COA Denial Where Padilla Claim Depends on Disputing Credibility-Based Findings of Deportation Advice

Introduction

In United States v. Akinsanya (10th Cir. Mar. 3, 2026), the Tenth Circuit denied Abdul Waris Akinsanya a certificate of appealability (COA) from the denial of his 28 U.S.C. § 2255 motion. Akinsanya, a Nigerian citizen, had pleaded guilty to conspiracy to commit mail fraud (18 U.S.C. §§ 371, 1341) arising from online “romance scams,” stipulated to a loss amount of $200,000, and received an 18-month sentence.

The central issue on appeal was a Sixth Amendment ineffective-assistance claim under Padilla v. Kentucky, 559 U.S. 356 (2010): whether trial counsel failed to advise him that his plea would trigger “mandatory” (i.e., virtually certain) deportation because the conviction qualified as an “aggravated felony” under immigration law. The district court held an evidentiary hearing, credited trial counsel’s testimony that he warned Akinsanya of deportation consequences, and denied relief and a COA. The Tenth Circuit’s order focuses on the COA standard and the deference owed to credibility-based factual findings at this threshold stage.

Summary of the Opinion

The court denied a COA and dismissed the appeal. Applying the COA standard from Buck v. Davis, 580 U.S. 100 (2017) and Miller-El v. Cockrell, 537 U.S. 322 (2003), the court held that no reasonable jurist could debate the district court’s key factual finding: that counsel advised Akinsanya he would be deported because the plea involved fraud and losses well above the $10,000 aggravated-felony threshold. Because Akinsanya’s argument depended on contradicting that factual finding, and because the record supported it, the court concluded there was “no room for debate” as to the outcome of the Padilla claim.

The court also noted, but declined to address, an oblique suggestion that the plea was not knowing and voluntary (a due process theory), because Akinsanya conceded he was pursuing only the Sixth Amendment Padilla issue, did not adequately brief due process, and had not raised it in his § 2255 motion.

Analysis

Precedents Cited

  • Padilla v. Kentucky, 559 U.S. 356 (2010)
    Padilla is the doctrinal foundation for the claim: defense counsel must advise a noncitizen client about immigration consequences of a plea when those consequences are clear. The Tenth Circuit assumed (without deciding) that a warning about “mandatory deportation consequences” is required, but it found the dispositive question was factual—whether counsel gave that warning.
  • Sessions v. Dimaya, 584 U.S. 148 (2018)
    Cited for the proposition that an aggravated-felony conviction can make deportation a “virtual certainty,” underscoring why immigration consequences are material in plea advice. The case supplies context for the severity and determinacy of removal consequences once the aggravated-felony label applies.
  • Kawashima v. Holder, 565 U.S. 478 (2012)
    Used in the opinion’s footnote to frame the first step of the aggravated-felony analysis—whether the offense “involv[es] fraud or deceit” under a categorical approach inquiry (and, for certain statutes, potentially a modified categorical approach).
  • Matter of Nemis, 28 I. & N. Dec. 250 (B.I.A. 2021)
    Cited to explain that 18 U.S.C. § 371 can be violated via distinct clauses (“offense” vs. “defraud”), and that a modified categorical approach may be used to determine which clause was involved. This supports looking to the predicate statute (here, mail fraud) to decide whether the conspiracy conviction involves “fraud or deceit.”
  • United States v. Ochoa, 861 F.3d 1010 (9th Cir. 2017)
    Cited as an example of analyzing the predicate criminal statute for a § 371 conviction to determine aggravated-felony status. Although not binding in the Tenth Circuit, it provides persuasive support for the analytic pathway used in the footnote.
  • Nijhawan v. Holder, 557 U.S. 29 (2009)
    Cited for the second step of the aggravated-felony inquiry under 8 U.S.C. § 1101(a)(43)(M)(i): examining the “specific circumstances” (a circumstance-specific approach) to determine whether the loss exceeded $10,000. Here, Akinsanya’s stipulation to $200,000 loss made this prong straightforward.
  • Buck v. Davis, 580 U.S. 100 (2017) and Miller-El v. Cockrell, 537 U.S. 322 (2003)
    These cases provide the COA threshold: the applicant must show that “jurists of reason could disagree” with the district court’s resolution or that the issues deserve encouragement to proceed. The Tenth Circuit applied this standard to the district court’s credibility and factual findings.

Legal Reasoning

  1. COA is a threshold inquiry, not a full merits appeal.
    The court treated the COA request as requiring a showing that reasonable jurists could debate the district court’s disposition. This framing is significant because it narrows the appellate lens to whether the district court’s resolution is debatably wrong, not whether the defendant can re-litigate facts anew.
  2. The claim’s viability turned on a single factual premise.
    Akinsanya’s Padilla theory—counsel failed to warn of mandatory deportation—depended on establishing that counsel did not provide such advice. The district court found the opposite: counsel advised Akinsanya “that he would be deported” due to a fraud conviction with substantial loss and tried to negotiate an immigration-safe disposition.
  3. Credibility determinations and record support foreclosed “reasonable debate.”
    The district court credited trial counsel’s testimony, including that he consulted immigration resources, referred Akinsanya to immigration attorneys, understood that aggravated felonies trigger mandatory removal, and repeatedly explained that a fraud conviction with losses over $10,000 would cause mandatory deportation. The Tenth Circuit emphasized that this finding was “well-supported by the record,” and therefore no reasonable jurist could debate it.
  4. Assuming a legal duty, the court resolved the case on factual grounds.
    The court expressly assumed (without deciding) that Padilla requires warning about “mandatory deportation consequences.” Even under that assumption, the claim failed because the factual predicate (no warning) could not be reasonably disputed on this record.
  5. Scope control: the court declined to reach due process.
    The court avoided expanding the case into a plea-voluntariness/due process inquiry because it was not properly preserved or briefed, and was not raised in the § 2255 motion. This reflects appellate discipline: COA review is limited to the claim actually pursued and developed.

Impact

Although the order is designated nonprecedential, it conveys a practical rule for COA litigation in immigration-advice cases: where the district court has held an evidentiary hearing and made credibility-based findings that counsel did, in fact, warn a defendant about deportation consequences, a COA is unlikely if the appeal merely re-argues the facts. The decision underscores that:

  • COA denials can hinge on factual deference. Even potentially substantial constitutional questions (like Padilla duties) may not proceed when the asserted constitutional violation depends on a factual premise resolved against the movant after live testimony.
  • Record-building is decisive. Trial counsel’s detailed testimony about advising on the $10,000 threshold, the “aggravated felony” trigger, and efforts to negotiate non-fraud alternatives created a record that insulated the district court’s ruling from “reasonable debate.”
  • Immigration-safe plea negotiation efforts matter. The court treated counsel’s search for alternative charges and discussion of strategy as corroboration of competent performance concerning immigration consequences.

Complex Concepts Simplified

  • Certificate of Appealability (COA): A permission slip to appeal the denial of a § 2255 motion. The appellant must show that reasonable judges could debate the district court’s decision or that the issue is substantial enough to proceed.
  • § 2255 motion: A post-conviction procedure allowing a federal prisoner to challenge the legality of a conviction or sentence, often based on constitutional violations like ineffective assistance of counsel.
  • Padilla duty: Defense counsel must advise a noncitizen client about the immigration consequences of a guilty plea when those consequences are clear (e.g., deportation will be mandatory).
  • Aggravated felony (immigration law): A category of offenses that triggers severe immigration consequences, commonly including mandatory removal and significant restrictions on relief from deportation. Relevant here: fraud or deceit offenses with losses exceeding $10,000. See 8 U.S.C. § 1101(a)(43)(M)(i).
  • Categorical vs. modified categorical approach: Methods used to decide whether a conviction fits an immigration category. The categorical approach looks to the statutory elements; the modified categorical approach allows a limited look at certain conviction records when a statute lists alternative ways to commit the offense.
  • Circumstance-specific loss inquiry: For the $10,000 loss requirement, courts may consider the actual loss tied to the offense, including stipulations, rather than only the statute’s elements. This is the principle associated with Nijhawan.

Conclusion

United States v. Akinsanya illustrates that, at the COA stage, the decisive question may be factual rather than doctrinal: even assuming Padilla requires explicit advice about mandatory deportation, an appellant cannot obtain a COA by re-litigating an adverse, credibility-based finding that counsel actually provided that advice. The order reinforces the gatekeeping role of the COA standard and the practical importance of evidentiary-hearing records in ineffective-assistance claims involving immigration consequences.