Valdiviezo-Galdamez v. Attorney General of the United States

HARDIMAN, Circuit Judge,

concurring in the judgment.

I agree with my colleagues that the BIA’s decision in this case raises concerns warranting remand, but I write separately to express my understanding of the scope of the BIA’s discretion upon remand. In my view, the BIA is free to adopt the additional requirements of “particularity” and “social visibility,” exactly as the Board has defined and rationalized them over the last five years. The only problem that I find with the BIA’s evolving approach to “particular social group” cases is that the Board has failed to acknowledge a change in course and forthrightly address how that change affects the continued validity of conflicting precedent. Accordingly, remand is necessary so the Board can either choose between its reasonable new requirements and its older but equally reasonable precedents, or reconcile the two interpretations in a coherent way.

In addition, I am troubled by the BIA’s factfinding in this case. Should the BIA choose to adopt new requirements for “particular social group,” I believe that it must also remand to the IJ for further factual development.

I

A

It is settled law that Chevron deference applies to BIA interpretations of “ ‘ambiguous statutory terms’ ” in the INA. Negusie v. Holder, 555 U.S. 511 , 129 S.Ct. 1159, 1163-64 , 173 L.Ed.2d 20 (2009) (quoting INS v. Aguirre-Aguirre, 526 U.S. 415, 425 , 119 S.Ct. 1439 , 143 L.Ed.2d 590 (1999)) (citing Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43 , 104 S.Ct. 2778 , 81 L.Ed.2d 694 (1984)). We have recognized that “particular social group,” as used in the INA’s definition of “refugee,” 8 U.S.C. § 1101 (a)(42)(A), is so ambiguous that “[b]oth courts and commentators have struggled to define [it],” and “[rjead in its broadest literal sense, the phrase is almost completely open-ended.” Fatin v. INS, 12 F.3d 1233, 1238 (3d Cir.1993). There should be no question, then, that Chevron deference applies, as long as “the agency’s [reading of the statute] is based on a permissible construction.” Chevron, 467 U.S. at 843 , 104 S.Ct. 2778 .

The Supreme Court recognized in Chevron that “[a]n initial agency interpretation is not instantly carved into stone.” 467 U.S. at 863 , 104 S.Ct. 2778 . It is therefore possible for the BIA’s current interpretation of the statute to conflict with prior *613 decisions without constituting an “impermissible” or “unreasonable” reading of the INA. The BIA must, however, provide “explanation or an ‘avowed alteration,’ ” or its change “could be viewed as ‘arbitrary, capricious, [or] an abuse of discretion.’ ” Johnson v. Ashcroft, 286 F.3d 696, 700 (3d Cir.2002) (quoting INS v. Yang, 519 U.S. 26, 32 , 117 S.Ct. 350 , 136 L.Ed.2d 288 (1996)); see also Administrative Procedure Act, 5 U.S.C. § 706 (2)(A). We have held that an agency may change course and reinterpret statutes “as long as it can justify its change with a ‘reasoned analysis.’ ” Horn v. Thoratec Corp., 376 F.3d 163, 179 (3d Cir.2004) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 42 , 103 S.Ct. 2856 , 77 L.Ed.2d 443 (1983)). In FCC v. Fox, Television Stations, Inc., 556 U.S. 502 , 129 S.Ct. 1800 , 173 L.Ed.2d 738 (2009), the Supreme Court described the “reasoned analysis” requirement this way:

To be sure, the requirement that an agency provide reasoned explanation for its action would ordinarily demand that it display awareness that it is changing position. An agency may not, for example, depart from a prior policy sub silentio or simply disregard rules that are still on the books. And of course the agency must show that there are good reasons for the new policy. But it need not demonstrate to a court’s satisfaction that the reasons for the new policy are better than the reasons for the old one; it suffices that the new policy is permissible under the statute, that there are good reasons for it, and that the agency believes it to be better, which the conscious change of course adequately indicates. This means that the agency need not always provide a more detailed justification than what would suffice for a new policy created on a blank slate. Sometimes it must — when, for example, its new policy rests upon factual findings that contradict those which underlay its prior policy; or when its prior policy has engendered serious reliance interests that must be taken into account. It would be arbitrary or capricious to ignore such matters. In such cases it is not that further justification is demanded by the mere fact of policy change; but that a reasoned explanation is needed for disregarding facts and circumstances that underlay or were engendered by the prior policy.

129 S.Ct. 1800, 1811 (2009) (emphasis in original; citations omitted).

B

I agree with my colleagues that the BIA’s “particularity” and “social visibility” requirements are changes in position from the longstanding test the BIA articulated in Matter of Acosta, 19 I. & N. Dec. 211 (B.I.A.1985), which for over twenty years — from 1985 until 2006 1 — provided the most widely-adopted definition of “particular social group.” See generally Castillo-Arias, 446 F.3d at 1196 (listing the six circuit courts of appeals, including the Third Circuit, that “deferred to the Acosta formulation,” and two others that, “while not expressly deferring[,] ... viewed Acosta favorably”). In Acosta, the BIA stated the requirements for establishing “persecution on account of membership in a particular social group” as follows:

[W]e interpret the phrase ‘persecution on account of membership in a particular social group’ to mean persecution that is directed toward an individual who *614 is a member of a group of persons all of whom share a common, immutable characteristic. The shared characteristic might be an innate one such as sex, color, or kinship ties, or in some circumstances it might be a shared past experience such as former military leadership or land ownership. The particular kind of group characteristic that will qualify under this construction remains to be determined on a case-by-case basis. However, whatever the common characteristic that defines the group, it must be one that the members of the group either cannot change, or should not be required to change because it is fundamental to them individual identities or consciences.

19 I. & N. Dec. at 233 . “Particularity” and “social visibility,” as the BIA currently defines them, were not independent elements. 2

The BIA introduced “particularity” as a stand-alone requirement in In re C-A-, finding that “noncriminal informants” is a group “too loosely defined to meet the requirement.” 23 I. & N. Dec. 951, 957 (B.I.A.2006). “Social visibility,” on the other hand, was first mentioned in In re R-A- as a non-determinative factor — not a mandatory requirement — in the “particular social group” analysis. 22 I. & N. Dec. 906, 918-19 (B.I.A.1999) (using indefinite terms like “frequently,” “generally,” “less likely” to describe the requirement and its justifications, and stating that “[t]he factors we look to in this case, beyond Acosta’s ‘immutableness’ test, are not prerequisites”). 3 Like “particularity,” “social visibility” was applied again in In re C-A-, 23 I. & N. Dec. at 959-961 , and both became absolute requirements in In re A-M-E & J-G-U-, 24 I. & N. Dec. 69, 74 (B.I.A.2007) (referring to “the requirements of a particular social group” and “the requirement that the shared characteristic of the group generally be recognizable to others in the community.” (emphasis in original)), affd sub nom., Ucelo-Gomez v. Mukasey, 509 F.3d 70 (2d Cir.2007). Since In re A-M-E- & J-G-U- the BIA has treated these newly-minted elements as established precedent. See In re A-T- 24 I & N. Dec. 296, 303 (B.I.A.2007); Matter of S-E-G-, 24 I. & N. Dec. 579, 582 (B.I.A.2008); Matter of E-A-G- 24 I. & N. Dec. 591, 594-95 (B.I.A.2008).

Although the BIA frames “particularity” and “social visibility” as merely “additional considerations” within the Acosta framework, In re R-A- 22 I. & N. Dec. at 920 , or the products of “evolving case law,” Matter of E-A-G- 24 I. & N. Dec. at 593, in practice, they have become stringent requirements that can be outcome-determinative in cases like this appeal. Where, as here, an applicant seems to meet the Acosta requirements but is denied asylum because he fails to show “particularity” and “social visibility,” it appears that the *615 BIA has changed course from its earlier, less stringent Acosta approach.

C

The Majority holds that: (1) because “the requirement of ‘social visibility’ is inconsistent with a number of the BIA’s prior decisions[J ... [it] is therefore not entitled to deference under Chevron ” and “is an unreasonable addition to the requirements for establishing refugee status ... [based on] membership in a particular social group,” and (2) because “ ‘[particularity’ appears to be little more than a reworked definition of ‘social visibility[,]’ ... it suffers from the same infirmity as ‘social visibility.’ ” Maj. Op. at 603, 604, 608. I disagree. As discussed above, agencies are free to change their interpretations of statutes, so the fact that there is a conflict between “social visibility” and prior BIA decisions does not necessarily mean Chevron deference does not apply. Nor does it mean that the BIA’s definition of “social visibility” is “unreasonable.”

In keeping with Fox, I would hold that the BIA may reinterpret “particular social group” to include whatever new requirements it sees fit — including “particularity” and “social visibility,” defined exactly as they are in the line of cases from In re R-A- through Matter of S-E-G- and Matter of E-A-G — as long as it “displayfs] awareness that it is changing position” and “provide[s] reasoned explanation for its action.” 129 S.Ct. at 1811 (emphasis in original). In order to exhibit such an awareness, I believe the Board must make a choice between these new requirements and its prior decisions granting “particular social group” status to applicants who would likely have been unable to show “particularity” or “social visibility.” The Majority is correct that “the BIA’s present interpretation [of ‘particular social group’] would have excluded the asylum claims that were granted in” cases decided in 1996, 1990, and 1988. Maj. Op. at 607. However, the problem arises not because of that fact, but rather because of the BIA’s failure to recognize and address it.

I note that the BIA has, in my opinion, adequately explained the utility of adding “particularity” and “social visibility” to the Acosta test, and unlike the Majority, I am not convinced that the two requirements are identical. According to In re R-A-, requiring “social visibility” allows the BIA to limit asylum to those individuals whose “potential persecutors in fact see persons sharing the [applicant’s ‘social group’] characteristic as warranting suppression or the infliction of harm.” 22 I. & N. Dee. at 918. As the BIA noted, “[i]f a characteristic is important in a given society, it is more likely that distinctions will be drawn within that society between those who share and those who do not share the characteristic.” Id. at 919. In addition, “the social group concept would virtually swallow the entire refugee definition if common characteristics, coupled with a meaningful level of harm, were all that need be shown.” Id. “Social visibility” therefore serves to limit the scope of “particular social group” to more closely match the other protected characteristics of race, religion, nationality, and political opinion. Id. at 918. Likewise, “particularity” was explained in In re A-M-E- & J-G-U- as necessary to “delimit ... potential members” of the purported “social group” and to allow the BIA to deny asylum claims based on membership in groups, the defining characteristics of which are “simply too subjective, inchoate, and variable.” 24 I. & N. Dec. at 76. If the Board were writing on a blank slate, I would find that it has provided a reasoned explanation for a permissible interpretation of the law. 4

*616 But the BIA’s analysis comes undone when it states in conclusory fashion that all of the groups recognized as “particular social groups” in earlier cases would meet the “particularity” and “social visibility” requirements. See In re C-A-, 23 I. & N. Dec. at 960 (listing “young women of a particular tribe who were opposed to female genital mutilation,” “persons listed by the government as having the status of a homosexual,” “former members of the national police,” and “former military leadership or land ownership” as “social groups [that] involved characteristics that were highly visible and recognizable by others in the country in question”). If this is true — that all of the groups that have been recognized under the Acosta standard would be recognized under the new approach-then the otherwise reasonable definitions and applications of “particularity” and “social visibility” become, at best, muddled, and, at worst, incoherent.

The BIA has said that “[t]he essence of the ‘particularity’ requirement ... is whether the proposed group can accurately be described in a manner sufficiently distinct that the group would be recognized, in the society in question, as a discrete class of persons.” Matter of S-E-G-, 24 I. & N. Dec. at 584. This allows the BIA to weed out groups that are “too subjective, inchoate, and variable.” In re A-M-E- & J-G-U- 24 I. & N. Dec. at 76. In rejecting Galdamez’s proposed group based on a lack of particularity, though, the BIA described it as “ ‘potentially large and diffuse.’ ” App. at 11 (quoting Matter of S-E-G-, 24 I. & N. Dec. at 585). This suggests that “particularity” also embodies some kind of numerical or geographical limitation. If there are no such limitations, then it is unclear why it matters how “large” or “diffuse” a proposed group is. If such limits do exist, then it is unclear how the BIA can be sure, without hearing any argument on the matter, that “young women of a particular tribe who were op *617 posed to female genital mutilation,” “persons listed by the government as having the status of a homosexual,” “former members of the national police,” and “former military leaders[] or land owners[]” are any less numerous or widespread than “Honduran youth who have been actively recruited by gangs but have refused to join because they oppose the gangs.”

“Social visibility” has been defined as “the extent to which members of a society perceive those with the characteristic in question as members of a social group,” Matter of E-A-G-, 24 I. & N. Dec. at 594, and requires that “the group ... generally be recognizable by others in the community,” Matter of S-E-G-, 24 I. & N. Dec. at 586. It is unclear whether this means that the group’s shared characteristic must be visible to the naked eye (ie., pass the “eyeball test”) or just that the applicant’s society must understand individuals with the shared characteristic (visible or invisible) to be members of a group. In In re C-A-, the BIA suggested that “social visibility” is an eyeball test when it rejected a proposed social group because its shared characteristic is one “that is generally out of the public view.” In re C-A- 23 I. & N. Dec. at 960 . The BIA seemed to reaffirm this approach in Matter of E-A-G- , when it found that an applicant’s proposed group lacked “social visibility” because he “d[id] not allege that he possesses any characteristics that would cause others in [his] society to recognize him” as a member. But if “social visibility” is, or somehow accounts for, an eyeball test, then it is unclear how “young women of a particular tribe who were opposed to female genital mutilation,” “persons listed by the government as having the status of a homosexual,” “former members of the national police,” or “former military leaders[ ]” would qualify.

Announcing a new interpretation while at the same time reaffirming seemingly irreconcilable precedents suggests that the BIA does not recognize, or is not being forthright about, the nature of the change its new interpretation effectuates. It also unfairly forces asylum applicants to shoot at a moving target. 5 It is up to the BIA to bring some stability to its interpretation of the law by committing either to the Acosta line of cases or to the “particularity” and “social visibility” requirements, both of which are permissible and reasonable.

I agree with the Majority that the BIA “can[ ] add new requirements to, or even change, its definition of ‘particular social group,’ ” Maj. Op. at 608, but I also note that the BIA’s change can be the adoption of “particularity” and “social visibility.” Were it not for the seemingly irreconcilable conflict with prior decisions that the BIA has not yet disavowed, I would see no “reasonableness” problem with the “particularity” and “social visibility” requirements. Thus, the BIA may, upon further review, decide to jettison Acosta and its progeny or open them up to reconsideration. Conversely, the BIA may decide that its precedent should remain intact, in which case, “particularity” and “social visibility” must be refined or eliminated. 6 It *618 is not for us to make this choice between new requirements and precedent, nor is it our place to impose our own readings of the statute in an attempt to reconcile the two. Rather, the BIA should address these issues on remand, and we should defer to whatever conclusion it reaches— even if it is to reject precedent and move ahead with “particularity” and “social visibility” in their current forms — provided it explains itself in a way that exhibits expert consideration and logical, reasonable, permissible interpretations of the INA. See N.L.R.B. v. Curtin Matheson Scientific, Inc., 494 U.S. 775, 800 , 110 S.Ct. 1542 , 108 L.Ed.2d 801 (1990) (Blackmun, J., dissenting) (“Confronted with a court’s conclusion that two of its policy pronouncements are inconsistent, the agency may choose for itself which path to follow, or it may attempt to explain why no contradiction actually exists.”); Gatimi, 578 F.3d 611, 616 (7th Cir.2009) (“When an administrative agency’s decisions are inconsistent, a court cannot pick one of the inconsistent lines and defer to that one, unless only one is within the scope of the agency’s discretion to interpret the statutes it enforces or to make policy as Congress’s delegate. Such picking and choosing would condone arbitrariness and usurp the agency’s responsibilities.” (citations omitted)).

II

I also write separately to take issue with the BIA’s factfinding. We first remanded this case in 2007 “so the agency [could] address the issues that it did not reach ... [including] whether the group identified by Galdamez is a ‘particular social group’ within the meaning of the Act.” Valdiviezo-Galdamez v. Att’y Gen. of the U.S., 502 F.3d 285, 291 (3d Cir.2007) (Valdiviezo- Galdamez I). We did not authorize the BIA to usurp the IJ’s role as factfinder.

According to 8 C.F.R. § 1003.1 (d)(3)(iv), “[e]xcept for taking administrative notice of commonly known facts such as current events or the contents of official documents, the [BIA] will not engage in fact-finding in the course of deciding appeals.” See also Negusie, 129 S.Ct. at 1168 (“If the BIA decides to adopt a standard that [differs from the existing standard], it may be prudent and necessary for the Immigration Judge to conduct additional factfinding based on the new standard.” (emphasis added)); Padmore v. Holder, 609 F.3d 62, 67 (2d Cir.2010) (“[W]hen the BIA engages in factfinding in contravention of 8 C.F.R. § 1003.1 (d)(3)(iv), it commits an error of law, which we have jurisdiction to correct.”); Hashmi v. Att’y Gen. of the U.S., 531 F.3d 256, 262 (3d Cir.2008) (“[T]o the extent that the BIA’s decision rests in the alternative on its own finding of fact ..., it erred.” (footnote omitted)).

As of June 2005, when the IJ first heard Galdamez’s case, the only BIA decision discussing “social visibility” was In re R-A-, and that opinion had been vacated in early 2001. In re C-A-, with its enhanced definition of “particularity” and application of the “social visibility” requirement, was not published as precedential until June 15, 2006, exactly one year after the IJ rejected Galdamez’s petition. Thus, Galdamez had no reason to present evidence or argue facts relating to, for instance, whether “he possesses any characteristics that would cause others in Honduran society to recognize him as one who has refused gang recruitment” or whether his proposed group is too “large and diffuse” *619 to be considered “particular.” Nor did the IJ have any reason to make factual findings on these points.

On remand from our 2007 decision, the BIA papered over the lack of factual development by relying on factual findings made in Matter of S-E-G- and Matter of E-A-G- , both of which also post-dated Galdamez’s hearing before the IJ. The only “evidence” to which the BIA cited for its “particularity” and “social visibility” findings were quotations from those cases, coupled with eonclusory statements that the same facts apply to Galdamez’s case. For instance, the BIA quoted Matter ofEA-G- at length for propositions such as: “[T]here is no showing that membership in a larger body of persons resistant to gangs is of concern to anyone in Honduras, including the gangs themselves.” Id. (quoting Matter of E-A-G-, 24 I. & N. Dec. at 594-95). I pass no judgment on the truth of this statement or its application in Matter of E-A-G- , but I do not accept its blanket application to Galdamez’s case, where the evidentiary record was not developed with any notion of “particularity” or “social visibility” requirements. It remains to be seen whether Galdamez might be able to produce testimony, affidavits, or some other evidence not presented by the applicants in Matter of S-E-G- or Matter ofE-A-G-.

Under 8 C.F.R. § 1003.1 (d)(3)(iv), the IJ is the factfinder. Here, the BIA implicitly determined that Galdamez is identically situated to the asylum applicants in Matter of S-E-G- and Matter of E-A-G- , and that he would be unable to show “particularity” and “social visibility” because those two earlier applicants were unable to do so. That may be true, but that decision is for the IJ in the first instance. If, on remand, the BIA does change course and adopts new rules for asylum applicants, it must also remand Galdamez’s case to the IJ for additional factfinding. Galdamez’s appeal before the BIA should not be limited by an evidentiary record compiled under an outdated law, or by the facts and arguments raised by other applicants in cases other than his own. He should have the opportunity to present evidence with an eye towards the law under which his case is being decided.

Ill

In sum, I agree with the Majority that a remand is in order. The BIA now has a choice of either remaining faithful to its precedents or adopting new requirements that would likely produce different outcomes for future applicants claiming to be members of the same “particular social groups” as were recognized in earlier Board decisions. If it chooses the latter course, I would also instruct the BIA to remand the matter to the IJ so Galdamez can have a full and fair opportunity to be heard under the new legal standards.

. In re C-A- was originally decided in August 2004, but it was not published or designated as precedent until June 2006. 23 I. & N. Dec. 951 , 951 n. 1 (B.I.A.2006), aff'd sub nom., Castillo-Arias v. U.S. Att’y Gen., 446 F.3d 1190, 1196 (11th Cir.2006), cert. denied sub nom., Castillo-Arias v. Gonzales, 549 U.S. 1115 , 127 S.Ct. 977 , 166 L.Ed.2d 709 (2007).

. The word '‘particular” in the phrase "particular social group” was given no independent or operative meaning under the Acosta formulation. In my view, the BIA's recent decisions elevating "particularity” to its own requirement — along with the traditional Acosta requirements and "social visibility”— amounts to a change in the agency’s interpretation of "particular social group,” as a term of art.

. Although the opinion in In re R-A- was vacated by Attorney General Janet Reno in 2001, it remains instructive when considering the history of, and reasoned explanation for, the "social visibility” requirement. It is also worth noting that the BIA explicitly limited In re R-A- to its facts, stating that it "d[id] not intend any categorical rulings as to analogous social group claims arising under any other conceivable set of circumstances.” 22 I. & N. Dec. at 914 .

. I have two lingering questions about the provenance of "social visibility” that the BIA might address on remand. Although the BIA can sufficiently explain and justify "social visibility” without answering these questions, any light the BIA can shed on these issues might help courts of appeals in the future and alleviate some of the remaining doubt about the reasonableness of these new requirements.

The most convincing justification for the “social visibility” requirement is that "if the alleged persecutor is not even aware of the [asylum applicant's] group's existence, it becomes harder to understand how the persecutor may have been motivated by the victim’s 'membership' in the group to inflict the harm on the victim.” In re R-A-, 22 I. & N. Dec. at 919. This helps explain why “social visibility” would be a factor in determining whether persecution is "on account of” membership in a group. It remains unclear, however, why “social visibility” should be used to define the group in the first place.

Another oft-used justification for "social visibility” is that it is derived from the United Nations High Commissioner for Refugees's (UNHCR) interpretation of the INA. In re CA-, 23 I. & N. Dec. at 956 , 960 (citing UNHCR, Guidelines on International Protection: "Membership of a particular social

group" within the context of Article 1A(2) of the 1951 Convention and/or its 1967 Protocol relating to the Status of Refugees, U.N. Doc. HCR/GIP/02/02 (May 7, 2002), ("UNHCR Guidelines ”)); In re A-M-E- & I-G-U-, 24 I. & N. Dec. at 74 (same). The UNHCR Guidelines, however, treat "social visibility” as an alternative to Acosta as a way to establish a "particular social group”; it is not a requirement in addition to Acosta. See In re C-A-, 23 I. & N. Dec. at 956 ("The UNHCR Guidelines define a 'particular social group' as 'a group of persons who share a common characteristic other than their risk of being persecuted, or who are perceived as a group by society.' " (emphasis added) (quoting UNHCR Guidelines at ¶ 11)). Why, then, has the BIA decided to turn the Guidelines’ disjunctive into a conjunctive, essentially creating an “Acosta-plus” test, rather than adopt the “Acosta-or” test endorsed by the UNHCR?

. Although the BIA noted in Acosta that "[t]he particular kind of group characteristic that will qualify under th[e Acosta ] construction remains to be determined on a case-by-case basis,” 19 I. & N. Dec. at 233 , I assume that the BIA was referring to case-by-case evaluation of individual applicants' proposed groups and whether they meet the established legal standard. If the BIA is permitted to engage in case-by-case ad hoc revisions of the entire legal framework — without acknowledgement or explanation — then it would be free to arbitrarily pick and choose whatever statutorily-permissible construction of "particular social group” it finds agreeable at the moment.

. I acknowledge that there may also be some way for the BIA to reconcile its new interpretation and its precedent, and I do not mean to *618 suggest that such a reconciliation automatically renders the BIA’s explanation defective. My point is simply that more analysis is needed before I can conclude that the BIA has provided a reasoned explanation of how its new rules fit — or, it appears, do not fit — with the old ones.