United States Court of Appeals
                            For the Eighth Circuit
                        ___________________________

                                No. 16-1680
                        ___________________________

                                   Roberto Barajas

                       lllllllllllllllllllllPetitioner - Appellant

                                           v.

                             United States of America

                      lllllllllllllllllllllRespondent - Appellee
                                      ____________

                    Appeal from United States District Court
                  for the Northern District of Iowa - Sioux City
                                 ____________

                          Submitted: September 22, 2017
                             Filed: December 5, 2017
                                  ____________

Before SMITH, Chief Judge, WOLLMAN and GRUENDER, Circuit Judges.
                             ____________

GRUENDER, Circuit Judge.

       Roberto Barajas appeals the denial of his 28 U.S.C. § 2255 petition alleging
ineffective assistance of counsel. Because the district court1 correctly found that



      1
        The Honorable Leonard T. Strand, Chief Judge, United States District Court
for the Northern District of Iowa.
Teague v. Lane, 

(1989), bars the application of the Supreme Court’s
decision in Padilla v. Kentucky, 

(2010), we affirm.

                                          I.

       In 2009, Barajas pleaded guilty to knowingly possessing a stolen firearm in
violation of 18 U.S.C. § 922(j), an offense categorized as an “aggravated felony”
under federal immigration law. 8 U.S.C. § 1101(a)(43)(E)(ii). Federal law provides
that “[a]ny alien who is convicted of an aggravated felony at any time after admission
is deportable.” 

Because Barajas is a citizen of Mexico and
otherwise satisfied § 1227(a)(2)(A)(iii), he was deported after serving his four-month
term of imprisonment.

       In September 2010, Barajas filed a 28 U.S.C. § 2255 petition.2 He claims that
his trial counsel provided ineffective assistance by failing to inform him of the
deportation consequences of his guilty plea. At the time Barajas’s conviction became
final, and even after his sentence ended, it was unclear if an attorney’s failure to

      2
        The Supreme Court explained in Chaidez v. United States that a petition for
a writ of coram nobis is the proper method “to collaterally attack a criminal
conviction for a person . . . who is no longer ‘in custody’ and therefore cannot seek
habeas relief under § 2255 or § 2241.” 

, 345 n.1 (2013). Barajas styled
his petition as a request for relief under § 2255, believing he was still “in custody”
because he had not yet completed his term of supervised release and because less than
one year had elapsed since his sentence. Alternatively, he requested a writ of error
coram nobis. The Government did not contest the § 2255 label, and the case has
proceeded as though Barajas filed a valid habeas petition. Because we have held that
“a federal District Court is not bound by the label attached to the remedy pursued,”
see Burns v. United States, 

, 896 (8th Cir. 1963), and because the parties
do not argue to the contrary, we assume, as the Supreme Court did in Chaidez, “that
nothing in this case turns on the difference between a coram nobis petition and a
habeas 

n.1.


                                         -2-
inform a client of collateral consequences like deportation could constitute ineffective
assistance of counsel. In Padilla, however, the Supreme Court held that criminal
defense attorneys have a duty to inform clients about the possible immigration
consequences of pleading guilty in order to provide effective counsel under the Sixth

. Barajas relied on Padilla in his § 2255 petition. In
2012, citing Padilla, the district court granted Barajas’s § 2255 petition and found
that his counsel was ineffective for failing to inform him about the immigration
consequences of pleading guilty. The Government timely appealed.

       In 2013, while the appeal was pending, the Supreme Court decided Chaidez,
which held that Padilla announced a “new rule” under the Teague standard and thus
cannot be applied retroactively to cases on collateral review. 

.
In light of that decision, we reversed the district court’s order and remanded for
further consideration. On remand, the district court denied Barajas’s § 2255 petition,
finding that Teague bars the application of Padilla’s new rule in this case.3 Barajas
now appeals the district court’s decision, arguing that Teague’s bar on retroactive
application of new rules does not apply to collateral review of federal convictions,
especially in the context of ineffective assistance of counsel claims.




       3
        In addition to denying the petition because trial counsel’s failure to inform did
not violate the law at the time, the district court also found that Barajas’s trial counsel
did not affirmatively misrepresent the consequences of his guilty plea—a finding that
Barajas attempts to challenge on appeal. However, the certificate of appealability
issued by the district court does not encompass this issue. Therefore, we do not
address whether counsel affirmatively misrepresented the consequences of the plea.
See Harris v. Bowersox, 

, 748 (8th Cir. 1999) (“[W]e limit our appellate
review to the issues specified in the certificate of appealability.”).

                                           -3-
                                          II.

      We review determinations of 28 U.S.C. § 2255 relief de novo and any
underlying findings of fact for clear error. Pierce v. United States, 

, 531
(8th Cir. 2012).

       In Teague, the Supreme Court adopted a uniform rule to “clarify how the
question of retroactivity should be resolved for cases on collateral 

. It held that “new rules” of criminal procedure “should not be applied
retroactively to criminal cases on collateral review.” 

A case announces
a new rule “when it breaks new ground[,] imposes a new obligation on the States or
the Federal Government,” or when “the result [is] not dictated by precedent.” 

The Teague Court grounded its decision to bar retroactive application of new
rules in systemic concerns about ensuring the finality of criminal convictions and
respecting principles of comity and federalism. 

       When the Supreme Court held in Padilla that “advice regarding deportation is
not categorically removed from the ambit of the Sixth Amendment right to 

, and thus that Strickland v. Washington’s standard of “reasonable
professional assistance” applies to such advice, 

it did not address
whether it was announcing a “new rule” for Teague purposes. Three years later, in
Chaidez, the Court clarified that Padilla indeed announced a new 

. The Court noted that, although it had applied Strickland “in diverse contexts
without ever suggesting that doing so required a new rule,” 

Padilla was not
simply an application of Strickland, 

Indeed, “had Padilla merely made
clear that a lawyer who neglects to inform a client about the risk of deportation is

      4
        The Court identified two exceptions to this general bar: substantive rules
related to the authority to proscribe specific conduct and watershed rules of criminal
procedure. See 

. This case does not implicate either
exception.

                                         -4-
professionally incompetent,” it would not have constituted a new rule. 

to asking how the Strickland test applied (‘Did this attorney act
unreasonably?’), Padilla asked whether the Strickland test applied (‘Should we even
evaluate if this attorney acted unreasonably?’).” 

The answer to “that
preliminary question about Strickland’s ambit . . . required a new rule,” which cannot
be applied retroactively. 

this conclusion, the Chaidez Court explained that “defendants whose
convictions became final prior to Padilla . . . cannot benefit from its holding.” 

Although at first blush this statement would seem to resolve the present appeal,
the scope of this general proposition is somewhat uncertain. In a footnote, Chaidez
explicitly declined to address whether Teague applies when a petitioner challenges
a federal conviction, especially when such a challenge is based on a claim of
ineffective assistance of counsel. 

n.16 (explaining that those arguments
were not raised properly below). Barajas’s appeal argues that Teague should not
apply in such cases. He suggests that federal cases do not involve Teague’s interest
in comity and that claims for ineffective assistance of counsel do not raise Teague’s
concern about the finality of criminal judgments. We address the issues in turn.

                                         A.
       Teague involved collateral review of a state conviction under 28 U.S.C. § 2254.
Although the Supreme Court has assumed that Teague’s bar on retroactivity applies
to collateral review of federal convictions under § 2255, see, e.g., Welch v. United
States, 

, 1264 (2016), it has not decided the question. Nor has this
court.5 Every circuit to have addressed the question directly has concluded that


      5
       We have considered the Teague doctrine when reviewing § 2255 petitions
without addressing the question. See Sun Bear v. United States, 

, 703-04
(8th Cir. 2011); see also Abraham v. United States, 

, 1052-53 (8th Cir.
2012).

                                          -5-
Teague applies to both federal and state convictions. See, e.g., United States v.
Sanchez-Cervantes, 

, 667 (9th Cir. 2002) (explaining, consistent with
the findings of the Second, Fourth, Seventh, and Tenth Circuits, that Teague applies
to § 2255 cases).6 Today, we join those circuits.

       Barajas contends that all courts to have considered the issue “misread the intent
of the Supreme Court in Teague.” Though he acknowledges the importance of
finality, Barajas argues that comity is a second, necessary “pillar” supporting
Teague’s limit on retroactivity. Indeed, Teague champions the importance of
preventing “intrusive” incursions into state sovereignty and avoiding the “costs
imposed upon the State[s] by retroactive application of new rules of constitutional
law on habeas corpus.” 

On many occasions, the Supreme Court has
reiterated the gravity of this interest. See, e.g., Lockhart v. Fretwell, 

,
372 (1993) (explaining that Teague “was motivated by a respect for the States’ strong
interest in the finality of criminal convictions, and the recognition that a State should
not be penalized for relying on the constitutional standards that prevailed at the time
the original proceedings took place” (internal quotation marks omitted)). Thus,
Barajas contends that because his challenge to his federal conviction does not
implicate comity or federalism, the Teague bar should not apply.

       We reject this argument for three reasons. First, while Teague emphasized the
particularly egregious consequences of upsetting the finality of state convictions, the
Supreme Court also, and independently, rooted the decision in the importance of
finality more generally. 

. The Court explained that “[a]pplication
of constitutional rules not in existence at the time a conviction became final seriously
undermines the principle of finality which is essential to the operation of our criminal


      6
       That said, some courts have begun to express doubt. See Reina-Rodriguez v.
United States, 

, 1190 (9th Cir. 2011); Duncan v. United States, 

, 444 n.2 (6th Cir. 2009).

                                          -6-
justice system.” 

the criminal law is deprived of much of its
deterrent effect.” 

concerns are conceptually distinct from concerns
about federalism and comity. Second, and relatedly, Teague explicitly “adopt[ed]
Justice Harlan’s approach to retroactivity for cases on collateral review.” 

Years before the Teague decision, Justice Harlan repeatedly urged the Court to
recognize that “it is sounder, in adjudicating habeas petitions, generally to apply the
law prevailing at the time the conviction became final than it is to seek to dispose of
all these cases on the basis of intervening changes in constitutional interpretation.”
See Mackey v. United States, 

, 689 (1971) (Harlan, J., concurring in the
judgment); see also Desist v. United States, 

, 262-63 (1969) (Harlan, J.,
dissenting). Notably, he raised these concerns in cases involving collateral review
of federal convictions. Third, the Supreme Court has explained in a similar context
that it sees “no basis for affording federal prisoners a preferred status when they seek
postconviction relief” because “the Federal Government, no less than the States, has
an interest in the finality of its criminal judgments.” United States v. Frady, 

, 166 (1982) (rejecting an argument that post-conviction rules promoting finality
should receive less weight in the federal context simply because “considerations of
comity” are not present).

       Given the importance of protecting the finality of criminal convictions, we join
our sister circuits and hold that the Teague limit on retroactivity applies to collateral
review of both state and federal convictions.

                                            B.
       We next consider Barajas’s argument that even if finality provides an
independently sufficient basis for applying Teague to federal convictions, we need
not apply Teague in the ineffective assistance context because “Strickland already
protects finality interests” in such cases. Barajas points out that Strickland claims for
ineffective assistance of counsel are unique because they “are almost always brought
on collateral review, which means they almost always involve a final conviction and

                                          -7-
thus implicate finality.” Therefore, he suggests, Strickland’s high bar must serve as
sufficient protection to the finality of convictions in the ineffective assistance context.

       More precisely, Strickland functions as an exception to the general, finality-
promoting rule that “claims not raised on direct appeal may not be raised on collateral
review.” Massaro v. United States, 

, 504-05 (2003). Petitioners may,
and in many cases must, wait to raise ineffective assistance claims for the first time
on collateral review. See 

This rule reflects a willingness to upset
finality by granting post-conviction relief but only if a petitioner can satisfy
Strickland’s rigorous standard for proving ineffective assistance. See 

,
687 (1984) (laying out the two-pronged test for proving ineffective assistance of
counsel, which requires a showing of deficient performance and prejudice); see also
Kimmelman v. Morrison, 

, 381 (1986) (emphasizing the “rigorous
standard which Strickland erected for ineffective-assistance claims”). In other words,
although successful Strickland claims actually undermine the finality of criminal
convictions, we tolerate this narrow exception in part because Strickland’s rigorous
standard sufficiently “reflects the profound importance of finality in criminal
proceedings.” See 

. Thus, Barajas argues, because
generally we rely only on Strickland’s rigorous standard to protect the finality of
convictions against claims for ineffective assistance of counsel—and, in the federal
context, finality of convictions is the only interest Teague serves—relying on Teague
to achieve the same purpose would be unnecessary and inappropriate. We disagree.

       Though both Teague’s bar and Strickland’s rigorous standard functionally
protect the finality of convictions, it does not follow that Strickland renders Teague
superfluous. This becomes clear when considering Strickland in context as a limited
exception to the finality-promoting doctrine of procedural default. In this sense, the
proper inquiry is not about the extent to which Strickland is able to protect finality
but rather about the reason we allow Strickland’s qualified protection in the first
place. The Strickland exception to procedural default derives from context-specific

                                           -8-
rationales that do not justify creating a similar exception to Teague’s bar on the
retroactive application of new rules of criminal procedure.

       The doctrine of procedural default prevents criminal defendants from reviving
forfeited claims without a showing of cause and prejudice in order to “conserve
judicial resources and to respect the law’s important interest in the finality of
judgments.” 

. It also promotes the timely resolution of
constitutional issues. See id.; see also Wainwright v. Sykes, 

, 89 (1977)
(suggesting that the doctrine prevents “sandbagging” and encourages the earliest
possible resolution of constitutional issues). In Massaro v. United States, however,
the Supreme Court found that requiring a criminal defendant to bring ineffective
assistance claims on direct appeal does not promote the objectives of the procedural-
default doctrine. 

. This is because, practically speaking, it is
difficult to identify and evaluate counsel’s relevant conduct on direct review, and it
is often impossible to determine prejudice prior to a final judgment. See 

06. Thus, defendants are permitted to raise ineffective assistance claims for the first
time on collateral review, leaving Strickland’s high threshold to guard finality in this
context. 

(explaining that the standard for proving ineffective
assistance of counsel reflects concerns about the finality of criminal convictions).

       It would be inappropriate to extend the exception created for the procedural-
default context to the Teague context because of the different purposes underlying the
two doctrines. Teague’s bar on the retroactive application of new rules prevents
convictions from being upended by every subsequent change in the law. 

. Unlike in the procedural-default context, ineffective assistance claims are
not uniquely situated when it comes to achieving this goal. In other words, there is
a difference between (1) allowing an exception to the finality of a decision for
criminal defendants to raise an attorney’s deficient performance, the grounds for
which were unknown or unreviewable on direct review, and (2) allowing an
exception for criminal defendants to claim an attorney’s performance was deficient

                                          -9-
even though the attorney complied fully with the standards of performance in
existence at the time. Cf. Toledo v. United States, 

, 681 (8th Cir. 2009)
(“We do not evaluate counsel’s performance using the clarity of hindsight, but in light
of the facts and circumstances at the time of trial.” (internal quotation marks
omitted)). Teague protects convictions that faithfully conform to existing
constitutional law by fixing in time the procedural standards to which they are held.

. That objective provides no basis for treating ineffective
assistance claims differently. See In re Ifenatuora, 528 F. App’x 333, 335 n.3 (4th
Cir. 2013) (unpublished) (declining to create an ineffective assistance exception to
Teague “given Teague’s emphasis on ensuring that retroactivity principles would not
vary from rule to rule”).

       Finally, Barajas argues that applying the Teague bar in this case serves to
“undermine Massaro’s collateral review regime” by encouraging defendants to bring
their ineffective assistance claims on direct review. We do not share this concern.
The Supreme Court explained in Chaidez that “virtually all” ineffective assistance
claims can be raised on collateral review without implicating 

. Thus, even assuming that applying Teague here would have some marginal
effect on incentives to bring ineffective assistance claims on direct review, that effect
would not warrant disregarding Teague in this context.

       We hold that Teague’s bar applies to federal petitioners raising ineffective
assistance of counsel claims.

                                          III.

      For the foregoing reasons, we affirm the denial of Barajas’s § 2255 petition.
                      ______________________________




                                          -10-