severally only.

B. Double Recovery

Lionel argues that the damage award is excessive because the amount calculated for Lionel’s unjust enrichment ($12,834,820) duplicates much of MTH’s lost past profits (totaling $11,978,887). See American Anodco, Inc. v. Reynolds Metals Co., 743 F.2d 417, 424 (6th Cir.1984) (“[L]ike any item of damages, loss of profits may only be recovered once.”). Under M.C.L. § 445.1904, a plaintiff in a trade-secret misappropriation case can recover damages for the actual loss caused by the misappropriation, as well as the unjust enrichment caused by the misappropriation that is not taken into account by the actual-loss calculation. Michigan law, however, generally prohibits double recovery. See McAuley v. General Motors Corp., 457 Mich. 513, 578 N.W.2d 282, 285 (1998) (“Because the purpose of compensatory damages is to make the injured party whole for the losses actually suffered, the amount of recovery for such damages is inherently limited by the amount of the loss.”).

Lionel notes that MTH offered evidence that Lionel’s unjust enrichment was $16,043,525, and that the jury found MTH’s actual past profit losses were $11,978,887. Lionel argues that the portion of Lionel’s unjust enrichment not duplicating MTH’s lost past profits is at most the $4,064,638 difference between these amounts. Accordingly, Lionel asserts that because this $4,064,638 is the maximum amount of Lionel’s unjust enrichment that MTH may recover, the jury’s $12,834,820 award on this point must be reduced to $4,064,638. MTH responds that the jury was specifically instructed not to award duplicate damages, and that juries are presumed to follow such instructions. Cf. SM, 259 F.3d at 600. Although juries are presumed to follow instructions, it is quite clear that the jury did not do so here. There is no evidence in the record that Lionel’s enrichment exceeded MTH’s losses by $12,834,820, nor does MTH identify the evidence in the record that could support that award. We therefore vacate the $12,834,820 award regarding Lionel’s unjust enrichment.

Moreover, although Lionel compares the jury’s award for MTH’s lost past profits of $11,978,887 with MTH’s evidence that Lionel’s unjust enrichment was $16,043,525, the proper comparison — and one that would be in Lionel’s favor — would be with the jury’s actual award of $12,834,820 for that unjust enrichment. The maximum amount of Lionel’s unjust enrichment that MTH could have recovered as non-duplicative based on the jury’s findings regarding that unjust enrichment ($12,834,820) is only that portion of the unjust enrichment exceeding MTH’s past lost profits ($11,978,887), a net of $855,933 (not $4,064,638). On remand, any award for Lionel’s unjust enrichment may not exceed MTH’s actual past profit losses. 6

V. CONCLUSION

For the preceding reasons, we AFFIRM the district court’s order denying Lionel’s motion for judgment as a matter of law; REVERSE the district court’s orders denying Lionel’s motion for a new trial and granting MTH’s request for an injunction; and REMAND this case to the district court for further proceedings consistent with this opinion.