CARLOS GRANDA v. UNITED STATES OF AMERICA
No. 17-15194
United States Court of Appeals, Eleventh Circuit
March 11, 2021
D.C. Docket Nos. 1:16-cv-23426-DMM; 1:07-cr-20155-DMM-3
[PUBLISH]
Aрpeal from the United States District Court for the Southern District of Florida
(March 11, 2021)
Before WILLIAM PRYOR, Chief Judge, JORDAN and MARCUS, Circuit Judges.
This is Carlos Granda‘s second trip to our Court. In this reverse sting case, Granda served as the lookout for a criminal crew that attempted to rob a tractor trailer purportedly filled with sixty to eighty kilograms of cocaine. The scheme netted him several serious criminal convictions. On direct appeal, we affirmed his convictions and sentence. Now, Granda appeals the denial of his
Granda‘s claim turns entirely on the observation that one of these offenses -- conspiracy to commit Hobbs Act robbery -- no longer qualifies as a valid crime-of-violence predicate after United States v. Davis, 139 S. Ct. 2319, 2336 (2019) and Brown v. United States, 942 F.3d 1069, 1075 (11th Cir. 2019)§ 924(o). Granda says we must vacate his conviction because of the possibility that the jury relied on an invalid predicate and, therefore, convicted him of a non-existent crime.
Granda faces two insuperable problems: he cannot overcome the procedural default of his claim (which he raises for the first time on collateral attack), nor could he otherwise prevail on the merits. Among the shortcomings that defeat his claim is a fundamental one that cuts across both the procedural and merits inquiries: all of the
Accordingly, we affirm.
I.
A.
In 2007, Carlos Granda‘s (“Granda“) brother Paulino told a confidential informant (“CI“) that he was putting together a criminal crew and looking for work. On January 26, 2007, the CI introduced Paulino to a detective who was working in an undercover capacity. The undercover agent claimed to be a disgruntled drug trafficker who wanted to rob his boss; and he knew where in Miami his boss kept a tractor-trailer full of some sixty to eighty kilograms of cocaine. Paulino orchestrated a plan to rob the tractor-trailer at gunpoint. He and his crew planned to approach the truck while armed and dressed as a police S.W.A.T. team, remove and detain the drivers, and steal the truck and the drugs. Paulino explained that he would secure the perimeter by having his associates do surveillance. He added that, if it became necessary, he and his crew were willing to shoot the driver and anyone else who resisted. Appellant Granda‘s role was to serve as a lookout. Paulino also asked him to bring a gun to the robbery.
Granda and the crew met up at his parents’ apartment, where members of the crew packed up guns, as well as construction ties to use as handcuffs. The crew then drove to a warehouse where they expected to find the stash truck. Granda stationed himself in his car near the entrance to the warehouse. Just as a member of the crew approached the truck, police officers revealed themselves and initiated a takedown. One of the would-be robbers raised his revolver, triggering a shootout. In the ensuing minutes, the police killed two members of the crew and wounded two others. Granda attempted to flee in his car but encountered a police blockade. He tried to evade the blockade by turning in front of a police car, striking it. The police arrested Granda, and found handcuffs, but not a firearm, in his car. They did find a loaded Hi-Point 9mm semi-automatic pistol in Paulino‘s car.
B.
A grand jury sitting in the Southern District of Florida returned a superseding indictment against Carlos Granda and his co-defendants. Granda was сharged with conspiracy to possess with intent to distribute cocaine in violation of
At trial, Granda‘s counsel insisted that Granda lacked the requisite criminal intent
Granda filed several unsuccessful collateral attacks, including an earlier
Proceeding pro se, Granda then sought leave to file the instant
A magistrate judge issued a Report and Recommendation concluding that Granda could not establish actual prejudice to overcome procedural default, and that, while the conspiracy to commit Hobbs Act robbery (Count 3) did not independently support Granda‘s
Proceeding pro se, Granda filed this appeal, and a full round of briefing ensued. Then the Supreme Court decided Davis, 139 S. Ct. at 2336, concluding that the
II.
To begin, we have (and the district court had) jurisdiction over Granda‘s
To be sure, we gave Granda leave to file a Johnson challenge -- the Supreme Court had not yet decided Davis, so we did not (nor could we then) certify that Granda‘s second petition “contain[ed]” the new rule of constitutional law Davis announced. See
But to resolve the Johnson claim we did authorize, we can, indeed we must, apply the controlling Supreme Court law of Davis. As this Court held en banc in United States v. Johnson, “when ‘a precedent of the Supreme Court has direct application,’ we must follow it.” 921 F.3d 991, 1001 (11th Cir.) (en bаnc) (alteration accepted) (citation omitted), cert. denied, 140 S. Ct. 376 (2019); see also Brown, 942 F.3d at 1072 (applying Davis to resolve a pre-Davis petition that raised a Johnson claim). Davis extended the reasoning of Johnson, providing us with the answer to a question central to Granda‘s petition: whether the
III.
A.
Granda challenges his conviction on Count 6, conspiracy to use or carry a firearm during and in relation to, and to possess a firearm in furtherance of, a drug-trafficking crime or a crime of violence in violation of
Except to the extent that a greater minimum sentence is otherwise provided by this subsection or by any other provision of law, any person who, during and in relation to any crime of violence or drug trafficking crime for which the person may be prosecuted in a court of the United States, uses or carries a firearm, or who, in furtherance of any such crime, possesses a firearm, shall, in addition to the punishment provided for such crime of violence or drug trafficking crime--
(i) be sentenced to a term of imprisonment of not less than 5 years . . . .
For purposes of this subsection the term “crime of violence” means an offense that is a felony and--
(A) has as an element the use, attempted use, or threatened use of physical force against the person or property of another, or
(B) that by its nature, involves a substantial risk that physical force against the person or property of another may be used in the course of committing the offense.
Since Granda filed the instant
- Count 1, conspiracy to possess with intent to distribute cocaine, is a “drug trafficking crime” and is, therefore, a
valid predicate for Granda‘s Count 6 conviction. (Granda does not dispute that Counts 1 and 2 are “drug trafficking crimes,” nor could he.) - Count 2, attempt to possess with intent to distribute cocaine, is a “drug trafficking crime” and is, therefore, a valid predicate for Granda‘s Count 6 conviction.
- Count 3, conspiracy to commit Hobbs Act robbery, is not a valid potential predicate for Granda‘s Count 6 conviction. It is not a drug-trafficking offense, so it must qualify, if аt all, as a crime of violence. But the
§ 924(c)(3) residual clause is unconstitutionally vague under Davis, 139 S. Ct. at 2336, and conspiracy to commit Hobbs Act robbery categorically is not a crime of violence under the§ 924(c)(3) elements clause. Brown, 942 F.3d at 1075. - Count 4, attempted Hobbs Act robbery, categorically qualifies as a crime of violence under the
§ 924(c)(3) elements clause and is, therefore, a valid predicate for Granda‘s Count 6 conviction. United States v. St. Hubert, 909 F.3d 335, 351-53 (11th Cir. 2018), abrogated on other grounds by Davis, 139 S. Ct. at 2336. - Count 5, attempted carjacking, categorically qualifies as a crime of violence under the
§ 924(c)(3) elements clause and is, therefore, a valid predicate for Granda‘s Count 6 conviction. Ovalles v. United States, 905 F.3d 1300, 1305-07 (11th Cir. 2018). - Davis applies retroactively to Granda‘s case. Hammoud, 931 F.3d at 1037-39.
Thus, any one of Counts 1, 2, 4, or 5 could have validly predicated Granda‘s Count 6 conviction. Count 3 could not have. But we cannot tell from the general verdict whether the jury relied on Counts 1, 2, 4, or 5 (alone or in combination with one another), or on any of them in addition to Count 3, or on Count 3 alone. Nor can we tell from the superseding indictment, which simply alleged in Count 6:
[Granda and his co-defendants] did knowingly and intentiоnally combine, conspire, confederate, and agree with each other to use and carry a firearm during and in relation to a crime of violence and a drug trafficking crime, and to possess a firearm in furtherance of a crime of violence and a drug trafficking crime, which are felonies prosecutable in a court of the United States, specifically . . . Counts 1, 2, 3, 4, and 5 of this Indictment . . . .
The jury instructions likewise did not shed any light on the problem. The district court told the jury that in order to sustain a conviction on Count 6, they had to find beyond a reasonable doubt that “the object of the unlawful plan was to use or carry a firearm during and in relation to, or to possess a firearm in furtherance of, one of the federal drug trafficking crimes, or one of the federal crimes of violence, or both, as charged in counts 1, 2, 3, 4, or 5 of the Superseding Indictment.”
If the jury relied on Count 3 alone, Granda would have been convicted of violating an unconstitutionally vague law, which “[i]n our constitutional order . . . is no law at all.” Davis, 139 S. Ct. at 2323. Granda claims that we cannot definitively rule out this possibility and, therefore, must vacate his conviction on Count 6. We remain unpersuaded.
B.
First, procedural default. This problem arises because Granda did not argue in the trial court, or on direct appeal, that his
We review de novo whether procedural default precludes a
1. Granda‘s claim is not sufficiently novel to establish cause.
“[W]here a constitutional claim is so novel that its legal basis is not reasonably available to counsel, a defendant has cause for his failure to raise the claim.” Howard v. United States, 374 F.3d 1068, 1072 (11th Cir. 2004) (quoting Reed v. Ross, 468 U.S. 1, 16 (1984)) (internal quotation marks omitted). “That an argument might have less than a high likelihood of success has little to do with whether the argument is available or not. An argument is available if there is a reasonable basis in law and fact for it.” Pitts v. Cook, 923 F.2d 1568, 1572 n.6 (11th Cir. 1991). “[T]he question is not whether subsequent legal developments have made counsel‘s task easier, but whether at the time of the default the claim was available at all.” McCoy v. United States, 266 F.3d 1245, 1258 (11th Cir. 2001) (internal quotation marks and citation omitted).
Granda brought his claim under Johnson (arguing that Johnson‘s reasoning should be extended to
The third category -- when a Supreme Court decision disapproves of “a practice [the Supreme Court] arguably has sanctioned in prior cases” -- is more complicated. Id. (internal quotation marks and citation omitted). For this category, we analyze “whether others were recognizing
Granda‘s best argument that his defaulted vagueness claim was not available on direct appeal is that at the time of that appeal,1 James v. United States, 550 U.S. 192, 210 n.6 (2007) had directly rejected the argument that the ACCA‘s residual clause was unconstitutionally vague. See Johnson, 576 U.S. at 594 (In James, “the Court rejected suggestions ... that the [ACCA‘s] residual clause violates the Constitution‘s prohibition of vague criminal lаws.“).
However, James did not consider the
These claims did not succeed. But if James did not deprive litigants of the tools to challenge even the ACCA‘s residual clause on vagueness grounds, it surely did not deprive them of the tools to challenge the
Moreover, the case law extant at the time of Granda‘s appeal confirms that he did not then lack the “building blocks of” a due process vagueness challenge to the
That few, if any, litigants had contended that the
2. Granda cannot show actual prejudice.
Even if Granda could show cause, however, he still could not overcome the procedural default of his vagueness claim because he cannot show actual prejudice.
“To prevail on a cause and prejudice theory, a petitioner must show actual prejudice. Actual prejudice means more than just the possibility of prejudice; it requires that the error worked to the petitioner‘s actual and substantial disadvantage, infecting his entire trial with error of constitutional dimensions.” Fordham, 706 F.3d at 1350 (alteration accepted) (internal quotation marks and citation omitted). The actual prejudice standard is “more stringent than the plain error standard.” Parks v. United States, 832 F.2d 1244, 1245 (11th Cir. 1987) (internal quotation marks omitted). The “ultimate inquiry” is: “Did the intrusion affect the jury‘s deliberations and thereby its verdict?” Ward v. Hall, 592 F.3d 1144, 1178-79 (11th Cir. 2010) (internal quotation marks and citation omitted).
So it is not enough for Granda to show that the jury may have relied on the Count 3 Hobbs Act conspiracy conviction as the predicate for his Count 6
the case, there was “no substantial likelihood” that the jury would have mitigated a first-degree murder conviction to manslaughter if “only the malice instructions had been better framed“).
Granda cannot make this showing. Based on his role as a lookout in a conspiracy and an attempt to rob at gunpoint a truck carrying some sixty to eighty kilograms of cocaine, the jury unanimously found Granda guilty of conspiracy and attempt to possess cocaine with intent to distribute, attempted carjacking, conspiracy to commit Hobbs Act robbery, and attempt to commit Hobbs Act robbery. Granda, 346 F. App‘x at 525. The trial record makes it abundantly clear that all of these findings rested on the same operative facts and the same set of events -- the jury found beyond a reasonable doubt that Granda had conspired and attempted to rob the truck in order to possess and distribute the cocaine it held.
The evidence supporting Granda‘s alleged conspiracy to possess a firearm in furtherance of these crimes included Paulino‘s instruction to Granda, who was to act as a lookout during the robbery, to bring a pistol. Notably, Granda also attended a meeting at which the co-conspirators packed guns and makeshift handcuffs for the robbery and discussed the execution of the crime.
As for Granda‘s knowledge of the drug-trafficking objective of the robbery, this Court explained in affirming his convictions on direct appeal:
[T]he jury was presented with evidence that (1) Paulino . . . specifically identified Granda in discussing the drug theft . . . ; (2) Granda was present during two-drug related meetings, one of which involved explicit discussion of marijuana; (3) Granda had telephone contact with various members of the conspiracy during critical times in planning and executing the drug theft; (4) vague references to the cocaine robbery were made to Granda or in his presence; (5) and Granda was apprehended within close proximity to the scene of the robbery.
Id. at 527. The objective of the robbery and the carjacking was the same: to obtain and sell the multi-kilogram quantity of cocaine that was to be taken by force from the truck. So the jury could not have concluded that Granda conspired to possess a firearm in furtherance of his robbery conspiracy without also finding at the same time that he conspired to possess the firearm in furtherance of his conspiracy and attempt to obtain and distribute the cocaine, his attempt at carjacking, and the attempt at the robbery itself.
United States v. Cannon, No. 16-16194, manuscript op. at 40-47 (11th Cir. Feb. 3, 2021) (“Lance Cannon“), which presents very similar facts, underscores and compels this conclusion. There, the defendants were convicted of several crimes -- including using and carrying a firearm during and in relation to a drug trafficking crime in violation of
[T]he trial evidence established the same facts and time period underlying each [predicate offense]. The cocaine the defendants were planning to rob from the narcotics traffickers was the same cocaine they were planning to possess with the intent to distribute. Undisputedly, the goal of the robbery scheme was to steal cocaine from a stash house so they could then distribute it themselves. . . . No reasonable juror could have found that Cannon and Holton carried their firearms in relation to the Hobbs Act robbery conspiracy but not the cocaine conspiracy.
Id. at 42-43. And the jury had convicted the defendants of both the underlying Hobbs Act conspiracy and the conspiracy to possess cocaine with intent to distribute, which “conclusively establish[ed] thаt the jury unanimously found beyond a reasonable doubt that the defendants were conspiring to rob the stash house of cocaine so they could then possess and distribute the cocaine themselves.” Id. at 42.
The “trial record ma[de] clear that the two predicate . . . crimes were so inextricably intertwined that no rational juror could have found that [the defendants] carried a firearm in relation to one predicate but not the other.” Id. at 41. So, too, with Granda‘s predicates. See also In re Cannon, 931 F.3d 1236, 1243-44 (11th Cir. 2019) (observing in dicta that, even on the limited record available at the second or successive
Granda cites no record evidence to suggest otherwise. At trial, he did not offer a theory that would have permitted the jury to drive a wedge between any plan to possess a firearm in furtherance of his conspiracy-to-rob and his attempt-to-rob. As we‘ve noted, Granda‘s defense at trial was that he believed Paulino, the leader of the conspiracy, was working as a government informant; Granda thus argued that he thought he was assisting a legitimate law enforcement effort and lacked the requisite criminal intent. But the jury plainly did not credit this account, finding Granda guilty of conspiracy and attempt to commit robbery and to possess cocaine with intent to distribute, along with attempted carjacking. If the jury did not believe Granda‘s
Granda did offer a theory that might have permitted jurors to distinguish between his conspiracy-to-rob and the drug-trafficking offenses -- he argued “that he knowingly participated in a robbery, but not a cocaine robbery,” Granda, 346 F. App‘x at 527 -- but the jury rejected this defense, too, finding beyond a reasonable doubt that he conspired and attempted to possess cocaine with the intent to distribute.3 And we know that the jurors did not split into two camps, one of which found that Granda conspired to possess a firearm in furtherance of one or more valid predicates while the other found Granda conspired to possess a firearm only in furtherance of the Hobbs Act conspiracy. Besides the lack of any record evidence to suggest such a split, the district court instructed the jury that it must “unanimously agree upon the way in which the defendant” violated
jurors found that they did so only during the cocaine conspiracy“); In re Price, 964 F.3d 1045, 1049 (11th Cir. 2020) (“We not only can, but we must, presume that juries follow their instructions.“).
Granda argues, nevertheless, that the jury‘s acquittal on Count 7 -- which charged Granda with using and carrying a firearm during, and with possessing a firearm in furtherance of, a crime of violence and drug-trafficking crime in violation of
In short, just as in Lance Cannon, the alternative predicate offenses are inextricably intertwined -- each arose from the same plan and attempt to commit armed robbery of a tractor-trailer full of cocaine. The tightly bound factual relationship of the predicate offenses precludes Granda from showing a substantial likelihood that the jury relied solely on Count 3 to predicate its conviction on Cоunt 6. Cf. Lomelo v. United States, 891 F.2d 1512, 1516, 1519 (11th Cir. 1990).
3. Granda cannot show actual innocence.
Inasmuch as Granda cannot establish both cause and prejudice, his only way around procedural default would
To demonstrate actual innocence of the
Since Granda can show neither cause, nor prejudice, nor actual innocence, he cannot overcome procedural default.
C.
Even were we to assume that the overlap between the valid and invalid рredicates did not bar Granda‘s claim on procedural default grounds, the same problem would rear its head again on the merits. The inextricability of the alternative predicate crimes compels the conclusion that the error Granda complains about -- instructing the jury on a constitutionally invalid predicate as one several of potential alternative predicates -- was harmless.
On collateral review, the harmless error standard mandates that “relief is proper only if the . . . court has grave doubt about whether a trial error of federal law had substantial and injurious effect or influence in determining the jury‘s verdict. There must be more than a reasonable possibility that the error was harmful.” Davis v. Ayala, 576 U.S. 257, 267-68 (2015) (internal quotation marks and citations omitted); Ross v. United States, 289 F.3d 677, 682 (11th Cir. 2002) (per curiam) (applying this standard to a
“We do not phrase the [Brecht] requirement as a burden of proof, for it is not.” Trepal v. Sec’y, Fla. Dep‘t of Corr., 684 F.3d 1088, 1111 n.26 (11th Cir. 2012); see O‘Neal v. McAninch, 513 U.S. 432, 436, 438 (1995) (explaining that Brecht‘s use of language suggesting the petitioner must establish prejudice “is not determinative” and “deliberately phras[ing] the issue . . .
This record does not provoke grave doubt about whether Granda‘s
Granda marshals several arguments against the application of Brecht. None are persuasive. First, Granda says that it is improper to rely on the presence of alternative, valid grounds for conviction to conduct a harmless error anаlysis. This argument largely rests on Stromberg v. California, 283 U.S. 359, 367-68 (1931), which stands for “the principle that, where a provision of the Constitution forbids conviction on a particular ground, the constitutional guarantee is violated by a general verdict that may have rested on that ground.” Griffin v. United States, 502 U.S. 46, 53 (1991); see also Yates v. United States, 354 U.S. 298, 312 (1957) (“[W]e think the proper rule to be applied is that which requires a verdict to be set aside in cases where the verdict is supportable on one ground, but not on another, and it is impossible to tell which ground the jury selected.“), overruled in part on other grounds by Burks v. United States, 437 U.S. 1, 10-18 (1978).
In Parker v. Sec’y for Dep‘t of Corr., also cited by the petitioner, we held that Stromberg barred the conclusion that an allegedly deficient felony murder instruction -- the judge had given a written instruction on felony murder but had failed to mention felony murder during his oral instructions -- was harmless solely because there was a valid, alternative basis for the petitioner‘s first-degree murder conviction
But, notably, after we decided Parker, the Supreme Court held in Hedgpeth v. Pulido that Stromberg error is subject to the Brecht harmless error standard. Hedgpeth, 555 U.S. 57, 58, 62 (2008) (per curiam). The Court concluded that a Stromberg error is not structural; it does not “require[] automatic reversal.” Id. at 60-62. It expressly rejected the standard the Ninth Circuit had applied in that case, which required that the conviction be overturned unless the court was “absolutely certain” that the jury relied on a valid ground. Id. at 62 (citation omitted). This standard was “plainly inconsistent with” Brecht, id., which held that reversal is warranted only when the petitioner suffered “actual prejudice” from the error, Brecht, 507 U.S. at 637 (citation omitted). “An instructional error arising in the context of multiple theories of guilt” does not “vitiate[] all the jury‘s findings.” Hedgpeth, 555 U.S. at 61.
Thus, it is proper to look at the record to determine whether the invalid predicate actually prejudiced the petitioner -- that is, aсtually led to his conviction -- or whether the jury instead (or also) found the defendant guilty under a valid theory. Limiting harmless error review to the invalid basis alone, as Granda suggests, would in this case effectively entitle Granda to relief if the jury may have relied on the invalid predicate to convict. But this result cannot be reconciled with with Brecht‘s requirement that “[t]here must be more than a reasonable possibility that the error was harmful,” Ayala, 576 U.S. at 268 (internal quotation marks and citation omitted), or with the Supreme Court‘s rejection in Hedgpeth of a standard requiring automatic reversal if there is any chance that the jury relied on an invalid basis, see id. (explaining that Brecht rejects the idea that relief is appropriate based on “mere speculation that the defendant was prejudiced by trial error“) (internal quotation marks and citation omitted).
Supreme Court cases decided since Hedgpeth confirm this conclusion. See Skilling v. United States, 561 U.S. 358, 414 n.46 (2010) (“The Fifth Circuit appeared to prejudge this issue, noting that, ‘if any of the three objects of Skilling‘s conspiracy offers a legally insufficient theory,’ it ‘must set aside his convictiоn.’ That reasoning relied on the mistaken premise that [Hedgpeth] governs only cases on collateral review. Harmless-error analysis, we clarify, applies equally to cases on direct appeal. Accordingly, the Fifth Circuit, on remand, should take a fresh look at the parties’ harmless-error arguments.” (citations omitted)); see also Black v. United States, 561 U.S. 465, 467, 474 (2010) (recognizing the possibility that error in instructing an honest-services fraud defendant‘s jury on two alternative
Next, Granda claims that “[t]he categorical approach requires the Court to presume that [his] conviction [on] Count 6 rests on [the] Hobbs Act conspiracy.” This argument misapprehends the categorical approach, which courts use to determine whether a particular offense qualifies under crime-of-violence-type elemеnts clauses such as those found in
Granda reasons that because the general verdict does not reveal which predicate the jury relied on to find guilt for the
Granda claims that the categorical approach must apply because determining that the jury did not rely solely on the Hobbs Act conspiracy predicate to convict would constitute impermissible judicial factfinding in violation of Alleyne v. United States, 570 U.S. 99, 114-16 (2013). Alleyne applies to findings of fact that increase a mandatory minimum, thereby “alter[ing] the prescribed range of sentences to which a defendant is exposed . . . in a manner that aggravates the punishment.” Id. at 108. Such facts, Alleyne holds, must be proven to a jury beyond a reasonable doubt. Id. For example, a sentencing judge may not look at the evidence afresh to determine by a preponderance of the evidence whether a defendant “brandished” a firearm during an offense and is therefore subject to an increased mandatory minimum. See id. at 117. But a judge conducting a Brecht harmless error analysis does not find a fact at all; instead, the judge asks as a matter of law whether there is grave doubt about whether an instruction on an invalid predicate substantially influenced what the jury already found beyond a reasonable doubt. See Phillips, 849 F.3d at 993. Absent any sound reason for doing so, we decline Granda‘s invitation to adopt what we see as an unprecedented expansion of the categorical approach.
But this prima facie decision is not inconsistent with a holding that, on the merits, the harmless error inquiry requires an examination of whether alternative, valid predicates grounded Granda‘s conviction. In Gomez, this Court expressly noted that it was for the district court to consider the merits of Gomez‘s petition -- so it left open the possibility that the district court would apply Brecht on the merits. Id. at 1228; see also id. at 1229 (Carnes, C.J., concurring) (“The district court will determine, in the first instance, which crimes constitute the basis for Gomez‘s
We, therefore, hold, in addition to our conclusion that procedural default bars Granda‘s claim, that Granda did not suffer harm from the erroneous jury instruction.
Accordingly, we AFFIRM.
JORDAN, Circuit Judge, concurring in part and concurring in the judgment.
I join Parts I, II, III.A, and III.C of the court‘s opinion and concur in the judgment. Because Mr. Granda cannot prevail on the merits, I would not address the issue of procedural default and therefore do not join Part III.B.