dissenting:
On our panel rehearing, my friends in the majority assume for the sake of argument that the improper and prejudicial testimony elicited and relied upon by the Government to convict Appellant Alejandro Garcia-Lagunas amounts to constitutional error. There is no need for assumptions; this is unequivocally, and admittedly, a case of constitutional error. Moreover, the majority concludes that the Government’s error, as compounded by the district court’s failure to correct it, even if of a constitutional magnitude, was harmless beyond a reasonable doubt. I, however, remain compelled to conclude that the Government did not carry its burden of proving beyond a reasonable doubt that its clearly unconstitutional use of a blatant ethnic generalization did not contribute to the jury’s verdict. Accordingly, I respectfully dissent from the majority’s decision to refuse, once again, to order a new trial.
I.
During trial, Garcia-Lagunas’s counsel sought to show that Garcia-Lagunas was, at most, a common drug abuser and not a sophisticated drug distributor who traf*498ficked large volumes of cocaine as alleged in the indictment. To make this distinction, counsel strategically questioned Government witnesses on cross-examination about Garcia-Lagunas’s meager lifestyle, a lifestyle devoid of the flamboyant trappings derived from drug proceeds that one might expect to surround a high-volume narcotics distributor. For example, the cross-examinations of convicted drug dealer Ronnie Réed and Detectives Shawn Collins and Pedro Orellano tended to establish that Garcia-Lagunas lived a life of truly limited means. Reed testified that he never knew Garcia-Lagunas to have any “fancy things” such as jewelry, firearms, or vehicles. J.A. 222. Detectives Collins and Orel-lano testified that, on the evening of his arrest, Garcia-Lagunas was found shirtless and shoeless in the “kitchen/living room area” of a small trailer at 353 West-cott Drive in which he had recently begun renting a room for less than $350 per month. J.A. 103-04, 315. Detective Collins described the bedroom belonging to Garcia-Lagunas as in “disarray” and explained to the jury that it looked as though Garcia-Lagunas had yet to unpack since moving to the trailer, as his belongings were scattered throughout the small room in laundry baskets. J.A. 120.
Further law-enforcement testimony showed that detectives searched the 353 Westcott trailer, as well as three other trailers in and around Robeson County, North Carolina, where it was alleged that Garcia-Lagunas had previously sold cocaine, and not one of the searches uncovered evidence of profits consistent with an individual allegedly trafficking hundreds of thousands of dollars’ worth of cocaine. In fact, the only items of value that the searches uncovered, a .32 caliber revolver and $600 in U.S. currency, were described as having been brought to the 353 West-cott trailer the night of Garcia-Lagunas’s arrest by Brian Jacobs, allegedly in exchange for three-quarters of an ounce of powder cocaine. J.A. 298, 320-21, 342. However, no powder cocaine was actually found at that trailer or any other of the trailers linked to Garcia-Lagunas.
Testimony revealed that the only substances discovered by law enforcement to lab-test positive for the presence of cocaine were two baggies containing user amounts of crack cocaine, for which Garcia-Lagu-nas was not charged. J.A. 117, 122, 124, 404. Counsel for Garcia-Lagunas provided an explanation for the presence of those drugs by questioning the Government witnesses about his client’s personal drug use. Three different Government witnesses testified that they had observed Garcia-Lagu-nas use drugs, J.A. 320, 349, 355, 376, and Jacobs testified that, for a number of years, Garcia-Lagunas had actually purchased small amounts of cocaine from him for Garcia-Lagunas’s personal use, J.A. 354. Further, several detectives explained to the jury that, on the night of his arrest, Garcia-Lagunas had white powder under his nose, which, together with his dilated pupils and erratic movements, suggested that he had ingested cocaine immediately before the arrival of the law-enforcement officers. J.A. 103-04, 248, 283.
To bring home the defense theory of the case, counsel emphasized during cross-examination that Garcia-Lagunas’s meager lifestyle did not square with the portrait that the Government was painting of a sophisticated, large-volume drug trafficker. Counsel astutely presented the theory by offering the jury the opportunity to contrast Garcia-Lagunas’s lifestyle with that of Reed, one of Garcia-Lagunas’s alleged purchasers. Counsel questioned Detective Collins and Reed on cross-examination about Reed’s drug-trafficking operation and the proceeds that Reed had amassed during the four .years that he sold drugs prior to his 2012 arrest on federal drug *499trafficking charges. J.A. 153-55, 225-30. During searches of Reed’s family home and stash house, officers found more than $100,000 in U.S. currency, multiple telephones, a 2008 Infíniti, a 2006 Chevy Impala, a 2004 Acura, a 2004 BMW, a 2002 Lincoln Navigator, and multiple firearms. J.A. 154-55. The officers also found contraband consistent with a large-scale drug-trafficking operation, including more than 180 grams of crack cocaine, more than three-and-a-half kilos of powder cocaine, 240 grams of marijuana, money that the Fayetteville police department had used to conduct controlled buys from Reed, a cocaine press, and a money counter. J.A. 225-29.
The upshot of all of this, contrary to the majority opinion’s one-sided spin on the evidence, is that there were two, competing narratives before the jury. And it was the jury’s call, not the job of the members of this appellate panel, to judge the credibility of all of the evidence, weigh it accordingly, and reach a fair and impartial verdict in accordance with law.
Ultimately, counsel for Garcia-Lagunas hoped the testimony he elicited would prompt the following question from at least one juror (because of course, he only needed to garner the interest of one juror to raise a possibility of a more beneficial outcome than the one he got): how can a man who is allegedly responsible for selling hundreds of thousands of dollars’ worth of cocaine1 not have on hand any discernable direct or indirect proceeds of any kind on the day of his arrest, with zero indication from any source that his arrest was imminent? Any experienced (and even an inexperienced) Assistant United States Attorney prosecuting cases in this Circuit would fully expect (and be prepared for) this kind of defense tack on this record.
My friends in the majority may not think much of defense theories in general, or of Garcia-Lagunas’s theory in particular, but that is what it was, fully supported by admissible evidence, and well within the realm of plausible disputation by a lawyer committed to her Fifth- and Sixth-Amendment-based obligations to her client.2 As in any prosecution, whether for a crime involving the infliction of unspeakable violence upon actual victims, or in the prosecution of the most plain-vanilla so-called “white collar” offense, and any prosecution in between, the defendant in our system is entitled to have the jury grapple, if it must, with his defense theory, unaided by blatantly foul blows delivered by the prosecution, abetted by the trial judge, in the use of racial or ethnic entreaties aimed at undermining or dismissing outright the defense theory of the case. But that is precisely what happened here.
As Garcia-Lagunas’s defense theory became apparent during trial, the Govern*500ment seemingly recognized for the first time the absence of drug trafficking proceeds as a potential weakness in its case, a case in which it now argues the evidence of guilt was always overwhelming. The Government opted not to cure the ostensible weakness through the introduction of admissible evidence by, for example, moving to admit proof of wire transfers from Gar-da-Lagunas to individuals in Mexico. Either because such evidence did not exist3 or because the Government failed to adequately prepare its case, the Government instead sought to counter the defense theory by eliciting an outrageous ethnic stereotype about the propensity of “Hispanic drug traffickers” to live modestly while sending “the majority .if not all the proceeds back to their native countries.” J.A. 270. The Government then highlighted this irrelevant and unsupported racial generalization at the outset of its rebuttal closing argument, stating:
Ladies and Gentlemen, what did Detective Orellano tell you about Hispanic drug trafficking organizations [sic] and about what they do with their money? He told you that they package that money and they send it back to their home country as part of the drug trafficking organization. That’s why we don’t have an extravagant lifestyle associated with this Defendant, fancy cars, any of the things like Ronnie Reed has talked about.
J.A. 520.
The relative ability of this particular stereotype to sway one or more jurors is evidenced by its extraordinary confirming effect on the presiding judge. In response to a renewed objection to Detective Orella-no’s testimony, the trial judge held a bench conference and admitted that he “wasn’t quite sure the relevance of’ the Detective’s testimony regarding Hispanic drug traffickers, but that, “based on [his] experience, ... most Latins [sic] send money home whether they’re drug dealers or not.”4 J.A. 273. The Government admittedly hoped the jurors would draw a similar inference when rendering a verdict. J.A. 273.
To counter Garcia-Lagunas’s primary defense theory and cure a perceived hole in its case, the Government offered up a generalization about Garcia-Lagunas’s ethnicity to the jury. The Government hoped that, like the presiding judge, the *501jurors would believe that Garcia-Lagu-nas’s modest lifestyle could not rationally undermine allegations that he distributed hundreds of thousands of dollars’ worth of cocaine because he assuredly had been sending his significant proceeds back to his native country, electing to live like a pauper in the United States.
Tellingly, even the Government concedes that the elicitation of Detective Orel-lano’s testimony during re-direct and the recitation of the testimony at the outset of the rebuttal closing argument amounted to a constitutional error. Oral Argument at 20:38-20:51, United States v. Garcia-Lagunas, No. 14-4370 (Sept. 17, 2015), available at http://coop.ca4.uscourts.gov/OAarchive/ mp3/14-4370-20150917.mps. During oral argument, when asked whether the error amounted to constitutional error, counsel for the Government responded unequivocally, “Yes.” Id. The Panel then asked, as a result of the Government’s belief that constitutional error had occurred, whether it was the Government’s burden “to prove beyond a reasonable doubt that the error had no substantial effect on the jury’s verdict.” Id. In response, counsel for the Government firmly stated, “That’s correct.” Id.
Accordingly, because it is clear that “[a]ppeals to racial, ethnic, or religious prejudice during the course of a trial violate a defendant’s Fifth Amendment right to a fair trial,” United States v. Cabrera, 222 F.3d 590, 594 (9th Cir. 2000), I see no reason to resort to assumptions in addressing Garcia-Lagunas’s appeal. See, e.g., United States v. Vue, 13 F.3d 1206, 1213 (8th Cir. 1994) (concluding that a constitutional error occurs when the Government “invite[s] the jury to put [a defendant’s] racial and cultural background into the balance in determining their guilt”). The Government’s appeal to an unabashed ethnic generalization was plainly a constitutional error, and as a result, it is the Government’s burden to prove that its error was harmless beyond a reasonable doubt. And, for the reasons set forth below, I cannot conclude that the Government carried that burden in this case.
II.
As the majority explains, not all constitutional errors mandate reversal. However, when a non-structural constitutional error occurs, the reviewing court may only disregard the error so long as the Government can carry its burden of demonstrating that the error was “harmless beyond a reasonable doubt.” Neder v. United States, 527 U.S. 1, 7, 119 S.Ct. 1827, 144 L.Ed.2d 35 (1999) (quoting Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967)). Here, the majority concludes that the Government has met its harmless-error burden because “even without the [Government’s improper use of an ethnic stereotype[,] a rational jury still would have arrived at that verdict.” Ante at 489.5 For several reasons, I believe this analysis grievously misses the mark.
*502Rule 52(a) of the Federal Rules of Criminal Procedure mandates that “[a]ny error, defect, irregularity, or variance that does not affect substantial rights must be disregarded.” Fed. R. Crim. Pro. 52(a). Rule 52(a)’s “emphasi[s] [on] ‘substantial rights’ ” serves two important purposes. Chapman, 386 U.S. at 22, 87 S.Ct. 824. First, it stresses the significance of the factfinding process, recognizing that, at its heart, “the central purpose of a criminal trial is to decide the factual question of the defendant’s guilt or innocence.” Neder, 527 U.S. at 18, 119 S.Ct. 1827 (quoting Delaware v. Van Arsdall, 475 U.S. 673, 681, 106 S.Ct. 1431, 89 L.Ed.2d 674 (1986)). Second, it “promotes public respect for the criminal process by focusing on the underlying fairness of the trial rather than on the virtually inevitable presence of immaterial error.” Van Arsdall, 475 U.S. at 681, 106 S.Ct. 1431. Accordingly, in practice, Rule 52(a) works to “save the good” — those convictions that, while the product of an imperfect trial, were the subject of “constitutional errors which in the setting of [the] particular case [were] so unimportant and insignificant that they may ... be deemed harmless” — while excising the bad — those convictions that might have been impacted by the complained of error. Chapman, 386 U.S. at 22-24, 87 S.Ct. 824.
The Supreme Court applied Rule 52(a)’s harmless-error analysis in Neder when a criminal defendant challenged a district court’s failure to submit the materiality element of the defendant’s tax-fraud charges to the jury. 527 U.S. at 4, 119 S.Ct. 1827. The Supreme Court began by stating the overarching test for determining whether a constitutional error is harmless: “[Wjhether it appears ‘beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained.’ ” Id. at 15, 119 S.Ct. 1827 (quoting Chapman, 386 U.S. at 24, 87 S.Ct. 824). To answer that question, the Court first considered the ways in which the Government could carry its burden of establishing .the materiality element. Id. at 16, 119 S.Ct. 1827. It explained that, “[i]n general, a false statement 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” but noted that “several courts have determined that any failure to report income is material.” Id. (alterations in original) (citations and internal quotation marks omitted). The Court then described how, at trial, the Government had “introduced evidence that Neder failed to report over $5 million in income from the loans he obtained,” and that “[t]he failure to report such substantial income incontrovertibly” established the materiality element of his charges. Id.
After emphasizing that Neder did not even attempt to contest materiality, either before the jury or on appeal, the Supreme Court concluded that, “[i]n this situation, where a reviewing court concludes beyond a reasonable doubt that the omitted element was uncontested and supported by overwhelming evidence, such that the jury *503verdict would have been the same absent error, the erroneous instruction is properly found to be harmless.” Id. at 16-17, 119 S.Ct. 1827. And specifically applying the test set forth in Chapman, the Court further noted that, “We think it beyond cavil here that the error ‘did not contribute to the verdict obtained.’ ” Id. at 17, 119 S.Ct. 1827 (quoting Chapman, 386 U.S. at 24, 87 5.Ct. 824).
It is trae that reading portions of Neder in isolation and out of context from the remainder of the Supreme Court’s extensive harmless-error jurisprudence, as the majority does in this case, could lead one to conclude that all a reviewing court must do to satisfy itself of an error’s harmlessness is ask whether it is beyond a reasonable doubt that a jury would have found the defendant guilty if the error had never occurred. Such an approach, however, is misplaced and ill-advised.
First, it fails to give proper credence to the narrow nature of the holding in Neder. In summarizing why its harmless-error inquiry reached “an appropriate balance between society’s interest in punishing the guilty [and] the method by which decisions of guilt are to be made,” the Court took care to explain that,
[ i]n a case such as this one, where a defendant did not, and apparently could not, bring forth facts contesting the omitted element, answering the question whether the jury verdict would have been the same absent the error does not fundamentally undermine the purpose of the jury trial guarantee.
Id. at 18-19, 119 S.Ct. 1827. Unlike Neder, the present appeal clearly does not fit within the narrow subset of cases where the fact that a rational jury could have found the defendant guilty absent the erroneous omission necessarily dictates that the error did not contribute to the verdict.
Second, merely assuring oneself that a rational jury would have nonetheless convicted the criminal defendant absent the error fails to heed important guidance from the Supreme Court. The Supreme Court has explained that, in the case of affirmative error, a reviewing court should not simply confine itself to the abstract and ask “whether, in a trial that occurred without the error, a guilty verdict would surely have been rendered.” Sullivan v. Louisiana, 508 U.S. 275, 279, 113 S.Ct. 2078, 124 L.Ed.2d 182 (1993). Rather, a Rule 52(a) harmless-error analysis requires us to consider “what effect [the error] had upon the guilty verdict in the case at hand” and assure ourselves that “the guilty verdict actually rendered in [the] trial was surely unattributable to the error.” Id. This is so because, when we frame the harmless-error analysis in the abstract and remain content to imagine a world where the Government exclusively relied upon admissible evidence, we not only fail to consider the error’s actual effect on the jury, but we also “improperly conflate[ ] sufficiency-of-the-evidence review with the appropriate Chapman standard.” United States v. Holness, 706 F.3d 579, 598 (4th Cir. 2013) (quoting Virgin Islands v. Martinez, 620 F.3d 321, 338 (3d Cir. 2010)). Here, when I consider the specifics of the Government’s prosecution of Garcia-Lagunas and the unique nature of the unconstitutional testimony and the prosecution’s arguments based thereon, I am unable to conclude beyond a reasonable doubt that the complained of error did not contribute to the jury’s verdict.6
*504From opening statements through closing arguments, Garcia-Lagunas’s trial lasted a mere thirteen hours spread over the course of three days. Within those thirteen hours, when confronted with a gaping hole in its confident characterization of .Garcia-Lagunas as a sophisticated drug trafficker responsible for the distribution of more than 39 kilos of cocaine valued at more than $1 million, the Government knowingly and purposefully elicited inadmissible and prejudicial testimony from a law-enforcement officer. While the Government did not qualify Detective Or-ellano as an expert, it repeatedly requested that he testify pursuant to his “training and experience” investigating Hispanic drug trafficking organizations. J.A. 272-74. Accordingly, when Detective Orellano explained to the jury that it did not need to be concerned that the investigation into Garcia-Lagunas recovered no proceeds and instead revealed a man of abject poverty because such evidence was actually “consistent with the method of operation of Hispanic drug traffickers,” he did so with an authority that any juror would have had difficulty discounting.7 The Government no doubt hoped that Detective Orellano’s years of experience investigating “Hispanic drug traffickers” would carry weight with the jury, and the import of his testimony to the Government’s case is evidenced by the Government’s decision to begin its rebuttal closing remarks by asking, “What did Detective Orellano tell you about Hispanic drug trafficking organizations and about what they do with their money?” J.A. 520.
The weighty impact of this unconstitutional testimony and argument is illuminated further when one considers that the Government’s case against Garcia-La-gunas relied almost exclusively upon criminal defendants testifying pursuant to plea agreements and circumstantial evidence. As Garcia-Lagunas pointed out to the jury, the Government was unable to present any direct evidence that Garcia-Lagunas participated in a drug trafficking conspiracy through law-enforcement testimony. When cross-examining Detective Collins, Garcia-Lagunas confirmed that, despite a lengthy investigation into a' “Mexican drug trafficker named Alex,” the Government did not have direct evidence of any hand-to-hand transactions or controlled buys involving Garcia-Lagunas. J.A. 152-58. This lack of direct evidence is especially probative when one considers that law-enforcement officers had Reed, Garcia-Lagunas’s alleged purchaser, under “intense surveillance” while Reed was allegedly visiting Garcia-Lagunas at least three times a week to purchase cocaine. J.A. 152-53, 204-06. Despite these frequent and consistent rendezvous, law enforcement never saw Reed with Garcia-Lagunas and did not become aware of the locations of the meetings until after Reed was arrested on his own federal drug trafficking charges. Id.
Of greatest significance to this appeal’s harmless-error analysis, however, is not the highly prejudicial method by which the unconstitutional evidence was presented to the jury, the Government’s repeated and strategic reliance upon the evidence, or *505the strength vel non of the Government’s case against Garda-Lagunas. The most critical factor here is the uniquely troublesome nature of the unconstitutional testimony. Not only do “[a]ppeals to racial, ethnic, or religious prejudice during the course of a trial violate a defendant’s Fifth Amendment right to a fair trial,” but on a broader note, they also place the public’s trust in “[t]he fairness and integrity of [our] criminal” justice system at risk. Cabrera, 222 F.3d at 594, 597; see also Pena-Rodriguez v. People, 350 P.3d 287, 294 (Colo. 2015) (Marquez, J., dissenting), cert. granted sub nom. Pena-Rodríguez v. Colorado, - U.S. -, 136 S.Ct. 1513, 194 L.Ed.2d 602 (2016) (“Racial discrimination in our jury trial system not only violates our Constitution and the laws enacted under it but is at war with our basic concepts of a democratic society and a representative government,” and “the harm caused by such discrimination is not limited to the defendant — there is injury to the jury system, to the law as an institution, to the community at large, and to the democratic ideal reflected in the processes of our courts.” (internal citations and quotations marks omitted)). And it is in recognition of this fact that several of our sister circuits have unequivocally condemned the use of impermissible ethnic or racial generalizations and reversed the convictions of criminal defendants, even where the reviewing panel believed that the non-erroneous evidence was sufficient to convict. See, e.g., Cabrera, 222 F.3d at 596-97 (reversing defendants’ convictions after noting that, “[although we find that the evidence was sufficient to convict Cabrera and Mulgado, Detective Brook’s repeated references to their Cuban origin and his generalizations about the Cuban community prejudiced Cabrera in the eyes of the jury”); Vue, 13 F.3d at 1213 (reversing defendants’ convictions despite finding that the evidence was sufficient to sustain the convictions because the panel believed that “the injection of ethnicity into the trial clearly invited the jury to put the [defendants’] racial and cultural' background into the balance of determining their guilt,” thereby undermining the bedrock principle of our legal system — “[fjormal equality before the law”). I agree that it is “much too late in the day to treat lightly the risk that racial bias may influence a jury’s verdict in a criminal case.” United States v. Doe, 903 F.2d 16, 21 (D.C. Cir. 1990).
Here, because the Government repeatedly encouraged the jury to consider Garcia-Lagunas’s ethnicity and draw inferences contrary to Garcia-Lagunas’s interest in reliance upon an ethnic generalization, I am unable to conclude that the constitutional error did not contribute to the jury’s verdict. Specifically, the effect of the error was to eviscerate the sole plausible defense theory of the case, one with ample evidentiary support in the record. Indeed, I am baffled how any reviewing court could consider an error of this magnitude harmless beyond a reasonable doubt, either to a criminal defendant’s conviction or our criminal justice system on the whole. By presenting to the jury its unconstitutionally constructed racial taxonomy of the universe of North Carolina drug traffickers — African-American drug dealers like Reed, who live the high life and spend lavishly and ostentatiously, with lots of cash and drugs lying about, in contrast to “Hispanic drug traffickers,” whose members, even long-time residents in this country, like Garcia-Lagunas, habitually choose to live in abject poverty— the Government blatantly bolstered its case in contravention of well known and well settled constitutional norms.
Moreover, as the majority opinion correctly and comprehensively explains, the jury knew, for lack of a timely objection or motion in limine by defense counsel, al*506though it should not have been told, that Gareia-Lagunas was present in this country illegally. The majority refuses to treat that error as one remediable under our plain error doctrine. But plain error as to that singular issue to one side, the jury’s knowledge of that irrelevant and highly prejudicial fact renders the prosecution’s resort to racial and ethnic animus more, not less, condemnable, and should factor into the harmless-error analysis. It blinks reality not to do so.
III.
It is ironic that, in a break from our sister circuits, and at a moment in our country’s history when uncommon attention is being paid to issues of racial and ethnic stereotyping and consequent mistreatment, actual or threatened, this Court chooses to privilege the Government to employ, without consequence, irrelevant, prejudicial, and factually unwarranted evidence of blatant racial stereotyping to obtain a criminal conviction. In this moment, not even the ethnic heritage of distinguished federal judges is beyond trashing in the public sphere, and by á prominent candidate for the most powerful office on the planet, no less. All this at a time when this Court has otherwise stood firmly against manifestations of insidious racial and ethnic animus in voting, N.C. State Conference of NAACP v. McCrory, 831 F.3d 204, 2016 WL 4053033 (4th Cir. July 29, 2016), employment, Boyer-Liberto v. Fontainebleau Corp., 786 F.3d 264 (4th Cir. 2015) (en banc), and many other domains of civic, economic, and political life.
After this published opinion, future panels of this Court will be required to struggle with the issue of just how much evidence of guilt is enough evidence of guilt to permit the Court to give the Government a pass when it bolsters its pursuit of a conviction through resort to gratuitous racial and ethnic evidence intended to spur one or more jurors to convict. This case sets a very low bar, considering that the level of certainty that the constitutional violation had no effect upon any juror is agreed to be “beyond a reasonable doubt,” a standard that, interestingly, this Court has long refused to allow trial judges to define for ordinary jurors. See United States v. Walton, 207 F.3d 694, 699 (4th Cir. 2000) (en banc) (“We find no reason to alter our current practice of not requiring a jury instruction defining reasonable doubt in criminal cases.”). Perhaps, as we approach the 50th anniversary of the seminal teachings of Chapman v. California, the time has come for this Court to undertake an examination of just what “beyond a reasonable doubt” means, after all.
I would vacate the conviction on the conspiracy count of the indictment and order a new trial.
. According to the testimony of the four drug dealers testifying pursuant to plea agreements, Garcia-Lagunas sold to them, in the aggregate, at least 39 kilos of cocaine, with each kilo of cocaine valuing approximately $30,000 to $32,000 during the relevant time frame. J.A. 205, 208, 239, 340-42, 360-61, 388.
. Recall that the indictment in this case charged a greater-included offense of conspiring to distribute or possess with the intent to distribute 500 grams or more of cocaine, in violation of 21 U.S.C. §§ 841(a), 846. Importantly, therefore, the defense theory in this case not only militated in favor of an acquittal, but perhaps even more important from the defense perspective, it laid the basis for the jury's consideration of a lesser included offense involving a lesser amount of narcotics and thus a lower potential sentence. Cf. United States v. Hickman, 626 F.3d 756, 763-71 (4th Cir. 2010) (holding that evidence was insufficient to support the jury's guilty verdict on the indicted conspiracy involving greater drug amount but remanding for resentencing on conspiracy involving lesser drug amount).
. As my colleagues in the majority point out, Garcia-Lagunas has resided in the United States since he was a teenager, and the majority of his family, including his parents, spouse, and two of his children, also live in the United States, making it improbable that he was sending large amounts of money back to individuals in Mexico.
. The majority chooses not to address how the trial judge's statements could have independently affected the jury's thinking because they were voiced during a bench conference and there is no affirmative evidence that the jury heard the trial judge’s reinforcing remarks. However, it is not Garcia-Lagunas’s burden to demonstrate the rippling effects of the Government's unconstitutional testimony. Rather, the Government is tasked with establishing that its constitutional error did not contribute to the juiy’s verdict. Here, the Government has not attempted to show that the trial judge’s statements did not affect the jury’s consideration of Garcia-Lagunas’s defense theory. Moreover, I note that, having both served as a juror on three occasions in criminal cases tried in Maryland state courts, and having presided for 14 years over federal jury trials employing "white noise” to keep jurors in the dark, I know full well that statements made during bench conferences not infrequently remain within earshot of nearby and attentive jurors. Accordingly, because there is nothing in the record here to suggest that the judge’s remarks went unheard in this instance, it undeniably falls on this Panel, in conducting a harmless-error review, to fully consider the trial judge's statements and their potential, if not likely, impact on the jury's verdict.
*502Harmless Error in Criminal Trials, 99 Nw. U. L. Rev. 1053, 1085-98 (2005) (arguing that judges should look at evidence of influence on jury rather than focusing primarily on untainted evidence of guilt).
Brandon L. Garrett, Judging Innocence, 108 Colum. L. Rev.55, 108 n.195 (2008); see also John M. Greabe, The Riddle of Harmless Error Revisited, 54 Hous. L. Rev. (forthcoming 2016) (manuscript at 12 n.70) ("The [Supreme] Court has at ... times ... suggested that the presence of overwhelming evidence of guilt alone renders an error harmless. But these statements — which are akin to a 'correct result’ test of the sort rejected in Chapman— are contradicted by the Court’s more carefully reasoned cases and should not be taken to express the proper formulation.” (internal citations omitted)).
. As discussed fully infra, in framing the issue as it does, the majority commits a fundamental error that has been identified and warned against by distinguished legal scholars and others for decades:
Properly applied, harmless error analysis should ask only whether the state can demonstrate that error did not sufficiently affect the outcome at trial and not, conversely, whether evidence of guilt outweighed the impact of any error. See Sullivan v. Louisiana, 508 U.S. 275, 279, 113 S.Ct. 2078, 124 L.Ed.2d 182 (1993) ("The inquiry ... is ... whether the guilty verdict actually rendered in this trial was surely unattributable to the error. That must be so, because to hypothesize a guilty verdict that was never in fact rendered — no matter how inescapable the findings to support that verdict might be— would violate the jury-trial guarantee.”); Jason M. Solomon, Causing Constitutional Harm: How Tort Law Can Help Determine
. In this regard, it bears mention that not all constitutional infringements visited upon defendants in criminal cases stand on equal footing. That is to say, as one scholar argues, "judicial proceedings marred by unconstitutional discrimination on the basis of race, religion, ethnicity, national origin, or gender and intentional misconduct by government *504officials such as ... prosecutors” deserve a heightened level of scrutiny in the analysis of harmless error. Greabe, supra note 5, at 5.
. Importantly, further exacerbating the impact of this improper testimony, the trial judge aslced Orellano, in open court before the jury, to state the basis of his opinion. The magic words "training and experience” were quickly forthcoming, and the trial judge promptly overruled counsel’s renewed objection. J.A. 272-73. One can easily understand the remarkable impact on a juror who observes such a display of judicial approval of a law-enforcement witness.