Singh v. Garland

FOR PUBLICATION UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT AJAY PAL SINGH, No. 22-211 Agency No. Petitioner, A208-189-362 v. OPINION MERRICK B. GARLAND, Attorney General, Respondent. On Petition for Review of an Order of the Board of Immigration Appeals Argued and Submitted April 17, 2023 San Francisco, California Filed March 22, 2024 Before: Lawrence VanDyke and Gabriel P. Sanchez, Circuit Judges, and Kathryn H. Vratil, * District Judge. Opinion by Judge Vratil; Dissent by Judge VanDyke * The Honorable Kathryn H. Vratil, United States District Judge for the District of Kansas, sitting by designation. 2 SINGH V. GARLAND SUMMARY ** Immigration Granting in part and denying in part Ajay Singh’s petition for review of a decision of the Board of Immigration Appeals affirming the denial of asylum and related relief, and remanding, the panel held that the record compelled the conclusion that Singh experienced past harm rising to the level of persecution, the BIA erred in its internal relocation analysis for purposes of asylum and withholding of removal, and substantial evidence supported the denial of protection under the Convention Against Torture. The panel concluded that the BIA erred when it read the immigration judge’s decision as having shifted the burden to the government to rebut the presumption of future persecution. The panel explained that because the BIA expressly adopted the IJ’s reasons for finding that internal relocation was safe and reasonable, it also adopted the IJ’s flawed relocation analysis, which did not afford Singh the presumption of past persecution or shift the burden to the government to prove that Singh can safely and reasonably relocate within India. The panel wrote that the BIA compounded its mistake by failing to conduct a reasoned analysis of Singh’s individualized situation to determine if he could safely relocate to another area of India. The panel explained that the BIA’s reliance on evidence that Singh never successfully filed a police report, or that the landlord-tenant identification ** This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.  SINGH V. GARLAND 3 system is not uniformly enforced, did not address whether Singh would be substantially safer in a new location if he were to continue expressing his support for the Khalistan secession movement or maintain his advocacy for the Mann party. Moreover, the BIA’s analysis was inadequate because it failed to specifically address Singh’s stated intent to continue proselytizing for his party. The panel held that substantial evidence supported the BIA’s findings that Singh did not suffer past torture and is not likely to suffer future torture, and that he failed to show that any torture would be inflicted or consented to by public officials or persons acting in official capacities. Dissenting, Judge VanDyke disagreed with the majority that the IJ or BIA failed to place the burden regarding internal relocation on the government. Judge VanDyke wrote that the panel manufactured a non-existent conflict between the IJ’s and BIA’s decisions, and explained that even if there was a conflict between the BIA’s and IJ’s analyses and conclusions, this court reviews the BIA’s decision, which appropriately placed the burden for internal relocation on the government, adequately considered the appropriate factors in its individualized analysis, and was supported by substantial evidence. COUNSEL Inna Lipkin (argued), Law Offices of Inna Lipkin, Redwood City, California, for Petitioner. Jesi J. Carlson (argued) and Nancy K. Canter, Senior Litigation Counsel; Sarah S. Wilson; Micah Engler, Trial Attorney; Brian Boynton, Principal Deputy Assistant 4 SINGH V. GARLAND Attorney General; Office of Immigration Litigation, Civil Division, United States Department of Justice, Washington, D.C.; for Respondents. OPINION VRATIL, District Judge: Ajay Pal Singh, a native and citizen of India, petitions for review of an order of the Board of Immigration Appeals (“BIA”) dismissing his applications for asylum, withholding of removal, and protection under the Convention Against Torture (“CAT”). Singh asserts that he suffered past persecution and has a well-founded fear of future persecution due to his membership in the Shiromani Akali Dal Amritsar (“Mann party”), which advocates for the creation of a sovereign state for Sikh people. Two of India’s major political parties, the Bharatiya Janata Party (“BJP”) and the Shiromani Akali Dal Badal (“Badal”) party, oppose the Mann party. The BIA affirmed the Immigration Judge’s (“IJ”) determination that Singh did not qualify for asylum or withholding of removal because the injuries and threats that he suffered at the hands of BJP and Badal party members were not sufficiently severe to constitute past persecution. The IJ and the BIA further found that even if Singh had established past persecution, he was not entitled to relief because he could reasonably relocate within India to avoid persecution in the future. The BIA also affirmed the IJ’s finding that Singh was not eligible for CAT relief because he had not established that he would suffer torture by or with the acquiescence of public officials.  SINGH V. GARLAND 5 For the reasons set forth below, we grant the petition in part, deny the petition in part, and remand to the BIA for further proceedings consistent with this opinion. I. BACKGROUND On or about August 24, 2015, Singh entered the United States without inspection or admission. On September 9, 2015, the Department of Homeland Security (“DHS”) issued a Notice to Appear which charged Singh with removability. On March 21, 2019, Singh appeared at a hearing before an IJ, conceded removability, and filed applications for asylum, withholding of removal, and protection under CAT. A. Singh’s Testimony At his removal hearing on February 16, 2018, Singh testified about the circumstances that he faced prior to coming to the United States. Singh stated that because of his affiliation with the Mann party, members of the BJP and the Badal party verbally and physically attacked him on multiple occasions in 2014 and 2015. Singh worked for the Mann party while living in India. He attended rallies, participated in social work activities, hung political party posters, and encouraged others to join the party. In November of 2014, he received two threatening telephone calls from callers who identified themselves as BJP and Badal party members and told him to quit the Mann party or be killed. On December 30, 2014, BJP and Badal party members attacked Singh and beat him with hockey sticks and baseball bats. His attackers stated that they would kill him if he resumed work for the Mann party. Singh spent one day in the hospital and remained on bed rest for 15 days. He went to a police station to report the attack and told them that he was a Mann party member. The police told him that they 6 SINGH V. GARLAND would not file a complaint against the BJP and threatened to jail him if he returned to the station to complain again. On June 10, 2015, BJP and Badal party members again attacked Singh as he returned home after prayer at a Sikh temple. The attackers beat him for about six minutes with wooden sticks and metal knuckle dusters. During the beating, his attackers said “you have not quit [the] Mann Party yet, you will be killed.” He spent one day in the hospital, then lived in hiding with his grandparents for several weeks before fleeing India. After Singh left India, his family told him that BJP and Badal party members harassed Singh’s family members and inquired about his whereabouts. Members of the BJP and the Badal party had previously harassed Singh’s father, a Mann party member, before he died. Singh testified that he could not safely relocate within India because he works for the Mann party and intends to keep doing so wherever he lives. He fears that BJP and Badal party members will kill him if he returns to India because BJP members live all over India and he will have to provide identification to rent lodgings, which would trigger a police check and reveal his whereabouts. Singh does not have any relatives who live outside of Punjab, and he does not speak Hindi, the predominant language outside of Punjab in India. B. The IJ’s Decision The IJ found Singh removable as charged and denied his applications for asylum, withholding of removal, and CAT relief. Although the IJ deemed Singh credible, the IJ concluded that the two beatings and threatening phone calls that Singh experienced did not rise to the level of persecution. The IJ noted that the beatings did not result in  SINGH V. GARLAND 7 “any lasting injuries requiring extensive medical treatment,” the two phone calls were “vague,” and Singh “failed to demonstrate that the callers have the will or ability to carry out such threats.” Because Singh had not demonstrated past persecution, the IJ stated that he was not entitled to a presumption of a well-founded fear of future persecution, citing 8 C.F.R. § 1208.13(b)(3). Accordingly, the IJ did not shift the burden to the DHS to show that Singh could safely and reasonably relocate to another part of the country to avoid persecution and instead placed the burden on Singh to demonstrate that he could not reasonably relocate. The IJ observed that Singh was only a low-level Mann party worker and “there was no persuasive evidence in the record that a member of the Mann party similarly situated” to Singh had been harmed or targeted by BJP or Badal party members or police outside of Punjab. The IJ noted that when Singh went to the police after his first attack, he never filed a police report and the police did not take his picture or fingerprints. The IJ concluded that Singh had “not offered more than speculative assertions that members of an opposing party would seek him out for persecution.” Regarding Singh’s claim that persecutors could track him through the landlord-tenant identification system, the IJ found that “the evidence [Singh] has submitted shows that the identification system is not uniformly enforced and does not always involve the police,” and that the record provided no evidence that the identification system is used to target Mann party members. Given Singh’s young age and good health, the IJ concluded that internal relocation would be reasonable and “the record does not demonstrate that there would be unreasonable social or cultural constraints, or other practical barriers.” 8 SINGH V. GARLAND The IJ also denied Singh’s application for withholding of removal, stating that because Singh was not eligible for asylum, it necessarily followed that he was not eligible for withholding of removal, which requires a heightened showing that he more likely than not would be persecuted if removed. Finally, the IJ denied Singh’s application for CAT protection. The IJ found that Singh’s two beatings did not amount to torture and that he did not establish that any public official consented or acquiesced to the attacks or would consent or acquiesce to future torture. The IJ noted evidence that “only those considered by police to be high-profile militants are at risk of persecution” and “the Indian government has made concrete efforts towards holding government officials responsible for wrongdoings.” C. The BIA’s Decision Singh appealed the IJ’s decision to the BIA. In dismissing the appeal, the BIA affirmed the findings of the IJ, first agreeing that Singh had not shown harm sufficiently severe to constitute persecution. The BIA cited Gu v. Gonzales, 454 F.3d 1014, 1019 (9th Cir. 2006), for the proposition that persecution is an “extreme concept” that “does not include every sort of treatment our society regards as offensive.” It further cited Prasad v. INS, 47 F.3d 336, 339–40 (9th Cir. 1995), for the conclusion that “the cumulative physical, psychological and emotional harm the respondent suffered, while abhorrent, does not rise to the level of persecution.” The BIA also affirmed the IJ’s relocation analysis, stating that “even assuming [Singh] established past persecution, which he did not, the Immigration Judge properly determined that the DHS met its burden of rebutting the  SINGH V. GARLAND 9 presumption of future persecution based on [Singh’s] ability to relocate within India.” The BIA emphasized the IJ’s findings that it was unlikely that police or anyone else would harm Singh or track him down due to his low-profile status as a low-level worker of the Mann party. As we discuss below, the BIA misinterpreted the IJ’s opinion, which did not analyze relocation under a presumption that Singh had established past persecution and therefore did not shift to the DHS the burden of showing that Singh could safely and reasonably relocate outside of Punjab. The BIA nevertheless affirmed the IJ’s holding that Singh could safely and reasonably relocate outside Punjab and that he was therefore ineligible for asylum and withholding of removal. The BIA also affirmed the IJ’s denial of protection under CAT. It determined that Singh had not suffered past torture and had not established state action as to any future torture, because “[g]eneral country conditions evidence that some government officials are corrupt and commit abuses [was] insufficient to show [that Singh] would face an individualized risk of torture inflicted by or at the instigation of or with the consent or acquiescence of a public official.” II. ANALYSIS A. Standards Of Review “[O]ur review is limited to the BIA’s decision, except to the extent that the IJ’s opinion is expressly adopted.” Soriano-Vino v. Holder, 653 F.3d 1096, 1099 (9th Cir. 2011) (quotation marks and citation omitted). We review questions of law de novo. Rodriguez v. Holder, 683 F.3d 1164, 1169 (9th Cir. 2012). We review factual findings under the substantial evidence standard. Singh v. Garland, 57 F.4th 643, 651 (9th Cir. 2022). Under this standard, “[a] factual finding is ‘not supported by substantial evidence when any 10 SINGH V. GARLAND reasonable adjudicator would be compelled to conclude to the contrary based on the evidence in the record.’” Aden v. Wilkinson, 989 F.3d 1073, 1079 (9th Cir. 2021) (quoting Bringas-Rodriguez v. Sessions, 850 F.3d 1051, 1059 (9th Cir. 2017) (en banc)). When the BIA determines whether particular acts constitute persecution for purposes of asylum, we have held alternatively that the BIA’s determination is reviewed de novo or for substantial evidence. See Kaur v. Wilkinson, 986 F.3d 1216, 1221 (9th Cir. 2021) (reviewing de novo); Sharma v. Garland, 9 F.4th 1052, 1060 (9th Cir. 2021) (reviewing for substantial evidence). We need not address which standard should apply because the harm suffered by Singh rose to the level of persecution even under the substantial evidence standard, which affords greater deference to the BIA’s determinations. See Singh v. Garland, 57 F.4th at 652. B. Asylum At the discretion of the Attorney General, asylum is available to an applicant who demonstrates that he is a refugee. 8 U.S.C. § 1158(b)(1). A refugee is a person who is unable or unwilling to return to the country of origin “because of persecution or a well-founded fear of persecution on account of race, religion, nationality, membership in a particular social group, or political opinion.” 8 U.S.C. § 1101(a)(42)(A). “Either past persecution or a well-founded fear of future persecution provides eligibility for a discretionary grant of asylum.” Ratnam v. INS, 154 F.3d 990, 994 (9th Cir. 1998). An individual “who establishes past persecution is presumed to have a well-founded fear of persecution.” Id. (citing 8 C.F.R. § 208.13(b)(1)(i)). “The source of the persecution  SINGH V. GARLAND 11 must be the government or forces that the government is unwilling or unable to control.” Ahmed v. Keisler, 504 F.3d 1183, 1191 (9th Cir. 2007) (citation omitted). 1. Past Persecution To establish past persecution, an asylum applicant must demonstrate (1) serious harm “ris[ing] 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, such as political opinion.” Kaur, 986 F.3d at 1221–22 (quoting Bringas- Rodriguez, 850 F.3d at 1062). The BIA affirmed the IJ’s determination that Singh did not experience serious harm rising to the level of persecution. After “comparing the facts of [Singh’s] case with those of similar cases,” we conclude that the record compels the conclusion that the harm Singh suffered rose to the level of persecution. Singh v. Garland, 57 F.4th at 654 (9th Cir. 2022) (quoting Singh v. INS, 134 F.3d 962, 967–68 (9th Cir. 1998)). In Singh v. Garland, we held that the BIA erred in dismissing a Mann party member’s application for asylum in circumstances which are strikingly similar to those presented here. Id. at 653–54. In doing so, we articulated five factors which compelled the conclusion that petitioner had experienced past persecution: (1) he was forced to flee his home after being repeatedly assaulted; (2) one of those incidents involved a death threat; (3) he was between the ages of 16 and 18 when the attacks occurred; (4) his brother also 12 SINGH V. GARLAND experienced this violence; and (5) we have already recognized that Mann Party members have faced persistent threats in the region of India [Punjab] where [the Mann party member] was twice attacked. Id. at 653. Four of those five factors support a finding of past persecution here. 1 As to the first factor, Singh was forced to flee his home and live in hiding with his grandparents after BJP and Badal party members assaulted him and threatened to kill him on multiple occasions. As we explained in Singh, “being forced to flee from one’s home in the face of an immediate threat of severe physical violence or death is squarely encompassed within the rubric of persecution.” Id. (quoting Flores Molina v. Garland, 37 F.4th 626, 633–34 (9th Cir. 2022)). As to the second factor, BJP and Badal party members repeatedly threatened to kill Singh. He received two threatening phone calls in November of 2014 and was threatened with death during the attacks in December of 2014 and June of 2015. “Repeated death threats, especially when those threats occurred in conjunction with other forms of abuse, require a finding of past persecution.” Smolniakova v. Gonzales, 422 F.3d 1037, 1049 (9th Cir. 2005); see also Aden, 989 F.3d at 1082 (“[W]hen the incidents have involved physical harm plus something more, such as credible death threats, we have not hesitated to 1 The third factor is not on point because Singh was 22 and not a minor at the time of his attacks.  SINGH V. GARLAND 13 conclude that the petitioner suffered persecution.” (emphasis in original)). 2 The fourth factor applies because, like the petitioner in Singh, Singh’s family experienced mistreatment from BJP and Badal party members. Singh testified that they harassed his father for being a Mann party member and harassed his family to discover Singh’s whereabouts after he left India. “[H]arms that have befallen a petitioner’s family members or close friends strengthen an applicant’s past-persecution claim.” Singh v. Garland, 57 F.4th at 654 ( quotation marks and citation omitted). Finally, as to the fifth factor, “we have recognized in multiple cases that Mann Party members have faced persistent harassment, intimidation, threats, and violence in Punjab,” and “an asylum applicant’s claim of persecution is further strengthened when evidence that the applicant was physically beaten and threatened with his life is presented in conjunction with evidence of the country’s political and social turmoil.” Id. (citing Kaur, 986 F.3d at 1219–20; Aden, 989 F.3d at 1083) (cleaned up). That is the precise situation here. As in Singh, the evidence compels the conclusion that Singh experienced serious harm rising to the level of persecution. See Flores Molina, 37 F.4th at 636 (“Any reasonable adjudicator would be compelled to hold that the repeated and specific death threats that [petitioner] experienced, amid the violence and menacing confrontations to which he was subjected, amount to persecution.”). 2 Even in the absence of physical violence, we have “consistently held that death threats alone can constitute persecution.” Canales-Vargas v. Gonzales, 441 F.3d 739, 743 (9th Cir. 2006) (quoting Navas, 217 F.3d at 658). 14 SINGH V. GARLAND We are not persuaded by the agency’s reasons for finding that Singh did not suffer sufficiently serious harm. The IJ cited the fact that Singh did not show evidence of “lasting injuries requiring extensive medical treatment.” However, “we do not require severe injuries to meet the serious-harm prong of the past-persecution analysis.” Singh v. Garland, 57 F.4th at 654; see Flores Molina, 37 F.4th at 636 (“[I]t is the conduct of the persecutor that is relevant to evaluating whether past treatment rises to the level of persecution—not the level of harm or subjective suffering the petitioner experienced.” (quotation marks and citation omitted)). In particular, “[w]here an applicant suffers [physical] harm on more than one occasion, and as in this case is victimized at different times over a period of years, the harm is severe enough that no reasonable fact-finder could conclude that it did not rise to the level of persecution” necessary to sustain an asylum claim. Chand v. INS, 222 F.3d 1066, 1073–74 (9th Cir. 2000). Here, Singh experienced multiple physical attacks and death threats over an eight-month period, from November of 2014 to June of 2015. No reasonable factfinder would conclude that Singh did not experience serious harm rising to the level of persecution. The BIA cited Gu v. Gonzales for the proposition that persecution is an “extreme concept” that “does not include every sort of treatment our society regards as offensive.” 454 F.3d at 1019. But Gu did not involve multiple instances of physical violence coupled with death threats. Instead, the petitioner in Gu experienced one brief detention, interrogation, and beating by Chinese police because he participated in an unsanctioned religious practice. Id. at 1020. Gu concluded that this single incident did not compel a finding of past persecution, “distinguishing cases in which the persecutor had some ‘continued interest’ in the petitioner  SINGH V. GARLAND 15 from those cases involving only ‘a single, isolated encounter.’” Id. Here, unlike Gu, BJP and Badal party members repeatedly targeted Singh for months. They continued to visit his family and inquire about him even after he left India. Singh’s attackers knew his identity and displayed a continuing interest in him, and his mistreatment was not a single isolated encounter. See Singh v. Garland, 57 F.4th at 655 (distinguishing Gu on similar basis). The BIA’s reliance on Prasad v. INS, 47 F.3d 336 (9th Cir. 1995), is similarly inapt. In Prasad, a group of ethnic Fijians detained the petitioner, on account of his race, hit him in the stomach, kicked him, and questioned him about his support for a particular political party. Id. at 339. Four to six hours later, the group set Prasad free. Id. Unlike the present case, the crowd subjected Prasad to a single attack, did not harm him severely enough to require medical attention, and did not expressly threaten to harm him again. Id. We concluded that “the cumulative physical, psychological and emotional harm the respondent suffered, while abhorrent, does not rise to the level of persecution.” 47 F.3d at 339–40. We later distinguished Prasad from another case, Desir v. Ilchert, 840 F.2d 723 (9th Cir. 1988), that bears a far closer resemblance to the circumstances of this petition. In Desir, petitioner had four separate encounters with his persecutors, members of a Haitian paramilitary called the Ton Ton Macoutes, involving his failure to pay bribes to the military force and his political opposition to the Haitian government. Id. at 724–25. In those four encounters, which spanned from 1979 to 1981, Macoutes members repeatedly arrested Desir, beat him, shot at him, and threatened him with death. Id. Desir fled Haiti, entered the United States, and sought asylum. Id. at 725. The agency denied asylum in part 16 SINGH V. GARLAND because it found that Desir had not experienced persecution based on his political opinion. Id. After determining that the Macoutes’ mistreatment of Desir was politically motivated, we determined that he experienced serious harm that rose to the level of persecution and that the BIA had erred in concluding otherwise. Id. at 729. We held that Desir had established eligibility for asylum because he had “presented evidence of (1) successive and specific threats on his life; (2) on the basis of imputed political opinion; (3) in the context of systemic human rights abuses linked to extortion by the Ton Ton Macoutes . . . .” Id. Like the petitioner in Desir and unlike the petitioner in Prasad, Singh had four separate encounters with his persecutors. Over an eight-month period, Singh received two phone calls where BJP and Badal party members threatened to kill him, and BJP and Badal party members beat him on two other occasions while threatening to kill him. These threats were expressly based on his political opinion, as his attackers demanded that he stop working for the Mann party. In addition, Singh submitted extensive evidence of systematic human rights abuses against Mann party members in Punjab. See id. at 729 (“Where evidence of a specific threat on an [applicant’s] life is presented in conjunction with general corroboration which describes political and social turmoil, the [applicant] has succeeded in establishing prima facie eligibility for asylum.”).  SINGH V. GARLAND 17 For all these reasons we find that the record compels a finding that Singh suffered harm rising to the level of persecution. 3 2. Fear Of Future Persecution If a petitioner demonstrates past persecution on account of statutorily protected grounds at the hands of individuals whom the government was unable or unwilling to control, he is entitled to a presumption of a well-founded fear of future persecution. Mashiri v. Ashcroft, 383 F.3d 1112, 1119 (9th Cir. 2004). The burden then shifts to the government to “show by a preponderance of the evidence that the applicant either no longer has a well-founded fear of future persecution in the country of his nationality, or that he can reasonably relocate internally to an area of safety.” Singh v. Whitaker, 914 F.3d 654, 659 (9th Cir. 2019). To meet this burden, the government must demonstrate either a “fundamental change in circumstances” or that Singh could “avoid future persecution by relocating to another part of [India], and under all the circumstances, it would be reasonable to expect [him] to do so.” Boer-Sedano v. Gonzales, 418 F.3d 1082, 1089 (9th Cir. 2005) (quoting 8 C.F.R. § 1208.13(b)(1)(i)(A)–(B)). Because the IJ determined that Singh had not established past persecution, he found that Singh was not entitled to a presumption of a well-founded fear of future persecution and therefore did not shift to the DHS the burden of showing that 3 On remand, the agency must determine in the first instance whether Singh established the two remaining elements of past persecution, that “the persecution was committed by the government, or by forces that the government was unable or unwilling to control” and “the persecution was on account of one or more protected grounds.” Kaur, 986 F.3d at 1221; Parussimova v. Mukasey, 555 F.3d 734, 738 (9th Cir. 2009). 18 SINGH V. GARLAND Singh could safely and reasonably relocate outside of Punjab. The BIA, however, concluded that “even assuming [Singh] established past persecution, which he did not, the Immigration Judge properly determined that the [DHS] met its burden of rebutting the presumption of future persecution based on [Singh’s] ability to relocate within India.” The BIA erred in finding that the IJ had shifted the burden to the government to demonstrate the reasonableness of internal relocation. As the record makes clear, the burden remained with Singh to prove that he could not safely and reasonably relocate to another part of the country. The IJ began its internal relocation analysis by stating, “[t]he Court finds, by a preponderance of the evidence, the respondent could avoid any future persecution by relocating to another part of India, and it would be reasonable to expect him to do so. See 8 C.F.R. § 1208.13(b)(3)(i).” That the IJ did not afford Singh the presumption of a well-founded fear of future persecution is made clear by the regulation it cited, 8 C.F.R. § 1208.13(b)(3)(i), which provides, “[i]n cases in which the applicant has not established past persecution, the applicant shall bear the burden of establishing that it would not be reasonable for him or her to relocate.” (emphasis added). 4 The IJ found that Singh could reasonably relocate within India, not because of evidence the government offered to rebut the presumption, but because “it is unlikely that an individual like [Singh] would be harmed by police or by 4 Conversely, the regulation the BIA cited and wrongly attributed to the IJ states the correct standard for analyzing internal relocation when an applicant has established past persecution and is entitled to the presumption of a well-founded fear of future persecution. See 8 C.F.R. § 1208.13(b)(1)(i)–(ii).  SINGH V. GARLAND 19 someone the police is unable or unwilling to control.” The IJ reasoned that Singh was a low-level worker and supporter of the Mann party, and “there is no persuasive evidence in the record that a member of the Mann party similarly situated to the respondent has been harmed or targeted by opposing party members or police outside of Punjab.” In other words, the IJ’s determination was based on Singh’s failure to offer evidence that low-level Mann party members have been targeted or harmed outside of Punjab. The IJ also faulted Singh for failing to establish that the national identification system used to register tenants would be used by landlords to pass along information either to police or opposition party members. The IJ found that “[t]he evidence [Singh] has submitted shows that the identification system is not uniformly enforced and does not always involve the police.” “Nor does the record evidence indicate that the identification system is used to target Mann party members.” This discussion reflects that the burden was on Singh to show that the landlord-tenant identification system would be utilized by landlords at the behest of police or opposition party members to persecute him in areas outside of Punjab. The IJ concluded that Singh “ha[d] not offered more than speculative assertions that members of an opposing party would seek him out for persecution,” and therefore his testimony “was insufficient to demonstrate that he could not relocate to another part of India.” The BIA thus erred when it found that the IJ had applied a rebuttable presumption of future persecution and shifted the burden to the government to demonstrate that Singh could safely and reasonably relocate within India. The BIA compounded its mistake by failing to conduct a “reasoned analysis with respect to [Singh’s] individualized situation” to determine if he could safely relocate within 20 SINGH V. GARLAND another area of India. Singh v. Whitaker, 914 F.3d at 661. In Singh v. Whitaker, the government bore the burden of showing by a preponderance of the evidence that the petitioner could safely and reasonably relocate internally. Id. at 659. We concluded that the BIA erred in its relocation analysis for two reasons. First, the BIA “erred by failing to address the potential harm [opposition party] members, or other local authorities, might inflict upon Singh in a new state.” Id. at 661. Second, we concluded that the BIA “failed to specifically address Singh’s stated intent to continue proselytizing for his party wherever he went.” Id. “Thus, the BIA’s analysis regarding whether Singh could reasonably relocate was inadequate.” Id. We remanded for the BIA to conduct a sufficiently individualized relocation analysis for petitioner’s asylum and withholding of removal claims. As in Singh v. Whitaker, the BIA here analyzed whether Punjabi police or other actors would try to pursue Singh to other parts of India based on his prior Mann party activities, without considering Singh’s stated intent to continue his political advocacy and support for the Mann party wherever he goes. For the government to rebut the presumption of future persecution, it is not enough to show that Punjabi police or other actors are unlikely to follow Singh outside of Punjab because he is a low-level Mann party member and not a “high-profile militant.” Id. at 660–61. The BIA’s reliance on evidence that Singh never successfully filed a police report, or that the landlord-tenant identification system is not uniformly enforced, fails to address whether Singh “would be substantially safer in a new location if he were to continue expressing his support for the Khalistan secession movement” or maintain his advocacy for the Mann party. Id. at 660. The BIA’s failure to specifically address  SINGH V. GARLAND 21 in its decision Singh’s “stated intent to continue proselytizing for his party wherever he went” rendered the BIA’s relocation analysis inadequate under our precedent. Id. Contrary to our dissenting colleague’s contention, the BIA did not independently analyze relocation and determine that the government met its burden. Rather, the BIA expressly adopted the IJ’s reasons for finding that internal relocation was safe and reasonable. In doing so, the BIA adopted the IJ’s flawed relocation analysis, which did not afford Singh the presumption of past persecution or shift the burden to the government to prove that Singh can safely and reasonably relocate within India. See 8 C.F.R. § 1208.13(b)(1)(i)–(ii). For example, the IJ’s finding that it was unlikely Singh would be harmed by police or opposition party members was based entirely on Singh’s failure to offer evidence that other similarly situated, low-level Mann party members have been targeted or harmed outside of Punjab. Under a rebuttable presumption, the DHS was required to demonstrate that individuals like Singh who participate in Mann party activities or advocate for a separatist Khalistan state in other regions of India are unlikely to suffer persecution at the hands of police or opposition party members. Id. The record is devoid of any such evidence. 5 5 Even though Singh was a low-level party member, opposition party members specifically targeted him and had a particularized interest in harming him. They made multiple threatening phone calls to his house, physically confronted and attacked him twice, and repeatedly visited his family’s home to inquire about him after he left India. His low-level status in the party did not protect him from past persecution and does not 22 SINGH V. GARLAND Our dissenting colleague asserts that the government did offer evidence concerning the persecution of Mann party members outside of Punjab. But the Law Library of Congress 2012 report submitted by the government in these proceedings is the same report found wanting in Singh v. Whitaker. See 914 F.3d at 660–61. That report concludes that the Punjabi police are likely to pursue only “high profile militants” outside of Punjab. As we observed in Singh v. Whitaker, the report does not address the risk of persecution at the hands of local authorities or opposition party members, nor does the report shed light on the likelihood of harm when an individual continues to advocate for Mann party activities in a new state. Id. at 661. On this record, the government has not carried its burden to show that Singh is unlikely to be harmed or targeted by local officials or opposition party members upon relocation based on his future political activities and advocacy for Khalistan succession. Even if the government had presented such evidence, the BIA failed to specifically address these questions in an individualized relocation analysis. The dissent’s reasoning falters for a second reason as well. Where, as here, the BIA erred in its relocation analysis, we do not ignore the error to see if substantial evidence nevertheless supports the agency’s determination. See, e.g., Singh v. Whitaker, 914 F.3d at 661 (remanding because “BIA’s analysis regarding whether Singh could reasonably relocate was inadequate”); Knezevic v. Ashcroft, 367 F.3d 1206, 1214–15 (9th Cir. 2004) (remanding issue of reasonableness where agency failed to take into consideration several regulatory factors); Mashiri v. support a conclusion that his low status will protect him from future persecution outside of Punjab.  SINGH V. GARLAND 23 Ashcroft, 383 F.3d 1112, 1123 (9th Cir. 2004) (remanding where IJ erroneously “placed the burden of proof regarding internal relocation on the petitioner”); cf. Garcia v. Wilkinson, 988 F.3d 1136, 1147 (9th Cir. 2021) (remanding where BIA applied erroneous legal standard to its nexus analysis). In sum, because the BIA erred in its relocation analysis, we grant Singh’s petition to review his claim for asylum and remand to the BIA for consideration in light of Singh v. Whitaker, 914 F.3d 654. C. Withholding Of Removal A petitioner is entitled to withholding of removal if he can establish a clear probability that his life or freedom will be threatened upon return on account of “race, religion, nationality, membership in a particular social group, or political opinion.” Shrestha v. Holder, 590 F.3d 1034, 1039 (9th Cir. 2010) (quoting Ahmed, 504 F.3d at 1199); 8 U.S.C. § 1231(b)(3)(A). The clear probability standard for withholding of removal is more stringent than the well-founded fear standard for asylum because withholding of removal is a mandatory form of relief. Ahmed, 504 F.3d at 1199. Here, the BIA affirmed the IJ’s denial of Singh’s application for withholding of removal because it affirmed the IJ’s denial of his application for asylum. It noted that “[w]hen the government rebuts an applicant’s well-founded fear of persecution, it defeats the asylum claim, and his or her claim for withholding of removal.” We therefore remand this claim to the BIA so that it can determine whether Singh has established an asylum claim, and thus benefits from a “presumption of entitlement to withholding of deportation.” See Canales-Vargas, 441 F.3d at 746. 24 SINGH V. GARLAND D. Convention Against Torture Finally, we analyze Singh’s claim for relief under CAT. To prevail on this claim, Singh must establish that more likely than not, he will be tortured if removed to India. See Kamalthas v. INS, 251 F.3d 1279, 1282 (9th Cir. 2001) (citing 8 C.F.R. § 208.16(c)(2)). Singh must also show that any torture would 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. 8 C.F.R. § 208.18(a)(1). The regulations implementing CAT define torture as follows: [A]ny act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as . . . punishing him or her for an act he or she or a third person has committed or is suspected of having committed, . . . or for any reason based on discrimination of any kind, 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). Here, the IJ found that Singh’s two beatings did not amount to torture and that he did not establish that any public official consented or acquiesced to the attacks or would consent or acquiesce to future torture. The BIA affirmed these findings.  SINGH V. GARLAND 25 Substantial evidence supports the BIA’s findings that Singh did not suffer past torture and is not likely to suffer future torture. See Singh, 914 F.3d at 663 (reversing agency’s denial of asylum and withholding of removal due to a flawed relocation analysis but upholding denial of CAT because record did not compel finding of torture). Specifically, substantial evidence supports the BIA’s determination that Singh did not experience pain or suffering that rose to the level of past torture or that more likely than not, he will experience future torture if removed to India. See Fon v. Garland, 34 F.4th 810, 815–16 (9th Cir. 2022) (record compelled finding of past persecution, but substantial evidence supported BIA determination that petitioner failed to show likelihood of future torture). Further, substantial evidence supports the BIA’s determination that Singh did not show that any torture would be inflicted or consented to by public officials or persons acting in official capacities. We therefore deny the petition as to CAT relief. III. CONCLUSION For the reasons set forth above, we GRANT Singh’s petition in part and REMAND to the BIA to consider (1) whether Singh is eligible for asylum because he suffered past persecution on account of statutorily protected grounds by the government or individuals whom the government was unable or unwilling to control; (2) if so, whether the DHS rebutted the presumption of a well-founded fear of future persecution; and (3) whether Singh is entitled to withholding of removal. Singh’s petition is otherwise DENIED. 26 SINGH V. GARLAND VANDYKE, Circuit Judge, dissenting. The agency in this case denied asylum and withholding of removal to Ajay Pal Singh because it concluded that (1) his past harm didn’t rise to the level of past persecution, and (2) even if it did, the government “met its burden of rebutting the presumption of future persecution based on [his] ability to relocate within India.” The majority disagrees on both accounts. There’s no reason to debate the first question because even if Singh did suffer past persecution, and the government thus bore the burden to rebut a presumption of future persecution, here the agency unmistakably determined the government met that burden. 1 And substantial evidence supports that determination. The majority concludes otherwise by (1) insisting that the BIA misunderstood the IJ’s analysis, and (2) cursorily second- guessing the agency’s weighing of ample evidence in this case supporting Singh’s ability to relocate. I will say it again: this is not how we should be reviewing the agency’s decisions. I respectfully dissent. 1 As a matter of first impression, this case presents a debatable question of whether Singh suffered past persecution. And as our court often does when an immigration case presents a debatable question, the majority here again imposes our court’s own view of what constitutes past persecution on the agency. See Molina v. Garland, 37 F.4th 626, 646 (9th Cir. 2022) (VanDyke, J., dissenting). But of course, our review in these cases is not supposed to be de novo, but instead extraordinarily deferential. Consistent with that standard, if required to reach it I would conclude that substantial evidence supports the agency’s conclusion that Singh did not suffer past persecution. But as I explain, it’s not necessary to address that issue to correctly decide this case.  SINGH V. GARLAND 27 DISCUSSION The majority’s main attack on the agency’s relocation analysis is that the agency failed to put the burden on the government. That is a strange reading of the BIA’s decision. The BIA clearly stated as an alternative conclusion that, “even assuming [Singh] established past persecution,” the government “met its burden of rebutting the presumption of future persecution based on [his] ability to relocate within India.” And the BIA cited three pages of the IJ’s decision in support. The majority argues the BIA was wrong, however, because the IJ never put the burden on the government. It’s the majority that is wrong, and doubly so. First, while the IJ never directly said it put the burden on the government, it did say that it “finds, by a preponderance of the evidence, [Singh] could avoid any future persecution by relocating ….” That statement makes no sense unless the IJ was putting the burden on the government to show, “by a preponderance of the evidence, [Singh] could avoid any future persecution by relocating.” Singh would never need to prove—by the preponderance of the evidence or otherwise—that he “could avoid any future persecution by relocating.” If the IJ had kept the burden on Singh for the entirety of its analysis, it would have said that Singh failed to show, by a preponderance of the evidence, that he would suffer (not avoid) future persecution even if he relocated. The BIA’s reading of the IJ as having flipped the burden to the government in parts of the IJ’s decision is not only a reasonable reading of that decision, but arguably the only reasonable reading. The majority strains mightily to impose its own eccentric reading of the IJ’s decision in lieu of the BIA’s. 28 SINGH V. GARLAND Moreover, even if the majority was right that the BIA must have read the IJ’s decision wrong (it isn’t), that still wouldn’t matter. If there was necessarily a conflict between the BIA’s and IJ’s analyses and conclusions (there isn’t), then obviously it is the BIA’s that we review for substantial evidence. There is no doubt the BIA put the burden to prove relocatability on the government and, reviewing the evidentiary record, concluded the government met its burden. The majority thus badly errs in first manufacturing a nonexistent conflict between the BIA and IJ, and then focusing on its misreading of the IJ’s decision instead of properly reviewing the BIA’s decision for substantial evidence. There is more than a bit of irony that we purport to hold the agency to such a persnickety standard when we can’t even get our own method of review right. Perhaps recognizing that its errant focus on the IJ’s decision is flawed, the majority also briefly contends the BIA’s conclusion that the government met its burden is not supported by substantial evidence. As discussed in more depth below, the evidence supporting the government is substantial, and includes a government report that directly addresses the “Feasibility of Relocation for Sikhs and Mann Party Members,” the IJ’s individualized finding the Singh was likely a “low-level worker,” and the absence of any evidence that other Mann party members similarly situated to Singh have been “harmed or targeted … outside of Punjab.” The majority relies on the fact that Singh was previously targeted in Punjab as inconsistent with the agency’s conclusion that he is unlikely to be targeted outside Punjab, failing to engage the agency’s actual rationale. Beyond that strawman, the majority misreads Singh v. Whitaker as imposing a magic words requirement mandating that the agency must talismanically state it has considered  SINGH V. GARLAND 29 that the petitioner will continue to engage in political activity once he relocates. 914 F.3d 654 (9th Cir. 2019). There is no reason to think the agency here assumed anything different, given that (1) it said it considered “all of the relevant evidence, including country conditions information and [Singh’s] testimony,” (2) it specifically cited our Whitaker decision, and (3) the norm is that the agency in the relocation analysis expects the petitioner will continue to exhibit the characteristic for which he was previously persecuted. We don’t assume, for example, that a petitioner who was persecuted for his religion will stop practicing that religion after relocating. This case is not meaningfully distinguishable from our very long list of other decisions where we denied petitions involving Mann Party Sikhs from Punjab claiming persecution and an inability to relocate. 2 The agency’s 2 See, e.g., Singh v. Garland, No. 22-1013, 2023 WL 4585960, at *1 (9th Cir. July 18, 2023) (“Substantial evidence supports the BIA’s conclusion that Petitioner could relocate to Mumbai or Kolkata because Petitioner lived in Mumbai and Kolkata without persecution for significant periods of time.”); Singh v. Garland, No. 22-171, 2023 WL 3220907, at *1 (9th Cir. May 3, 2023) (“First, substantial evidence supports the BIA’s conclusion that Singh can safely relocate in India. Singh argues that he cannot safely relocate because he is a member of the Mann Party and faces threats by both the Bharatiya Janata Party (‘BJP’) and Akali Dal Badal Party (‘Badal Party’). The record supports the conclusion that Singh is a low-level member of the Mann Party, and the central authorities controlled by the BJP target ‘high-profile militants.’ Singh is also unlikely to suffer persecution by Badal Party outside of Punjab since different political parties control other states.” (citation omitted)); Singh v. Garland, No. 21-1005, 2023 WL 2945312, at *2 (9th Cir. Apr. 14, 2023) (“Moreover, the agency reasonably found, based on the evidence, that the particular kind of work Singh engaged in was unlikely to result in harm to him. Though hard-core militants, or those who have drawn 30 SINGH V. GARLAND the interest of central Indian authorities, may be unable to feasibly relocate within India, political-party workers like Singh who do not draw governmental scrutiny appear able to do so. … And though Singh reiterated his fear of future persecution in India based on two attacks that he experienced as a Congress Party member in Uttar Pradesh, nothing in the record compels the conclusion that he has a well-founded fear of persecution in Punjab.”); Singh v. Garland, No. 22-286, 2023 WL 2401263, at *1 (9th Cir. Mar. 8, 2023) (“The agency further found that although the police with whom Singh spoke did not help him, they did not take his picture or fingerprint, and Singh indeed testified that he has no reason to believe the police would seek him out elsewhere in India. The agency considered whether Singh’s stated intent to continue his political activity would cause other individuals to harm him outside of Punjab, see Singh v. Whitaker, 914 F.3d 654, 661 (9th Cir. 2019), and concluded that country conditions evidence indicated that he would not be harmed if he expressed his views peacefully, and that Singh did not testify that he would engage in violent activity. The agency also pointed to country conditions evidence indicating that low-profile Mann Party members who expressed their views peacefully have been able to relocate successfully within India. The record therefore does not compel the conclusion that internal relocation would not be safe.”); Singh v. Garland, No. 17-71809, 2022 WL 3998574, at *1 (9th Cir. Sept. 1, 2022) (“The IJ cited sufficient evidence that Singh could relocate to ‘other states’ within India beyond Punjab. See Singh v. Whitaker, 914 F.3d 654, 659 (9th Cir. 2019). For example, the IJ found that ‘holding pro- Khalistani views would not make someone a high-profile militant,’ and that Sikhs from Punjab may relocate internally to escape the attention of local police.”); Singh v. Garland, No. 21-70289, 2022 WL 819524, at *1 (9th Cir. Mar. 17, 2022) (“The agency reasonably determined that the government sufficiently rebutted the presumption of future persecution with evidence that Singh could safely and reasonably relocate outside of Punjab (including a 2018 report from the Library of Congress titled ‘India: Feasibility of Relocation of Sikhs and Members of the Shiromani Akali Dal (Mann) Party,’ which concludes that relocation is feasible as long as the individual is not a high-profile militant of interest to the central authorities, together with Singh’s own testimony that he is not a high-profile member of the Mann Party and has never been linked to any terrorism or extremism in India).”); Singh v. Garland, No. 16-73511,  SINGH V. GARLAND 31 relocation finding is easily supported by substantial evidence, and the majority follows a clearly erroneous if too- familiar path in concluding otherwise. A. The BIA Did Not Err When Interpreting the IJ’s Decision. Assuming that Singh did experience past persecution, the government bears the burden of overcoming the presumption that Singh has a well-founded fear of future persecution. Popova v. INS, 273 F.3d 1251, 1259 (9th Cir. 2001) (citing 8 C.F.R. § 208.13(b)(1)(i)). The government can overcome this presumption by proving, as relevant here, 2022 WL 414247, at *1 (9th Cir. Feb. 10, 2022) (“The evidence, including a 2012 report from the Library of Congress, supported the conclusion that relocation is feasible given Singh’s own testimony that he is not a high-profile member of the Mann Party. Although Singh testified that he continues to support the Mann party and believes he will be ‘traceable’ as a Sikh if he relocates, the record does not compel a conclusion different than the agency’s because substantial evidence— including reports in the administrative record that refute Singh’s concern—supports the finding that Singh could safely and reasonably relocate within India.”); Singh v. Garland, No. 19-71638, 2021 WL 5277081, at *1 (9th Cir. Nov. 21, 2021) (“Substantial evidence supports the agency’s determination that, even if Singh established past persecution in Punjab, his presumption of a clear probability of future persecution was rebutted by evidence that he could safely and reasonably relocate to another part of India.” (citation omitted)); Singh v. Garland, 843 Fed. App’x 973, 974 (9th Cir. 2021) (“Substantial evidence also supports the agency’s determination that Singh was able to safely relocate in India to avoid future persecution.” (citation omitted)); Singh v. Garland, No. 19-70396, 2022 WL 445515, at *1 (9th Cir. Feb. 14, 2022) (“Singh’s speculation that police will harm him on return, his mother’s interactions with police, and the documentary evidence do not compel the conclusion that he is more likely than not to be tortured if removed.”). 32 SINGH V. GARLAND that Singh can avoid future persecution by relocating somewhere within India. Id. Here, the BIA provided the benefit of the presumption to Singh by requiring the government to prove by a preponderance of the evidence that Singh (1) can relocate within India to a place where he wouldn’t have a well- founded fear and (2) that it would be reasonable to require him to relocate there. As explained below, the BIA offered a reasonable interpretation of the IJ’s decision as it conducted this analysis, and even if the BIA had misinterpreted the IJ’s decision, it would be immaterial to the BIA’s analysis, which is the decision we review for substantial evidence. 1. The BIA Reasonably Interpreted the IJ’s Decision as Offering Two Alternative Analyses. Although admitting that both “[t]he IJ and the BIA … found that even if Singh had established past persecution, he … could reasonably relocate within India to avoid future persecution,” the majority insists the BIA misunderstood the IJ’s decision as performing two alternative analyses, one of which put the burden on the government. The majority insists the entirety of the IJ’s decision put the burden on Singh to prove that it would be unreasonable for him to relocate within India, and thus the IJ never shifted the burden to the government to overcome the presumption that Singh has a well-founded fear of future persecution. The majority’s privileging of its own reading of the IJ’s decision over the BIA’s alternative interpretation is inappropriate. The IJ’s decision, while not an exemplar of clarity, can certainly be read the way the BIA read it—  SINGH V. GARLAND 33 indeed, that might be the only way it should be read. The BIA construed the IJ’s decision as reaching two alternative conclusions: First, that Singh did not experience past persecution and thus had the burden to prove he couldn’t relocate, and that he failed to meet that burden. Second, that even if Singh did experience past persecution, and the government thus bore the burden to prove Singh could relocate to avoid persecution, the government met its burden and proved Singh could relocate. That is a reasonable reading of what the IJ did. a. The IJ’s Decision The majority quotes several statements from the IJ’s decision, claiming they demonstrate that the IJ never placed the burden on the government. One of those statements does put the burden on Singh, but another puts the burden on the government. The rest are statements that say nothing as to who has the burden. i. One Statement Places the Burden on Singh. The IJ’s decision contains one statement that clearly places the burden on Singh to prove that it would be unreasonable for him to relocate within India: As found above, the respondent has not established by a preponderance of the evidence that it would be unreasonable for the respondent to relocate within India to avoid future persecution. Of course, the fact that the IJ put the burden on Singh for part of its analysis—the part where it had determined Singh failed to demonstrate past persecution and thus bore the 34 SINGH V. GARLAND burden—is entirely unsurprising and consistent with the BIA’s reading of the IJ’s decision as containing two alternative analyses. 3 It could be argued—and presumably this is the majority’s position—that the “[a]s found above” part of the IJ’s statement indicates that earlier parts of the IJ’s decision also necessarily placed the burden on Singh. But it’s not necessary to read “found above” to mean that. As a logical matter, if the IJ earlier in its decision reached a finding that Singh could safely relocate after having placed the burden on the government—which, as I’ll explain next, is the best reading of the IJ’s decision—then a fortiori those earlier findings would support that Singh could relocate once the burden was switched to Singh. Thus, the BIA reasonably interpreted the language from the IJ’s decision quoted above as merely indicating that the same evidence supporting a finding that the government satisfied its burden of proving that Singh could relocate also supports a finding on relocation if Singh bore the burden. That, of course, makes sense. Regardless of who has the burden, the IJ and BIA were required to take into consideration all evidence submitted by the parties and determine whether Singh has a well-founded fear of future 3 The majority relatedly makes much of the IJ’s citation to the regulation that places the burden on the petitioner to show unreasonableness of relocation when he experienced no past persecution. But again, the IJ’s citation to that regulation is entirely consistent with the BIA’s interpretation of the IJ’s decision, since nobody disagrees that the IJ put the burden on the petitioner for at least part of the IJ’s analysis. The mere fact that the IJ cited a regulation relevant to one part of its analysis that everyone agrees it conducted doesn’t help the majority in showing that the IJ did not also do an alternative analysis with the burden on the government.  SINGH V. GARLAND 35 persecution. If a preponderance of the evidence shows that the government met its burden of showing that Singh can relocate, then the same evidence shows that Singh did not prevail on relocation if he bore the burden. Both the IJ and the BIA considered the whole record and reached a conclusion supported by substantial evidence that a preponderance of the evidence demonstrated Singh could relocate within India to avoid persecution. The single sentence in the IJ’s decision that Singh failed to meet his burden after the IJ found that he had not suffered past persecution does not change this fact and has no impact on whether the IJ had an alternative finding where it considered the same “evidence as a whole” while placing the burden on the government. ii. One Statement Places the Burden on the Government. The IJ also determined that “by a preponderance of the evidence, [Singh] could avoid any future persecution by relocating to another part of India, and it would be reasonable to expect him to do so.” Although the IJ did not explicitly say that it was placing the burden on the government in this portion of its analysis, that statement just quoted is only consistent with the IJ putting the burden on the government. The only circumstance where someone had to prove, “by a preponderance of the evidence,” that Singh “could avoid … future persecution by relocating,” is where the IJ had put the burden of proof for relocation on the government. 4 4 Even assuming arguendo that this statement by the IJ could be read consistently with putting the burden on Singh (which it can’t), reading it as putting the burden on the government is certainly a reasonable 36 SINGH V. GARLAND After finding that Singh had not faced past persecution, the IJ considered whether Singh had proven that he has a well-founded fear of future persecution. The IJ (and the BIA) concluded he did not. Past persecution is a close call in this case. Presumably recognizing that, both the IJ and the BIA offered an alternative conclusion that the government proved “by a preponderance of the evidence, [Singh] could avoid any future persecution by relocating to another part of India, and it would be reasonable to expect him to do so.” iii. The Other Statements Are Irrelevant as to Who Has the Burden. The majority points to other statements in the IJ’s decision as supporting that the IJ kept the burden on Singh for the entirety of its analysis, but none of those statements support that conclusion. The majority argues that the IJ found Singh could reasonably relocate within India “not because of evidence the government offered” but because of “Singh’s failure to offer evidence that low-level Mann party members have been targeted or harmed outside of Punjab.” But the government did offer evidence, including a government report specifically about persecution of Mann party members outside of Punjab. That report is evidence the government offered to show that someone like Singh—a low level party member—is unlikely to be harmed outside of Punjab. It is both (1) evidence and (2) individualized because the agency needed to determine whether Singh was a low-level party member or not. The IJ’s consideration of interpretation of the sentence. The IJ does not explicitly say who it is placing the burden on in this sentence, and nothing in the record belies the BIA’s reading that here the IJ was putting the burden on the government.  SINGH V. GARLAND 37 “Singh’s failure to offer evidence that low-level Mann party members have been targeted or harmed outside of Punjab” therefore does not demonstrate that the IJ placed the burden on Singh. The government had already provided evidence that low-level members like Singh were rarely targeted. The IJ’s and BIA’s comments about Singh’s evidentiary shortcomings cannot be divorced from the backdrop of the government’s evidentiary showing about Singh’s ability to relocate. The government had already met its burden, and the agency was merely observing that Singh had failed to offer any counter evidence that undermined the government’s showing. The majority’s highlighting of other statements by the IJ face the same problem: those statements only show that the IJ considered the evidence offered by Singh (or, more accurately, the lack thereof) as a part of its consideration of the totality of the evidence. The IJ could reasonably be read—as the BIA did—as having done so in deciding whether the Petitioner had provided any evidence rebutting the evidence given by the government to meet its burden. Each of the IJ’s references to Singh’s lack of evidence are against the backdrop of the evidence already offered by the government. For example, the majority points to the IJ’s finding that “[t]he evidence [Singh] has submitted shows that the identification system is not uniformly enforced and does not always involve the police.” The majority next highlights the IJ’s statement that “the record evidence [does not] indicate that the identification system is used to target Mann party members” and concludes that “[t]his discussion reflects that the burden was on Singh.” But these statements reflect nothing of the sort. The IJ considered Singh’s evidence on this issue in light of the evidence the government had already 38 SINGH V. GARLAND offered. This shows nothing other than that the IJ considered all the evidence—something it was required to do regardless of who bore the burden. Similarly, the IJ’s observation that Singh had “not offered more than speculative assertions that members of an opposing party would seek him out for persecution,” merely shows that the IJ weighed the evidence offered by both the government and Singh to make its determination, and that it determined the evidence offered by the government was more concrete than Singh’s. If additional evidence in the record had corroborated Singh’s bare assertions (and overcame the strength of the evidence the government provided), the IJ might have found the government did not meet its burden of showing Singh could relocate. The simple fact that a factfinder points out the paucity of evidence provided by one of the parties does not evince that the factfinder is necessarily putting the burden on that party. It just means that party hasn’t put forward anything that would rebut the other party’s evidence on that issue. The IJ’s statements that some of Singh’s assertions were “speculative” and “unsupported” does not show the IJ put the burden on Singh; it simply shows the IJ satisfied its responsibility of considering all the evidence—including the absence thereof in some instances—before deciding if the party who bore the burden had met it. In short, none of the IJ’s statements that the majority relies on clearly demonstrate that the IJ put the burden on Singh for the entirety of its analysis. Indeed, at least one of the IJ’s statements is inconsistent with the conclusion that Singh bore the burden. It is the panel majority, not the BIA, that has manufactured a strained reading of the IJ’s decision.  SINGH V. GARLAND 39 b. The BIA’s Opinion As explained, the BIA interpreted the IJ’s decision as offering two alternative conclusions. First, that Singh had not suffered past persecution and failed to prove that he had a well-founded fear of future persecution. And alternatively, that even if Singh suffered past persecution, the government proved by a preponderance of the evidence Singh could relocate within India. The BIA did not clearly err in reaching this conclusion. Although the IJ never explicitly stated that it was making alternative findings, as described above, the IJ’s decision is consistent with the BIA’s conclusion that it did. Indeed, the BIA’s reading is probably the best interpretation of the IJ’s decision—better than the majority’s, which as explained simply ignores a statement from the IJ’s decision that only makes sense if the burden was put on the government for that part of the decision. The IJ never stated in the section with that statement that the burden was on Singh. And the IJ’s only statement clearly placing the burden on Singh is not inconsistent with the BIA’s “two-alternatives” interpretation since that interpretation allows that the IJ did properly put the burden on Singh for part of his analysis. Nothing within the IJ’s decision forecloses the BIA’s interpretation of it. 2. Even if the BIA Had Misread the IJ’s Decision, the BIA’s Independent Analysis and Conclusion That the Government Met Its Burden Controls and Is Supported by Substantial Evidence. It is undisputed that the BIA placed the burden on the government to prove by a preponderance of the evidence that Singh could relocate within India. No one can dispute that. The BIA’s opinion is crystal clear that the BIA reached an 40 SINGH V. GARLAND alternative conclusion: that even if Singh suffered past persecution, the government proved by a preponderance of the evidence that Singh does not have a well-founded fear of future persecution because he could relocate within India. If that conclusion is supported by substantial evidence (as explained below, it is), we should deny Singh’s petition for review. Instead, the majority gets caught up in a fallacy: that the BIA’s supposed misunderstanding of the IJ’s decision affected the BIA’s analysis and ultimate conclusion. But even if the IJ did not offer alternative conclusions like the BIA said it did, that reading of the IJ’s decision had no impact on the BIA’s own analysis regarding whether the government proved by a preponderance of the evidence that Singh could relocate. When, as here, the BIA conducts its own analysis and does not merely adopt the IJ’s analysis, the only relevant question for our review is whether the BIA’s decision was supported by substantial evidence. See Yan Liu v. Holder, 640 F.3d 918, 925 (9th Cir. 2011). Of course, if the BIA did adopt portions of the IJ’s decision, we review those as if they are part of the BIA’s decision. See Shrestha v. Holder, 590 F.3d 1034, 1039 (9th Cir. 2010); Melkonian v. Ashcroft, 320 F.3d 1061, 1065 (9th Cir. 2003). But to the extent the BIA’s and IJ’s decisions conflict, we review only the BIA’s conclusion—even if the BIA itself did not recognize a conflict existed. See Rodriguez-Zuniga v. Garland, 69 F.4th 1012, 1022–23 (9th Cir. 2023) (“If the IJ erred in applying some … rule, the BIA did not …. And to the extent the BIA and IJ part ways, we review the BIA’s findings for substantial evidence.”). Here, even if the IJ failed to give Singh the benefit of the presumption, the BIA clearly did. The BIA explained, “even assuming the respondent established past persecution, which  SINGH V. GARLAND 41 he did not, … the DHS met its burden of rebutting the presumption of future persecution based on [Singh’s] ability to relocate within India.” Unless the record compels a conclusion otherwise—which it does not—we must uphold the BIA’s conclusion. The majority argues that my reasoning falters because where “the BIA erred in its relocation analysis, we do not ignore the error to see if substantial evidence nevertheless supports the agency’s determination.” But the majority rebuts a strawman. My argument is not that we should ignore the agency’s legal error to see if its conclusion is still otherwise supported by substantial evidence. My argument is that the BIA’s reading of the IJ as offering alternative analyses is reasonable, but even if the BIA were wrong about the IJ’s reasoning, that error did not affect the BIA’s independent conclusion, which itself is still supported by substantial evidence. Whether the IJ failed to give Singh the benefit of the presumption is irrelevant to the BIA’s conclusion in this case because it had no substantive impact on the evidentiary record. Imagine, for example, that in some hypothetical case an IJ found the petitioner had suffered past persecution. By law, that finding should give rise to a presumption of a well- founded fear of future persecution. But imagine the IJ messed up and completely biffed it on the law by putting the burden on the petitioner to prove that he could not relocate (even though he faced past persecution). That legal error could be corrected by the BIA on appeal. The BIA could apply the correct standard—affording the petitioner the presumption that he has a well-founded fear of future persecution and placing the burden on the government to prove that the petitioner could relocate to avoid persecution—and properly reach the conclusion that the 42 SINGH V. GARLAND preponderance of the evidence demonstrated the petitioner could relocate. The only way the IJ’s misapplication of the law would impact the BIA’s analysis is if the IJ’s error somehow prevented the petitioner from completing the evidentiary record. If, for example, the IJ in our hypothetical prohibited the government from putting any evidence into the record (or if the government chose not to) because the burden, in the IJ’s mistaken view, was on the petitioner to prove that he could not relocate, then that would of course affect the evidentiary record and eliminate the BIA’s ability to later conduct a proper analysis with the burden properly on the government. No one in this case claims such an evidentiary gap exists. Both the government and Singh put all the evidence they wished into the record to support their positions regarding Singh’s ability to relocate. As both the IJ and the BIA concluded, the preponderance of all that evidence supported that Singh could relocate within India. That conclusion is not affected in this case by who bore the burden. In some cases, who bears the burden can make the difference. But this is not one of those cases. Here, both parties offered evidence the agency could consider relevant to relocation. Singh’s evidence was primarily in the form of a self-serving affidavit. The government’s evidence was more substantial. The agency’s job was to sift through all the evidence to determine whether Singh could relocate or not. Because the record does not compel a contrary conclusion to the BIA’s answer to that question, we should have denied Singh’s petition for review of his asylum and withholding of removal claims. There is simply no need to remand to the BIA for reconsideration of “its relocation analysis” when the BIA has already properly considered the evidence while placing the burden of proof  SINGH V. GARLAND 43 on the government, and it’s obvious nothing will change on remand. B. The Agency Reached a Conclusion Supported by Substantial Evidence and Did Not Err in Its Individualized Analysis Regarding Whether Singh Can Relocate. The agency “must conduct a reasoned analysis with respect to a petitioner’s individualized situation” when determining whether the petitioner can avoid future persecution by relocating. Singh v. Whitaker, 914 F.3d at 661. Included in this individualized analysis is the question of whether the police, or people that the police cannot control, would persecute the petitioner for his future political activities in the location where the petitioner would relocate. Id. When reviewing the agency’s analysis, we are supposed to presume that the agency considered all evidence in the record, even if some evidence is not explicitly mentioned. Hernandez v. Garland, 52 F.4th 757, 770–71 (9th Cir. 2022); see also Najmabadi v. Holder, 597 F.3d 983, 991 (9th Cir. 2010) (concluding that the agency “adequately considered” the record even though it “did not directly reference” certain evidence). Here, applying this well-established presumption and the agency’s reasoned analysis of the evidence, I conclude that the agency’s determination is supported by substantial evidence. 1. Substantial Evidence Supports the Agency’s Conclusion That Singh Can Relocate Within India. Here, the agency had to determine whether Singh could relocate outside of Punjab to avoid future persecution. Citing Singh v. Whitaker, the BIA acknowledged that the IJ conducted an individualized analysis regarding whether 44 SINGH V. GARLAND Singh could relocate within India, and then the BIA proceeded to conduct its own individualized analysis. Both the IJ and BIA considered “all of the relevant evidence, including country conditions information and [Singh’s] testimony,” and concluded that Singh could relocate within India to avoid future persecution. Substantial evidence supports this conclusion. As the agency explained, record evidence in this case establishes that a low-level Mann party member like Singh who relocates within India typically does not have a well- founded fear of persecution from police, people from Punjab, or those outside of Punjab. Absent something unusual, low-level Mann party workers like Singh are not of interest to the police or people outside of Punjab; only hard- core militants are. Because Singh is admittedly not a hard- core militant, and is merely a low-level Mann party worker, it is unlikely that someone would harm him outside of Punjab. Moreover, the record is void of any evidence that anything about Singh in particular would be of interest to central Indian authorities, and it is unlikely that “the police in one state would share information about his location with opposition party supporters elsewhere in India.” Instead, the record contains evidence that “the Punjab police would require a court order … from the other state’s police in order to track someone who moves to a different state,” and the police “would likely only track someone in ‘extreme’ cases.” In addition, Singh is young, in good health, and was already able to relocate to the United States even though he does not speak the predominant language here. Hussain v. Rosen, 985 F.3d 634, 649 (9th Cir. 2021). To conclude that a finding that Singh is unable to safely relocate within India is compelled by the record, the majority first notes that “the DHS was required to demonstrate that  SINGH V. GARLAND 45 individuals like Singh who participate in Mann party activities or advocate for a separatist Khalistan state in other regions of India are unlikely to suffer persecution at the hands of police or opposition party members.” The majority then inexplicably claims that the “record is devoid of any such evidence,” simply ignoring the primary evidence the government submitted on this issue: a governmental report directly addressing the feasibility of relocation for Sikhs and Mann party members. That report specifically states that “only those considered by police to be high-profile militants are at risk of persecution even if they were to relocate.” And “simply holding pro-Khalistani views—favoring an independent Sikh state in Punjab—would not make an individual a high-profile suspect.” This evidence, offered by the government, directly addresses the issue the majority identifies. It is also individualized once it is coupled with Singh’s personal low- level involvement in pro-Khalistani activities. Indeed, our court has repeatedly relied on precisely this same evidence in reviewing and affirming agency decisions concluding that other low-level Mann party members like Singh here could safely relocate. See, e.g., Singh v. Garland, No. 22-171, 2023 WL 3220907, at *1 (9th Cir. May 3, 2023); Singh v. Barr, 808 F. App’x 408 (9th Cir. 2020); Singh v. Barr, 786 F. App’x 682 (9th Cir. 2019) (mem.); Gill v. Whitaker, 750 F. App’x 587, 588 (9th Cir. 2019) (mem.). The majority’s naked assertion that the record here is “devoid of any … evidence” that “individuals like Singh who participate in Mann party activities or advocate for a separatist Khalistan state in other regions of India are unlikely to suffer persecution” is not just obviously wrong. It also implies that many other panels of our court have been improperly denying petitions in other indistinguishable cases despite the 46 SINGH V. GARLAND fact that the agency decisions in those cases were supposedly similarly “devoid of any evidence” supporting those decisions. Absurd. And insulting to those panels. The majority suggests that the government’s report cannot serve as evidence that Singh would not be targeted by local authorities in a new state because this court in Singh v. Whitaker found this “same report wanting.” Again, the majority misreads Singh v. Whitaker. Nothing in that case found the report itself “wanting”—which would be strange indeed since, as documented above, our court has repeatedly relied on this same report as constituting substantial evidence supporting the agency’s decisions in many other cases involving Mann-party petitioners. What the panel in Singh v. Whitaker found inadequate was the BIA’s “analysis” in that particular case—not the report. See 914 F.3d at 661. The panel in Singh v. Whitaker wrote that “[a]lthough the BIA discussed the … Report and its conclusion that the police will likely pursue only ‘high-profile militants’ outside of Punjab, [the BIA] erred by failing to address the potential harm Congress Party members, or other local authorities, might inflict upon Singh in a new state.” Id. In other words, we faulted the BIA for not considering whether local authorities in a new state would persecute a petitioner, not for relying on the report (coupled with individual evidence about how the petitioner fits within the categories discussed by the report) to conclude that they wouldn’t. The report states that relocation is “feasible where the applicant’s fear is of local [Punjabi] police and the individual is not of interest to the central authorities.” Therefore, the report does provide evidence that Singh—who is a low-level party member, and has provided no reason why he would be of particular interest to authorities outside of Punjab—could feasibly relocate.  SINGH V. GARLAND 47 The majority next points out that, “even though Singh was a low-level party member, opposition party members specifically targeted him and had a particularized interest in harming him.” The majority argues that because “[h]is low- level status in the party did not protect him from past persecution” it therefore “does not support a conclusion that his low status will protect him from future persecution outside of Punjab.” But this argument rests on a mischaracterization of the BIA’s findings. The BIA did not find that opposing party members lacked a particularized interest in harming Singh in the abstract; the BIA concluded they lacked a particularized interest in him outside Punjab if he relocated. Therefore, the targeted incidents with opposing party members in Punjab, on which the majority relies, cannot compel a finding that such incidents would continue if he left Punjab. In sum, substantial evidence supports the agency’s conclusion that Singh could relocate within India to avoid future persecution, and the record does not compel a conclusion otherwise. 2. Singh v. Whitaker Does Not Require the Agency to Explicitly Address Whether a Petitioner Plans to Continue Proselytizing. The majority relies on Singh v. Whitaker to fault the BIA for not considering Singh’s “stated intent to continue proselytizing for his party wherever he went.” Given that (1) the BIA expressly said that it considered Singh’s testimony (which is the only evidence indicating he’ll continue proselytizing), and (2) our assumption that the BIA considered all relevant evidence unless something suggests or indicates to the contrary, the majority’s opinion can only be read as requiring the agency to explicitly mention whether 48 SINGH V. GARLAND a person will continue proselytizing. Indeed, the majority concludes that the “BIA’s failure to specifically address” Singh’s intent to continue proselytizing “rendered the BIA’s relocation analysis inadequate under” Singh v. Whitaker. That is an extreme and unwarranted reading of Singh v. Whitaker. Requiring explicit consideration of these matters would not only be inconsistent with Singh v. Whitaker itself, but also would be contrary to basic principles in our immigration precedent. In Singh v. Whitaker, the court faulted the BIA for considering only whether the petitioner would suffer harm from the Punjabi police outside of Punjab. 914 F.3d at 661. The BIA had not conducted any reasoned analysis regarding persecution the petitioner might experience by non-Punjabi police or other people outside of Punjab as a result of his future political activities. Id. Instead, the BIA completely ignored that Singh might face persecution in the future by people outside of Punjab. Id. In contrast, both the IJ and the BIA in this case expressly considered (1) whether police or other people from Punjab would persecute Singh in the future and (2) whether police or other people outside of Punjab would. The agency ultimately concluded that Singh’s individualized circumstances made him of no interest to opposition party members anywhere other than in Punjab. Moreover, the fact that Singh will continue proselytizing is not highly probative or potentially dispositive in this case. Najmabadi, 597 F.3d at 991. And although the agency did not explicitly address this fact, there is nothing to suggest that the agency did not consider it—to the contrary, there is indication that it did. The BIA specifically cited Singh v. Whitaker. Yet the majority strangely concludes that the BIA cited Whitaker  SINGH V. GARLAND 49 and then blatantly ignored what it says. Instead of applying the normal presumption that the agency followed the law, Kohli v. Gonzales, 473 F.3d 1061, 1068 (9th Cir. 2007), the majority applies the opposite presumption that even when the agency explicitly references the correct law, it’s probably not following it. Ultimately, the majority’s conclusion in this case makes no sense unless you read Singh v. Whitaker as imposing a “magic words” requirement that the agency must explicitly mention that someone intends to continue their political advocacy. I would not lightly read Whitaker that way, since that would be obviously contrary to basic principles in immigration law. We always presume the agency considered all relevant evidence even if it’s not explicitly mentioned. See Hernandez, 52 F.4th at 771; Najmabadi, 597 F.3d at 991. And underlying the presumption of a well- founded fear of future persecution is the assumption that the petitioner will continue being politically active in the new location. If the petitioner did not continue his political activity, that would arguably give rise to a changed circumstance finding that would overcome the presumption that the petitioner has a well-founded fear of future persecution. And even the majority does not appear fully committed to its position that the BIA’s failure to expressly mention that Singh intended to continue proselytizing is fatal to the BIA’s decision on its own. In the majority’s own words, any error here simply “compound[s] its mistake.” Ultimately, this case is yet another example of our court inventing trivial and nitpicky supposed shortcomings in the agency’s analyses in order to buy petitioners another round of proceedings before the immigration authorities, and all the resultant delay that entails. The record does not compel a conclusion contrary to the BIA’s. Rather, substantial 50 SINGH V. GARLAND evidence supports the conclusion that Singh can relocate within India to avoid future persecution, just as it has supported the same conclusion for literally dozens of other Mann Party Sikhs from Punjab seeking asylum. 5 5 I agree with the majority that Singh’s petition for review of his CAT claim should be denied.