Sixth Circuit Forecloses Merits-Panel Expansion of COAs and Distinguishes Katz and Jones as Separate, Exhaustion-Required Fourth Amendment Theories

Introduction

In a published decision with significant implications for federal habeas practice in the Sixth Circuit, the court affirmed the denial of habeas relief to Andrew Maurice Randolph and, in doing so, set two important procedural markers. First, the panel held that a merits panel lacks authority to expand a Certificate of Appealability (COA) beyond the issues previously certified by a motions panel and after an en banc denial, at least where the State objects and absent a local rule authorizing such expansion. Second, the court clarified that the “Katz” reasonable-expectation-of-privacy theory and the “Jones” property-trespass theory describe distinct Fourth Amendment claims for exhaustion purposes; raising only Katz in state court does not fairly present Jones.

The case arises from Randolph’s state conviction for second-degree murder and firearms offenses following the fatal shooting of his girlfriend’s mother. The evidentiary pillars at trial were ammunition found in Randolph’s belongings at his father’s home (searched with the father’s consent) and a handgun recovered from his brother’s home (searched under the brother’s parole conditions), later identified as the murder weapon. Randolph argued his trial counsel rendered ineffective assistance by failing to move to suppress both sets of evidence. After a partial COA was granted limited to the suppression-based ineffective assistance claim, Randolph sought, for a fifth time, to expand the COA to reach other issues. The panel declined, then rejected the certified claim under AEDPA’s deferential standards.

Summary of the Opinion

Writing for the court, Judge Bush affirmed the district court’s denial of habeas relief and denied Randolph’s request to expand the COA. The court:

• Held that a Sixth Circuit merits panel lacks authority to expand a COA beyond what a motions panel granted, especially after an en banc denial and where the State objects, citing AEDPA’s text and structure, the absence of a Sixth Circuit local rule authorizing expansion, the law-of-the-case doctrine, and respect for en banc dispositions.
• Limited the merits to the certified “Katz theory” of Fourth Amendment expectation of privacy; deemed Randolph’s “Jones theory” (property-based trespass) unexhausted because it was never fairly presented to the Michigan courts.
• Concluded that the Michigan Court of Appeals’ factual findings were not unreasonable within the meaning of § 2254(d)(2), and its legal conclusions were neither contrary to nor an unreasonable application of clearly established Supreme Court precedent under § 2254(d)(1).
• Emphasized that reliance on Sixth Circuit decisions such as United States v. Waller could not establish entitlement to relief under AEDPA, which requires a misapplication of Supreme Court holdings.

Judge Thapar concurred, agreeing with the disposition but preferring to resolve the Fourth Amendment prejudice question on attenuation grounds: the two-month gap, intervening circumstances (the brother’s parole search condition), and non-flagrant conduct severed the causal chain between the initial search of Randolph’s belongings and discovery of the murder weapon. Judge Mathis also concurred in part and in the judgment, agreeing the COA should not be expanded in this case but disputing the majority’s categorical holding; he would preserve a merits panel’s authority to expand a COA in appropriate circumstances.

Analysis

Precedents Cited and Their Roles

The panel’s opinion is steeped in AEDPA’s framework and draws several important lines:

• COA scope and authority
  • 28 U.S.C. § 2253(c): Requires a COA and specifies it “shall indicate [the] specific issue or issues” to be reviewed. The majority treats this specification as a mandatory limitation when invoked by the State (citing Gonzalez v. Thaler, 565 U.S. 134, 146 (2012), recognizing § 2253(c)(3) as nonjurisdictional but enforceable upon timely objection) and emphasizes AEDPA’s design to limit collateral review.
  • Hohn v. United States, 524 U.S. 236 (1998): Identifies the Supreme Court as the remaining avenue if a COA is denied. The majority notes Randolph did not seek Supreme Court relief after the Sixth Circuit denied expansion en banc.
  • Third and Ninth Circuit local rules (Villot v. Varner; Olivas v. Neven) allow merits panels to expand COAs; the Sixth Circuit has no analogous local rule, and the majority uses the expressio unius canon (Bittner v. United States) to stress that omission.
  • Law-of-the-case doctrine: Daunt v. Benson; Wallace v. FedEx Corp.; Rouse v. DaimlerChrysler; Edmonds v. Smith. The majority treats prior COA denials (including en banc denial) as binding in the same appeal, absent recognized exceptions.
• Exhaustion
  • 28 U.S.C. § 2254(b)(1)(A): Requires “fair presentation” of the claim’s factual and legal bases to the state courts, through the state appellate system, including the state supreme court (Wagner v. Smith; Whitman v. Gray).
  • Miller v. Genovese, 994 F.3d 734 (6th Cir. 2021): The court distinguishes Miller’s Confrontation Clause reasoning. In Miller, a general Crawford cross-examination argument subsumed the specific Van Arsdall bias-impeachment limitation. Here, by contrast, Katz and Jones are independent Fourth Amendment theories; one does not subsume the other, so only raising Katz failed to fairly present Jones.
• AEDPA merits standards
  • 28 U.S.C. § 2254(d)(1)-(2): Bars relief unless the state-court decision is contrary to or an unreasonable application of clearly established Supreme Court law, or rests on an unreasonable factual determination.
  • Harrington v. Richter; White v. Woodall; Knowles v. Mirzayance: Frame the high bar for establishing unreasonableness under AEDPA; fairminded disagreement precludes relief; no relief for failure to follow rules not “squarely established.”
  • Johnson v. Williams: State courts are not bound by lower federal court decisions; reliance on Sixth Circuit cases (e.g., Waller) cannot prove an unreasonable application of Supreme Court precedent.
  • Wilson v. Sellers: Review focuses on the last reasoned state-court decision (here, the 2019 Michigan Court of Appeals decision).
• Findings-of-fact deference
  • § 2254(d)(2): State-court factual findings are presumed correct and can be displaced only if “objectively unreasonable by clear and convincing evidence” (Miller-El v. Cockrell; Hodge v. Plappert; Shoop v. Twyford; Hale v. Cool). The petitioner must also show the factual error was the but-for cause of the decision (Hale; Rice v. White).
• Fourth Amendment frameworks
  • Katz v. United States vs. United States v. Jones: The court underscores these as distinct doctrinal tracks—privacy-based expectations versus property-trespass concepts—each potentially triggering different analyses and each requiring separate exhaustion.
  • Attenuation (concurring opinion): Utah v. Strieff; Brown v. Illinois; United States v. Leon; Herring v. United States; Pennsylvania Bd. of Prob. & Parole v. Scott; Wong Sun. Judge Thapar applies these to explain why the murder weapon would be admissible even if the initial search were unlawful.

Legal Reasoning

1) No merits-panel authority to expand the COA in this case

The majority grounds its holding in four strands:

• Text and structure of AEDPA: Congress requires issue-specific certification, and the statute’s design emphasizes finality and gatekeeping. Allowing merits panels to relitigate COA denials would effectively collapse the motions-stage screen, diluting AEDPA’s limitations.
• No local rule authorizing expansion: Unlike the Third and Ninth Circuits, the Sixth Circuit’s local rules have no mechanism empowering a merits panel to expand a COA. Where Congress or local rules speak elsewhere and are silent here, the majority reads that silence as intentional.
• Law-of-the-case: A single judge, a motions panel, and the en banc court each refused to expand the COA. Absent a recognized exception, a later merits panel does not revisit the same COA question—particularly when the State has objected to consideration beyond the certified issue (see Gonzalez v. Thaler’s admonition that claim-processing rules, though nonjurisdictional, cannot be ignored when invoked).
• Even under more permissive circuits’ practices: The majority notes other circuits caution against lightly overturning motions-panel COA decisions (e.g., Duka v. United States), especially where the merits briefing simply repackages arguments already rejected at the COA stage.

The opinion expressly “holds that a merits panel of this court lacks the authority to expand a COA beyond what was granted by the motions panel,” reiterating that Randolph’s remaining recourse after en banc denial lay with the Supreme Court (Hohn).

2) Exhaustion: Katz and Jones are independent claims

The court cabined review to the “Katz theory” (reasonable expectation of privacy) because Randolph never fairly presented a “Jones theory” (trespass to effects) in state court. The opinion explains that the two are not merely different routes to the same destination:

• A scenario may negate privacy expectations (e.g., an open handbag visible to officers) while still involve physical intrusion amounting to trespass if officers reach into the bag. That possibility shows the claims can diverge in outcome and must be independently exhausted.
• Miller v. Genovese does not help the petitioner; Van Arsdall’s bias-impeachment right is a narrow subset of Crawford’s cross-examination guarantee, but Jones is not a subset of Katz. The court therefore refuses to deem Jones “fairly presented” by virtue of a Katz framing.

3) AEDPA review of the state court’s decision on the Katz theory

On the certified ineffective-assistance claim tied to a suppression motion under Katz, the court takes a two-step approach:

• Factual findings: The Michigan Court of Appeals found (a) Randolph left Fant’s home fearing police involvement, and (b) he failed to timely retrieve his belongings. The panel holds those are reasonable readings of the record and, importantly, are not the but-for cause of the state court’s outcome. The state court also found that Randolph failed to show a subjective expectation of privacy, which independently defeats a Katz claim; the challenged findings relate to the objective reasonableness prong. Thus, even if the challenged findings were wrong, they would not have changed the result under § 2254(d)(2).
• Legal conclusions: Under § 2254(d)(1), the petitioner identified no Supreme Court precedent the state court unreasonably applied. His reliance on Sixth Circuit case law (e.g., United States v. Waller) cannot, under Johnson v. Williams, establish a violation of clearly established federal law. Given Richter and Woodall, fairminded jurists could disagree with the petitioner’s position, so habeas relief is unavailable.

A footnote underscores that the petitioner’s separate challenge to whether ammunition would independently support a warrant was waived in state court and thus unexhausted, so the panel does not consider it.

Concurrences: Alternative paths and limits to the majority’s rule

• Judge Thapar (attenuation): He would resolve the prejudice component of the Strickland claim by holding that any suppression motion aimed at the murder weapon would have failed on attenuation grounds. Three factors point to admissibility:
  • Temporal gap: Two months elapsed between the search of Randolph’s belongings and discovery of the gun. Cases like Elmore and Gross treat such a gap as strongly favoring attenuation; by contrast, Strieff and Brown involved minutes or less than two hours.
  • Intervening circumstances: The brother’s parole status and search condition—predating and unconnected to the December search—provided an independent, lawful basis to search the home where the weapon was found, analogous to the valid arrest warrant in Strieff.
  • Non-flagrancy: The initial search occurred after the father consented (even if arguably unauthorized), and the officers exhibited solicitude for Fourth Amendment requirements; this was not systemic or egregious misconduct “most in need of deterrence.”
  Because any motion to suppress would fail, counsel was not ineffective and the petitioner suffered no Strickland prejudice.
• Judge Mathis (COA authority): While agreeing that expansion should be denied here, he disputes the majority’s categorical rule. He stresses:
  • § 2253(c)(3) is nonjurisdictional (Gonzalez v. Thaler), so a panel is not powerless to expand a COA.
  • Sixth Circuit authority has recognized merits-panel flexibility to expand COAs (Kraus v. Taylor; Post v. Bradshaw; and analogous authority in Howard v. United States), and motions-panel orders are “interlocutory” and “subject to revision” (Wallace; Kraus).
  • Law-of-the-case doctrine is a poor fit for interlocutory COA orders and does not rigidly bar reconsideration before final judgment.
  • The en banc court did not decide anything on the merits; it merely declined rehearing. Accordingly, future merits panels should retain discretion to expand COAs in appropriate cases.

Impact

The decision carries significant practical and doctrinal consequences:

• COA practice in the Sixth Circuit
  • Litigants should assume that a motions panel’s COA scope is the last word within the Sixth Circuit. After a denial of expansion by a motions panel and en banc, a merits panel will not revisit COA scope, particularly where the State objects to consideration of uncertified issues.
  • Practitioners must present their best case at the COA stage, including any request for expansion, and, if denied, promptly consider seeking Supreme Court review under Hohn.
  • Although Judge Mathis points to Kraus and Post to preserve flexibility, the majority’s published holding will likely govern future panels unless and until the Sixth Circuit revisits the issue en banc or the Supreme Court intervenes.
• Exhaustion strategy for Fourth Amendment claims
  • The opinion removes any doubt that Katz and Jones theories must be separately raised in state court. Defense counsel should plead and brief both expectation-of-privacy and trespass-to-effects theories where both are colorable, to avoid procedural default of one or the other.
  • Attempts to argue that one theory “subsumes” the other, as in Miller’s Crawford/Van Arsdall logic, will not succeed; the court’s handbag hypothetical lucidly demonstrates how outcomes can diverge.
• AEDPA merits: Supreme Court holdings or bust
  • Ineffective-assistance claims premised on unfiled suppression motions must anchor their underlying Fourth Amendment theory in clearly established Supreme Court law, not lower court decisions. Reliance on circuit precedent (e.g., Waller) is insufficient to satisfy § 2254(d)(1).
  • The court’s suggestion that failure to cite Sixth Circuit authority is unlikely to be constitutionally deficient—because state courts are free to reject it—reinforces strategic emphasis on Supreme Court doctrine.
• Attenuation roadmap
  • Judge Thapar’s concurrence offers a prosecution- and law-enforcement-friendly template: extended time gaps, intervening lawful authorities (such as parole/probation search conditions), and non-flagrant conduct will often defeat suppression—even if an earlier step was unlawful.
  • Defense counsel should be prepared to address attenuation head-on when crafting suppression arguments and Strickland prejudice showings.

Complex Concepts Simplified

• Certificate of Appealability (COA): A gatekeeping device requiring a habeas petitioner to obtain permission to appeal, and specifying the particular issues that may be raised. Without a COA, the court of appeals cannot reach the merits. In the Sixth Circuit after this case, a merits panel will not expand a COA beyond what a motions panel granted, especially over the State’s objection and absent a local rule authorizing expansion.
• AEDPA deference: Federal courts cannot grant habeas relief simply because they think the state court was wrong. Relief is permitted only if the state decision is contrary to or an unreasonable application of Supreme Court holdings, or based on an unreasonable factual determination. Reasonable disagreement among jurists is enough to deny relief.
• Exhaustion: A habeas petitioner must “fairly present” both the legal theory and factual basis of each claim to the state appellate courts, including the state’s highest court. Raising only a privacy-based Katz argument does not exhaust a property-based Jones argument.
• Katz vs. Jones: Katz asks whether the defendant had a subjective expectation of privacy that society recognizes as reasonable; Jones asks whether the government committed a physical trespass on “persons, houses, papers, and effects.” They are distinct routes to finding a Fourth Amendment violation.
• Strickland ineffective assistance: The petitioner must show deficient performance and prejudice. For failure-to-suppress claims, prejudice requires showing that a suppression motion would have succeeded and that excluding the evidence would likely have affected the outcome. Under AEDPA, the standard is “doubly deferential.”
• Attenuation doctrine: Even if police commit a Fourth Amendment violation, evidence may still be admissible if the connection between the illegality and the evidence is too remote. Courts weigh temporal proximity, intervening circumstances, and the flagrancy of the initial misconduct.
• Law-of-the-case: A doctrine counseling courts not to revisit earlier rulings in the same case absent special circumstances. Here, the majority treats prior COA-limitation rulings as binding in the same appeal.
• “Clearly established law”: Only Supreme Court holdings count for AEDPA purposes; lower federal court decisions cannot create “clearly established” law binding on state courts.
• § 2254(d)(2) factual errors and causation: It is not enough to find a state-court factual error; the petitioner must show it was objectively unreasonable and that, but for that error, the result would have been different.

Conclusion

Randolph v. Macauley is a consequential habeas decision on two fronts. Procedurally, it announces a firm rule in the Sixth Circuit: merits panels may not expand COAs beyond the issues certified by motions panels, particularly after en banc denial and in the face of a State’s objection, absent a local rule authorizing such expansion. Substantively, it underscores AEDPA’s rigorous demands. The court declined to reach an unexhausted Jones trespass theory, held that the state court’s factual determinations were not unreasonably made (and, in any event, not outcome-determinative), and reiterated that only Supreme Court holdings can sustain relief under § 2254(d)(1). The concurrences deepen the opinion’s practical importance: Judge Thapar charts an attenuation path that would have independently defeated suppression-based prejudice, while Judge Mathis cautions against broadly foreclosing merits-panel COA expansion in all future cases.

The key takeaways are clear. Habeas petitioners must frame and exhaust discrete Fourth Amendment theories in state court; COA litigation in the Sixth Circuit is front-loaded and largely final at the motions stage; and AEDPA demands that ineffective-assistance suppression claims be anchored in Supreme Court precedent and survive attenuation analysis. Collectively, these principles advance AEDPA’s core aim of finality while offering a detailed roadmap for future litigants on both sides of the “v.”