Opinion
20-30728
04-22-2022
United States of America, Plaintiff-Appellee, v. Oliver Pierre, Defendant-Appellant.
Appeal from the United States District Court for the Eastern District of Louisiana No. 2:17-CR-132-2
Before Smith, Elrod, and Oldham, Circuit Judges.
Per Curiam [*]
Oliver Pierre pleaded guilty to conspiring to traffic at least 280 grams of cocaine. In exchange, the government dropped a less serious charge and agreed not to pursue a sentencing enhancement. In hindsight, it is clear that the sentencing enhancement could not have applied to Pierre even if the government had pursued it. So, Pierre gave up his right to stand trial to avoid a punishment that could never have come. Because he has shown that the district court committed plain error by accepting a plea based on an illusory benefit, we VACATE the district court's judgment and the plea and REMAND for further proceedings consistent with this opinion.
I.
In the spring of 2014, local law enforcement began investigating the drug trafficking activities of a person called "B.P." The officers purchased cocaine base from B.P. three times in April of that year for quantities totaling about ten grams. The DEA collected more information and by the next year learned that one of B.P.'s suppliers was named James Bickham. They purchased cocaine from B.P. three more times for quantities totaling about twenty-five grams. The DEA officers soon intercepted phone calls between B.P. and Bickham and learned that the two of them hoped to make a drug deal in Houston. Specifically, they discussed having a courier, Pierre, assist them in the deal.
The officers staked out B.P.'s home at the scheduled time. They followed B.P. to a gas station, where he parked next to Pierre. Pierre tossed two bags of cocaine base into B.P.'s car, and the two of them drove away separately. The officers stopped and detained B.P. and recovered 156.1 grams of cocaine base.
Pierre was indicted for conspiring to traffic 280 grams or more of cocaine base (Count 1) and distributing 28 grams or more of cocaine base (Count 2). He was assigned counsel in late 2017 and over a year later successfully moved for the appointment of new counsel. Eventually a plea agreement was reached. Under it, Pierre pleaded guilty to Count 1. In return, the government agreed to request dismissal of Count 2 and agreed that it would not charge Pierre as a multiple offender under 21 U.S.C. § 851-a provision which, Pierre was told, would have raised his mandatory minimum sentence from ten years to fifteen years.
The government also agreed not to bring "any other charges in the Eastern District of Louisiana arising from [Pierre's] violations of the Federal Controlled Substances Act prior to October 8, 2015," as long Pierre provided the details of those crimes.
At Pierre's re-arraignment, the district court read Count 1 of the indictment, which charged that Pierre conspired "to distribute and possess with the intent to distribute 280 grams or more of cocaine base . . . in violation of [21 U.S.C. § 841(a)(1), (b)(1)(A)] . . . ." It then explained that for him to be convicted of Count 1, the government would have to prove beyond a reasonable doubt that: (1) two or more persons reached an agreement to possess and distribute cocaine base; (2) Pierre knew of the agreement's unlawful purpose; and (3) Pierre willfully joined the agreement. At that point the court did not specifically explain that the government would have to prove that Pierre's participation in the conspiracy foreseeably involved the 280-gram quantity.
Because this charge was based on an alleged conspiracy, Pierre was also charged under 21 U.S.C. § 846.
The factual basis for the plea provided by the government contained the facts surrounding the incident described above. It also included a stipulation for the purposes of sentencing that Pierre had participated in the trafficking of at least 280 grams, but less than 840 grams, of cocaine base. Pierre affirmed that the information pertaining to him was true.
The presentence report also noted that both the government and Pierre stipulated to the 280- to 840-gram quantity for the purposes of sentencing. But it likewise specifically noted Pierre's role as courier for "at least 156.1 grams of crack cocaine," and that Pierre was a "minor participant in any criminal activity." The criminal history portion of the presentence report identified five prior state convictions, and Pierre's sentencing range was thus calculated as 120 to 137 months. The district court sentenced Pierre to the statutory minimum 120 months. Pierre appealed the judgment and sentence to this court, arguing for the first time that the district court erred by accepting his guilty plea and that he was unconstitutionally deprived of effective assistance of counsel.
II.
Pierre argues that his plea was not knowing and voluntary and thus should be vacated because, among other things, it was conditioned on the government agreeing not to pursue a sentencing enhancement under a provision that could not apply to him. We agree with Pierre and hold that that the district court plainly erred by accepting his plea under these circumstances.
A.
Pierre argues that the district court erred by accepting his guilty plea. Because he raises this challenge for the first time on appeal, we review for plain error. See United States v. Hicks, 958 F.3d 399, 401 (5th Cir. 2020); United States v. Hughes, 726 F.3d 656, 659 (5th Cir. 2013). Under the plain error standard, Pierre must show (1) an error, (2) that is clear or obvious, and (3) that affected his substantial rights. United States v. Trejo, 610 F.3d 308, 319 (5th Cir. 2010). When challenging the acceptance of a guilty plea, specifically, he must demonstrate "a reasonable probability" that he would not have pleaded guilty without the error. Hicks, 958 F.3d at 401-02. And "[i]n making this determination, we may consult the whole record on appeal." United States v. Reyes, 300 F.3d 555, 559 (5th Cir. 2002). Finally, if Pierre satisfies each criterion, we may vacate the judgment if the error "seriously affect[s] the fairness, integrity or public reputation of judicial proceedings." United States v. Young, 470 U.S. 1, 15 (1985) (quoting United States v. Atkinson, 297 U.S. 157, 160 (1936)).
B.
Pierre argues that his guilty plea was not knowing and voluntary because it was induced by the government's promise not to pursue a sentencing enhancement under a provision which could not have applied to him. The government concedes that the sentencing enhancement provision could not have applied but argues that Pierre cannot satisfy the plain error standard because he cannot show that there is a reasonable probability he would not have pleaded guilty without the error. We agree with Pierre.
The dissenting opinion attempts to cabin such a challenge to a collateral action under 28 U.S.C. § 2255 instead of in a direct appeal. But we regularly consider involuntary-plea claims on direct appeal, and we will not depart from our standard practice here. See, e.g., United States v. Diaz, 989 F.3d 390, 392-94 (5th Cir. 2021); United States v. Scott, 857 F.3d 241, 244-46 (5th Cir. 2017); United States v. Washington, 480 F.3d 309, 315-16 (5th Cir. 2007).
A guilty plea must be knowing and voluntary. Parke v. Raley, 506 U.S. 20, 28 (1992). "That is so because a guilty plea constitutes a waiver of three constitutional rights: the right to a jury trial, the right to confront one's accusers, and the privilege against self-incrimination." Id. at 29. A guilty plea is not sufficiently voluntary if it is induced by "deception, an unfulfillable promise, or misrepresentation . . . ." United States v. Amaya, 111 F.3d 386, 389 (5th Cir. 1997). "A plea of guilty that is based on the fear of a non-existent penalty can be neither knowing nor intelligent, and this flaw colors the fundamental fairness of the entire proceeding." Kennedy v. Maggio, 725 F.2d 269, 273 (5th Cir. 1984).
Under this standard, Pierre's plea was not knowing and voluntary, so the district court erred by accepting it. His guilty plea was induced in part by the government's promise not to charge him as a multiple offender "having at least one prior felony drug conviction, which would have resulted in a mandatory sentence of at least fifteen (15) years" under 21 U.S.C. §§ 841(b)(1)(A) and 851. The current version of § 841(b)(1)(A) explains that a defendant's mandatory minimum sentence may be increased from ten to fifteen years if he has a "prior conviction for a serious drug felony." A "serious drug felony" is one described in 18 U.S.C. § 924(e)(2) for which the defendant served a term of imprisonment of over a year and for which the defendant's release from imprisonment occurred within fifteen years of the current offense. 21 U.S.C. § 802(57). And 18 U.S.C. § 924(e)(2)(A)(ii) limits the provision to offenses which, "under State law, involv[e] manufacturing, distributing, or possessing with intent to manufacture or distribute, a controlled substance . . . [and] for which a maximum term of imprisonment of ten years or more is prescribed by law."
Pierre had three prior convictions for possessing a controlled substance. As the government concedes, those offenses "do not include the necessary elements of manufacturing, distributing, or possessing with intent to do so," so they do not provide a basis for the § 841(b)(1)(A) enhancement. And Pierre's one conviction for distribution carries a maximum sentence of less than ten years. So, Pierre's plea was induced by the promise of an illusory benefit and thus was not sufficiently knowing and voluntary.
The sentencing enhancement would have applied to Pierre under the previous version of that provision. Before 2018, § 841(b)(1)(A) provided for an enhanced mandatory minimum sentence of 20 years and simply applied to "felony drug offense[s]," which included any drug offense "punishable by imprisonment for more than one year." 21 U.S.C. § 802(44). This change to the statute was made after Pierre was indicted but before he pleaded guilty. Presumably, that is the reason this error was made and neither the district court nor the parties caught it initially.
This error flatly contradicted the applicable statutory language of the sentencing enhancement provision, and so was "clear or obvious." Cf. Trejo, 610 F.3d at 319. To prove that any error was "clear or obvious," a defendant need not identify a past decision of this court directly holding in the defendant's favor on the exact same issue. Instead, our rule is that a defendant must show that the "error is clear under existing law." United States v. Dunigan, 555 F.3d 501, 506 (5th Cir. 2009) (quoting United States v. Maturin, 488 F.3d 657, 663 (5th Cir. 2007)). An undisputed contradiction of statutory language is an error that is clear under existing law. United States v. Aderholt, 87 F.3d 740, 744 (5th Cir. 1996) ("The error is evident from a plain reading of the statute and thus, is obvious."). There is no dispute that the sentencing enhancement provision could not have applied to Pierre, and the parties agree on that point. Thus, prongs one and two of the plain error analysis are satisfied. See Trejo, 610 F.3d at 319 (identifying prongs one and two of the plain error analysis as (1) whether an error occurred, and (2) whether the error was clear or obvious).
But Pierre must also show that absent the error there is a reasonable probability that he would not have pleaded guilty. Hicks, 958 F.3d at 401-02. He has made this showing because the promise not to seek the inapplicable sentence enhancement was the primary benefit he was to receive in exchange for pleading guilty to Count 1. If applicable, the enhancement would have increased his mandatory minimum sentence by fifty percent, from ten years to fifteen years. Avoiding that enhancement is a substantial benefit and, indeed, the presentence report described it as the "[i]mpact" of the plea agreement. Moreover, as Pierre points out, he and his counsel participated in plea negotiations for quite a long time, and Pierre only agreed to plead guilty after the government offered not to seek the enhancement. The plea's timing does not prove a direct causal relationship between the government's promise and Pierre's plea, but it certainly weighs in Pierre's favor; it suggests that the promise made a difference for his decision. And, finally, Pierre conceivably could have prevailed at trial had he chosen to proceed. Certainly we cannot determine at this stage what additional evidence against him the government might have gathered if proceedings continued towards trial. But it is likely that the promise not to pursue a substantially enhanced sentence played a significant role in his decision to plead guilty.
Of course, in exchange for Pierre's guilty plea to Count 1, the government also agreed to move to dismiss Count 2. At the surface level, then, Pierre had another incentive to plead guilty besides the promise not to seek the inapplicable sentence enhancement. But the surface is as deep as that additional benefit goes. Count 2 charged Pierre with distributing twenty-eight grams or more of cocaine base. That crime is less serious than the Count 1 offense and in this case is based on the same conduct as that to which the alleged conspiracy was aimed; thus, had Pierre been convicted of both Count 1 and Count 2, the Count 2 conviction likely would have been subsumed by Count 1 for the purposes of sentencing. See U.S.S.G. § 3D1.2 cmt. 4 ("When one count charges a conspiracy or solicitation and the other charges a substantive offense that was the sole object of the conspiracy or solicitation, the counts will be grouped together under [U.S.S.G. § 3D1.2(b)]."). And if at trial Pierre had been convicted on Count 2 only, that offense carries a lower sentencing range than Count 1. See 21 U.S.C. § 841(b)(1)(A)(iii), (b)(1)(B)(iii); 21 U.S.C. § 846. In other words, it is unlikely that a Count 2 conviction would have affected his guidelines range for sentencing. The government therefore conveyed little (if any) benefit by dismissing the Count 2 charge in exchange for his guilty plea to Count 1, leaving as the primary "benefit" the illusory promise not to pursue an inapplicable sentencing enhancement.
The Count 1 offense to which Pierre pleaded guilty, conspiracy to traffic 280 grams or more of cocaine base, carries a standard sentencing range of 10 years to life imprisonment. 21 U.S.C. § 841(b)(1)(A)(iii); 21 U.S.C. § 846. The Count 2 offense, distributing 28 grams or more of cocaine base, carries a standard sentencing range of 5 to 40 years imprisonment. 21 U.S.C. § 841(b)(1)(B)(iii).
The government also agreed not to bring "any other charges in the Eastern District of Louisiana arising from [Pierre's] violations of the Federal Controlled Substances Act prior to October 8, 2015," as long Pierre provided the details of those crimes. The statute of limitations applicable to such offenses is five years. 18 U.S.C. § 3282 (a) . Becau se Pierre pleaded guilty in June of 2019, the government's promise could have applied only to an offense committed between mid-2014 and early October of 2015. The record does not show that the government was in a position to charge Pierre with any such offense. Thus, it is unlikely that the promise gave Pierre a significant benefit. Notwithstanding this additional promise the government made, there is still a reasonable probability that the illusory promise not to pursue the sentencing enhancement affected Pierre's decision to plead guilty.
It is true that Pierre was ultimately sentenced to the statutory minimum for the Count 1 offense to which he pleaded guilty. That means that if he had not pleaded guilty, and then had proceeded to trial and been convicted on either count, he could have been sentenced to a longer term. But, contrary to the government's position, Pierre nonetheless could have been prejudiced by the misconception about the applicability of the sentencing enhancement. When reviewing the acceptance of a guilty plea for plain error, the question is not whether the plea likely caused the defendant to be issued a longer sentence, but whether there is a reasonable probability that absent the error the defendant would not have entered the plea. United States v. Dominguez Benitez, 542 U.S. 74, 83 (2004). Pierre has shown that the core benefit he received from the guilty plea was the government's promise not to charge him as a multiple offender. Thus, there is a reasonable probability that he would not have entered the plea if he understood that the benefit was illusory. Under Dominguez Benitez, he has satisfied the third element of plain error review.
Finally, Pierre must show that the error seriously affects the fairness, integrity, or public reputation of judicial proceedings. Young, 470 U.S. at 15. If he does, we "may then exercise [our] discretion to notice the error." United States v. Serna-Villarreal, 352 F.3d 225, 231 (5th Cir. 2003). Under our precedent, we conclude that Pierre has satisfied this criterion. The right to a jury trial is one of the primary constitutional safeguards for fairness and integrity in judicial proceedings. Thus, as we have explained, "[t]he integrity of the plea bargaining system is 'vital to our national system of criminal justice, '" and "[m]aintaining that integrity requires diligently policing its failure to function properly." United States v. Palmer, 456 F.3d 484, 491 (5th Cir. 2006) (quoting United States v. Ashburn, 20 F.3d 1336, 1347 (5th Cir. 1994)).
When a plea is induced by the government's promise of an illusory benefit, and so the defendant forgoes a fundamental right because he wrongly believes doing so will bring him a substantial benefit, the criminal justice system has "fail[ed] to function properly." See Palmer, 456 F.3d at 491. In this case Pierre lost the opportunity to pursue a more favorable outcome in a manner consistent with his fundamental rights-either by going to trial or by negotiating a different plea agreement founded on accurate assumptions-so we exercise our discretion to vacate the judgment and his plea.
We consider this case appropriate for us to exercise our discretion under prong four of the plain error analysis in part because Pierre could potentially negotiate a significantly more favorable plea deal if given the chance. If, for example, he pleaded guilty to participating in the trafficking of anything less than the 280-gram quantity of cocaine base, he would face a five-year mandatory minimum sentence instead of a ten-year mandatory minimum. Compare 21 U.S.C. § 841(b)(1)(A), (b)(1)(A)(iii), with id. § 841(b)(1)(B), (b)(1)(B)(iii). Such a charge would also correspond to a lower base offense level. Compare U.S.S.G. § 2D1.1(c)(5), with U.S.S.G. § 2D1.1(c)(6). With all else equal, those changes could reduce Pierre's guidelines minimum sentence to 92 months (a sentence nearly twenty-five percent shorter than the 120-month sentence he received) and his guidelines maximum to 115 months (which is less than the 120-month sentence he received).
* * *
Thus, the district court plainly erred when it accepted Pierre's guilty plea that was induced by a promise not to pursue a sentencing enhancement that could not have applied. We therefore VACATE the district court's judgment and the plea and REMAND to the district court for further proceedings consistent with this opinion.
Pierre also argues that his plea was not knowing and voluntary because the district court failed to follow the requirements of Rule 11 of the Federal Rules of Criminal Procedure. He claims, first, that the district court failed to properly advise him of the nature of the charge against him as required by Rule 11(b)(1)(G). Relatedly, he claims that the court erred by accepting his guilty plea without a sufficient factual basis for the drug quantity as required by Rule 11(b)(3). Because we vacate the judgment entered on his guilty plea and remand for further proceedings, we need not address these alternative arguments for vacating. For the same reason, we decline to address Pierre's argument that he was deprived of effective assistance of counsel. But we note that rarely will we consider such claims on direct appeal. See United States v. Gulley, 526 F.3d 809, 821 (5th Cir. 2008) ("[T]he 'general rule in this circuit is that a claim of ineffective assistance of counsel cannot be resolved on direct appeal when the claim has not been raised before the district court since no opportunity existed to develop the record on the merits of the allegations.'" (quoting United States v. Higdon, 832 F.2d 312, 313-14 (5th Cir. 1987))); United States v. Gordon, 346 F.3d 135, 136 (5th Cir. 2003) ("[A] § 2255 motion is the preferred method for raising a claim of ineffective assistance of counsel.").
Andrew S. Oldham, Circuit Judge, dissenting:
Pierre bears the burden of showing plain error in this court. That means he must establish "(1) that the district court committed an error (2) that is plain and (3) affects his substantial rights and (4) that failure to correct the error would 'seriously affect the fairness, integrity or public reputation of judicial proceedings.'" United States v. Sanchez-Hernandez, 931 F.3d 408, 410 (5th Cir. 2019) (quoting Johnson v. United States, 520 U.S. 461, 467 (1997)). Pierre cannot make that showing as to the third or fourth prongs. I therefore respectfully dissent.
One overarching procedural problem afflicts the majority's analysis: Pierre is raising his involuntary-plea claim, not by way of a 28 U.S.C. § 2255 motion, but on direct appeal. That matters because a § 2255 motion would allow the district court to "grant a prompt hearing" on Pierre's claim and to "determine the issues and make findings of fact and conclusions of law with respect thereto." 28 U.S.C. § 2255(b) (emphasis added). That factfinding mechanism is crucial for claims, like ineffective-assistance-of-counsel claims, that are predicated on facts outside the trial record. For that very reason, "the general rule in this circuit is that a claim of ineffective assistance of counsel cannot be resolved on direct appeal when the claim has not been raised before the district court since no opportunity existed to develop the record on the merits of the allegations." United States v. Gulley, 526 F.3d 809, 821 (5th Cir. 2008) (quotation omitted); see also Massaro v. United States, 538 U.S. 500, 504 (2003) ("[I]n most cases a motion brought under § 2255 is preferable to direct appeal for deciding claims of ineffective assistance."); ante, at 11 n. 8 (citing this rule favorably). Pierre's involuntary-plea claim, no less than his ineffective-assistance claim, turns on facts that he has not proved. Both claims turn on who said what and when in the plea negotiations between his lawyer and the prosecutor. The majority simply assumes that Pierre's assertions are true and then vacates his conviction based on nothing more than his assertions. There is zero basis in law or logic for that approach.
I.
The legal standard for our prong-three prejudice inquiry is undisputed. We must ask whether there is "a reasonable probability that, but for the error, [the defendant] would not have entered the plea." United States v. Dominguez Benitez, 542 U.S. 74, 83 (2004) (in the context of Federal Rule of Criminal Procedure 11); see also United States v. Reyes, 300 F.3d 555, 558 (5th Cir. 2002) ("Rule 11 ensures that a guilty plea is knowing and voluntary."). So Pierre must show a "reasonable probability" that, if the district court had noticed the plea involved an unfulfillable promise, he wouldn't have pleaded guilty to Count 1 (the conspiracy charge). See Dominguez Benitez, 542 U.S. at 83; United States v. Hicks, 958 F.3d 399, 401 (5th Cir. 2020).
That's tough to do. Consider, for example, United States v. Alvarado-Casas, 715 F.3d 945 (5th Cir. 2013). In that case, the district court had wrongly told the defendant that, if he pleaded guilty, "he faced a maximum of ten years of imprisonment." Id. at 953-54. His counsel told him the same thing. Id. at 954. In fact, his maximum exposure was twice that-twenty years. Ibid. After the defendant pleaded guilty, he was in fact sentenced to 190 months-"70 months greater than the 10-year maximum possible prison sentence of which his counsel and the district court advised him." Id. at 954 (quotation omitted). Despite all of that, this court held the defendant failed to show prejudice. See id. at 954-55 (considering, inter alia, the otherwise-generous plea agreement, the fact that the PSR did not share the court's error, and the defendant's failure to show he had been ready to go to trial); see also United States v. Molina, 469 F.3d 408, 412 (5th Cir. 2006) (holding the third prong unsatisfied and explaining the defendant did not "direct this court to any portion of the record supporting the proposition that the maximum sentence for count three affected his plea decision. Because [the defendant] has not satisfied his burden, we affirm his conviction.").
The majority opinion holds Pierre has shown "that absent the error there is a reasonable probability that he would not have pleaded guilty." See ante, at 7-10. The majority gives five main reasons for that holding. Each of those reasons, naturally, focuses on the Government's promise not to seek an inapplicable sentencing enhancement. At no point does the majority acknowledge that it is climbing uphill against precedents like Alvarado-Casas-nor does it acknowledge that the absence of a § 2255 factual hearing renders Pierre incapable of making the required showing in this case.
I reproduce the majority's stated reasons for that holding here, in my own order. (1) The PSR described the Government's meaningless promise as the plea agreement's "impact." Id. at 8. (2) Pierre and his first lawyer had been negotiating with the Government for a long time, and Pierre "only agreed to plead guilty after the government offered not to seek the enhancement." Ibid. (citing only Pierre's brief in support). (3) "Pierre conceivably could have prevailed at trial had he chosen to proceed." Ibid. But see ibid. (acknowledging that "we cannot determine at this stage what additional evidence against him the government might have gathered if proceedings continued towards trial"). (4) "If applicable, the enhancement would have increased his mandatory minimum sentence by fifty percent, from ten years to fifteen years." Ibid. And finally, (5), the other promises in Pierre's plea agreement went only as deep as the "surface," so when the no-sentencing-enhancement promise turned out to be empty, Pierre got little out of the deal. See id. at 8-9. So goes the reasoning.
The first three reasons aren't probative at all. (1) I'm not sure what the majority means by its quotation of the "impact" phrase in the PSR. Is the idea that Pierre must have relied on the PSR to understand what the plea deal entailed, and that the PSR framed the deal as being solely about the sentencing enhancement? If so, that's descriptively inaccurate: Elsewhere, the PSR acknowledged the Government's other two promises, which I address in more detail below. And it makes little sense to treat the PSR's characterization of the plea agreement as more probative than the agreement itself-which clearly made three promises, not just one.
Reason (2) is baseless because the record says nothing about the inner workings of the plea negotiations. The record shows that Pierre and the Government repeatedly, and successfully, asked the district court to push back the trial date while they hammered out a plea agreement. The record shows that, in December 2018, Pierre wrote the district judge a letter asking for new counsel, and that he got his wish in February 2019. And the record shows that the parties notified the court of their plea agreement a few months later, in June 2019. That's it. That's all the record shows.
The majority simply assumes that Pierre was holding out for a better deal all that time-and that he agreed to plead guilty only after the Government sweetened the pot by promising not to seek the sentencing enhancement. They act as if we know the Government had offered both to drop Count 2 and to forego other charges all along, that Pierre was repeatedly rejecting that offer throughout the negotiations, and that he agreed to plead guilty only after the Government added the sentencing-enhancement promise. But the record suggests no such thing. Maybe Pierre could prove the Government's promise was material. Maybe he couldn't. But that's why we have § 2255 proceedings and evidentiary hearings. I am unaware of any authority that says we can simply take the prisoner's allegations as fact and grant a new trial based on no evidence.
Reason (3) is the weakest. Here, the majority indulges in pure speculation that Pierre may have fared well if he'd decided to go to trial. As the majority seems to acknowledge, we have no idea how strong the Government's case would have been if it had tried to prove Count 1. See ante, at 7-8.
Reason (4) also does nothing. The Government promised not to seek the sentencing enhancement, and the enhancement would've bumped Pierre's minimum sentence from 10 to 15 years. The question is: Can we infer from the existence of the promise, res ipsa loquitur, that there's a "reasonable probability" Pierre wouldn't have pleaded guilty without it? See Dominguez Benitez, 542 U.S. at 83. The answer is no. Because the Government gave Pierre more than one promise in exchange for his guilty plea, the mere existence of the no-sentencing-enhancement promise doesn't let us make any counterfactual inferences at all.
Consider a monopsony situation where one seller ("S") agrees to sell a good ("G") to one buyer ("B") for $100. After the deal goes through, can we infer that S wouldn't have agreed to sell for $75? What about $50? What about a mere $1? Nope. Without more information, the mere fact of the deal tells us only two things: that S valued G at something less than $100, and that B valued G at something more than $100. So we have no clue what S would've done if B had offered less than $100.
We likewise have no clue who would've bargained for what in the counterfactual world imagined by Pierre and the majority. The Government offered Pierre three things in exchange for his guilty plea: the promise not to seek the sentencing enhancement, the promise to drop the indictment's second charge ("Count 2"), and the promise not to prosecute Pierre for certain crimes, on the condition that he disclose those crimes. Pierre accepted. Does it follow that Pierre wouldn't have accepted if the Government had offered only to drop Count 2? Absolutely not. We don't have enough information to say anything about that hypothetical. Reason (4), like reasons (1)-(3), tells us nothing useful.
Nor does my view entail that a defendant can never satisfy the third plain-error prong after pleading guilty. Imagine this was a § 2255 case. If so, Pierre would have had the opportunity for a factual hearing in district court. See 28 U.S.C. § 2255(b). Imagine that, at the hearing, Pierre showed the Government had first offered only to drop Count 2, and Pierre refused that offer. And suppose the Government had only then came back with a promise to drop Count 2, paired with a promise not to seek the sentencing enhancement. In light of those facts, we would have very strong reason to believe Pierre wouldn't have pleaded guilty without the no-sentencing-enhancement promise. And the absence of such showings in the case before us perfectly illustrates why this claim cannot be resolved without facts. Cf. Gulley, 526 F.3d at 821 (ineffective-assistance claims generally not cognizable on direct appeal). In all events, the mere existence of the Government's promise can't answer the reasonable-probability question all by itself.
That leaves reason (5). It's the oddest of them all. The majority recognizes that the Government made three promises in the plea agreement: (A) a promise to dismiss Count 2; (B) a promise not to bring future charges; and (C) an illusory promise not to seek an inapplicable sentencing enhancement. All agree that the value of (C) is zero. But for the majority's reason (5) to make sense, the value of A and B also must equal zero. Yet the majority effectively concedes that (A) and (B) were not worth zero. The most the majority can say is that (C) was Pierre's "primary" benefit under the plea agreement, and that (A) + (B) conferred "little (if any)" additional benefit on him. See ante, at 9. All that matters, however, is that (A) + (B) conferred some additional benefit; because they did-as the majority effectively concedes-Pierre undisputedly benefited from his plea.
And the majority is correct not to dispute the value of (A) + (B). As to (A), the Government agreed to dismiss Count 2 entirely. The majority says that might not matter because a conviction on both Counts 1 and 2 would not affect Pierre's Guidelines range. See id. at 8-9 (citing U.S.S.G. § 3D1.2). But the majority ignores the fact that the sentencing court could've determined the applicable range was inappropriate for a multi-count conviction and departed accordingly. In United States v. Saldana, 427 F.3d 298 (5th Cir. 2005), for example, § 3D1.2 of the Guidelines required the defendant's convictions be grouped together. Compare id. at 311 (explaining how the Guidelines work and explaining, "no [multiple-convictions] adjustment was available in this case" under the Guidelines), with ante, at 8-9 (explaining the same in this case). But in Saldana, the district court consciously chose to impose consecutive sentences anyway, "quadrupl[ing] the maximum sentence allowable . . . under the Guidelines." 427 F.3d at 308-12. We upheld that decision as an upward departure from the Guidelines. Id. at 311-12. It's not my view that such upward departures are categorically permissible. But that doesn't matter: For a defendant facing prison, the prospect of a multi-count conviction is a serious risk to be mitigated. The Government's promise to drop Count 2 didn't just mitigate that risk for Pierre, it eliminated it entirely. It blinks reality to dismiss the obvious benefit of (A) as basically nothing.
As to (B), the Government promised not to bring certain charges against Pierre in the future, so long as Pierre disclosed those offenses in full. The majority attempts to dismiss this promise in a footnote, arguing that "[t]he record does not show that the government was in a position to charge Pierre with any such offense." Ante, at 9 n. 6. But that's not good enough: It's Pierre's burden to establish plain error, including the third prong. See Sanchez-Hernandez, 931 F.3d at 412 n. 3 ("Under prong three, it remains the defendant's burden (not the government's) to prove prejudice."). And that means it's Pierre's burden to establish the Government's no-charge promise was meaningless. He hasn't done so; instead, his briefing ignored this promise entirely. That alone is fatal to his plain-error argument. See Molina, 469 F.3d at 412 ("Moreover, [the defendant] does not direct this court to any portion of the record supporting the proposition that the maximum sentence for count three affected his plea decision. Because [the defendant] has not satisfied his burden, we affirm his conviction." (emphasis added)).
In sum: The majority's key claim (to the extent it even makes the claim) on the third prong is that the only benefit of the plea agreement was the Government's illusory promise. See ante, at 9. But the majority's five reasons to believe that are themselves illusory-and that's because the majority is substituting its own hypotheses for the factual development a § 2255(b) hearing is supposed to provide.
II.
In any event, this is not the kind of extreme case that warrants an exercise of our discretion under the fourth plain-error prong. This prong involves "discretion," Henderson v. United States, 568 U.S. 266, 272 (2013), and "a plain error affecting substantial rights does not, without more," satisfy this standard, United States v. Olano, 507 U.S. 725, 737 (1993). Indeed, Rule 52(b) "authorizes the Courts of Appeals to correct only particularly egregious errors." United States v. Young, 470 U.S. 1, 15 (1985) (quotation omitted).
The majority gives short shrift to one key fact: Pierre got the 10-year, minimum sentence for his Count 1 offense-an offense he admitted under oath. It's not as if Pierre was sentenced under the inapplicable mandatory-minimum of 15 years. Nor did the sentencing court impose a life sentence (which the statute would've allowed). There's nothing "particularly egregious" about that. See ibid. (quotation omitted); cf. Alvarado-Casas, 715 F.3d at 954-55 (denying plain-error relief partially because the Government had dropped many charges as part of the plea agreement, which suggested the bargain was a fair one).
The majority's response, see ante, at 11 n.7, is to form yet another hypothesis: that Pierre might be able to get a still better deal on remand. In the majority's words, "[i]f, for example, [Pierre] pleaded guilty to participating in the trafficking of anything less than the 280-gram quantity of cocaine base [the Count 1 offense], he would face a five-year mandatory minimum sentence instead of a ten-year mandatory minimum." See ibid. What conceivable basis is there for this speculation? Pierre already got the minimum sentence for his crime. The only way to get a better deal is to suggest-again without a single piece of factual support-that Pierre somehow, maybe, could've beaten the rap for the crime he admittedly committed. Or perhaps the Government would've allowed him to plead to something less than conspiring to traffic 280 grams or more of cocaine base. But criminal appeals don't turn on perhapses, would'ves, could'ves, and maybes. We ordinarily require prisoners to prove their claims before vacating their convictions. I would've followed our usual course.
I respectfully dissent.
[*]Pursuant to 5th Circuit Rule 47.5, the court has determined that this opinion should not be published and is not precedent except under the limited circumstances set forth in 5th Circuit Rule 47.5.4.