United States v. Greenlaw

Before Stewart, Dennis, and Southwick, Circuit Judges. Carl E. Stewart, Circuit Judge: In January 2022, a jury convicted United Development Funding (“UDF”) executives Hollis Greenlaw, Benjamin Wissink, Cara Obert, and Jeffrey Jester (collectively “Appellants”) of conspiracy to commit wire fraud affecting a financial institution, conspiracy to commit securities fraud, and eight counts of aiding and abetting securities fraud. 18 U.S.C. §§ 1343, 1348, 1349 & 2. Jurors heard evidence that Appellants were involved in what the Government deemed “a classic Ponzi-like scheme,” in which Appellants transferred money out of one fund to pay distributions to another fund’s investors, without disclosing this information to their investors or the Securities Exchange Commission (“SEC”). Appellants did not refute that they conducted these transactions. They instead pointed to evidence that their conduct did not constitute fraud because it amounted to routine business transactions that benefited all involved without causing harm to their investors. On appeal, they urge this court to view this evidence as proof that they did not intend to deprive their investors of money or property as a conviction under the fraud statutes requires. Appellants each filed separate appeals, challenging their convictions on several grounds. Considered together, they argue that (1) the jury verdict should be vacated because the evidence at trial was insufficient to support their convictions or alternatively, (2) they are entitled to a new trial because the jury instructions were improper. As explained below, Appellants have demonstrated at least one error in the jury instructions—the intent to defraud instruction. Because this error was harmless, and thus, does not warrant a new trial, we also address Appellants’ remaining challenges on the merits. Appellants also argue that the district court erred in (3) limiting cross-examination regarding a non-testifying government informant; (4) allowing the Government to constructively amend the indictment and include certain improper statements in its closing argument; (5) imposing a time limit during trial; and (6) failing to apply the cumulative-error doctrine. Because these arguments also do not warrant a new trial, we AFFIRM the jury verdict in its entirety. I. Background UDF finances residential real estate developments, which entails buying land, building the infrastructure, and selling lots or homes built on those lots. Each phase of this development cycle increases the value of the real estate and, in turn, the developer profits when the finished product is sold for more than the costs of development. Real estate developers typically need loans to finance these construction projects, and when they do, they can call UDF. Greenlaw co-founded the company,[1] which offers a group of investment funds—UDF III, UDF IV, and UDF V—to support the developments at each stage of the process.[2] In return for its loan to developers, the investment fund receives liens on the land, and developers are required to pay back the loans with interest. The money from the interest is then disbursed to the funds’ investors as distributions. A. The Trial i. Evidence of Undisclosed Advances Evidence presented at trial revealed that, between January 21, 2011 and December 29, 2015, the process in which Appellants paid UDF III investors their distributions changed. Developers were not paying back the loans quick enough, so UDF III was short on funds to pay its investors’ its “general rate of return” of 9.75% per year. Even though this rate was not promised, it was advertised in marketing materials to “broker/dealers and financial advisors.” To remedy this, Appellants transferred money, by way of an advance, from UDF IV and UDF V to UDF III to cover the distributions, maintain a high distribution rate, and ensure that UDF III continued to appear lucrative to the investing public. A Federal Bureau of Investigation (“FBI”) forensic accountant testified that $66.8 million was transferred to UDF III from UDF IV and UDF V during the relevant period. Jurors heard evidence about how UDF was able to conduct these transactions. UDF asset manager, Jeff Gilpatrick, testified that developers, like Centurion and Buffington, that had loans from UDF entities would generally submit a request when they needed an advance, and the approval for the advance would come from Jester or Wissink. But emails revealed that, as the date of each distribution drew near, Appellants diverted money from UDF IV and UDF V to UDF III unbeknownst to the developers. As a means to do this, Appellants relied on a clause in the loan agreement between UDF IV and UDF V and its developers that gave Appellants authorization to make advances without notice or input from the developer. Later, UDF relied on this clause to control the advance requests which were funneled into UDF III, even over the objection of the developers. Obert, Greenlaw, and Jester testified in their own defense at the trial. They contended that these advances amounted to a process which they likened to refinancing a loan with common borrowers. Each time a fund loaned money to another fund it received a specified amount of collateral that was worth more than the loan. Appellants further contended that the advances were beneficial because they allowed UDF IV and UDF V to garner collateralized loans that would generate interest and allowed UDF III to have its loan repaid so it could make distributions to investors and pay its own debts. Ultimately, as UDF’s auditor explained at trial, these advances made it appear as though UDF IV and UDF V had more notes receivable because it was issuing new loans, and it also made it appear as though UDF III’s loans were getting paid down successfully, when they were not. To further exacerbate these allegations, UDF’s SEC filings stated that UDF V would not engage in affiliate transactions[3] and that the source of funds in UDF III would be “cash . . . from operations.” The Government’s theory was that the cash was not from operations, but from investors in UDF IV and UDF V that Appellants used to pay distributions to investors and to repay loans from banks. Moreover, according to the Government, affiliate transactions were exactly what Appellants conducted when they transferred money from one UDF fund to another, even though UDF V’s SEC filings stated that it “would not participate in any investments with . . . any of [its] affiliates.” ii. Evidence of Other Sources Funding the Scheme Along with conducting undisclosed advances, jurors heard evidence that Jester and Wissink manipulated developers’ cash-flow statements before submitting them to auditors, which made the developers appear financially capable of paying off their loans earlier than projected. Appellants also used loan funds for unauthorized purposes and obtained loans from various banks that relied on UDF’s SEC filings when deciding whether to issue loans. Upon the conclusion of evidence, the district court delivered instructions to the jury. After which, they deliberated for “a day and a half” and found Appellants guilty of all charges. The trial lasted a total of seven days. B. Post Trial At the presentencing stage, the United States Probation Office collected various victim-impact statements and calculated that Appellants caused an “intended loss” of over one million dollars. The Government argued that this amount was properly calculated and sought restitution reflecting that loss. However, the district court rejected this request and determined that the Government failed to meet its burden to show that restitution should be imposed on Appellants as part of their sentence. Nevertheless, the district court still imposed an enhancement for intended loss and for substantial hardship caused to 25 or more victims. As a result, Greenlaw was sentenced to 84 months’ imprisonment, Wissink and Obert were sentenced to 60 months’ imprisonment, and Jester was sentenced to 36 months’ imprisonment. Appellants then filed motions for acquittal and for a new trial in which they challenged, inter alia, the jury instructions and several other issues they appeal herein. When those arguments proved unsuccessful before the district court, they appealed. II. Discussion A. Sufficiency of the Evidence “We review challenges to the sufficiency of the evidence de novo, applying the same standard as applied by the district court: could a rational jury find that all elements of the crime were proved beyond a reasonable doubt?” United States v. Chapman, 851 F.3d 363, 376 (5th Cir. 2017) (citation omitted). That means, rather than “reweigh theevidence,” we must “search the record for evidence to support the convictions beyond a reasonable doubt.” United States v. Swenson, 25 F.4th 309, 316 (5th Cir. 2022) (quotations omitted). In doing so, we “view[] all evidence in the light most favorable to the government and draw[] all reasonable inferences in favor of the jury’s verdict.” United States v. Harris, 821 F.3d 589, 598 (5th Cir. 2016) (citing United States v. Eghobor, 812 F.3d 352, 361–62 (5th Cir. 2015)). “This standard is ‘highly deferential to the verdict.’” Id. (quoting United States v. Roetcisoender, 792 F.3d 547, 550 (5th Cir. 2015)). Recall that Appellants were convicted of one count of conspiracy to commit wire fraud affecting a financial institution, in violation of 18 U.S.C. § 1349 (18 U.S.C. § 1343); one count of conspiracy to commit securities fraud, in violation of 18U.S.C. § 1349(18U.S.C.§ 1348); and eight counts of securities fraud and aiding and abetting, in violation of 18 U.S.C. §§ 1348 & 2. Pertinent here, each conspiracy count requires that the act be completed with a specific “intent to defraud.”[4] As for the substantive aiding and abetting securities fraud counts, the statutes both require, inter alia, that the act be completed with a specific “intent to defraud” and prohibit the execution of a “scheme to defraud.” See 18 U.S.C. § 1348;[5] United States v. Hussain, 972 F.3d 1138, 1145–46 (9th Cir. 2020).[6] We have described a scheme to defraud, as including “any false or fraudulent pretenses or representations intended to deceive others in order to obtain something of value, such as money, from the null to be deceived.” United States v. Evans, 892 F.3d 692, 711–12 (5th Cir. 2018), as revised (July 6, 2018) (alteration in original); United States v. Scully, 951 F.3d 656, 671 (5th Cir. 2020) (“To establish that [the defendant] engaged in a scheme to defraud, the Government must prove that he ‘made some kind of a false or fraudulent material misrepresentation.’” (quotation omitted)). As for the “intent to defraud” element, the Government must prove “an intent to (1) deceive, and (2) cause some harm to result from the deceit.” Evans, 892 F.3d at 712 (quotation omitted). This element is generally satisfied “when [a defendant] acts knowingly with the specific intent to deceive for the purpose of causing pecuniary loss to another or bringing about some financial gain to himself.” Id. (quotation omitted); Scully, 951 F.3d at 671. Appellants argue that (1) the Government failed to prove that they made material misrepresentations and (2) those misrepresentations were made with the intent to deprive investors of money or property. Separately, Jester argues (3) that the evidence failed to show that he had the requisite knowledge necessary to support his convictions. We address each argument in turn. i. Material Misrepresentation “The essence of fraud is that its perpetrator has persuaded his victim to believe, beyond the dictates of reason or prudence, what is not so.” United States v. Perez-Ceballos, 907 F.3d 863, 868 (5th Cir. 2018) (quoting United States v. Church, 888 F.2d 20, 24 (5th Cir. 1989)). As stated, this principle manifests under the “scheme to defraud” prong which requires that the defendant have “made some kind of a false or fraudulent material misrepresentation.” Scully, 951 F.3d at 671 (quotation omitted). A misrepresentation is material if “it ‘has a natural tendency to influence, or is capable of influencing, the decision of the decision-making body to which it was addressed.’” United States v. Lucas, 516 F.3d 316, 339 (5th Cir. 2008) (quoting United States v. Harms, 442 F.3d 367, 372 (5th Cir. 2006)). The Government’s principal allegation at trial was that Appellants used investor money from UDF IV and UDF V to pay distributions to UDF III investors and to repay loans from banks, and then lied about it to the investing public and the SEC. This allegation was based on two central alleged misrepresentations. First, the Government argued that Appellants represented on UDF III’s quarterly and annual filings that the source of distributions to UDF III investors and lenders would be “cash . . . from operations”—i.e., the “interest the developers were paying back on [their UDF III loans].” But the Government asserted that the source of the funds was really “cash from investors in UDF IV and UDF V.” Second, the Government argued that Appellants represented in their UDF V filings and advertised in marketing presentations to brokers, dealers, and financial advisors that UDF V would not engage in affiliate transactions. However, the Government argued that the money movements from UDF V to UDF III to pay distributions and a bank loan constituted affiliate transactions because UDF controlled both sides of the transaction. Throughout the trial and again on appeal, Appellants contend that the money movement was loans to common borrowers, rather than affiliate transactions. According to them, as projects in UDF III’s portfolio reached later stages of development, UDF IV and UDF V provided subsequent financing to developers for specific projects to fund new phases of development. In such instances, the developer would become a common borrower because they would, for example, take out a loan from UDF III and later in the development process take out a loan from UDF IV and UDF V. Appellants argue that when it moved money from UDF IV and UDF V to UDF III, it was actually paying down the developer’s existing loan with UDF III. And in return, the lender, UDF IV or UDF V, obtained an enforceable security interest in collateral released through repayment to UDF III. As such, Appellants argue that the statements in UDF III’s SEC filings were not “false or fraudulent” because the money transferred to UDF III was the repayment of an existing UDF III borrower’s debt, and thus cash from operations. Moreover, they emphasize that the transactions were not between affiliates because they were between UDF and a common borrower, as the transferred funds were advanced to the same UDF III borrower pursuant to a collateralized loan from UDF IV and UDF V. They maintain that, although UDF V represented that it would not engage in certain transactions with “affiliates,” that term was ambiguous, not objectively false. Further, they assert that UDF’s filings disclosed that it might “invest in multiple mortgage loans that share a common borrower.” Appellants therefore urge that there was insufficient evidence to support the jury’s finding that they made false or fraudulent material misrepresentations beyond a reasonable doubt. We disagree. a. UDF III’s SEC Filings There is overwhelming evidence showing that the cash used to pay UDF III’s investors was not cash from operations as purported in its annual and quarterly SEC filings. One of the Government’s key witnesses, Scott Martinez, a forensic accountant at the FBI, exhibited a cash-tracing mechanism and generated a report showing that the source of the money— over $66millionof the UDF III distributions—was solely cash from investors in UDF IV and UDF V. Although Appellants’ witness, Dale Kitchens, a certified specialist in financial forensics, testified that the transactions were a legitimate business practice intended to pay common borrower’s loans, the jury was free to weigh the contrary evidence before it and our role is not to reweigh that evidence. See Swenson, 25 F.4th at 316. The record shows that Martinez’s cash-tracing testimony was heavily supported by emails from several UDF employees explaining that the purpose of the money movement was to afford UDF III’s distributions. For example, an email of a UDF employee showed that the advances were completed “to pay investor distributions so we can ensure it goes out today” and another stated that the “bottom line of the transaction was to get cash into UDF III in order to fund distributions.” From this evidence, a reasonable juror could conclude that the money was not transferred to pay a common developer’s loan as Appellants urge. Furthermore, Appellants fail to refute extensive evidence showing that they paid past due payments on a loan from Legacy Texas Bank using investor money from UDF V and not interest from UDF III’s operations as represented in the SEC filings. Testimony revealed that, after Greenlaw and Wissink spoke with employees from the bank about the past due payment, Wissink and Jester worked together to approve two draw requests from UDF V to UDF III’s Legacy Texas bank account to pay the past due amount. This evidence exemplifies that their representations were not only false, but also material. See Lucas, 516 F.3d at 339–41. The undisclosed advances allowed Appellants to mask UDF III’s true financial health from the investing public, as the investors in UDF IV and UDF V, along with the re-investors in UDF III, believed that they were buying into a more successful fund. In fact, auditors testified that they were alarmed by this practice and it would have been salient information for their reports. b. UDF V’s SEC Filings As to Appellants’ filings for UDF V, the statement—that it would not participate in affiliate transactions—is likewise materially false. The term “affiliate” is defined by the SEC as one “that directly, or indirectly through one or more intermediaries, controls, or is controlled by, or is under common control with the issuer.” 17 C.F.R. § 230.144(a)(1). The term “control” is defined as “the possession, direct or indirect, of the power to direct or cause the direction of the management and policies of a person, whether through ownership of voting securities, by contract or otherwise.” 17 C.F.R. § 230.405. The jury was instructed on these definitions, and testimony further guided them in their interpretation and application of these terms at trial. The Government’s witness, Michael Wilson, UDF’s marketing director, explained that “an affiliate transaction is when one UDF fund would engage in a transaction directly with another UDF fund. So[,] a loan to another UDF fund or a profit participation in a loan to another UDF fund or a credit enhancement to another UDF fund.” Thomas Carocci, an attorney for the Financial Industry Regulatory Authority, described UDF III, IV, and V as affiliates and testified that transactions between them must be disclosed in SEC filings. Moreover, William Kahane, one UDF V’s own board members, testified that “[h]e would have considered money going from UDF V to UDF III to pay investors to be an affiliated transaction, and he would not have approved.” A reasonable juror could deduce from these explanations that UDF V’s transfer of money to UDF III to pay distributions was a “loan to another UDF fund,” and thus an affiliate transaction.[7] Appellants’ argument—that the Government failed to prove an objectively false statement—is debunked by the evidence. They rely on United States v. Harra, a case in which the Third Circuit reversed a false statement conviction under 18 U.S.C. § 1001. 985 F.3d 196, 225 (3d Cir. 2021). It ultimately held that “in the face of ambiguous reporting requirements, . . . fair warning demands that the Government prove a defendant’s statement false under each objectively reasonable interpretation of the relevant requirements.” Id. at 211, 213. We see no reason to apply this false statement rule here. The term “affiliate” is not undisputedly ambiguous. Unlike the term analyzed in Harra, which had “no statutory or regulatory definitions illuminating the definition,” the term “affiliate” is expressly defined within the statutory and regulatory definitions. Id. at 206. Indeed, the defining factor in determining whether these transactions constituted affiliate transactions was control, and evidence revealed that UDF fully controlled the transaction on each end. Instead of developers prompting an advance when they needed money to fund a short-term special project, Appellants initiated the advances, determined the amount of these advances, and determined which developer would fund the transfer without any involvement from the developer. An email from an alarmed auditor explained the process: UDF III needed money to pay distributions, UDF IV sent them $1.2M. On UDF IV it was recorded as a debit to [the developer's] notes receivable and credit to cash. On UDF III they recorded as a debit to cash and credit to one of [the developer's notes]. They don’t get any kind of approval from [the developer] before they do stuff like this. There is even evidence of instances where Appellants initiated an advance on the part of a developer who expressly requested that UDF stop the practice. Such evidence strongly supports a determination that UDF V’s statements were false. In the same vein, the evidence demonstrates that UDF V’s misrepresentations were material. See Lucas, 516 F.3d at 339. Multiple witnesses testified that the industry had shifted away from affiliate transactions because they were disfavored and that a no-affiliate-transaction policy in UDF V would enable it to participate in a larger network of broker, dealers, and investors. Appellants, specifically Greenlaw and Obert, participated in meetings with brokers, dealers, and financial advisors who represented UDF’s products to their retail investors. To garner business, they advertised that UDF V would not conduct affiliate transactions. Considering this evidence and drawing all reasonable inferences in the light most favorable to the verdict, a reasonable juror could have determined that Appellants made material misrepresentations in UDF III and UDF V’s filings that were sufficient to uphold their convictions. See Scully, 951 F.3d at 671. ii. Intent to Deprive of Money or Property Appellants’ second argument is that there was insufficient evidence to prove beyond a reasonable doubt that they intended to deprive the investors of money or property.[8] Specifically, Appellants state that the “scheme to defraud” and “intent to defraud” elements require that a juror find not only that they intended to deceive, but also that they intended to deprive victims of money or property. According to Appellants, the Government introduced no evidence showing that any of the investors were harmed or that they intended to harm investors. Appellants emphasize that the “uncontradicted evidence establishe[d] that all three funds received significant value from the transactions at issue.” Specifically, Appellants assert that the advances benefited all involved, including the investors. The Government does not refute that it was required to prove that Appellants intended to deprive investors of money or property to satisfy these elements. It points to evidence that Appellants participated in a scheme to defraud investors of their money, emphasizing that when UDF III lacked sufficient money in its bank account to pay distributions to its investors, a mad scramble ensued moving money between funds. The Government further argues that Appellants’ concealment of these transactions from the investing public, its current investors, and auditors, demonstrates their intent to defraud. The Government points to evidence showing that because distributions were not tied to the funds’ actual performance, Appellants caused investors to believe that they were putting their money into a successful business when they were not. As an initial matter, Appellants are correct that the “scheme to defraud” and “intent to defraud” elements must be based on property interests. Although neither the wire nor securities fraud statutes provide a definition of what constitutes fraud, we are aided by an extensive line of cases construing these provisions from the Supreme Court and this court. These cases draw a fine line differentiating conduct that is merely deceitful, from conduct that “wrong[s] one in his property rights.” Cleveland v. United States, 531 U.S. 12, 19 (2000) (citation omitted) (explaining that the fraud statutes “protect[] property rights only”). Only the latter falls within the “common understanding” of defraud. Id. (citation omitted). This is not a new principle, though recent Supreme Court cases have brought it to the forefront. In Kelly v. United States, the Court relied on its 1987 decision in McNally for the proposition that fraud statutes are “‘limited in scope to the protection of property rights.’” 140 S. Ct. 1565, 1571 (2020) (quoting McNally v. United States, 483 U.S. 350, 360 (1987)). In doing so, the Court reversed the wire fraud convictions of defendants that “used deception to reduce the city’s access lanes to the George Washington Bridge” for political gain. Id. at 1574. It explained that the Government had to prove “not only that [the defendants] engaged in deception, but that an object of their fraud was property.” Id. at 1571 (cleaned up). Because commandeering the bridge did not deprive victims of their “property,” the defendant’s actions were not a crime under the fraud statutes. Id. at 1572. Likewise, in Shaw v. United States, the Supreme Court remanded a case involving the bank fraud statute for the Ninth Circuit to determine whether the jury instructions were erroneous. See 580 U.S. 63, 72 (2016). The instructions defined a “scheme to defraud” as “any deliberate plan of action or course of conduct by which someone intends to deceive, cheat, or deprive a financial institution of something of value.” Id. Although the Court did not decide the question, it stated that it agreed with the parties that “the scheme must be one to deceive the bank and deprive it of something of value.” Id. (emphasis in original). Our case law has also consistently modeled this principle. See United States v. Hoeffner, 626 F.3d 857, 863 (5th Cir. 2010) (per curiam) (explaining that the mail and wire fraud statutes’ proscriptions reach schemes to deprive another person of “money or property by means of false or fraudulent pretenses, representations or promises” (quoting 18 U.S.C. §§ 1341, 1343)); United States v. Ratcliff, 488 F.3d 639, 645 (5th Cir. 2007) (affirming the district court’s order dismissing an indictment where it alleged only deceitful conduct and failed to allege a scheme that wronged the victim’s property rights); United States v. McMillan, 600 F.3d 434, 449 (5th Cir. 2010) (“The issue is whether the victims’ property rights were affected by the misrepresentations.”).[9] As applied to this case, the Government was required to prove that the scheme was one in which Appellants intended to deprive the investors of money or property through misrepresentations, thereby wronging the investors’ property rights. See Ratcliff, 488 F.3d at 645. We hold that it met its burden. Evidence strongly supports a finding that Appellants intended to conduct a scheme to deprive investors of their money. There is proof that they purposefully advertised a desired rate of return to brokers and continued to solicit investors to invest their money into UDF III despite knowing that UDF III did not have enough money to sustain its current investors. Evidence also shows that they purposefully did not invest UDF IV and UDF V investors’ money into the business or otherwise use the money to further fund developer’s projects. Their intention is further displayed through several emails evincing that they were aware that they needed to use new investor money to fund their distributions or risk deterring current investors from selling stock and new investors from buying stock. A rational trier of fact could readily infer from this evidence that new investor money was the object of Appellants’ operation because it was only after the money was transferred that they were able to pay distributions to UDF III investors. See Chapman, 851 F.3d at 376. Along with evidence of the undisclosed advances, there is evidence that Appellants used bank loans for unapproved purposes and masked the developers’ financial health by manipulating the developers’ cash-flow statements prior to submitting it to their auditors. By concealing information, investors were lured into investing in a business that could not pay its distributions in 53 out of the 60 months relevant here. We also agree with the Government that Appellants’ concealment of these actions provide support for an inference of intent. Scully, 951 F.3d at 671; see United States v. Dobbs, 63 F.3d 391, 396 (5th Cir. 1995) (explaining that evidence of a defendant’s attempt to conceal his actions can support an inference of intent to defraud). Jurors also saw that this operation was dependent on each co-appellant. For example, they saw an email from Wissink instructing a UDF employee to “advance $1.75 million” from UDF IV to UDF III, adding: “This will cover the distribution for Monday.” One UDF employee testified that Wissink regularly told him to carry out such advance requests. As another UDF employee explained: “The bottom line of the transaction was to get cash into UDF III in order to fund distributions.” Along with Wissink, Jester also directed most of these advances, with Obert included on many of the emails. Moreover, during cross-examination, Obert admitted that she knew UDF personnel initiated the advance requests and that she was aware of this practice of using advance requests to transfer money between UDF entities. In fact, Obert and Greenlaw signed all of the SEC filings which concealed the true source of UDF III’s funds and misrepresented the funds’ performance. Based on this avalanche of evidence, a rational trier of fact could infer that Appellants participated in a scheme to obtain something of value—namely, money. See United States v. Jonas, 824 F. App’x 224, 232 (5th Cir. 2020) (per curiam) (unpublished) (holding that a reasonable jury could infer that the defendant engaged in a scheme to defraud where he used misrepresentations “to obtain money from the investors”). Appellants’ arguments fail to overcome the sufficient evidence at hand. First, their characterization of these acts as normal business transactions that benefited the investors is unpersuasive. They emphasize that Martinez admitted that he did not assess the economic rationale for the transactions or the value of the collateral that secured the loans from UDF IV and UDF V. But even if we assumed that evidence of an economic benefit to UDF funds would somehow refute that investors were deprived of their money, a jury was entitled to reject that evidence and instead credit the contrary testimony of the Government’s witnesses. See Scully, 951 F.3d at 671 (citing United States v. Spalding, 894 F.3d 173, 181 (5th Cir. 2018)). Second, the fraud convictions are not undermined by the fact that the Government did not present evidence showing that investors incurred monetary loss. This is because success of the scheme is immaterial. Take Shaw for example. There, the Supreme Court rejected a bank-fraud defendant’s argument that “he did not intend to cause the bank financial harm.” 580 U.S. at 67. In that case, it was known that “due to standard business practices in place at the time of the fraud, no bank involved in the scheme ultimately suffered any monetary loss.” Id. The Court reasoned that it is “sufficient that the victim . . . be deprived of its right to use of the property, even if it ultimately did not suffer unreimbursed loss.” Id. at 67–68 (citing Carpenter v. United States, 484 U.S. 19, 26–27 (1987)); see id. at 68 (“[W]here cash is taken from a bank but the bank is fully insured, the theft is complete when the cash is taken; the fact that the bank has a contract with an insurance company enabling it to shift the loss to that company is immaterial.” (cleaned up)). It follows that Appellants’ argument is unconvincing.[10] It does not matter that UDF IV and UDF V had collateral on the loans that it transferred to UDF III. Nor does it matter that they did not intend to cause investors financial loss. See Shaw, 580 U.S. at 67 (“[T]he [fraud] statute, while insisting upon ‘a scheme to defraud,’ demands neither a showing of ultimate financial loss nor a showing of intent to cause financial loss.”). Appellants exposed investors to risks and losses that, if publicly disclosed, would have decreased its value and investment power. That is enough to support a fraud conviction. See United States v. McCauley, 253 F.3d 815, 820 (5th Cir. 2001); United States v. Saks, 964 F.2d 1514, 1519 (5th Cir. 1992) (holding that a fraudulent loan transaction exposed financial institutions and lenders to risk of loss even though the loan was “secured, and [defendants] assumed a legal obligation to repay it.”). Lastly, Appellants argue that the Government failed to show that they affected investors’ property rights because it improperly argued that they deprived UDF investors of accurate information about UDF III’s financial health. The Government does not refute that it relied on this theory; it instead argues that the accurate information theory was subsidiary to its trial theory, i.e., that “Appellants defrauded the investing public and their own shareholders with the intent to obtain tangible property—money—from investors in the later funds to pay distributions on the earlier funds.” In making this argument, Appellants highlight the Ninth Circuit’s decision in United States v. Yates which held that “[t]here is no cognizable property interest in the ethereal right to accurate information.” 16 F.4th 256, 265 (9th Cir. 2021) (citation and quotation omitted). The Ninth Circuit explained that if it were to “recogniz[e] accurate information as property,” that “would transform all deception into fraud.” Id. This is because “[b]y definition, deception entails depriving the victim of accurate information about the subject of the deception.” Id. We agree with this interpretation of the accurate information theory. As a matter of logic, to deprive someone of accurate information is to deprive them of the truth, i.e., to deceive. See United States v. Sadler, 750 F.3d 585, 591 (6th Cir. 2014) (explaining that it cannot “plausibly be said that the right to accurate information amounts to an interest that ‘has long been recognized as property.’” (quoting Cleveland, 531 U.S. at 23)).[11] Nonetheless, the foremost scheme alleged here was for the Appellants to obtain money from investors, and the Government’s mountain of evidence supporting this theory is sufficient, regardless of the invalidity of its subsidiary theory. See United States v. Stalnaker, 571 F.3d 428, 437–38(5th Cir. 2009) (upholding verdict upon finding that it did not rest on an insufficient legal theory). Accordingly, we are convinced that the evidence is sufficient beyond a reasonable doubt from which a rational jury could infer that Appellants participated in a scheme to defraud and acted with an intent to defraud investors of their money. See Chapman, 851 F.3d at 376. iii. Jester’s Knowledge Jester separately asserts that he did not have the requisite knowledge to support his convictions for conspiracy to commit wire fraud, conspiracy to commit securities fraud, and aiding and abetting securities fraud. Along with a specific intent to defraud, to prove a conspiracy the Government must prove that “(1) two or more persons made an agreement to commit an unlawful act; (2) the defendant knew the unlawful purpose of the agreement; and (3) the defendant joined in the agreement willfully, with the intent to further the unlawful purpose.” United States v. Simpson, 741 F.3d 539, 547 (5th Cir. 2014) (citing United States v. Grant, 683 F.3d 639, 643 (5th Cir. 2012)); see also Brooks, 681 F.3d at 699. Jester’s principal contention is that the Government’s case is built on two central misrepresentations in UDF V and UDF III’s SEC filings, but, because of his position as the director of asset management, he did not have any knowledge of the SEC filings or of its contents. We are unpersuaded and will not reverse a conviction for “lack of evidence that [he] knew each detail of the conspiracy.” United States v. Vergara, 687 F.2d 57, 61 (5th Cir. 1982) (citing United States v. Rosado-Fernandez, 614 F.2d 50 (5th Cir. 1980)). Rather, “knowledge may be inferred from surrounding circumstances.” Simpson, 741 F.3d at 547; United States v. Sanders, 952 F.3d 263, 273 (5th Cir. 2020) (“[A] jury can infer from the surrounding circumstances whether a defendant participated in and knew of the conspiracy.”). Reviewing what is now familiar, the evidence at trial established that Jester’s role was prevalent in several parts of the overall scheme in at least three central ways. Jester directed many of the undisclosed advance transactions at issue; he manipulated developer’s cash flow projection spreadsheets by inflating their amount of development projects to make it appear as though developers were paying off their loans quicker; and he directed that money from loans be used to pay UDF III investor’s distributions. Put simply, Jester’s actions were the lifeline to the misrepresentations at issue here. See United States v. Thompson, 761 F. App’x 283, 291 (5th Cir. 2019) (per curiam) (unpublished) (determining that “repeated exposure to the fraud” was probative of the defendant’s knowledge). Moreover, during his testimony he revealed that he knew that the spreadsheet with the inflated projected cash-flow spreadsheets were “going to an independent auditor that was going to rely upon it for its audit of the financial statements.” His knowledge about the unlawfulness of the scheme also emanates from his leadership in the fraudulent actions and his propensity to commit deceitful acts that were outside of UDF’s general practice, such as directing the undisclosed advances when UDF III had insufficient funds and misapplying investor money when UDF was behind on loan payments. See United States v. Willett, 751 F.3d 335, 340–41 (5th Cir. 2014) (determining that a “position of authority” was probative of knowledge). He also played a role in concealing UDF III’s true financial condition. For example, in August 2013, Jester suggested that distributions could be funded from a UDF loan and told a UDF asset manager to “remember to fix this once we finalize some of the new deals.” See United States v. Martinez, 921 F.3d 452, 470 (5th Cir. 2019) (“[E]fforts to assist in the concealment of a conspiracy may help support an inference that an alleged conspirator had joined the conspiracy while it was still in operation.”) (citation omitted)). Despite the prevalence of Jester’s actions throughout this case, he analogizes the evidence here to that which was insufficient to show that the defendant acted “knowingly and willfully” in United States v. Nora, 988 F.3d 823 (5th Cir. 2021). Jester asserts that, like the defendant in Nora, although he was entangled in the fraudulent transactions, he did not know about the specific misrepresentations in the filings as he merely “performed his job duties without visibility into that separate part of UDF’s business.” In Nora, we held that generalized evidence that a staff member received training on compliance which would have alerted him to the unlawful nature of his work was insufficient to support his health care fraud and kickback scheme convictions. 988 F.3d at 831–34. The evidence here is a far cry from the slim evidence in Nora. Here, there is evidence that Jester was present in meetings with developers, bank personnel and co-appellants, and had a key role in developing the documentation that the auditors relied upon for their reports to the SEC. Thus, the evidence was sufficient to support the knowledge element of Jester’s conspiracy to commit wire and securities fraud convictions. See United States v. Scott, 892 F.3d 791, 797–98 (5th Cir. 2018). The analysis is similar for Jester’s aiding and abetting securities fraud counts. Id. at 799 (“Typically, the same evidence will support both a conspiracy and an aiding and abetting conviction.”). Aiding and abetting “is not a separate offense, but it is an alternative charge in every indictment, whether explicit or implicit.” United States v. Neal, 951 F.2d 630, 633 (5th Cir. 1992). It is therefore “not necessary that [Jester] commit the overt acts that . . . accomplish the offense or that he have knowledge of the particular means his principals . . . employ to carry out the criminal activity.” United States v. Austin, 585 F.2d 1271, 1277 (5th Cir. 1978). There likewise does not need to “be proof ‘that the defendant was present when the crime was committed or that he actively participated therein.’” Sanders, 952F.3d at277 (quoting United States v. James, 528 F.2d 999, 1015 (5th Cir. 1976)). “[T]he government need only establish that [he] ‘assisted the actual perpetrator of the [fraud] while sharing the requisite criminal intent.’” Stalnaker, 571 F.3d at 437 (quoting United States v. Rivera, 295 F.3d 461, 466 (5th Cir. 2002)). Considering the “extremely deferential review of jury verdicts,” the Government has met this standard as to the aiding and abetting securities fraud counts as well. Nora, 988 F.3d at 834; see Stalnaker, 571 F.3d at 437. B. Jury Instructions Appellants advance two issues pertaining to the jury instructions. They argue that the district court (1) misstated elements of the law and (2) abused its discretion in denying a proposed instruction. We address each argument in turn. i. Misstatement of Law First, Appellants argue that even if this court determines that there was sufficient evidence to support the “scheme to defraud” and “intent to defraud” elements, they deserve a new trial because the district court’s instructions defining those elements were incorrect. They insist that the instructions did not require the jury to determine that they intended to deprive investors of money or property. This issue is one of statutory construction, which we review de novo. United States v. Gas Pipe, Inc., 997 F.3d 231, 236 (5th Cir. 2021). “Generally, failure to instruct the jury on every essential element of the offense is error.” United States v. Williams, 985 F.2d 749, 755 (5th Cir. 1993). That error is then “subject to harmless error” review. Gas Pipe, Inc., 997 F.3d at 236 (citation omitted). “Erroneous jury instructions are harmless if a court, after a thorough examination of the record, is able to conclude beyond a reasonable doubt that the jury verdict would have been the same absent the error.” United States v. Stanford, 823 F.3d 814, 828 (5th Cir. 2016) (internal quotation marks and citations omitted); see also Pope v. Illinois, 481 U.S. 497, 501–02 (1987). Before the trial, Appellants requested jury instructions which stated that a “‘scheme to defraud’ [was] a plan intended to deprive another of money or property” and that a “‘specific intent to defraud’ [was] a willful, conscious, knowing intent to cheat someone out of money or property.” The district court rejected their request and instead relied on language in the Fifth Circuit Criminal Pattern Jury Instructions.[12] The district court instructed the jury that: A “scheme to defraud” means any plan, pattern, or course of action intended to deprive another of money or property or bring about some financial gain to the person engaged in the scheme . . . [and] A “specific intent to defraud” means a conscious, knowing intent to deceive or cheat someone. See Fifth Circuit, Pattern Jury Instructions (Criminal Cases) § 2.57 (2019) (emphasis added). Appellants objected to these instructions before the district court. Specifically, they argued that in both charges, the second clause after the disjunctive “or” results in a misstatement of the law. See Shaw, 580 U.S. at 72. They further argued that a case in this circuit applying the same jury instruction was distinguishable, see United States v. Baker, 923 F.3d 390, 402– 03 (5th Cir. 2019), cert. denied, 140 S. Ct. 2565 (2020), and emphasized that other circuits have changed their pattern jury instructions after the Supreme Court’s decision in Kelly clarified that the focus of the inquiry is property rights. These objections were overruled with little to no discussion. On appeal, Appellants expound on the same arguments, stating that the current model jury instructions for a “scheme to defraud” and “intent to defraud” are “inconsistent” with the law. Model jury instructions are only proper if they are a “correct statement of the law.” United States v. Toure, 965 F.3d 393, 403 (5th Cir. 2020) (citing United States v. Whitfield, 590 F.3d 325, 354 (5th Cir. 2009)). As explaine d supra, under the fraud statutes, the proper question is “whether the victims’ property rights were affected by the misrepresentations.” McMillan, 600 F.3d at 449. Our next step then is to determine whether the “scheme to defraud” and “intent to defraud” instructions were aligned with our interpretation of the fraud statute’s required inquiry. As to the “intent to defraud” instruction, we agree with Appellants that the disjunctive “or” makes it a misstatement of law. Under a plain reading of the instruction given, the jury could find that the Government proved an “intent to defraud” if Appellants merely exhibited a “conscious, knowing intent to deceive . . . someone.” But we have already explained that deception is not synonymous with depriving another of their property interests. The Government provides no argument refuting this construction of the intent to defraud language which is directly at odds with our caselaw holding that a jury cannot convict a defendant under the fraud statutes based on deceit alone. Indeed, it has long been our understanding that an “‘intent to defraud’ requires ‘an intent to (1) deceive, and (2) cause some harm to result from the deceit.’” Evans, 892 F.3d at 712 (emphasis added) (quoting United States v. Moser, 123 F.3d 813, 820 (5th Cir. 1997) (quoting in turn United States v. Jimenez, 77 F.3d 95, 97 (5th Cir. 1996))); Ratcliff, 488 F.3d at 645–49. We find support for our interpretation of the “deceive or cheat” construction when reviewing our sister circuits’ decisions analyzing similar challenges. See United States v. Miller, 953 F.3d 1095, 1101–04 (9th Cir. 2020). In line with then-existing Ninth Circuit pattern instructions, the district court in Miller charged the jury “that, to be guilty of wire fraud, a defendant must have acted with the intent to ‘deceive or cheat.’” Id. at 1101. (emphasis in original). On appeal, the defendant argued that this instruction misstated the law.[13] Id. The Ninth Circuit agreed, holding that several of its cases applying that instruction “were no longer tenable in light of the Supreme Court’s intervening ruling” in Shaw. Id. at 1102; 580 U.S. at 72. It thus held “that wire fraud requires the intent to deceive and cheat—in other words, to deprive the victim of money or property by means of deception.” Miller, 953 F.3d at 1103 (emphasis in original). Likewise, the Eleventh Circuit’s decision in Takhalov is consistent with our viewhere.827 F.3dat1312, 1314. In that case, the defendants tricked men to come into their bars and nightclubs by hiring “[women] to pose as tourists, locate visiting businessmen, and lure them” in. Id. at 1310. The defendants asked the district court to instruct the jury that the “[f]ailure to disclose the financial arrangement between the [women] and the Bar, in and of itself, [was] not sufficient to convict a defendant of any offense[.]” Id. at 1314. The Eleventh Circuit determined that the defendant’s instruction was a correct statement of the law and rejected the Government’s attempt to instruct the “jurors that they could convict only if they found that the defendants had schemed to lie about the quality or price of the goods sold to the[men].” Id. It therefore made clear that a defendant “cannot be convicted of wire fraud on the basis of [a] lie alone . . . [because] deceiving is a necessary condition of defrauding but not a sufficient one.” Id. at 1312, 1314. Appellants point to these decisions as persuasive authority and we agree with our sister circuit’s interpretation of similar instructions of an intent to defraud.[14] This element is present in each count of this case, and when confronted with this instruction, a reasonable juror could conclude that evidence of Appellants’ misrepresentations and lies alone obligated the jury to infer an intent to defraud. Because deception, alone, will not suffice, the intent to “deceive or cheat” instruction was erroneous. See Miller, 953 F.3d at 1103. Our analysis under the “scheme to defraud” instruction, however, is much less clear. Appellants argue that, under a plain reading, the jury could find that the Government proved a “scheme to defraud” if Appellants executed a “plan . . . intended to . . . bring about some financial gain to the person engaged in the scheme.” They further assert that this instruction is inconsistent with the Supreme Court’s recent holding in Kelly. 140 S. Ct. at 1574. But the Government rejects this perspective and counters that the second phrase after the disjunctive “or” simply modifies the first phrase because “[d]epriving a victim of a property right and obtaining financial gain from fraud are two sides of the same coin.” Moreover, the Government highlights that this court approved of an instruction with identical language in Baker, 923 F.3d at 402–03, and this court has relied on Baker and similar “financial gain” language even after Kelly. As an initial matter, we recognize, as the Government points out, that this court has long embraced the notion that an “[i]ntent to defraud exists if the defendant acts knowingly with the specific intent to deceive for the purpose of causing financial loss to another or bringing about some financial gain to himself.” Jimenez, 77 F.3d at 97 (citing United States v. St. Gelais, 952 F.2d 90, 95 (5th Cir. 1992)); see also Evans, 892 F.3d at 712 (same); Moser, 123 F.3d at 820 (same); United States v. Akpan, 407 F.3d 360, 370 (5th Cir. 2005) (same); Scully, 951 F.3d at 671 (same). This statement is often construed alongside, and read as consistent with, the understanding that “[n]ot only must a defendant intend to defraud or deceive, but he must intend for some harm to result from the deceit.” St. Gelais, 952 F.2d at 95; Evans, 892 F.3d at 712. Moreover, we have explained that a “scheme to defraud” is insufficient “insofar as victims were left without money that they otherwise would have possessed.” Baker, 923 F.3d at 405. In Baker, when interpreting a “scheme to defraud” instruction with identical language, we held that it “allowed for a conviction if [the defendant] intended to deceive the victims out of their money for his own financial benefit.” Id. Nevertheless, after our review of the arguments and the accompanying caselaw, we decline to decide herein whether the “scheme to defraud” instruction was erroneous, as we have done with the “intent to defraud” instruction. Even if the instruction was another error, there are overwhelming facts supporting these elements. In our harmless analysis, the question is whether the record contains evidence from which a rational juror could find that Appellants participated in a scheme to defraud investors of money or property and had the requisite intent to do so. See Stanford, 823 F.3d at 828. The parties’ arguments on harmlessness are the same as those expressed in the sufficiency of the evidence analysis. And as we have explained supra, there was substantial evidence from which a jury could find beyond a reasonable doubt that the object of Appellants’ scheme was money. Accordingly, “even if the jury had been properly instructed, . . . we are certain beyond a reasonable doubt that the jury would still have found that” Appellants met the “scheme to defraud” and “intent to defraud” elements. United States v. Allende-Garcia, 407 F. App’x 829, 836 (5th Cir. 2011) (unpublished). Because any error “did not contribute to the verdict obtained,” Neder v. United States, 527 U.S. 1, 15 (1999) (quotation omitted), “the conviction can stand.” United States v. Foster, 229 F.3d 1196, 1197 (5th Cir. 2000). ii. Requested Falsity Instruction Appellants next argue that the district court abused its discretion by rejecting their proposed falsity instruction. This argument is unpersuasive. “We ‘review challenges to jury instructions for abuse of discretionand afford the trial court great latitude in the framing and structure of jury instructions.’” Matter of 3 Star Properties, L.L.C., 6 F.4th 595, 609 (5th Cir. 2021) (quoting Young v. Bd. of Supervisors, 927 F.3d 898, 904(5th Cir. 2019)). In doing so, “we must test the instructions given not against those [that the defendant] requested—for a criminal defendant lacks the right to have [his or her] requests adopted word for word—but against the law.” United States v. Hunt, 794 F.2d 1095, 1097 (5th Cir. 1986) (citation omitted). “[A] trial judge’s refusal to deliver a requested instruction constitutes reversible error only if three conditions exist: (1) the instruction is substantively correct; (2) it is not substantially covered in the charge actually given to the jury; and (3) it concerns an important point in the trial so that the failure to give it seriously impairs the defendant’s ability to present a given defense effectively.” Id. (citation omitted). We have interpreted this to mean that “an abuse of discretion occurs only when the failure to give a requested instruction serves to prevent the jury from considering the defendant’s defense.” Id. Appellants requested that the district court instruct the jury that “[a] statement of reasonable opinion is not a false statement.” Appellants again cite the Third Circuit’s decision in Harra, see supra Part II.1.A, for the proposition that the Government has a burden to prove that its interpretation of Appellants’ misrepresentations in the SEC filings were “the only reasonable interpretation[s].” 985 F.3d at 216 (quotation omitted). The district court denied Appellants’ request because it determined that the rest of the charge adequately covered the issue. Consistent with the Fifth Circuit pattern instructions, the district court instructed the jury that “[a] representation is ‘false’ if it is known to be untrue or it is made with reckless indifference as to its truth or falsity” and “would also be ‘false’ if it constitutes a half truth, or effectively omits or conceals a material fact, provided it is made with the intent to defraud.”[15] Appellants argue that their requested instruction was not adequately covered by the other instructions given by the district court. Specifically, they take issue with the district court stating that a true statement is “false” whenever it is “made with reckless indifference as to its truth or falsity.” They emphasize that “[u]nder [these] instructions it made no difference that reasonable minds could deem the disputed statements accurate, or even whether they were accurate—so long as the defendants were ‘reckless.’” The Government, on the other hand, argues that the district court did not commit reversible error in denying Appellants’ request because they thoroughly argued throughout trial that there was a reasonable difference of opinion as to whether the relevant representations were false.” As support, the Government cites to United States v. Brooks, 681 F.3d 678, 705 n.22, 708 n.26 (5th Cir. 2012) and United States v. Gray, 751 F.2d 733, 736–37 (5th Cir. 1985) and contends that the district court allowed Appellants to provide evidence of their reasonable-opinion defense and testimony explaining that they did not believe that the transactions at issue were affiliate transactions. In Gray, the defendant argued that the district court “erred in refusing his requested good-faith instruction.” 751 F.2d at 736. On review, this court acknowledged that the defendant had ample opportunity to present evidence regarding his good faith during the trial to support his defense that he lacked the requisite intent to defraud his customers. This court held that the district court’s refusal to give the instruction was not error. Id. at 737. It concluded: Taken together, the trial, charge, and closing argument laid Gray’s theory squarely before the jury. The court’s charge enabled the jury to recognize and understand the defense theory, test it against the evidence presented at trial, and then make a definitive decision whether, based on that evidence and in light of the defense theory, the defendant was guilty or not guilty Id. at 736–37 (internal quotation and citation omitted). Like in Gray, Appellants had ample opportunity to present their reasonable opinion defense to the jury and they did so “through evidence, closing arguments, or other jury instructions.” United States v. Williams, 774 F. App’x 247, 248 (5th Cir. 2019) (per curiam) (unpublished). In fact, on direct examination, one of UDF’s auditors testified that affiliate transactions were “a judgmental area.” On cross, Appellants questioned him about that statement, and the auditor responded in the affirmative when asked whether his statement meant that “reasonable accountants both acting in good faith could reach a conclusion that is different regarding whether a particular transaction is an [affiliate] transaction.” Appellants also elicited testimony of this nature from at least two other auditors that were before the jury. Further, as mentioned supra, Appellants brought a witness, Kitchens, to testify that UDF V’s money transfers to UDF III were not affiliate transactions because they were transactions amongst common borrowers. See supra Part II. A.i.a. Moreover, during their closing arguments, Appellants concluded based on the culmination of the trial that there was a reasonable difference of opinion as to whether the representations in the SEC filings were false. In one instance, they stated: [Affiliate transactions are] an area where reasonable people, objective people, acting in good faith could look at the same transaction and reach a different conclusion . . . They are not [affiliate] transactions. But even if there was testimony going both ways, you could still have all of the executives acting in good faith. Why? Because it’s an area of judgment. [The Government] cannot get there on the evidence because of this. Lastly, the district court presented to the jury a good-faith instruction stating: In determining whether or not a defendant acted with criminal intent to defraud or deceive, you may consider whether or not the defendant had a good faith belief that what he or she was doing was legal. If you have a reasonable doubt as to whether or not the defendant had a good faith belief that what he or she was doing was legal, you must acquit the defendant and say so by your verdict [of] not guilty. This instruction aided in supporting the crux of Appellants’ argument, which is that they were operating under that good faith belief that the transactions were not “affiliate” transactions, and such was possible because reasonable opinions could differ in good faith on the interpretation of the term. See Lucas, 516 F.3d at 324 (holding that the district court did not abuse its discretion in omitting a requested instruction because the instructions substantially covered the defendants requested instructions). Accordingly, even though the district court denied Appellants’ requested instruction on falsity, they were able to argue that reasonable opinions could differ on the misrepresentations through evidence, arguments, and other jury instructions. See Williams, 774 F. App’x at 248 (“Williams has not shown the omission of his requested instruction impaired his good-faith defense because he raised the defense in several ways during trial.” (citations omitted)). Thus, we hold that the district court did not err in denying the addition of Appellants’ reasonable opinion charge. See id. C. Cross-Examination Appellants next assert that the district court improperly limited their cross-examination regarding a non-testifying informant in violation of their constitutional rights. We review alleged constitutional violations of the Sixth Amendment’s Confrontation Clause de novo. See United States v. Bell, 367 F.3d 452, 465 (5th Cir. 2004). Likewise, we review “alleged violations of a defendant’s Sixth Amendment right to present a complete defensede novo.” United States v. Skelton, 514 F.3d 433, 438 (5th Cir. 2008) (emphasis omitted). Both claims, however, are subject to harmless error review. See Bell, 367 F.3d at 465. And if no constitutional violation has occurred, “we review a district court’s limitation on cross-examination for an abuse of discretion, which requires a showing that the limitations were clearly prejudicial.” Skelton, 514 F.3d at 438. Kyle Bass was the initial informant that contacted the FBI to provide information about UDF, which helped prompt the FBI’s investigation.[16] Appellants argue that the district court’s decision to exclude evidence of Bass’s role in the Government’s investigation violated their constitutional right to present a complete defense and cross-examine key witnesses. They contend that Bass’s testimony would have invalidated the prosecution’s star witness, Martinez, by demonstrating that his cash-tracing report was tainted by Bass’s biased version of Appellants’ actions. We disagree. We have explained that a defendant’s right to present a complete defense and cross-examine witnesses are “closely related.” United States v. Ramos, 537 F.3d 439, 447–48 (5th Cir. 2008). While these two rights are “essential,” they are still subject to limitations. See United States v. Mizell, 88 F.3d 288, 294 (5th Cir. 1996) (explaining that these protections are “limited and must be weighed against the countervailing interests in the integrity of the adversary process . . . the interest in the fair and efficient administration of justice . . . and the potential prejudice to the truth-determining function of the trial process”) (internal quotations and citations omitted). One example of such limitation is the district court’s broad discretion to constrain or deny the cross-examination of a witness whose testimony offers little probative value. See Delaware v. Van Arsdall, 475 U.S. 673, 679 (1986) (permitting trial courts to “impose reasonable limits on cross-examinations based on . . . prejudice, confusion of the issue, . . . or interrogation that is . . . only marginally relevant”). “[T]he Confrontation Clause [only] guarantees an opportunity for effective cross-examination, not cross-examination that is effective in whatever way, and to whatever extent, the defense might wish.” Delaware v. Fensterer, 474 U.S. 15, 20 (1985) (per curiam) (emphasis in original). “To establish a violation of the confrontation right, the defendant need only establish that a reasonable jury might have received a significantly different impression of the witness’s credibility had defense counsel been permitted to pursue his proposed line of cross examination.” Skelton, 514 F.3d at 439–40 (internal quotation marks and citations omitted). Ultimately, “[t]he determination of whether the exclusion of evidence is of a constitutional dimension depends on the district court’s reason for the exclusion and the effect of the exclusion.” Id. This determination often entails a Rule 403 analysis. See id.; Fed. R. Evid. 403 (“The court may exclude relevant evidence if its probative value is substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence.”). At trial, Appellants sought to cross-examine Martinez about his relationship with Bass. They contended that cross-examination was necessary because: (1) Martinez was the key witness and responsible for the cash-tracing theory that the Government relied on in prosecuting Appellants; (2) Martinez could prove that Bass played a substantial role in UDF’s alleged financial scheme; and (3) Bass was the primary reason for Martinez’s decision to trace UDF’s cashflow and he had a tainted motive for compelling the FBI to investigate UDF. In contrast, the Government asked the district court to prohibit questioning related to Bass during Martinez’s cross-examination because it was irrelevant, time consuming, and substantially risked confusing the jury. The district court agreed with the Government and decided that Martinez could not be questioned about Bass on cross-examination.[17] It reasoned that Appellants failed to demonstrate how Bass’s alleged conduct related to or negated their potential guilt. It also explained that Appellants had not shown how evidence of Bass’s conduct affected any “material fact of the crimes charged” against them. Finally, after completing a Rule 403 balancing, the district court stated that this “evidence would be extremely time consuming and confusing,” while only providing minimal “probative value.” Here, the district court’s decision to deny the use of Bass-related information during Martinez’s cross-examination was not a constitutional violation because it would not have altered the jury’s impression of Martinez’s credibility. First, that Bass informed the FBI of potentially illegal conduct at UDF because he intended to benefit from the investigation has no effect on the credibility of Martinez’s cash-tracing theory. As the district court explained, Bass’s conduct “has no bearing on whether [Appellants] committed the crimes charged in the indictment.” Bass, regardless of his motive, simply reported what Appellants were doing with numerous UDF funds. The FBI chose to act on the information Bass relayed to them, but still had to conduct its own investigation and build a case against Appellants. Second, Appellants argue that the Bass-related information could have demonstrated that he was the primary culprit behind UDF’s allegedly illicit activities. That position is unpersuasive because even if Bass had contributed to some of the allegedly fraudulent behaviors, that would not absolve Appellants of their individual roles in the fraudulent scheme. Furthermore, any criminal conduct by Bass would not have tainted the discoveries Martinez and the FBI made following their discussions with him. Third, Appellants would have this court believe that the Bass-related information tainted Martinez’s cash-tracing model by compelling him to ignore the potential economic benefits UDF and the UDF IV fund provided to its investors. But whether Appellants financially harmed their investors is not material of their guilt. See Shaw, 580 U.S. at 67–68. Furthermore, they still presented a line of questioning regarding the potential benefits their funding mechanisms provided to their investors. Specifically, they asked Martinez whether he considered the economic benefit provided by UDF’s funding scheme, to which, Martinez responded in the negative. So, the jury was still able to consider the information that they assert the district court deprived them of presenting. Finally, the district court’s Rule 403 analysis was correct.[18] There was very little probative value in parsing Martinez’s relationship with Bass. The reality is that all of Appellants’ concerns were still considered by the jury, even if those concerns were not raised during Martinez’s cross-examination. For example, the jury considered the novelty of Martinez’s framework at different stages in the litigation and the lack of economic benefit considerations in the cash-tracing model. At best, adding the Bass-related information into the mix would be redundant. At worst, it could have transformed this case in a trial about Bass instead of Appellants.[19] In sum, the possibility that the Bass-related information could have caused an undue delay or substantially confused the jury far outweighed the probative value from using it during Martinez’s cross-examination. See Fed. R. Evid. 403. Because Rule 403 supports the district court’s decision to exclude the Bass-related evidence and prohibiting the information did not significantly affect the jury’s impression of Martinez’s credibility, we hold that the district court did not violate Appellants’ constitutional rights or abuse its discretion in doing so. See Fed. R. Evid. 403; Skelton, 514 F.3d at 440. D. Closing Arguments Appellants bring two issues which arise out of the closing arguments. They argue that the district court (1) allowed the Government to constructively amend the indictment and (2) abused its discretion in allowing the Government to include certain improper statements in its closing argument. We address each argument in turn. i. Constructive Amendment of the Indictment Appellants first argue that the district court improperly allowed, over their objection, a constructive amendment of the indictment during the Government’s closing argument. We review constructive amendment claims de novo. McMillan, 600 F.3d at 450. Appellants take issue with the Government’s assertion that “they made themselves affiliates with Centurion and Buffington.” They aver that this sentence improperly led the jury to convict based on a finding that two of UDF’s developers were UDF V’s prohibited “affiliates.” Appellants argue that this was a theory different than what was alleged in the indictment, i.e., that UDF III and UDF IV were the prohibited affiliates of UDF V. The Government responds that this statement was read out-of-context and that it presented a consistent theory throughout trial. We agree. The constructive amendment doctrine’s core is in the Fifth Amendment, “which provides for criminal prosecution only on the basis of a grand jury indictment.” United States v. Doucet, 994 F.2d 169, 172 (5th Cir. 1993). “[A] court cannot permit a defendant to be tried on charges that are not made in the indictment against him.” Stirone v. United States, 361 U.S. 212, 217 (1960). “Only the grand jury can amend an indictment to broaden it.” Doucet, 994 F.2d at 172. “This court has held that ‘an implicit or constructive amendment . . . occurs when it permits the defendant to be convicted upon a factual basis that effectively modifies an essential element of the offense charged or permits the [G]overnment to convict the defendant on a materially different theory or set of facts than that with which she was charged.’” United States v. Hoover, 467 F.3d 496, 500–01 (5th Cir. 2006) (quoting United States v. Reasor, 418 F.3d 466, 475 (5th Cir. 2005)). As explained supra, the Government’s theory throughout the case was that Appellants conducted “affiliate” transactions when they transferred investor’s money from UDF V to UDF III to pay distributions and loans. Appellants argue that this theory shifted, but we see no indication, outside of this singular statement, that would suggest that the Government altered its theory from what was alleged in the indictment. See United States v. Thompson, 647 F.3d 180, 186 (5th Cir. 2011) (holding that the there was no constructive amendment where the Government maintained “a single, consistent theory of conviction throughout” the trial). Our review of the closing arguments indicates that the Government was not meaningfully shifting its theory to suggest that Centurion and Buffington were affiliates, but instead explaining to the jury that UDF had control of both ends of the transaction, and developers like Centurion and Buffington were just “conduits” or “shells” through which Appellants were able to paper over their unilateral transactions between UDF entities.[20] ii. Improper Statements by the Prosecutor Appellants next argue that prosecutors made three statements during closing arguments that constituted reversible error. First, Appellants rehash their constructive amendment argument, asserting that they were prejudiced when prosecutors stated that Centurion and Buffington were affiliates of UDF V. Second, Appellants argue that the district court committed a similar prejudicial error by allowing prosecutors, during their rebuttal, to misstate the evidence and claim that Susan Powell, UDF III’s auditor, did not understand the transactions, when she in fact knew about the transactions and how they were structured. Lastly, Appellants argue that prosecutors misstated the law regarding the jury’s consideration of Appellants’ good character evidence. This court reviews challenges to the statements made by the prosecutor for abuse of discretion when the defendant objects to them, but the district court still admits them over the objection.[21] United States v. Alaniz, 726F.3d 586, 615 (5th Cir. 2013) (citation omitted).First, we “decide whether the prosecutor made an improper remark and, if an improper remark was made, we must determine whether the remark affected the substantial rights of the defendant.” Id. (internal quotation and citation omitted). “To determine whether a remark prejudiced the defendant’s substantial rights, we assess the magnitude of the statement’s prejudice, the effect of any cautionary instructions given, and the strength of the evidence of the defendant’s guilt.” Id. (internal quotation and citation omitted). We have made these assessments of the prosecutor’s remarks at trial, and we conclude that even if we were to interpret these statements as Appellants do and hold that they were improper, the statements did not prejudice their substantial rights. See id. The prejudicial effect of any one of these comments, considered alone or together, was minimal. For instance, Appellants argue that prosecutors wrongfully stated that the jury could not consider evidence of their good character when determining whether they possessed an intent to defraud. But they concede that the district court charged the jury to consider Appellants’ “evidence of good general reputation or opinion testimony concerning: truth and veracity, honesty and integrity, or character as a law-abiding citizen” alongside “other evidence in the case.” This court has held that “for the ‘improper comment or questioning to represent reversible error, it generally must be so pronounced and persistent that it permeates the entire atmosphere of the trial.’” Alaniz, 726 F.3d at 616 (quotation omitted). Additionally, “[a] prosecutor’s closing remarks are reversible error when they ‘cast serious doubt on the correctness of the jury’s verdict.’” United States v. Bush, 451 F. App’x 445, 451 (5th Cir. 2011) (per curiam) (unpublished) (quoting United States v. Mares, 402 F.3d 511, 515 (5th Cir. 2005). Appellants have not met this high standard given that the effect of the statements was insignificant and the evidence against them was strong.[22] E. Time Limits Appellate courts typically review a district court’s implementation of time limits under an abuse of discretion standard. See, e.g., United States v. Hay, 122 F.3d 1233, 1235 (9th Cir. 1997) (citations omitted). But because no objections or requests for additional time were raised at trial, we proceed under plain error review. United States v. Gray, 105 F.3d 956, 965 (5th Cir. 1997). “Plain error is error which, when examined in the context of the entire case, is so obvious and substantial that failure to notice and correct it would affect the fairness, integrity or public reputation of judicial proceedings.” United States v. Vonsteen, 950 F.2d 1086, 1092 (5th Cir. 1992) (en banc) (internal quotations and citation omitted). Jester asserts that the district court’s imposition of a shared fifteen-hour time limit among all Appellants violated his constitutional right to present a complete defense. He concedes that the district court is entitled to a degree of control over the length of each party’s argument but argues that it implemented time limits on an unreasonable and arbitrary basis. We disagree. The Fourteenth Amendment’s Due Process Clause guarantees criminal defendants the right to “a meaningful opportunity to present a complete defense.” Holmes v. South Carolina, 547 U.S. 319, 324 (2006). Furthermore, the Sixth Amendment’s Confrontation Clause ensures a defendant’s right to cross-examine witnesses that the Government puts forth against him. See United States v. Moparty, 11 F.4th 280, 293 n.18 (5th Cir. 2021) (internal citations omitted) (explaining that a Confrontation Clause violation occurs where “a reasonable jury might have received a significantly different impression of the witness’s credibility had defense counsel been permitted to pursue his proposed line of cross examination”). Taken together, these constitutional guarantees protect a defendant from arbitrary, unreasonable restrictions on his right to present his case-in-chief and cross-examine the Government’s witnesses. See United States v. Morrison, 833 F.3d 491, 504 (5th Cir. 2016). District courts do not run afoul of either guarantee if an implemented time limit (1) allows a defendant the “meaningful opportunity to present a complete defense,” and (2) does not deprive the jury of the opportunity to “receive[ ] a significantly different impression of [a] witness’s credibility.” Holmes, 547 U.S. at 324; Moparty, 11 F.4th at 293 n.18. Jester alleges that the district court’s time limit violated his constitutional right to “put on a complete defense, confront the witnesses against him, and testify in his own defense.” He offers three examples in support of his argument. First, he contends that the time constraints reduced his testimony to “mere minutes.” Second, he asserts that the limitations demanded cursory discussions of the Government’s witnesses during cross-examinations. Finally, he argues that the composition of the defendants left him isolated in terms of how much time would be dedicated to his unique arguments. On the latter point, he notes that each defendant, except for him, was an executive at UDF, so he was outnumbered by the executive-defendants, who had no reason to prioritize his defense in the fifteen hours allotted to their case. In support, Jester relies substantially on our decision in Morrison. See 833 F.3d at 503. There, the district court was dissatisfied with the pace of the defendant’s trial and imposed a time limit on remaining witness examinations to expedite the proceedings. See id. It required the attorneys to estimate how long direct examinations would take for the remaining witnesses and enforced those estimations on the Government and the defense for the remainder of the trial. Id. The defendant ultimately appealed, arguing that the district court’s time limits deprived her of the opportunity to present a complete defense to her charges. On appeal, we rejected her argument because she failed to make an offer of proof sufficient to preserve her objection to the district court’s limits.[23] Id. at 505. Here, Jester’s arguments fail because he did not preserve his objection during the trial and the record states that the district court would have granted more time if he had requested it. First, Jester objected to the fifteen-hour time limit one time before the trial but failed to do so again at any other point during the trial.[24] Like in Morrison, the alleged injustice of the time constraint was substantially curbed by Jester’s failure to make an offer of proof to preserve his objection. See 833 F.3d at 505–06. Accordingly, his “lack of a contemporaneous offer of proof limits our ability on appellate review to determine whether the exclusion was harmful.” Id. at 505. Because he made no offer of proof during the trial, thus failing to preserve the objection, the district court did not err by implementing time constraints. Second, the record supports that Jester’s decision not to object to the time constraints was potentially a strategic maneuver by Appellants. As the district court explained, Appellants likely decided against requesting additional time during cross-examinations in hopes of making the Government’s burden more difficult. (“The reasonable inference is that [Appellants] strategically made no request for additional time to ensure the Government received no additional time so as to limit the Government’s opportunity to engage in full [cross-examination].”). Indeed, our precedent supports the district court’s inference. See Morrison, 833 F.3d at 504 (“If anything, it seems that time limits in criminal cases will generally pose more of a challenge for the prosecution as it typically presents far more of the evidence given that it has the burden of proof.”). Jester cannot benefit from the decision not to object to the time constraints during the trial, only to assert those same constraints as a reason for overturning his conviction. Ultimately, Jester fails to prove that the district court committed reversible error by imposing a fifteen-hour limit on both parties. Because the district court’s time constraint was not an obvious or substantial error that jeopardized the fairness or integrity of his trial, we hold in favor of the Government on this issue. See Vonsteen, 950 F.2d at 1092. F. Cumulative Error Doctrine Finally, Appellants maintain that they are entitled to relief under the cumulative error doctrine. We disagree. “The cumulative error doctrine . . . provides that an aggregation of non-reversible errors (i.e., plain errors failing to necessitate reversal and harmless errors) can yield a denial of the constitutional right to a fair trial, which calls for reversal.” United States v. Delgado, 672 F.3d 320, 343–44 (5th Cir. 2012) (internal quotations omitted) (citing United States v. Munoz, 150 F.3d 401, 418 (5th Cir. 1998)). “Cumulative error justifies reversal only when errors so fatally infect the trial that they violated the trial’s fundamental fairness.” Delgado, 672 F.3d at 344 (quotation marks omitted). We have explained that this doctrine is only applied in the “unusual case in which synergic or repetitive error[s] violate [ ] the trial’s fundamental fairness.” Id. Appellants argue that the cumulation of errors throughout their trial prejudiced the outcome and deprived them of constitutional rights. In response, the Government asserts that the cumulative error doctrine does not apply because Appellants have identified no errors and it has presented substantial evidence of guilt. The Government’s arguments are persuasive. Here, Appellants fail to highlight the multiple errors that they allege occurred throughout their trial. The only error they point to involves the district court’s jury instructions, and we have already determined that this error was, at most, harmless and does not warrant reversal of their convictions. See supra Part II.B.i. Absent additional errors, there is nothing for Appellants to cumulate. Thus, the doctrine is inapplicable here. See Delgado, 672 F.3d at 344. III. Conclusion For the foregoing reasons, the jury verdict is AFFIRMED in its entirety.