FILED
                                                            United States Court of Appeals
                                                                    Tenth Circuit

                                                                 September 1, 2011
                                     PUBLISH                     Elisabeth A. Shumaker
                                                                     Clerk of Court
                    UNITED STATES COURT OF APPEALS

                           FOR THE TENTH CIRCUIT


 UNITED STATES OF AMERICA,

       Plaintiff - Appellee,

 v.                                                       No. 10-6294

 CHANG HONG,

       Defendant - Appellant.



                                      ORDER



Before O’BRIEN, McKAY, and TYMKOVICH, Circuit Judges.


      This matter is before the court to amend, sua sponte, the Opinion issued

originally in this appeal on August 30, 2011. The amendment is limited to adding

a citation to Chaidez v. United States, ___ F.3d ___, 

(7th Cir.

Aug. 23, 2011). The clerk is directed to file the amended decision, which is

attached to this order, nunc pro tunc to the original filing date.


                                        Entered for the Court,




                                        ELISABETH A. SHUMAKER
                                        Clerk of Court
                                                                      FILED
                                                          United States Court of Appeals
                                                                  Tenth Circuit

                                                                August 30, 2011
                                      PUBLISH                 Elisabeth A. Shumaker
                                                                  Clerk of Court
                    UNITED STATES COURT OF APPEALS

                                 TENTH CIRCUIT



 UNITED STATES OF AMERICA,

               Plaintiff-Appellee,
          v.                                            No. 10-6294
 CHANG HONG,

               Defendant-Appellant.


           APPEAL FROM THE UNITED STATES DISTRICT COURT
              FOR THE WESTERN DISTRICT OF OKLAHOMA
                        (D.C. NO. CV-10-00978)


Submitted on the brief *:

Joan L. Lopez, Oklahoma City, Oklahoma, for Appellant.

Jonathon E. Boatman, Assistant United States Attorney, Office of the United
States Attorney, Oklahoma City, Oklahoma, for Appellee.



Before O’BRIEN, McKAY, and TYMKOVICH, Circuit Judges.




      *
         No brief was filed on behalf of the Plaintiff-Appellee pursuant to 10th
Cir. R. 22.1(B). After examining the brief and the appellate record, this three-
judge panel has determined unanimously that oral argument would not be of
material assistance in the determination of this appeal. See Fed. R. App. P. 34(a);
10th Cir. R. 34.1(G). The cause is therefore ordered submitted without oral
argument.
TYMKOVICH, Circuit Judge.



      Chang Hong seeks to appeal the district court’s denial of his motion for

relief under 28 U.S.C. § 2255 as untimely. He asserted claims of ineffective

assistance of counsel, alleging his counsel failed to advise him of the immigration

consequences of his guilty plea as required by Padilla v. Kentucky, 

(2010). Hong argues Padilla is a new rule of constitutional law that applies

retroactively to cases on collateral review, making his § 2255 motion timely. We

construe Hong’s notice of appeal and opening brief as a request for a certificate

of appealability (COA) to appeal the district court’s order.

      Exercising jurisdiction under 28 U.S.C. §§ 1291 and 2253(a), we find

Padilla is a new rule of constitutional law, but it does not apply retroactively to

cases on collateral review. Therefore, Hong’s § 2255 motion was untimely, and

we conclude Hong has not made a substantial showing of the denial of a

constitutional right.

      Accordingly, we DENY Hong’s request for a COA and DISMISS his

appeal.

                                  I. Background

      Hong is a citizen of South Korea and was a permanent legal resident of the

United States. In September 2007, he pleaded guilty to one count of conspiracy

to possess with intent to distribute and to distribute ecstasy, marijuana, and


                                          2
hydro-marijuana. In February 2008, he was sentenced to 37 months’

imprisonment and did not file a direct appeal.

       In August 2010, while in federal prison, Hong received a Notice to Appear

from the United States Department of Homeland Security, which stated Hong was

being placed in immigration removal proceedings. The notice asserted Hong was

subject to removal from the United States because of his drug conspiracy

conviction. See 8 U.S.C. § 1227(a)(2)(A)(iii) (“Any alien who is convicted of an

aggravated felony at any time after admission is deportable.”), (a)(2)(B)(i) (“Any

alien who at any time after admission has been convicted of a violation of (or a

conspiracy or attempt to violate) any law or regulation of . . . the United States

. . . relating to a controlled substance . . . is deportable.”).

       In September 2010, Hong filed a motion under 28 U.S.C. § 2255 seeking to

vacate his conviction and sentence as well as to withdraw his guilty plea on the

grounds of ineffective assistance of counsel. 2 Hong alleged his counsel failed to

advise him of the immigration consequences of his guilty plea. In support, he

cited Padilla v. Kentucky, 

(2010), wherein the Supreme Court

held that “before a non-citizen criminal defendant enters a guilty plea, his counsel

has a duty under the Sixth Amendment to inform him ‘whether his plea carries a


       2
         Hong also requested, in the alternative, that the court grant a writ of
coram nobis. The district court denied this alternative request, finding Hong was
ineligible for such relief because he was still in custody. See United States v.
Torres, 

, 1245 (10th Cir. 2002). Hong does not challenge this
decision on appeal, and we do not address it.

                                             3
risk of deportation.’” Waugh v. Holder, 

, 1283 (10th Cir. 2011)

(quoting 

).

      On December 15, 2010, the district court denied Hong’s § 2255 motion as

untimely because (1) it was filed outside the one-year statute of limitations period

under § 2255(f)(1), and (2) Padilla was not a new rule of constitutional law and

did not apply retroactively to cases on collateral review, so § 2255(f)(3) did not

provide the correct starting date for the statute of limitations. This appeal

followed.

      Meanwhile, on December 28, 2010, Hong was released from federal prison

after completing his sentence for drug conspiracy. 3 Due to an immigration

detainer, Hong was transferred to the custody of United States Immigration and

Customs Enforcement upon his release. He later appeared at an immigration

removal hearing, and a removal order was entered against him on April 26, 2011.

Hong did not appeal the removal order, and on June 28, 2011, he was removed

from the United States.


      3
         Although Hong has completed his sentence and has been deported, his
habeas petition challenging his conviction is not moot. See Prost v. Anderson,

, 582 n.3 (10th Cir. 2011) (“The Supreme Court has told us that a
habeas petition challenging a conviction isn’t mooted by a prisoner’s release from
incarceration because the Court is ‘willing to presume’ that the fact of conviction
‘has continuing collateral consequences.’”) (quoting Spencer v. Kemna, 

, 7–8 (1998)). Hong’s drug conspiracy conviction renders him ineligible to
receive a visa or for admission to the United States. See 8 U.S.C.
§ 1182(a)(2)(A)(I) (“[A]ny alien convicted of . . . a violation of (or a conspiracy
or attempt to violate) any law . . . [of] the United States . . relating to a controlled
substance . . . is inadmissible.”).

                                           4
                                   II. Discussion

      A defendant may not appeal the denial of a § 2255 motion unless we first

issue a COA. 28 U.S.C. § 2253(c)(1)(B). We will issue a COA only when the

defendant “has made a substantial showing of the denial of a constitutional right.”

To meet this burden, Hong must show “that reasonable jurists

could debate whether (or, for that matter, agree that) the petition should have

been resolved in a different manner or that the issues presented were adequate to

deserve encouragement to proceed further.” Allen v. Zavaras, 

,

1199 (10th Cir. 2009) (quoting Slack v. McDaniel, 

, 484 (2000)).

Hong fails to make this showing.

      A. Section 2255 Motions and Statute of Limitations

      A defendant must file a § 2255 motion within one year of the date his

conviction became final. See § 2255(f)(1). Hong’s conviction became final on

February 29, 2008, but his § 2255 motion was filed over two years later, on

September 10, 2010. Therefore Hong’s § 2255 motion was untimely under

§ 2255(f)(1).

      Nonetheless, Hong argues his petition was timely under § 2255(f)(3).

Under that provision, the one-year limitations period to file a § 2255 motion runs

from “the date on which the right asserted was initially recognized by the

Supreme Court, if that right has been newly recognized by the Supreme Court and

made retroactively applicable to cases on collateral review.” 

                                         5
Hong contends the Supreme Court’s decision in Padilla created a new rule that

applies retroactively for the purposes of § 2255(f)(3). Under his theory, his

§ 2255 motion is timely because Padilla was decided on March 31, 2010, and his

§ 2255 motion was filed within one year of that decision, on September 10, 2010.

      Hong is incorrect. Although Padilla establishes a new rule of

constitutional law, under the Supreme Court’s rubric for determining retroactivity

established in Teague v. Lane, 

(1989), 4 Padilla does not apply

retroactively to cases on collateral review. Therefore, § 2255(f)(3) does not



      4
         Although Teague arose in the context of a § 2254 petition and Hong’s
motion arises under § 2255, in accord with other circuits, we have held that
“Teague’s nonretroactivity doctrine applies equally to habeas petitions brought
under sections 2254 and 2255.” Daniels v. United States, 

, 1194
(10th Cir. 2001) (en banc). While clearly bound by our prior en banc decision,
we note the Supreme Court has never applied Teague to a § 2255 petition. See
Danforth v. Minnesota, 

, 269 n.4, 281 n.16 (2008) (noting the
opinion does not consider whether Teague applies to cases under § 2255 but that
lower federal courts have applied Teague to § 2255 motions and that much of the
reasoning applicable to petitions under § 2254 is “equally applicable” to § 2255
motions); cf. Reina-Rodriguez v. United States, No. 08-16676, – F.3d –, 

, at *5 (9th Cir. June 22, 2011) (“[T]here is some doubt under current
Supreme Court jurisprudence whether Teague applies to federal prisoners . . .
who seek federal habeas relief.”); Duncan v. United States, 

, 444 n.2
(6th Cir. 2009) (“It is not entirely clear that Teague’s framework is appropriate
for federal habeas petitions under 18 U.S.C. § 2255 because many of the comity
and federalism concerns animating Teague are lacking.”); see also Valentine v.
United States, 

, 341–45 (6th Cir. 2007) (Martin, J., concurring in
part and dissenting in part) (questioning the applicability of Teague in federal
habeas cases); J. Thomas Sullivan, Danforth, Retroactivity, and Federalism, 61
O KLA . L. R EV . 425, 463 (2008) (“The . . . question deliberately left unanswered
by the Danforth Court involves the application of the Teague retroactivity
doctrine to post-conviction actions brought by federal defendants pursuant to 28
U.S.C. § 2255.”).

                                         6
apply, and Hong’s § 2255 motion was untimely under § 2255(f)(1).

      B. Retroactivity

      In Teague, and subsequent cases interpreting its analysis, the Supreme

Court constructed and refined a three-step process to determine retroactivity. We

apply it here to decide whether Padilla announced a new rule of constitutional

law and whether it applies retroactively to Hong’s conviction. Whorton v.

Bockting, 

, 416 (2007). We first must determine whether Hong’s

conviction was final before the Supreme Court’s decision in Padilla. 5 Beard v.

Banks, 

, 411 (2004).

      Second, we assess whether the rule in Padilla is actually “new,” based on

whether a “court considering [Hong]’s claim at the time his conviction became

final would have felt compelled by existing precedent to conclude that the rule

[announced in Padilla] was required by the Constitution.” O’Dell v. Netherland,

, 156 (1997).

      Third, if we determine the rule in Padilla is new, we will apply it to Hong’s

conviction only if that new rule “falls within either of the two narrow exceptions

to nonretroactivity.” 

; see 

(“Unless they fall within an exception to the general rule, new constitutional rules


      5
         A conviction and sentence becomes final for the purposes of Teague
“when the availability of a direct appeal has been exhausted and the time for
filing a petition for certiorari with the Supreme Court has elapsed or the Court has
denied a timely petition for certiorari.” United States v. Dago, 

,
1243 (10th Cir. 2006) (citing Caspari v. Bohlen, 

, 390 (1994)).

                                         7
of criminal procedure will not be applicable to those cases which have become

final before the new rules are announced.”). We discuss the exceptions in more

detail below.

      1. Final Conviction

      Because Hong declined to file a direct appeal, his conviction became final

14 days after the district court entered judgment on February 29, 2008. See Fed.

R. App. P. 4(b)(1)(A)(I) (giving a defendant 14 days after entry of judgment to

file a notice of appeal). Padilla was decided more than two years later, on March

31, 2010. Hong’s conviction was final before the Supreme Court’s decision in

Padilla. Therefore, we are left to determine whether Padilla announced a new

rule of constitutional law and, if so, whether it falls within either of the Teague

exceptions to the retroactivity bar.

      2. New Rule

      Step two of the Teague analysis requires us to assess whether Padilla

represents a “new rule” of constitutional law. While a closer question, we

conclude Padilla is a new rule of constitutional law because it was not compelled

by existing precedent at the time Hong’s conviction became final.

      a. Padilla

      In Padilla, the Court considered whether defense counsel has an obligation

to advise his client that a guilty plea would make him subject to automatic

deportation. Jose Padilla had pleaded guilty to drug trafficking in Kentucky state


                                           8
court, and as a lawful permanent resident of the United States, he was subject to

virtually mandatory removal because of his drug conviction.

      Padilla sought state post-conviction relief from his guilty plea, alleging

ineffective assistance of counsel. He argued he entered his guilty plea in reliance

on his counsel’s erroneous advice that the plea would not affect Padilla’s

immigration status. Ultimately, the Kentucky Supreme Court denied Padilla post-

conviction relief and held the Sixth Amendment’s guarantee of effective

assistance of counsel did not protect him from erroneous advice regarding

collateral consequences of a conviction, such as deportation or removal. 6

      The Supreme Court reversed and remanded the case. It found

“constitutionally competent counsel would have advised [Padilla] that his

conviction for drug distribution made him subject to automatic deportation.”

. Before addressing Padilla’s claims, the Court

surveyed the development of federal immigration law and noted “deportation or

removal is now virtually inevitable for a vast number of noncitizens convicted of

crimes.” 

Because of this connection between conviction

and removal, “deportation is an integral part—indeed, sometimes the most

important part—of the penalty that may be imposed on noncitizen defendants who

plead guilty to specified crimes.” 

Therefore, the Court


      6
        With changes in immigration law, came changes in nomenclature from
“deportation” to “removal.” See 

n.6. For our
purposes, we treat the terms as interchangeable.

                                          9
determined the “importance of accurate legal advice for noncitizens accused of

crimes has never been more important.” 

      The Court turned to Padilla’s claims and considered whether his counsel

rendered effective assistance as required by Strickland v. Washington, 

(1984). Before applying Strickland’s familiar two-part test of deficient

performance and prejudice, the Court first clarified that it had “never applied a

distinction between direct and collateral consequences to define the scope of

constitutionally ‘reasonable professional assistance’ required under Strickland.”

. The Kentucky Supreme Court had rejected Padilla’s

arguments because he claimed ineffective assistance with regard to collateral

consequences of his plea—deportation—that the Kentucky Supreme Court ruled

were outside the requirements of the Sixth Amendment. 7 The Supreme Court in

Padilla stated the “collateral versus direct distinction” was not useful when

considering a Strickland claim regarding the risk of deportation. 

Even though deportation proceedings are civil proceedings, based on the “unique

nature of deportation,” they were “nevertheless intimately related to the criminal

process.” 

This “close connection to the criminal process” made



      7
         In accord with many other state and federal courts, before Padilla we
also held that “deportation remains a collateral consequence of a criminal
conviction, and counsel’s failure to advise a criminal defendant of its possibility
does not result in a Sixth Amendment deprivation.” Broomes v. Ashcroft, 

, 1257 (10th Cir. 2004) abrogated by Padilla v. Kentucky, 

, 1481 n.9 (2010).

                                         10
deportation “uniquely difficult” to classify as either a direct or collateral

consequence. 

Rather than force deportation into one category, the

Court concluded that Strickland applied to Padilla’s claims because “advice

regarding deportation is not categorically removed from the ambit of the Sixth

Amendment right to counsel.” 

Strickland applied, the Court then analyzed whether

Padilla’s counsel rendered assistance that fell below an objective standard of

reasonableness. Prevailing professional norms long required defense counsel to

advise their noncitizen clients of the risk of deportation, and with this, the Court

stated it was “not a hard case in which to find deficiency.” 

Padilla’s

counsel could have consulted the removal statutes and easily determined Padilla’s

guilty plea would make his removal virtually mandatory. Instead, his counsel

gave Padilla the false assurance that his plea would not affect his immigration

status.

          While immigration consequences may have been clear in Padilla’s case, the

Court acknowledged immigration law is a complex subject matter. And in

situations where the deportation consequences are unclear, defense counsel would

still have a duty to advise a noncitizen client, but only the limited duty to advise

him that pending criminal charges may have negative immigration consequences.

But, as in Padilla’s case, “when the deportation consequence is truly clear . . . the

duty to give correct advice is equally clear.” 

sum, the Supreme Court in Padilla held that the seriousness and severity

of deportation as a consequence of a guilty plea makes it critical that defense

counsel “inform her client whether his plea carries a risk of deportation.” 

Having distilled the Supreme Court’s holding in Padilla, we turn to our

Teague analysis and consider whether Padilla is a new rule that retroactively

applies to cases on collateral review.

      b. New vs. Old Rule

      When the Supreme Court announces a rule, that rule’s effect on a

defendant’s conviction will differ based on whether the rule is “new” or “old” and

whether his case is pending on direct or collateral review. A “new” rule will

generally apply only to criminal cases pending on direct review, while an “old”

rule will apply to cases on both direct and collateral review. 

(citing Griffith v. Kentucky, 

, 328 (1987)). The key question

is whether a rule is old or new.

      The Supreme Court admits it has “stated variously the formula for

determining when a rule is new.” 

; see also United States

v. Christensen, 

, 1207 (10th Cir. 2006) (“The term new rule is

somewhat imprecise.”). 8 A rule is new “within the meaning of Teague if it

‘breaks new ground,’ ‘imposes a new obligation on the States or the Federal


      8
         In some sense, every rule announced in a Supreme Court decision could
be considered “new” if the Court had not previously ruled on the specific issue.
But for the purposes of a Teague analysis, our inquiry is more limited.

                                         12
Government,’ or was not ‘dictated by precedent existing at the time the

defendant’s conviction became final.’” Graham v. Collins, 

, 467

(1993) (quoting 

). Conversely, a rule is old if a “court

considering the defendant’s claim at the time his conviction became final would

have felt compelled by existing precedent to conclude that the rule he seeks was

required by the Constitution.” 

(quotation and brackets

omitted).

      When making this judgment we consider whether “reasonable jurists could

have differed as to whether” a rule was compelled or dictated by existing

precedent. 

. “The new rule principle . . . validates

reasonable, good-faith interpretations of existing precedents, . . . even if those

good-faith interpretations are . . . contrary to later decisions.” United States v.

Price, 

, 847 (10th Cir. 2005) (quoting 

).

The inquiry recognizes that reasonable jurists may differ on the import of a

Supreme Court decision.

      “[T]he fact that a court says that its decision is within the ‘logical compass’

of an earlier decision, or indeed that it is ‘controlled’ by a prior decision, is not

conclusive for purposes of deciding whether the current decision is a ‘new rule’

under Teague.” Butler v. McKellar, 

, 415 (1990). “While there can

be no dispute that a decision announces a new rule if it expressly overrules a prior

decision, ‘it is more difficult . . . to determine whether [the Court] announce[s] a


                                           13
new rule when a decision extends the reasoning of [its] prior cases.’” 

(quoting Saffle v. Parks, 

, 488 (1990)).

      With these principles in mind, we turn to Padilla. The starting point is

whether Padilla represents a new rule or merely represents an application of the

venerable Strickland rule to new facts. The central holding of Padilla is that

defense counsel “must inform her client whether his [guilty] plea carries a risk of

deportation” if those consequences are 

. Even if the

consequences are not clear, defense counsel must still “advise a noncitizen client

that pending criminal charges may carry a risk of adverse immigration

consequences.” 

The majority’s opinion focused on whether Padilla’s

counsel provided representation that fell below an objective standard of

reasonableness and citations to Strickland run throughout. Without doubt, Padilla

is a Strickland case.

      While grounded in Strickland, we still conclude Padilla is a new rule of

constitutional law. Before Padilla, most state and federal courts had considered

the failure to advise a client of potential collateral consequences of a conviction

to be outside the requirements of the Sixth Amendment. See 

n.9 (collecting cases); see also Gabriel J. Chin & Richard W. Holmes, Jr.,

Effective Assistance of Counsel and the Consequences of Guilty Pleas, 87

C ORNELL L. R EV . 697, 699 (2002) (“[E]leven federal circuits, more than thirty

states, and the District of Columbia have held that lawyers need not explain


                                          14
collateral consequences [under the Sixth Amendment].”). All of these

courts—including our own—thought the rule in Padilla was not dictated or

compelled by Court precedent. It goes without saying these are some of the

“reasonable jurists” we must survey to determine if Padilla is a new rule.

      In addition to this precedent, when assessing whether a rule is new, we

have also considered the concurring or dissenting views of other Justices in the

case. For example, in 

, we examined whether

Shepard v. United States, 

(2005), announced a new rule. We looked

to Justice O’Connor’s dissent—which stated the Court’s rule was not compelled

by statute or Court precedent—and found it “dispositive” of whether Shepard

announced a new rule of constitutional law. 

. We

apply the same logic here for further guidance.

      Padilla, a 7-2 decision, generated both a strong concurrence and dissent.

In a concurrence, Justice Alito (joined by Chief Justice Roberts) stated “the

Court’s decision marks a major upheaval in Sixth Amendment law” and noted the

majority failed to cite any precedent for the premise that a defense counsel’s

failure to provide advice concerning the immigration consequences of a criminal

conviction violated a defendant’s right to counsel. 

(Alito, J., concurring in judgment); see also 

(noting the majority’s

“dramatic departure from precedent”); 

(“[T]he Court’s view has been

rejected by every Federal Court of Appeals to have considered the issue thus


                                         15
far.”); 

(“The majority seeks to downplay its dramatic expansion of the

scope of criminal defense counsel’s duties under the Sixth Amendment.”).

      Similarly, Justice Scalia in a dissent (joined by Justice Thomas), argued the

Sixth Amendment right to counsel does not extend to “advice about the collateral

consequences of conviction” and that the Court, until Padilla, had limited the

Sixth Amendment to advice directly related to defense against criminal

prosecutions. 

(Scalia, J., dissenting); see also 

(“There

is no basis in text or in principle to extend the constitutionally required advice

regarding guilty pleas beyond those matters germane to the criminal prosecution

at hand.”). We take the concurrence and dissent as support for our conclusion

that reasonable jurists did not find the rule in Padilla compelled or dictated by the

Court’s prior precedent.

      We acknowledge those counter-arguments that Padilla is not a new

rule—that is, why a court would have felt compelled to apply Strickland to

deportation consequences of a guilty plea. For example, Padilla did not overturn

any of the Court’s prior precedent and is grounded in Strickland. In addition,

even before Padilla was decided, the Court had already recognized the

importance of considering potential immigration consequences when entering into

a plea agreement. See INS v. St. Cyr, 

, 322–23 (2001).

      Despite these arguments, we think the better argument is that Padilla

announced a new rule of constitutional law. While the Supreme Court had never


                                          16
foreclosed the application of Strickland to collateral consequences of a

conviction, it had never applied Strickland to them either. And lower courts had

adhered to this direct versus collateral dichotomy. The departure from that

longstanding legal distinction, and the application of Strickland to immigration

consequences of a guilty plea, was an extension of Strickland into previously

untread grounds. See Stephanos Bibas, Regulating the Plea-Bargaining Market:

From Caveat Emptor to Consumer Protection, 99 C AL . L. R EV . 1117, 1118 (2011)

(“Padilla v. Kentucky marks a watershed in the Court’s approach to regulating

plea bargains.”). We find a reasonable jurist at the time of Hong’s conviction

would not have considered Supreme Court precedent to compel the application of

Strickland to the immigration consequences of a guilty plea. Indeed, we as a

court did not feel so compelled prior to Padilla.

      In sum, we find Padilla announced a new rule of constitutional law. 9

      We note one other circuit court to consider Padilla’s retroactivity reached

the opposite conclusion, finding “Padilla followed directly from Strickland and



      9
          The Seventh Circuit recently reached the same conclusion. In Chaidez
v. United States, No. 10-3623, 

, at *8 (7th Cir. Aug. 23, 2011),
it concluded that Padilla established a new rule, finding that the narrow definition
of what constitutes an old rule in Teague “tips the scales” toward a new rule.
Another circuit court, in an unpublished decision, doubted Padilla applied
retroactively. See United States v. Hernandez-Monreal, 404 F. App’x 714, 715
n.1 (4th Cir. 2010) (“[N]othing in the Padilla decision indicates that it is
retroactively applicable to cases on collateral review.”). See also Commonwealth
v. Clarke, 

(Mass. 2011) (concluding Padilla did not apply a new
rule but merely extended Strickland).

                                         17
long-established professional norms” and therefore was “an ‘old rule’ for Teague

purposes.” United States v. Orocio, No. 10-1231,– F.3d –, 

, at

*7 (3d Cir. June 29, 2011). We disagree and believe Padilla marked a dramatic

shift when it applied Strickland to collateral civil consequences of a

conviction—a line courts had never crossed before.

      The Third Circuit, while acknowledging deportation consequences never

were considered within the scope of Strickland before Padilla, found that history

unhelpful. Orocio, 

, at *4. Rather than rest its analysis on an

“incomplete approach,” it concluded “Padilla followed from the clearly

established principles” of Strickland. 

*6. The case “broke no new

ground” and was “hardly novel” based on the prevailing professional norms. 

(quotation omitted). Padilla’s application of Strickland was not new

because defense counsel was “long required” to provide effective assistance

regarding all “important decisions” that could affect the plea process, which

included the immigration consequences of a guilty plea. 

Padilla is

“best read as merely recognizing that a plea agreement’s immigration

consequences” is the type of information a defendant would need when making

important decisions about a plea bargain. 

from extending

the Strickland rule into uncharted territory, Padilla reaffirmed defense counsel’s

obligations to the criminal defendant during the plea process.” 

above, we disagree. Padilla extended the Sixth Amendment


                                         18
right to effective counsel and applied it to an aspect of a plea bargain previously

untouched by Strickland. See 

(Alito, J., concurring in

judgment) (“[T]he majority does not cite a single case, from this or any other

federal court, holding that criminal defense counsel’s failure to provide advice

concerning the removal consequences of a criminal conviction violates a

defendant’s Sixth Amendment right to counsel.”); cf. Bibas, 99 C AL . L. R EV . at

1139 (“[T]he early signs are that Padilla was not a one-off decision but may have

heralded the dawn of a new era.”). Padilla is a new rule of constitutional law not

because of what it applies—Strickland—but because of where it applies—

collateral immigration consequences of a plea bargain.

      3. Exceptions to the Retroactivity Bar

      Having determined Padilla represents a new rule of constitutional law, we

move to step three of the Teague analysis and consider whether Padilla applies

retroactively to cases on collateral review. 10 As a new rule of criminal procedure,


      10
          It is not readily apparent which court must declare that a new rule
applies retroactively. We have held “Teague’s retroactivity analysis . . .
determines whether the new rule is applicable to an initial motion for collateral
habeas relief.” Browning v. United States, 

, 1264 (10th Cir. 2001)
(en banc). That is, we may apply Teague in the first instance and determine
ourselves whether a new rule applies retroactively to initial habeas petitions. By
contrast, for second or successive habeas petitions, “a new rule is made
retroactive to cases on collateral review only when the Supreme Court explicitly
holds that the rule it announced applies retroactively to such cases.” Bey v.
United States, 

, 1268 (10th Cir. 2005).

      Other circuit courts have compared the language governing retroactivity in
                                                                   (continued...)

                                          19
Padilla will apply retroactively only if it falls within one of the two narrow

exceptions to the retroactivity bar outlined in Teague. We find Padilla does not

fit within either Teague exception and therefore does not apply retroactively to

cases, like Hong’s, on collateral review.

      A new rule will apply retroactively to a final conviction only under very

limited circumstances. Schriro v. Summerlin, 

, 351 (2004). “A new

rule applies retroactively in a collateral proceeding only if (1) the rule is

substantive, or (2) the rule is a watershed rule of criminal procedure implicating

the fundamental fairness and accuracy of the criminal proceeding.” 11 Whorton,

      10
         (...continued)
§ 2255(f)(3) and § 2255(h)(2) to determine whether a circuit court may decide if a
new rule applies retroactively to initial habeas petitions. Compare § 2255(f)(3)
(“[T]hat right has been newly recognized by the Supreme Court and made
retroactively applicable to cases on collateral review.”) with § 2255(h)(2) (“[A]
new rule of constitutional law, made retroactive to cases on collateral review by
the Supreme Court.”). Those courts have concluded they can because
§ 2255(f)(3) “does not require that the initial retroactivity question be decided in
the affirmative only by the Supreme Court.” United States v. Thomas, 

, 536–37 (4th Cir. 2010); see also Wiegand v. United States, 

,
892 (6th Cir. 2004); Dodd v. United States, 

, 1278 (11th Cir. 2004)
aff’d 

(2005); United States v. Swinton, 

, 486–87 (3d
Cir. 2003); Fischer v. United States, 

, 599–600 (7th Cir. 2002);
United States v. Lopez, 

, 432 (5th Cir. 2001).
      11
          Previously, the Supreme Court described the first Teague exception as
applying to “rules forbidding punishment of certain primary conduct or to rules
prohibiting a certain category of punishment for a class of defendants because of
their status or offense.” 

. Recently, however, the Court
explained “[r]ules that fall within what we have referred to as Teague’s first
exception ‘are more accurately characterized as substantive rules not subject to
[Teague’s] bar.’” 

n.3 (quoting 

n.4). Padilla’s
rule does not implicate this category of rules, whether such rules are deemed
                                                                       (continued...)

                                            

(quotation and alteration omitted). A substantive rule is one that

“alters the range of conduct or the class of persons that the law punishes.” 12

. By contrast, a procedural rule “regulate[s] only the

manner of determining the defendant’s culpability.” 13 

in Padilla is procedural, not substantive. It regulates the manner

in which a defendant arrives at a decision to plead guilty. Thus, only the second

Teague exception might apply here—as a watershed rule of criminal procedure

implicating the fundamental fairness and accuracy of the criminal proceeding. 

The exception is quite narrow, and since Teague, the Court has rejected

every attempt to fit a case within the exception. See 

–18

      11
        (...continued)
substantive or procedural falling within Teague’s first exception. Cf. United
States v. Price, 

, 848 n.2 (10th Cir. 2005).
      12
          New substantive rules generally apply retroactively “because they
‘necessarily carry a significant risk that a defendant stands convicted of an act
that the law does not make criminal’ or faces a punishment that the law cannot
impose upon him.” 

(quoting Bousley v. United States,

, 620 (1998)) (other quotations omitted). Substantive rules include
“decisions that narrow the scope of a criminal statute by interpreting its terms, as
well as constitutional determinations that place particular conduct or persons
covered by the statute beyond the State’s power to punish.” 

–52 (citation omitted).
      13
          New procedural rules, unlike new substantive rules, “do not produce a
class of persons convicted of conduct the law does not make criminal,” but simply
raise the potential that a defendant who was convicted under improper procedure
may have been otherwise acquitted. 

. Based on this
“more speculative connection to innocence,” the Supreme Court gives retroactive
effect to a very small set of procedural rules that “implicat[e] the fundamental
fairness and accuracy of the criminal proceeding” as represented in Teague’s
second exception. 

cases).

      To surmount this “watershed” requirement, a new rule (1) “must be

necessary to prevent an impermissibly large risk of an inaccurate conviction,” and

(2) “must alter our understanding of the bedrock procedural elements essential to

the fairness of a proceeding.” 

(quotations omitted). Elevating the

standard even more, a “showing that a new procedural rule is based on a

‘bedrock’ right” is insufficient because “a new rule must itself constitute a

previously unrecognized bedrock procedural element that is essential to the

fairness of a proceeding.” 

      The Supreme Court has repeatedly identified its decision in Gideon v.

Wainwright, 

(1963)—recognizing an indigent defendant’s right to

counsel—as the only rule which, if Gideon had been decided after Teague, might

have fallen within the second Teague exception. See 

–19; 

–18. The Court has “not hesitated to hold that

less sweeping and fundamental rules [than Gideon] do not fall within Teague’s

second exception.” 

; Steven W. Allen, Toward a Unified

Theory of Retroactivity, 54 N.Y.L. S CH . L. R EV . 105, 128 (2009-2010) (noting

Teague’s second exception only “relates to errors of the highest magnitude”).

      Applying this rubric here, we conclude Padilla did not announce a

watershed rule of criminal procedure and does not fall within Teague’s second

exception to the retroactivity bar. Therefore, Padilla does not retroactively apply


                                         22
to Hong’s case on collateral review.

      Simply put, Padilla is not Gideon. Padilla does not concern the fairness

and accuracy of a criminal proceeding, but instead relates to the deportation

consequences of a defendant’s guilty plea. The rule does not affect the

determination of a defendant’s guilt and only governs what advice defense

counsel must render when his noncitizen client contemplates a plea bargain.

Padilla would only be at issue in cases where the defendant admits guilt and

pleads guilty. In such situations, because the defendant’s guilt is established

through his own admission—with all the strictures of a Rule 11 plea

colloquy—Padilla is simply not germane to concerns about risks of inaccurate

convictions or fundamental procedural fairness.

      Despite all this, Hong contends the Supreme Court did implicitly decide the

issue of retroactivity in Padilla. We find Hong’s argument unpersuasive.

Nowhere in Padilla does the Supreme Court state even tangentially that its

holding applies retroactively to cases on collateral review. Hong tacitly

acknowledges this point but argues the Court’s decision “strongly implies”

Padilla applies retroactively. Aplt. Br. at 9.

       In Padilla, the Court acknowledged it gave “serious consideration to the

concerns” that its ruling might undermine the finality of convictions by opening

the “floodgates” to challenges of convictions obtained through guilty 

. The Court allayed these concerns by noting the same concerns


                                          23
were raised in prior Strickland cases. Those past decisions did not lead to a flood

of cases, likely because “[s]urmounting Strickland’s high bar is never any easy

task”—a defendant must show deficient representation as well as prejudice. 

The Court went on to conclude

      It seems unlikely that our decision today will have a significant
      effect on those convictions already obtained as the result of plea
      bargains. . . . Those who collaterally attack their guilty pleas lose the
      benefit of the bargain obtained as a result of the plea. Thus, a
      different calculus informs whether it is wise to challenge a guilty
      plea in a habeas proceeding because, ultimately, the challenge may
      result in a less favorable outcome for the defendant, whereas a
      collateral challenge to a conviction obtained after a jury trial has no
      similar downside potential.

(first emphasis added and citation omitted). Hong points to this

language as evidence the Court decided the issue of Padilla’s retroactivity. He

argues there would be no need to discuss pleas “already obtained” if the case did

not apply retroactively. See Orocio, 

, at *7 (“[I]t is not

unlikely that the Padilla Court anticipated the retroactive application of its

holding on collateral review when it considered the effect its decision would have

on final convictions.”); United States v. Hubenig, No. 6:03-mj-040, 

, at *8 (E.D. Cal. July 1, 2010) (“If the Court intended Padilla to be a

new rule which would apply only prospectively, the entire ‘floodgates’ discussion

would have been unnecessary.”).

      We disagree. We interpret the Court’s statement to simply recognize that

past decisions enumerating the contours of Strickland have not led to a surfeit of

collateral attacks on guilty pleas. The force of the Court’s argument is that

                                          24
Padilla would have a similar (lack of) effect on guilty pleas. In addition, we

think it unwise to imply retroactivity based on dicta—and abandon the Teague

analysis entirely. The Teague framework exists to promote the finality of

convictions by shielding them from collateral attacks mounted on new procedural

rules of constitutional law. To imply retroactivity from an isolated phrase in a

Supreme Court opinion would completely ignore this goal.

      In sum, we find Padilla did not announce a new watershed rule of criminal

procedure that affects the fundamental fairness and accuracy of a criminal

proceeding. It is not within either of the extremely narrow Teague exceptions to

the retroactivity bar. Therefore, Padilla is a new rule of constitutional law but

does not apply retroactively to cases on collateral review.

      C. Padilla and § 2255(f)(3)

      Because Padilla does not apply retroactively to cases on collateral review,

the limitations period under § 2255(f)(3) does not apply here. Hong’s motion was

untimely because it was not filed within one year of his conviction becoming

final, as required by § 2255(f)(1). Accordingly, no reasonable jurist could

conclude the district court erred when it dismissed Hong’s § 2255 motion as

untimely.

                                  III. Conclusion

      For the foregoing reasons, we DENY Hong’s application for a COA and

DISMISS his appeal.


                                         25