Jeffrey S. Chase | Opinions/Analysis on Immigration Law

Blog

To listen to the articles press the play button.

Nexus Update: Matter of M-R-M-S- is Vacated

On August 26, 2025, the U.S. Court of Appeals for the Tenth Circuit issued its decision in O.C.V. et. al. v. Bondi.1 This case was the direct appeal of the BIA’s precedent decision in Matter of M-R-M-S-2. In O.C.V., the Tenth Circuit thus vacated the Board’s precedent decision itself. In doing so the Tenth Circuit stated: “We agree with Petitioners that the BIA stated an erroneous standard for determining whether family membership ‘was or will be at least one central reason for’ persecution under the Immigration and Nationality Act (INA).

This is big news. According to a practice advisory issued a few years ago by CLINIC and Duke Law School’s Immigrant Rights Clinic, when a BIA precedent decision itself is vacated on appeal by a circuit court, “the Board’s decision has no effect anywhere in the United States.” The advisory cited Harmon v. Thornburgh, 878 F.2d 484, 495 n.21 (D.C. Cir. 1989) (stating “When a reviewing court determines that agency regulations are unlawful, the ordinary result is that the rules are vacated—not that their application to the individual petitioners is proscribed.”)”3

I want to give credit to the legal team involved in this decision. Anne Dutton and Anne Peterson of the Center for Gender and Refugee Studies, and Victoria Nielson of the National Immigration Project of the National Lawyers Guild represented the Petitioners. Special thanks are offered to Erik Kundu and Rebecca Human of Perkins Coie LLP for representing our Round Table of Former Immigration Judges on our amicus brief filed in the case. And Peter Alfredson and Amelia Dagen filed an amicus brief on behalf of the Amica Center for Immigrant Rights.

Even if Immigration Judges and the BIA ignore the above guidance and seek to continue to apply the holding of M-R-M-S- in circuits that have not yet rejected its nexus standard, it is worth discussing what might be left of the decision at this point to apply.

I say this because the Board itself admitted in M-R-M-S- that its nexus standard is at odds with the Fourth Circuit’s views on the subject. The Board offered the Tenth Circuit’s view as conflicting with that of the Fourth Circuit, and concluded “In our view, the Tenth Circuit’s approach is the proper way to analyze whether membership in a family-based particular social group is one central reason for harm.”4  But now the Tenth Circuit has rejected the Board’s standard as well (although for different reasons than the Fourth Circuit).

But there’s more. The viability of the Board’s standard suffered additional blows in published decisions by the Third and Seventh Circuits that preceded the Tenth Circuit’s decision in O.C.V.

Furthermore, in its decision, the Tenth Circuit also acknowledged that the Sixth and First Circuits were in disagreement with the Board’s standard. So that’s a lot of rejection.

I will review all of these decisions below for the benefit of practitioners and adjudicators in the relevant circuits. But let’s first take a step back and review what M-R-M-S- held.

Matter of M-R-M-S-: Did the BIA apply a secret “shadow standard” for nexus?

I wrote an in depth analysis of M-R-M-S- shortly after its issuance (the link is here). I want to focus at present on the decision’s reassertion of both an animus and a subordination test for nexus where an asylum claim is based on a particular social group of family.

A recent law review article used the term “shadow docket” in reference to the BIA’s issuance of thousands of unpublished decisions.5 I believe the current BIA also applies “shadow law,” consisting of principles not found in current statute, regulation, or binding case law, and which in fact may contradict applicable law, but which the Board nevertheless subscribes to and tries to slip by the censors, so to speak.

One example is found in the legislative history of section 208(b)(1)(B)(i) of the Immigration and Nationality Act (“the Act”), which was added by Congress in 2005 as part of the REAL ID Act. As the book I co-author with Deborah Anker points out, an earlier version of this statute that was introduced in 2004 required that a protected ground ‘was or will be the central motive for persecuting the applicant.’”6 That version was not acted on by Congress.

The following year, the 109th Congress changed “the central motive” to “at least one central reason,” which is the language used in the statute. The change eliminated the need for the protected ground to be the sole central reason, allowing instead for multiple reasons for the persecution, of which the protected ground need not top of the list. Furthermore, the change of wording from “motive” to reason” shifted the focus away from examining the persecutor’s specific mindset to a broader consideration of reason without regard to motive. As I like to point out, if I slip and fall on ice, the ice is the reason I fell, even though it had no motive to cause me to do so.

However, the BIA seems to prefer applying the draft language rejected by Congress, requiring the protected ground to be “the central motive” for the persecution. That was the unstated standard applied by the Board in M-R-M-S-.

My reasons for saying this are as follows.

First, as I’ve written previously, in its initial precedent decision addressing the REAL ID Act’s “at least one central reason” standard, the Board recognized that that the standard did not require a central reason to be “dominant” in relation to other reasons for persecution; the Board actually conceded that “Congress purposely did not require that the protected ground be the central reason for the actions of the persecutors.”7

Yet later in the same decision, the Board tried to slip a dominance test by the censors, holding that a protected ground "cannot be incidental, tangential, superficial, or subordinate to another reason for harm” in order to satisfy the standard.8  Note the inclusion of the word “subordinate.”

The U.S. Court of Appeals for the Third Circuit called out this conflict in rejecting the Board’s standard, stating that there was no difference between saying a protected ground could not be subordinate to another ground and requiring it to be the dominant ground (which the Board had already noted Congress intentionally did not require).9

In subsequent precedent decisions, the BIA dropped the “subordinate” requirement when stating the standard for nexus. It should be noted that the Board did not officially declare that it was amending its previously stated standard from J-B-N- & S-M- for the reasons flagged by the Third Circuit; rather, it elided over the word “subordinate” in quoting the standard from that case  so as to read that the protected ground “cannot be incidental [or] tangential . . . to another reason for harm.”10

But in its decision in M-R-M-S-, the Board returned the word “subordinate” to the standard it quoted for nexus, which in practical terms meant it was applying the shadow “the central motive” standard that Congress intentionally chose not to use.11

And I say “motive” as opposed to “reason” because, in spite of holding way back in 1996 in Matter of Kasinga that a “subjective ‘punitive’ or ‘malignant’ intent is not required for harm to constitute persecution” for asylum purposes,12 the current BIA’s requirement of looking to the persecutor’s “ultimate goal” and its stated “animus test” for determining nexus to a group consisting of family is entirely focused on motive.

The Tenth Circuit rejected the subordination requirement.

Interestingly, the Tenth Circuit in O.C.V. did not find the Board to have erred for focusing on motive. This is because the court adopted a very broad view of the meaning of the term “animus.” Instead of the common definition (such as that found in the Cambridge online dictionary of “a feeling of hate or anger toward someone or something”)13, the Tenth Circuit took the term to mean an intent “to overcome an undesired characteristic related to a protected group” even where the persecutor harbors no ill will towards the victim, or even believes “that the persecution is in the applicant’s best interest.”14 Based on that last point, we might refer to the Tenth Circuit’s approach as the “killing them with kindness” standard.

But in spite of giving the Board a free pass on “animus,” the court vacated the decision for applying an erroneous legal standard because of the Board’s reinsertion of the word “subordinate” into its stated legal standard for nexus. So once again (as with the Third Circuit’s decision in Ndayshimiye), the Board failed to slip that word past the censors.

The Tenth Circuit read the standard stated by the Board in M-R-M-S- as follows:

The BIA’s new rule says an asylum applicant cannot show nexus if a motive unrelated to a protected ground exists. By contrast, the INA accommodates the understanding that— as will be often the case—a persecutor may be centrally motivated by both a protected ground and an unprotected one.15

In other words, the court called out the Board for applying a “the central motive” test - the Board’s preferred “shadow law” standard.

How the BIA’s own Romanov example helped sink its preferred standard.

When the BIA first put forth its “animus” requirement for families, it provided only one example of when that standard might be satisfied. Wrote the Board: “For illustrative purposes, we consider the well-known historical scenario of the Bolshevik assassination of Czar Nicholas II, his wife, Czarina Alexandra, and their five children after he abdicated the throne in 1917.”16

Rather than the Romanovs, the Board’s standard more closely resembles Rasputin in its staying power after so many circuit rejections. However, the Tenth Circuit agreed with the argument made in our Round Table’s amicus brief that in fact, the Romanovs themselves would fail the nexus standard applied by the Board in M-R-M-S-.

After first stating that the Board’s nexus standard “articulated an if-then formulation: ‘If a persecutor is targeting members of a certain family as a means of achieving some other ultimate goal unrelated to the protected ground, family membership is incidental or subordinate to that other ultimate goal and therefore not one central reason for the harm,” the court then turned to the Board’s Romanov example:

A hypothetical demonstrating how the BIA's rule works in practice helps further show why it is incorrect. Consider what the BIA in L-E-A- used as a quintessential example of nexus to a family-based PSG: “the well-known historical scenario of the Bolshevik assassination of Czar Nicholas II, his wife, Czarina Alexandra, and their five children after he abdicated the throne in 1917.” 27 I. & N. Dec. at 44. As the BIA explained, “This is a classic example of a persecutor whose intent, for at least one central reason, was to overcome the protected characteristic of the immediate family.” Id. Yet this persecution could also be framed as a means to other, unprotected ends, like fomenting political change. Id.

Consider what happens when this factual scenario is evaluated under the BIA's new rule. The antecedent (“if”) condition is satisfied, because “a persecutor is targeting members of a certain family as a means of achieving some other ultimate goal unrelated to the protected ground”: political change…Under the BIA's formulation, upon finding that motive, the inquiry prematurely ends, and the agency must then simply conclude “family membership is incidental or subordinate to that other ultimate goal and therefore not one central reason for the harm…Thus, even this “classic example” of nexus in a mixed-motive case, L-E-A-, 27 I. & N. Dec. at 44, fails under the BIA's stated rule.17

The Board’s approach to nexus in family-based cases was dealt two additional blows in published decisions by the Third and Seventh Circuits that preceded the Tenth Circuit’s decision in O.C.V.

The Third Circuit rejects both the subordination and animus requirements.

On June 30, the Third Circuit in Tipan Lopez v. Att’y Gen. of the U.S. remanded a religious-based asylum claim with instructions to the BIA not to apply either a subordination or an animus test for nexus. In its decision, the court took M-R-M-S- to task for reinserting the word “subordinate” into its stated test for nexus, noting that the standard is directly in conflict with the court’s earlier decision in Ndayshimiye.18

But the court continued that  “[t]he  IJ indeed  applied  an  impermissible  animus-based test for nexus.” In finding the animus requirement inconsistent with prior BIA precedent (i.e. its holding in Matter of Kasinga that “subjective ‘punitive’ or ‘malignant’ intent is not required for  harm  to  constitute  persecution”), as well as with DHS guidance to its asylum officers, and the UNHCR Handbook (“There is no need for the persecutor to have a punitive intent to establish the causal link.”), the court focused on the “reason versus motive” distinction, providing two examples (one involving gang members seeking to recruit an indigenous person, the other involving a terrorist targeting the family of an escapee) in which the reason requirement would be satisfied for nexus purposes in spite of the absence of animus.19

And interestingly, the majority on the panel rejected a dissenting judge’s suggestion that the term “animus” could have a meaning similar to that later employed by the Tenth Circuit in O.C.V., respondent “We are operating in a field of law—immigration—in which ‘animus’ has a specific meaning,” that meaning being “hatred or hostility,” and not “intent or motivation.”20

Although the Board in M-R-M-S- claimed that the Third Circuit “generally agrees” with its standard (apart from its use of the word “subordinate” rejected in Ndayshimiye),21 that claim does not survive the court’s decision in Tipan Lopez, which clearly rejects both the subordination and animus requirements of the Board’s nexus standard.

The Seventh Circuit agrees with the Fourth Circuit’s approach

On July 17, in Mejia-Hernandez v. Bondi, the Seventh Circuit reversed the agency’s denial of asylum in a case involving persecution on account of family membership. Without referencing Matter of M-R-M-S-, the court rejected the agency’s view that the reason for the feared persecution was a “private dispute,” holding to the contrary that the petitioner’s “family relationships were the reason she, and not another individual in Honduras, was targeted…”22

In its legal analysis, the Seventh Circuit referenced its prior finding of nexus where the petitioner’s “relationship to his wife was the reason he, and not someone else, was targeted.”23

The standard cited is the exact one that has been repeatedly applied by the Fourth Circuit, which is directly at odds with the Board’s animus test for nexus.24

This is significant, because in M-R-M-S-, the Board treated the Fourth Circuit as something of  an outlier in its application of the traditional “but for” test for nexus, which looks to whether the asylum seeker would not have been at risk of persecution “but for” their family membership. But that is the exact standard the Seventh Circuit applied in Mejia. And the Seventh Circuit noted that it was not applying this standard for the first time, referencing its 2019 decision in Gonzales-Ruano v. Barr.25  Quoting the standard it had applied in that earlier case, the court stated:  “As we put it, ‘Gonzalez Ruano's relationship to his wife was the reason he, and not someone else, was targeted.’”26 Again, the quoted language invokes the same standard applied by the Fourth Circuit.

And what about those SIxth and First Circuit Cases mentioned in O.C.V.?

Sixth Circuit

In O.C.V., the Tenth Circuit referenced the Sixth Circuit’s decision in Mazariegos-Rodas v. Garland,27 which concluded that the nexus rule applied by the BIA in M-R-M-S- “is ‘overly restrictive’ in terms of what asylum claims it allows and ‘unnecessarily broad’ in terms of what it disallows.”28

Looking at the Sixth Circuit’s decision itself, the court further stated that “an overly restrictive nexus standard contradicts the BIA's own pronouncements that ‘[i]n adjudicating mixed motive cases, it is important to keep in mind the fundamental humanitarian concerns of asylum,” and that “[s]uch an approach is designed to afford a generous standard for protection in cases of doubt.’”29

First Circuit

The court in O.C.V. also acknowledged a First Circuit decision that found the animus requirement of M-R-M-S- inconsistent with its circuit precedent, citing to Mayancela v. Bondi.30 In a footnote contained in that decision, the First Circuit stated that the nexus standard of M-R-M-S-  “requiring proof of ‘animus' has been definitively rejected by this circuit since the agency's decision.”

Conclusion

So besides from M-R-M-S- itself being vacated and, according to the above-cited practice advisory and the case law it cites, therefore being inapplicable anywhere, the principles it expresses - i.e. the subordination and animus requirements (which effectively combine to form the “shadow law” standard wrongly requiring a protected ground to be “the central motive”) have been rejected either in whole or in part by the First, Third, Fourth, Sixth, Seventh, and Tenth Circuits.

I hope that this post, in conjunction with the one I wrote just after M-R-M-S- was issued (which suggests that M-R-M-S- conflicts with case law of additional circuits), will prove useful in litigating particular social group claims based on family.

Copyright 2025, Jeffrey S. Chase. All rights reserved.

Notes:

  1. No. 23-9609, __ F.4th __, 2025 WL 2447603 (10th Cir., 2025).

  2. 28 I&N Dec. 757 (BIA 2023).

  3. Practice Advisory of CLINIC and Duke Law School’s Immigrant Rights Clinic, December 2020: https://www.cliniclegal.org/resources/asylum-and-refugee-law/matter-e-r-l-vacated.

  4. M-R-M-S- at 761 (contrasting Hernandez-Avalos v. Lynch, 784 F.3d 944 (4th Cir. 2015) (finding nexus where th applicant’s relationship to her son is why she, and not another person, was targeted), and Orellana-Recinos v. Garland, 993 F.3d 851, 858 (10th Cir. 2021)).

  5. See Faiza W. Said, “The Immigration Shadow Docket,” 117 Nw. U. L. Rev. 893 (2023), available at https://scholarlycommons.law.northwestern.edu/nulr/vol117/iss4/1.

  6. Deborah E. Anker and Jeffrey S. Chase, Law of Asylum in the United States (2025 Ed.) (Thomson Reuters) at § 5:13 (quoting H.R. 10, 108th Cong. § 3007 (2004) (emphasis added).

  7. Matter of J-B-N- & S-M-, 25 I&N Dec. 208, 212 (BIA 2007).

  8. Id. at 214.

  9. Ndayshimiye v. Att’y Gen. of U.S., 557 F.3d 124, 129-30 (3rd Cir., 2009).

  10. Matter of L-E-A-, 27 I&N Dec. 40, 44 (BIA 2017).

  11. M-R-M-S- at 759.

  12. Matter of Kasinga, 21 I&N Dec. 357, 365 (BIA 1996).

  13. https://dictionary.cambridge.org/us/dictionary/english/animus#google_vignette,

  14. O.C.V. at *13.

  15. Id. at *7.

  16. L-E-A- I at 44.

  17. O.C.V. at *8. See Amicus Brief of Roundtable of Former Immigration Judges at 18-19 (“In L-E-A- I, the BIA cited the Romanov family of Czar Nicholas II and Czarina Alexandra as the epitome of true family persecution, where “a persecutor is seeking to harm the family members because of an animus against the family itself.” 27 I. & N. Dec. 40 at 44. Yet applying the reasoning in Matter of M-R-M-S- to those facts, the Romanov family could be denied asylum based on family status: the Romanovs were arguably not killed simply because someone had a personal grudge against them, but as a symbolic act of anti-monarchic violence…Because the dominant underlying reason for the assassination of the Romanovs was symbolic and political, not to specifically hurt the Romanov family, it is possible to reason using the logic of Matter of M-R-M-S- that the Romanovs were not targeted for their family membership.”).

  18. 142 F.4th 162, 171 (3d Cir. 2025).

  19. Id. at 173.

  20. Id. at 174.

  21. M-R-M-S- at 759, n. 6.

  22. Mejia v. Bondi, 144 F.4th 965, 975 (7th Cir. 2025).

  23. Id. at 975.

  24. See, e.g., Alvarez-Lagos v. Barr, 927 F.3d 236, 250 (4th Cir. 2019) (holding that “the record evidence compels the conclusion that Alvarez Lagos’s membership in her proposed social group…was one reason why she, and not another person, was threatened”); Cruz v. Sessions, 853 F.3d 122, 129 (4th Cir. 2017).

  25. 922 F.3d 346 (7th Cir. 2019).

  26. Mejia at 975 (quoting Cruz v. Sessions, supra at 356).

  27. 122 F.4th 655 (6th Cir. 2024).

  28. O.C.V. at *8 (quoting Mazariegos-Rodas at 671).

  29. Mazariegos-Rodas at 671 (quoting Matter of  S-P-, 21 I. & N. Dec. 486, 492 (B.I.A. 1996)).

  30. 136 F.4th 1, 13, n.8 (1st Cir. 2025).

JEFF CHASE