Doris Rodriguez-Zuniga v. Merrick Garland ( 2023 )


Menu:
  •                     FOR PUBLICATION
    UNITED STATES COURT OF APPEALS
    FOR THE NINTH CIRCUIT
    DORIS AMANDA RODRIGUEZ-                           No. 19-72024
    ZUNIGA; NELSON GABRIEL
    TOBAR-RODRIGUEZ,                                 Agency Nos.
    Petitioners,                        A209-217-710
    A209-217-711
    v.
    MERRICK B. GARLAND, Attorney                        OPINION
    General,
    Respondent.
    On Petition for Review of an Order of the
    Board of Immigration Appeals
    Argued and Submitted October 20, 2022
    San Francisco, California
    Filed June 7, 2023
    Before: Ronald Lee Gilman,* Consuelo M. Callahan, and
    Lawrence VanDyke, Circuit Judges.
    Opinion by Judge VanDyke;
    Dissent by Judge Gilman
    *
    The Honorable Ronald Lee Gilman, United States Circuit Judge for the
    U.S. Court of Appeals for the Sixth Circuit, sitting by designation.
    2                 RODRIGUEZ-ZUNIGA V. GARLAND
    SUMMARY**
    Immigration
    The panel denied Doris Amanda Rodriguez-Zuniga’s
    and her son Nelson Gabriel Tobar-Rodriguez’s petition for
    review of the Board of Immigration Appeals’ dismissal of
    their appeal of an immigration judge’s denial of asylum and
    related relief.
    Rodriguez-Zuniga testified that she was afraid to return
    to Guatemala because a woman had attempted to rob her
    after she withdrew money from a bank. The woman told
    Rodriguez-Zuniga that she targeted her because Rodriguez-
    Zuniga had family in the United States and a lot of
    money. The woman also threatened that Rodriguez-
    Zuniga’s son would “pay for it” due to Rodriguez-Zuniga’s
    refusal to give her the money. Rodriguez-Zuniga and her
    son asserted that she had suffered past persecution and had a
    well-founded fear of future persecution on account of her
    political opinion of refusing to submit to violence by
    criminal groups or gangs, and their claimed membership in
    three particular social groups: “Guatemalan families that
    lack an immediate family male protector,” “Guatemalan
    women,” and “immediate family members of Doris Amanda
    Rodriguez-Zuniga.”
    Because the record did not compel the conclusion that
    Guatemalan society perceived it as a distinct group, the panel
    held that Rodriguez-Zuniga failed to show that the agency
    erred in concluding that her proposed social group
    **
    This summary constitutes no part of the opinion of the court. It has
    been prepared by court staff for the convenience of the reader.
    RODRIGUEZ-ZUNIGA V. GARLAND                 3
    comprised of “Guatemalan families that lack an immediate
    family male protector” was not cognizable. The panel also
    concluded that substantial evidence supported the agency’s
    determination that Rodriguez-Zuniga had not expressed a
    political opinion. The panel explained that Rodriguez-
    Zuniga’s refusal to give money to the threatening robber was
    not evidence of a “conscious and deliberate” decision that
    would naturally result in attributing a political position to
    her, and that she instead simply reacted to being robbed.
    Absent some evidence that Rodriguez-Zuniga expressed a
    political opinion beyond merely her resistance to being
    robbed, the panel concluded that the agency did not err in
    determining that she failed to establish nexus to a political
    opinion.
    Turning to Rodriguez-Zuniga’s family social group
    claim, the panel concluded that the murder of Rodriguez-
    Zuniga’s cousin and her cousin’s son because they refused
    to pay the gangs did not compel any conclusion about the
    robber’s motivation in Rodriguez-Zuniga’s case. The panel
    further concluded that Rodriguez-Zuniga failed to establish
    a nexus to her family membership based on the robber’s
    threats to her son to get her to pay money. The panel wrote
    that to establish a nexus between her family membership and
    her harm, Rodriguez-Zuniga had to show that her family
    membership was a reason motivating the robber to target
    her. The panel explained that where the record indicates that
    the persecutor’s actual motivation for threatening a person is
    to extort money from a third person, the record does not
    compel finding that the persecutor threatened the target
    because of a protected characteristic such as family relation.
    Even assuming for the sake of argument that the harm
    against Rodriguez-Zuniga’s son could support her own
    asylum claim, the panel held that Rodriguez-Zuniga failed to
    4                RODRIGUEZ-ZUNIGA V. GARLAND
    make the required showing of nexus. The panel explained
    that substantial evidence supported the agency’s finding that
    the robber threatened Rodriguez-Zuniga’s son only as an
    instrumental means to obtain money, and that the robber was
    not motivated intrinsically by the son’s familial relationship
    to Rodriguez-Zuniga. Rather, the robber targeted the son for
    the same reason she would target, for example, the
    petitioner’s life-long friend if the opportunity arose—merely
    because she thought Rodriguez-Zuniga cared about that
    person and thus the robber could use threats against that
    person as a means of obtaining money from Rodriguez-
    Zuniga. The panel explained that the extorted person’s
    motivation to give the money because they care for their
    family member does not transform the persecutor’s
    motivation from money to actual animus against a protected
    characteristic.
    The panel also rejected Rodriguez-Zuniga’s “extortion-
    plus” claim under Ayala v. Sessions, 
    855 F.3d 1012
     (9th Cir.
    2017). The panel explained that “extortion-plus” is simply
    the recognition that a persecutor can hold multiple motives
    for harming someone. However, unlike in Ayala, in this case
    the agency made no erroneous legal conclusion that
    extortion could not constitute persecution regardless of other
    motives. Instead, the agency expressly concluded there were
    no other such motives.
    The panel held that the remainder of Rodriguez-Zuniga’s
    evidence from country conditions reports did not compel the
    conclusion that Rodriguez-Zuniga established an objectively
    reasonable fear of future persecution. While recognizing
    that there is no categorical rule that the failure to establish a
    nexus for past persecution forecloses nexus for future
    persecution, the panel did not read the IJ’s decision as
    necessarily applying such a categorical rule. The panel
    RODRIGUEZ-ZUNIGA V. GARLAND                  5
    wrote that even if the IJ erred in applying some categorial
    rule, the BIA did not, and instead addressed them as two
    independent inquiries.
    Finally, the panel concluded that by failing to adequately
    address the issue in her opening brief, Rodriguez-Zuniga had
    forfeited any argument that the agency incorrectly found she
    would not suffer torture with the consent or acquiescence of
    the government. Because a petitioner can state a claim for
    CAT relief only if she shows that the government would
    acquiesce in her torture, the panel concluded that Rodriguez-
    Zuniga’s failure to contest this point was fatal to her claim.
    Dissenting, Judge Gilman disagreed with the majority’s
    conclusion that petitioners failed to establish nexus based on
    their family membership. Because Nelson’s would-be
    persecutors were interested in him only because of his
    relationship to his mother, in Judge Gilman’s view, he
    satisfied both the “a reason” nexus standard for withholding
    of removal and the “one central reason” nexus standard for
    asylum.
    Judge Gilman also wrote that the majority’s nexus
    holding with respect to Rodriguez-Zuniga conflicted with
    this court’s decision in Ayala. Judge Gilman wrote that
    nowhere in Ayala does the court suggest that a showing of
    “animus” on the part of the persecutor is necessary. Judge
    Gilman also explained that, unlike in Ayala, in this case there
    was no ambiguity as to why Rodriguez-Zuniga was
    targeted. Rodriguez Zuniga’s potential persecutors knew
    her identity and the identities of her family members, and
    their representative targeted Rodriguez-Zuniga using her
    relationship to her son and because of her relationship to her
    husband. Additionally, Judge Gilman wrote that by
    eliminating a petitioner’s ability to establish a nexus to a
    6               RODRIGUEZ-ZUNIGA V. GARLAND
    protected ground where the persecutor’s actual motivation
    for threatening a person is to extort money from a third
    person, the majority departs from this court’s precedents that
    affirm the principle that people, including persecutors, often
    have mixed motives.
    Judge Gilman wrote that the majority also failed to hold
    the agency accountable for several procedural missteps,
    including the agency’s failure to independently analyze the
    likelihood that petitioners would be subjected to future harm,
    failure to properly consider Rodriguez-Zuniga’s political
    opinion claim, failure to consider all evidence relevant to the
    possibility of future torture, including based on her gender,
    and the agency’s application of too narrow a government
    acquiescence standard. Judge Gilman also disagreed that
    petitioners had forfeited the government acquiescence issue.
    COUNSEL
    Roger Sismaet Green (argued) and Jenny Tsai, Green &
    Tsai, San Francisco, California, for Petitioners.
    Timothy B. Stanton (argued) and Kristen H. Blosser, Trial
    Attorneys; Sabatino F. Leo, Senior Litigation Counsel;
    Joseph H. Hunt, Assistant Attorney General, Civil Division;
    Office of Immigration Litigation, United States Department
    of Justice; Washington, D.C.; for Respondent.
    RODRIGUEZ-ZUNIGA V. GARLAND                  7
    OPINION
    VANDYKE, Circuit Judge:
    Doris Amanda Rodriguez-Zuniga petitions for review of
    the decision of the Board of Immigration Appeals (BIA)
    dismissing her appeal from the removal order of the
    Immigration Judge (IJ). The heart of Rodriguez-Zuniga’s
    petition is her fear that, because she experienced an
    attempted robbery in her native country, she will be subject
    to persecution in the future. But fear of generalized crime is
    not a sufficient basis for asylum or withholding of removal,
    nor do her other arguments show that she is entitled to relief.
    We have jurisdiction under 
    8 U.S.C. § 1252
     and deny her
    petition.
    I.     BACKGROUND
    In June 2016, Rodriguez-Zuniga and her son, Nelson
    Gabriel Tobar-Rodriguez, entered the United States without
    valid entry documents. They are both citizens and natives of
    Guatemala. Soon after, the United States initiated removal
    proceedings against Rodriguez-Zuniga. She was charged
    with being an “immigrant who, at the time of application for
    admission, [was] not in possession of a … valid entry
    document.”
    Rodriguez-Zuniga conceded both the allegations against
    her and removability, but applied for asylum, withholding of
    removal, and protection under the Convention Against
    Torture (CAT). She claimed that she was entitled to asylum
    because she “suffered past persecution and ha[d] a well-
    founded fear of future persecution based upon her political
    opinion and membership in a particular social group.” Her
    purported political opinion was her “refusal to submit to
    8               RODRIGUEZ-ZUNIGA V. GARLAND
    violence by criminal groups/gangs.” She claimed to be a
    member of three particular social groups: “Guatemalan
    families that lack an immediate family male protector,”
    “Guatemalan women,” and “immediate family members of
    Doris Amanda Rodriguez-Zuniga.” She alleged she suffered
    past persecution generally because of the “bad state of gang
    affairs” in Guatemala and, more specifically, because a
    woman had attempted to rob her.
    Rodriguez-Zuniga also argued that she had a well-
    founded fear of future persecution because of her past
    persecution, the murder of her cousin and her cousin’s child,
    and the generally “precarious current state of Guatemala.”
    Rodriguez-Zuniga finally claimed that she was entitled to
    withholding of removal and CAT relief for the same reasons
    she was entitled to asylum.
    At her hearing before the IJ, Rodriguez-Zuniga testified
    that she was afraid to return to Guatemala because a woman
    had attempted to rob her when she lived there. In March
    2016, just a few months before she entered the United States,
    Rodriguez-Zuniga had gone to the bank to withdraw $150,
    money that her husband in the United States “had sent [her].”
    At that time, Rodriguez-Zuniga regularly received money
    from her husband. As she exited the bank, a woman
    threatened Rodriguez-Zuniga if she did not give her the
    money. Rodriguez-Zuniga believed the woman “belonged
    to a group of … gangs or a group who harms people.” The
    woman told Rodriguez-Zuniga that if she did not give her
    the money, she was going to hurt Rodriguez-Zuniga or her
    son. Rodriguez-Zuniga refused, and the woman told her that
    Rodriguez-Zuniga’s “son was going to pay for it and that she
    was going to come and find [Rodriguez-Zuniga] again.” The
    woman told Rodriguez-Zuniga that she targeted her because
    Rodriguez-Zuniga’s “family was [in the United States] and
    RODRIGUEZ-ZUNIGA V. GARLAND                 9
    [because she] had a lot of money that [she] could give them.”
    Rodriguez-Zuniga never saw the woman again.
    Rodriguez-Zuniga also told the IJ about her female
    cousin who had been killed by gangs when “she refused to
    give them money.” Her cousin’s son was also killed, but
    Rodriguez-Zuniga did not know whether it “was the gang
    members that killed him.”
    The IJ found that Rodriguez-Zuniga “testified credibly
    and accord[ed] her testimony full evidentiary weight,” but
    denied relief. As for asylum and withholding, the IJ found
    that Rodriguez-Zuniga’s past harms and fear of future harms
    lacked the requisite nexus to her statutorily protected
    grounds. The IJ rejected Rodriguez-Zuniga’s proposed
    particular social group of “Guatemalan families that lack an
    immediate family male protector.” The IJ found that
    Rodriguez-Zuniga’s family was a particular social group
    because “family relationships are generally ‘easily
    recognizable and understood by others to constitute social
    groups.’” The IJ also found that Guatemalan women were a
    particular social group because of the “high level of violence
    committed against Guatemalan women” and their “need
    [for] specialized protection” indicated they are viewed as a
    distinct group as compared to the general population in
    Guatemala. But the IJ found no nexus between the harm
    Rodriguez-Zuniga suffered and her membership in either
    cognizable particular social group, observing that the female
    robber “did not mention [Rodriguez-Zuniga’s] gender or her
    lack of a ‘male protector’ in the family; rather, the
    perpetrator seemed to only want money.” And because she
    presented no additional evidence for her feared future
    persecution beyond her nexus evidence for her past
    persecution, the IJ found that Rodriguez-Zuniga likewise
    lacked a nexus for her feared future harm. Finally, the IJ
    10             RODRIGUEZ-ZUNIGA V. GARLAND
    denied Rodriguez-Zuniga’s claim for CAT relief because
    she did not establish that it was “more likely than not she
    will be tortured by or with the acquiescence of the
    government if [she] returned to Guatemala.”
    Rodriguez-Zuniga appealed to the BIA, and the BIA
    dismissed her appeal. It first “adopt[ed] and affirm[ed] the
    decision of the Immigration Judge for the reasons stated [in
    the IJ’s opinion]” before explaining the its own additional
    reasons for dismissing the appeal. As for asylum and
    withholding, the BIA agreed that one of Rodriguez-Zuniga’s
    proposed social groups was not cognizable and that she
    failed to establish a nexus between her past harm or feared
    future harm and “either her family membership[] or her
    status as a Guatemalan woman.” The BIA rejected
    Rodriguez-Zuniga’s claim that she was persecuted because
    of her political opinion, concluding that “there [was] no
    evidence she ever expressed a political opinion.” It also
    rejected her CAT claim because Rodriguez-Zuniga did not
    show that she was sufficiently likely to suffer torture with
    the consent of the government.
    II.    STANDARD OF REVIEW
    “Where, as here, the BIA agrees with the IJ’s reasoning,
    we review both decisions.” See Garcia-Martinez v.
    Sessions, 
    886 F.3d 1291
    , 1293 (9th Cir. 2018). We review
    the agency’s decision under the highly deferential
    substantial evidence standard. Ruiz-Colmenares v. Garland,
    
    25 F.4th 742
    , 748 (9th Cir. 2022). Under that standard, the
    agency’s findings of fact are considered “conclusive unless
    any reasonable adjudicator would be compelled to conclude
    to the contrary.” 
    Id.
     We review questions of law de novo.
    
    Id.
    RODRIGUEZ-ZUNIGA V. GARLAND                     11
    III.      DISCUSSION
    The agency did not err in denying Rodriguez-Zuniga’s
    asylum and withholding claims.1 Her arguments do not
    show that the agency erred in rejecting one of her proposed
    particularized social groups or in concluding that she failed
    to present evidence that she expressed a political opinion.
    And substantial evidence supports the agency finding of no
    nexus between Rodriguez-Zuniga’s membership in the
    particularized social groups that the agency accepted as
    cognizable and any harm she experienced or feared.
    Substantial evidence also supports the agency’s finding that
    Rodriguez-Zuniga did not establish it was more likely than
    not that the government would torture her upon her return to
    Guatemala. We therefore deny her petition for review.
    a. The Agency Did Not Err in Denying Rodriguez-
    Zuniga’s Asylum and Withholding of Removal
    Claims.
    For both asylum and withholding claims, a petitioner
    must prove a causal nexus between one of her statutorily
    protected characteristics and either her past harm or her
    objectively tenable fear of future harm. See Garcia v.
    Wilkinson, 
    988 F.3d 1136
    , 1143 (9th Cir. 2021) (asylum);
    Flores-Vega v. Barr, 
    932 F.3d 878
    , 886–87 (9th Cir. 2019)
    (withholding). These statutorily protected characteristics
    include “race, religion, nationality, membership in a
    particular social group, [and] political opinion.” 
    8 U.S.C. § 1158
    (b)(1)(B)(i) (asylum); see also 
    8 U.S.C. § 1231
    (b)(3)(A) (withholding).
    1
    Nelson, Rodriguez-Zuniga’s son, is a derivative beneficiary on this
    petition.
    12              RODRIGUEZ-ZUNIGA V. GARLAND
    The agency denied asylum and withholding relief
    because Rodriguez-Zuniga failed to make a showing of
    nexus for either her past harm or feared future harms.
    Rodriguez-Zuniga contends that the agency erred in
    numerous respects: first, by rejecting one of her proposed
    particular social groups; second, in finding that she never
    expressed a political opinion; and third, in finding that there
    was no nexus between Rodriguez-Zuniga’s cognizable
    protected social groups—family membership and
    “Guatemalan women”—and either her past harms or feared
    future harms.
    i.   Rodriguez-Zuniga does not show that the
    agency erred in rejecting her proposed
    particular social group.
    The agency concluded that Rodriguez-Zuniga’s
    proposed particular social group of “Guatemalan families
    that lack an immediate family male protector” was not
    cognizable because the evidence did not establish that such
    families “are perceived as a group by society.” “The
    [agency’s] conclusion regarding social distinction—whether
    there is evidence that a specific society recognizes a social
    group—is a question of fact that we review for substantial
    evidence.” Conde Quevedo v. Barr, 
    947 F.3d 1238
    , 1242
    (9th Cir. 2020). Rodriguez-Zuniga fails to show that the
    record compels the conclusion that her proposed group is
    “perceived as a group by society.”
    Rodriguez-Zuniga contends that the agency failed to
    recognize that the group it rejected is built upon the
    foundations of the groups that were previously accepted.
    Although somewhat unclear, she appears to contend that the
    same evidence that makes both her family—which lacks an
    immediate male protector—and “Guatemalan women”
    socially distinct necessarily renders every family without an
    RODRIGUEZ-ZUNIGA V. GARLAND                 13
    immediate male protector socially distinct. But the evidence
    the agency relied on to find these two other groups
    cognizable—the “easily recognizable” nature of family units
    and people’s general understanding that they “constitute
    social groups,” as well as the recognition of Guatemalan
    women’s need for special protection—doesn’t compel the
    conclusion that families without an “immediate family male
    protector” are separately perceived as “set apart, or distinct,
    from other persons within a society.” Diaz-Reynoso v. Barr,
    
    968 F.3d 1070
    , 1077 (9th Cir. 2020).
    The three groups are different, and the agency could
    reasonably find that the same evidence doesn’t render all
    three groups distinct. Beyond this failed argument,
    Rodriguez-Zuniga points to no record evidence indicating
    that “families that lack an immediate family male protector”
    is perceived as a distinct group. She thus fails to show that
    the agency lacked substantial evidence in rejecting her
    membership in “families that lack an immediate family male
    protector” or that the record compels a different conclusion.
    ii.   Substantial evidence supports the agency’s
    conclusion that Rodriguez-Zuniga did not
    present evidence that she expressed a political
    opinion.
    Rodriguez-Zuniga claims she was persecuted because of
    her political opinion, which she frames as her refusal to
    “submit to violence by criminal groups/gangs.” The agency
    below rejected her claim, concluding that she presented no
    evidence “she ever expressed a political opinion.”
    Rodriguez-Zuniga argues that the agency erred because she
    presented evidence of a political opinion when she testified
    to the agency regarding her refusal to give the female robber
    the demanded money. But because that act was not a
    14               RODRIGUEZ-ZUNIGA V. GARLAND
    “sufficiently conscious and deliberate” expression of a
    political opinion, our precedents make clear that it cannot
    support Rodriguez-Zuniga’s claim. De Valle v. INS, 
    901 F.2d 787
    , 791 (9th Cir. 1990) (citation omitted).
    A person’s deeds express a political opinion only when
    they are “‘sufficiently conscious and deliberate’ decisions or
    acts” such that society would naturally “attribute[] certain
    political opinions to [the petitioner]” based on those acts. 
    Id.
    (citation omitted). Rodriguez-Zuniga’s refusal to give
    money to the threatening robber is not evidence of a
    “conscious and deliberate” decision that would naturally
    result in attributing a political position to her. 
    Id.
     Rather,
    she simply reacted to being robbed. If merely resisting a
    robbery could constitute expressing a political opinion, then
    every person who avoided being the victim of a crime could
    seek asylum. But most people who resist criminal activity
    directed towards them do so for obvious non-political self-
    interested reasons—they don’t want to be the victim of a
    crime. See Santos-Lemus v. Mukasey, 
    542 F.3d 738
    , 747
    (9th Cir. 2008) (relying on the fact that “opposition to [a]
    gang’s criminal activity” is not necessarily “based on
    political opinion”), abrogated on other grounds by
    Henriquez-Rivas v. Holder, 
    707 F.3d 1081
     (9th Cir. 2013).
    Nor is Rodriguez-Zuniga’s mere unwillingness “to
    cooperate with a potential persecutor … necessarily
    expressive conduct constituting a political opinion.” Chen
    v. INS, 
    95 F.3d 801
    , 806 (9th Cir. 1996) (citing INS v. Elias-
    Zacarias, 
    502 U.S. 478
    , 480–83 (1992)); see also Regalado-
    Escobar v. Holder, 
    717 F.3d 724
    , 730 (9th Cir. 2013)
    (explaining that “fail[ing] to cooperate in [a political party’s]
    recruitment efforts” does not necessarily express “principled
    opposition to [the political party] or its violence”).
    RODRIGUEZ-ZUNIGA V. GARLAND               15
    Rodriguez-Zuniga presents no other evidence in support
    of her political opinion argument, nor does she present
    additional evidence as to any nexus between her purported
    political opinion and her harm. Absent some evidence that
    Rodriguez-Zuniga expressed a political opinion beyond
    merely her resistance to being robbed, the record does not
    require the conclusion the agency erred. See Garcia-Milian
    v. Holder, 
    755 F.3d 1026
    , 1032 (9th Cir. 2014) (concluding
    that the record “does not compel” a different conclusion than
    the agency’s when the petitioner “provided no evidence” on
    the point). Substantial evidence thus supports the agency’s
    rejection of Rodriguez-Zuniga’s claim for asylum and
    withholding on the basis of a political opinion. See
    Rodriguez Tornes v. Garland, 
    993 F.3d 743
    , 752 (9th Cir.
    2021) (requiring a showing of nexus for both asylum and
    withholding claims based on political opinion).
    iii.   Substantial evidence supports the agency’s
    finding that Rodriguez-Zuniga’s harms
    lacked a nexus to a protected characteristic.
    For both her asylum and withholding claims,
    Rodriguez-Zuniga must show a nexus between her past
    harms or feared future harm and her statutorily protected
    characteristics. See Garcia, 988 F.3d at 1143 (asylum);
    Flores-Vega, 932 F.3d at 887 (withholding). For asylum,
    she must provide evidence showing that her protected
    characteristics were “one central reason” for either her past
    harms or her feared future harms.                 
    8 U.S.C. § 1158
    (b)(1)(B)(i). For withholding, she must provide
    evidence showing it is more likely than not that her life or
    freedom will be threatened, consisting in part of evidence
    indicating that her protected characteristics will be “a
    reason” for her suffering harm in the future. See 
    8 U.S.C. § 1231
    (b)(3)(A); Garcia, 988 F.3d at 1146. The reasons
    16              RODRIGUEZ-ZUNIGA V. GARLAND
    needed to prove a nexus refer to the persecutor’s
    motivations for persecuting the petitioner.        See id.
    “Because a persecutor’s actual motive is a matter of fact,
    we review that finding for substantial evidence.” Vasquez-
    Rodriguez v. Garland, 
    7 F.4th 888
    , 893 (9th Cir. 2021)
    (cleaned up).
    The agency found Rodriguez-Zuniga had failed to
    establish any nexus whatsoever between past harm she
    suffered and either of her proposed social groups. Circuit
    precedent requires a lower nexus showing for withholding of
    removal than asylum, requiring only “a reason” for
    withholding of removal as compared to a “central reason”
    for asylum. Barajas-Romero v. Lynch, 
    846 F.3d 351
    , 360
    (9th Cir. 2017). So it is possible that a petitioner who failed
    to show a sufficient nexus for asylum might nonetheless
    meet the lower nexus requirement for withholding of
    removal. But where, as here, the agency concludes that the
    petitioner has not shown any nexus whatsoever, then the
    petitioner fails to establish past persecution for both asylum
    and withholding. See 
    id.
     (observing there is “no distinction
    between the ‘one central reason’ phrase in the asylum statute
    and the ‘a reason’ phrase in the withholding statute …
    [where] there was no nexus at all” (quoting Zetino v. Holder,
    
    622 F.3d 1007
    , 1016 (9th Cir. 2010) (emphasis added)).
    The sole harm that Rodriguez-Zuniga contends
    amounted to past persecution was the female robber’s threat
    against her and her son. She claims the robber was
    motivated to make this threat by Rodriguez-Zuniga’s
    membership in two protected groups: Rodriguez-Zuniga’s
    family and “Guatemalan women.” The agency found that
    the robber was not motivated by her membership in either of
    those groups, but was instead solely motivated by money.
    RODRIGUEZ-ZUNIGA V. GARLAND               17
    The agency first found that the woman did not target
    Rodriguez-Zuniga because of her status as a “Guatemalan
    woman,” noting that the woman “did not mention
    [Rodriguez-Zuniga’s] gender” at all during the attempted
    robbery. Rodriguez-Zuniga contends this was an error, but
    the only evidence she references is that “Guatemala ha[s] a
    notorious record in lacking in Guatemalan female
    protection.” The general vulnerability of women in
    Guatemala tells us nothing about the female robber’s
    particular motivations, and certainly does not compel the
    conclusion that the robber threatened Rodriguez-Zuniga
    because she is a woman. See Barajas-Romero, 
    846 F.3d at 357
     (explaining that “the persecutor’s motive” is what
    matters for nexus); cf. Lalayan v. Garland, 
    4 F.4th 822
    , 840
    (9th Cir. 2021) (explaining that while “country reports and
    news articles” indicate problems in the petitioner’s home
    country, “they in no way establish that [he] would ‘more
    likely than not’ be persecuted upon removal … on account
    of his [protected ground]”).
    The agency next found that the woman did not threaten
    Rodriguez-Zuniga’s son because of or on account of his
    kinship to Rodriguez-Zuniga. Instead, she “appeared to be
    motivated exclusively by monetary interest.” Substantial
    evidence supports the agency’s nexus finding because
    Rodriguez-Zuniga “did not demonstrate that the gang
    members who sought to extort money from [her] … were
    motivated by anything other than an economic interest.”
    Iraheta-Osorio v. Holder, 
    445 F. App’x 8
    , 9 (9th Cir. 2011)
    (citing Parussimova v. Mukasey, 
    555 F.3d 734
    , 740 (9th Cir.
    2009)).
    Rodriguez-Zuniga contends that the agency erred in
    finding no nexus between the threat and her family
    membership. She relies on the fact that gangs murdered her
    18                 RODRIGUEZ-ZUNIGA V. GARLAND
    cousin and her cousin’s son because they refused to pay the
    gangs, but this does not compel any conclusion about the
    robber’s motivation in Rodriguez-Zuniga’s case. See
    Tamang v. Holder, 
    598 F.3d 1083
    , 1094 (9th Cir. 2010)
    (explaining that “vague threats made against [the
    petitioner’s] family” did not compel the conclusion that the
    petitioner’s “perceived fear of future persecution
    is … objectively reasonable”).
    The heart of Rodriguez-Zuniga’s argument, instead, is
    that the woman threatened her son to get her to pay money,
    which “displays the gangs’ specifically targeting … a
    specific family member to get petitioner to comply.” To
    establish a nexus between her family membership and her
    harm, Rodriguez-Zuniga must show that her family
    membership was a reason motivating the robber to target her.
    See Barajas-Romero, 
    846 F.3d at 357
    . Where the record
    indicates that the persecutor’s actual motivation for
    threatening a person is to extort money from a third person,
    the record does not compel finding that the persecutor
    threatened the target because of a protected characteristic
    such as family relation.2 See Baballah v. Ashcroft, 
    367 F.3d 2
     The dissent laments our use of the modifier “actual” before motivation,
    criticizing our failure to recognize that a petitioner may establish asylum
    or withholding even when the persecutor holds multiple or mixed
    motivations. We of course agree that a petitioner may be entitled to relief
    when a persecutor holds multiple or mixed motivations, and one of those
    motivations is that the persecutor actually wants to harm the petitioner
    based on her protected characteristic. We do not deny that basic
    principle in emphasizing that the motivation must be “actual.” Instead,
    the modifier “actual” distinguishes between where a protected
    characteristic intrinsically motivates the persecutor to harm a victim—
    for example, because the persecutor has animus towards people who
    profess a certain religion—and where a protected characteristic is simply
    RODRIGUEZ-ZUNIGA V. GARLAND                         19
    1067, 1075 n.7 (9th Cir. 2004) (explaining that this court’s
    precedent precludes relief when persecution is “solely on
    account of an economic motive”); Zetino, 
    622 F.3d at 1016
    (“An alien’s desire to be free from harassment by criminals
    motivated by theft or random violence by gang members
    bears no nexus to a protected ground.”); see also Sanchez v.
    Sessions, 
    706 F. App’x 897
    , 899 (9th Cir. 2017) (rejecting a
    nexus because threats to the petitioner’s family were only a
    means of obtaining desired information, not “because of the
    family relationship per se”). In such a situation, the extorted
    person may be motivated to give the money because they
    care for their family member, but that doesn’t transform the
    persecutor’s motivation from money to actual animus
    against a protected characteristic.
    an instrumentality for the persecutor to accomplish his goals, such as
    here, where there is no evidence the gang member cared at all that
    Rodriguez-Zuniga was related to her son or that her son was related to
    Rodriguez-Zuniga. The dissent similarly asserts that our distinction
    would preclude a petitioner from “establish[ing] that family membership
    is even ‘a reason’ for any potential persecution where financial gain also
    motivates the persecutor.” That worry is unfounded. Our caselaw
    permits someone to establish a nexus when one of the persecutor’s
    motivations is financial so long as another motivation, either the
    “central” reason or simply “a reason” depending on what claim is being
    pressed, is actually motivating the persecutor and is based on a victim’s
    protected characteristic. See, e.g., Baghdasaryan v. Holder, 
    592 F.3d 1018
    , 1025–26 (9th Cir. 2010). It should thus be clear from the foregoing
    that we are not—contrary to the dissent—saying that a victim persecuted
    because of “the petitioner’s family membership” is prevented from
    showing a nexus just because the “persecutor’s motives also contain a
    financial dimension.” (Emphasis added.) But this case does not present
    such mixed motives—substantial evidence supports the agency’s finding
    that here financial motivation was not in addition to a motivation based
    on family membership, but was instead the persecutor’s exclusive
    motivation.
    20                 RODRIGUEZ-ZUNIGA V. GARLAND
    Assuming for the sake of argument that harm against
    Rodriguez-Zuniga’s son could support her own asylum
    claim, Rodriguez-Zuniga has failed to make the required
    showing of nexus. Substantial evidence supports the
    agency’s finding that the robber threatened Rodriguez-
    Zuniga’s son only as an instrumental means to obtain money,
    and was not motivated intrinsically by his familial
    relationship to his mother. Rodriguez-Zuniga repeatedly
    testified that the reason the woman targeted her was because
    she “had a lot of money that [she] could give them.” “[I]f
    [Rodriguez-Zuniga] didn’t give her that money[,] she was
    going to hurt” her or her son. The record in this case thus
    does not compel the finding that the robber’s motivation for
    threatening to hurt Rodriguez-Zuniga’s son was his familial
    relationship to Rodriguez-Zuniga. Rather, the robber
    targeted the son for the same reason she would target, say,
    the petitioner’s life-long friend if the opportunity arose—
    merely because she thought Rodriguez-Zuniga cared about
    that person and thus the robber could use threats against that
    person as a means of obtaining money from Rodriguez-
    Zuniga.3
    Rodriguez-Zuniga attempts to refute this point by citing
    Ayala v. Sessions, where our court stated that “economic
    extortion on the basis of a protected characteristic can
    constitute persecution.” 
    855 F.3d 1012
    , 1020 (9th Cir.
    2017). Rodriguez-Zuniga misreads Ayala. Like this case,
    Ayala involved extortion. But as the court explained, it
    involved an “extortion-plus” claim, that is, a claim that the
    petitioner was independently targeted, not just for money,
    3
    We do not suggest one way or another whether the threat to the son
    could apply to Rodriguez-Zuniga. There is no need to reach that issue in
    this case.
    RODRIGUEZ-ZUNIGA V. GARLAND                     21
    but also because of a protected ground. 
    Id. at 1021
    . An
    “extortion-plus” nexus is simply one instantiation of our
    precedent’s recognition that a persecutor can hold multiple
    motives for harming someone. See Parussimova, 
    555 F.3d at
    739 (citing Borja v. INS, 
    175 F.3d 732
     (9th Cir. 1999) (en
    banc), where our court first recognized an “extortion-plus”
    motive, as an example of a “mixed-motive case[]”). To that
    end, the court in Ayala held that it was “legal error for the IJ
    to hold that extortion could not constitute persecution for the
    purposes of withholding[] where the petitioner’s
    membership in a particular social group … is at least ‘a
    reason’ for the extortion.” Ayala, 
    855 F.3d at 1021
    .
    Our court identified two errors by the agency in Ayala:
    one legal and one factual. The agency’s legal error was
    categorically holding that, if a persecutor is motivated by a
    financial goal, i.e., to extort, he cannot also be motivated by
    a petitioner’s protected characteristic. See 
    id.
     As Ayala
    recognizes, that is incorrect. Logically, a persecutor who
    extorts someone could in theory be motivated not just by the
    prospect of obtaining money but also by a petitioner’s
    protected characteristic. The first error the agency in Ayala
    made was thus holding that a petitioner could never prove a
    nexus when the persecutor extorted the petitioner. See 
    id.
    But it also bears noting that just because an “extortion-plus”
    persecution is possible—someone could be motivated to
    extort a particular person by, say, actual animus towards
    their family—common sense tells us that will often, indeed
    usually, not be the case.4
    4
    The dissent quibbles over our example of an “extortion-plus” claim
    because we use the word “animus.” But our court often uses “animus”
    as an example of the type of “nexus” required for an asylum or
    22                 RODRIGUEZ-ZUNIGA V. GARLAND
    Regardless of the likely infrequency of an “extortion-
    plus” claim, it remains true that, when required to decide
    whether the petitioner established a nexus, the agency must
    determine whether the petitioner showed that a protected
    characteristic motivated the petitioner’s persecutor. See,
    e.g., Garcia, 988 F.3d at 1143–44 (citation omitted) (noting
    that the petitioner must show that a protected characteristic
    motivated the persecutor). The agency’s second error in
    Ayala, its factual error, was its failure to do just that.
    Because the agency held that a petitioner “could not” prove
    a nexus where a persecutor extorts her, the agency
    necessarily never made a factual finding on whether the
    petitioner established her persecutor was motivated by
    anything other than financial ends. Ayala, 
    855 F.3d at 1021
    ;
    see Regalado-Escobar, 
    717 F.3d at
    729–30 (explaining that
    the BIA’s holding that opposition to a political party’s use of
    violence “could not be a political opinion” was an “err[or] as
    a matter of law” and meant that the agency “did not conduct
    the necessary factual inquiry as to whether [the petitioner]
    had a protected political opinion”).
    Given these two errors, our court remanded the case. But
    to be clear, even the panel in Ayala expressed reservations—
    reservations the dissent here seems to share—over whether
    the petitioner would actually succeed before the agency. See
    withholding claim. See, e.g., Sinha v. Holder, 
    564 F.3d 1015
    , 1022 (9th
    Cir. 2009) (faulting the IJ “in his nexus analysis” because a reasonable
    IJ would have found that the attackers “were motivated at least in part
    by racial animus”); Garcia, 988 F.3d at 1145 (noting that, “over time,”
    certain conduct “can demonstrate a kind of animus …. sufficient to
    demonstrate nexus” (citation omitted)). And ultimately, the dissent’s
    aversion to the word animus does not change our main point—Ayala was
    about two errors, one legal and one factual, and neither error was made
    by the agency in this case.
    RODRIGUEZ-ZUNIGA V. GARLAND                23
    Ayala, 
    855 F.3d at 1021
     (“Whatever the merits of her claim,
    it was legal error for the IJ to hold that extortion could not
    constitute persecution ….” (emphasis added)). The reason
    for the court’s doubts is obvious from reading Ayala: the
    petitioner there presented barely any evidence, perhaps none
    at all, that family membership actually motivated her
    persecutor. See 
    id.
     at 1016–17. Thus, even if the record in
    Ayala seems unlikely to compel a finding of nexus, the
    agency’s legal error and its failure to find the persecutor’s
    actual motivations required remand in that case.
    Ultimately, it is the majority’s approach in this case and
    not the dissent’s that is consistent with Ayala. To borrow
    from the dissent’s discussion of the petitioner in Ayala,
    Rodriguez-Zuniga has not “present[ed] the court with factual
    evidence to support her theoretically valid nexus theory.”
    And unlike in Ayala, there was no erroneous legal
    conclusion by the agency in this case that “extortion could
    not constitute persecution” regardless of other motives—the
    underlying error the court faulted the agency for in Ayala.
    
    Id. at 1021
    . Instead, the agency here expressly concluded
    there were no other such motives—precisely what our court
    found missing in Ayala. In short, there is no reason to
    remand in this case because the agency here didn’t make the
    categorical legal mistake it made in Ayala, and it has already
    made the factual conclusion that was missing in Ayala. And
    nothing compels the conclusion that the robber in this case
    was motivated by anything other than underlying economic
    reasons, even though those economic motivations also
    resulted in threats to Rodriguez-Zuniga’s son. See Juarez
    Morales v. Wilkinson, 
    836 F. App’x 603
    , 605 (9th Cir. 2021)
    (explaining that Ayala “is irrelevant …. [where the
    petitioner] offers no evidence that a protected ground was a
    reason for [her] extortion”).
    24              RODRIGUEZ-ZUNIGA V. GARLAND
    As for Rodriguez-Zuniga’s fear of future harm, the
    agency again found no nexus. The BIA explained that she
    “did not establish a nexus between any feared harm and a
    protected ground.” Instead, the BIA noted that her “claim
    [was] based on a fear of general violence and criminal
    activity in Guatemala.” The record compels no different
    conclusion. As explained above, the record supports the
    agency’s finding that the attempted robbery bore no nexus to
    a protected characteristic. The remainder of Rodriguez-
    Zuniga’s evidence is from country conditions reports that do
    not compel the conclusion that her fear of future persecution
    was objectively reasonable.
    The dissent argues that the agency erred in its analysis of
    Rodriguez-Zuniga’s fear of future harms because the IJ
    applied an improper categorical rule to reject the claim. The
    IJ noted that, having found Rodriguez-Zuniga could not
    “claim past persecution on account of a protected ground, it
    necessarily follow[ed] she also cannot claim a well-founded
    fear of future persecution on account of the same ground.”
    The dissent reads the IJ as applying as a categorical rule that
    every time a petitioner fails to establish a nexus for past
    persecution, she would necessarily fail to establish a nexus
    for future persecution. We agree that such a categorical rule
    would not be supported by our precedent, but we do not read
    the IJ as necessarily applying such a categorical rule.
    Instead, the IJ was noting that in this case, because
    Rodriguez-Zuniga offered evidence going only to her past
    persecution, the fact that she failed to establish a nexus for
    past persecution means that the same evidence would not
    show future persecution.
    That’s certainly how the BIA read the IJ, and we need
    not read the decision any differently. The BIA explained
    that Rodriguez-Zuniga “did not establish a nexus between
    RODRIGUEZ-ZUNIGA V. GARLAND                25
    the single incident she experienced,” i.e., her past harm, “or
    the harm she fears,” i.e., her feared future harm, “and either
    her family membership, or her status as a Guatemalan
    woman.” If the IJ erred in applying some categorical rule,
    the BIA did not—it instead addressed them as two
    independent inquiries. And to the extent the BIA and IJ part
    ways, we review the BIA’s findings for substantial evidence.
    In short, substantial evidence supports the agency’s
    finding that Rodriguez-Zuniga’s two protected grounds were
    not “a reason” for her past persecution or feared future
    persecution, necessarily defeating both her asylum and
    withholding claims. We therefore deny Rodriguez-Zuniga’s
    petition for review of her asylum and withholding claims.
    b. Substantial Evidence Supports the Agency’s
    Denial of CAT Relief.
    “For CAT relief, the alien must prove that it is ‘more
    likely than not that he or she would be tortured if removed
    to the proposed country.’” Barajas-Romero, 
    846 F.3d at 361
    . And that “torture must be ‘inflicted by or at the
    instigation of or with the consent or acquiescence of a public
    official or other person acting in an official capacity.’” 
    Id.
    The agency found that Rodriguez-Zuniga had not
    “demonstrate[d] the government of Guatemala would
    consent or acquiesce to her torture.”
    Although she argues that the agency should have granted
    her CAT relief, Rodriguez-Zuniga does not argue that the
    agency erred in finding that she presented insufficient
    evidence that the Guatemalan government would consent to
    her torture. Indeed, the closest Rodriguez-Zuniga comes to
    addressing acquiescence is in her background section, when
    she explains that she did not tell the police about the
    threatening woman “because the police are connected to the
    26               RODRIGUEZ-ZUNIGA V. GARLAND
    gangs.” This reference, found only in her background
    section and not directly connected with any argument, does
    not prevent Rodriguez-Zuniga from forfeiting the argument
    that the agency incorrectly found she would not suffer
    torture with the consent or acquiescence of the government.
    See Johnson v. Baker, 
    23 F.4th 1209
    , 1216 n.3 (9th Cir.
    2022) (considering “abandoned” assertions made only in the
    background section of a brief).
    The dissent does not agree that Rodriguez-Zuniga
    forfeited the argument, but it also does not contend that
    Rodriguez-Zuniga         ever      mentioned          government
    acquiescence, per se. Instead, the dissent notes, quoting her
    briefing, that Rodriguez-Zuniga asserted that she had
    “articulated a specific individualized threat of torture.” The
    dissent extrapolates from this that she must have been
    implicitly arguing that the government would acquiesce in
    her torture if she returned, because acquiescence is a part of
    the definition of torture. But that just substitutes one
    problem for another: even if she meant to implicitly argue
    that the government would acquiesce, she must still
    “specifically and distinctly” raise an argument and support it
    with citations to the record to raise it on appeal. Hayes v.
    Idaho Corr. Ctr., 
    849 F.3d 1204
    , 1213 (9th Cir. 2017);
    accord Greenwood v. FAA, 
    28 F.3d 971
    , 977 (9th Cir. 1994);
    see also United States v. Graf, 
    610 F.3d 1148
    , 1166 (9th Cir.
    2010) (“Arguments made in passing and not supported by
    citations to the record or to case authority are generally
    deemed waived.”). Even assuming there was no need for
    Rodriguez-Zuniga to use the term “acquiescence” or one of
    its cognates to raise the issue, she still failed to “specifically
    and distinctly” raise the argument. See Hayes, 
    849 F.3d at 1213
    .
    RODRIGUEZ-ZUNIGA V. GARLAND                     27
    Because a petitioner can state a claim for CAT relief only
    if she shows that the government would acquiesce in her
    torture, Rodriguez-Zuniga’s failure to contest this point is
    fatal to her claim. See, e.g., Madrigal v. Holder, 
    716 F.3d 499
    , 508 (9th Cir. 2013).5 We therefore deny Rodriguez-
    Zuniga’s petition for review of her CAT claim.
    c. The Dissent’s Remaining Concerns Lack Merit.
    The dissent vigorously disagrees with the foregoing
    analysis—in what seems to be practically every regard. But
    before we respond to those disagreements one-by-one and
    explain why they lack merit, it is worth taking a step back to
    consider the uncontested facts of this case. This is a case
    about an attempted robbery of a woman and her son who,
    upon leaving a bank, were threatened with violence if they
    didn’t hand over some money. That’s it. Relying on that
    unfortunate event, plus the regrettably unenviable conditions
    prevalent in Guatemala, Rodriguez-Zuniga seeks
    immigration relief that Congress made available for refugees
    who, if they are returned to their home country, face a
    particularized risk of persecution because of their status, or
    else face a greater-than-50% risk of being tortured. An oft-
    recognized corollary is that such relief is not available to
    those who have simply had the misfortune of becoming a
    victim of criminal misconduct abroad, motivated by the sorts
    of things (money, generally) that motivate criminals. See,
    e.g., Zetino, 
    622 F.3d at 1016
    . Immigration law can be
    complicated, especially because courts have manufactured a
    byzantine and ever-increasing maze of procedural and
    substantive standards that are difficult for everyone—
    5
    Even if she had argued this point, we would still deny her petition
    because substantial evidence supports the agency’s denial of her CAT
    claim.
    28              RODRIGUEZ-ZUNIGA V. GARLAND
    asylum-seekers, immigration officials, and courts alike—to
    navigate. And much of the discussion that follows relates to
    such arcane requirements. But something has gone terribly
    wrong when judges conclude that relief for persecution and
    torture is mandated just because someone was the victim of
    a brief and failed robbery attempt in their home country. Is
    that really what anyone thinks Congress meant by providing
    relief for refugees?
    i.   The dissent’s concerns regarding our nexus
    holding are unwarranted.
    The dissent offers several criticisms regarding both our
    explanation of this circuit’s precedent on nexus
    determinations and how we apply that precedent in the above
    analysis. None of these concerns are warranted.
    First, the dissent criticizes our statement that the “record
    does not compel finding that the persecutor threatened the
    target because of a protected characteristic such as family
    relation” when “the record indicates that the persecutor’s
    actual motivation for threatening a person is to extort money
    from a third person.” The dissent finds it difficult to identify
    who the “target” is. But the dissent’s demand that we
    precisely define whether the “target” is Rodriguez-Zuniga or
    her son contradicts its later assertion that we must treat
    Rodriguez-Zuniga and her son as one undifferentiated claim.
    In any event, any ambiguity that might persist in the identity
    of the “target” is because we assume that the threat could
    count as a harm to either person, and then decide whether
    there was a nexus between that harm and a protected
    characteristic. The dissent’s assertion that we must
    differentiate between mother and son thus lacks merit.
    Second, the dissent argues that the record compels the
    conclusion that family membership was both “a reason” and
    RODRIGUEZ-ZUNIGA V. GARLAND               29
    “a central reason” petitioner was targeted for attempted
    robbery. The dissent strangely claims that we have, “in
    effect,” concluded that the family membership of Rodriguez-
    Zuniga’s son is a but-for cause of the robber threatening the
    son. That is the opposite of what we’re saying—which is
    that there is zero evidence the robber targeted the son based
    on the son’s family membership per se. The but-for cause of
    the robber targeting the son is not family membership, it is
    that the robber thought Rodriguez-Zuniga cared about her
    son.
    To illustrate with an example, imagine the robber
    attempted to rob a woman who was accompanied, not by her
    son, but by a pet dog. This dog sports an ornately bejeweled
    collar and leash and its fur indicates frequent grooming, so
    the robber can infer the woman cares deeply for her pet. In
    an attempt to rob the woman, the robber might
    understandably threaten the dog. No one would think that
    such a threat was motivated by any animus toward the
    animal. It would instead be obvious that that the but-for
    cause of the robber threatening the dog was the robber’s
    belief that the woman cared about the dog and would give
    the robber money if the dog was threatened. So too here: the
    but-for cause of the robber threatening the son is not his
    family membership but because the robber thought,
    probably correctly, that Rodriguez-Zuniga cared for her son.
    Under the dissent’s view, every kidnapping where a
    kidnapper demands money from the family of the kidnapped
    individual would necessarily establish a nexus. If that’s
    what Congress had wanted, it would have made family
    membership an enumerated category—instead, family
    membership is sometimes, but not always, a particularized
    social group. See Rios v. Lynch, 
    807 F.3d 1123
    , 1128 (9th
    Cir. 2015) (explaining that “family membership may
    30                  RODRIGUEZ-ZUNIGA V. GARLAND
    constitute membership in a ‘particular social group’”
    (emphasis added)).6
    Third, the dissent argues that the nexus inquiry is not
    about “whether the persecutors’ acts were motivated by an
    unprotected characteristic.” The dissent tries to ground this
    innovation in the Supreme Court’s decision in Elias-
    Zacharias, 
    502 U.S. at 478
    . In that case, the purported
    persecution was a guerilla military organization attempting
    to conscript the petitioner into its forces. The Court
    explained that even if the persecutors were motivated
    politically to conscript people because they wanted to fill
    their ranks and “carry on their war against the government,”
    that was “irrelevant.” 
    Id.
     at 481–82. That is, a nexus could
    not be established by a motive “irrelevant” to whether the
    persecutor harmed a victim because of a victim’s protected
    characteristic. See 
    id.
     Instead of relying on Elias-Zacharias,
    the dissent’s argument that the persecutor’s financial
    motivation is irrelevant ignores the plain import of the
    decision. The obvious takeaway is that if, like here, a
    persecutor is motivated exclusively by a consideration
    “irrelevant” to a victim’s protected ground, that motivation
    is emphatically “relevant”—indeed, that motivation is
    6
    The dissent faults us for noting that family membership is not
    necessarily a particularized social group. But it’s unclear where the
    dissent derives its belief that a petitioner need not show a family is
    “defined with particularity” and “socially distinct within the society in
    question.” Rios, 
    807 F.3d at 1127
     (describing the BIA’s “rubric” for
    establishing a particular social group). Our caselaw establishes that
    family membership “may constitute membership in a ‘particular social
    group,’” not that families are automatically a particular social group. 
    Id. at 1128
     (emphasis added) (quotation omitted); see Gonsalez Padilla v.
    Barr, 
    830 F. App’x 182
    , 184 n.2 (9th Cir. 2020) (noting that precedent
    “recognize[s] that ‘family’ could be the basis of a particular social group
    and it [is] error to not even consider it” (citing Rios, 
    807 F.3d at 1128
    )).
    RODRIGUEZ-ZUNIGA V. GARLAND                31
    decisive—for nexus. A persecutor that is exclusively
    motivated by something unrelated to a victim’s protected
    characteristic is, tautologically, not motivated by the
    victim’s protected characteristic.      Thus here, where
    substantial evidence supports the agency’s finding that the
    actual motivation of the persecutors was “exclusively”
    financial, not any protected characteristic, that exclusive
    financial motivation cannot establish a nexus.
    Finally, in criticizing the panel’s nexus conclusion, the
    dissent appears to argue that because the IJ found Rodriguez-
    Zuniga credible it was obligated to also credit as true her
    speculation about her persecutor’s motives for targeting her.
    This argument flows from our court’s now defunct “deemed-
    true-or-credible” rule, where we had required that an IJ find
    true any testimony found to be credible. See Garland v.
    Ming Dai, 
    141 S. Ct. 1669
    , 1676–77 (2021). The Supreme
    Court squarely rejected this rule in Ming Dai. 
    Id. at 1677
    (“The Ninth Circuit’s rule has no proper place in a reviewing
    court’s analysis.”). The IJ was thus free to find Rodriguez-
    Zuniga credible without finding persuasive her subjective
    beliefs about why the robber attempted to extort her.
    The dissent similarly contends that Rodriguez-Zuniga
    did not speculate about her persecutor’s motives because she
    testified the robber stated that the robber targeted her
    because “her family was here and that [she] had a lot of
    money that [she] could give them.” But that contention
    simply repackages the dissent’s error discussed above:
    evidence that a persecutor targeted someone is not
    necessarily evidence of the underlying motivation for doing
    so. To the extent this is evidence of the robber’s motives, it
    is direct evidence that the persecutor targeted Rodriguez-
    Zuniga, not because of family membership, but because her
    32              RODRIGUEZ-ZUNIGA V. GARLAND
    family was in the United States and so they assumed she
    “had a lot of money that [she] could give them.”
    ii.    The dissent fails to show a flaw in the agency’s
    reasoning regarding Rodriguez-Zuniga’s
    political opinion—or lack thereof.
    The dissent argues that the agency erred in rejecting
    Rodriguez-Zuniga’s claim that she was persecuted because
    of her political opinion because the agency failed to offer
    sufficient reasons for its decision and because the BIA
    engaged in impermissible factfinding. Both arguments fail.
    First, the dissent argues that the BIA abused its discretion
    by offering only a “single-sentence resolution of the issue”
    that fails to “explain what factors it has considered or relied
    upon.” Kalubi v. Ashcroft, 
    364 F.3d 1134
    , 1140 (9th Cir.
    2004). It is revealing that the dissent never references the
    content of Rodriguez-Zuniga’s asserted political opinion.
    Her claimed political opinion is her “refusal to submit to
    violence by criminal groups/gangs.” As far as the record
    reveals, she never stated this opinion to the gang member or
    to anyone before the robbery, and the only actions that she
    argues displayed this opinion was her noncompliance with a
    gang member’s extortion attempt.
    Although an agency must give reasons sufficient for us
    to review, the reasons that the agency must offer are
    certainly coextensive with the complexity of the analysis
    required by the issue. See Marcu v. INS, 
    147 F.3d 1078
    ,
    1083 (9th Cir. 1998) (denying a petition because the BIA
    “demonstrate[d] that it heard the claim, considered the
    evidence, and decided against” the petitioner). And here,
    given the complete lack of evidence supporting Rodriguez-
    Zuniga’s political opinion claim, it is hardly surprising the
    agency decided to dispose of it in a sentence. Nor do we
    RODRIGUEZ-ZUNIGA V. GARLAND                         33
    have any difficulty ascertaining why the agency denied the
    claim.
    The dissent complains that we have explained the BIA’s
    resolution of the political opinion claim in too much detail.
    According to the dissent, we have “expand[ed] on the
    BIA’s … rationale for nearly five pages” and are using ex
    post rationales to justify the BIA. But notice what
    explanation the BIA gave and what explanation we give
    now: the BIA rejected the claim because “there is no
    evidence [Rodriguez-Zuniga] ever expressed a political
    opinion.” We deny the petition for review of that claim
    because there was no evidence she ever expressed a political
    opinion. We have not conjured up an ex post facto
    rationalization of the BIA’s decision; we have reviewed for
    substantial evidence the reason expressly provided by the
    BIA.
    Second, the dissent contends that the BIA engaged in
    improper factfinding in rejecting Rodriguez-Zuniga’s
    political claim because the IJ did not express any factual
    findings regarding the claim. But observing the absence of
    evidence is not a factual finding. If it was, then this court of
    appeals would frequently be a factfinder—which, of course,
    it cannot be.7 The BIA never found facts, it simply looked
    at the record and observed the petitioner had provided none.
    7
    See, e.g., Furnace v. Sullivan, 
    705 F.3d 1021
    , 1031 (9th Cir. 2013)
    (affirming a grant of summary judgment against an equal protection
    claim “[b]ecause [the plaintiff] adduce[d] no evidence that he was treated
    differently than any other [comparator]”); Freyd v. Univ. of Oregon, 
    990 F.3d 1211
    , 1229 (9th Cir. 2021) (affirming a grant of summary judgment
    “because [the plaintiff] has presented no evidence of intentional
    discrimination”); Fisher v. City of San Jose, 
    558 F.3d 1069
    , 1085 (9th
    34                 RODRIGUEZ-ZUNIGA V. GARLAND
    IV.      CONCLUSION
    The majority is not unsympathetic to Rodriguez-Zuniga
    and her son’s desire to stay in this country. But all that she
    has provided in support of her petition are country condition
    reports and one failed, non-violent robbery that the agency
    reasonably concluded was wholly economically motivated
    (as robberies usually are). Our legal system understandably
    places primary authority for immigration policy in Congress
    and the executive branch. If we stretched our law to grant
    the petition here, it would be clear that we have substituted
    ourselves for the immigration officials. Because she failed
    to show a nexus between her past or feared future harms and
    any protected grounds, we deny Rodriguez-Zuniga’s petition
    for review of her asylum and withholding of removal claims.
    And because she forfeited any challenge to the agency’s
    finding that she offered insufficient evidence that the
    Guatemalan government would acquiesce or consent to her
    torture, we deny her petition for review of her CAT claim.
    PETITION DENIED.
    Cir. 2009) (reversing the district court’s grant of the plaintiff’s renewed
    motion for judgment as a matter of law on a 
    42 U.S.C. § 1983
     claim
    because “there [was] no evidence that the [defendant] had a
    policy … that was the ‘moving force’ behind any alleged constitutional
    violation”) (en banc); Gilbrook v. City of Westminster, 
    177 F.3d 839
    , 868
    (9th Cir. 1999), as amended on denial of reh’g (July 15, 1999) (reversing
    a Rule 50(a) grant of judgment as a matter of law by concluding, among
    other things, that there was “no evidence in the record suggesting” that
    a factor weighed in favor of affirmance (emphasis added)); see also
    Burley v. Gagacki, 
    729 F.3d 610
    , 620 (6th Cir. 2013) (affirming a grant
    of summary judgment against a “failure-to-intervene theory….
    “[b]ecause no evidence place[d] the state and local defendants inside
    plaintiffs’ home at the appropriate time to witness or respond to any
    unconstitutional conduct that may have occurred”).
    RODRIGUEZ-ZUNIGA V. GARLAND                 35
    GILMAN, Circuit Judge, dissenting:
    INTRODUCTION
    This court has repeatedly emphasized that “the family
    remains the quintessential particular social group.” See
    Parada v. Sessions, 
    902 F.3d 901
    , 910 (9th Cir. 2018)
    (quoting Rios v. Lynch, 
    807 F.3d 1123
    , 1128 (9th Cir.
    2015)). “That is, an asylum-seeker who has suffered
    persecution ‘on account of th[eir] familial relationship’ has
    suffered persecution by reason of membership in a particular
    social group.’” 
    Id.
     (alteration in original) (quoting Rios, 
    807 F.3d at 1128
    ).
    The majority apparently disagrees with the above legal
    principle by stating that “family membership is sometimes,
    but not always, a particularized social group.” See Majority
    Op. at 29. It therefore discounts persecution that occurs by
    reason of the petitioner’s family membership if the
    persecutor’s motives also contain a financial dimension. In
    the majority’s view, such a petitioner would have to provide
    an alternative, “actual” reason for the alleged harm. See id.
    at 18, 29. This places an unjustified burden on those seeking
    relief based on their family membership by harkening back
    to the much-maligned (and now vacated) regime of Matter
    of L-E-A- II, 
    27 I. & N. Dec. 581
     (A.G. 2019), which held
    that, “in the ordinary case, a nuclear family will not, without
    more,” qualify as a particular social group. See 
    id. at 589
    ,
    vacated by Matter of L-E-A- III, 
    28 I. & N. Dec. 304
    (A.G. 2021).
    Congress passed a statute in 1952—the Immigration and
    Nationality Act—that (1) offers a discretionary pathway to
    relief for those who reasonably fear persecution on account
    of their membership in a particular social group, see 8 U.S.C.
    36              RODRIGUEZ-ZUNIGA V. GARLAND
    § 1158(b), and (2) outright prohibits the removal of
    noncitizens to countries where they face a clear probability
    of persecution because of the same, see 
    8 U.S.C. § 1231
    (b)(3). I believe that the majority’s holding is
    inconsistent with this statutory scheme and Ninth Circuit
    precedent.
    A. An inherent contradiction exists in the majority’s
    treatment of the nexus element with respect to
    Rodriguez-Zuniga’s son Nelson because family
    membership cannot be both the primary reason for
    Nelson’s persecution and no reason at all
    The majority holds that, “[w]here the record indicates
    that the persecutor’s actual motivation for threatening a
    person is to extort money from a third person, the record
    does not compel finding that the persecutor threatened the
    target because of a protected characteristic such as family
    relation.”     Majority Op. at 18. This proposition is
    ambiguous—intentionally so, the majority acknowledges,
    see 
    id.
     at 26-27—because it does not make clear whether the
    “target” is the direct target of the extortion (Rodriguez-
    Zuniga and others similarly situated) or the indirect target of
    the threat (Nelson and others similarly situated). In other
    words, the majority’s holding forecloses not only
    Rodriguez-Zuniga’s ability to satisfy the nexus requirement,
    but Nelson’s as well.
    On the one hand, the majority states that “the robber
    targeted [Nelson] for the same reason she would target, say,
    [Rodriguez-Zuniga’s] life-long friend if the opportunity
    arose—merely because she thought Rodriguez-Zuniga cared
    about that person and thus the robber could use threats
    against that person as a means of obtaining money from
    Rodriguez-Zuniga.” 
    Id. at 20
     (emphasis added). This in
    RODRIGUEZ-ZUNIGA V. GARLAND                 37
    effect makes Nelson’s family membership not only a but-for
    cause, but also the primary cause of his being placed in
    harm’s way. On the other hand, the majority inexplicably
    holds that Nelson and those in like circumstances cannot
    satisfy the nexus requirement.        This is inherently
    contradictory and unsupported by precedent.
    The majority claims that it does not mean to say that
    Nelson’s relationship to his mother is a but-for cause of his
    threatened harm—this is apparently “the opposite of what [it
    is saying], 
    id.
     at 29—and yet it then repeats the contradiction
    by positing that “there is zero evidence the robber targeted
    the son based on the son’s family membership per se. The
    but-for-cause of the robber targeting the son is not family
    membership, it is that the robber thought Rodriguez-Zuniga
    cared about her son.” 
    Id. at 29
     (emphasis added). This
    strikes me as double-talk.
    The majority tries once more, this time likening Nelson
    to Rodriguez-Zuniga’s hypothetical pet dog. If the robber
    had threatened to harm the dog, the majority suggests, “the
    but-for cause of the robber threatening the dog was the
    robber’s belief that the woman cared about the dog.” 
    Id. at 29
    . Here again, the majority’s error is clear: the target of
    the threatened harm is a target only because of his
    relationship—in Nelson’s case, his family relationship—to
    another person.
    To satisfy the nexus requirement, an asylum applicant
    must show that a protected characteristic is “one central
    reason” for the feared harm. Garcia v. Wilkinson, 
    988 F.3d 1136
    , 1144 (9th Cir. 2021). That “central reason” may be
    one among many, and “an asylum applicant need not prove
    which reason was dominant” so long as the protected
    characteristic is likely to be “a cause of the persecutors’
    38             RODRIGUEZ-ZUNIGA V. GARLAND
    acts.” 
    Id. at 1144
     (quoting Parussimova v. Mukasey, 
    555 F.3d 734
    , 741 (9th Cir. 2009)). For an applicant seeking
    withholding of removal, an even “weaker motive” will
    suffice: a protected characteristic need only be “a reason”
    for the feared harm. Barajas-Romero v. Lynch, 
    846 F.3d 351
    , 359 (9th Cir. 2017).
    Because Nelson’s would-be persecutors were interested
    in him only because of his relationship to his mother, he
    satisfies both the “a reason” nexus standard for withholding
    of removal and the “one central reason” nexus standard for
    asylum. Yet the majority holds that Nelson and others in his
    position cannot establish that family membership is even “a
    reason” for any potential persecution where financial gain
    also motivates the persecutor.
    The IJ made the same error in holding that Rodriguez-
    Zuniga “presented no evidence that her son was threatened
    on account of his kinship to her,” and that the threat to
    Nelson was instead “motivated exclusively by monetary
    interest.” But the record compels the opposite conclusion:
    that Nelson was targeted, as the majority puts it, “merely
    because [the robber] thought Rodriguez-Zuniga cared”
    about Nelson. Majority Op. at 20. We should therefore
    reverse the agency’s decision for lack of substantial
    supporting evidence.
    I further note that the agency’s disposition of Nelson’s
    claim has a direct effect on the agency’s disposition of
    Rodriguez-Zuniga’s claim. In Tchoukhrova v. Gonzales,
    
    404 F.3d 1181
     (9th Cir. 2005), vacated on other grounds,
    
    549 U.S. 801
     (2006), this court made clear that the
    threatened harm against Nelson can support Rodriguez-
    Zuniga’s own asylum application:
    RODRIGUEZ-ZUNIGA V. GARLAND              39
    [W]hen it is only the child who is the direct
    victim, a narrow interpretation of our asylum
    laws could have devastating practical effects:
    Facing imminent removal, parents could be
    forced to make a choice between abandoning
    their child in the United States or taking him
    to a country where it is likely that he will be
    persecuted.
    ....
    Our precedent supports the pragmatic
    approach applied here by the agency. When
    confronting cases involving persecution of
    multiple family members, we have not
    formalistically divided the claims between
    “principal” and “derivative” applicants but
    instead, without discussion, have simply
    viewed the family as a whole . . . .
    
    Id. at 1191-92
    .
    Thus, should Nelson succeed on his asylum claim,
    Rodriguez-Zuniga herself may also be afforded relief. As I
    explain in the next section, however, this court’s decision in
    Ayala v. Sessions, 
    855 F.3d 1012
     (9th Cir. 2017), supports
    the conclusion that Rodriguez-Zuniga independently
    satisfies the nexus requirement for withholding of removal.
    40              RODRIGUEZ-ZUNIGA V. GARLAND
    B. The majority’s nexus holding with respect to
    Rodriguez-Zuniga conflicts with this court’s decision
    in Ayala, which held that an extortionist’s financial
    motivation does not preclude a nexus finding based
    on family membership
    In Ayala, the petitioner claimed that “she and her
    husband were the subjects of extortion because of his
    family’s ownership of hotels.” 
    855 F.3d at 1020
    . The court
    held that the IJ erred in concluding that “the only motivation
    indicated throughout is extortion” despite Ayala’s testimony
    that she was afraid of being targeted on the basis of her
    marriage to a hotel owner. 
    Id.
     The persecutors’ financial
    motivation was insufficient to defeat a nexus finding because
    Ayala testified that she “faced extortion[] and threats of
    violence[] not only for economic reasons, but also because
    of her family ties.” 
    Id.
     at 1021 (citing Rios v. Lynch, 
    807 F.3d 1123
    , 1128 (9th Cir. 2015)).
    The majority attempts to distinguish Ayala, alleging that
    Ayala brought “an ‘extortion-plus’ claim, that is, a claim that
    the petitioner was independently targeted, not just for
    money, but also because of a protected ground.” Majority
    Op. at 20-21 (emphasis in original). But the majority asserts
    that “common sense tell us that will often, indeed usually,
    not be the case.” although it “is possible—someone could
    be motivated to extort a particular person by, say, animus
    towards their family.” Id. at 21.
    But nowhere in Ayala does the court suggest that a
    showing of “animus” on the part of the persecutor is
    necessary and, indeed, Ayala herself made no such showing.
    See Ayala, 
    855 F.3d at 1016, 1020-21
    . The IJ’s error lay in
    discounting Ayala’s testimony that she had been extorted in
    the past on the basis of her family membership, not in
    RODRIGUEZ-ZUNIGA V. GARLAND                  41
    discounting testimony that she had been extorted on the
    basis of hatred or animus toward her family. See 
    id.
     Ayala
    is thus directly on point because we have before us a similar
    case of “extortion-plus.”
    Having failed to distinguish Ayala, the majority attempts
    to diminish its weight by claiming that the IJ in that case
    committed legal error by “categorically holding that, if a
    persecutor is motivated by a financial goal, i.e., to extort, he
    cannot also be motivated by a petitioner’s protected
    characteristic.” Majority Op. at 21. But contrary to the
    majority’s characterization, the IJ in Ayala made no such
    categorical pronouncement. The Ayala court was, in fact,
    quite clear that the IJ’s mistake was akin to that made in this
    case:
    During the hearing, Ayala claimed that a
    “group of people” was targeting her because
    “[m]y husband’s family owned hotels and I
    believe they wanted to extort us and that is
    why we were being followed.” At the end of
    the hearing, the IJ stated that he was
    affirming the asylum officer’s decision
    “because the only motivation indicated
    throughout is extortion, criminal acts.” He
    did not offer any other explanation.
    Ayala, 
    855 F.3d at 1012
     (emphasis added).
    In a last-ditch effort to escape Ayala’s clear implications,
    the majority argues that “even the panel in Ayala expressed
    some reservations . . . over whether the petitioner would
    actually succeed before the agency.” Majority Op. at 22.
    But such reservations are hardly surprising considering
    Ayala’s failure to present the court with factual evidence to
    42              RODRIGUEZ-ZUNIGA V. GARLAND
    support her theoretically valid nexus theory. The Ayala
    court summarized the record as follows:
    [Ayala] first entered the United States in
    1991 . . . . [S]he remained in the United States
    until December 1998, when she left with her
    husband for Guatemala.
    Ayala stayed in Guatemala for only one
    month. Soon after returning to Guatemala,
    she and her husband were followed by a car
    while riding their motorcycle. Although
    Ayala got off the motorcycle at her husband’s
    urging, he continued riding, and the car
    followed him. Later that day, he was found
    badly beaten. Her husband then told her to
    return to the United States with their child.
    During that same month in Guatemala, Ayala
    also received threatening phone calls at her
    house.
    Ayala returned to the United States in
    January 1999 . . . . While she has been in the
    United States, her family in Guatemala has
    continued to face threats. In 2007, her
    husband was murdered, and at some point in
    2012, unknown assailants shot at her
    mother’s house.
    Ayala, 
    855 F.3d at 1016
    . Based on these facts, “Ayala
    claimed that a ‘group of people’ was targeting her because
    ‘[m]y husband’s family owned hotels and I believe they
    wanted to extort us and that is why we were being
    followed.’” 
    Id.
     (alteration in original). On the record before
    it, the agency might well have found that Ayala had not
    RODRIGUEZ-ZUNIGA V. GARLAND                43
    provided “some evidence . . . , direct or circumstantial” that
    her persecutors were in fact motivated by her relationship to
    her husband’s family. See Elias-Zacharias, 
    502 U.S. at 483
    (emphasis in original).
    But in the present case, the record does not require that
    we simply infer a nexus between persecution and a protected
    ground. The agency here “accord[ed] . . . full evidentiary
    weight” to Rodriguez-Zuniga’s testimony that the robber
    warned her “that if she didn’t give [the robber] . . . money,
    [the robber] was going to hurt [her] son,” and that she was
    being targeted because her “family was here and that [she]
    had a lot of money that [she] could give them.” And when
    Rodriguez-Zuniga refused, the robber said that Rodriguez-
    Zuniga’s “son was going to pay for it.”
    Rodriguez-Zuniga did not guess at the woman’s motives;
    she instead credibly testified that her would-be persecutor
    told her why she was being targeted. And “there was no
    testimony or other evidence inconsistent with [Rodriguez-
    Zuniga’s] recounting of her experiences, and there was no
    reason to doubt the truth, or ‘persuasiveness,’ of her
    narrative” concerning the words that were uttered to her by
    her would-be persecutor. See Plancarte Sauceda v.
    Garland, 
    23 F.4th 824
    , 827 (9th Cir. 2022).
    Unlike in Ayala, then, there is no ambiguity as to why
    Rodriguez-Zuniga was targeted.          Rodriguez-Zuniga’s
    potential persecutors knew her identity and the identities of
    her family members, and their representative targeted
    Rodriguez-Zuniga using her relationship to her son and
    because of her relationship to her husband. Rodriguez-
    Zuniga has therefore satisfied her burden of establishing that
    her family membership was at least “a reason” for her
    persecutors’ actions. See Barajas-Romero v. Lynch, 846
    44              RODRIGUEZ-ZUNIGA V. GARLAND
    F.3d 351, 357-58 (9th Cir. 2017) (holding that, to meet the
    nexus requirement for a withholding-of-removal claim, an
    applicant need show only that a protected ground is “a
    reason” for her feared harm).
    C. The nexus standard for family-based particular
    social groups is not dependent on the persecutor’s
    singular “actual” motivation
    In determining whether a nexus exists between
    persecution and a protected ground, the majority erroneously
    limits consideration to the persecutor’s singular “actual” or
    “intrinsic” motivation. See Majority Op. at 18, 18 n.2. But
    an asylum seeker need prove only that any prospective
    persecution “would be ‘on account of’ one of the five
    [protected grounds],” or that those protected grounds would
    be “one central reason” for the harm. Barajas-Romero, 
    846 F.3d at 357-58
    . In contrast, a person seeking withholding of
    removal must prove that her “life or freedom will be
    threatened in [her] home country . . . ‘because of’ one of the
    five [protected grounds],” or that those protected grounds
    would be “a reason” for the harm. 
    Id.
    Neither asylum claims nor withholding-of-removal
    claims require that a protected ground be “the persecutor’s
    actual motivation,” see Majority Op. at 18, for inflicting the
    harm. Even under asylum’s more stringent “at least one
    central reason” standard, “persecution may be caused by
    more than one central reason, and an asylum applicant need
    not prove which reason was dominant.” Parussimova, 
    555 F.3d at 741
    . By eliminating a petitioner’s ability to establish
    a nexus to a protected ground where “the persecutor’s actual
    motivation for threatening a person is to extort money from
    a third person,” see Majority Op. at 18, the majority departs
    from this court’s precedents that affirm the principle that
    RODRIGUEZ-ZUNIGA V. GARLAND                 45
    “[p]eople, including persecutors, often have mixed
    motives.” Garcia v. Wilkinson, 
    988 F.3d 1136
    , 1143 (9th
    Cir. 2021) (quoting Barajas-Romero, 
    846 F.3d at 357
    ).
    The majority states that it does not deny the “basic
    principle” that “a petitioner may be entitled to relief when a
    persecutor holds multiple or mixed motivations.” See
    Majority Op. at 18 n.2. But the effect of failing to consider
    alternative motives for persecutory acts when the persecutor
    also holds a financial motivation (which, in the majority’s
    view, is “the persecutor’s actual motivation,” see id. at 18)
    is to deny that basic principle.
    The crux of the majority’s rationale seems to be that
    family membership in a case such as this is not “the actual
    motivation” for persecution because it is a means to an end.
    See id. at 31-32 But “[a] person may have ‘a reason’ to do
    something that is not his ‘central’ reason or even ‘one central
    reason.’” Barajas-Romero, 
    846 F.3d at 359
    . And so, to the
    extent that the majority’s holding is directed at Rodriguez-
    Zuniga and others similarly situated (rather than at Nelson),
    it is inconsistent with the more lenient nexus requirement for
    withholding-of-removal claims.
    Moreover, a motive is not only “a reason” but also “a
    ‘central reason’ if the persecutor would not have harmed the
    applicant if such motive did not exist.” Parussimova, 
    555 F.3d at 741
    . To satisfy asylum’s nexus standard, “an
    applicant must prove that such ground was a cause of the
    persecutors’ acts.” 
    Id.
     And a but-for cause is certainly a
    cause. See 
    id.
     Thus, for someone like Nelson, who becomes
    the indirect target of extortionist threats presented to his
    mother, both nexus standards are satisfied.
    The majority’s narrow focus on a persecutor’s financial
    motivation is also difficult to reconcile with binding
    46              RODRIGUEZ-ZUNIGA V. GARLAND
    Supreme Court precedent. In INS v. Elias-Zacharias, 
    502 U.S. 478
    , 481-82 (1992), the Supreme Court held that
    whether a persecutor’s motives are themselves political is
    “irrelevant” to establishing a nexus to a protected political
    opinion. Instead, reviewing courts must ask whether the
    persecutor is motivated by what they perceive to be the
    victim’s protected political opinion. 
    Id. at 482
    .
    That same principle applies here. Our inquiry should not
    be based on whether the persecutors’ acts were motivated by
    an unprotected characteristic, such as a desire for financial
    gain, but rather on whether they were related to one of
    Rodriguez-Zuniga’s or Nelson’s protected grounds—such as
    family membership. See 
    id.
     (“[T]he ordinary meaning of the
    phrase ‘persecution on account of [a protected ground]’ in [
    8 U.S.C. § 1101
    ](a)(42) is persecution on account of the
    victim’s [protected ground], not the persecutor’s.”)
    (emphasis in original). That the potential persecutors in the
    present case also had an economic motivation is thus an
    insufficient basis for us to dismiss their interest in
    Rodriguez-Zuniga’s and Nelson’s family membership, just
    as the nonpolitical motivations of the persecutors in
    Elias-Zacharias were insufficient to allow the court to
    dismiss their interest in the petitioner’s protected
    characteristics. See 
    id.
    D. The majority fails to hold the agency accountable for
    several procedural missteps
    1. The agency failed to independently analyze the
    likelihood that Rodriguez-Zuniga and Nelson
    would be subject to future harm
    Both the IJ and the BIA erred in relying on a negative
    past-persecution finding to reflexively dispose of
    Rodriguez-Zuniga’s and Nelson’s future-persecution claim.
    RODRIGUEZ-ZUNIGA V. GARLAND                  47
    The majority outlines the IJ’s well-founded-fear finding as
    follows: “The IJ noted that, having found Rodriguez-Zuniga
    could not ‘claim past persecution on account of a protected
    ground, it necessarily follow[ed] she also cannot claim a
    well-founded fear of future persecution on account of the
    same protected ground.’” See Majority Op. at 24 (alteration
    in original) (quoting the IJ).
    No additional analysis was provided by the BIA.
    Instead, the BIA stated that Rodriguez-Zuniga “did not
    establish a nexus between the single incident she
    experienced or the harm she fears and either her family
    membership, or her status as a Guatemalan woman.” 
    Id. at 25
    . And it did so without considering Rodriguez-Zuniga’s
    extensive evidence in support of her well-founded fear of
    future persecution. The BIA’s single sentence is not an
    analysis; it is a conclusion, and an unsupported one at that.
    Settled law clearly provides that the failure to establish a
    nexus for past harm does not preclude a petitioner from
    establishing a nexus with respect to likely future harm, even
    if the claim rests upon the same proffered evidence and
    protected grounds. See, e.g., Regalado-Escobar v. Holder,
    
    717 F.3d 724
    , 729-30 (9th Cir. 2013) (remanding for the BIA
    to consider whether a petitioner had established a well-
    founded fear of future persecution on account of a protected
    ground, despite affirming the BIA’s conclusion that the
    petitioner had failed to establish a nexus to that same
    protected ground for past persecution); Spesovska v. Holder,
    
    311 F. App’x 946
    , 948-49 (9th Cir. 2009) (granting the
    petition for review and remanding “[b]ecause the BIA did
    not address the question of [the petitioner]’s individualized
    risk of future persecution based on her religion,”
    notwithstanding its conclusion that substantial evidence
    48              RODRIGUEZ-ZUNIGA V. GARLAND
    supported the BIA’s determination that her past harm had
    not occurred “on account of” religion).
    A contrary interpretation is untenable because the past-
    persecution and well-founded-fear inquiries are distinct
    from one another and encompass different factors. To
    establish past persecution, a petitioner must provide
    evidence that “(1) he has endured serious harm such that his
    ‘treatment rises to the level of persecution’; (2) ‘the
    persecution was committed by the government, or by forces
    that the government was unable or unwilling to control’; and
    (3) ‘the persecution was on account of one or more protected
    grounds.’” Singh v. Garland, 
    48 F.4th 1059
    , 1067 (9th Cir.
    2022) (quoting Kaur v. Wilkinson, 
    986 F.3d 1216
    , 1221-22
    (9th Cir. 2021)).
    By contrast, to demonstrate a well-founded fear of future
    persecution, a petitioner need not make such a showing. The
    petitioner may instead either (1) “establish[] ‘a pattern or
    practice of persecution of people similarly situated,’” or (2)
    “prove that she is a member of a ‘disfavored group’ coupled
    with a showing that she, in particular, is likely to be targeted
    as a member of that group.” Sael v. Ashcroft, 
    386 F.3d 922
    ,
    925 (9th Cir. 2004) (quoting Knezevic v. Ashcroft, 
    367 F.3d 1206
    , 1213 (9th Cir. 2004), and Mgoian v. INS, 
    184 F.3d 1029
    , 1035 n.4 (9th Cir. 1999)); accord, e.g., Halim v.
    Holder, 
    590 F.3d 971
    , 977 (9th Cir. 2009). Under the latter
    approach, “the ‘more serious and widespread the threat’ to
    the group in general, ‘the less individualized the threat of
    persecution needs to be.’” Sael, 
    386 F.3d at 925
     (quoting
    Mgoian, 
    184 F.3d at
    1035 n.4).
    The distinction between the past-persecution and the
    well-founded-fear analyses bears directly on the nexus issue.
    Suppose an IJ found that an LGBTQ petitioner who faced
    RODRIGUEZ-ZUNIGA V. GARLAND                  49
    removal to a country that criminalizes same-sex relations
    had faced a single incident of past harm that was not “on
    account of” her sexual identity or orientation, but rather was
    an act of random violence. The law does not allow the IJ to
    entirely bypass the pattern-or-practice and disfavored-group
    inquiries by categorically declaring, as she did here, that
    “because the Court has previously found [that] [the LGBTQ
    petitioner] cannot claim past persecution on account of a
    protected ground, it necessarily follows [that] she also
    cannot claim a well-founded fear of persecution on account
    of the same protected ground.”
    Elsewhere in the decision, the IJ acknowledged the
    evidence that Rodriguez-Zuniga had presented in support of
    her claim of a well-founded fear of future persecution when
    the IJ stated that “the record shows that there is a high level
    of violence against Guatemalan women,” including “sexual
    assault, torture, and mutilation,” and recognizing the “very
    high” levels of “impunity for the perpetrators of such
    crimes” due to “the government[’s] fail[ure] to enforce its
    laws against rape and domestic abuse.” Yet the IJ
    inexplicably failed to consider Rodriguez-Zuniga’s claim
    that, “as an alternative to past persecution, [she] satisfies the
    requirement of well-founded fear of future persecution” on
    account of her membership in the cognizable social group of
    “Guatemalan women.”
    “IJs and the BIA are not free to ignore arguments raised
    by a petitioner.” Antonio v. Garland, 
    58 F.4th 1067
    , 1075
    (9th Cir. 2023) (quoting Sagaydak v. Gonzales, 
    405 F.3d 1035
    , 1040 (9th Cir. 2005)). And the “[f]ailure to address a
    social group claim, or failure to analyze such a claim under
    the correct legal standard, ‘constitutes error and requires
    remand.’” See 
    id.
     (quoting Rios v. Lynch, 
    807 F.3d 1123
    ,
    1126 (9th Cir. 2015)).
    50              RODRIGUEZ-ZUNIGA V. GARLAND
    Remand is required even if, as the majority suggests, see
    Majority Op. at 22-23, the IJ did not intend to announce a
    categorical rule, but instead inartfully summarized her
    implicit findings on nexus with respect to future persecution.
    “[T]he basis for an agency determination ‘must be set forth
    with such clarity as to be understandable. It will not do for
    a court to be compelled to guess at the theory underlying the
    agency’s action.’” Recinos de Leon v. Gonzales, 
    400 F.3d 1185
    , 1189 (9th Cir. 2005) (quoting SEC v. Chenery Corp.,
    
    332 U.S. 194
    , 196-97 (1947)).
    2. The agency failed to properly consider Rodriguez-
    Zuniga’s political-opinion claim
    In addition to claiming before the agency that she feared
    persecution by reason of her family membership and her
    status as a Guatemalan woman, Rodriguez-Zuniga
    contended that she was at risk due to her political opinion.
    Yet the IJ did not address this claim in any way, much less
    in a meaningful one. And again, “IJs and the BIA are not
    free to ignore arguments raised by a petitioner.” Antonio v.
    Garland, 
    58 F.4th 1067
    , 1075 (9th Cir. 2023) (quoting
    Sagaydak v. Gonzales, 
    405 F.3d 1035
    , 1040 (9th Cir. 2005)).
    The majority states that the BIA addressed Rodriguez-
    Zuniga’s political-opinion argument by “concluding that she
    presented no evidence [that] ‘she ever expressed a political
    opinion.’” Majority Op. at 13 (quoting the BIA). But the
    BIA abused its discretion in so doing because its single-
    sentence resolution of the issue—in a footnote, no less—did
    not satisfy its duty to “explain what factors it has considered
    or relied upon.” Kalubi v. Ashcroft, 
    364 F.3d 1134
    , 1140
    (9th Cir. 2004) (citing Rodriguez-Matamoros v. INS, 
    86 F.3d 158
    , 160 (9th Cir. 1996)); see also, e.g., Movsisian v.
    Ashcroft, 
    395 F.3d 1095
    , 1098 (9th Cir. 2005) (“We have
    RODRIGUEZ-ZUNIGA V. GARLAND                  51
    long held that the BIA abuses its discretion when it fails to
    provide a reasoned explanation for its actions.”); Mattis v.
    INS, 
    774 F.2d 965
    , 967 (9th Cir. 1985) (“[W]e require that
    [the BIA’s] stated reasons evidence its consideration of all
    relevant factors.     Cursory, summary or conclusory
    statements are inadequate.” (citations omitted)).
    The majority expands on the BIA’s single-sentence,
    footnoted rationale for nearly five pages. See Majority Op.
    at 13-15, 32-33. But “reviewing courts remain bound by
    traditional administrative law principles, including the rule
    that judges generally must assess the lawfulness of an
    agency’s action in light of the explanations the agency
    offered for it rather than any ex post rationales a court can
    devise.” Garland v. Ming Dai, 
    141 S. Ct. 1669
    , 1679 (2021).
    The BIA’s cursory ruling on Rodriguez-Zuniga’s
    political-opinion claim was also erroneous because the BIA
    arrived at its evidentiary conclusion without the benefit of a
    factual finding by the IJ. The BIA’s regulations are
    unequivocal: the BIA cannot engage in its own factfinding.
    
    8 C.F.R. § 1003.1
    (d)(3)(iv)(A) (2022) (“The Board will not
    engage in factfinding in the course of deciding cases . . . .”).
    This court, in Rodriguez v. Holder, 
    683 F.3d 1164
     (9th Cir.
    2012), has said the same:
    Where the IJ has not made a finding of fact
    on a disputed matter, and such a finding is
    necessary to resolution of the case, the BIA
    must remand to the IJ to make the required
    finding; it may not conduct its own fact-
    finding. Where the BIA fails to follow its
    own regulations and makes factual findings,
    “it commits an error of law, which we have
    jurisdiction to correct.”
    52              RODRIGUEZ-ZUNIGA V. GARLAND
    
    Id. at 1170
     (citations omitted) (quoting Padmore v. Holder,
    
    609 F.3d 62
    , 67 (2d Cir. 2010)); see also, e.g., Brezilien v.
    Holder, 
    569 F.3d 403
    , 413 (9th Cir. 2009) (“[W]here the IJ
    has not made a necessary factual finding, the regulation
    requires the BIA to remand the factual inquiry to the IJ rather
    than making its own factual finding on the matter.”).
    Here, the IJ entirely neglected to consider, even
    cursorily, that Rodriguez-Zuniga had asserted a political
    opinion as a protected ground. The BIA thus had no lawful
    basis to reach a conclusion on this issue. See Solorio Mejia
    v. Barr, 
    833 F. App’x 455
    , 457 n.2 (9th Cir. 2020) (“The
    Board lacked authority to correct the IJ’s failure to make a
    factual determination about whether the cartel imputed a
    political opinion to Solorio Mejia.”). And the defect is not
    curable by the majority conducting independent factfinding
    of its own or by offering rationales that were not put forth by
    the BIA itself. The case should therefore be remanded.
    3. The agency erred in its analysis of Rodriguez-
    Zuniga’s request for protection           under    the
    Convention Against Torture (CAT)
    Finally, the agency erred in its analysis of Rodriguez-
    Zuniga’s request for protection under the CAT. The
    majority avoids this issue by opining that Rodriguez-Zuniga
    has waived any argument regarding this claim because she
    “does not argue that the agency erred in finding that she
    presented insufficient evidence that the Guatemalan
    government would consent to her torture.” Majority Op. at
    25.
    To the contrary, Rodriguez-Zuniga argues that the
    agency erred in failing to consider evidence favorable to her
    CAT claim, and that she has “articulated a specific
    individualized threat of torture.” Government acquiescence
    RODRIGUEZ-ZUNIGA V. GARLAND                53
    is part and parcel of the legal meaning of torture, which is
    defined under the CAT not just as an act imposing severe
    pain or suffering, but rather
    any act by which severe pain or suffering,
    whether physical or mental, is intentionally
    inflicted on a person . . . when such pain or
    suffering is inflicted by, or at the instigation
    of, or with the consent or acquiescence of, a
    public official acting in an official capacity or
    other person acting in an official capacity.
    
    8 C.F.R. § 1208.18
    (a)(1); see also, e.g., Ornelas-Chavez v.
    Gonzales, 
    458 F.3d 1052
    , 1059 (9th Cir. 2006) (“To qualify
    for protection under CAT, Ornelas-Chavez must establish
    that he suffered torture, i.e., severe pain or suffering
    intentionally inflicted for discriminatory purposes “by or at
    the instigation of or with the consent or acquiescence of a
    public official or other person acting in an official
    capacity.”); Hernandez v. Barr, 
    830 F. App’x 804
    , 807 (9th
    Cir. 2020) (Hunsaker, J., dissenting) (“[T]he definition of
    ‘torture’ encompasses government ‘acquiescence’ . . . .”).
    By challenging the agency’s finding that she was not more
    likely than not to be tortured, Rodriguez-Zuniga therefore
    necessarily challenges the agency’s findings both as to the
    severity of the harm that she was likely to suffer and as to
    acquiescence by the Guatemalan government.
    Once the obstacle of waiver is removed, the agency’s
    analytical errors are readily apparent. As a threshold matter,
    this court has held that “a CAT applicant may satisfy his
    burden with evidence of country conditions alone.” Aguilar-
    Ramos v. Holder, 
    594 F.3d 701
    , 705 (9th Cir. 2010). The
    IJ’s guiding principle that “a pattern of human rights
    54              RODRIGUEZ-ZUNIGA V. GARLAND
    violations alone is insufficient to show a particular applicant
    would be in danger of torture if returned to that country” is
    therefore contrary to the law of this circuit.
    The agency also erred in failing to consider “all evidence
    relevant to the possibility of future torture.” See 
    8 C.F.R. § 1208.16
    (c)(3). Specifically, it did not discuss, in analyzing
    Rodriguez-Zuniga’s CAT claim, the evidence that
    Rodriguez-Zuniga had presented with the respect to the
    danger that she was likely to face as a woman in Guatemala.
    The agency’s failure to discuss that evidence suggests that
    the agency ignored its own regulation:
    When nothing in the record or the BIA’s
    decision indicates a failure to consider all the
    evidence, a “general statement that [the
    agency] considered all the evidence before
    [it]” may be sufficient. But, where there is
    any indication that the BIA did not consider
    all of the evidence before it, a catchall phrase
    does not suffice, and the decision cannot
    stand. Such indications include misstating
    the record and failing to mention highly
    probative or potentially dispositive evidence.
    Cole v. Holder, 
    659 F.3d 762
    , 771-72 (9th Cir. 2011)
    (brackets in original) (emphasis added) (quoting Almaghzar
    v. Gonzales, 
    457 F.3d 915
    , 922 (9th Cir. 2006)).
    Moreover, “in assessing a CAT claim from an applicant
    who has posited multiple theories for why he might be
    tortured, the relevant inquiry is whether the total probability
    that the applicant will be tortured—considering all potential
    sources of and reasons for torture—exceeds 50 percent.”
    Velasquez-Samayoa v. Garland, 
    49 F.4th 1149
    , 1154 (9th
    RODRIGUEZ-ZUNIGA V. GARLAND                  55
    Cir. 2022) (emphasis in original). The agency was thus
    required to consider not only the country-conditions
    evidence, but also to consider it in conjunction with evidence
    of the prior threats against Rodriguez-Zuniga and her son.
    Finally, “the IJ and BIA erred by construing
    ‘government acquiescence’ too narrowly.” See Aguilar-
    Ramos, 
    594 F.3d at 705
    . The IJ based her acquiescence
    conclusion solely on “the Guatemalan government[’s
    having] taken steps to combat criminal violence and combat
    human rights,” and that “[t]he Guatemalan constitution and
    laws also prohibit torture or cruel, inhuman, and degrading
    treatment.” But this court has found error where “the BIA
    focused only on the national government’s efforts and not
    their efficacy.” See Barajas-Romero v. Lynch, 
    846 F.3d 351
    ,
    363 (9th Cir. 2017). “[T]he ‘efficacy of the government’s
    efforts to stop the . . . violence,’ not just the willingness of
    the national government to do so, must be examined.” 
    Id.
    (quoting Madrigal v. Holder, 
    716 F.3d 499
    , 509 (9th Cir.
    2013)).
    Indeed, in the same opinion where the IJ lauded the
    Guatemalan government’s efforts, she also observed that
    “[i]mpunity for the perpetrators of . . . crimes against women
    remain[s] very high, and the government is failing to enforce
    its laws against rape and domestic abuse,” and that “[t]he
    record details often brutal violence against women by gangs,
    government authorities, and society in general.” (emphasis
    added). A State Department report concerning human rights
    in Guatemala remarked that “[p]rincipal human rights
    abuses included widespread institutional corruption,
    particularly in the police and judicial sectors; police and
    military involvement in serious crimes, such as kidnapping,
    drug trafficking, trafficking in persons, and extortion; and
    societal violence, including lethal violence against women.”
    56              RODRIGUEZ-ZUNIGA V. GARLAND
    The report further noted that “[g]angs, organized crime, and
    narcotics trafficking organizations committed considerable
    violence; corruption and inadequate investigation made
    prosecution of such crimes difficult.”
    The agency therefore erred in its consideration of
    government acquiescence by limiting its analysis to the
    formalized existence of governmental efforts to protect its
    citizens. Whatever the merits of her CAT claim, Rodriguez-
    Zuniga is entitled to a procedurally adequate adjudication.
    CONCLUSION
    The majority hesitates to apply what it characterizes as
    the “arcane requirements” of our settled law because it is
    concerned that doing so might lead to results that are not
    “what . . . Congress meant by providing relief for refugees.”
    Majority Op. at 28. But our duty is to apply the law, not to
    rewrite it.
    In any event, the majority’s fear that “[s]omething has
    gone terribly wrong when judges conclude that relief for
    persecution and torture is mandated just because someone
    was the victim of a brief and failed robbery attempt in their
    home country,” Majority Op. at 28, is unfounded. To apply
    this court’s precedent would not require the BIA to afford
    relief to Rodriguez-Zuniga and Nelson. Instead, “[w]hen we
    remand due to the BIA’s legal error, we allow the BIA to
    exercise its judgment and administrative expertise using the
    appropriate legal standards. In such cases . . . we do not
    instruct the BIA as to any required outcome on remand.”
    Kotasz v. INS, 
    31 F.3d 847
    , 851 (9th Cir. 1994) (citation
    omitted).
    Although I believe that substantial evidence should
    compel the agency to conclude that Rodriguez-Zuniga and
    RODRIGUEZ-ZUNIGA V. GARLAND                57
    Nelson have satisfied the nexus requirement, the agency
    must still determine whether they are likely to suffer future
    harm rising to the level of persecution. It must also consider
    whether the Guatemalan government would be unable or
    unwilling to protect them from such persecution, and
    whether they would be able to reasonably relocate within
    Guatemala. They thus have many hurdles yet to clear, but in
    my view they have cleared enough to be entitled to
    reconsideration. I therefore respectfully dissent.