Opinion
No. 19-50092
02-08-2022
UNITED STATES of America, Plaintiff-Appellee, v. Henry MENDOZA, aka Hank, aka Pelon, aka Spanks, aka Spanky, Defendant-Appellant.
Ethan A. Balogh (argued) and Narai Sugino, Balogh & Co. APC, San Francisco, California, for Defendant-Appellant. Lindsay M. Bailey (argued), Assistant United States Attorney, International Narcotics, Money Laundering & Racketeering Section; Bram M. Alden, Acting Chief, Criminal Appeals Section; Tracy L. Wilkison, Acting United States Attorney; United States Attorney's Office, Los Angeles, California; for Plaintiff-Appellee.
Ethan A. Balogh (argued) and Narai Sugino, Balogh & Co. APC, San Francisco, California, for Defendant-Appellant.
Lindsay M. Bailey (argued), Assistant United States Attorney, International Narcotics, Money Laundering & Racketeering Section; Bram M. Alden, Acting Chief, Criminal Appeals Section; Tracy L. Wilkison, Acting United States Attorney; United States Attorney's Office, Los Angeles, California; for Plaintiff-Appellee.
Before: Marsha S. Berzon, Carlos T. Bea, and Jacqueline H. Nguyen, Circuit Judges.
OPINION
BEA, Circuit Judge: Every parent knows that teenagers make mistakes. A fifteen-year-old Henry Mendoza was no exception, though his error was far more serious than most. At that young age, Mendoza joined the Canta Ranas Organization ("CRO"), a Californian gang known for violent extortion and drug distribution. Mendoza was a member of the CRO for at least eight years and served under the leadership of two of his childhood friends, the gang's heads. All agree so far. But after this point, Mendoza's path becomes less clear.
The government alleges that Mendoza continued as an active gang member until 2016, when he was arrested as part of federal law enforcement's wide-ranging takedown of the CRO. As its key evidence, the government cites two incidents in June 2013 and December 2016 when law enforcement caught Mendoza with a handgun and methamphetamine (16.2 grams in 2013; 3.3 grams in 2016) and a handful of phone and text message conversations between Mendoza and CRO members, at least two of which involve Mendoza asking for methamphetamine.
Mendoza admits to a long-standing methamphetamine addiction but denies membership in the CRO. As he tells it, he left the CRO after eight years in its membership, and is now an addict, not a dealer. He argues that the methamphetamine with which he was found in 2013 and 2016 was for his own consumption, not for re-sale, and that the contacts between him and CRO members were sporadic attempts to purchase drugs from childhood friends rather than evidence of committed gang membership.
The jury, for its part, appears to have believed both Mendoza and the government, which also means that it fully believed neither. As to Mendoza's drug possession in June 2013, the jury acquitted him of possession of methamphetamine with intent to distribute and convicted him only of simple possession. But as to Mendoza's relationship with the CRO, and despite its apparent view of the evidence from the June 2013 incident, the jury convicted Mendoza of conspiracy to distribute methamphetamine, RICO conspiracy, and possession of a firearm during and in relation to or in furtherance of a crime of violence or drug-trafficking offense.
The jury also convicted Mendoza of being a felon in possession of a firearm, a conviction that Mendoza does not challenge.
Mendoza appeals the two conspiracy convictions and the firearm possession conviction, challenging both the sufficiency of the evidence underlying them and several jury instructions and decisions by the district court. With jurisdiction under 28 U.S.C. § 1291, we vacate these three challenged convictions for insufficient evidence and remand with instructions to resentence Mendoza accordingly. I. BACKGROUND
A. Factual Background
The Canta Ranas Organization (again, the "CRO") was a street gang active in Santa Fe Springs, California. The CRO was managed on the street by Jose Loza and David Gaitan, engaged in extortion and drug trafficking, among other crimes, and at its peak had dozens of members.
Mendoza joined the CRO at age fifteen, and since childhood he has been close personal friends with CRO leaders Loza and Gaitan. Mendoza testified at trial, however, that he left the gang around eight years after he joined and subsequently moved away from the gang's territory, taking with him only a lifelong addiction to methamphetamine and a series of gang-related tattoos. (The government disputes that Mendoza left the gang.) Mendoza also introduced testimony that in the years after he claims he left the CRO, he worked a regular job and supported his ongoing methamphetamine addiction with that job's earnings.
In the 2010s, the federal government began a sustained investigation of the CRO; this investigation yielded the government's evidence against Mendoza. As part of the investigation, law enforcement surveilled CRO leader Gaitan's home, the "central hub for all of the gang's drugs and guns," and wiretapped Gaitan's phone. Over a seven-month period in 2013, the police intercepted about 21,000 calls and texts. Of the 21,000, Mendoza participated in four. Separately, the police also later found one text message conversation between Mendoza and a CRO member.
Mendoza puts this number at three but apparently omits one call.
Two of the four calls that included Mendoza were from June 2013. In the first, Mendoza called Gaitan to ask if he could purchase methamphetamine; in the second, Gaitan called Mendoza's house the next day to tell Mendoza that the drugs were ready. That evening, the police intercepted Mendoza in his car, and upon searching him and the car, found $31 in cash, a phone, a handgun, a police scanner, and about 16.2 grams of pure meth. The police found no evidence of drug selling, packaging, or cutting (i.e. , diluting it with another substance) but also no needles or pipes that Mendoza could use to consume the drug.
Several weeks later, the police intercepted two calls between CRO members that were about Mendoza. One was between CRO head Loza and CRO member Robert McAfee; the other was between Loza and CRO head Gaitan. In both, Loza and the other member (McAfee or Gaitan) discussed their unsuccessful efforts to contact Mendoza. And a day or two later came a third call involving Mendoza. This time, Mendoza called Jose Loza and the two discussed Loza's efforts to contact Mendoza.
The fourth call involving Mendoza was from May 2013. In it, Mendoza called CRO leader Gaitan and the two spoke briefly about an upcoming gathering in Gaitan's backyard. The government did not cite this call in its briefing as evidence for any incriminating proposition about Mendoza.
Three years later, in April and May 2016, came the text conversation involving Mendoza. In it, Mendoza asked a CRO member, Antolin, for meth, and used several CRO-related phrases and images when making his request. During the next few weeks, Mendoza haggled with Antolin over the price of the methamphetamine he requested, and informed Antolin that if Antolin could not provide the drugs, Mendoza would purchase them from someone else. In June 2016, Mendoza was indicted along with around 50 CRO members. Within a few months, the police had arrested nearly all the indicted individuals, including CRO leaders Gaitan and Loza. The police were originally not able to find Mendoza, but eventually arrested him in December 2016. At the time, law enforcement found with him a handgun, 3.5 grams of methamphetamine, and some cash. Again, the police found no evidence of drug selling, packaging, or cutting, but also no needles or pipes that could be used to consume methamphetamine.
B. Procedural History
At trial, Mendoza was charged with: 1) possession of methamphetamine with intent to distribute under 21 U.S.C. § 841(a)(1) (with simple possession of methamphetamine under 21 U.S.C. § 844 as a lesser-included offense), 2) being a felon in possession of a firearm under 18 U.S.C. § 992(g), 3 ) conspiracy to possess methamphetamine with intent to distribute it under 21 U.S.C. § 846, 4) RICO conspiracy under 18 U.S.C. § 1962(d), and 5) carrying a firearm "during and in relation to" or "in furtherance of" a crime of violence or drug-trafficking crime under 18 U.S.C. § 924(c)(1)(A). Mendoza was convicted of the two conspiracy charges and the two gun-possession charges. He was acquitted, however, of possession of methamphetamine with the intent to distribute and convicted of only the lesser-included offense of simple possession of methamphetamine. Mendoza was sentenced to a total of 180 months' imprisonment and he timely appealed his convictions for conspiracy to distribute meth, RICO conspiracy, and possession of a firearm in relation to or in furtherance of a drug-trafficking crime or crime of violence.
Mendoza does not appeal his convictions for simple possession of methamphetamine nor for being a felon in possession of a firearm.
II. DISCUSSION
Mendoza's primary argument on appeal is that the government offered insufficient evidence to support three charges of which he was convicted: 1) conspiracy to distribute methamphetamine 2) RICO conspiracy, and 3) carrying a firearm during and in relation to or in furtherance of a crime of violence or drug-trafficking crime. In addition, Mendoza argues: 1) that he was entitled to a sua sponte jury instruction for his drug conspiracy charge on the "buyer-seller rule," which instructs that a "conviction for conspiracy cannot be based solely on the purchase of an unlawful substance," United States v. Moe , 781 F.3d 1120, 1123 (2015) ; 2) that the trial judge gave erroneous supplementary jury instructions on the RICO conspiracy charge that confused the jury; and 3) that the jury may have convicted him under § 924(c) based on a legally invalid predicate offense.
Mendoza also originally argued that the judge improperly instructed the jury on the elements required to convict him under § 924(c) but later conceded that argument.
We discuss below Mendoza's argument that insufficient evidence supports his three challenged convictions. As we agree with Mendoza, we vacate those convictions and decline to reach Mendoza's other alleged errors.
A. Standard of Review
When this Court reviews a challenge to the sufficiency of the evidence supporting a criminal conviction, we perform a two-step analysis. First, we "consider the evidence presented at trial in the light most favorable to the prosecution." United States v. Nevils , 598 F.3d 1158, 1164 (9th Cir. 2010) (en banc). In other words, we cannot fashion an "exculpatory explanation" for admittedly incriminating evidence but need make only "reasonable inference[s]" in the prosecution's favor; we need not heed evidentiary theories, or affirm jury verdicts, that are based on "mere speculation." Id. at 1167. And second, we must determine whether the evidence, including any "evidence of innocence" or "lack of evidence of guilt," "could allow any rational trier of fact to find the essential elements of the crime beyond a reasonable doubt." Id. at 1164–65.
B. Mendoza's Conviction for Conspiracy to Distribute Methamphetamine
Mendoza first challenges the evidence supporting his conviction for conspiracy to distribute methamphetamine. To prove a conspiracy, the government must prove: "1) an agreement to accomplish an illegal objective; and 2) the intent to commit the underlying offense." United States v. Mincoff , 574 F.3d 1186, 1192 (9th Cir. 2009) (quoting United States v. Barragan , 263 F.3d 919, 922 (9th Cir. 2001) ). Circumstantial evidence can suffice to prove a conspiracy. See id.
Here, we must distinguish between a mere drug buyer, as Mendoza contends he was, and a participant in a drug-distribution conspiracy, as the government alleges he was. In cases like this, the buyer-seller rule dictates that "mere sales to [or purchases from] other individuals do not establish a conspiracy to distribute or possess with intent to distribute." United States v. Lennick , 18 F.3d 814, 819 n.4 (9th Cir. 1994). Rather, the government must show that Mendoza and the CRO "had an agreement to further distribute the drug in question": methamphetamine. Id.
We will find such an agreement and "uphold a conviction for conspiracy between buyer and seller where there is ‘evidence of a prolonged and actively pursued course of sales coupled with the seller's knowledge of and a shared stake in the buyer's illegal venture.’ " Moe , 781 F.3d at 1125 (quoting United States v. Ramirez , 714 F.3d 1134, 1140 (9th Cir. 2013) ). If we instead see only "a casual sale [or purchase] of drugs, of a quantity consistent with personal use on the part of the buyer, with no evidence of any subsequent (or planned) redistribution of purchased drugs," the evidence is generally insufficient to support a conspiracy conviction. Id. We assess " ‘the entire course of dealing’ between alleged co-conspirators," and consider whether the "drugs were sold on credit," the "frequency" and "quantity" of sales, and "whether the transactions were standardized," among other factors. Id. at 1125, 1126 (quoting Mincoff , 574 F.3d at 1194 ).
At trial, the government offered the following evidence of Mendoza's guilt:
• Mendoza was admittedly a member of the CRO when he was a teenager, and gang membership is typically "for life."
• Mendoza has multiple tattoos with CRO imagery, one of which commemorates a deceased CRO member and post-dates when Mendoza claims he left the gang.
• At Mendoza's arrest on June 2, 2013, he had with him 16.2 grams of pure methamphetamine in two separate bags, a police scanner, and a loaded gun, all consistent with drug "sales as opposed to personal use." Further, Mendoza was not found with either syringes or pipes that he could use to inject or smoke meth.
Mendoza argues that we cannot consider this evidence when analyzing the sufficiency of the evidence supporting Mendoza's convictions because, when the jury considered Mendoza's conduct in June 2013, the jury found him guilty of simple possession of methamphetamine but acquitted him of possession with intent to distribute. We rejected an analogous argument in United States v. Johnson , 804 F.2d 1078 (9th Cir. 1986). There, a jury acquitted the defendant of bank robbery but found him guilty of possessing stolen bank property. We held that evidence pertaining to the acquitted count could be considered in a sufficiency-of-the-evidence analysis of the guilty count. Id. at 1079–80, 1083. Mendoza argues that Johnson rejected an approach based on "inconsistent verdicts—rather than an assessment of all the evidence"—but he misreads the case. See id. at 1083 ("Therefore, the eyewitness testimony should not have been excluded from the assessment of the sufficiency of the evidence supporting Johnson's conviction under § 2113(c).").
• Mendoza purchased the 16.2 grams of methamphetamine he was caught with in 2013 from CRO leader Gaitan, and did so on credit rather than paying in cash.
• Mendoza had several phone or text conversations with CRO members. Mendoza called Gaitan once in 2013 to ask if he should come to his house, and then called a few weeks later to request methamphetamine. The day after the second call, Mendoza received a call from Gaitan on his mother's home phone to tell him that the drugs were ready. Mendoza also received a call from Loza on his mother's home phone several weeks later in 2013, and in 2016, Mendoza texted Antolin, a CRO foot soldier, to ask him for methamphetamine and used several CRO-related images in the conversation.
• When Mendoza was arrested in December 2016, he had a gun and 3.5 grams of methamphetamine in two baggies.
• The gun that Mendoza had in 2013 had been used to shoot at a car that was driven by a member of another gang just over a week before Mendoza was found with it.
The prosecution asserts that these phone calls establish that Loza collected "taxes" from Mendoza, but fairly read, the calls do not so establish. The calls do not allude to any obligation or debt that Mendoza owed Loza. We must make "reasonable inference[s]" in the prosecution's favor, Nevils , 598 F.3d at 1167, but we cannot give those conversations meaning that their words do not bear.
The government is incorrect when it asserts that Mendoza requested "42 grams of methamphetamine in the span of just three weeks, an amount well in excess of the quantity [Mendoza] claimed he personally used in a month's time." Viewing the texts between Mendoza and Antolin in the light most favorable to the prosecution, Mendoza asked for half an ounce (about 14 grams), which would apparently last him a month, and then asked for another half an ounce three weeks later. So, Mendoza actually asked for 28 grams in a span of three weeks. In other words, he twice requested a month's supply and his two requests were nearly a month apart.
Mendoza offered innocent interpretations of the above evidence, but we must ignore them, see Nevils , 598 F.3d at 1167.
Mendoza also offered the following as evidence of his innocence:
• When Mendoza was found with methamphetamine in June 2013 and in December 2016, he had with him no materials that could be used to cut (again, to dilute with another substance), weigh, package, or otherwise sell drugs.
• Mendoza was absent at the "mandatory" CRO meeting that the government surveilled and apparently suffered no consequences as a result.
• CRO members apparently did not speak with him regularly or even have his phone number.
• Mendoza participated in just four communications over seven months with CRO members out of the 21,000 gang communications the police intercepted.
• The two consummated or attempted drug sales between Mendoza and CRO members were ad hoc rather than standardized transactions. Specifically, Mendoza negotiated price, and threatened to buy methamphetamine from another seller (who the government did not establish was a CRO member) if Mendoza could not obtain a good price from his CRO contact.
In our view, the above evidence is insufficient to convict Mendoza of conspiracy to distribute methamphetamine beyond a reasonable doubt. True, the government offered some circumstantial evidence linking Mendoza to the CRO (for instance, Mendoza's 2013 tattoo and his conversations with CRO members) and to objects that are consistent with drug trafficking more generally (for instance, Mendoza's guns and police scanner). And Mendoza did once purchase methamphetamine from CRO leader Gaitan once without paying immediately in cash, and he attempted to buy drugs another time from CRO member Antolin. But even "ample proof that [Mendoza] possessed and [bought] drugs" is insufficient on its own for a conspiracy conviction. Ramirez , 714 F.3d at 1140. The government must prove with sufficient evidence "an agreement" between Mendoza and CRO co-conspirators under which Mendoza "would ‘further distribute the drugs’ " that he bought from the CRO. Id. (quoting Lennick , 18 F.3d at 819 ). When we rely on circumstantial evidence to establish an agreement, as we do here, "what we are looking for is evidence of a prolonged and actively pursued course of sales" and Mendoza's "knowledge of" and "shared stake in" the CRO's drug operation. Id.
Again, Mendoza offers plausible explanations for much of this evidence that do not support the conspiracy and gun possession counts. For instance, he says the methamphetamine he had in his possession was for his personal use, argues that he kept the gun only for personal protection, and contends that Gaitan sold him methamphetamine on credit due only to their long-time friendship. But we must draw all "reasonable inference[s]" in the prosecution's favor, and all this evidence is susceptible to "reasonable" incriminating inferences, so we consider it as circumstantial evidence of Mendoza's guilt. See Nevils , 598 F.3d at 1164.
Assessed under these criteria, the evidence here of any agreement or shared stake is lacking compared to what we have previously found sufficient. Consider our two decisions in United States v. Mincoff and United States v. Loveland , 825 F.3d 555 (9th Cir. 2016). In Mincoff , we found sufficient evidence for a drug conspiracy conviction because the evidence, including live testimony and multiple recorded calls outlining the drug buyer's future plans to re-sell the product to another buyer, "demonstrated an agreement to further distribute the cocaine, rather than the ‘mere purchase’ of large quantities of drugs." 574 F.3d at 1194. Here, in contrast, we have no "recorded calls" or testimony that Mendoza was to "further distribute" methamphetamine (or actually did), and Mendoza was never found with typical implements of drug sales like cutting agents, scales, or numerous small baggies.
In Loveland , we vacated a conspiracy conviction and held that even evidence of "repeated sales and large quantities could not sustain a conspiracy conviction" absent evidence of a defendant's "involvement" in future drug sales. 825 F.3d at 560. Here, we have no evidence of repeated, large-quantity sales and barely any evidence linking Mendoza to future drug sales. Mendoza was caught with 16.2 grams of methamphetamine in June 2013 and just 3.5 grams in December 2016; the government's evidence proves that Mendoza purchased drugs from the CRO at most three times in three years. The only evidence even suggesting that Mendoza might be involved in any future drug sales is the single time he purchased methamphetamine from CRO leader Gaitan without immediately paying in cash. See Moe , 781 F.3d at 1125 & n.1 (recognizing that drug sales on credit suggests an agreement to further distribute the drugs). The evidence that Mendoza and the CRO had the requisite "agreement" to distribute methamphetamine falls well short of the evidence we found sufficient in Mincoff , and short as well of the evidence we found insufficient in Loveland.
While this gap in the government's case might not be fatal on its own, Mendoza also presented multiple items of evidence that affirmatively contradict the government's theory. For instance, the government argues that Mendoza was a "senior foot soldier" in the CRO. But Mendoza missed a "mandatory" gang meeting and suffered no consequences, and the CRO's leaders did not even have his phone number. In fact, one of the CRO's leaders, Mendoza's close friend from childhood, resorted to messaging Mendoza using a text-messaging feature on a videogame the two played, and he often failed to reach Mendoza even through this method because the two rarely played at the same time. That is not how co-conspirators usually communicate. Cf. Moe , 781 F.3d at 1126 (finding sufficient evidence to support a drug conspiracy conviction where the buyer and seller "communicated closely together and coordinated their actions"). The government submits that Mendoza was "required to pay taxes" (i.e. , the CRO's share of drug sale profits) on drugs he acquired from the CRO and then re-sold and was then "hounded ... for repayment." But the communications between Mendoza and CRO members the government cites make no mention of "taxes" or a debt owed on drugs obtained by Mendoza, and after CRO leader Loza told Mendoza that he tried to reach him for "seven days straight," he did not berate Mendoza for not paying taxes but instead told him to "take care." That is hardly "hound[ing]," and certainly not what one would expect a gang leader to tell his in-debt inferior. And the government contends that in May and June 2016, Mendoza "communicated with [a CRO member] for the purpose of obtaining methamphetamine for sale on behalf of the CRO." But Mendoza had to pester that CRO member for almost a month and then threaten to purchase drugs from someone else to convince the gang member to sell to him, and even after Mendoza convinced the gang member to make a sale, Mendoza haggled over price and quantity. There was no mention in the communication of resale. That is not how co-conspirators in a drug-trafficking operation transact. See Moe , 781 F.3d at 1126 (recognizing that drug co-conspirators typically engage in "standardized" transactions). Indeed, we have held that a buyer-seller relationship (as opposed to conspiracy) is particularly likely when, as here, the downstream buyer called the upstream seller (rather than vice versa) and when the downstream buyer was "free to shop elsewhere." Loveland , 825 F.3d at 563 ; see id. at 562.
We find only more support for our conclusion when we step back to consider the "entire course of dealing" between Mendoza and the CRO. Moe , 781 F.3d at 1125 (quoting Mincoff , 574 F.3d at 1194 ). That "entire course of dealing" consisted of four phone calls and one short text conversation out of 21,000 communications and seven months of intensive audio surveillance on the CRO, along with one methamphetamine purchase from a CRO source, a second attempted purchase, and a later third purchase, spread across over three years' time and totaling just 47.5 grams. That may be more conversations with drug kingpins and purchases from methamphetamine dealers than most people who are not addicted to methamphetamine have had in their lives, but it is still a thin circumstantial basis for a drug conspiracy conviction, especially for an addict. Compare Moe , 781 F.3d at 1123, 1126 (finding sufficient evidence to support a drug conspiracy conviction when a drug buyer was "not just a casual or occasional buyer" but participated in "at least seven" transactions together involving 140 grams of meth, "94 cell phone contacts," and "51 text messages" with the seller), with Ramirez , 714 F.3d at 1140 (finding insufficient evidence to support a drug conspiracy conviction even with four sales of "large quantities" of methamphetamine because there was little evidence of an "agreement ... to distribute meth"). Compared to the evidence in Moe , the "course of dealing" between Mendoza and the CRO was more of a trickle.
This assumes the government's theory that Mendoza obtained from the CRO the methamphetamine that he had with him at his December 2016 arrest.
Given all the evidence just discussed, and even after making all "reasonable inference[s]" in the prosecution's favor, Nevils , 598 F.3d at 1167, the government simply did not establish the "prolonged and actively pursued course of [drug] sales" for which we look when deciding, in the absence of direct evidence of an agreement, if there is "sufficient evidence of an agreement" to distribute drugs. Loveland , 825 F.3d at 560 (quoting Ramirez , 714 F.3d at 1140 ). Even if the evidence of Mendoza's relationship with the CRO raises "a reasonable suspicion or probability" of his guilt, that level of certainty "is not enough." United States v. Espinoza-Valdez , 889 F.3d 654, 659 (9th Cir. 2018). "Guilt, according to the basic principles of our jurisprudence, must be established beyond a reasonable doubt." Id. Given this strict standard, no reasonable jury could determine beyond a reasonable doubt that Mendoza was part of a conspiracy to distribute methamphetamine. The evidence the government offered at trial as to the conspiracy count was insufficient.
C. Mendoza's Conviction for RICO Conspiracy
Mendoza next challenges the sufficiency of the evidence supporting his conviction for RICO conspiracy. Of the five elements of a RICO conspiracy set out in the jury instructions, Mendoza challenged only one: whether he actually "became a member of the conspiracy knowing of its object and intending to help further or facilitate" it. And as with the drug conspiracy count, we may rely on circumstantial evidence. See Mincoff , 574 F.3d at 1192.
The government's theory, at trial and on appeal, is that Mendoza was a part of the CRO's racketeering conspiracy because he sold drugs for the gang—the same theory the government pursued as to Mendoza's drug conspiracy charge. For these parallel theories, the government offered parallel evidence: the same evidence supported both the drug conspiracy charge and the RICO conspiracy charge. And unsurprisingly, Mendoza countered with the same argument that he used to challenge the drug conspiracy charge—that he was a mere drug user, not a conspirator in distributing drugs to others—and with the same counterevidence. The government's RICO conspiracy case turns on the same element of proof and on the same evidence as did its drug conspiracy case. So, the outcome here is the same as with the drug conspiracy: insufficient evidence supports Mendoza's conviction for RICO conspiracy. Cf. Espinoza-Valdez , 889 F.3d at 659 (analyzing as a whole the overlapping evidence for two separate conspiracy convictions and finding insufficient evidence for either).
D. Mendoza's Conviction for Possessing a Gun in Relation to or in Furtherance of a Drug-Trafficking Crime or Crime of Violence
Last, Mendoza challenges the sufficiency of the evidence underlying his conviction under 18 U.S.C. § 924(c) for carrying or possessing a firearm "in relation to" or "in furtherance" of a drug-trafficking crime or crime of violence. To obtain a conviction under § 924(c), the government must prove that the defendant "committed [an] underlying crime" of violence or drug trafficking, United States v. Hunter , 887 F.2d 1001, 1003 (9th Cir. 1989) (per curiam), and also "possessed [a] weapon to promote or facilitate th[at] underlying crime," United States v. Krouse , 370 F.3d 965, 967 (9th Cir. 2004). The government must have sufficient evidence of "all elements of the crime created by section 924(c)(1)." Hunter , 887 F.2d at 1003.
We focus on the first element of proof: whether Mendoza committed an "underlying crime" of violence or drug trafficking. The government suggested at trial that the jury could find that Mendoza committed any or all of three possible underlying crimes: 1) possession of methamphetamine with intent to distribute, which the judge instructed the jury was a drug-trafficking crime; 2) conspiracy to distribute methamphetamine, which the judge instructed the jury was a drug-trafficking crime; and 3) RICO conspiracy, which the judge instructed the jury was "crime of violence." The jury acquitted Mendoza of possession of methamphetamine with intent to distribute, so that charge cannot serve as an underlying crime supporting Mendoza's § 924(c) conviction.
We note that the Supreme Court recently held in United States v. Davis , ––– U.S. ––––, 139 S. Ct. 2319, 204 L.Ed.2d 757 (2019), that § 924(c)(3)(B) is unconstitutionally vague, see id. at 2324, causing some courts to conclude that "RICO conspiracy is not a crime of violence" and can no longer support a conviction under § 924(c). E.g., United States v. Capers , 20 F.4th 105, 120 (2d Cir. 2021). This Court has yet to publish an opinion that follows Davis and reaches the same conclusion as Capers , and we need not take that step here.
This leaves Mendoza's charges for conspiracy to distribute methamphetamine and RICO conspiracy. We concluded above that the government presented insufficient evidence to justify those two convictions. So, by the same token, we conclude that the government offered insufficient evidence to prove that Mendoza committed either of those two crimes as an "underlying crime" of drug trafficking or violence for the purposes of § 924(c). Hunter , 887 F.2d at 1003. Of course, the government need not "separately charge[ ] ... and convict[ ]" a defendant charged under § 924(c) "of the underlying offense," id. , so whether or not Mendoza is convicted of either conspiracy is irrelevant. But because the government failed to prove the conspiracy charges beyond a reasonable doubt, the government failed to prove that Mendoza "committed" either conspiracy offense. As a result, neither conspiracy to distribute methamphetamine nor RICO conspiracy can serve as the underlying crime for Mendoza's conviction under § 924(c). The government thus failed to prove an essential element of Mendoza's § 924(c) offense, leaving that conviction unsupported by sufficient evidence.
* * *
Given our conclusion that sufficient evidence did not support Mendoza's convictions for drug-trafficking conspiracy, RICO conspiracy, and possession of a firearm in furtherance of or in relation to a violent or drug-trafficking offense, we need not address Mendoza's other arguments.
Mendoza's argument that he was entitled to a sua sponte instruction on the "buyer-seller rule" provides one explanation for the apparently inconsistent jury verdicts below. Without that instruction, the jury may have believed that it could convict Mendoza of conspiracy to distribute methamphetamine and of RICO conspiracy based only on an agreement to purchase methamphetamine from CRO members. If the jury so believed, but did not believe that Mendoza had any agreement with the CRO to re-sell or otherwise distribute the methamphetamine for the gang's profit, then the jury could reasonably have reached the ostensibly inconsistent verdict below: not guilty of possession with intent to distribute but guilty of both simple possession and conspiracy to distribute. Still, we need not address the buyer-seller rule issue, even though it might give us a chance to reconcile the jury's verdicts. We conclude above that the evidence adduced was insufficient to support the three convictions that Mendoza challenges, incorporating into our analysis the buyer-seller rule that evidence of a buyer-seller relationship alone is not sufficient evidence of conspiracy. This appraisal resolves Mendoza's appeal and dispenses with any need to address his other arguments, including whether Mendoza was entitled to a sua sponte instruction on the buyer-seller rule. We note that any such instructional error would entitle Mendoza only to a retrial, a lesser remedy than the vacation, remand, and resentencing that Mendoza's insufficiency of evidence arguments have merited him.
III. CONCLUSION
We thus VACATE Henry Mendoza's convictions for conspiracy to distribute drugs under 21 U.S.C. § 846, RICO conspiracy under 18 U.S.C. § 1962(d), and carrying a firearm "during and in relation to" or "in furtherance of" a crime of violence or drug-trafficking crime under 18 U.S.C. § 924(c)(1)(A), and we REMAND to the district court to grant a judgment of acquittal on those charges and to resentence Mendoza accordingly.