United States Ex Rel. Smith v. Boeing Co.

FILED United States Court of Appeals PUBLISH Tenth Circuit UNITED STATES COURT OF APPEALS June 13, 2016 Elisabeth A. Shumaker FOR THE TENTH CIRCUIT Clerk of Court _________________________________ UNITED STATES OF AMERICA, ex rel. TAYLOR SMITH; JEANNINE PREWITT; JAMES AILES, Plaintiffs - Appellants, v. No. 14-3247 THE BOEING COMPANY; DUCOMMUN, INC., f/k/a AHF - Ducommun, Defendants - Appellees. _________________________________ Appeal from the United States District Court for the District of Kansas (D.C. No. 6:05-CV-01073-MLB-KMH) _________________________________ Don R. Sampen, Clausen Miller, P.C., Chicago, Illinois (Edward M. Kay, Michael R. Grimm, Dean S. Rauchwerger, and Joseph J. Ferrini, Clausen Miller, P.C., Chicago, Illinois, Corlin J. Pratt and Terry L. Unruh, Sherwood, Harper, Dakan, Unruh & Pratt, LC, Wichita, Kansas, and William J. Skepnek, The Skepnek Law Firm, P.A., Lawrence, Kansas, with him on the briefs), for Plaintiffs-Appellants. Eric B. Wolff, Perkins Coie LLP, Seattle, Washington (Steve Y. Koh and Paul Graves, Perkins Coie LLP, Seattle, Washington, Lyndon W. Vix, Stephen E. Robison, Ron Campbell, Fleeson, Gooing, Coulson & Kitch, LLC, Wichita, Kansas, and James M. Armstrong and Jeffery A. Jordan, Foulston Siefkin, LLP, Wichita, Kansas, with him on the brief), for Defendant-Appellee. _________________________________ Before GORSUCH, MURPHY, and MORITZ, Circuit Judges. _________________________________ MORITZ, Circuit Judge. _________________________________ Three former employees of The Boeing Company, referred to as relators in this qui tam action, brought suit under the False Claims Act (FCA) against Boeing and one of its suppliers, Ducommun, Inc. The relators claimed Boeing falsely certified that several aircraft it sold to the government complied with all applicable Federal Aviation Administration (FAA) regulations, even though it knew parts manufactured by Ducommun and incorporated into the aircraft didn’t conform to FAA-approved designs. The district court granted Boeing’s and Ducommun’s respective motions for summary judgment on the relators’ FCA claims, finding no genuine dispute of material fact as to the falsity, scienter, and materiality elements of those claims. The district court also denied the relators’ motion to strike two FAA investigative reports, which the court then relied on in granting the motions for summary judgment. The relators appeal. We conclude the district court properly admitted the FAA reports under the Federal Rules of Evidence and the relators failed to establish the scienter element of their FCA claims. Accordingly, we affirm. BACKGROUND At the heart of this appeal are Boeing’s alleged violations of FAA regulations arising from aircraft Boeing sold or leased to the government. So to provide context, we first briefly review the FAA regulatory scheme governing the aircraft at issue. 2 Through the Federal Aviation Act of 1958, Congress directed the Secretary of Transportation to establish minimum standards for aircraft design, materials, workmanship, construction, and performance in order to promote the safety of air transportation in the United States. United States v. S.A. Empresa de Viacao Aerea Rio Grandense (Varig Airlines), 467 U.S. 797, 804 (1984). The Act established a multi-step certification process for new aircraft, and granted FAA authority to implement the process. Id. at 804-05. The certification process requires a manufacturer to first obtain a type certificate before it commences full-scale production of a new design. Id. at 805. The type certificate signifies FAA’s approval of the basic design of the aircraft, referred to as the type design.1 Id. at 806. The manufacturer must submit to FAA aircraft drawings, test reports, and other computations necessary to show the type design comports with all FAA regulations. Id. at 805. The manufacturer also must produce a prototype aircraft and conduct ground and flight tests on the prototype. Id. at 805-06. Once FAA is satisfied the type design meets all applicable regulations and the prototype is airworthy, FAA issues a type certificate for the proposed design. Id. 1 FAA’s regulations in force at all relevant times defined type design as consisting of, in part, (a) The drawings and specifications, and a listing of those drawings and specifications, necessary to define the configuration and the design features of the product shown to comply with the requirements of that part of this subchapter applicable to the product; [and] (b) Information on dimensions, materials, and processes necessary to define the structural strength of the product. 14 C.F.R. § 21.31 (2008). 3 The manufacturer must then obtain a production certificate before it can begin duplicating the prototype for sale. Id. at 806. This requires the manufacturer demonstrate that it has established and can maintain quality control systems that assure each aircraft it produces complies with FAA-approved type design. Id. A production certificate thus signifies the manufacturer has shown it can reliably duplicate the prototype so that each aircraft complies with the approved type design. See id. Finally, each individual aircraft receives final approval from FAA in the form of an airworthiness certificate before it’s placed into service. Id. The airworthiness certificate is FAA’s designation that the aircraft in question conforms to the type design and is otherwise in condition for safe operation. Id. This final approval is performed by either an FAA employee or, more often, an FAA-designated representative. Id. at 807. That representative—typically an employee of the manufacturer with detailed knowledge of the aircraft design—acts as a surrogate of FAA. Id. Boeing is a manufacturer of commercial aircraft. In accordance with FAA’s certification process, Boeing first obtained a type certificate in 1967 for a model 737 jet, designated as a 737-100 series jet. Boeing has since updated the 737 model and has received a type certificate for each derivative, spanning from 1967 (the 737-100 series) to 2007 (a series designated as 737-900ER). Beginning with a model designated as the 737-600 series, Boeing referred to the 737 aircraft as a “Next Generation” aircraft, or “737NG.” The 737NG differed 4 from its “737 Classic” predecessor by virtue of the methods Boeing used in the design, manufacture, and quality control of the aircraft. Whereas Boeing created the 737 Classic aircraft using traditional design and manufacturing methods, including labor-intensive, hand-directed machine tools and manual measurement and inspection of parts for purposes of quality control, Boeing used newer computer-aided technologies and advanced assembly techniques in designing and manufacturing the 737NG aircraft. Between 1997 and 2002, Boeing entered into contracts with the federal government for the manufacture and sale or lease of several 737NG aircraft. In relevant part, the contracts required Boeing to obtain for each aircraft a type certificate and an airworthiness certificate or, for some of the contracts, the military equivalent of an airworthiness certificate, called a conformity certificate. The contracts also required Boeing to maintain a production certificate for each of its production facilities. Boeing in turn contracted with Ducommun to supply more than 200 parts for the 737NG aircraft, most for the fuselage. The contracts required Ducommun to implement a manufacturing process known as Advanced Technology Assembly (ATA). ATA typically uses a Computer Numerical Control (CNC) machine to precisely locate and drill mating holes at various locations on parts of the aircraft. During assembly, these mating holes are aligned with one another and the various parts are affixed via temporary fasteners until they can be permanently joined. ATA 5 thus results in a quicker, more accurate assembly by reducing or eliminating the need to rely on bulky, labor-intensive locating tools. Ducommun initially produced ATA parts using computerized machinery to achieve first article inspection—a first production run that allowed Boeing to verify that Ducommun was complying with ATA production methods. At some point after receiving first article inspection clearance, however, Ducommun reverted to using traditional manufacturing processes. For example, Ducommun began drilling ATA holes on fuselage components using manual drill jigs. In light of suspected irregularities with Ducommun-supplied parts, Boeing conducted a tooling audit of Ducommun in 1999. The audit revealed anomalies with Ducommun’s manufacturing processes and tooling procurement procedures, potentially entitling Boeing to several million dollars in restitution. The audit team found that Ducommun was using tools that didn’t conform to Boeing specifications, using tools designed for use with CNC machinery as fixtures for manual tools, and implementing labor-intensive, hand-directed processes rather than CNC processes. These hand-directed processes included the use of ball-peen steel hammers, scribes, and hand-driven belt sanders, even though the contracts between Boeing and Ducommun required the use of CNC machines and other computerized processes. Boeing and Ducommun entered into a settlement agreement to resolve disputes arising from the tooling audit. Under the terms of the agreement, Ducommun implemented several changes to address the manufacturing irregularities, and Boeing agreed to accept parts that Ducommun produced using manual manufacturing 6 techniques rather than CNC machines in return for a three-percent price reduction. Accordingly, Ducommun continued to supply parts for the 737NG aircraft that it manufactured using drill jigs and other hand-operated tools. Boeing in turn sold or leased the 737NG aircraft that incorporated these Ducommun parts to the government. Several members of the Boeing audit team—relators Taylor Smith, Jeannine Prewitt, and James Ailes—believed that Ducommun’s manufacturing processes violated FAA regulations, particularly the 737NG’s type design. Accordingly, these members brought a qui tam action against Boeing and Ducommun under the FCA in 2002.2 In pertinent part, the FCA provides that a party can be held liable for treble damages if it “knowingly presents, or causes to be presented, a false or fraudulent claim for payment or approval” to the government, or “knowingly makes, uses, or causes to be made or used, a false record or statement material to a false or fraudulent claim.” 31 U.S.C. § 3729(a)(1)(A)-(B). Because Boeing submitted claims for payment even though it knew the aircraft didn’t comply with the 737NG type certificate, the relators argued, Boeing violated the FCA. 2 The FCA allows a private person, referred to as a relator, to bring an FCA suit on behalf of the government in what is known as a qui tam action. 31 U.S.C. § 3730(b)(1); see United States ex rel. Lemmon v. Envirocare of Utah, Inc., 614 F.3d 1163, 1167 (10th Cir. 2010). The relator serves the government with a copy of the complaint, and the government has 60 days to intervene and proceed with the action once served; otherwise, the relator conducts the action. 31 U.S.C. § 3730(b)(4). If the relator conducts the action and is ultimately successful, she receives between 25 and 30 percent of the proceeds of the action or settlement. Id. § 3730(d)(2). 7 After the relators served their 2002 complaint on the government, FAA opened a Suspected Unapproved Parts (SUP) investigation into the Ducommun parts. FAA explained that it was investigating Ducommun because the relators suspected thousands of structural parts it manufactured for Boeing aircraft didn’t conform to type design. Over a period of two years, FAA investigated the relators’ claims by conducting a surprise inspection of Ducommun’s manufacturing facilities and inspecting the allegedly unapproved parts. Because FAA found no type-design violations, it closed its investigation in 2004 and the government declined to intervene. In turn, the relators voluntarily dismissed their initial FCA action. The relators then filed another FCA action in 2005, which is the subject of this appeal. At the heart of the 2005 FCA action—and this appeal—is the relators’ allegation that Ducommun was required to manufacture ATA parts using computerized technologies, including tooling capable of collecting statistical process control (SPC) data. Because Ducommun manually drilled ATA holes using drill jigs incapable of collecting SPC data, the relators contended the Ducommun parts didn’t conform to type design. To support this allegation, the relators relied in part on two experts’ opinions that an engineering note on the 737NG drawings, flag note S3, required the use of SPC in order to conform to type design. That flag note appears on most of the 737NG’s drawings for ATA parts and states, FEATURES IDENTIFIED AS STATISTICALLY TOLERANCED SHALL BE PRODUCED WITH STATISTICAL PROCESS CONTROLS. THE DRAWING TOLERANCE APPLIES ONLY WHEN PROCESS 8 MEASUREMENTS MEET THE FOLLOWING REQUIREMENTS: 1) THE PROCESS CONTROL CHARTS SHOW THAT THE ASSOCIATED MANUFACTURING PROCESS IS IN CONTROL. 2) THE MEAN DEVIATES FROM NOMINAL NO MORE THAN 10 PERCENT OF THE SPECIFIED TOLERANCE. 3) THE MINIMUM CPK IS 1.0, WITH 90 PERCENT CONFIDENCE. WHEN THESE REQUIREMENTS ARE NOT MET, INDIVIDUAL PRODUCT MEASUREMENTS MUST FALL WITHIN +/- THIRTY PERCENT OF THE SPECIFIED TOLERANCE, CENTERED ON NOMINAL. Aplee. App. vol. 2, 346. The relators’ experts focused on the first sentence of flag note S3 (“FEATURES IDENTIFIED AS STATISTICALLY TOLERANCED SHALL BE PRODUCED WITH STATISTICAL PROCESS CONTROLS,” id. (emphasis added)) and claimed that Ducommun violated the 737NG’s type design by using hand tools— which are incapable of collecting SPC data—to drill ATA holes. And when Boeing sold or leased nearly a billion dollars’ worth of aircraft to the government with knowledge of these nonconformities, the relators contended, it submitted a false claim for payment to the government in violation of the FCA. The relators thus sought trebled damages on behalf of the government. But Boeing’s design engineers insisted the relators’ interpretation of flag note S3 was incorrect. The design engineers explained that, as with all manufacturing processes, the drilling of ATA holes—even if performed by a CNC machine—results in random variations that can’t always be controlled. The design engineers thus accounted for acceptable tolerances in the aircraft design, allowing ATA holes to vary from their nominal location by a specified distance while still conforming to engineering requirements. To meet engineering requirements and thus type design, 9 the engineers explained, each ATA hole must be located within the specified tolerance from its nominal location. The design engineers further explained that newer, computerized tooling is capable of collecting detailed data regarding the precise location of each drilled hole and its corresponding variance from the nominal location. If the collected data is normally distributed—that is, if the variances resemble a bell curve centered near the nominal location—then the process is said to be “in SPC” and there is a high probability that any variances for a given assembly will be randomly distributed around the nominal location. Accordingly, the design engineers allowed for larger tolerances if a manufacturer used sophisticated machinery capable of showing its manufacturing process was in SPC. The engineers explained that the first portion of flag note S3 was thus intended to mean that if a manufacturer chose to use the tolerances appearing on the face of the drawings, it had to demonstrate its manufacturing process was in SPC. The design engineers also indicated, however, that they understood manufacturers wouldn’t always manufacture parts using sophisticated machinery capable of collecting SPC data. So the design engineers included the second part of flag note S3, beginning with “WHEN THESE REQUIREMENTS ARE NOT MET,” id., to note that if a manufacturer didn’t utilize tooling capable of collecting SPC data, its parts had to meet tighter tolerances. The 2005 action prompted another FAA SUP investigation. Once again, FAA found no indications of nonconforming parts. The report indicated that FAA 10 interpreted flag note S3 as allowing Ducommun to use SPC for product acceptance if the process control chart showed that the process was in control, but alternatively allowed Ducommun to deviate from SPC requirements, albeit at tighter tolerances. Noting that Ducommun had sufficient controls in place resulting in parts complying with these tighter tolerances, and that none of the investigating agencies discovered any evidence supporting the relators’ allegations that the Ducommun parts violated FAA regulations, FAA recommended closing the case. The government again declined to intervene. This time the relators maintained their FCA action notwithstanding FAA’s failure to find any violations in its SUP investigation. The district court ultimately granted summary judgment for Boeing and Ducommun on the FCA claims. In doing so, the court acknowledged that the relators’ interpretation of flag note S3 was reasonable, but found the flag note could alternatively and reasonably be interpreted as allowing the use of noncomputerized machinery (such as a drill jig) at tighter tolerances. The district court concluded the conflicting indications of whether SPC was required undermined the relators’ assertion that Boeing acted with the requisite scienter necessary to sustain an FCA claim.3 Accordingly, the district court granted Boeing’s and Ducommun’s respective motions for summary judgment. 3 The district court also concluded that even if Boeing acted with the requisite scienter (that is, if it knowingly submitted a false claim for payment), the relators failed to establish any false representations were material to the government’s purchasing decision. 11 The relators appeal. They contend the district court erred by admitting the SUP reports and, in turn, by deferring to FAA’s determinations that the Ducommun parts complied with type design. They also contend the district court erroneously found a lack of falsity, scienter, and materiality with respect to the FCA claims, and thus erred in granting summary judgment.4 DISCUSSION We review the district court’s grant of summary judgment dismissing the relators’ claims de novo. United States ex rel. Burlbaw v. Orenduff, 548 F.3d 931, 944 (10th Cir. 2008). Summary judgment is appropriate “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). When considering a defendant’s motion for summary judgment on FCA claims, we accept as true the relators’ evidence and draw all reasonable inferences in their favor. Burlbaw, 548 F.3d at 944. We review a district court’s decision to admit evidence under the Federal Rules of Evidence for an abuse of discretion. United States v. Rodella, 804 F.3d 1317, 1329 (10th Cir. 2015). “Under this standard, we will not reverse unless the district court’s decision exceeded the bounds of permissible choice in the circumstances or was arbitrary, capricious or whimsical.” Id. 4 The district court also granted Boeing’s motion for summary judgment on retaliation claims brought by one of the relators, Prewitt, under an FCA retaliation provision and Kansas state law, and granted the relators’ motion to strike Robert Eastin’s deposition testimony. But the relators appeal only the rulings relating to FCA liability and denial of their motion to strike the SUP reports, and Boeing hasn’t cross-appealed the court’s ruling striking Eastin’s deposition testimony. We thus limit our discussion to the rulings concerning FCA liability and the SUP reports. 12 I. The district court neither abused its discretion in admitting the SUP reports nor abdicated its judicial function in considering those reports. A. The district court properly admitted the reports. The district court admitted both the 2004 and 2005 SUP reports under Federal Rule of Evidence 803(8)(A)(iii), an exception to the rule against hearsay that permits admission of a public record that sets out, “in a civil case or against the government in a criminal case, factual findings from a legally authorized investigation.” The relators argue the district court abused its discretion in admitting the SUP reports because they are untrustworthy. See Fed. R. Evid. 803(8)(B) (explaining public records are inadmissible if opponent demonstrates “a lack of trustworthiness”). We’ve previously provided a framework for considering the trustworthiness of public reports, explaining that the court should consider “the timeliness of the investigation; the special skill or experience of the investigator; whether a hearing was held and the level at which it was conducted; and any possible motivation problems in the preparation of the report.” Perrin v. Anderson, 784 F.2d 1040, 1047 (10th Cir. 1986). After considering these factors, the district court concluded the SUP reports were trustworthy because they weren’t untimely; the relators failed to show FAA was unqualified to make the findings; FAA considered the statements of the relators during its investigation; and FAA prepared the SUP reports in accordance with its legal obligation to investigate upon receiving notice of suspected unapproved parts, rather than for purposes of litigation. The relators nonetheless argue on appeal 13 that one or both of the reports are untrustworthy for several reasons, none of which we find persuasive. First, the relators argue the 2004 report is untrustworthy because approximately six of the 34 lines on page 3 are redacted. But the relators cite no authority suggesting partial redaction of a public record is a sign of untrustworthiness, and we decline to find so here. Next, the relators argue the reports are untrustworthy because they are based on multiple layers of hearsay. The reports reference the findings of other investigative agencies, including the Defense Criminal Investigative Service (DCIS) and FAA’s Manufacturing Inspection District Office (MIDO). But we have previously explained, in examining Rule 803(8)’s predecessor, that the rule necessarily covers “findings . . . based in part on matters outside the personal knowledge of the preparers of those reports.” Denny v. Hutchinson Sales Corp., 649 F.2d 816, 821 (10th Cir. 1981). We see no reason why FAA’s efforts to consider all relevant material in its investigation, including the reports of other investigative agencies, diminishes the trustworthiness of its findings. The relators also argue the reports are untrustworthy because FAA didn’t conduct a hearing when preparing them. But FAA did consider the relators’ claims and statements during its investigation. And although the lack of a hearing may tend to suggest untrustworthiness in certain cases, it’s just one of several factors a district court should consider in evaluating trustworthiness. See Perrin, 784 F.2d at 1047. We can’t say the lack of a hearing decisively tipped the scales in favor of 14 untrustworthiness here, particularly because FAA considered the relators’ input in preparing its reports. The relators next argue the reports are untrustworthy because the investigators lacked skill and expertise. They argue the investigators violated FAA’s own investigatory guidelines by failing to take photographs, secure evidence, and record witness statements. But even though the reports don’t contain photographs or other physical evidence, the record indicates that the investigators conducted a thorough investigation that included performing a surprise inspection of Ducommun’s facilities and physically inspecting suspect parts. We find no basis for questioning the investigators’ skill and experience in view of this thorough investigation. The relators also advance several arguments attacking the content of the reports. Specifically, the relators argue the reports are untrustworthy because (1) the reports don’t focus on the processes in place at the time of the alleged type design violations; (2) the reports don’t address the quality acceptance processes in place at Boeing or Ducommun; and (3) Boeing’s audit report contradicts the reports’ findings. But we fail to see how these arguments, which each go to the reports’ alleged lack of credibility, demonstrate the reports’ lack of trustworthiness for purposes of admission under Rule 803(8)(A)(iii). See Moss v. Ole S. Real Estate, Inc., 933 F.2d 1300, 1307 (5th Cir. 1991) (explaining that a report’s trustworthiness—which is a question of the report’s reliability—is the test for admission under the rule, not its credibility— 15 which is a question of what weight the factfinder should give the report once admitted).5 Accordingly, we agree the relators have not met their burden in “show[ing] that the source of information or other circumstances indicate a lack of trustworthiness.” Fed. R. Evid. 803(8)(B). The district court thus didn’t abuse its discretion in admitting the reports under Rule 803(8)(A)(iii). See Rodella, 804 F.3d at 1329. B. The district court didn’t abdicate its judicial function when it considered the reports in ruling on FCA liability. The relators next argue that the district court abdicated its judicial function by deferring to FAA’s position in the SUP reports. First, they contend the court erred in concluding that FAA had primary jurisdiction over the matter and that this allegedly erroneous finding “colored its entire summary judgment analysis.” Aplt. Br. 24. Second, they argue the district court erroneously found FAA’s regulatory authority over Boeing insulated Boeing from FCA liability. And third, the relators contend the district court erroneously gave FAA’s findings preclusive weight. 5 In any event, we find these additional arguments meritless. Although the 2005 Report limits some of its findings to current manufacturing processes, it also states that investigators reviewed historical shop orders and records, suggesting the investigators reviewed the manufacturing processes in place at the time of the alleged violations. The reports also concluded that Boeing’s Material Review Board (MRB) recorded and approved any deviation from type design, and that flag note S3 provides an exception to using SPC for product acceptance, indicating the investigators did consider quality acceptance processes. And the audit report, which was directed to tooling irregularities, didn’t address flag note S3 and whether Ducommun’s tooling violates FAA regulations, and thus didn’t contradict the SUP reports’ findings. 16 But we see nothing in the district court’s opinion indicating the court believed Boeing was somehow insulated from judicial review. And even assuming the district court erred in concluding FAA had primary jurisdiction, we see nothing that suggests it erroneously deferred to the agency’s SUP report findings on that basis or applied the reports in a preclusive manner. Indeed, the opposite is true. If the district court had concluded that FAA’s regulation insulated Boeing from liability or that FAA’s prior conclusions controlled the matter, these conclusions likely would’ve ended the district court’s inquiry. But the district court expressly declined to decide whether FAA’s finding of regulatory compliance precluded the relators’ claims. Instead, the district court discussed and applied each element of the FCA claims at length, including a 13-page discussion of the scienter element. Thus, the relators’ contention that the district court treated FAA as “running interference” for Boeing, see Aplt. Br. 27, lacks merit. In short, the district court neither abused its discretion in admitting the SUP reports under Rule 803(8)(A)(iii), nor abdicated its judicial function by considering the reports when evaluating FCA liability. II. The district court didn’t err in granting summary judgment on the relators’ FCA claims because there is no evidence Boeing knowingly presented a false claim to the government for payment. In pertinent part, the FCA makes any person liable who (1) “knowingly presents, or causes to be presented, a false or fraudulent claim for payment or approval” to the government, or (2) “knowingly makes, uses, or causes to be made or used, a false record or statement material to a false or fraudulent claim.” 31 U.S.C. 17 § 3729(a)(1)(A)-(B). Thus, it is not enough to simply submit a false claim. To give rise to liability under the FCA, the submitted claim must be both knowingly and materially false. Id. § 3729(a)(1)(A)-(B), (b)(1); see Burlbaw, 548 F.3d at 952-53 (explaining that “[t]he proper focus of the scienter inquiry under § 3729(a) must always rest on the defendant’s ‘knowledge’ of whether the claim is false”); United States ex rel. Thomas v. Black & Veatch Special Projects Corp., --- F.3d ----, ----, 2016 WL 1612857, at *7 (10th Cir. Apr. 22, 2016) (explaining “an FCA plaintiff may establish materiality by demonstrating that the defendant violated a contractual or regulatory provision that ‘undercut the purpose of the contract[],’” or, “where a defendant violates only a tangential or minor contractual provision, . . . by coming forward with evidence indicating that, despite the tangential nature of the violation, it may have persuaded the government not to pay the defendant” (first alteration in original) (quoting Lemmon, 614 F.3d at 1169)). A claim is false for purposes of the FCA if it’s either factually or legally false. United States ex rel. Conner v. Salina Reg’l Health Ctr., Inc., 543 F.3d 1211, 1217 (10th Cir. 2008). A payee makes a factually false claim by either (1) submitting an incorrect description of the goods or services provided; or (2) requesting reimbursement for goods or services never provided. Id. And a payee makes a legally false claim by certifying compliance with a statute or regulation as a condition to government payment but knowingly failing to comply with that statute or regulation. Id. 18 Moreover, there are two forms of legally false certification—express and implied. Lemmon, 614 F.3d at 1168. Express false certification occurs when a government contractor falsely certifies compliance with a particular statute, regulation, or contract term and compliance is a prerequisite to payment. Id. Implied false certification occurs when a government contractor doesn’t expressly certify compliance, but knowingly and falsely implies that it is entitled to payment when it submits a claim. Id. at 1169. In this case, the relators allege both express and implied false certification.6 Generally, the relators assert Boeing submitted an express legally false claim by certifying the aircraft complied with FAA regulations despite its knowledge that Ducommun’s processes didn’t conform to type design. And they assert Boeing submitted an implied legally false claim by knowingly failing to comply with FAA regulations including the 737NG type design even though it knew compliance with the regulations was a prerequisite to government payment.7 6 On April 19, 2016, the Supreme Court heard argument in Universal Health Services, Inc. v. United States ex rel. Escobar, No. 15-7. At issue there is whether implied false certification is a viable theory under the FCA. Because we conclude the relators haven’t established a triable issue with respect to implied false certification—whether viable or not—the Court’s ultimate resolution in Escobar won’t affect our holding. 7 The relators also assert Boeing made a factually false claim because Boeing allegedly knew before executing its first government contract that Ducommun was unable to meet the computerized requirements of type design, and thus knew Ducommun couldn’t satisfy the contracts’ requirements. But an FCA violation requires presenting a false claim for payment. See 31 U.S.C. § 3729(a)(1)(A)-(B); Conner, 543 F.3d at 1217. The relators haven’t established that Boeing’s mere knowledge—prior to entering into any contracts with the government—is actionable under the FCA. 19 But the relators haven’t shown the requisite scienter necessary to establish FCA liability. The relators must show more than a falsehood—they must show that Boeing knowingly presented a false claim for payment. 31 U.S.C. § 3729(a)(1)(A)- (B); Burlbaw, 548 F.3d at 945. Although “no proof of specific intent to defraud” is required, the relators must at least show that Boeing acted with “reckless disregard of the truth or falsity of the information.” 31 U.S.C. § 3729(b)(1) (defining “knowingly” as acting with “actual knowledge of the information,” or with “deliberate ignorance . . . or . . . reckless disregard of the truth or falsity of the information”); see also Burlbaw, 548 F.3d at 945. Here, the relators present no evidence that anyone at Boeing knew the Ducommun parts didn’t comply with FAA regulations—or, alternatively, was deliberately ignorant of, or acted with reckless disregard to, FAA violations—yet submitted a claim to the government for payment anyway. The relators’ naked assertions, devoid of any evidence of scienter, can’t survive summary judgment. Indeed, the relators’ claims are analogous to those brought in Burlbaw. There, we held the relators’ accusations that a university knowingly misrepresented that it was a minority institution in order to receive Department of Defense (DOD) set-aside contracts couldn’t survive summary judgment because the relators hadn’t put forth any evidence showing anyone from the university knowingly provided a false certification. Burlbaw, 548 F.3d at 948-51. We accepted as true the premise that the university didn’t meet the DOD criteria and thus any certification to the contrary was false. Id. at 937. But we nonetheless concluded summary judgment was appropriate 20 because, among other reasons, the relators failed to show anyone from the university knew the university didn’t satisfy the DOD criteria. Id. at 949-51. As in Burlbaw, we are “struck” here “by what is not in the record.” Id. at 949. Even if we assume that the 737NG aircraft Boeing sold to the government didn’t comply with FAA regulations, there are simply no facts in the record supporting the relators’ contention that Boeing knew about the nonconformities when submitting the claims for payment. The relators appear to recognize this deficiency, but they argue the 737NG’s type design so clearly required Ducommun to manufacture the 737NG parts using CNC machines and SPC that Boeing’s submission of the aircraft for payment was “knowingly false as a matter of law.” Aplt. Br. 54. For example, the relators suggest that type design so clearly required computerized processes that Boeing must have known the aircraft didn’t conform to the type certificate when it incorporated Ducommun’s hand-fabricated parts into the 737NG aircraft. In support, the relators principally rely on flag note S3. Flag note S3 provides that parts identified as statistically toleranced, which include the ATA parts Ducommun produced, “SHALL BE PRODUCED WITH STATISTICAL PROCESS CONTROLS.” Aplee. App. vol. 2, 346 (emphasis added). The relators contend use of the word “shall” required Ducommun to use CNC machines capable of SPC in order to comply with type design, and Boeing thus must have known its claims were false when submitted for payment. 21 The relators proffer the opinion of their expert, Dr. Michael Dreikorn, who testified that because the “first sentence of flag note S3 contains the term ‘shall,’ which denotes a nondiscretionary requirement,” the flag note “specifically requires the application of [SPC] to verify product conformity.” Aplt. App. vol. 17, 3628-29. The relators also point to the opinion of another expert, Pat Duggins, who similarly concluded that flag note S3’s use of the word “shall” required suppliers to produce the part using SPC. Id. at 3853. Although their interpretation of flag note S3 is at the heart of their scienter argument, the relators also offer other evidence that they insist shows Boeing’s knowledge. For example, the relators point to the tooling audit report, which suggests Ducommun’s manufacturing processes violated its contracts with Boeing, and to the declarations of each relator rehashing these findings. The relators also point to testimony of various witnesses who believed it was mandatory for Ducommun to implement computerized manufacturing and quality control processes. And they argue that, by definition, the “Next Generation” designation signified the 737NG aircraft differed from its 737 Classic counterpart and required the use of computerized technologies during manufacture. The problem with these arguments is that FAA and the design engineers who wrote flag note S3 disagree with this interpretation of the note. For example, the authors of flag note S3 explained that they included the additional condition, “WHEN THESE REQUIREMENTS ARE NOT MET, INDIVIDUAL PRODUCT MEASUREMENTS MUST FALL WITHIN +/- THIRTY PERCENT OF THE 22 SPECIFIED TOLERANCE, CENTERED ON NOMINAL,” see Aplee. App. vol. 2, 346, to explain that different methods of manufacture would result in different tolerances. One of the authors, Michael Kuss, stated that using drill jigs was and remains common in the airline industry, but because drill jigs are not conducive to data collection for SPC purposes, flag note S3 required tighter tolerances when using these jigs rather than sophisticated machinery. Kuss explained: We included the provision (“When these requirements are not met . . .”) in the text of the flag note to explain that different methods of manufacture, depending on whether SPC data was generated, would result in different tolerances. Flag note S3 was not meant to require SPC in every instance, and in my view, the text states that fact clearly. Aplt. App. vol. 3, 603 (omission in original). A co-author of the flag note, Robert Atkinson, similarly explained: I am told that certain experts in this case have asserted that flag note S3 mandates automated hole drilling using [a CNC] machine and the collection and use of statistical data. That is incorrect. As the note itself explains, there are three requirements for suppliers to take advantage of the wider tolerances offered by [SPC]. But “[w]hen these requirements are not met,” the product must measure within the thirty percent tighter tolerance. As one of the authors of flag note S3, it is simply wrong to claim that automated drilling and statistical processes were required. Id. at 607 (third alteration in original). Likewise, FAA concluded that the latter portion of flag note S3 allows for a deviation from the SPC requirements if the parts meet tighter tolerances than those shown on the face of the drawing. Contemporaneous Boeing design documents also support this interpretation of flag note S3. For example, the ATA design guide in effect at the relevant time 23 explained the purpose of the “S” series flag notes (including flag note S3) was to provide different tolerances depending on the manufacturing process elected: When statistical tolerancing is used on an engineering drawing, the corresponding arithmetic tolerances may also be shown. The statistical tolerances will be identified with an “S” series Flag Note. If Manufacturing elects to build to statistical tolerances rather than arithmetic tolerances, the part features must be fabricated using statistical process controls; and Quality Assurance shall accept/reject parts based on statistical acceptance methods. Aplt. App. vol. 9, 2101-02 (emphasis added). Thus, at best, the evidence shows conflicting opinions regarding whether Ducommun was required to use computerized manufacturing and quality control processes when manufacturing parts for the 737NG aircraft. In light of these conflicting opinions, we reject the relators’ contention that their interpretation is so indisputably correct as to render Boeing’s certifications “knowingly false as a matter of law.” See Aplt. Br. 54. And because the relators offer no other evidence that anyone at Boeing knowingly made a false certification, we conclude no genuine dispute of material fact exists as to whether Boeing acted with the requisite scienter to trigger FCA liability. Because this conclusion is fatal to the relators’ claims, we need not address the remaining elements of falsity and materiality. CONCLUSION The district court didn’t abuse its discretion in admitting the two FAA SUP reports. Nor did the court abdicate its judicial responsibility by considering those reports at the summary judgment stage. And because we find no evidence that Boeing 24 knowingly submitted a false claim to the government for payment, we conclude the district court properly granted summary judgment to Boeing and Ducommun. Affirmed. 25