Johana Cece v. Eric Holder, Jr.

EASTERBROOK, Chief Judge,

dissenting.

Cece defines, as the “social group” at risk of persecution, “young Albanian women in danger of being trafficked as prostitutes.” At earlier stages of these proceedings she made different proposals, but this is the definition in her appellate briefs. My colleagues hold that the Board of Immigration Appeals erred by not treating “young Albanian women who live alone” as her social group. Put to one side the fact that Cece does not ask us to define a social group that way. Whether the group is “young Albanian women in danger of being trafficked as prostitutes” or “young Albanian women who live alone”, Cece isn’t in it. Her own expert defined “young” as 16 to 26 or 27. Cece is 34. The basis for her claim of asylum is future risk; she does not argue that she suffered persecution before leaving Albania, so the fact that she is not a member of her own proposed group should be dispositive. (Perhaps Cece looks younger than her age and would be targeted by mistake, but she does not argue this.)

Then there is the question “how much risk is too much?” Cece’s expert did not attempt to quantify the risk that young Albanian women living alone face, nor does the majority. That many of Western Europe’s prostitutes are Albanians does not tell us how many are in the sex trade involuntarily. The State Department tries to estimate risk. Its Human Trafficking Report 2012 finds that 84 complaints about trafficking were made to Albanian public agencies during 2011. Report at 64. Nongovernmental groups (NGOs) reported more: they counted 132 Albanian traffick *679 ing victims in 2011. Ibid. The number of young women living alone in Albania is substantially higher. The State Department ranks nations into four tiers (1, 2, 2 Watch List, and 3), with Tier 1 representing the best performance. Id. at 51. Albania is in Tier 2, as are Greece, Hong Kong, Japan, Switzerland, and more than 60 other nations. Id. at 52. Fifty-one nations are in tiers 2WL or 3, below Albania. Ibid. Deplorable as human trafficking is, any given woman’s danger in Albania may be modest.

Whatever risk Cece faces comes from criminals, not from the government, yet “persecution” means mistreatment at public hands. See Hor v. Gonzales, 421 F.3d 497 (7th Cir.2005); Bitsin v. Holder, 719 F.3d 619, 628-31 (7th Cir.2013). Crime may be rampant in Albania, but it is common in the United States too. People are forced into prostitution in Chicago. See, e.g., United States v. Cephus, 684 F.3d 703 (7th Cir.2012). Must Canada grant asylum to young women who fear prostitution in the United States, or who dread the risk of violence in or near public-housing projects? If there were reason to think the Albanian government in cahoots with the traffickers, Cece would have a better case; but when the record shows no more than ineffective law enforcement, there’s no basis to infer persecution. Meghani v. INS, 236 F.3d 843, 847 (7th Cir.2001).

I can see why we ought not make anything turn today on the facts that Cece is 34 years old, that the number of trafficking victims in Albania may be under 150 annually, and that any risk comes from private criminals rather than public policy: the BIA did not do so, and the Chenery doctrine (SEC v. Chenery Corp., 318 U.S. 80, 87 , 63 S.Ct. 454 , 87 L.Ed. 626 (1943)) limits reviewing courts to the agency’s grounds of decision. Perhaps the Board will consider these issues on remand.

When tackling this subject, the Board may wish to consider whether a government’s inability to protect people from criminals is a form of persecution. Equating inability to control crime with unwillingness to do so (a form of persecution because it reflects public policy disfavoring a person or group) first appeared in a decision of the Board in 1964. See Matter of Eusaph, 10 I. & N. Dec. 453, 454 (1964). The formula has been repeated many times, e.g., Matter of Pierre, 15 I. & N. Dec. 461, 462 (1975); Matter of Kasinga, 21 I. & N. Dec. 357, 365 (1996), without explaining why “unable” has the same effect as “unwilling” — or quantifying what “unable” means. The Board appears to be happy with this formula, but its utility is limited when we do not know how much shortfall in law enforcement counts as “inability” to protect citizens.

A remand is unnecessary even on the majority’s views about the social-group question, however. The Board found that Cece could live safely in Tirana, though perhaps not in her parents’ city. Part III of the majority’s opinion declares that this decision is not supported by substantial evidence. As Judge Manion shows, however, the Board had, and gave, the best of all possible reasons: that Cece had moved to Tirana and, was not followed or accosted there. Indeed, Cece does not even contend that the person who pursued her in Korgé learned that she was in Tirana or attempted to locate her (or anyone else) outside of Korgé. Cece’s expert witness testified that Tirana is a collection of enclaves and that people find things out; this might have led the Board to conclude that Cece was at risk there after her sister left. But the inference is permissive, not compulsory. An agency is entitled to give more weight to what actually happened than to what could have happened. Cece’s untroubled time in Tirana is “substantial *680 evidence” for the Board’s conclusion. See INS v. Elias-Zacarias, 502 U.S. 478, 481 , 112 S.Ct. 812 , 117 L.Ed.2d 38 (1992) (if the record allows reasonable disagreement, the Board’s decision must stand).

Cece’s brief in conjunction with the rehearing en banc makes a different argument: that it is not reasonable to require her to relocate to Tirana, even if she would be safe. See 8 C.F.R. § 1208.13 (b)(2)(ii). According to Cece, the Board should have deemed relocation unreasonable because she had no relatives in Tirana after her sister left. Yet every year millions of persons move to cities where they are strangers; they make new friends (or acquire new relatives) afterward. A person who left Tirana for Rome, and then left Rome for Chicago, is hard pressed to contend that it would not be “reasonable” to think that she could have lived closer to her relatives, even if none was in the neighborhood. (Korgé and Tirana are 110 miles apart.)

Although I think the majority mistaken in its treatment of Cece’s specific claim, I am more concerned by its treatment of the Board’s doctrine. My colleagues recognize that the statute does not define the term “social group” and that Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 , 104 S.Ct. 2778 , 81 L.Ed.2d 694 (1984), therefore applies to the Board’s gap-filling work. See Holder v. Martinez Gutierrez, — U.S.-, 132 S.Ct. 2011, 2017 , 182 L.Ed.2d 922 (2012); INS v. Aguirre-Aguirre, 526 U.S. 415, 424-26 , 119 S.Ct. 1439 , 143 L.Ed.2d 590 (1999). Yet the dispositions of this and other cases demonstrate that the Seventh Circuit has rejected the Board’s approach and established its own — one under which everyone belongs to a “social group” and the question whether that membership caused the persecution drops out of consideration. (It drops out because, when the asserted criteria of persecution are used to define the “social group,” the process is circular.)

The Immigration and Nationality Act permits federal officials to grant asylum to aliens who seek refuge here “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). Cece does not contend that her race, nationality, or politics mattered to the traffickers (or to any Albanian public official), and although she initially argued that she was at risk because of her religion she has abandoned that contention. This leaves “social group.”

What the Seventh Circuit has done in this and other recent cases is read “because of ... membership in a particular social group” in a way that includes everyone threatened by criminals, rebels, or anyone else a nation’s government does not control. This makes eligible for asylum everyone who faces a substantial risk of harm in his native land, no matter the reason.

The Board of Immigration Appeals has established, in decisions the court concedes we must respect, several requirements of social-group status. One is that a “social group” entails a characteristic that is either immutable or so important that no one should be required to change it. See Matter of Acosta, 19 I. & N. Dec. 211, 233-34 (1985); Matter of Kasinga, 21 I. & N. Dec. 357, 365-66 (1996). Age changes; that’s why decisions such as Vance v. Bradley, 440 U.S. 93 , 99 S.Ct. 939 , 59 L.Ed.2d 171 (1979), reject arguments that age must be treated like race or sex for legal purposes. Whether a person lives alone also is subject to change. People may marry, live with relatives, or join forces with similarly situated persons. Many single women live with other single *681 women. A group such as “young Albanian women who live alone” therefore flunks the Board’s test on multiple grounds, even if we treat marital status as the sort of thing that no one should be required to change.

Another series of cases in this circuit expands the scope of “social group” by a different route. It asks whether the alien used to be at risk. For example, Escobar v. Holder, 657 F.3d 537 (7th Cir.2011), holds that an alien is eligible for asylum as a member of a “social group” that comprises truckers who ever opposed a band of Colombian rebels. Events of a decade ago cannot be changed; the past is “immutable”; thus the Board’s primary defense against limitless expansion of “social group” vanishes. Everyone who seeks asylum in the United States can point to some event in the past, and as the past can’t be changed this event becomes the basis for a claim based on an “immutable characteristic.” See, e.g., Sepulveda v. Gonzales, 464 F.3d 770 (7th Cir.2006) (former employees of a public agency are a social group); Benitez Ramos v. Holder, 589 F.3d 426 (7th Cir.2009) (former criminal-gang members are a social group).

Some people might be inclined to ask: Why not just treat everyone as belonging to a social group and skip to the question whether persecution occurred? The answer is that the statute makes membership in a group (or classification by race, religion, or politics) essential to analysis of the supposed persecution. The agency must decide whether a person has been persecuted “on account of’ membership in the group (or because of politics, race, etc.). You can’t sensibly ask about cause without deciding what differentiates the applicant from other persons. To know whether X has been persecuted on account of Y, it is essential to know what Y is. The Board has tried to define groups by fixed attributes (such as “member of the Yoruba tribe” or “born in Korge”). It is only after defining a group that it becomes possible to ask the statutory question: whether membership in that group is the reason for the adverse treatment.

Cece is a member of one social group that the Board probably would acknowledge: Albanian women. But she does not contend that she fears persecution because of those characteristics. She does not say that the government of Albania persecutes Albanian women. Indeed, she does not contend that Albania discriminates in any way by national origin or sex. She does not maintain that police and courts protect male victims of crime but not female victims; instead she tells us that Albania’s system of law enforcement is weak. Failure to achieve optimal deterrence is unfortunate but not “persecution” by any useful understanding. That’s why Cece proposed a group such as “young Albanian women in danger of being trafficked as prostitutes.” The qualifications that distinguish this proposal from “all Albanian women” — age, living alone, and the criminal enterprise of sex traffickers — all fail the Board’s filters.

The BIA has held that a “social group” cannot be identified by asking who was mistreated. Matter of C-A-, 23 I. & N. Dec. 951, 956 (2006). For if the persecutors’ acts define social groups, then again § 1101(a)(42)(A) effectively offers asylum to all mistreated persons, whether or not race, religion, politics, or some extrinsically defined characteristics (such as tribal membership) account for the persecution. And again this court professes to accept the Board’s position — though with the proviso that it applies only when the persecutors’ acts are the entire definition (opinion at 671-72, which uses “only” or “solely” or “merely” four times, putting three of the four in italics for emphasis). Thus although the Board concluded that “young *682 Albanian women in danger of being trafficked as prostitutes” flunks, the majority-rules otherwise because “danger of being trafficked as prostitutes” is not the sole component of the definition. That is not what Chevron requires. We have not applied the Board’s definition. We have rewritten it.

Under this court’s approach, any person mistreated in his native country can specify a “social group” and then show in circular fashion that the mistreatment occurred because of membership in that ad hoc group. Anyone threatened or injured in the past has an “immutable” characteristic (the past can’t be changed), and the selection criteria used by the persecutor (here, people who want to force others into prostitution) become the defining characteristics of the “social group”. The structure of § 1101(a)(42)(A) unravels.

The majority accuses the Board of inconsistency (opinion at 676-77), but the BIA has been inconsistent by the court’s standard, not its own. For example, the “social group” in Kasinga, which the court calls “not unlike” Cece’s (opinion at 676), was a tribe. The Board sees a big difference between tribal groups (membership is immutable and extrinsic to the choices made by criminals) and “young Albanian women in danger of being trafficked as prostitutes” (defined in part by changeable characteristics and in part by who criminals target). The majority says that it does not see a difference, so the Board must be inconsistent. That’s a statement about judicial rejection of the Board’s doctrine, not about how the Board administers its own approach.

I grant that some of the inconsistency is real — it has been forced on the Board by judicial refusal to accept its approach to defining “social group.” Our court has discarded not only the immutability and don’t-use-the-wrongdoers’-perspeetive rules but also another component of the Board’s definition: social visibility. See, e.g., Gatimi v. Holder, 578 F.3d 611, 614-15 (7th Cir.2009); majority opinion at 668 n. 1. Other circuits have allowed the Board to use the “visibility” criterion but have revised or rejected different parts of the Board’s definition of “social group.” See, e.g., Ucelo-Gomez v. Mukasey, 509 F.3d 70 (2d Cir.2007) (approving the “visibility” standard and adding that changeable attributes such as wealth do not identify a social group); Castillo-Arias v. Attorney General, 446 F.3d 1190 (11th Cir.2006) (denying a petition to set aside the decision in CA-). The Board tries to apply circuit law — and, when it does so, is accused of abandoning its own definition or applying it inconsistently.

The majority says that it accepts the Board’s approach. Yet in case after case, of which today’s is just a sample, we set aside the Board’s decisions. I have already mentioned Escobar, Gatimi, Sepulveda, and Benitez Ramos . Here are a few more: Sarhan v. Holder, 658 F.3d 649 (7th Cir.2011); Torres v. Mukasey, 551 F.3d 616 (7th Cir.2008); Agbor v. Gonzales, 487 F.3d 499 (7th Cir.2007); Tapiero de Orejuela v. Gonzales, 423 F.3d 666 (7th Cir.2005); Yadegar-Sargis v. INS, 297 F.3d 596 (7th Cir.2002). The meaning of a legal standard lies in its application to concrete facts. That the BIA and this court regularly reach different decisions on identical facts shows that they are applying different legal standards.

This is a particularly poor case for our court to nix the Board’s approach, because at least two other circuits have held that the agency properly denied asylum claims identical to Cece’s. The majority concedes (opinion at 672) that its decision conflicts with Rreshpja v. Gonzales, 420 F.3d 551, 555-56 (6th Cir.2005). It also conflicts with Gjura v. Holder, 502 Fed.Appx. 91 *683 (2d Cir.2012). The Second Circuit’s initial opinion, 695 F.3d -223, held that the Board acted within its discretion in concluding that “young, unmarried Albanian women-is not a social group. The court- then issued a replacement opinion denying the petition on the ground that sex traffickers are private actors whose criminal conduct does not demonstrate persecution by public officials.- Cece would have lost in the Second Circuit for either of these reasons. As far as I can see this circuit stands alone in dismantling the BIA’s approach so thoroughly that the agency must recognize social groups such as “young Albanian women in danger of being trafficked as prostitutes” or “young Albanian women who live alone” and treat members of that group as victims of persecution.

The majority expresses sympathy for Cece and. other applicants for asylum. Yet the choice whether to be strict or lenient belongs to the agency, not to the court. See, e.g., INS v. Phinpathya, 464 U.S. 183 , 104 S.Ct. 584 , 78 L.Ed.2d 401 (1984); INS v. Jong Ha Wang, 450 U.S. 139 , 101 S.Ct. 1027 , 67 L.Ed.2d 123 (1981). The Board has chosen to make “social group” do real work. This court effectively reads it out of the statute and directs the agency to ask only whether an alien faces a significant risk for any reason. This intrudes on the sort of choice Congress has committed .to the Executive Branch. The Attorney General could direct the Board to ditch Acosta and C-A- but as long as the political branches of government stand by them, Chevron requires the judiciary to implement their choices.

One final observation. Cece entered the United States by fraud, pretending to be from a nation whose citizens do not need visas to visit. See Bayo v. Napolitano, 593 F.3d 495 (7th Cir.2010) (en banc). She started that journey from Rome and has never contended that she feared sex trafficking in Italy. Her fraud thus is not mitigated by a need to escape from danger. The Board has concluded that entering the United States by fraud, when danger is not imminent, is a strongly adverse factor in the discretionary decision whether to grant asylum. See Alsagladi v. Gonzales, 450 F.3d 700 (7th Cir.2006); Matter of Pula, 19 I. & N. Dec. 467 (1987). Although the court decides today that Cece is eligible for asylum, it does not hold that she is entitled to it; that question, at least, remains open to decision on remand.