Tom Kondash v. Kia Motors Am., Inc

NOT RECOMMENDED FOR FULL-TEXT PUBLICATION File Name: 19a0150n.06 Case No. 18-3181 UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT FILED Mar 28, 2019 TOM KONDASH, ) DEBORAH S. HUNT, Clerk ) Plaintiff-Appellee, ) ) ON APPEAL FROM THE UNITED v. ) STATES DISTRICT COURT FOR ) THE SOUTHERN DISTRICT OF KIA MOTORS AMERICA, INC., and KIA ) OHIO MOTORS CORP., ) ) Defendants-Appellants. ) ) BEFORE: GRIFFIN and DONALD, Circuit Judges; and BERTELSMAN, District Judge. BERNICE BOUIE DONALD, Circuit Judge. In this interlocutory appeal, the parties dispute whether the district court properly declined to seal records that allegedly contain trade secrets. Because we find that the district court did not abuse its discretion in determining not to seal the documents and did not err in determining that the documents did not contain trade secrets, we AFFIRM. I. Plaintiff Tom Kondash brought a class action lawsuit against Defendants Kia Motors America, Inc. and Kia Motors Corporation (collectively, “Kia”), claiming that the panoramic  The Honorable William O. Bertelsman, United States District Judge for the Eastern District of Kentucky, sitting by designation. Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al sunroofs in certain Kia vehicles are defective because the glass breaks without warning. In support of his motion for class certification, Kondash sought to file numerous internal documents he had received from Kia during discovery. Pursuant to the protective order put in place for the discovery process, Kondash initially sought to file the documents under seal. However, the district court denied Kondash’s motion, directing that the documents be filed unsealed. Kia filed an emergency motion for reconsideration, arguing that many of its records contained trade secrets. In response, the district court ordered Kondash to file the records conditionally under seal. The district court then held an evidentiary hearing, at which a witness for Kia testified that the records were confidential and contained trade secrets. At the end of the hearing, the district court ordered the parties to jointly review the documents and identify anything that both parties agreed contained trade secrets. The court ordered the parties to analyze in detail, document by document, the propriety of secrecy, and provide reasoning and legal citations to support their recommendations. After conferring, the parties withdrew the request to seal as to some of the records, agreed that some records should be filed unsealed but with redactions and some should be sealed entirely, and disagreed as to whether some of the documents should be sealed. The parties presented their positions to the district court for review. Upon review, the court accepted the withdrawn requests and ordered those records to be filed unsealed. The court then reviewed the documents the parties agreed should be filed unsealed but with redactions and concluded that the redactions were appropriate. Next, the district court ordered sealed certain documents that it found contained trade secrets. As to the records on which the parties disagreed, the district court largely determined that the documents did not contain trade secrets and ordered those files filed unsealed. Kia now appeals the district court’s decision to file the disputed documents unsealed. -2- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al II. Kia contends that the district court erred in finding the disputed documents should be unsealed, setting forth two arguments: (1) the district court applied the incorrect standard for sealing records; and (2) the district court erred in holding that the disputed documents did not contain protectable trade secrets. We address each of these arguments in turn. A. Standard of Review for Sealing Documents We review decisions of the district court to seal court documents or records, as well as orders lifting or modifying a seal, for abuse of discretion. Meyer Goldberg, Inc. v. Fisher Foods, Inc., 823 F.2d 159, 161 (6th Cir. 1987). “An abuse of discretion exists when the district court applies the wrong legal standard, misapplies the correct legal standard, or relies on clearly erroneous findings of fact.” First Tech. Safety Sys., Inc. v. Depinet, 11 F.3d 641, 647 (6th Cir. 1993) (citing Fleischut v. Nixon Detroit Diesel, Inc., 859 F.2d 26, 30 (6th Cir. 1988)) A court’s discretion to seal records is bounded by a “long-established legal tradition” of the “presumptive right of the public to inspect and copy judicial documents and files.” In re Knoxville News–Sentinel Co., Inc., 723 F.2d 470, 474 (6th Cir. 1983). Because of the importance of the rights involved, we have held that “the district court’s decision [to seal documents] is not accorded the traditional scope of ‘narrow review reserved for discretionary decisions based on first-hand observations.’” Id. at 476 (quoting United States v. Criden I, 648 F.2d 814, 818 (3d Cir. 1981)). As a result, “[o]nly the most compelling reasons can justify non-disclosure of judicial records.” Id. (internal citations omitted). The party seeking to seal records has the heavy burden of overcoming the “strong presumption in favor of openness.” Shane Grp., Inc. v. Blue Cross Blue Shield, 825 F.3d 299, 305 (6th Cir. 2016) (citing Brown & Williamson Tobacco Corp. v. F.T.C., 710 F.2d 1165, 1180 (6th -3- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Cir. 1983)). To meet this burden, the party must show three things: (1) a compelling interest in sealing the records; (2) that the interest in sealing outweighs the public’s interest in accessing the records; and (3) that the request is narrowly tailored. See id. (citations omitted). Where a party can show a compelling reason for sealing, the party must then show why those reasons outweigh the public interest in access to those records and that the seal is narrowly tailored to serve that reason. Id. To do so, the party must “analyze in detail, document by document, the propriety of secrecy, providing reasons and legal citations.” Id. at 305-06 (citing Baxter Int’l, Inc. v. Abbott Lab., 297 F.3d 544, 548 (7th Cir. 2002)). The presumption in favor of public access is strong when public safety is implicated. See Brown & Williamson Tobacco Corp. v. F.T.C., 710 F.2d 1165, 1180-81 (6th Cir. 1983) (vacating the district court’s sealing order because the litigation potentially involved the information regarding the true contents of cigarettes). “It is well-established that confidentiality provisions, protective orders, and the sealing of cases are appropriate litigation tools in some circumstances. However, the interests of public health and safety will often outweigh any confidentiality interests that might be implicated.” NHTSA Enforcement Guidance Bulletin 2015-01: Recommended Best Practices for Protective Orders and Settlement Agreements in Civil Litigation, 81 Fed. Reg. 13026- 02, 13027 (March 11, 2016). This is particularly true in class actions, where, because of the interest of a broader public outside of the named parties, the standards for overcoming the presumption of openness “should be applied [. . .] with particular strictness.” Shane Group, Inc. v. Blue Cross Blue Shield of Michigan, 825 F.3d 299, 305 (6th Cir. 2016) (quoting In re Cendant Corp., 260 F.3d 183, 194 (3d Cir. 2001)). -4- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Where a district court finds that sealing is appropriate, the court must set forth its specific findings and conclusions “which justify nondisclosure to the public.” Id. at 306 (quoting Brown & Williamson Tobacco Corp. v. F.T.C., 710 F.2d 1165, 1176 (6th Cir. 1983)). Kia contends that the district court erred by impermissibly including in its trade secret determination consideration of whether access to the documents was of public importance. This is not the case. The district court properly conducted its analysis by first determining whether a trade secret existed and then, after finding no trade secret, explained the public’s interest in access to the relevant documents. See, e.g. Kondash v. Kia Motors America, Inc., No. 1:15-cv-506, 2018 WL 770418, at *4 (S.D. Ohio Feb. 7, 2018) (“Kia has not demonstrated that the reports and investigation materials are trade secrets. . . . Furthermore, the documents’ contents are of great public interest, particularly to drivers concerned about their safety and class members interested in their legal claims in this case.”). Not only was it proper for the district court to address whether the information was of public importance, it was a necessary step in the analysis. While the existence of a trade secret will generally satisfy a party’s burden of showing a compelling reason for sealing documents, even if a trade secret does not exist, a court may still find a compelling reason exists; further, even if a district court finds that a trade secret exists, it must still determine whether public interest outweighs the moving party’s interests in protecting their trade secret. See Shane Group, Inc. 825 F.3d at 308 (“‘[I]n civil litigation, only trade secrets, information covered by a recognized privilege (such as attorney-client privilege), and information required by statute to be maintained in confidence (such as the name of a minor victim of a sexual assault),’ is typically enough to overcome the presumption of access.”) (emphasis added) (quoting Baxter, 297 F.3d at 546). -5- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Though the district court’s discussion was sparse, it is clear from the opinion that the court determined whether the documents contained trade secrets independent of its consideration regarding the public’s interest in the information. Accordingly, we find that the district court applied the correct legal standard and therefore did not abuse its discretion. B. Trade Secret Determination Next, Kia argues that the district court erred in determining that the documents at issue contained no trade secrets. Under Ohio law, a trade secret is defined as “information . . . that satisfies both of the following: (1) It derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use[; and] (2) [i]t is the subject of efforts that are reasonable under the circumstances to maintain its secrecy.” Ohio Rev. Code § 1333.61(D)(1)-(2). “This Court reviews for clear error the district court’s factual conclusion that [no trade secrets exist].” Hickory Specialties, Inc. v. Forest Flavors Int’l, Inc., No. 99-5003, 2000 WL 687681 at *7 (6th Cir. May 19, 2000) (citing Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 474 (1974)); see also Kewanee Oil Co., 416 U.S. at 474 (1974) (noting that the Sixth Circuit had reviewed a finding that a trade secret existed for clear error). The Ohio Supreme Court has set forth six factors to be considered in determining whether an item constitutes a trade secret: “(1) The extent to which the information is known outside the business; (2) the extent to which it is known to those inside the business . . .; (3) the precautions taken by the holder of the trade secret to guard the secrecy of the information; (4) the savings effected and the value to the holder in having the information as against competitors; (5) the amount of effort or money expended in obtaining and developing the information; and (6) the amount of time and expense it would take for others to acquire and duplicate the information.” Heartland Home Fin., Inc. v. Allied Home Mortg. Capital Corp., 258 F. App’x 860, 861-62 (6th Cir. 2008) (quoting State ex rel. Plain Dealer v. Ohio Dep’t of Ins., 687 N.E.2d 661, 672 (1997)). -6- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al A finding as to one factor is not dispositive, but “‘[a] business or possessor of a potential trade secret must take some active steps to maintain its secrecy in order to enjoy presumptive trade secret status.’” Id. at 682 (quoting State ex rel. Plain Dealer, 687 N.E.2d at 672). The fact that a document will reveal “competitively-sensitive financial and negotiating information” is not an adequate justification for sealing—rather, “the proponents of closure bears the burden of showing that ‘disclosure will work a clearly defined and serious injury.’” Shane Group, Inc., 825 F.3d at 307 (quoting In re Cendant Corp., 260 F.3d 183, 194 (3d Cir. 2001)). Kia has not shown that the district court committed clear error in determining the disputed documents do not contain trade secrets. Although the district court’s explanation was sparse, it is evident the court applied Ohio’s standard for determining whether a trade secret exists. See, e.g. Kondash, 2018 WL at *4 (“Kia has not demonstrated that the reports and investigation materials are trade secrets. . . . [T]he Court fails to see how these documents constitute a ‘procedure’ or ‘process’ that derives independent economic value from not being known and is the subject of efforts to maintain its secrecy. . . . The knowledge that ceramic paint weakens tempered glass is already known throughout the automotive industry.”). Additionally, Kia contends that the district court erred in its factual determinations because it distinguished testing for product development and manufacturing from testing in response to a problem; but its cited authority does not support Kia’s proposition. Kia cites to two cases where Ohio courts found that a company’s testing and analysis of its product performance post-market entrance are trade secrets. See Procter & Gamble Co. v. Stoneham, 747 N.E.2d 268, 271-72 (Ohio Ct. App. 2000) (noting that a former P&G employee had knowledge of, among other things, the company’s plans for “revitalization of existing products,” “the perceived weakness of products,” and relevant “market research” and finding that the employee was “privy to massive amounts of -7- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al information that constituted trade secrets”); Valco Cincinnati, Inc. v. N&D Machining Service, Inc., 492 N.E.2d 814, 817 (Ohio 1986) (finding the company’s materials and manufacturing processes were trade secrets because they were unique and “developed only after a great deal of experimentation, testing and field experience”). However, neither case dealt with investigative reports created in response to, and for the purpose of, investigating an incident that resulted in litigation. Further, in both cases, the court found that the Ohio’s six-factor test weighed in favor of the proponent of sealing the records because, among other reasons, the information was not widely known, and the company had taken steps to keep the information private. In contrast, the district court in this case found the opposite; a finding that Kia has not demonstrated was clearly in error. Finally, although Kia cites to four cases outside the Sixth Circuit where the same type of information—internal investigation processes pertaining product issues that resulted in litigation—were considered trade secrets, we are not bound by those decisions, especially where, as here, the district court specifically applied Ohio-trade-secret law, and those cases did not.1 III. Because the district court applied the correct standard when determining whether to seal the disputed documents and did not commit clear error in determining whether the disputed documents contained trade secrets, we AFFIRM the district court’s decision. 1 See McDonnell v. Southwest Airlines Co., 292 F. App’x 679, 680 (9th Cir. Sep. 11, 2008) (district court did not abuse its discretion in sealing documents showing “procedures and communications geared toward investigating the cause of the airline crash”); Velasco v. Chrysler Grp. LLC, No. CV 13-08080 DDP, 2017 WL 388797, at *3 (C.D. Cal. Jan. 26, 2017) (sealing “technical data pertaining to efforts to understand failures in [the product that] are inextricably intertwined with information about Chrysler’s technical capacity”); Grayson v. Gen. Elec. Co., No. 3:13-cv-1799, 2017 WL 923907, at *3 (D. Conn. Mar. 7, 2017) (same); McCrary v. Elations Co. LLC, No. EDCV 13-0242, 2014 WL 12561600, at *5-*6 (C.D. Cal. Dec. 8, 2014) (same). -8- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al GRIFFIN, Circuit Judge, dissenting. In my view, the district court committed multiple legal errors and thus abused its discretion. Given these errors, I would vacate the district court’s order and remand for a ruling under the correct legal standards. I respectfully dissent. I. Kia’s challenge takes us to the intersection of two robust legal interests: (1) the public’s interest in access to court records and (2) a party’s interest in protecting its trade secrets. Access to Court Records. The public has a strong interest in knowing what information court records contain. Brown & Williamson Tobacco Corp. v. F.T.C., 710 F.2d 1165, 1180 (6th Cir. 1983). Such knowledge allows the public to assess the conduct giving rise to a lawsuit, the lawsuit’s result, and the result’s merits. Shane Grp., Inc. v. Blue Cross Blue Shield, 825 F.3d 299, 305 (6th Cir. 2016). This interest brings with it a “strong presumption in favor of openness.” Id. (quoting Brown & Williamson, 710 F.2d at 1179). And this presumption imposes a heavy burden on a party seeking to seal court records. Id. (citing In re Knoxville News-Sentinel Co., 723 F.2d 470, 476 (6th Cir. 1983)). To meet that burden, a party must show three things. First, it must show a compelling interest in sealing the records. Id. This requires a party to provide, for each record, detailed reasons why the record warrants sealing and legal citations to support those reasons. Id. Second, the party must show why its interest in sealing the records outweighs the public’s interest in accessing them. Id. at 307. And third, the party must show that its request is narrowly tailored. Id. Trade Secrets. The concept of trade secrets originated in England in the early 1800s and came to the United States a few decades later. See Sharon K. Sandeen & Christopher B. Seaman, Toward a Federal Jurisprudence of Trade Secret Law, 32 Berkeley Tech. L.J. 829, 835 (2017). In an era of industrialization, trade secrets protected proprietary manufacturing knowledge. Id. -9- Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Since then, much has changed. The source of trade-secret law has expanded—starting in the common law and moving to the Restatement of Torts, state law, and eventually federal law. See id. at 835–53. The scope of what constitutes a trade secret has expanded as well; it now comprises financial, business, scientific, technical, economic, and engineering information. See 18 U.S.C. § 1839(3). With this expansion has come a three-part test for determining whether a record includes trade secrets. The record must contain the right type of information. Id. The party that owns the record must take reasonable measures to keep the information secret. Id. at § 1839(3)(A). And the information must derive independent economic value from not being generally known. Id. at § 1839(3)(B). Intersection of the Interests. Whether a record contains trade secrets and whether it should be sealed are distinct inquiries. The former requires factual findings; the latter requires legal ones. Thus, when a party asks a court to seal records because they contain trade secrets, the court must conduct a bifurcated analysis. First, the court must make factual findings as to which types of information the records contain, whether the party takes reasonable measures to keep the information secret, and whether the information derives value from its secrecy. Then, the court must reach legal conclusions as to whether the party has a compelling interest in sealing the records, whether that interest outweighs the public’s interest in accessing them, and whether the party has narrowly tailored its request. The answer to the first inquiry affects the answer to the second. Trade secrets are an exception to the presumption in favor of open court records. Shane Grp., 825 F.3d at 308; Brown & Williamson, 710 F.2d at 1180. But the exception is neither automatic nor universal. The existence of a trade secret within a record shows that a party has a compelling interest in sealing it - 10 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al that outweighs the public’s interest in accessing it but speaks not to whether the request to seal is narrowly tailored. Consider, for example, a lengthy internal memorandum that mentions a single trade secret in a single sentence. A request to seal the entirety of the record would lack narrow tailoring because redaction could preserve the trade secret while providing access to the rest of the record. Put differently, a party may not invoke the trade-secret exception to shield from the public eye other, less sensitive information. The Result. After Kondash filed the records conditionally under seal, Kia asked the district court to keep many of them sealed because they contained investigation findings, pricing information, government communications, and other sensitive data Kia believed were trade secrets. Regarding the records the court ordered unsealed, it grouped them into five categories, each of which it analyzed separately. Kondash v. Kia Motors America, Inc., No. 1:15-cv-506, 2018 WL 770418, at *4–5 (S.D. Ohio Feb. 7, 2018). For some categories, it discussed at length whether it would seal the records. Id. For others, it provided only a few sentences devoid of legal support. Id. at *4. Unlike the majority, I would hold that the district court, in ruling as it did, abused its discretion by committing four legal errors. First, the district court conflated the trade-secret inquiry with the records-sealing inquiry. The majority would have us think otherwise, but they misrepresent the district court’s decision. They say “[t]he district court properly conducted its analysis by first determining whether a trade secret existed and then, after finding no trade secret, explained the public’s interest in access to the relevant documents.” For support, they excerpt this quote from the district court’s opinion: Kia has not demonstrated that the reports and investigation materials are trade secrets. . . . Furthermore, the documents’ contents are of great public interest, particularly to drivers concerned about their safety and class members interested in their legal claims in this case. - 11 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Notice the ellipsis. It omits four sentences. And those sentences reveal the error: Kia has not demonstrated that the reports and investigation materials are trade secrets. Unlike in [Proct[e]r & Gamble Co. v. Ranir, LLC, No. 1:17-CV-185, 2017 WL 3537195 (S.D. Ohio Aug. 17, 2017)], Kia does not propose narrowly-tailored [sic] redactions to these documents. More importantly, however, the Court fails to see how these documents constitute a “procedure” or “process” that derives independent economic value from not being known and is the subject of efforts to maintain its secrecy. Unlike in [Procter & Gamble Co. v. Stoneham, 747 N.E.2d 268 (Ohio Ct. App. 2000)], there is no evidence that these reports were created with the significant outlay of time and money. Rather, all appear to be routine forms or documents, or reports generated in response to potential litigation over sunroof shattering. Under the logic presented by Kia, virtually all internal documents would fall under the category of a “trade secret.” Furthermore, the documents’ contents are of great public interest, particularly to drivers concerned about their safety and class members interested in their legal claims in this case. Id. at *4 (emphasis added). The district court begins by stating a conclusion: Kia’s investigation records don’t contain trade secrets. One reason it gives for this conclusion is that the records don’t contain information that derives independent economic value from not being generally known. To be sure, that’s part of a trade-secret inquiry. 18 U.S.C. § 1839(3)(B). But the first reason the district court gives is that Kia failed to propose narrowly tailored redactions. And that’s not part of a trade-secret inquiry; it’s part of a record-sealing one. See Shane Grp., 825 F.3d at 307. The district court made the same error elsewhere as well. For example, it said: The knowledge that ceramic paint weakens tempered glass is already known throughout the automotive industry. To the extent the document focuses on how much of the Kia sunroof is covered in ceramic paint, that information can be discerned by looking at a vehicle. Thus, sealing is not appropriate. - 12 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Kondash, 2018 WL 770418, at *4. This ties the sealing inquiry to whether the information was publicly known, which is a part of the trade-secret inquiry. The district court also said: The warranty and/or goodwill payments are not the type of pricing and negotiation information that would reveal highly sensitive data about Kia. Rather, the goodwill payments and information about warranty payments are extremely relevant to this action and thus of great public interest. The Court, therefore, will unseal these documents. Id. at * 5. This ties the data’s sensitivity, which is a part of the trade-secret inquiry, to the public’s interest in accessing it, which is a part of the sealing inquiry. The district court further said: The Court is persuaded that dealer price information is private. Therefore, the Court documents should be unsealed except that the dealer price information may be redacted[.] Id. This ties the sealing inquiry to whether the information was publicly known, which is a part of the trade-secret inquiry. Finally, the district court said: Kia asserts this information is confidential because it includes ball[- ]drop[-] testing results and the documents are treated as confidential by law. The Court finds this investigative testing as a result of the shattering incidences is distinguished from testing performed while developing a new product. As such [sic], the Court will not seal [the exhibits]. Id. This ties the sealing inquiry to whether the information is the trade-secret type. These portions of the district court’s decision don’t comport with the bifurcated, sequential analysis the law requires. First, a court must ask whether a record contains trade secrets. To do so, it must examine the information the records contain, the steps the record’s owner takes to keep the information secret, and what value, if any, that information derives from being secret. 18 U.S.C. § 1839(3). Then, with the answer to the trade-secret inquiry in mind, a court must ask whether it should seal the records. To do so, the court must examine the interest in nondisclosure of the party seeking sealing, the public’s interest in accessing the record, and the scope of the - 13 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al request to seal. Shane Grp., 825 F.3d at 305–07. By mixing and matching the prongs of these distinct inquires, the district court erred. The district court’s second legal error occurred when it ruled that Kia’s records didn’t contain trade secrets without making the factual findings that the law requires. As discussed above, that inquiry involves findings as to which types of information the records contain, whether the records’ owner takes reasonable measures to keep that information secret, and whether the information derives value from its secrecy. See 18 U.S.C. § 1839(3). If a court concludes that records fail as to one of the factors, it need not address the others. But it may not skip the process entirely, which is what the district court did. Of fifteen potential factual findings, it made one: Independent Reasonable economic value Right type of measures to keep from not being information? information secret? generally known? Internal- Investigation Records Ceramic-Paint Records Warranty, Goodwill, and No. Parts-Data Records Vehicle-Package- Pricing Records Government- Communication Records Kondash, 2018 WL 770418, at *4–5. To be sure, the district court more than once mentioned whether the records contained private information. See, e.g., id. at *4 (“The knowledge that ceramic paint weakens tempered glass is already known throughout the automotive industry. To the extent the document focuses on how much of the Kia sunroof is covered in ceramic paint, that - 14 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al information can be discerned by looking at a vehicle.”) (emphasis added); id. at *5 (“The warranty and/or goodwill payments are not the type of pricing and negotiation information that would reveal highly sensitive data about Kia.”) (emphasis added); id. at *5 (“The Court is persuaded that dealer price information is private.”) (emphasis added). But a trade secret must be more than private; it must derive independent economic value from its privacy. See 18 U.S.C. § 1839(3)(B). So to the extent the district court made factual findings as to the privacy of Kia’s records, those findings, alone, could not establish the existence of a trade secret. To analyze whether the district court correctly declined to seal the records because they contained no trade secrets (which would mean Kia lacked a compelling interest in filing them under seal), then, we would need to make numerous factual findings in the first instance. This we may not do. As an appellate court, our task is to review the district court’s factual findings, not to supply our own findings when the district court hasn’t made them. See, e.g., Taglieri v. Monasky, 907 F.3d 404, 408 (6th Cir. 2018) (en banc) (noting the different roles that district courts and appellate courts play when it comes to fact finding). Yet my colleagues overlook this glaring deficiency. The district court erred a third time when it created a distinction between pre-sale and post- sale product testing that lacks any basis in law or logic. For example, the court ordered unsealed one record, which contained Kia’s projection of how many sunroofs will shatter in the future, because the court saw a difference between “product development and manufacturing” and the response to “product problems.” Kondash, 2018 WL 770418, at *5. The court also rejected Kia’s request to keep sealed two records related to sunroof testing because the court characterized the testing as “investigative” rather than testing for “developing a new product.” Id. I can find no legal support for this distinction. “[T]he term ‘trade secret’ means all forms and types of financial, - 15 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al business, scientific, technical, economic, or engineering information,” 18 U.S.C. § 1839(3) (emphasis added), not just what a company creates before a product’s release. Indeed, this distinction makes little sense. Like many companies, Kia offers warranties for its products. Warranties create ongoing customer relationships. To maintain those relationships, Kia must respond to customer dissatisfaction. And Kia has an interest in responding better than its competitors do: each time Kia turns a frown upside down, the more likely the company is to earn the customer’s repeat business. The majority rejects Kia’s arguments regarding the district court’s testing distinction because Kia hasn’t cited binding case law directly on point. Although the lack of such authority tells us we are not required to side with Kia, it says nothing about whether we should. If a statute’s plain language is clear (e.g., “‘trade secret’ means all forms and types of financial, business, scientific, technical, economic, or engineering information”), we cannot abandon our interpretive duties because another court hasn’t previously ruled on the issue. The district court’s fourth legal error came when it rejected some of Kia’s arguments without explanation. Kia sought to keep sealed records relating to warranty repairs and goodwill payments (which cover repairs of out-of-warranty vehicles) because the information revealed failure rates for parts and the company’s approach to addressing customer dissatisfaction outside the warranty period. The district court dismissed these arguments without discussing why they lacked merit; it simply “reject[ed]” them. Kondash, 2018 WL 770418, at *5. When we review a district court’s exercise of its discretion, we must have enough of an explanation to meaningfully review its decision. See, e.g., Mich. State AFL-CIO v. Miller, 103 F.3d 1240, 1248 (6th Cir. 1997) (faulting a district court for failing to explain the basis for its decision). A district court must do more than state the rule and result, id., and here the district court didn’t even state the rule. - 16 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al My colleagues neglect this error as well, rejecting Kia’s arguments in the same perfunctory, conclusory fashion as the district court. They say “it is clear from the opinion that the court determined whether the documents contained trade secrets independent of its consideration regarding the public’s interest in the information.” But do they explain why that is so? No. They also say “it is evident the Court applied Ohio’s standard for determining whether a trade secret exists.” Not only is Ohio law likely the wrong law to apply—as I explain below—but the majority yet again relies on a bare, unsupported conclusion. II. Next, we need to address the district court’s explanation for its decision to permit the redaction of some records and to keep others sealed. The parties agreed to redact or seal certain exhibits, and the district court accepted those compromises without much scrutiny. Kondash, 2018 WL 770418, at *2–3. As to the records that would be redacted, the district court explained what they were (dealer invoices and design drawings) and then said only that it was “satisfied that the[] redactions are narrowly-tailored [sic] and appropriate.” Id. at *3. And as to the exhibits that would remain sealed, the district court again described what they were (technical-design details and replacement-parts revenue information) and said that the technical records would remain sealed because they contained “information about engineering modifications, Kia’s engineering requirements, and related design elements” and that the revenue records would remain sealed because “the dollar figures depicted are not publicly known.” Id. Although the majority does not address this ruling, I would, and I would hold that the district court abused its discretion in four ways. First, when examining Kia’s technical records, the district court again conflated the trade-secret standard with the records-sealing one. Whether a record contains technical information matters when determining whether it contains trade secrets but, standing alone, says nothing about whether the record should be sealed. - 17 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al Second, when examining Kia’s revenue records, the district court premised its ruling on an incomplete view of what constitutes a trade secret. That information is private, standing alone, makes not a trade secret; the information must derive value from being private. 18 U.S.C. § 1839(3)(B). It must also be the right type of information, and it must be subject to reasonable measures to maintain its privacy. Id. at § 1839(3), (3)(A). Although a record may warrant sealing even if it doesn’t contain trade secrets, Kia never argued that its revenue records should remain sealed for other reasons. Nor did the district court say that the private nature of the information was itself a compelling reason to keep the records sealed—much less one that outweighed the public’s interest in seeing them. And even had the district court said as much, that the records are private says nothing about the scope of Kia’s request. Third, the district court failed to meet its burden of explanation. As we recognized in Shane Group, a court that seals records must provide “specific findings and conclusions” to justify its decision. 825 F.3d at 306. This obligation is independent of whether a party objects to sealing the records. Id. And it requires the court to address “why the interests in support of nondisclosure are compelling, why the interests supporting access are less so, and why the seal itself is no broader than necessary.” Id. A court’s failure to meet this obligation is grounds to vacate an order to seal. Id. Here, the district court addressed neither Kia’s interest in nondisclosure nor such an interest’s weight relative to the public’s interest in access. The court did say Kia’s request was narrowly tailored, but it never explained why that was so. The district court’s stated basis for its ruling was therefore “brief, perfunctory, and patently inadequate.” Id. Fourth, to the extent that the district court meant to rest its decision on the existence of trade secrets within the records, the district court again failed to make the required factual findings. Nowhere did the district court determine whether Kia took reasonable steps to keep the information - 18 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al private or whether the information derived independent economic value from not being generally known. 18 U.S.C. § 1839(3). Those are findings we may not make in the first instance. III. Finally, the district court, the parties, and now my colleagues have erred in assuming Ohio law governs whether Kia’s documents contain trade secrets. In the past, state law governed civil trade-secret inquires by default because there was no federal civil trade-secret law. But that changed in 2016 with the enactment of the Defend Trade Secrets Act, 18 U.S.C. § 1836. Thus, both federal law and Ohio law now define “trade secret” for the purposes of civil lawsuits. This overlap requires us to look to the Erie doctrine, which establishes that federal courts sitting in diversity jurisdiction apply state substantive law and federal procedural law. Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78 (1938); see also Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415, 427 (1996). Whether federal law or Ohio law applies to the trade-secret inquiry, then, depends on whether the issue is substantive or procedural. To identify the nature of a legal issue, the Supreme Court has adopted an “outcome- determination test,” which asks whether it would “significantly affect” a lawsuit’s outcome if a federal court applied federal law instead of state law that would have controlled had the lawsuit remained in state court. Gasperini, 518 U.S. at 427–28 (quoting Guaranty Trust Co. v. York, 326 U.S. 99, 109 (1945)). In other words, the outcome of a case should be “substantially the same, so far as legal rules determine the outcome of a litigation, as it would be if tried in a State court.” Guaranty Trust, 326 U.S. at 109. When applying this test, courts must consider Erie’s twin aims: “discouragement of forum-shopping and avoidance of inequitable administration of the laws.” Gasperini, 518 U.S. at 428 (quoting Hanna v. Plumer, 380 U.S. 460, 468 (1965)). Here, the issue appears to be procedural. Kondash has not sued Kia for trade-secret misappropriation; his claims are for negligence, breach of express warranty, and breach of implied - 19 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al warranty. And how he files Kia’s records—under seal or in the open—seems unlikely to affect the outcome of the case. He will still be able to file the records, and the district court will still consider them. So sealing records will leave untouched how the district court (or eventually a jury) resolves any ultimate issues in the case. Erie’s twin aims further suggest that the issue is procedural. Unless a party has the devious ulterior motive of forcing its opponent to disclose trade secrets, it would have no reason to forum shop—whether documents must be filed under seal is simply too far removed from a case’s merits. Even if a party was so devious, it isn’t clear that federal law defines “trade secret” advantageously. Every state except New York has adopted the Uniform Trade Secrets Act. See Legislative Bill Tracking - Trade Secrets Act, https://tinyurl.com/y87nmbf5, (last visited Mar. 13, 2019) (listing every state but New York and Massachusetts as having adopted the Act); H.B. 4868, 190th Gen. Ct. (Mass. 2018), subsequently codified at Mass. Ann. Laws ch. 93, § 42 (LexisNexis) (adopting the UTSA). And the Defend Trade Secrets Act and the Uniform Trade Secrets Act define “trade secret” similarly: both refer to similar types of information; both require the information to be subject to reasonable measures to keep it secret; and both require the information to derive value from not being generally known. Compare 18 U.S.C. § 1839(3) with UTSA § 1(4)(i) (available at https://tinyurl.com/ybqres3e (last visited Mar. 13, 2019)). Given these similarities, the distinction between federal law and state law might be one without a difference. If that is so, application of federal law would neither encourage forum shopping nor lead to inequitable administration of the law because both federal and state law would compel the same result. So despite the majority’s mechanical application of Ohio law, federal law appears to apply. That said, the parties have never had the opportunity to express their views on the issue. Thus, - 20 - Case No. 18-3181, Kondash v. Kia Motors Am., Inc., et. al because I would remand the case for application of the correct legal standard, I would also leave it to the district court to determine in the first instance which law applies. * * * In sum, I would hold that the district court abused its discretion by: (1) conflating the trade- secret standard with the record-sealing one; (2) failing to make the factual findings that a trade- secret analysis requires; (3) inventing an illogical distinction between pre-sale and post-sale product testing; (4) rejecting some of Kia’s arguments without explanation; (5) relying exclusively on a factor that, alone, answers neither the trade-secret inquiry nor the records-sealing one; and (6) failing to explain its basis for redacting some records and keeping others sealed. This does not mean the district court necessarily erred in determining what to keep sealed, to unseal, or to redact. Nor does this mean Kia necessarily met its burden of justifying its request to keep the records sealed. Instead, I would remand to the district court to apply the correct legal standard. - 21 -