Substantial-Evidence Review Governs the BIA’s “Persecution” Determination Even on Undisputed Facts

1. Introduction

Urias-Orellana v. Bondi (2026) resolves a circuit split over the standard of review that federal courts of appeals must apply when reviewing the Board of Immigration Appeals’ (BIA) determination whether a noncitizen’s (largely undisputed) experiences amount to “persecution” (or a “well-founded fear of persecution”) under the Immigration and Nationality Act (INA), 8 U.S.C. §1101(a)(42)(A).

Petitioners—Douglas Humberto Urias-Orellana, his wife Sayra Iliana Gamez-Mejia, and their minor child—are Salvadoran nationals who entered the United States without authorization and sought asylum defensively in removal proceedings. Urias-Orellana testified that a sicario had threatened his family after shooting two half-brothers, and that threats and extortion demands followed the family through multiple internal relocations in El Salvador. The Immigration Judge (IJ) found the testimony credible but concluded it did not establish past persecution or a well-founded fear of future persecution; the BIA and the First Circuit affirmed. The Supreme Court granted certiorari to decide how reviewing courts must evaluate the agency’s “persecution” determination: de novo (as a mixed question of law and fact) or under the INA’s substantial-evidence standard.

2. Summary of the Opinion

The Court (Jackson, J., unanimous) holds that 8 U.S.C. §1252(b)(4)(B) requires courts of appeals to apply substantial-evidence review to the agency’s determination whether a given set of facts rises to the level of “persecution” under §1101(a)(42)(A)—including when the historical facts are undisputed and the dispute is whether those facts satisfy the statutory standard.

The Court affirms the First Circuit’s decision in Urias-Orellana v. Garland, 121 F. 4th 327, which reviewed the BIA’s persecution determination for substantial evidence and upheld the denial of asylum.

The Court emphasizes it is not deciding the correct legal standard for when threats constitute persecution; petitioners did not challenge the agency’s First Circuit-derived standard (quoted from Bonilla v. Mukasey), and the Solicitor General agreed that purely legal questions about the appropriate standard remain subject to de novo review.

3. Analysis

3.1. Precedents Cited

A. The INA’s substantial-evidence framework

Nasrallah v. Barr, 590 U. S. 573 (2020): Interprets §1252(b)(4)(B) as prescribing the deferential “substantial-evidence” standard. Urias-Orellana relies on Nasrallah to anchor the statutory meaning of “conclusive unless any reasonable adjudicator would be compelled to conclude to the contrary.”
Biestek v. Berryhill, 587 U. S. 97 (2019): Restates the classic definition of substantial evidence as “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” The Court uses Biestek to operationalize §1252(b)(4)(B)’s phrasing.
Consolidated Edison Co. v. NLRB, 305 U. S. 197 (1938): Supplies the canonical administrative-law formulation of substantial evidence, which Biestek quotes and the Court applies by analogy to immigration review under the INA.

B. The controlling immigration precedent: codification of a standard

INS v. Elias-Zacarias, 502 U. S. 478 (1992): The centerpiece of the Court’s analysis. There, applying the pre-IIRIRA judicial-review provision, the Court held that reversal of the agency’s persecution determination is available only when the record evidence is “so compelling that no reasonable factfinder could fail to find the requisite fear of persecution.” Urias-Orellana treats Elias-Zacarias as having already extended deferential review to the overall persecution determination, not just “raw” historical facts.
Post-Elias-Zacarias circuit decisions cited as confirming the contemporaneous understanding that asylum eligibility/persecution determinations were reviewed deferentially, including: Klawitter v. INS; Huaman-Cornelio v. BIA; Chun v. INS; Milosevic v. INS; Yacoub v. INS; Kazlauskas v. INS.
Bartenwerfer v. Buckley, 598 U. S. 69 (2023): The Court uses the interpretive presumption that Congress legislates against the backdrop of Supreme Court precedent—supporting the conclusion that IIRIRA’s language in §1252(b)(4)(B) should be read to codify, not undo, Elias-Zacarias.

C. Jurisdiction vs. standard of review (and why mixed questions don’t change the answer)

Wilkinson v. Garland, 601 U. S. 209 (2024), and Guerrero-Lasprilla v. Barr, 589 U. S. 221 (2020): Petitioners invoked these “mixed question” cases, but the Court distinguishes them as addressing whether an issue qualifies as a “question of law” for purposes of the INA’s jurisdiction-stripping provisions in §1252(a), not what standard of review applies under §1252(b)(4)(B).
Garland v. Ming Dai, 593 U. S. 357 (2021): Cited to underscore that even when an applicant’s story is accepted, the system is built around IJ factfinding (including credibility determinations), reinforcing Congress’s choice of deference in §1252(b)(4).
Loper Bright Enterprises v. Raimondo, 603 U. S. 369 (2024): Raised by petitioners to argue for de novo review; the Court declines to reach it because §1252(b)(4)(B) is a statute-specific command for deference (and Loper Bright does not apply where Congress mandates deferential review or delegates authority within constitutional limits).

D. Background authority referenced in the opinion text

United States v. Detroit Timber & Lumber Co., 200 U. S. 321 (1906): Cited in the syllabus boilerplate to clarify the syllabus is not part of the Court’s opinion.
Bonilla v. Mukasey, 539 F. 3d 72 (CA1 2008): Provides the (unchallenged here) First Circuit standard that death threats may constitute persecution only when “so menacing as to cause significant actual suffering or harm.”

E. The circuit split the Court resolves

The opinion catalogs divergent approaches across circuits. Some review the entire persecution inquiry for substantial evidence (e.g., Gómez-Medina v. Barr; Ai Hua Chen v. Holder; Yu v. Ashcroft; Ahmed v. Gonzales; Vicente-Elias v. Mukasey). Others articulate a bifurcated approach—substantial evidence for fact, de novo for “legal conclusions” (e.g., Diallo v. INS; Herrera-Reyes v. Attorney Gen.; Lopez-Gomez v. Ashcroft; Njong v. Whitaker; Kaur v. Wilkinson; Mejia v. Attorney Gen.). Still others blur the line in practice (e.g., Scarlett v. Barr; Thayalan v. Attorney Gen.; Gjetani v. Barr; Brizuela v. Garland; Sharma v. Garland; Martinez v. Attorney Gen.).

The Court’s holding pushes the doctrine toward a uniform rule: the “persecution” determination—characterized as a “mixed” application of the INA to established facts—still receives substantial-evidence review under §1252(b)(4)(B).

3.2. Legal Reasoning

A. Text and structure of §1252(b)(4)

The Court begins with the statutory review framework in §1252(b)(4), emphasizing that it is expressly titled to address the “[s]cope and standard for review.” Subparagraph (B) states that “administrative findings of fact are conclusive unless any reasonable adjudicator would be compelled to conclude to the contrary.” The Court treats this as a statutory command of deference, which it has previously labeled the substantial-evidence standard.

B. The decisive move: reading §1252(b)(4)(B) through the lens of INS v. Elias-Zacarias

Rather than resolving the case by abstract classification (fact vs. law vs. mixed), the Court leans on institutional continuity: under the pre-IIRIRA regime, INS v. Elias-Zacarias applied deferential review to the BIA’s persecution determination, including to whether accepted facts met the statutory standard. Congress then enacted IIRIRA and inserted §1252(b)(4)(B) with language that the Court reads as effectively restating Elias-Zacarias’s “compelled to conclude to the contrary” formulation.

From this, the Court draws an interpretive inference: Congress codified the deferential approach. Therefore, even if “persecution” is a mixed application, the whole determination remains subject to substantial-evidence review.

C. Rejecting petitioners’ “mixed question implies de novo” argument

Petitioners argued that because §1252(b)(4)(B) refers to “findings of fact,” and because persecution is a “mixed question of law and fact,” de novo review should apply. The Court responds that this formal label cannot overcome (i) Elias-Zacarias’s demonstrated treatment of persecution determinations under a deferential standard, and (ii) the post-1996 statutory history.

The Court also adds a functional point: refugee-status adjudication is fact-intensive, and even when facts are “undisputed” the IJ’s credibility finding is an antecedent factual determination. Given Congress’s consistent emphasis on deference across §1252(b)(4), it would be “anomalous” to allow courts to re-weigh substantially similar persecution-related determinations de novo.

D. Clarifying what Wilkinson and Guerrero-Lasprilla do (and do not) decide

The Court distinguishes the jurisdictional question (“is this a reviewable question of law under §1252(a)?”) from the merits-review question (“what standard of review applies under §1252(b)?”). Even if a mixed issue can be categorized as a “question of law” to avoid a jurisdictional bar, that does not dictate that the court must review it de novo where another statutory provision prescribes deference.

3.3. Impact

A. Immediate doctrinal effect: uniformity and deference

Urias-Orellana v. Bondi substantially narrows the room for de novo appellate recharacterization of an applicant’s mistreatment as “persecution.” Courts of appeals must uphold the BIA’s determination unless the record compels the opposite conclusion. This will tend to reduce reversals in close cases where the dispute is whether threats, harassment, or episodic violence cross the legal threshold of persecution.

B. Litigation strategy in asylum cases

Front-loading the record becomes even more decisive. Because appellate courts are confined to the administrative record (§1252(b)(4)(A)) and apply deference to the persecution determination, applicants’ best chance is to build compelling, corroborated evidence at the IJ stage (medical/psychological evidence, expert testimony, country conditions tying threats to protected grounds, and proof on internal relocation).
Legal challenges shift toward “pure law.” The Court notes (and the Government agreed) that “purely legal questions about the appropriate standard” of persecution remain subject to de novo review. Future petitioners may frame disputes as challenges to the legal test (e.g., the proper treatment of death threats, cumulative harm, or nexus), rather than arguing only that their facts meet an accepted test.
Agency consistency gains importance. Deferential review can amplify the practical significance of how the BIA articulates persecution thresholds and cumulative-harm analysis, since appellate courts will be less likely to correct borderline applications unless compelled.

C. System-level consequences

The decision reinforces Congress’s IIRIRA-era design of constrained judicial review in immigration cases, particularly where determinations are fact-intensive and intertwined with credibility, internal relocation, and severity-of-harm evaluations. Over time, this may increase the value of administrative appellate advocacy (BIA briefing, motions to remand for additional evidence) and may also elevate the importance of circuit law on what counts as “persecution” as a matter of pure law—because once the legal standard is set, its application is more insulated.

4. Complex Concepts Simplified

“Substantial-evidence review”: A highly deferential appellate standard. The court does not ask whether it would reach the same result; it asks whether a reasonable adjudicator could reach the agency’s result on the record. Reversal is reserved for outcomes the record compels the court to reject.
“Persecution”: Not defined with mathematical precision in the INA; it is a legal threshold applied to facts (threats, violence, detention, severe discrimination, etc.). Different circuits have developed formulations, such as the First Circuit’s approach to death threats referenced via Bonilla v. Mukasey.
“Mixed question of law and fact”: A decision that applies a legal standard to established facts (e.g., whether conduct is “persecution”). This case holds that, even if “persecution” is mixed, Congress can require deferential review of that application.
Jurisdiction vs. standard of review: Jurisdiction asks whether a court may hear an issue at all; standard of review asks how much deference the court gives once it can hear it. Wilkinson v. Garland and Guerrero-Lasprilla v. Barr dealt with the former; this case decides the latter.
IIRIRA (1996): A major statutory overhaul that generally tightened immigration enforcement and limited judicial review. The Court reads §1252(b)(4)(B) as part of that restrictive structure and as codifying INS v. Elias-Zacarias.

5. Conclusion

Urias-Orellana v. Bondi establishes a clear rule: courts of appeals must apply substantial-evidence review to the BIA’s determination whether an applicant’s experiences amount to “persecution” under §1101(a)(42)(A), even where the historical facts are not disputed. The Court reaches this result by treating INS v. Elias-Zacarias as controlling in substance and reading IIRIRA’s §1252(b)(4)(B) as codifying that deferential approach.

The decision’s broader significance lies in its reinforcement of deference in immigration adjudication: “persecution” determinations are insulated from appellate reweighing unless the record compels the opposite conclusion. As a practical matter, the ruling shifts decisive battlegrounds toward (i) developing robust evidence before the IJ and (ii) litigating the “purely legal” content of the persecution standard rather than its application in close cases.