Brito Guevara v. Francisco Castro

139 F.4th 422

• Court: Court of Appeals for the Fifth Circuit
• Decided: June 2, 2025
• Docket: 24-10520
• Status: Published

Opinion

Case: 24-10520 Document: 82-1 Page: 1 Date Filed: 06/02/2025

United States Court of Appeals for the Fifth Circuit ____________ United States Court of Appeals Fifth Circuit

No. 24-10520 FILED June 2, 2025 ____________ Lyle W. Cayce Jose Leonardo Brito Guevara, Clerk

Plaintiff—Appellant,

versus

Samantha Estefania Francisco Castro,

Defendant—Appellee. ______________________________

Appeal from the United States District Court for the Northern District of Texas USDC No. 3:23-CV-1726 ______________________________

Before Richman, Willett, and Douglas, Circuit Judges. Don R. Willett: When A.F. was five years old, her mother, Samantha Estefania Francisco Castro, removed her from the lawful custody of her father, Jose Leonardo Brito Guevara, in Venezuela and then illegally immigrated with A.F. to the United States. 1 Brito petitioned for A.F.’s return under the Hague Convention on Civil Aspects of International Child Abduction.

_____________________ 1 A.F. was five years old at the time the district court decided this case. She is now seven. Case: 24-10520 Document: 82-1 Page: 2 Date Filed: 06/02/2025

No. 24-10520

The district court denied the petition, finding that although Brito had established a prima facie case of wrongful removal, A.F. was now well-settled in Texas. We REVERSE and REMAND with instructions that the district court enter an order that A.F. be returned to Venezuela. I A A.F. was born May 3, 2018 to Jose Leonardo Brito Guevara and Samantha Estefania Francisco Castro. ] Although never married, Castro and Brito lived together with A.F. in the home of Brito’s mother in Venezuela until they separated in July 2019. Following their separation, Brito was granted custody rights over A.F. During this period, A.F. maintained regular contact with both parents, though the record does not clearly indicate her primary residence. In August 2021, Brito relocated to Spain for a better-paying job. While in Spain, Brito continued to support A.F. financially, maintained regular contact through video calls and voice messages, and stayed in close contact with A.F.’s grandmother, who ensured that A.F. was provided for during Brito’s absence. The district court found that Brito was exercising his custody rights throughout his time in Spain. Until late 2021, A.F. had lived exclusively in Venezuela, and nothing in the record suggests she was not living a stable and safe life. 2 But in November 2021, Castro removed A.F. from Venezuela without Brito’s consent and unlawfully entered the United States. After presenting herself

_____________________ 2 The district court found that there was “next to zero evidence to prove the presence of ‘grave risk of harm’” if A.F. were to return to Venezuela.

2 Case: 24-10520 Document: 82-1 Page: 3 Date Filed: 06/02/2025

and A.F. to U.S. Border Patrol in San Luis, Arizona, Castro relocated to Lewisville, Texas. There, she lived with Castro’s boyfriend, Otton Rodriguez, for eleven months. In October 2022, Castro, A.F., and Rodriguez moved together to Dallas. Brito remained in contact with A.F. during this time and attempted to visit her in the United States, though his visa application was denied. The district court found that Castro “has been gainfully employed since arriving in the United States and provides for A.F.” Since arriving in the United States, Castro has worked for four different companies, averaging 40–45 hours a week, with hourly wages ranging from $12 to $16. Castro and A.F. lack permanent residence status in the United States. U.S. Citizenship and Immigration Services issued them employment authorization documentation, but their asylum applications remain pending. B Immediately upon learning that Castro had taken A.F. to the United States, Brito contacted the family attorney, Venezuelan authorities, and both the United States and Venezuelan embassies in Spain. He authorized his mother to file an application under the Hague Convention seeking A.F.’s return. Venezuelan authorities received the application on January 20, 2022—just under two months after Castro removed A.F. to the United States. The application languished until November 7, 2022, when the U.S. Department of State sent a letter to Castro, advising that the request had been forwarded from Venezuela and urging her to resolve the matter amicably or voluntarily return A.F. to Venezuela. Castro did not respond. After efforts to reach an agreement with Castro failed, Brito filed a petition in the Eastern District of Texas in April 2023. The district court

3 Case: 24-10520 Document: 82-1 Page: 4 Date Filed: 06/02/2025

issued a temporary restraining order, followed by a preliminary injunction barring Castro from leaving the jurisdiction and requiring her to disclose her address and contact information to both the court and Brito. Despite receiving actual notice, Castro failed to appear at the preliminary injunction hearing. A month later, in June 2023, Castro—through counsel—finally accepted service and disclosed her address, which was in fact located within the Northern District of Texas. By agreement of the parties, the action was transferred to the Northern District on August 1, 2023. Although Brito repeatedly requested expedited consideration, the Northern District did not hold a bench trial until March 2024—eight months after the transfer. Six weeks later, the court issued its findings of facts and conclusions of law. The court denied Brito’s petition, concluding that although he had established a prima facie case for A.F.’s return, Castro had sufficiently shown that A.F. was well-settled in Dallas and that remaining there was in her best interest. Brito timely appealed. II The Hague Convention mandates the return of “a child wrongfully removed from her country of habitual residence . . . upon petition.” 3 The Convention’s two chief objectives “are to restore the pre-abduction status quo and to deter parents from crossing borders in search of a more sympathetic court.” 4 The Convention rests on a core principle: “the best

_____________________ 3 England v. England, 234 F.3d 268, 270 (5th Cir. 2000). 4 Id. at 271 (quotations omitted).

4 Case: 24-10520 Document: 82-1 Page: 5 Date Filed: 06/02/2025

interests of the child are well served when decisions regarding custody rights are made in the country of habitual residence.” 5 Accordingly, the Convention’s default rule is that the child should be returned to her country of habitual residence. But the Convention “does not pursue that goal at any cost.” 6 It recognizes that, in certain cases, “the interests of the child may be better served by the child remaining” in her new environment, and it “provides ‘several narrow affirmative defenses to wrongful removal.’” 7 In the United States, the Hague Convention is implemented through the International Child Abduction Remedies Act (ICARA). 8 “Under ICARA, once a petitioner has established by a preponderance of the evidence that the child was wrongfully removed or retained, the burden shifts to the respondent to establish an affirmative defense.” 9 This case concerns one such defense: the “well-settled” exception found in Article 12. Article 12 provides that, “when a court receives a petition for return within one year after the child’s wrongful removal, the court ‘shall order the return of the child forthwith.” 10 But “where the proceedings have been commenced after the expiration of the period of one year,” the court “shall also order the return of the child, unless it is demonstrated that the _____________________ 5 Abbott v. Abbott, 560 U.S. 1, 20 (2010). 6 Hernandez v. Pena, 820 F.3d 782, 786 (5th Cir. 2016) (quoting Lozano v. Montoya Alvarez, 572 U.S. 1, 16 (2014)). 7 Id. (quoting Sealed Appellant v. Sealed Appellee, 394 F.3d 338, 343 (5th Cir. 2004)) (emphasis removed). 8 Galaviz v. Reyes, 95 F.4th 246, 251 (5th Cir. 2024); see also 22 U.S.C. § 9001(b)(1). 9 Galaviz, 95 F.4th at 521 (citing 22 U.S.C. § 9003(e)). 10 Lozano, 572 U.S. at 5 (quoting Hague Convention on the Civil Aspects of International Child Abduction, Oct. 25, 1980, T.I.A.S. No. 11670, art. 12).

5 Case: 24-10520 Document: 82-1 Page: 6 Date Filed: 06/02/2025

child is now settled in its new environment.” 11 “The underlying purpose of this defense is to recognize that at some point a child may become so settled in a new environment that return is no longer in the child’s best interests.” 12 To assess whether the well-settled defense applies, we consider seven factors: (1) the child’s age; (2) the stability and duration of the child’s residence in the new environment; (3) whether the child attends school or day care consistently; (4) whether the child has friends and relatives in the new area; (5) the child’s participation in community or extracurricular activities; (6) the respondent’s employment and financial stability; and (7) the immigration status of the respondent and child. 13 III “A district court’s determination of whether a child is well-settled presents a mixed question of law and fact.” 14 “We review the district court’s factual findings for clear error, and its legal conclusions de novo.” 15 Our precedent has long treated the balancing of factors under the well- settled defense as a legal question subject to de novo review. 16 The dissent

_____________________ 11 Id. (quoting Hague Convention, art. 12). 12 Hernandez, 820 F.3d at 787. 13 Id. at 787–88. 14 Id. at 787. 15 Id. 16 Id. at 790.

6 Case: 24-10520 Document: 82-1 Page: 7 Date Filed: 06/02/2025

contends that the Supreme Court’s recent decision Bufkin v. Collins 17 requires clear-error review. It does not. Bufkin addressed the standard of review the Veterans Court must apply in reviewing the Department of Veterans Affairs’ (VA) application of the statutory “benefit-of-the doubt rule.” 18 This “unique” rule, codified by Congress, instructs the VA to “give the benefit of the doubt to the claimant” whenever “there is an approximate balance of positive and negative evidence.” 19 The Supreme Court held that the Veterans Court must review the VA’s application of the rule “the same way it would any other determination—by reviewing legal issues de novo and factual issues for clear error.” 20 It further held that determining whether the evidence is approximately balanced is a “predominantly factual determination reviewed only for clear error.” 21 Like the VA’s approximate-balance test, our analysis of whether a child is well-settled presents a mixed question of fact and law. And the Supreme Court has made clear that the standard of review “depends ‘on whether answering it entails primarily legal or factual work.’” 22 Here is where the approximate-balance test and our well-settled test part ways. The VA’s determination of whether record evidence is approximately balanced is a textbook factual inquiry. It entails categorizing each piece of evidence based

_____________________ 17 604 U.S. __, 145 S.Ct. 728 (2025). 18 Id. at 733. 19 Id.; 38 U.S.C. § 5107(b). 20 Bufkin, 145 S.Ct. at 733. 21 Id. 22 Id. at 739 (quoting U.S. Bank N.A. v. Vill. at Lakeridge, LLC, 583 U.S. 387, 396 (2018)).

7 Case: 24-10520 Document: 82-1 Page: 8 Date Filed: 06/02/2025

on whether it supports or undermines the claim, comparing the relative strength and persuasiveness of the evidence on each side, then determining whether it is approximately balanced. 23 As the Supreme Court put it, this is “inherently a factual task.” 24 By contrast, our well-settled inquiry is primarily legal in character. We do not engage in a mathematical tallying of how the evidence aligns with each of the seven factors. Rather, the well-settled factors are a judicially crafted framework designed to inform a legal judgment: Is the child well-settled? None of the factors is dispositive. 25 We do not conduct a “head-to-head weighing” of the factors favoring one party versus the other. 26 Our review is holistic and guided—but not dictated—by the factors. 27 Our task is to assess whether, taken together, the evidence supports the district court’s legal conclusion. Congress might well have prescribed a different review standard had it envisioned evidence-balancing akin to the VA regime. But it did not. To be sure, the well-settled defense is a “creature of statute,” 28 deriving from the Hague Convention as implemented by Congress. 29 But the balancing _____________________ 23 Id. at 738–39. 24 Id. at 738. 25 Hernandez, 820 F.3d at 788. 26 See Bufkin, 145 S.Ct. at 749 (Jackson, J., dissenting). 27 See Hernandez, 820 F.3d at 787–88 (stating that “the following factors should be considered” and noting that immigration is “one relevant factor in a multifactor test” (emphasis added)) (citing Lozano v. Alvarez, 697 F.3d 41, 57 (2d Cir. 2012); In re B. Del C.S.B., 559 F.3d 999, 1009 (9th Cir. 2009)); see also Lozano, 697 F.3d at 57 (listing factors that courts should “generally” consider); In re B. Del C.S.B., 559 F.3d at 1009 (listing “a number” of non-exclusive factors courts should consider). 28 Bufkin, 145 S.Ct. at 740. 29 See post, at n.2.

8 Case: 24-10520 Document: 82-1 Page: 9 Date Filed: 06/02/2025

framework we use to assess that defense is not the product of statute. The factors are judicial constructs, not legislative commands. So we are not bound to the sort of calibrated factfinding Congress required under the VA’s benefit-of-the-doubt rule. Indeed, if Congress mandated anything here, it is this: courts “shall . . . order the return of the child” unless the respondent proves the well-settled defense applies. The factors we consult in applying that standard are just that—factors—not formulas that impose a duty of evidentiary calibration. They remain useful aids—but they are tools of our own making, crafted not to precisely quantify the weight of each piece of evidence, but to “generate guidance for . . . future courts” wrestling with the well-settled defense. 30 Our role, then, is not to duplicate the district court’s work in “compar[ing] the relative strength and persuasiveness of” the evidence. 31 Rather, we evaluate whether the district court properly applied the law—the requirements of the Hague Convention—to the facts before it. Unlike the evidentiary balancing required in Bufkin, this application of the factors is a legal inquiry, not a factual one. Bufkin thus leaves untouched our settled standard of review. We examine the district court’s factual findings for clear error, and we review de

_____________________ 30 Bufkin, 145 S.Ct. at 741. 31 See post, at 24 (quoting Bufkin, 145 S.Ct. at 738).

9 Case: 24-10520 Document: 82-1 Page: 10 Date Filed: 06/02/2025

novo the legal question whether, in light of holistic balance of the seven nondispositive factors, the evidence supports the district court’s conclusion. IV The parties do not dispute the district court’s finding that Brito established a prima facie case for A.F.’s return. The sole question on appeal is whether the well-settled defense bars that return. We conclude that the district court erred in both its legal framing and application of the well-settled exception. Balancing the relevant factors de novo, we are not persuaded that A.F. has formed such deep or enduring ties to her new environment that returning to her home in Venezuela would contravene her best interests. The first factor is A.F.’s age. She is seven years old—and was five at the time of the bench trial. The district court acknowledged, citing our precedent in Hernandez, that a child of this age is “a very young child not able to form the same level of attachments and connections to a new environment as an older child.” 32 Yet the district court described this factor as “lukewarm”—a characterization unsupported by the record. A.F.’s young age means it will take more time for her to become “so settled” in the United States that her best interests lie in remaining here rather than returning home to Venezuela. 33 At age seven, A.F. is not yet capable of forming the kind of enduring attachments that the Convention deems sufficient to override its default return remedy.

_____________________ 32 Hernandez, 820 F.3d at 789. 33 See id.; see also Hernandez v. Erazo, No. 23-50281, 2023 WL 3175471, *4 (5th Cir. May 1, 2023) (holding that young age can “discount[] the detrimental effect of being relocated” even where residential stability and daycare attendance cut against return).

10 Case: 24-10520 Document: 82-1 Page: 11 Date Filed: 06/02/2025

The second factor considers the stability and duration of A.F.’s residence in the United States. The district court found that over the past three years, Castro and A.F. have lived in two separate residences in the United States. It characterized this arrangement as stable and weighed the factor in favor of Castro. That conclusion was error. That A.F. has already moved multiple times in her brief time here undermines any claim of residential stability. 34 So too does the fact that Castro and A.F. currently reside in the home of Castro’s boyfriend. Should that relationship falter, Castro and A.F. would be forced to relocate once more. Castro conceded that if the relationship were to end, she and A.F. would need to downgrade to a cheaper apartment, as they rely—at least in part—on her boyfriend’s income. Even if A.F.’s present living situation appears stable, its long-term viability is far from assured. 35 The third factor examines whether the child attends school consistently. The district court rightly found that A.F. is enrolled in kindergarten and performing well. But that fact must be viewed in context and alongside the other factors. 36 At her young age, A.F. has ample time and

_____________________ 34 Cf. Belay v. Getachew, 272 F. Supp. 2d 553 (D. Md. 2003) (holding that stability factor weighed in favor of well-settled defense where child had lived at only one address since moving to the United States). 35 Cf. Ramirez v. Buyauskas, 2012 WL 606746 (E.D. Pa. 2012), opinion amended on other grounds, 2012 WL 699458 (E.D. Pa. 2012) (analyzing not just whether residence was currently stable but also whether it would remain stable). 36 The dissent claims that we impermissibly “bleed[] several factors together to circumvent the analysis of one.” Post, at 29. But it cites no support for the notion that each factor must be hermetically sealed and analyzed in isolation before the totality is considered. Our precedent says the opposite: the well-settled factors “should not be considered in the abstract.” Hernandez, 820 F.3d at 788. In any event, the dissent endorses the district court’s treatment of the fourth factor even though that analysis “overlapped” with its assessment of the third. Post, at 30. And the dissent itself folds delay into the factor- based framework, “incorporat[ing]” it rather than treating it as distinct. Post, at 41.

11 Case: 24-10520 Document: 82-1 Page: 12 Date Filed: 06/02/2025

opportunity to integrate into a new school community in Venezuela. 37 Moreover, A.F.’s school environment in United States is not especially secure, given the uncertainty of her immigration status, the nature and impermanence of Castro’s transient employment, and their reliance on Castro’s boyfriend for housing. These circumstances suggest a real possibility of future moves, which could disrupt A.F.’s schooling and undercut any sense of educational continuity. The fourth factor considers whether the child has formed meaningful relationships with friends and family in her new environment. A.F. does have at least six close relatives in the United States, as well as several friends she sees regularly. Still, most of A.F.’s extended family—including Castro’s parents, two brothers, a cousin, an aunt and uncle, and Brito’s mother, siblings, and additional relatives—remain in Venezuela. Most notably, A.F. cannot see her father in the United States. As discussed at oral argument, Brito attempted to visit her but was denied a visa. While the inquiry is not a numbers game, the fact that A.F. has a “large extended family” in Venezuela remains significant—particularly because her relationships in the United States are entirely derivative of her mother’s. 38 In addition, Castro’s boyfriend lacks lawful permanent resident status, and none of A.F.’s relatives in the United States are U.S. citizens. The unsettled immigration status of A.F.’s family here casts doubt on durability of those relationships and weighs against a finding that they are well-settled. The fifth factor examines A.F’s participation in community activities. The district court found that A.F. regularly attends church, visits a primary _____________________ 37 See Erazo, 2023 WL 3175471 at *4 (“Although [the child] has been in a stable home for over a year and attends daycare six days a week, his young age discounts the determinantal effect of being relocated.”). 38 See Hernandez, 820 F.3d at 789.

12 Case: 24-10520 Document: 82-1 Page: 13 Date Filed: 06/02/2025

care physician, goes on family vacations, has playdates with friends, uses community playgrounds, goes swimming, and attends birthday parties. The district court deemed this evidence “overwhelming” support for the well- settled defense. We disagree. Though it certainly weighs in Castro’s favor, this factor on its own does not demonstrate that A.F. is “so settled” in the United States that returning to Venezuela would be contrary to her best interests—especially since she could engage in many of these same activities there. 39 The sixth factor considers Castro’s economic and employment stability. The district court found that Castro “has been gainfully employed since arriving in the United States and provides for A.F.” But while the court acknowledged that Castro has changed jobs four times since her arrival, it failed to give appropriate weight to other facts that cast doubt on the stability of her employment. For instance, the court found that Castro was unemployed for at least two months between jobs. Nor does the record show that any of her jobs were permanent positions offering reliable income or benefits. The court further acknowledged that Castro shares both a car and an apartment with her boyfriend but overlooked the precariousness of that arrangement—namely, that if the relationship ended, Castro and A.F. would have to relocate. The end of the relationship would also leave them without transportation, impairing A.F.’s ability to attend school and participate in community life. While Castro is currently meeting A.F.’s basic needs, her

_____________________ 39 See id. at 789–90 (holding child was not well-settled despite evidence that he attended church regularly with his mother); Vite-Cruz v. Sanchez, 360 F.Supp.3d 346, 358 (D.S.C. 2018) (holding child was not well-settled despite evidence that he regularly spends time with his friends and is “very active in his school’s jazz ensemble”).

13 Case: 24-10520 Document: 82-1 Page: 14 Date Filed: 06/02/2025

financial circumstances are not “so settled” that it would be against A.F.’s best interest to return to her life in Venezuela. 40 The seventh and final factor concerns immigration status. The district court acknowledged that neither Castro nor A.F. has lawful permanent residence status in the United States and that both have pending asylum applications. But the court deemed this factor merely “lukewarm.” That conclusion was error. Castro presented no evidence suggesting their asylum claims are likely to succeed. Indeed, the court found no evidence that A.F. would face a “grave risk of harm” if returned to Venezuela—a finding that undercuts any suggestion that her asylum claim will succeed. 41 We acknowledge that “immigration status is not dispositive” and that lacking lawful permanent resident status “does not necessarily prevent a child from developing significant connections in a new environment.” 42 Still, “immigration status should not be analyzed in the abstract,” and the Convention requires “an individualized, fact-specific inquiry.” 43 The district court erred by evaluating immigration status in isolation, rather than

_____________________ 40 See Vite-Cruz, 360 F.Supp.3d at 358 (considering financial dependence on mother’s boyfriend evidence opposing a well-settled determination). 41 Asylum is available only “where 1) a person is ‘unwilling to return to’ their home country ‘because of persecution or a well-founded fear of persecution’; and 2) the applicant has demonstrated that ‘race, religion, nationality, membership in a particular social group, or political opinion was or will be at least one central reason for persecuting the applicant.’” Tamara-Gomez v. Gonzales, 447 F.3d 343, 348 (5th Cir. 2006) (quoting 8 U.S.C. § 1101(a)(42), 1158(b)). The fact that there is no “grave risk of harm” if A.F. returned to Venezuela strongly suggests that she does not face persecution there. Of course, the immigration court will reach its own findings in adjudicating A.F.’s asylum claim. But based on the record before us, she appears unlikely to satisfy the statutory requirements for asylum. 42 Hernandez, 820 F.3d at 788. 43 Id. at 788–89.

14 Case: 24-10520 Document: 82-1 Page: 15 Date Filed: 06/02/2025

assessing how it interacts with and undermines the other well-settled factors. 44 Here, the uncertainty surrounding Castro’s and A.F.’s immigration status permeates every aspect of their life in the United States, rendering it fundamentally unstable. This factor weighs heavily against finding that A.F. is well-settled. Overall, balancing the factors de novo, we disagree with the district court’s assessment that factors one and seven are merely “lukewarm,” and that the remaining factors “overwhelmingly” support a “well-settled” finding. The court failed to give due weight to A.F.’s young age—which favors her ability to readjust to life in Venezuela—and to her uncertain immigration status, which erodes any stability she may have developed in the United States. The district court also gave more weight to the remaining factors than is supported by the record. Certainly, as both the dissent and the district court observe, the record reflects that A.F. has enjoyed a stable and loving life with her mother in the United States. But that is not the legal question before us. Our task is to determine whether A.F. is “so settled in a new environment that return is no longer in [her] best interests.” 45 On balance, the answer is no. The factors do not support the conclusion that A.F. is so deeply rooted in the United States that returning her to Venezuela would contravene her best interests. This decision is not easy, nor is it without sorrow. But it accords with the Convention’s core objective: “to restore the pre-abduction status quo and to

_____________________ 44 Once again, the dissent contends that we improperly “bleed” our analysis of immigration status into the other factors. Post, at 35. But nothing in our precedent requires that each factor be assessed in hermetic isolation. See supra, at n.36. A holistic inquiry necessarily contemplates how various aspects of a child’s life—legal status included— interact to shape her connection to a new environment. 45 Hernandez, 820 F.3d at 787 (emphasis added).

15 Case: 24-10520 Document: 82-1 Page: 16 Date Filed: 06/02/2025

deter parents from crossing borders in search of a more sympathetic court.” 46 Because Brito established a prima facie case for return—and because the well-settled exception does not apply—the district court erred in denying his petition. 47 V Concluding that A.F. is not well-settled in her new environment, we VACATE the district court’s order and RENDER judgment in favor of Brito. We REVERSE and REMAND with instructions that the district court enter an order directing A.F.’s return to Venezuela.

_____________________ 46 England, 234 F.3d at 271 (quotations omitted). 47 Brito also argues that the district court erred in failing to consider its own delay in trying the case and Castro’s previous delays in responding to his petition. Because we render judgment in Brito’s favor on the grounds that the well-settled defense does not apply, we need not reach the delay issue.

16 Case: 24-10520 Document: 82-1 Page: 17 Date Filed: 06/02/2025

Dana M. Douglas, Circuit Judge, dissenting: The majority opinion reverses the district court’s order and renders judgment in Brito’s favor. In doing so, it reweighs evidence, bleeds various factors together in violation of established law, and assumes imminent failure of an undecided asylum claim. Today’s decision punishes A.F.—who is well- settled in her new home—for her mother’s decisions. I respectfully dissent. I A Brito and Castro are Venezuelan citizens and former romantic partners. On May 3, 2018, their daughter, A.F., was born. Brito and Castro were never married, but they lived together in Brito’s mother’s home in Venezuela when A.F. was born. Approximately two years after the couple split up, in August 2021, Brito moved from Venezuela to Madrid, Spain, for a new job; he has not returned to Venezuela since then. At some point after Brito moved, Castro mentioned to Brito and/or his mother that she was considering traveling to the United States, but never informed them of a desire to bring A.F. with her; accordingly, Brito never consented to A.F.’s removal from Venezuela. Nevertheless, about three months after Brito moved, Castro left Venezuela with A.F., entering the United States without documentation. They immediately presented themselves to the United States Border Patrol in San Luis, Arizona, and applied for asylum. Both of their asylum applications remain open; although they do not have Lawful Permanent Residence status in the United States, both are awaiting asylum interviews with the U.S. Citizenship and

17 Case: 24-10520 Document: 82-1 Page: 18 Date Filed: 06/02/2025

Immigration Services (“USCIS”). 1 Approximately six months before moving, Castro began a romantic relationship with Otton Rodriguez, who has Temporary Protected Status under the Immigration and Nationality Act (“INA”). Rodriguez has resided with Castro and A.F. in Texas since they arrived. They married during the pendency of this appeal. Since moving to Texas, Castro has held positions at four companies and has earned enough money to open a bank account in the United States. While Castro worked, she either hired a caretaker for A.F. or left A.F. in the care of family. Eventually, A.F. began attending kindergarten full-time in Addison, Texas, and began seeing a primary care physician. B On January 20, 2022, Brito’s mother filed a petition under the Hague Convention on the Civil Aspects of International Child Abduction (“Hague Convention” or “Convention”) with Venezuelan authorities, seeking A.F.’s return. That application, however, was not transferred to the U.S. Department of State until November 7, 2022, at which point the State Department attempted to contact Castro. Because Castro did not consent to return to Venezuela, Brito petitioned the United States District Court for the Eastern District of Texas on April 19, 2023, for A.F.’s return under the Hague Convention, Oct. 25, 1980, T.I.A.S. No. 670, and the International Child Abduction Remedies Act (“ICARA”), 22 U.S.C. §§ 9001–9011. The matter was ultimately transferred to the Northern District of Texas. Following the transfer, Brito repeatedly requested status conferences to schedule a trial. The court scheduled an off-the-record status conference

_____________________ 1 Castro’s brief notes that she and A.F. were provided Temporary Protected Status under the INA during the pendency of this appeal. She again represented this fact at oral argument.

18 Case: 24-10520 Document: 82-1 Page: 19 Date Filed: 06/02/2025

for November 7, 2023. Brito asserts that the judge did not address his concerns at that conference, instead requesting a recitation of facts before ending the conference due to a scheduling conflict. On November 8, the court reset the conference. On November 10, the parties attended a Zoom conference at which the court allegedly indicated that trial would be set no sooner than March 2024. In a later status report, Brito requested an expedited trial setting and reserved the right to request a formal statement of delay or judicial transfer. In other filings, Brito compared the court’s delay with the time other Hague Convention cases in the district took to reach trial. The court denied all requests without a hearing and scheduled the final trial. The court held a two-day bench trial beginning on March 21, 2024. On May 8, 2024, the district court issued findings of fact and conclusions of law, finding that A.F. has stable housing in the United States, Castro is financially secure and amply provides for A.F., and Rodriguez cares deeply for and serves as a father figure to A.F. The district court also found that A.F. has formed significant connections to her environment in Texas— stronger than those to Venezuela. Therefore, it concluded that, while Venezuela is A.F.’s country of habitual residence, Castro successfully demonstrated that A.F. is well-settled in Texas. It issued a final judgment denying Brito’s complaint and petition for A.F.’s return. Brito appealed. II The Hague Convention addresses “the problem of international child abductions during domestic disputes.” Lozano v. Montoya Alvarez, 572 U.S. 1, 4 (2014) (quoting Abbott v. Abbott, 560 U.S. 1, 8 (2010)). “The Convention states two primary objectives: ‘to secure the prompt return of children wrongfully removed to or retained in any Contracting State,’ and ‘to ensure that rights of custody and of access under the law of one Contracting State are effectively respected in the other Contracting States.’” Id. at 4–5

19 Case: 24-10520 Document: 82-1 Page: 20 Date Filed: 06/02/2025

(quoting Hague Convention, art. 1). So, the focus “is the return of the child,” which lays the venue for the ultimate custody determination. Id. at 5. For a petitioner to make a prima facie showing that the child should be returned to the country of habitual residence, they must demonstrate that (1) the child was removed from the country of habitual residence; (2) the removal violated petitioner’s rights of custody under the laws of the country of habitual residence; and (3) petitioner was exercising those rights at the time of removal. Larbie v. Larbie, 690 F.3d 295, 307 (5th Cir. 2012). If a petitioner demonstrates these three elements, the child shall be returned to the country of habitual residence. See id. at 306–07. Nevertheless, the Convention’s remedy of return “is not absolute.” Lozano, 572 U.S. at 5. The Convention provides several affirmative defenses to the respondent, typically proven by a preponderance of the evidence, to refute a petitioner’s prima facie showing that the child was wrongfully removed from their country of habitual residence. Many of these defenses are housed in Article 13 of the Convention, but Article 12 holds the one we consider today: “[W]hen a court receives a petition for return within one year after the child’s wrongful removal, the court ‘shall order the return of the child forthwith.’” Id. (quoting Hague Convention, art. 12). But “where the proceedings have been commenced after the expiration of the period of one year,” the court “shall also order the return of the child, unless it is demonstrated that the child is now settled in its new environment.” Id. (quoting Hague Convention, art. 12). Courts consider the following factors for the well-settled defense:

(1) the child’s age; (2) the stability and duration of the child’s residence in the new environment; (3) whether the child attends school or day care consistently; (4) whether the child has friends and relatives in the new area; (5) the child’s participation in community or extracurricular activities; (6) the

20 Case: 24-10520 Document: 82-1 Page: 21 Date Filed: 06/02/2025

respondent’s employment and financial stability; and (7) the immigration status of the respondent and child. Hernandez v. Garcia Pena, 820 F.3d 782, 787–88 (5th Cir. 2016). While “[c]ourts diverge . . . with regard to the significance of immigration status,” we have concluded “that immigration status is neither dispositive nor subject to categorical rules, but instead is one relevant factor in a multifactor test.” Id. at 788. “The underlying purpose of this defense is to recognize that at some point a child may become so settled in a new environment that return is no longer in the child’s best interests.” Id. at 787. Ultimately, even if an affirmative defense applies, “a federal court has ‘and should use when appropriate’ the discretion to return a child to his or her place of habitual residence ‘if return would further the aims of the Convention.’” England v. England, 234 F.3d 268, 271 (5th Cir. 2000) (quoting Friedrich v. Friedrich, 78 F.3d 1060, 1067 (6th Cir. 1996)). III Before considering the merits, we must adopt the proper standard of review. The majority opinion takes the traditional path of reviewing factual findings for clear error and legal determinations de novo. However, the Supreme Court recently altered this determination in Bufkin v. Collins, 145 S. Ct. 728 (2025), which demands a clear-error review of both the fact-finding and the weighing of evidence in certain instances. Bufkin should apply here. In Bufkin, the Court considered a Board of Veterans’ Appeals decision denying a veteran’s claim to disability benefits for PTSD. The Court accepted that legal conclusions are subject to de novo review. “For example, if the veteran argues that the [Department of Veterans’ Affairs (“VA”)] misunderstood the definition of ‘approximate balance,’ the Veterans Court would construe the challenge as a legal one and review it de novo.” Id. at 738. But typically, “a veteran challenges the VA’s determination that the

21 Case: 24-10520 Document: 82-1 Page: 22 Date Filed: 06/02/2025

evidence on a particular material issue is not in approximate balance.” Id. The Supreme Court concluded that this “is a predominantly factual question and thus subject to clear-error review.” Id. The method of weighing the factors in Bufkin was strikingly similar to the case at hand. “First, the VA reviews each item of evidence in the record and assigns weight to it.” Id. The parties agreed that this was reviewed for clear error. Then, “the VA assesses the weight of the evidence as a whole,” deciding whether there was an approximate balance on any material issue. Id. The Court noted that the second step had “both legal and factual components,” considering “marshaling and weighing evidence” factual. Id. “The appropriate standard of review for a mixed question depends ‘on whether answering it entails primarily legal or factual work.’” Id. at 739 (quoting U.S. Bank N.A. v. Vill. at Lakeridge, LLC, 583 U.S. 387, 396 (2018)). “When applying the law involves developing legal principles for use in future cases, appellate courts typically review the decision de novo.” Id. But, critically, “[w]hen the tribunal below is ‘immerse[d]’ in facts and compelled to ‘marshal and weigh evidence’ and ‘make credibility judgments,’ the appellate court ‘should usually review a decision with deference.’” Id. (second alteration in original) (quoting U.S. Bank, 583 U.S. at 396). “Reviewing a determination whether record evidence is approximately balanced is ‘about as factual sounding’ as any question gets.” Id. (quoting U.S. Bank, 583 U.S. at 397). Because the Board had to weigh evidence, the work was “fact intensive” and its determinations received deference. Id. 2

_____________________ 2 The Supreme Court also distinguished between those evidence-weighing determinations that are constitutional, and those that are statutory. Constitutional standards are entitled to a presumption of de novo review that statutory standards do not

22 Case: 24-10520 Document: 82-1 Page: 23 Date Filed: 06/02/2025

The inquiry before us today is one that required the district court to be “‘immerse[d]’ in facts and compelled to ‘marshal and weigh evidence’ and ‘make credibility judgments.’” Id. (alteration in original) (quoting U.S. Bank, 583 U.S. at 396). The majority focuses on the fact that our analysis does not include a “head-to-head weighing” of the factors, and that “[o]ur review is holistic and guided—but not dictated—by the factors.” Ante, at 8. But our court borrowed this test from the Second Circuit, which “formally adopt[ed]” this “fact-specific multi-factor test.” Lozano v. Alvarez, 697 F.3d 41, 57 (2d Cir. 2012), aff’d, Lozano, 572 U.S. 1; accord Lozano, 572 U.S. at 17 (referring to whether a child is settled as “a factual determination”). And while the language in Hernandez may not be mandatory, stating that we “should” consider the seven factors we look to today, Hernandez, 820 F.3d at 787, the court used softer language because the determination is inherently factually driven and context dependent. See Duarte v. Bardales, 526 F.3d 563, 576 (9th Cir. 2008) (Bea, J., dissenting) (noting that “a court may consider any factor relevant to a child’s connection to his living environment” before listing the seven factors that courts “generally” consider). But courts widely consider these same factors. See Hernandez, 820 F.3d at 787 (“We join the

_____________________ receive. See Bufkin, 145 S. Ct. at 740. The Hague Convention undoubtedly falls in the “statutory” realm. Not only was it implemented by congressional act, but basic Supremacy Clause jurisprudence puts treaties on par with statutes. See, e.g., Reid v. Covert, 354 U.S. 1, 18 (1957) (“This Court has also repeatedly taken the position that an Act of Congress, which must comply with the Constitution, is on a full parity with a treaty . . . .”); Whitney v. Robertson, 124 U.S. 190, 194 (1888) (“By the constitution, a treaty is placed on the same footing, and made of like obligation, with an act of legislation. Both are declared by that instrument to be the supreme law of the land, and no superior efficacy is given to either over the other.” (emphases added)).

23 Case: 24-10520 Document: 82-1 Page: 24 Date Filed: 06/02/2025

circuits that have addressed this issue and hold that the following factors should be considered . . . .”). 3 When considering the well-settled defense, the district court makes credibility judgments and considers evidence produced by each side, and, ultimately, “compares the relative strength and persuasiveness of” that evidence in determining whether a child’s best interests would be served by remaining in their current environment. Bufkin, 145 S. Ct. at 738. Considering a child’s best interests through record evidence “is ‘about as factual sounding’ as any question gets.” Id. at 739 (quoting U.S. Bank, 583 U.S. at 397). Today we do not consider any inherently legal issue, such as whether the district court properly interpreted a statute. See id. at 738 (“[I]f the veteran argues that the VA misunderstood the definition of ‘approximate balance,’ the Veterans Court would construe the challenge as a legal one and review it de novo. So too if the veteran argues that the VA gave the benefit of the doubt to the wrong party.”). But the “approximate balance” determination brought before the Supreme Court was “case specific and fact intensive.” Id. at 740. The same is true of the well-settled defense as it is before us today. Therefore, Supreme Court precedent dictates that clear-error review dominates the entire analysis—not merely the facts underlying each

_____________________ 3 While the majority rightly notes that these factors are judicially created, see ante, at 9, nothing in Bufkin demands that the factors be statutorily created to compel a clear- error review. Indeed, our court conducts clear-error review in other judicially created doctrines—even ones that create exceptions to the constitutional standards that typically receive de novo review. See United States v. Newman, 472 F.3d 233, 237 (5th Cir. 2006) (“The district court’s determination as to whether exigent circumstances existed is fact- specific, and we will not reverse it unless clearly erroneous.”). Similar to our inquiry here, the exigent-circumstances test considers a “non-exhaustive” list of five factors. See United States v. Aguirre, 664 F.3d 606, 611 (5th Cir. 2011). That does not render the determination legal.

24 Case: 24-10520 Document: 82-1 Page: 25 Date Filed: 06/02/2025

factor. Nevertheless, even if the majority opinion is correct in weighing the factors de novo, I respectfully disagree with its conclusion. IV The majority opinion identifies and considers the seven well-settled defense factors, but it blends them together without providing proper deference to the district court. I discuss each factor below. A. Age A.F. was five years old at the time of trial. I agree with the majority opinion that “A.F. is not yet capable of forming the kind of enduring attachments” that would weigh in favor of applying the well-settled defense. Ante, at 10. B. Stability and Duration of Residence The district court found that A.F. has lived with Castro and Rodriguez at two locations over the course of nearly three years. It concluded that she had stable housing during that time. The majority opinion reverses course on this, stating: “That A.F. has already moved multiple times in her brief time here undermines any claim of residential stability.” Ante, at 11. But there is no support for this statement. First, A.F. has moved once since arriving to the United States. And second, the only cited legal support—a non-binding opinion from the District of Maryland—is inapposite. The district court in that case weighed the stability factor in favor of a well-settled conclusion where the child had lived at only one location. See Belay v. Getachew, 272 F. Supp. 2d 553, 561 (D. Md. 2003). Under the majority’s reading of Belay, the only way that a child may have “stability” in their new environment is if they live in one place for some indeterminate period of

25 Case: 24-10520 Document: 82-1 Page: 26 Date Filed: 06/02/2025

time. 4 But Belay does not stand for this restrictively narrow proposition, and the majority fails to cite to a decision of any court supporting that conclusion. It is unclear what more Castro could have done to provide A.F. with a settled home life. Moving once in three years is far from unstable, and holding otherwise creates an unrealistic hurdle. Simultaneously, the majority opinion concludes that A.F. lacks housing stability because “Castro and A.F. currently reside in the home of Castro’s boyfriend,” and, “[s]hould that relationship falter, Castro and A.F. would be forced to relocate once more.” Ante, at 11. 5 But the Hague Convention does not work in theoretical possibilities. See, e.g., Farley v. Hill, 150 U.S. 572, 577 (1893) (“But a court cannot act upon such uncertain conjectures.”); DeFunis v. Odegaard, 416 U.S. 312, 320 n.5 (1974) (refusing

_____________________ 4 Similarly, the various cases that Brito cites in favor of his position that the home was unstable are distinguishable. Each considers a situation in which the child or children lived in several homes over a shorter period of time. See, e.g., Argueta v. Lemus, No. 21-cv- 209, 2022 WL 88039, at *3, 8 (N.D. Miss. Mar. 9, 2022) (finding that living in three locations in approximately eighteen months, including several months at the mother’s friend’s house, weighed against a well-settled finding); In re A.V.P.G., 251 S.W.3d 117, 126 (Tex. App. 2008) (“The children were living in Belgium with both parents when they were suddenly uprooted, resided for a time with Guajardo’s parents in Mexico, then relocated to a different home in Mexico with Guajardo; were placed in foster care in Texas; and then resided with their grandparents in Texas.”); Moretti v. Braga, No. 23-cv-0586, 2023 WL 3590690, at *19 (N.D. Tex. May 22, 2023) (noting that the child “had no less [sic] than four different residences, some as temporary as a tent in a campground,” over fourteen months). 5 In support of this speculative proposition, the majority cites loosely to a decision from the Eastern District of Pennsylvania, which considered whether the child’s residence would remain stable in the future. Ante, at 11 n.35. That case notes that future stability was promising “because the rent is government-subsidized and is far less than the amount respondent receives each month in governmental assistance.” Ramirez v. Buyauskas, No. 11-6411, 2012 U.S. Dist. LEXIS 24899, at *53–54 (E.D. Pa. Feb. 24, 2012). This is far from the speculative reasons provided in the majority opinion for finding A.F.’s future housing unstable.

26 Case: 24-10520 Document: 82-1 Page: 27 Date Filed: 06/02/2025

to consider “speculative contingencies” (quoting Hall v. Beals, 396 U.S. 45, 49 (1969))). Imagine the limitless possibilities that could weigh against stability if we allow speculation to creep into our analysis. Consider, for instance, a tenant that pays rent to a landlord, with no suggestion they have violated their lease. Is it not possible that rent increases, necessitating a move? Or that the landlord chooses not to relet to the tenant at the expiration of the initial lease term, for no discernable reason? Imagine they live in a hurricane-prone region. What if a natural disaster strikes, rendering the home uninhabitable, a scenario with which this circuit is tragically familiar? More pointedly, is it not possible that anyone could split up with their partner and need a new home? Does that make their housing unstable? What if they have lived together for twenty years? Fifty years? The majority opinion draws no lines as to what a “stable” environment is in the face of this hypothetical. Instead, it premises A.F.’s lack of stability on speculation, which does not further the Convention’s purpose. Indeed, any of these possibilities could just as likely arise in Venezuela, where A.F.’s father has never returned since his relocation to Spain. Nor do Brito’s arguments fare any better. He claims that Castro lived in four separate locations, asserting that her housing in temporary asylum facilities should weigh against stability. But he effectively stipulated to the fact that A.F. had only resided in two locations in the United States in the unopposed statement of facts. Moreover, the record supports the district court’s determinations: Castro testified that she “handed [herself] over” to border patrol agents upon her arrival, at which point they transferred Castro and A.F. “to one of the refugee camps.” They were only at that “family refugee center . . . for about one day” before they were sent to another refugee center for two days. They then were transferred to a hotel in Los Angeles by

27 Case: 24-10520 Document: 82-1 Page: 28 Date Filed: 06/02/2025

bus for one night, after which she flew to Texas. The well-settled defense— with good reason—establishes a one-year bar on its applicability. It would defy logic that one week’s worth of one- or two-day stops upon arrival in a new country should cut against a finding that a child has stable housing. I would defer to the district court’s well-reasoned and well-supported factfinding and weigh this factor in favor of a well-settled finding. C. Whether A.F. Attends School or Day Care The district court next found that A.F. received daily care from two individuals—a caretaker and a family member—from her arrival in November 2021 until August 2023, at which point she began kindergarten. She attends school each day with her cousin, and they both go to her aunt’s house after school. The district court also recognized her success in school, including her nomination for the Gifted and Talented Program (through which she receives a bilingual education) and her report cards, which demonstrate continued academic improvement. The majority opinion discounts these factual findings, stating that they “must be viewed in context and alongside the other factors.” Ante, at 11. In support, it cites only one unpublished opinion, Hernandez v. Erazo, which stated that, “[a]lthough [the child] ha[d] . . . been in a stable home for over a year and attend[ed] daycare six days a week, his young age discount[ed] the detrimental effect of being relocated.” No. 23-50281, 2023 WL 3175471, at *4 (5th Cir. May 1, 2023). Stepping beyond the fact that Erazo was in a vastly distinguishable procedural posture—a motion to stay the district court order pending appeal, id. at *1—the court made that statement while weighing all of the factors together, not while individually

28 Case: 24-10520 Document: 82-1 Page: 29 Date Filed: 06/02/2025

analyzing them. 6 Allowing the ultimate weighing of the factors to impact the individual analyses puts the cart before the horse. So, when the majority opinion states that A.F. “has ample time and opportunity to integrate into a new school community in Venezuela,” and that her situation here “is not especially secure, given the uncertainty of her immigration status, the nature and impermanence of Castro’s transient employment, and their reliance on Castro’s boyfriend for housing,” ante, at 12, it impermissibly bleeds several factors together to circumvent the analysis of one. Neither the Convention nor our case law endorses this approach, which would prevent a respondent from ever successfully invoking the defense. See Hernandez, 820 F.3d at 789–90 (considering each factor individually before weighing them together). 7

_____________________ 6 The statement clearly credits the child’s attendance at day care in favor of a well- settled finding before saying that it may be outweighed by a negative factor. 7 While the majority opinion states that there is no authority “for the notion that each factor must be hermetically sealed and analyzed in isolation before the totality is considered,” ante at 11 n.36, Hernandez itself took that very approach. For instance, Hernandez considered stability of residence separately from immigration status. 820 F.3d at 789–90 (“With regard to [stability and duration of the child’s new residence], although [the child’s] residence is stable, he has lived in New Orleans less than a year. . . . Finally, the seventh factor we consider is immigration status. [The child and his parent] are both illegally present in the United States and involved in active removal proceedings. This involvement in active removal proceedings and categorization as new immigration violators seriously threatens their ability to remain in the United States.”); id. at 790 (“[b]alancing the factors” against one another). And while the majority points to the determination that delay folds into the factor- based framework, ante at 11 n.36, not only is that conclusion compelled by Supreme Court precedent, as discussed below, but a parent’s concealment has no individual impact on whether a child is well-settled unless the parent’s actions directly impact the child. Those actions then fit squarely within our Hernandez framework. See post, at 39–43.

29 Case: 24-10520 Document: 82-1 Page: 30 Date Filed: 06/02/2025

For his part, Brito provides only a conclusory, threadbare challenge that the evidence was not overwhelming. But the record is replete with evidence to support the court’s findings on this factor. Thus, neither Brito nor the majority opinion shows that the district court committed clear error in reviewing this factor. I would accept its findings. D. A.F.’s Friends and Relatives The district court’s consideration of this factor largely overlapped with its consideration of the third given the crossover of A.F.’s school and family communities. But it also noted that she “has many friends outside of her family with whom she has been photographed.” Brito argues that while Castro testified that A.F. enjoys playdates and swimming, she failed to show the frequency or duration of those activities. He also complains that the court pointed to only three other schoolmates and one family member, which he considers insufficient. The record supports the district court’s findings. First, A.F. clearly has significant family ties to the area. She sees her cousin and aunt daily after school, took a family vacation to Disney World, and has several family members whom she sees “[a]lmost every day.” Moreover, Rodriguez teaches her how to swim; she goes to the movies, zoo, and aquarium; and she rides bikes with her family. She also “has very tight connections with her friends” and talks about her friends every day. Castro produced photos of her with her friends Manuela and Emma at trial. Contrary to Brito’s unsupported allegations, this evidence is significant. 8

_____________________ 8 Similarly, Brito’s citation to an unpublished decision from the Middle District of North Carolina for the proposition that making “only a few friends” can weigh against a

30 Case: 24-10520 Document: 82-1 Page: 31 Date Filed: 06/02/2025

The majority opinion considers none of this evidence. Instead, despite acknowledging that “the inquiry is not a numbers game,” it effectively counts the number of family members she has in both the United States and Venezuela. Ante, at 12. While it is certainly relevant that she has a large extended family in Venezuela, see Hernandez, 820 F.3d at 789, this fact alone does not conclusively demonstrate that the district court’s findings were clearly erroneous. The majority opinion goes on to note that Castro’s husband lacks Lawful Permanent Resident status, and that none of A.F.’s family members in the United States are citizens. First, Rodriguez has Temporary Protected Status under the INA. 9 Second, that A.F.’s family members are not citizens should not weigh against whether A.F. is well- settled. There is no evidence that any of these individuals face imminent removal, nor was any other information regarding their immigration statuses produced. 10 This factor should weigh in favor of a finding that A.F. is well-settled. E. Participation in Community or Extracurricular Activities The district court labeled the evidence of A.F.’s extracurricular and community participation “overwhelming.” Among other things, it found

_____________________ well-settled finding is readily distinguishable. There, the child was thirteen years old and had made significant connections in their home country before removal. See Chambers v. Russell, No. 20-cv-498, 2020 WL 5044036, at *1, 6 (M.D.N.C. Aug. 26, 2020). 9 Therefore, he “cannot be detained by DHS on the basis of his . . . immigration status in the United States,” and he is “not removable from the United States.” Temporary Protected Status, https://www.uscis.gov/humanitarian/temporary-protected-status (last visited April 29, 2025). 10 And, even if there was such evidence, there is no suggestion that we must consider their immigration statuses. Considering immigration statuses of the individuals with whom A.F. acquaints adds yet another formerly unrecognized consideration into the typical Hernandez factors.

31 Case: 24-10520 Document: 82-1 Page: 32 Date Filed: 06/02/2025

that she regularly attends church in Dallas with Castro and Rodriguez, sees a primary care physician, goes on trips with her family, has playdates with school friends, is learning English, plays at community playgrounds, swims, and attends birthday parties. The majority opinion does not show that the district court’s factual finding that this evidence is overwhelming is implausible in light of the record as a whole. Instead, after recounting the district court’s findings, it simply states: “[w]e disagree.” 11 Ante, at 13. Instead, the majority states that “this factor on its own” does not demonstrate that A.F. is well settled. Ante, at 13 (emphasis added). But the district court never made such a statement. And as it relates to the district court’s comment that the evidence was overwhelming in favor of Castro, neither the majority nor Brito offers any evidence that clearly undermines the district court’s finding. The trial testimony indicates that A.F. is involved in the community activities described above. Considering these facts, the district court correctly weighed this factor in favor of applying the well-settled defense. It is not the role of this court to conclude that the district court’s finding of fact was clearly erroneous based on a belief that, “had [we] been sitting as the trier of fact, [we] would have weighed the evidence differently.” Anderson v. City of Bessemer City, 470 U.S. 564, 573–74 (1985). F. Mother’s Employment and Economic Stability

_____________________ 11 True enough, the opinion cites two cases as demonstrating that a child may not be well-settled despite spending time with friends or attending church. But it relies on statements made during the ultimate weighing of the factors. These citations, if anything, cut against the majority opinion’s statement, as they demonstrate that the extracurriculars weighed in favor of a well-settled finding. That factor was merely outweighed by others under the specific facts of those cases.

32 Case: 24-10520 Document: 82-1 Page: 33 Date Filed: 06/02/2025

The district court found that Castro has held four jobs since arriving in the United States, “with each subsequent job having a higher salary than the one before.” 12 It noted that she has been gainfully employed since she arrived and that she sufficiently provides for A.F. Additionally, Rodriguez contributes financially and splits rent and utility payments with Castro. Castro makes monthly car payments for Rodriguez’s car, which they share, and she has healthcare covering both herself and A.F. Finally, Castro testified that if she and Rodriguez were to ever break up, she would move to a cheaper apartment so that she could provide for herself and A.F. The majority opinion focuses not on her employment history, but on the fact that she “was unemployed for at least two months between jobs” and that “the record [does not] show that any of her jobs were permanent positions offering reliable income or benefits.” Ante, at 13. As an initial matter, while Castro may have been unemployed for two months, she was employed for twenty-seven months, with her term of employment increasing at each company. Focusing on this two-month period is misleading. Nor does the majority opinion provide any law suggesting that “stability” in employment requires permanence or stable benefits. The majority then turns to the same concerns it espoused before: if Rodriguez and Castro break up, what of her economic stability? There is no support for considering hypothetical scenarios in determining that someone is not currently well-settled, absent some clear, imminent event. The majority cites only Vite-Cruz v. Sanchez, 360 F. Supp. 3d 346, 358 (D.S.C. 2018), for _____________________ 12 This finding was error. Castro’s wage changed as follows: (1) $12/hour, (2) $16/hour, (3) $14/hour, and (4) $16.20/hour. She worked an estimated (1) 40 hours per week, (2) 45 hours per week, (3) 45 hours per week, and (4) 40 hours per week. Therefore, she had an approximate weekly income of: (1) $480, (2) $720, (3) $630, and (4) $648. Regardless of the metric by which the district court measured Castro’s income, it did not consistently increase. Nevertheless, this error has no impact upon the analysis.

33 Case: 24-10520 Document: 82-1 Page: 34 Date Filed: 06/02/2025

the proposition that economic reliance on a partner may weigh against a well- settled finding, especially if the partner ceases to provide assistance. But that case is doubly distinguishable: there, the mother “testified she [did] not work,” so their economic stability was entirely dependent on the boyfriend, who was undocumented (with no indication that he had Temporary Protected Status). Id. As established above, Castro is employed. Moreover, Rodriguez has married Castro. Nothing in the majority opinion demonstrates how the district court clearly erred in this determination. Nor do Brito’s arguments carry the day. First, he claims that Castro’s expenses exceed her income. But a review of the record shows that such would be the case only if Rodriguez were to move out and all expenses remained the same. The district court found that Castro would reduce her expenses under those circumstances, so this argument fails. Second, he complains that she has held four positions since moving, so her employment cannot be stable. But she has held each position for a longer time than the one directly preceding it. And while her previous position at the Great Wolf Lodge netted the highest weekly income, she has found more temporal stability at her final two jobs, including having been employed at Paycom for approximately eleven months at the time of trial. Brito’s argument finds no support, and the non-binding case law he cites is distinguishable. 13

_____________________ 13 See, e.g., Vite-Cruz, 360 F. Supp. 3d at 358 (finding that if the mother’s romantic partner were to lose his job, be deported, or otherwise split with the mother, she would be unable to provide for the child because she “depend[ed] solely on [her] boyfriend . . . to provide financial support for food, housing, clothing[,] and other necessities” (emphasis added)); Moretti, 2023 WL 3590690, at *20 (noting that the respondent’s entrepreneurial activities were just “beginning to achieve a modicum of success,” but that she did not make a paycheck the prior month and had “not held any other employment”).

34 Case: 24-10520 Document: 82-1 Page: 35 Date Filed: 06/02/2025

G. Immigration Status As to the final factor, “it is undisputed that both [Castro] and A.F. do not have Lawful Permanent Residence status in the United States, but they both have actively pending asylum applications and are currently awaiting their asylum interview with USCIS.” The district court credited Castro and A.F. with immediately surrendering themselves to border patrol upon entry to the United States and endeavoring through “the proper procedures to achieve lawful status in the United States.” Finally, both Castro and A.F. have employment authorization documentation from the USCIS. The majority opinion properly notes that immigration status is not dispositive and that a child may still develop contacts in a new environment. It also correctly states that immigration status should not be considered in the abstract, but requires “an individualized, fact-specific inquiry.” Hernandez, 820 F.3d at 789. But it misreads this to mean that immigration status should not be considered “in isolation”; instead, it concludes that the court should consider how immigration status “interacts with and undermines the other well-settled factors.” Ante, at 14–15. This contradicts Hernandez twice over. First, it overlooks Hernandez’s statement that immigration is merely “one relevant factor in a multifactor test.” Hernandez, 820 F.3d at 788. This alone suggests that immigration does not bleed into other factors. The majority opinion, however, contrarily concludes that immigration status alone can “permeate[] every aspect of their life in the United States, rendering it fundamentally unstable,” thus “weigh[ing] heavily against a finding that A.F. is well-settled.” Ante, at 15. Second, it takes the statement that there must be an “individualized, fact-specific inquiry” out of context. Hernandez held that the district court should have “adequately examine[d] [the individual’s] actual immigration status.” Hernandez, 820 F.3d at 789. In other words, it should “take into

35 Case: 24-10520 Document: 82-1 Page: 36 Date Filed: 06/02/2025

account relevant, case-specific distinctions that may exist among and between different immigration statuses.” Id. At risk of repetition, Hernandez requires “a proper analysis of [the individual’s] specific immigration status.” Id. Nowhere does Hernandez suggest that we ought to consider immigration within each other factor. Rather, it merely held that broad statements suggesting unlikelihood of removal are insufficient. See id. The district court erred to the extent it failed to consider the statuses of A.F.’s and Castro’s asylum petitions. But the majority opinion does not stop here, instead pointing to the district court’s conclusion Castro did not face a grave risk of harm under the Hague Convention. This, it says, “undercuts any suggestion that her asylum claim will succeed.” Ante, at 14. This raises three concerns. First, there is no reason to believe that this is the only basis through which Castro seeks asylum. Second, it presupposes that the evidence provided in this proceeding is the same as that provided to USCIS—an assumption that the record does not unequivocally support. And third, it ignores that asylum seekers face different standards of proof and review than do those seeking to demonstrate a grave risk of harm under the Hague Convention. 14

_____________________ 14 Compare Hague Convention, art. 13(b) (providing that a contracting state is not bound to return a child if the person establishes that “there is a grave risk that his or her return would expose the child to physical or psychological harm or otherwise place the child in an intolerable situation”), and Soto v. Contreras, 880 F.3d 706, 712 (5th Cir. 2018) (“[T]he Hague Convention does not require objective evidence in proving the grave-risk defense by clear and convincing evidence.” (emphasis added) (citing 22 U.S.C. § 9003(e)(2)(A))), with 8 U.S.C. § 1101(a)(42) (defining “refugee” as one “who is unable or unwilling to return to, and is unable or unwilling to avail himself or herself of the protection of, [their home country] 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. § 1158(a) (granting the Attorney General discretion in such determinations), 8 C.F.R. § 208.13(a) (placing the burden of proof on the applicant “to

36 Case: 24-10520 Document: 82-1 Page: 37 Date Filed: 06/02/2025

Finally, after recognizing the district court’s failure to fully consider immigration, the majority opinion chooses not to vacate and remand. Instead, it vacates and renders judgment in favor of Brito. This leapfrogs USCIS’s review of Castro’s and A.F.’s asylum claims, based solely on the conclusion—from a limited record—that it is unlikely that “her asylum claim will succeed,” thus “erod[ing] any stability she may have developed in the United States.” Ante, at 15. No law or record evidence supports these statements. Even assuming that the outcome of Castro’s asylum application is woefully uncertain—a claim we are ill-suited to make given the lack of record evidence to support it—it cannot be said that the district court clearly erred in weighing the other five factors over immigration and age. After all, “immigration status is [not] dispositive.” Hernandez, 820 F.3d at 788. Allowing it to seep into every other factor makes it dispositive. Even if the proper review is de novo, these factors still support a finding that A.F. is well- settled in Texas. The district court did not err in these determinations.

_____________________ establish that he or she is a refugee as defined in” 8 U.S.C. § 1101(a)(42) and noting that the applicant’s testimony, “if credible, may be sufficient to sustain the burden of proof without corroboration”), 8 C.F.R. § 208.13(b)(1)(i)–(ii) (requiring a preponderance of the evidence, with the burden of proof on the INS, to overcome an asylum applicant’s showing of eligibility), 8 C.F.R. § 208.13(b)(1)(iii) (permitting an officer to grant asylum in their discretion if the applicant demonstrates “compelling reasons for being unwilling or unable to return to the country arising out of the severity of the past persecution” or they have “established that there is a reasonable possibility that he or she may suffer other serious harm upon removal to that country”), I.N.S. v. Cardoza-Fonseca, 480 U.S. 421, 431 (1987) (noting that a well-founded fear may be proven by “less than a 50% chance of the occurrence taking place” and differentiating “well-founded fear” from “clear probability”), and Orane v. Barr, 919 F.3d 904, 910 n.2 (5th Cir. 2019) (stating that the likelihood “need not be ‘more likely than not’” and that “a ‘reasonable possibility’ suffices,” but declining to select a specific percentage requirement (quoting Cardoza- Fonseca, 480 U.S. at 440)).

37 Case: 24-10520 Document: 82-1 Page: 38 Date Filed: 06/02/2025

V The question remains of whether the district court erred in failing to consider litigative delays in determining whether A.F. was well-settled. The Supreme Court has emphasized the importance of expeditious litigation of Hague Convention petitions. Chafin v. Chafin, 568 U.S. 165, 180 (2013). Brito, citing this support for expeditious disposition, argues that the Fifth Circuit has endorsed a general goal of “adjudicating Hague Convention petitions ‘within six weeks of the start of proceedings, or as expeditiously as possible within the context of the case.’” He further provides citations to courts from various other circuits that, at the least, properly identify the Hague Convention’s procedures to protect expeditious litigation. Brito argues that the district court improperly held that A.F. was well- settled by “ignoring the Hague Convention’s central pillar of expediency.” In essence, he asserts that, because Castro refused to inform him of her Texas address, and because the district court took months to try the case, it considered evidence that it would not have considered had the trial occurred sooner. Accordingly, he claims that the district court erred by not adequately considering the delay. In response, Castro argues that Brito waived this argument by failing to discuss it in the Pre-Trial Order, not objecting to evidence, and failing to press that the district court should have exercised its discretion and denied the well-settled exception, and that the ICARA advocates against his reading. Finally, she argues that Brito really brings an exclusionary argument that the court should not have considered the evidence that was brought about by Castro’s and the district court’s delays. As an initial matter, Castro misrepresents Brito’s argument. He does not request that the court exclude all evidence that accumulated between the filing of his suit (or the six-week goal) and the trial, but that the court instead

38 Case: 24-10520 Document: 82-1 Page: 39 Date Filed: 06/02/2025

must consider the passage of time and why the delays occurred. Moreover, he could not have pressed this objection at trial. “Even though [Castro’s] alleged paucity of pre-petition evidence certainly could have been fodder for [Brito’s] closing argument, it was not until the court rendered its decision that the alleged error was committed, affording [Brito] something concrete to challenge.” da Costa v. de Lima, 94 F.4th 174, 182 (1st Cir. 2024). For this reason, the First Circuit has “reject[ed] the suggestion of waiver” under these circumstances. Id. Although the trial transcript makes no mention of the alleged delays, the final judgment was rendered alongside the findings of fact and conclusions of law in which the district court failed to adequately consider its delays. There was no reasonable time during which Brito could raise such an argument, except on reconsideration. Assuming, therefore, that this argument was sufficiently preserved, we ask whether the district court needed to consider the delay. The Supreme Court provided insight in Lozano v. Montoya Alvarez, 572 U.S. 1. There, the child’s mother left the United Kingdom with her child without informing the father of her intended destination. Id. at 8. Because the mother did not inform the father of her whereabouts, and because he could not locate her, he was unable to file a Petition for Return of Child for over sixteen months. Id. Presented with an argument that the mother’s intentional concealment of the child should have equitably tolled the one-year filing timeline for the well- settled exception, the Court held that it was “unwilling to apply equitable tolling principles that would, in practice, rewrite the treaty.” Id. at 17. Instead, the Court expressed that, similar to the approaches of other signatory nations’ courts, “concealment may be taken into account in the factual determination whether the child is settled.” Id. After all, “steps taken to promote concealment can . . . prevent the stable attachments that make a child ‘settled.’” Id.

39 Case: 24-10520 Document: 82-1 Page: 40 Date Filed: 06/02/2025

Therefore, delay may be considered through the established factors. In finding that concealment could prevent a child from being well-settled, the Supreme Court cited various cases, all of which considered concealment within the Hernandez factors. E.g., Mendez Lynch v. Mendez Lynch, 220 F. Supp. 2d 1347, 1363–64 (M.D. Fla. 2002) (children lived in seven locations in eighteen months); Wigley v. Hares, 82 So. 3d 932, 942 (Fla. App. 2011) (mother intentionally kept the child from participation in community activities, sports, or church); In re Coffield, 644 N.E.2d 662, 666 (Ohio Ct. App. 1994) (child withheld from school and organized activity). These cases, and the Supreme Court’s favorable citation thereto, do not stand for the proposition that a delay dampens post-petition evidence presented in favor of a well-settled finding. To the contrary, if an abducting parent intentionally delays proceedings through active concealment (or otherwise), those acts are considered through the lens of the well-settled factors, including the child’s exposure to the environment and home stability. This interpretation tracks with Hernandez, our governing standard on the well-settled defense. 15 If delay were an independent consideration, it would have been listed among

_____________________ 15 It also comports with the approach of the only other circuit court to consider this issue. In da Costa, the First Circuit faced the argument that reliance solely on post-petition evidence “does not align with the reasoning behind the now settled defense.” 94 F.4th at 181. But the court held that “the Convention itself gives a strong indication that post- petition evidence remains important.” Id. at 182. Specifically, it noted that “[t]he phrase ‘now settled’—the wording of which itself suggests an emphasis on the present—is introduced in the context of post-petition circumstances without reference to pre-petition circumstances.” Id. It concluded that one would have expected the drafters to have “expressed that intent more explicitly in the text” if pre-petition evidence were required. Id. at 183. This comports with the plain text of the treaty.

40 Case: 24-10520 Document: 82-1 Page: 41 Date Filed: 06/02/2025

the several factors. It was not. To the extent that Lozano extended those considerations, it did so by incorporating time into the existing inquiry. 16 So, the question is whether the district court erred in failing to consider the contributions of the delays when analyzing the seven aforementioned factors. Brito asserts that some evidence would not be considered but for the delays, including that A.F. would have lived in Texas for a shorter period of time, would not have gone to school for as long, would not have participated in the community as much, and would not have seen a doctor or begun to learn English. But, for all of these arguments, Brito never mentions how the delays impacted the specific criteria that the court was to consider under Hernandez. None of his examples show that Castro kept A.F. from participating in the community, going to school, meeting friends, living in a stable home, or otherwise growing settled in Texas. Instead, the additional time resulted in A.F.’s schooling, at which she has excelled, her participation in community functions, and her establishment of friendships and relationships in Texas. The delay does not cut against a finding that A.F. was well-settled in Texas. As to the expediency requirement, Brito is correct that a six-week goal exists: “Article 11 of the Hague Convention contemplates an immediate emergency hearing in international child abduction cases and a judicial decision within six weeks.” Lops v. Lops, 140 F.3d 927, 944 (11th Cir. 1998). _____________________ 16 Lozano works hand-in-hand with a related federal regulation, shedding light on how to consider time delays. See Hague International Child Abduction Convention; Text and Legal Analysis, 51 Fed. Reg. 10494, 10509 (Mar. 26, 1986) (“The reason for the passage of time, which may have made it possible for the child to form ties to the new country, is also relevant to the ultimate disposition of the return petition. If the alleged wrongdoer concealed the child’s whereabouts from the custodian necessitating a long search for the child and thereby delayed the commencement of a return proceeding by the applicant, it is highly questionable whether the respondent should be permitted to benefit from such conduct absent strong countervailing considerations.”).

41 Case: 24-10520 Document: 82-1 Page: 42 Date Filed: 06/02/2025

But he makes no showing that the Convention or the ICARA requires a judicial determination in six weeks (or as close thereto as possible). As the Eleventh Circuit noted, Article 11 contemplates, but does not demand, an immediate emergency hearing. Id. The Convention states that the judicial authorities “shall act expeditiously in proceedings for the return of children.” Hague Convention, art. 11 (emphasis added). But, as for relief, it only provides that if the judicial authority has not yet reached a decision in six weeks, the applicant “shall have the right to request a statement of the reasons for the delay.” Id. There is no indication in the record or the briefing that Brito sought such relief. Even if he did request a statement, the answer would have been clear. His briefing and underlying requests for reassignment demonstrate as much. He filed the petition over one year after A.F.’s removal, litigated in the improper district for part of the time, and the judge to whom he was assigned had no trial availability for months. Ultimately, if a district court is too dilatory in setting the trial, a party can follow “familiar judicial tools” and petition for a writ of mandamus; after all, “courts can and should take steps to decide these cases as expeditiously as possible.” Chafin, 568 U.S. at 178– 79. 17

_____________________ 17 Brito did not move to expedite this appeal. While that does not impact the ultimate determination, Chafin advocates for prompt return of children “through the familiar judicial tools of expediting proceedings.” 568 U.S. at 178. It is true that “courts can and should take steps to decide these cases as expeditiously as possible.” Id. at 179. And “[e]xpedition will help minimize the extent to which uncertainty adds to the challenges confronting both parents and child.” Id. at 180. But “the Convention does not prescribe modes of, or time frames for, appellate review of first instance decisions.” Id. at 181 (Ginsburg, J., concurring). “It therefore rests with each Contracting State to ensure that appeals proceed with dispatch.” Id. Brito could have—but chose not to—follow this oft-trodden, established judicial path.

42 Case: 24-10520 Document: 82-1 Page: 43 Date Filed: 06/02/2025

The district court did not err by not considering its delay, nor was it improper to set its trial date outside of the aspirational six-week time frame. VI Hague Convention cases are difficult and sad matters. I sympathize with Brito’s inability to enter the United States to visit his child. But the Hague Convention does not permit a court to adjudicate the merits of custody disputes. See England, 234 F.3d at 271; 22 U.S.C. § 9001(b)(4). As difficult as it is to be separated from a child, A.F.’s relationship with Brito in Texas is no different than it was after he moved to Spain: all interactions are virtual. It is not our province to consider his ability to see A.F. She left Venezuela when she was three. Her entire life as she knows it—including the last three years—is in Texas, and she has grown well-settled. The majority opinion today fails to provide sufficient deference to the district court and reshapes Hague Convention jurisprudence by creating a new emphasis on immigration. It relies on inferential leaps based on limited evidence to determine that a child—who lives a stable, happy, and enriching life in Texas—should be uprooted because of various hypothetical possibilities. This does not comport with our case law, nor does it fit within the purposes of the Hague Convention. I respectfully dissent.