United States v. Andre De Moya

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued January 23, 2026 Decided July 7, 2026 No. 24-3013 UNITED STATES OF AMERICA, APPELLEE v. ANDRE DE MOYA, APPELLANT Consolidated with 24-3030, 24-3031 Appeals from the United States District Court for the District of Columbia (No. 1:19-cr-00158-1) (No. 1:19-cr-00228-1) Barry Coburn, appointed by the court, argued the cause and filed the brief for appellant Andre De Moya. Stephen C. Leckar, appointed by the court, argued the cause and filed the brief for appellant Anthony Merritt. Katherine M. Kelly, Assistant U.S. Attorney, argued the cause for appellee. With her on the brief were Jeanine Ferris  2 Pirro, U.S. Attorney, and Chrisellen R. Kolb and Eric Hansford, Assistant U.S. Attorneys. Before: MILLETT, PILLARD, and PAN, Circuit Judges. Opinion for the Court filed by Circuit Judge PAN. PAN, Circuit Judge: Heavy is the crime when a government official trades on his office for personal gain. “[A]nd no less heavy is the offense of the bribe giver.” Theodore Roosevelt, Third Annual Message to Congress (Dec. 7, 1903). While some crimes strike at individuals, public corruption strikes “at the foundation of all law.” Id. Andre De Moya and Anthony Merritt were tried and convicted on charges of quintessential acts of public corruption: bribing a District of Columbia official to reduce the tax liabilities of several local businesses — a scheme that cost the D.C. government about $2.3 million. On appeal, De Moya contends that the evidence was insufficient to support his conspiracy, bribery, and wire fraud convictions, and both defendants challenge the district court’s jury instructions. Merritt separately claims that his trial counsel was constitutionally ineffective and that the district court penalized him for exercising his right to stand trial. We find all of those arguments unpersuasive and affirm the judgments of the district court. I. Background The D.C. Office of Tax and Revenue (OTR) collects local taxes and conducts audits and criminal tax investigations. In addition, the OTR makes “adjustments” to tax bills that contain inaccuracies, upon the filing of a complaint by the taxpayer and the submission of supporting documentation. From 2011 to  3 2017, Vincent Slater worked at the OTR as the supervisor of the Adjustment Unit. Appellant Andre De Moya and his business partner, Arman Amirshahi, were part owners of several bars and nightclubs in the District, including “Echostage,” “Ultrabar,” and “Barcode.” Those businesses were required to collect sales-and-use taxes on items sold for immediate consumption, such as food and beverages, and to remit that tax revenue to the OTR. De Moya and Amirshahi deemed their tax bills too high and sought a shortcut to address the situation. They hired a self-employed “expediter,” appellant Anthony Merritt. Merritt publicly promoted himself as an expert who could help clients navigate D.C. business and tax regulations. But actually, Merritt served as an intermediary who reduced his clients’ tax obligations by passing cash bribes to Vincent Slater. Between September 2015 and December 2017, Merritt made multiple cash payments to Slater in exchange for decreasing the tax liabilities of De Moya’s and Amirshahi’s businesses. Slater made the tax reductions by accessing the OTR’s Integrated Tax System (ITS) and changing the system’s records to reflect lower amounts owed by the client businesses. Because Slater was a supervisor, he could not adjust the records in the ITS in his own name. So he used his coworkers’ ITS accounts to tamper with the system. To cover his tracks, Slater sometimes created false documents that purported to memorialize the tax reductions as settlements. Slater provided proof of the illegal adjustments to Merritt, who kept De Moya and Amirshahi apprised. The proof included photographs depicting computer screens that displayed diminished tax amounts. Merritt’s clients, including De Moya, typically paid half of their tax savings to Merritt and Slater, who split the  4 proceeds. The scheme ultimately cost the District approximately $2.3 million in lost tax revenue. The arrangement unraveled in January 2017, when the D.C. government conducted an audit that uncovered several large tax adjustments with no supporting documentation. The government’s investigation led it to Slater, and ultimately to Merritt and his clients. Amirshahi and another of Merritt’s clients, Charles Zhou, pled guilty to bribery in advance of being indicted, pursuant to cooperation agreements with the government. In May 2019, Slater, Merritt, and De Moya were indicted on charges of conspiracy (18 U.S.C. § 371), bribery (18 U.S.C. § 201(b)(1)–(2)), and wire fraud (18 U.S.C. §§ 1343, 1346). Slater and Merritt were also charged in a separate case involving similar conduct for another client. Slater pled guilty to one count of bribery, as well as several other offenses in the separate case, and he cooperated in the government’s prosecution of Merritt and De Moya. After a fourteen-day jury trial, Merritt was convicted on all counts. The jury found De Moya guilty of conspiracy to commit bribery, bribery, and two counts of wire fraud, but acquitted him on four other wire-fraud counts. The district court sentenced Merritt to 110 months’ imprisonment and three years of supervised release, and De Moya to thirty months’ imprisonment and three years of supervised release. De Moya and Merritt timely appealed. We have jurisdiction under 28 U.S.C. § 1291. II. Analysis De Moya and Merritt present a variety of arguments on appeal. First, De Moya contends that the evidence at trial was insufficient to prove that he was aware of the illicit nature of Merritt’s payments to Slater. He claims that, from his  5 perspective, all the cash payments to Slater appeared to be fees for legitimate “expediting” activities. Second, De Moya and Merritt jointly challenge the district court’s jury instruction on bribery, asserting that the district court inaccurately described the proper means of proving a quid pro quo. Third, Merritt argues that his trial counsel was constitutionally ineffective because she failed to object to the “unsoundness” of the Sentencing Guidelines’ economic Loss Table. And fourth, Merritt contends that the district court unlawfully “punished” him at sentencing because Merritt chose to go to trial, rather than plead guilty. We disagree on all fronts. A. Sufficiency of Evidence De Moya claims that the evidence at trial was insufficient to support his convictions. He argues that no reasonable jury could have concluded that he knew his payments to Slater “were bribes, or for the purpose of fraud,” as opposed to payments that he believed would be lawfully applied to his tax bills. Appellants Br. 10. But De Moya ignores evidence showing that he both had knowledge of the illicit nature of the tax-reduction scheme and intended to carry it out. That evidence included testimony from cooperating witnesses, which was corroborated by documents, text messages, and phone records. To prevail on a claim of insufficient evidence, De Moya “faces a high threshold and bears a heavy burden.” United States v. Borda, 848 F.3d 1044, 1053 (D.C. Cir. 2017) (cleaned up). In considering such a claim, we must “view[] the evidence in the light most favorable to the government” and affirm so long as “any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.” United States v. Littlejohn, 489 F.3d 1335, 1338 (D.C. Cir. 2007) (cleaned up). Moreover, we must “give full play to the right of  6 the jury to determine credibility, weigh the evidence, and draw justifiable inferences of fact.” Id. (cleaned up). Under that “highly deferential” standard of review, the evidence was sufficient. United States v. Brown, 125 F.4th 1186, 1195 (D.C. Cir. 2025). Both Amirshahi and Slater — two men at the heart of the illicit transactions — testified that De Moya played an important role in planning and carrying out the bribery scheme. Amirshahi testified that De Moya told him that they could pay bribes to reduce Ultrabar’s business-tax liabilities: De Moya “made it clear to [Amirshahi] that [they] ha[d] to pay cash” to an expediter (Merritt) who “had a connect at OTR” (Slater), and “that [their] taxes would be taken out,” even though “it was not proper.” Gov’t S.A. 239. Moreover, at a meeting in the OTR lobby, Slater explained to Merritt, in De Moya’s presence, what he “could do” to make “adjustments in the system” to lower Ultrabar’s taxes. J.A. 484–85. After the meeting, De Moya provided Ultrabar’s employer identification number to Merritt so that the plan could be executed. After a series of text messages and phone calls, Slater reduced Ultrabar’s tax liabilities by approximately $216,000. Amirshahi testified that he and De Moya then gave cash to Merritt, with the expectation that he would pay Slater, because “that was the arrangement [they all] had.” Gov’t S.A. 194. De Moya, Merritt, and Slater secured tax reductions in a similar fashion on three more occasions, and the evidence demonstrated De Moya’s corrupt intent. For example, after an abrupt reduction in Barcode’s tax bill, Amirshahi tried to make the change appear legitimate by asking De Moya to obtain a written “agreement” or “settlement in full letter” from the OTR. Gov’t S.A. 210–11, 216. He noted that “this could come back” and “if audited, we have no paperwork.” Id. at 211–12. De Moya said that he would try to secure such false  7 documentation, but it would be difficult, noting that Amirshahi’s request was “[k]illing [him]” because Slater “ha[d] already” reduced Barcode’s tax obligation in the OTR system. Id. at 87–88. Amirshahi and De Moya also exchanged text messages about making a $10,000 payment to Merritt and Slater for the Barcode transaction. Amirshahi explained at trial that the $10,000 payment was in exchange for Slater reducing their taxes “without making a legitimate payment.” Id. at 219. And still there is more. The government presented evidence that De Moya, Merritt, and Slater met in a Safeway Starbucks in June 2016. At that meeting, De Moya offered to pay Slater to decrease the tax obligations of Echostage. Slater testified that he made the requested adjustments because De Moya paid him, and not for any legitimate reason. De Moya later coordinated with Slater to remove all of Echostage’s outstanding tax obligations so that the club could renew its alcohol license. In short, a reasonable jury plainly could infer from the record evidence that De Moya knowingly and intentionally participated in the bribery scheme. B. Jury Instructions We next consider De Moya and Merritt’s challenge to the district court’s jury instructions about the offense of bribery. They note — as they did before the district court — that the government can prove bribery only by “directly, specifically, [and] causally” linking each cash payment to each specific official act. Appellants Br. 18. According to De Moya and Merritt, their convictions must be reversed because the district court lowered the bar, allowing the jury to convict them for mere payments to buy Slater’s “good will or his general disposition to help . . . in non-particularized ways.” Id. at 17 (citation omitted). We disagree: Although the jury instructions  8 diverged from the governing case law in one respect, that divergence was harmless error. Whether the jury was properly instructed is a legal question that we review de novo. United States v. Orenuga, 430 F.3d 1158, 1166 (D.C. Cir. 2005). Our task is to “determine whether, taken as a whole, the instructions accurately state the governing law.” Id. (cleaned up). “An error in a jury instruction does not require reversal if the error was harmless.” United States v. Cicero, 22 F.3d 1156, 1161 (D.C. Cir. 1994). Here, the government charged De Moya and Merritt with violating the federal bribery statute. That law “prohibits quid pro quo corruption — the exchange of a thing of value for an ‘official act.’” McDonnell v. United States, 579 U.S. 550, 574 (2016) (quoting 18 U.S.C. § 201). To convict a defendant of bribing a public official, the government must prove “that something of value was corruptly given, offered, or promised . . . with intent, inter alia, ‘to influence any official act.’” United States v. Sun-Diamond Growers of Cal., 526 U.S. 398, 404 (1999) (quoting 18 U.S.C. § 201(b)(1)). In McDonnell, the Supreme Court clarified that the “official act” in question must be “something specific and focused” that either is “pending” or “may by law be brought before a public official.” 579 U.S. at 574 (cleaned up). The Court further explained that the statute’s terms are “relatively circumscribed,” indicating that the “official act” must be “the kind of thing that can be put on an agenda, tracked for progress, and then checked off as complete.” Id. at 570. A lower court should instruct a jury that to find an individual guilty of bribery, it “must identify” the official act, as the Court defined that term in McDonnell, and find that the public official “made a decision  9 or took an action — or agreed to do so — on the identified” official act. Id. at 578–79 (emphasis in original). The district court’s jury instructions largely aligned with the governing law. The court described the elements of bribery, gave the statutory definition of “official act,” and correctly explained the bribery statute’s narrow scope, consistent with McDonnell. J.A. 521–24; see also Sun- Diamond Growers, 526 U.S. at 404–05 (describing the elements of bribery). But the district court then instructed the jury that the government could prove its allegations pursuant to a “course of conduct” theory of bribery, which it described in the following way: Now, the government does not need to connect each thing of value to a specific official act or violation of an official duty. Instead, the government can show a course of conduct, that is a pattern of giving things of value to Vincent Slater in exchange for Mr. Slater’s pattern of being influenced in his official act or acting in violation of his official duties. The government must show beyond a reasonable doubt that the things of value were given to Vincent Slater with the intent or accepted with the understanding that in return an official act would be influenced or Mr. Slater would be motivated or encouraged to violate an official duty. J.A. 525–26. De Moya and Merritt argue that the district court’s jury instruction mischaracterizes the requirements of the bribery  10 statute. They note that basing a bribery conviction on a “pattern” of payments in exchange for a “pattern” of “being influenced” departs from McDonnell, which requires that payments be exchanged for official acts that are known, specific, and circumscribed. They contend that the government had to prove the connection between each of their payments and each of Slater’s official acts, and that the district court’s instructions impermissibly lightened the government’s burden of proof. De Moya and Merritt’s argument has some force: The district court’s instruction relying on a “pattern of giving things of value” in exchange for a “pattern of being influenced” is in tension with established precedents. J.A. 525–26. The Supreme Court has made clear that bribery must feature a quid pro quo, and the government therefore must prove “a specific intent to give or receive something of value in exchange for an official act.” Sun-Diamond Growers, 526 U.S. at 404–05 (emphasis in original). In other words, a bribery conviction needs to rest on a “connection” between the intent to give or receive a thing of value and a specific official act. Id. at 405. And the official act must be “something specific and focused that is pending or may by law be brought before any public official.” McDonnell, 579 U.S. at 579 (cleaned up). 1 1 We note, however, that the bribery statute is not limited to addressing only one-for-one exchanges. There may be cases in which many things of value are exchanged for a single official act; one thing of value is exchanged for multiple official acts; or many things of value are exchanged for multiple official acts. At bottom, the government must prove that the thing or things of value were given “with intent . . . to influence” a specific and focused official act or acts. 18 U.S.C. § 201(b)(1)(A); see also Sun-Diamond Growers, 526 U.S. at 404–05; United States v. McCabe, 103 F.4th 259, 284–85 (4th Cir. 2024) (The bribery statute does not “reward  11 Nevertheless, we conclude that any error in the district court’s instructions was harmless. An error is harmless if it appears “beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained.” Chapman v. California, 386 U.S. 18, 24 (1967); see also Neder v. United States, 527 U.S. 1, 4 (1999) (applying the harmless-error rule of Chapman in the context of an erroneous jury instruction). Under that standard, we have held that an erroneous jury instruction is “harmless if the jury necessarily found facts that would have satisfied a proper instruction.” United States v. Johnson, 216 F.3d 1162, 1166 (D.C. Cir. 2000). Here, the instructional error did not contribute to the jury’s verdict because the government only presented evidence of a quid pro quo scheme: Discrete payments were exchanged for discrete adjustments of specific tax obligations. In other words, each bribe was a transaction involving an identifiable price for an identifiable official act. The evidence showed that De Moya and Merritt communicated with Slater and others about each illegal tax adjustment and followed up with corresponding communications about the payments owed for each event. Moreover, witnesses on both sides of the transactions testified that the cash payments were in exchange for Slater’s official acts. Thus, we conclude “beyond a reasonable doubt” that the jury convicted De Moya and Merritt on a proper quid pro quo theory. United States v. Rhone, 864 F.2d 832, 836 (D.C. Cir. 1989) (citation omitted). Stated differently, the evidence against De Moya and Merritt “was untouched by the instruction, and . . . that evidence showed beyond doubt that [they] committed the offenses for which the jury found [them] guilty.” United States v. Alston-Graves, 435 F.3d 331, 342 (D.C. Cir. 2006). corrupt bribery schemes” simply because they “involve multiple exchanges over a period of time.”).  12 C. Ineffective Assistance of Counsel Merritt next argues that he received ineffective assistance of counsel at his sentencing hearing. He contends that the Sentencing Guidelines’ economic Loss Table, U.S.S.G. § 2B1.1, had an outsized influence in the calculation of his Guidelines range, and that his counsel should have sought a downward variance by arguing that the policies underlying the Loss Table are flawed. He asks us to remand so that the district court may evaluate his policy arguments. Because Merritt has not demonstrated that the alleged error was prejudicial, we deny his ineffective-assistance claim. To succeed on a Sixth Amendment claim of ineffective assistance of counsel, a defendant must show two things: (1) “that counsel’s performance was deficient,” meaning “counsel’s representation fell below an objective standard of reasonableness,” and (2) “that the deficient performance prejudiced the defense,” meaning “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland v. Washington, 466 U.S. 668, 687–88, 694 (1984). That two- part test is “highly demanding.” Kimmelman v. Morrison, 477 U.S. 365, 382 (1986). And although we typically remand a “colorable and previously unexplored claim of ineffective assistance,” United States v. Rashad, 331 F.3d 908, 908 (D.C. Cir. 2003), “we decline to remand when the record ‘conclusively shows’ the defendant is not entitled to relief,” United States v. Marshall, 946 F.3d 591, 596 (D.C. Cir. 2020) (quoting Rashad, 331 F.3d at 910). Merritt’s claim of ineffective assistance of counsel centers on his 110-month prison sentence. To arrive at that sentence, the district court first calculated the Sentencing Guidelines range. The range reflected several enhancements based on the  13 nature and circumstances of the offenses of conviction, the largest of which related to the significant financial loss incurred by the District. Specifically, pursuant to the Sentencing Guidelines’ Loss Table, a sixteen-level enhancement was warranted because Merritt’s offenses resulted in a financial loss of more than $1.5 million. See U.S.S.G. § 2B1.1(b)(1)(I). At sentencing, the district court heard and addressed arguments about the Loss Table, including trial counsel’s argument that the sixteen-level enhancement overstated Merritt’s involvement in the offenses. Counsel also argued that, based on sentences imposed on other defendants in similar cases, Merritt should receive a below-Guidelines sentence. The district court, citing those comparator cases, varied downward from the Guidelines range of 188–235 months to impose a sentence of 110 months. That downward variance corresponded to a reduction in the loss amount from over $1.5 million to $150,000. See U.S.S.G. § 2B1.1. Merritt does not dispute that his Guidelines range was correctly calculated or that his counsel successfully argued for a substantial downward variance. Instead, he asserts that his counsel also should have argued for a downward variance on the ground that the Guidelines’ Loss Table is an “infirm,” “unsound,” and “irrational” tool that results in overly punitive Guidelines recommendations. Appellants Br. 25–60. Merritt contends that the U.S. Sentencing Commission, when formulating the Loss Table, committed a series of policy errors, including raising the enhancement levels “due to pressure from Congress and the Department of Justice” and wrongly “presum[ing] that increased sophistication and moral culpability correspond to greater losses and disproportionally lengthier ranges.” Id. at 26. Merritt concedes that it is “impossible to say” how the district court would have responded if those policy arguments had been made, id. at 56,  14 but he notes that some other courts have granted downward variances based on similar, policy-related considerations. Merritt has not shown that counsel’s failure to press the policy arguments in question was prejudicial. To establish prejudice, Merritt must demonstrate “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different” — here, that such arguments would have resulted in a lower sentence. Strickland, 466 U.S. at 694. Merritt cannot clear that bar by essentially putting a different label on an issue that his trial counsel effectively identified and argued — i.e., that a sentence below the loss-driven Guidelines range was appropriate. The court found counsel’s request persuasive and granted a substantial downward variance. Merritt’s sentence was reduced to 110 months, even though the low end of his Guidelines range was 188 months. A recasting of an argument that the court already considered and accepted does not establish a “substantial” likelihood of a different sentence. Harrington, 562 U.S. at 112. We also reject Merritt’s extraordinary proposal that we order the government to maintain a database of cases in which the Loss Table has impacted defendants’ sentences. Appellants Br. 59–60 (arguing that no database “reveals how the local trial judges have sentenced offenders impacted by the Table” and that the government should be responsible for maintaining such a database). We only “sparingly” exercise our supervisory power. United States v. Jones, 433 F.2d 1176, 1181–82 (D.C. Cir. 1970) (quoting Lopez v. United States, 373 U.S. 427, 440 (1963)). And we do so only “to implement a remedy for violation of recognized rights; to preserve judicial integrity by ensuring that a conviction rests on appropriate considerations validly before the jury; and finally, as a remedy designed to deter illegal conduct.” United States v. Hasting, 461 U.S. 499,  15 505 (1983) (citations omitted). None of those purposes is implicated here. D. Trial Penalty Finally, Merritt contends that the district court punished him for exercising his right to stand trial. He claims that it is unfair that he received a sentence that is much higher than what was contemplated by a plea bargain that he initially tried to accept. That argument rests on a fundamental misunderstanding of how plea bargaining works, and we easily reject it. The relevant facts are straightforward. On May 16, 2022, Merritt entered a guilty plea to all counts in both indictments. In the plea agreement that Merritt negotiated with the government, the parties agreed to an estimated offense-level calculation that (1) did not include a four-level enhancement for acting as an organizer or leader under U.S.S.G. § 3B1.1(a), and (2) included a two-level reduction for acceptance of responsibility under § 3E1.1. The plea agreement projected a Guidelines range of 78–97 months. But that negotiated plea agreement fell apart. At a November 2022 sentencing hearing, Merritt’s counsel represented that Merritt maintained his innocence and denied all criminal culpability. That led the district court to vacate Merritt’s guilty plea and to set a trial date, without any objection from Merritt. After Merritt was convicted on all counts, his probation officer prepared a presentence report with a new Guidelines- range calculation. The post-trial Guidelines range rested on an offense level that included a four-level enhancement under U.S.S.G. § 3B1.1(a) for being a leader/organizer; a two-level enhancement under § 3C1.1 for obstruction of justice based on  16 Merritt’s false trial testimony; and no reduction for acceptance of responsibility. The resulting range was 188–235 months. As noted, the court ultimately varied downward and imposed a sentence of 110 months. Merritt argues for the first time on appeal that the post-trial Guidelines range reflected an unconstitutional penalty for going to trial instead of pleading guilty. We review that unpreserved claim only for plain error. United States v. Locke, 664 F.3d 353, 357 (D.C. Cir. 2011). Under the stringent plain- error standard, Merritt must show a “clear” or “obvious” error that affected his substantial rights. United States v. Olano, 507 U.S. 725, 734 (1993) (citations omitted); see also United States v. Gomez, 431 F.3d 818, 822 (D.C. Cir. 2005) (applying the plain-error standard to a sentencing challenge). Even then, correcting the forfeited error is subject to the reviewing court’s discretion, which we exercise only when the error “seriously affect[ed] the fairness, integrity or public reputation of judicial proceedings.” Olano, 507 U.S. at 735–36. He falls far short of meeting that standard. Merritt’s theory is that, because his vacated plea agreement projected a Guidelines range of 78–97 months and his post-trial range “skyrocketed” to 188–235 months, the increase must have been a penalty for going to trial. Appellants Br. 71. Hardly. Plea bargaining routinely involves the government’s agreement to forgo otherwise-applicable Guidelines enhancements in exchange for a defendant’s agreement to plead guilty and accept responsibility. When the bargain disappears, so do the negotiated concessions, and the government is once again free to seek every enhancement that the law and the facts support. And “there is nothing impermissible (or even unusual)” about “the prosecutor recommending a longer sentence after trial than before trial.” United States v. Mejia, 597 F.3d 1329, 1344 (D.C. Cir. 2010).  17 Merritt has shown no error, let alone the clear or obvious error that plain-error review demands. * * * A jury convicted De Moya and Merritt of multiple counts related to their bribery scheme, and we discern no reversible error. The government’s evidence sufficiently proved that De Moya and Merritt traded cash for Slater’s official acts of reducing specific tax liabilities; and that evidence satisfied the definition of bribery. Moreover, Merritt was neither prejudiced by his lawyer’s failure to make policy-based objections to the Sentencing Guidelines’ Loss Table, nor punished by the district court for his decision to stand trial. We therefore affirm the judgments of the district court. So ordered.