Lopez Alfaro v. Blanche

NOT FOR PUBLICATION FILED UNITED STATES COURT OF APPEALS APR 22 2026 MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS FOR THE NINTH CIRCUIT CRISTIAN ANTONIO LOPEZ ALFARO, No. 25-1828 Agency No. Petitioner, A209-142-332 v. MEMORANDUM* TODD BLANCHE, Acting Attorney General, Respondent.

On Petition for Review of an Order of the Board of Immigration Appeals Argued and Submitted February 12, 2026 San Francisco, California Before: N.R. SMITH, NGUYEN, and SANCHEZ, Circuit Judges.

Partial Dissent by Judge N.R. SMITH.

Petitioner Christian Antonio Lopez Alfaro, a native and citizen of El Salvador, petitions for review of a Board of Immigration Appeals (“BIA”) order affirming an immigration judge’s (“IJ”) denial of his applications for asylum and withholding of removal. We have jurisdiction under 8 U.S.C. § 1252. Where, as here, the BIA adopts the IJ’s reasoning and adds its own analysis, we review both agency decisions. Aleman-Belloso v. Bondi, 128 F.4th 1031, 1039 (9th Cir. 2024).

We review the agency’s discretionary denial of a statutorily eligible asylum applicant for abuse of discretion. Andriasian v. INS, 180 F.3d 1033, 1040 (9th Cir. 1999). We review legal questions de novo. Rodriguez-Zuniga v. Garland, 69 F.4th 1012, 1016 (9th Cir. 2023). We grant the petition in part and remand to the BIA for further proceedings.

1. When the agency denies asylum relief on discretionary grounds, it must demonstrate that it has considered and properly weighed all relevant positive and negative factors. Kalubi v. Ashcroft, 364 F.3d 1134, 1140 (9th Cir. 2004); Gulla v. Gonzales, 498 F.3d 911, 916 (9th Cir. 2007). The agency must place particular importance on whether denial of asylum is likely to result in future persecution. Kalubi, 364 F.3d at 1141. In denying Lopez Alfaro’s application for asylum, the agency properly considered positive factors favoring asylum relief, including his past experiences with a homophobic church, his past persecution by gangs, and the risks that LGBTQ men face in El Salvador at the hands of both private and government actors. It weighed these positive factors against Lopez Alfaro’s recent criminal and drug history, including a felony firearms offense, and determined that the negative factors outweighed the positive factors. Accordingly, the agency acted properly within its discretion to deny petitioner’s asylum claim.

The agency misapplied the “more likely than not” standard in denying withholding of removal. The IJ noted that there were 692 documented cases of violence against LGBTQ people in El Salvador between 2015 and 2019. The court reasoned that, when compared to the country’s population of over six million, such a figure “falls far short of establishing” that Lopez Alfaro is likely to face persecution. Under the “more likely than not” standard, however, the agency must assess whether an individual applicant has a greater than fifty percent chance of persecution, not whether more than fifty percent of a particular population or subpopulation is at risk of persecution. See Marcos, 410 F.3d at 1121 (holding that individualized analysis “is required in this circuit,” and failure to conduct such analysis is an error of law). Requiring Lopez Alfaro to demonstrate that more than

The agency failed to acknowledge or incorporate probative evidence of Lopez Alfaro’s significant family ties to an “extremist” church and the church’s current homophobic ideology. As the IJ briefly mentioned, the record contains ample evidence of the church’s past violence toward LGBTQ individuals, including incidents in 2012 and 2013 where church members publicly beat two gay men, tied their feet to a truck, and dragged them through the street until their skin was “peeling.” Lopez Alfaro witnessed this violence and credibly testified that police observed or participated in these incidents without intervening. The agency acknowledged the church’s past violence toward gay men but concluded that this type of violence is not probable in the future. In doing so, the agency failed to address evidence that the church currently maintains “extremist” views toward LGBTQ people, as reflected in the church’s recent sermons and other anti-LGBTQ views, its support for the Bukele government, and its support for the government’s discrimination of LGBTQ people.

The agency likewise did not consider how Lopez Alfaro’s engagement to a transgender woman may put him at increased risk of persecution. Lopez Alfaro presented credible testimony that he fears that his family and the broader community will discover he is gay because of his engagement to his visibly transgender partner. He further testified that the “community’s dedication” to the church may motivate them to expose him and his fiancée to harm. The agency discounted his testimony, concluding that cisgender gay men are not subject to the same level of harm that transgender women in El Salvador face. But it failed to assess how Lopez Alfaro’s association with a transgender partner increases the

Accordingly, we remand to the agency to correctly apply the “more likely than not” standard and to consider evidence of both Lopez Alfaro’s connection to a transgender partner and the current homophobic sentiments of his family’s church.

PETITION DENIED IN PART, GRANTED IN PART, REMANDED.

As an initial matter, remand on this basis is unwarranted because Lopez Alfaro has not challenged the legal standard applied by the agency. Instead, Lopez Alfaro argues that the agency’s decision was not “backed by substantial evidence.”

By addressing an issue not raised by the parties, the majority departs from the proper scope of appellate review. 3 See Greenwood v. F.A.A., 28 F.3d 971, 977 (9th Cir. 1994) (“We will not manufacture arguments for an appellant, and a bare assertion does not preserve a claim....”); Miller v. Fairchild Industries, Inc., 797 F.2d 727, 738 (9th Cir. 1986). Further, even under the unchallenged legal standard, the agency did not legally err.

The agency correctly applied the heightened “more likely than not” standard required for withholding of removal, which demands a significantly greater showing than the “reasonable possibility” standard governing asylum. See Tamang v. Holder, 598 F.3d 1083, 1091 (9th Cir. 2010). The IJ outlined several reasons to support her conclusion that “there are multiple plausible scenarios in which [Lopez Alfaro] could be the victim of harm or discrimination on account of his LGBTQ identity in El Salvador, but the record demonstrates a relatively low possibility for each individual source of potential harm.” The IJ then considered all the risks cumulatively and determined that they did not meet the standard. Contrary to the majority’s reading of the decision, the IJ did not require proof that a majority of the population faces persecution. Rather, the IJ relied on the general country- conditions evidence to conclude that anecdotal and generalized evidence does not support a finding that Lopez Alfaro personally would more likely than not be targeted. Rather than considering the IJ’s opinion as a whole, the majority takes one of the IJ’s citations to the country-conditions evidence (submitted by Lopez Alfaro) out of context. The IJ did not suggest that Lopez Alfaro would have to demonstrate that over fifty percent of the country’s population or LGBTQ population is likely to face harm. Instead, the IJ reasoned that there was a low number of documented cases of abuse against LGBTQ people while taking into consideration “only a subset of the population is LGBTQ and that many crimes go unreported.” Read in context, the decision reflects an individualized and cumulative analysis that properly determines that, Lopez Alfaro (in his specific situation) would not “more likely than not” be targeted. See Singh v. Holder, 753 F.3d 826, 834–35 (9th Cir. 2014); Marcos v. Gonzales, 410 F.3d 1112, 1121 (9th Cir. 2005).

Additionally, the majority incorrectly concludes that the agency legally erred by failing to consider all probative evidence of Lopez Alfaro’s individualized risk of harm. The record shows that the agency explicitly addressed Lopez Alfaro’s family ties to a church with a homophobic ideology and relationship with his transgender, United States citizen partner. In doing so, the agency acknowledged that harm was possible but found that the likelihood fell short of the required probability. That the IJ weighed the evidence differently than the majority would have, and subsequently made different findings, does not warrant the conclusion that the agency failed to consider all probative evidence, especially given our presumption that the agency reviews all the evidence in the record. See Larita- Martinez v. I.N.S., 220 F.3d 1092, 1095–96 (9th Cir. 2000); Rodriguez-Jimenez v. Garland, 20 F.4th 434, 439 (9th Cir. 2021) (concluding that we do not overturn a decision by the agency based on a “mere disagreement” about how the facts should be evaluated).

2. The proper question for our consideration is whether substantial evidence supports the agency’s denial of withholding of removal. Even though the majority does not address this issue, I do so here. Under substantial evidence review, the IJ’s findings are conclusive unless the record compels the conclusion that persecution is more likely than not. See Manzano v. Garland, 104 F.4th 1202, 1204 (9th Cir. 2024). This demanding standard is not met here. Homophobic beliefs alone do not compel a finding that Lopez Alfaro would more likely than not face persecution. Likewise, generalized evidence of violence and speculative theories about future harm without more do not compel the conclusion that Lopez Alfaro personally would more likely than not face persecution. Even if alternative inferences could be drawn, the record evidence does not compel a contrary conclusion. See Rodriguez-Zuniga v. Garland, 69 F.4th 1012, 1024 (9th Cir. 2023); Barajas-Romero, 846 F.3d 351, 357 (9th Cir. 2017) (“For purposes of . . . withholding of removal, it is not enough that a person comes from a wretched place, where life will most probably be far worse than if he remains in the United States.”). Accordingly, Lopez Alfaro’s claim that substantial evidence does not support the agency’s determination fails.