DIANA GRIBBON MOTZ, Circuit Judge:

This case returns to us again, this time on appeal from an adverse jury verdict. Dr. Jon Oberg, as relator for the United States, brought this qui tam action against four student loan corporations, including the Pennsylvania Higher Education Assistance Agency ("PHEAA"). He alleged that the corporations had defrauded the Department of Education and so violated the False Claims Act ("FCA"), 31 U.S.C. § 3729 et seq.

Over the course of several appeals, we affirmed the dismissal of one defendant and two of the other defendants settled. See United States ex rel. Oberg v. Pa. Higher Educ. Assistance Agency, 804 F.3d 646, 650 (4th Cir. 2015); United States ex rel. Oberg v. Pa. Higher Educ. Assistance Agency, 745 F.3d 131, 145 (4th Cir. 2014); United States ex rel. Oberg v. Ky. Higher Educ. Student Loan Corp., 681 F.3d 575, 579-81 (4th Cir. 2012). The case proceeded to trial only against PHEAA. Oberg now appeals the jury's unanimous verdict in favor of PHEAA. For the reasons that follow, we affirm.

I.

Oberg's claim concerns a Department of Education subsidy program meant to encourage the issuance of low-interest federal student loans. It did so by offering Special Allowance Payments ("SAPs") to certain qualifying lenders. 20 U.S.C. § 1087-1. For one particular category of loans — those financed through tax-exempt bonds — Congress guaranteed lenders a 9.5 percent return. Id. § 1087-1(b)(2)(B). In the low-interest environment of the mid-2000s, this guaranteed rate made tax-exempt bonds a particularly attractive investment vehicle.

To take advantage of the favorable return offered by the program, Oberg claims that between 2002 and 2006, PHEAA submitted false claims for SAP subsidies by improperly transferring student loans from non-tax-exempt bonds into tax-exempt bonds. In doing so, PHEAA converted lower-interest floating-rate loans into loans that guaranteed a 9.5 percent return. This translated into millions of dollars in additional revenue for PHEAA.

During a five-day trial, the court admitted more than 100 exhibits and the jury heard testimony from more than a dozen witnesses. After deliberating for less than three hours, the jury returned a unanimous verdict in favor of PHEAA.

Oberg timely noted this appeal, asking that we vacate the judgment and remand for a new trial. He maintains that the district court substantially impeded his ability to prove his claims by improperly excluding critical evidence and rejecting Oberg's proposed jury instructions.

II.

Oberg first contends that the district court erred by excluding certain evidence at trial. We review a district court's evidentiary rulings for abuse of discretion. SAS Inst., Inc. v. World Programming Ltd., 874 F.3d 370, 384 (4th Cir. 2017). "A district court abuses its discretion if it relies on an error of law or a clearly erroneous factual finding." EEOC v. Freeman, 778 F.3d 463, 466 (4th Cir. 2015).

At trial, Oberg sought admission of a 2004-2007 Performance Audit of PHEAA performed by the Pennsylvania Auditor General. The Audit "evaluate[d] PHEAA's performance in improving access to higher education for Pennsylvania residents" and concluded that PHEAA had largely "failed its mission." The Audit found that PHEAA paid excessive salaries and bonuses to its executives and managers. The Audit also catalogued and strongly criticized PHEAA's lavish spending on employee benefits and "extravagant" expenditures on other unnecessary expenses. It "concluded that PHEAA was governed and managed within a culture that sometimes allowed self-reward to supersede fiscal prudence." The district court excluded the Audit as irrelevant under Federal Rule of Evidence 401.

Oberg contends that the Audit was relevant for several reasons. First, he argues that the Audit's critical findings tended to establish scienter — i.e., that desire for personal gain motivated PHEAA officers to submit false claims. This argument fails because unlike the securities fraud cases on which Oberg relies,1 FCA claims require a relator to show only that the defendant had knowledge of the illegality of its actions, rather than specific intent to defraud. See, e.g., United States ex rel. Harrison v. Westinghouse Savannah River Co., 352 F.3d 908, 921 (4th Cir. 2003) ("In establishing liability under the FCA, a plaintiff need not prove the defendant had a financial motive to make a false statement relating to a claim seeking government funds."); United States ex rel. K & R Ltd. P'ship v. Mass. Hous. Fin. Agency, 530 F.3d 980, 984 (D.C. Cir. 2008) (finding evidence of defendant's "motive to submit false claims — the need to bail itself out of financial trouble — could not ... support a finding of knowledge, be it actual, deliberate ignorance, or reckless disregard"). As the district court correctly explained: "It doesn't really make any difference whether they were operating well or not well or whatever. The only issue in this case is: Did they commit fraud and file a false claim?"

Oberg next maintains that the Audit would have allowed him to rebut PHEAA's own improper argument that its management acted with the "benevolent motive" to benefit borrowers in carrying out its scheme. Oberg lodged little objection to this evidence at trial. Indeed, Oberg himself elicited most of the "benevolent motive" testimony through his questions to PHEAA's management. In any event, whether PHEAA's management had benevolent motive was a collateral issue of limited relevance, making any error harmless. See Smith v. Balt. City Police Dep't, 840 F.3d 193, 200-01 (4th Cir. 2016).

Finally, Oberg contends that the Audit would have helped him attack the credibility of PHEAA's executives. He argues that because witness credibility is "always at issue," any "evidence concerning a witness's credibility is always relevant." United States v. Green, 617 F.3d 233, 251 (3d Cir. 2010). This argument also fails, for Oberg questioned PHEAA's executives on the company's compensation practices and elicited much of the same information contained in the Audit. Moreover, because nothing in the Audit contradicted the trial testimony of PHEAA executives, its admission would hardly have aided Oberg in impeaching the executives' credibility.

In sum, the district court did not abuse its discretion in excluding the Audit as irrelevant.2

III.

Oberg's remaining claims center on the district court's refusal to give several of his proposed jury instructions.

A.

Before we consider the merit of his claims, we must first determine our standard of review.

PHEAA contends that because Oberg did not object to the jury instructions before the district court, he has failed to preserve his challenge to them and so we can review only for plain error. Oberg counters that the district court rejected his properly requested instructions in a "definitive ruling on the record," and so we must instead review for abuse of discretion. Fed. R. Civ. P. 51(d)(1)(B); see also BMG Rights Mgmt. (US) LLC v. Cox Commc'ns, Inc., 881 F.3d 293, 305 (4th Cir. 2018).

To resolve this question, we turn to the language of Federal Rule of Civil Procedure 51(d)(1)(B), which provides:

Fed. R. Civ. P. 51(d)(1) (emphasis added).

This provision was added to Rule 51 in 2003. See Fed. R. Civ. P. 51 advisory committee's note to 2003 amendment (hereinafter, 2003 Advisory Committee Note). The Advisory Committee explained that the change was motivated by concern over the majority view "that a proper request for a jury instruction [was] not alone enough to preserve the right to appeal failure to give [an] instruction." Id. The Committee concluded that this rule was "appropriate when the court [had not] sufficiently focused on the request" or "believe[d] that the request ha[d] been granted in substance although in different words." Id. But in other circumstances, the doctrine created "a trap for the unwary who fail[ed] to add an objection after the court ... made it clear that the request [had] been considered and rejected on the merits." Id. To correct this, the drafters added the ruling-on-the-record language to "establish[] authority to review the failure to grant a timely request, despite a failure to add an objection, when the court has made a definitive ruling on the record rejecting the request." Id.

Notwithstanding the 2003 Amendment, some courts have seemed to continue to follow a more stringent preservation rule than Rule 51(d)(1)(B) now appears to command. See EEOC v. New Breed Logistics, 783 F.3d 1057, 1075 (6th Cir. 2015); C.B. v. City of Sonora, 769 F.3d 1005, 1032 (9th Cir. 2014); Consumer Prod. Research & Design, Inc. v. Jensen, 572 F.3d 436, 439 (7th Cir. 2009); Colon-Millin v. Sears Roebuck De P.R., Inc., 455 F.3d 30, 41 (1st Cir. 2006); Collins v. Alco Parking Corp., 448 F.3d 652, 656 (3d Cir. 2006).

In our view, the Rule's text and the Advisory Committee's direction compel a different approach. For a court's rejection of a proposed instruction to constitute "a definitive ruling on the record," the record must simply provide a reviewing court with a sufficient basis from which to determine the district court's rejection was "on the merits" — in other words, that the court rejected the substance of the proposed instruction, not merely the litigant's choice of words. And, to be "definitive," this rejection must be final rather than tentative.

It remains true, of course, that "the mere tendering of a proposed instruction will not preserve error for appeal." Bunn v. Oldendorff Carriers GmbH & Co. KG, 723 F.3d 454, 469 (4th Cir. 2013) (internal quotation marks omitted). But where a litigant proposes an instruction and the district court's final ruling on the record demonstrates that the court rejected it on the merits, the claim of error is preserved. This approach hews closely to the language and purpose of Rule 51(d)(1)(B) while remaining cognizant of the important interests embodied in requiring parties to raise objections in the district court. See Booker v. Mass. Dep't of Pub. Health, 612 F.3d 34, 41 (1st Cir. 2010).

Applied here, we must conclude that Oberg did not preserve his challenge to the failure to give his proposed jury instructions. At the charge conference, the district court outlined its proposed jury instructions in broad terms, explaining that it would give standard instructions on issues like the burden of proof and reasonable doubt, and would also instruct the jury on the essential elements of an FCA claim. PHEAA raised several clarifying questions. Oberg sought only to ensure that the instructions contained a provision not at issue on appeal, i.e., that proof of reckless disregard sufficed to establish an FCA claim. When the court called for any final comments, PHEAA asked whether the court would read any of the instructions Oberg had submitted before the charge conference. The court responded that it was "not reading anybody's instruction[s]" and it had "the areas" it was "going to instruct on." Oberg did not object. After instructing the jury, the court once again asked whether the parties had any objections to the instructions. PHEAA raised two; Oberg responded: "[n]o objection from the relator."

Oberg argues that the district court's statement that it would "not read anybody's instruction[s]" constitutes a "definitive ruling on the record" rejecting his instructions on the merits. Fed. R. Civ. P. 51(d)(1)(B). This argument falls far short. Even assuming that the court's asserted "denial" — a brief response to a question by PHEAA, not Oberg — constituted a final ruling that foreclosed further discussion of the issue, we cannot discern from the record whether the court considered Oberg's proposed instructions and rejected them for their substance, or simply believed that its own instructions were an adequate substitute. To preserve the issue, Oberg could have sought clarification as to the court's basis for rejecting his instructions, or he could have timely objected. Because he did neither, we review only for plain error.3

We note that the Tenth Circuit also recently addressed the 2003 amendment to Rule 51. There, appellants claimed to have preserved their challenge to jury instructions by submitting proposed instructions and receiving a responsive email from the district court with the court's tentative instructions, which did not incorporate their proposals. First Am. Title Ins. Co. v. Nw. Title Ins. Agency, 906 F.3d 884, 894-95 (10th Cir. 2018). The Tenth Circuit observed that the district court did not suggest the email included its final instructions and that counsel had ample opportunity to discuss its proposed instructions. Id. at 895. It thus held that the email did not constitute a "definitive" rejection as contemplated by Rule 51(d)(1)(B). Id.4 Our holding here does not conflict with the Tenth Circuit's opinion, but that case does not provide as clear a parallel as PHEAA suggests. The district court's statement that it would not read any party's instruction was arguably more final than that considered by the Tenth Circuit, but we cannot conclude that it was "on the merits," and so it fails to satisfy the Rule for that reason.

B.

Because Oberg failed to preserve his objections to the jury instructions, we review only for plain error. Oberg must show that in refusing to give his proposed instructions (1) the district court erred; "(2) the error is plain; (3) the error affects substantial rights; and (4) ... the error seriously affects the fairness, integrity or public reputation of judicial proceedings." Gregg v. Ham, 678 F.3d 333, 338 (4th Cir. 2012) (internal quotation marks omitted). Even if his claim had been properly preserved, Oberg "faces a heavy burden, for we accord the district court much discretion to fashion the charge." United States v. Raza, 876 F.3d 604, 614 (4th Cir. 2017) (internal quotation marks omitted).

Oberg first argues that the district court erred by failing to inform the jury that he did not need to prove PHEAA had a "specific intent to defraud" the government for his FCA claim to succeed. Oberg's proposed instruction did accurately restate a statutory element of the FCA, 31 U.S.C. § 3729(b)(1)(B), but the instructions given by the district court "substantially covered" this issue. United States ex rel. Drakeford v. Tuomey, 792 F.3d 364, 382 (4th Cir. 2015). The court's instructions correctly listed the three ways in which the jury could find that PHEAA acted with the requisite knowledge, none of which required Oberg to show specific intent to defraud. The court did not err, much less plainly err, in declining to provide any additional instruction. See Noel v. Artson, 641 F.3d 580, 590 (4th Cir. 2011) ("We have always left the choice between generality versus specificity in the charge to the sound discretion of the trial courts." (internal quotation marks and alteration omitted)).

Next, Oberg argues that the district court should have included an instruction stating that "benevolent motivation... is not a defense" to an FCA claim. Much like Oberg's previous objection, the instructions given by the court substantially covered this instruction. Moreover, Oberg's proposed instruction addresses only a collateral issue — attempting to "cure" any improper introduction of benevolent motive evidence — that would not have affected the outcome in the district court. See Gregg, 678 F.3d at 338 (requiring that proposed instruction must affect substantial rights to establish plain error).

Finally, Oberg contends that the court erred by refusing to instruct the jury on how it could find PHEAA's claims fraudulent. He asserts that because the court did not provide the jury with the specific statutory language defining whether PHEAA's claims were eligible for the 9.5 percent SAP payments, the jury lacked the necessary tools to decide an element of his FCA claim. Had the district court erred in its charge on this issue, it would have committed the type of error that would usually require reversal, even on plain error review. See United States v. Jennings, 160 F.3d 1006, 1019 (4th Cir. 1998) (noting that "failure to instruct the jury on any essential element of an offense constitutes plain error").

But yet again, the district court's instructions substantially covered the substance of Oberg's proposal. Contrary to Oberg's assertions, the court's instructions sufficiently explained that the jury had to consider whether PHEAA's claims were "false or fraudulent," and that the crux of Oberg's argument was that PHEAA "was not entitled to the 9.5 percent floor SAP on the claimed loans." Although the district court did not detail the precise steps the jury should follow to determine the eligibility of PHEAA's SAP claims, "where, as here, the instructions accurately covered all the issues in the case, the failure to reference specific aspects of a party's contentions cannot serve as a basis for a finding of error." Noel, 641 F.3d at 590 (internal quotation marks and alterations omitted).

In contrast, Oberg's proposed instruction provided a misleading framework that presumed his interpretation of the relevant statute and DOE's regulations was correct, essentially requiring the jury to find PHEAA's claims ineligible. See id. ("[W]e cannot fault the district court for declining to give a wink and a nod to the jury.").

Armed with extensive testimony and numerous exhibits, it was up to the jury to determine whether PHEAA's claims were eligible and, if they were not, whether PHEAA knew so when it submitted its claims. This is precisely what the district court's jury instructions asked of them.

IV.

For the foregoing reasons, the judgment of the district court is

AFFIRMED.