FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

EAST BAY SANCTUARY No. 23-16032

COVENANT; CENTRAL

AMERICAN RESOURCE CENTER; D.C. No. 4:18-cv-

TAHIRIH JUSTICE CENTER; 06810-JST

NATIONAL CENTER FOR

LESBIAN RIGHTS; IMMIGRANT

DEFENDERS LAW CENTER; ORDER

AMERICAN GATEWAYS,

Plaintiffs-Appellees,

v.

JOSEPH R. BIDEN, President of the United States; MERRICK B. GARLAND, Attorney General;

UNITED STATES DEPARTMENT

OF JUSTICE; DAVID NEAL;

EXECUTIVE OFFICE FOR

IMMIGRATION REVIEW;

ALEJANDRO N. MAYORKAS; U.S.

DEPARTMENT OF HOMELAND

SECURITY; UR M. JADDOU;

UNITED STATES CITIZENSHIP

AND IMMIGRATION SERVICES;

TROY A. MILLER; UNITED

STATES CUSTOMS AND BORDER PROTECTION,

Defendants-Appellants.

Filed May 22, 2024

Before: William A. Fletcher, Richard A. Paez, and Lawrence VanDyke, Circuit Judges.

Order; Dissent by Judge VanDyke

SUMMARY *

Immigration/Intervention

In a case in which the government appeals the district court’s order vacating the government’s rule, Circumvention of Lawful Pathways (“the Rule”), the panel issued an order denying a motion to intervene filed by the States of Alabama, Kansas, Georgia, Louisiana, and West Virginia (“the States”). Noting that the appeal is currently in abeyance for settlement discussions, the majority explained that the States sought to intervene to participate in settlement negotiations and possibly to object to any proposed settlement and to request that the court stay its order. The panel observed that

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

EAST BAY SANCTUARY COVENANT V. BIDEN 3

it was unaware of any instance where a court permitted intervention at the appellate stage while the case was held in abeyance for settlement negotiations. Because no statute or rule governs intervention on appeal, the policies underlying Federal Rule of Civil Procedure 24 guided the majority’s analysis. To intervene as of right under Rule 24(a)(2), the States must show that: (1) their motion is timely; (2) they have a significantly protectable interest relating to the property or transaction which is the subject of the action; (3) the disposition of the action may as a practical matter impair or impede their ability to protect that interest; and (4) their interest is inadequately represented by the parties to the action. The majority concluded that the States had not shown that they have a “significant protectible interest” in either: 1) maintaining the Rule or in reducing immigration into the United States; or 2) minimizing their expenditures and preserving their population-based political representation. Because that failure alone was a sufficient ground to deny intervention as of right, the majority did not reach the remaining factors. For similar reasons, the majority declined to exercise its discretion to allow the States to intervene permissively, explaining that the nature and extent of the States’ interest in this appeal are far too attenuated to support intervention. Dissenting, Judge VanDyke would grant the States’ motion to intervene as of right. Judge VanDyke explained that this court has continually acknowledged that it should construe Rule 24(a)’s requirements of intervention as of right liberally and broadly in favor of proposed intervenors. Under this lenient standard, Judge VanDyke concluded that: (1) the States’ motion was timely because they acted swiftly after the court granted the parties’ unexpected motion to stay; (2) the States’ interests would clearly be harmed by a settlement that negates the effect of the Rule because they would be forced to bear the additional costs of illegal immigration into their territories; and (3) the federal government’s openness to settlement discussion after months of aggressively defending the Rule evinces a change in position such that it no longer adequately represents the States’ interests.

COUNSEL

Brian M. Boynton (argued), Principal Deputy Assistant Attorney General; Daniel J. Tenny, Sean R. Janda, and Brian J. Springer, Appellate Staff Attorneys; Antitrust Division, United States Department of Justice, Washington, D.C.; Christina P. Greer and Patrick J. Glen, Senior Litigation Counsel; Erez Reuveni, Assistant Director; William C. Peachey, Director; Office of Immigration Litigation, Civil Division, United States Department of Justice, Washington, D.C.; for Defendants-Appellants. Spencer E. Wittmann Amdur (argued), Katrina L. Eiland, Morgan Russell, Oscar S. Roman, and Cody Wofsy, American Civil Liberties Union Foundation Immigrants’ Rights Project, San Francisco, California; Lee P. Gelernt, Omar C. Jadwat, Wafa Junaid, and Judy Rabinovitz, American Civil Liberties Union Foundation Immigrants’ Rights Project, New York, New York; Melissa E. Crow, Center for Gender & Refugee Studies, Washington, D.C.; Anne E. Peterson, Blaine Bookey, Julie B. Bourdoiseau, and Karen Musalo, Center for Gender & Refugee Studies, San Francisco, California; Robert Pauw, CGRS Cooperating

EAST BAY SANCTUARY COVENANT V. BIDEN 5

Attorney, Gibbs Houston Pauw, Seattle, Washington; Keren H. Zwick, Richard Caldarone, Colleen Cowgill, and Mary Georgevich, National Immigrant Justice Center, Chicago, Illinois; Michelle Y. Cho, American Civil Liberties Union Foundation of Northern California, Inc., San Francisco, California; Angelo Guisado, Center for Constitutional Rights, New York, New York; for Plaintiffs-Appellees. Matt A. Crapo and Christopher J. Hajec, Immigration Reform Law Institute, Washington, D.C., for Amicus Curiae Immigration Reform Law Institute. Katherine L. Evans and Charles S. Ellison, Duke University School of Law Immigrant Rights Clinic, Durham, North Carolina, for Amici Curiae Professors of Immigration Law. Ashley B. Vinson, Akin Gump Strauss Hauer & Feld LLP, San Francisco, California, for Amici Curiae Former Immigration Judges & Former Members of the Board of Immigration Appeals. Alice Farmer, Office of the United Nations High Commissioner for Refugees, Washington, D.C.; Robert R. Anderson, Arnold & Porter Kaye Scholer LLP, Denver, Colorado; Samuel M. Witten and Kaitlin Konkel, Arnold & Porter Kaye Scholer LLP, Washington, D.C.; for Amicus Curiae Office of the United Nations High Commissioner for Refugees. Kathleen R. Hartnett and Zoë Helstrom, Cooley LLP, San Francisco, California, for Amicus Curiae National Citizenship and Immigration Services Council 119. Cameron C. Russell and Rebecca Berman, Freshfields Bruckhaus Deringer US LLP, New York, New York; Justina Sessions and J. Mia Tsui, Freshfields Bruckhaus Deringer US LLP, Redwood City, California; Seve Kale, Freshfields Bruckhaus Deringer US LLP, Washington, D.C.; for Amici Curiae Asylum Access México A.C. and Instituto para las Mujeres en la Migración A.C. Anwen Hughes, Christina Asencio, Rebecca Gendelman, and Licha M. Nyiendo, Human Rights First, New York, New York; Farida Chehata, Human Rights First, Los Angeles, California; for Amici Curiae Human Rights and Legal Services Organizations. Matt A. Crapo and Christopher J. Hajec, Immigration Reform Law Institute, Washington, D.C., for Amicus Curiae Immigration Reform Law Institute. Neville S. Hedley, Hamilton Lincoln Law Institute, Washington, D.C., for Amicus Curiae Hamilton Lincoln Law Institute. Bridget K. O’Hickey, Assistant Solicitor General; James H. Percival, Chief of Staff; Henry C. Whitaker, Solicitor General, Antitrust Section; Ashley Moody, Attorney General of Florida; Office of the Attorney General, Tallahassee, Florida; for Amici Curiae States of Florida, Arkansas, Indiana, Iowa, Kentucky, Mississippi, Missouri, Montana, Nebraska, New Hampshire, North Dakota, Ohio, South Carolina, South Dakota, Texas, Utah, and Virginia.

EAST BAY SANCTUARY COVENANT V. BIDEN 7

ORDER

The States of Alabama, Kansas, Georgia, Louisiana, and West Virginia (“the States”) seek to intervene in this appeal. Dkt. No. 86. For the reasons explained below, we deny the motion. This appeal is currently held in abeyance. In July 2023, after the parties filed cross-motions for summary judgment, the district court granted summary judgment in Plaintiffs’ favor and vacated the challenged rule (the “Rule”). See E. Bay Sanctuary Covenant v. Biden, 683 F. Supp. 3d 1025, 1053 (N.D. Cal. 2023); Circumvention of Lawful Pathways, 88 Fed. Reg. 31314 (May 16, 2023) (codified at 8 C.F.R. §§ 208.33, 1208.33). The government sought a stay pending appeal, which we granted. E. Bay Sanctuary Covenant v. Biden, No. 21., slip op. at 1 (9th Cir. Aug. 3, 2023). We heard oral argument in November 2023. Dkt. No. 82. In February 2024, the parties jointly requested that we hold the appeal in abeyance while they explored settlement. Dkt. No. 83. We granted the motion. E. Bay Sanctuary Covenant v. Biden, 93 F.4th 1130, 1131 (9th Cir. 2024). As of May 2024, the parties remain engaged in settlement discussions. Dkt. No. 111. The States now seek to intervene in order to participate in settlement negotiations and possibly to object to any proposed settlement and to request that we vacate our stay order. In their motion, the States claim an interest in preserving the Rule in order to reduce unlawful immigration, thereby minimizing state expenditures and preserving their population-based political representation. The States contend that they did not seek to intervene earlier because they previously believed that their interests would be represented by the government. Only after they learned of the stay and settlement negotiations did they seek to intervene. “Intervention at the appellate stage is, of course, unusual and should ordinarily be allowed only for ‘imperative reasons.’” Bates v. Jones, 127 F.3d 870, 873 (9th Cir. 1997) (quoting Landreth Timber Co. v. Landreth, 731 F.2d 1348, 1353 (9th Cir. 1984), rev’d on other grounds, 471 U.S. 681 (1985)). 1 This Court disfavors putative intervenors who merely seek to “attack or thwart” a remedy. United States v. Alisal Water Corp., 370 F.3d 915, 922 (9th Cir. 2004). We are unaware of any instance where a court has permitted

1 Other circuits similarly distinguish between intervention at the district court and intervention on appeal. See Richardson v. Flores, 979 F.3d 1102, 1105 (5th Cir. 2020) (noting there is a “steep threshold for allowing intervention on appeal”); Pub. Serv. Co. of N. M. v. Barboan, 857 F.3d 1101, 1113 (10th Cir. 2017) (“Though we usually take a liberal view of Rule 24(a), when an applicant has not sought intervention in the district court, we permit it on appeal ‘only in an exceptional case for imperative reasons.’” (quoting Elliott Indus. Ltd. P’ship v. BP Am. Prod. Co., 407 F.3d 1091, 1103 (10th Cir. 2005))); Hall v. Holder, 117 F.3d 1222, 1231 (11th Cir. 1997) (“A court of appeals may, but only in an exceptional case for imperative reasons, permit intervention where none was sought in the district court.” (quoting McKenna v. Pan Am. Petroleum Corp., 303 F.2d 778, 779 (5th Cir. 1962))); Amalgamated Transit Union Int’l, AFL-CIO v. Donovan, 771 F.2d 1551, 1552–53, 1553 n.3 (D.C. Cir. 1985) (“A court of appeals may allow intervention at the appellate stage where none was sought in the district court ‘only in an exceptional case for imperative reasons.’” (quoting Landreth, 731 F.2d at 1353)); In re Grand Jury Investigation, 587 F.2d 598, 601 (3d Cir. 1978) (“Those courts which have considered the question have recognized that while a court of appeals has power to permit intervention that power should be exercised only in exceptional circumstances for imperative reasons.”); see also Ass’n for Educ. Fairness v. Montgomery Cnty. Bd. of Educ., 88 F.4th 495, 499 (4th Cir. 2023) (discussing the unique concerns raised by appellate intervention).

EAST BAY SANCTUARY COVENANT V. BIDEN 9

intervention at the appellate stage while the case was held in abeyance to allow settlement negotiations. Although “[n]o statute or rule provides a general standard to apply in deciding whether intervention on appeal should be allowed,” the “policies underlying” Federal Rule of Civil Procedure 24 guide our analysis. Cameron v. EMW Women’s Surgical Ctr., P.S.C., 595 U.S. 267, 276–77 (2022); Day v. Apoliona, 505 F.3d 963, 964 (9th Cir. 2007). To intervene as of right under Rule 24(a)(2), the States must show that: (1) their motion is timely; (2) they have a “significantly protectable interest relating to the property or transaction which is the subject of the action;” (3) “the disposition of the action may as a practical matter impair or impede [their] ability to protect that interest;” and (4) their “interest is inadequately represented by the parties to the action.” Wilderness Soc’y v. U.S. Forest Serv., 630 F.3d 1173, 1177 (9th Cir. 2011) (en banc) (quoting Sierra Club v. EPA, 995 F.2d 1478, 1481 (9th Cir. 1993)). The States “bear[] the burden of showing that these four elements are met.” W. Watersheds Project v. Haaland, 22 F.4th 828, 835 (9th Cir. 2022). 2 The States have not shown that they have a “significantly protectable interest” in the litigation. Wilderness Soc’y, 630

2 The dissent is correct that we generally construe Rule 24(a)(2)’s requirements for intervention as of right in the district court “broadly in favor of proposed intervenors.” Wilderness Soc’y, 630 F.3d at 1179. Even so, we see nothing to suggest that Bates and Landreth—which counsel that appellate intervention requires closer scrutiny—have been overruled. See Bates, 127 F.3d at 873; Landreth, 731 F.2d at 1353. This is not to say that intervention at the appellate stage is impossible. We have, in appropriate circumstances, permitted intervention when the relevant factors are satisfied. See Peruta v. Cnty. of San Diego, 824 F.3d 919, 940 (9th Cir. 2016); Day, 505 F.3d at 966. F.3d at 1179. A proposed intervenor “has a significant protectable interest if the interest is protected by law and there is a relationship between that interest and the claim or claims at issue.” Cooper v. Newsom, 13 F.4th 857, 865 (9th Cir. 2021). The States’ asserted interests are insufficient to support intervention in this appeal for purposes of exerting influence on settlement negotiations. First, the States do not have a significant protectible interest in maintaining the Rule or in reducing immigration into the United States. Although “federal policies frequently generate indirect effects on state revenues or state spending,” states have no legally protectible interest in compelling enforcement of federal immigration policies. United States v. Texas, 599 U.S. 670, 677–80, 680 n.3 (2023); see also Sure-Tan, Inc. v. NLRB, 467 U.S. 883, 897 (1984); Linda R.S. v. Richard D., 410 U.S. 614, 619 (1973). 3 Second, the States do not have a significant protectable interest in minimizing their expenditures and preserving their population-based political representation. While an “economic interest may be sufficient to support a right of intervention,” it “must be concrete and related to the underlying subject matter of the action.” Alisal Water Corp., 370 F.3d at 919 (citing Arakaki v. Cayetano, 324 F.3d 1078,

3 Although Texas is about Article III standing, it holds that absent other circumstances, states cannot assert an interest in procuring greater immigration enforcement. See 599 U.S. at 677, 681–83. The dissent cites no case for the proposition that the States can assert such an interest merely because the Rule’s lawfulness is at issue. We decline to reach a conclusion contrary to the principles articulated by the Supreme Court in Texas.

EAST BAY SANCTUARY COVENANT V. BIDEN 11

1085, 1088 (9th Cir. 2003)). 4 Even if disposition of this appeal might affect state expenditures and political representation, such incidental effects are not at issue in the litigation and are, in any event, attenuated and speculative. See id. at 920 (holding that an intervenor’s claimed interest cannot be “several degrees removed from the [issues] that are the backbone of [the] litigation”). 5 We therefore conclude that the States lack the requisite significant protectable interest to support intervention as of right under Rule 24(a). “Because that failure alone is a sufficient ground to deny intervention as of right,” we do not reach the remaining factors. See Cooper, 13 F.4th at 865 (citing Perry v. Proposition 8 Off. Proponents, 587 F.3d 947, 950 (9th Cir. 2009)). For similar reasons, we decline to exercise our discretion to allow the States to intervene permissively. See Fed. R. Civ. P. 24(b)(1)(B); Cameron, 595 U.S. at 278–79 (“Resolution of a motion for permissive intervention is

4 The States allege that elimination of the Rule would cause increased immigration, and that at least some immigrants would end up in their states and thus strain state resources. As this causal chain demonstrates, the States’ asserted interest in minimizing expenditures is thus removed from the “subject of the action”—and predicated on an alleged interest in procuring greater immigration enforcement. Fed. R. Civ. P. 24(a)(2). 5 See also Med. Liab. Mut. Ins. Co. v. Alan Curtis LLC, 485 F.3d 1006, 1008 (8th Cir. 2007)) (“Economic interests or interests contingent on a sequence of events are generally insufficient for mandatory intervention.”); United States v. Peoples Benefit Life Ins. Co., 271 F.3d 411, 415 (2d Cir. 2001) (“An interest that is remote from the subject matter of the proceeding, or that is contingent upon the occurrence of a sequence of events before it becomes colorable, will not satisfy the rule.” (quoting Wash. Elec. Coop., Inc. v. Mass. Mun. Wholesale Elec. Co., 922 F.2d 92, 97 (2d Cir. 1990)). committed to the discretion of the court before which intervention is sought.”). The “nature and extent” of the States’ interest in this appeal are far too attenuated to support intervention. Perry, 587 F.3d at 955. Accordingly, the States’ Motion to Intervene is

DENIED.

VANDYKE, Circuit Judge, dissenting:

Our court has continually acknowledged that we should construe Rule 24(a)(2)’s requirements for intervention as of right “broadly in favor of proposed intervenors.” Wilderness Soc’y v. U.S. Forest Serv., 630 F.3d 1173, 1179 (9th Cir. 2011) (en banc) (citation omitted); see also Sw. Ctr. for Biological Diversity v. Berg, 268 F.3d 810, 818 (9th Cir. 2001) (“In general, we construe Rule 24(a) liberally in favor of potential intervenors.”); U.S. ex rel. McGough v. Covington Techs. Co., 967 F.2d 1391, 1394 (9th Cir. 1992). We have explained that this “liberal policy in favor of intervention serves both efficient resolution of issues and broadened access to the courts.” Wilderness Soc’y, 630 F.3d at 1179 (quoting United States v. City of Los Angeles, 288 F.3d 391, 397–98 (9th Cir. 2002)). Yet the majority once again inexplicably throttles this standard by cursorily denying the States’ request to intervene in this case. This continues a troubling trend by our court of denying intervention whenever it might upset a possible collusive settlement resulting in a favored policy. See, e.g., City & County of San Francisco v. USCIS, 992 F.3d 742 (9th Cir.

EAST BAY SANCTUARY COVENANT V. BIDEN 13

2021); Cooper v. Newsom, 26 F.4th 1104 (9th Cir. 2022). 1 That is not what I think our court meant when it said “we construe Rule 24(a) liberally,” so I must respectfully dissent. When analyzing a motion to intervene as of right, we apply a four-part test: “(1) the motion must be timely; (2) the applicant must claim a ‘significantly protectable’ interest relating to the property or transaction which is the subject of the action; (3) the applicant must be so situated that the disposition of the action may as a practical matter impair or impede its ability to protect that interest; and (4) the applicant’s interest must be inadequately represented by the parties to the action.” Wilderness Soc’y, 630 F.3d at 1177 (quoting Sierra Club v. EPA, 995 F.2d 1478, 1481 (9th Cir. 1993)). The States’ motion here was timely because they acted swiftly after we granted the parties’ unexpected motion to stay. Their interests would clearly be harmed by a settlement that negates the effect of the Rule because they would be forced to bear the additional costs of illegal immigration into their territories. And the federal government’s openness to settlement discussions after months of aggressively defending the Rule evinces a change in position such that it no longer adequately represents the States’ interests. I therefore would grant the States’ motion to intervene as of right.

I. The States’ Motion Was Timely.

As an initial matter, only the organizational plaintiffs contest the timeliness of the States’ motion—the federal government does not. “Timeliness is to be determined from

1 Compare Peruta v. County of San Diego, 824 F.3d 919, 940–41 (9th Cir. 2016) (granting California’s motion to intervene at the appellate stage to argue against the Second Amendment). all the circumstances … by the court in the exercise of its sound discretion.” NAACP v. New York, 413 U.S. 345, 366 (1973). In making this determination, we generally look to “(1) the stage of the proceeding, (2) the prejudice to other parties, and (3) the reason for and length of the delay.” Day v. Apoliona, 505 F.3d 963, 965 (9th Cir. 2007) (cleaned up) (citation omitted). The majority relies heavily on the first factor, emphasizing the appellate posture of these proceedings. The majority makes much of this fact, citing cases in which, notwithstanding the generally permissive standard for intervention, our court and others have rejected attempts to intervene at the appellate stage after failing to seek intervention before the district court. But it is not at all clear from the cited cases that this or any other circuit has adopted a rule under which the mere fact that a proceeding has reached the appellate stage, standing alone, is a particularly relevant factor. In my view, there is nothing particularly unique about the “appellate” nature of the proceedings except that often it is a good proxy for the fact that a case has already been going on for quite a long time and the putative intervenors could have sought to intervene earlier. Under ordinary circumstances, a case that has already been appealed would have afforded a potential intervenor more than adequate time to intervene and protect its interests. Thus, there is often a good reason to conclude a motion to intervene is untimely where the intervenor had notice that its rights were not adequately protected by the parties before the district court, yet nonetheless declined to seek intervention at that stage. But, as illustrated by these proceedings, it can also be the case that an intervenor’s rights were adequately protected until a party abruptly changed its position at the appellate

EAST BAY SANCTUARY COVENANT V. BIDEN 15

stage. Under such circumstances, it makes little sense to conclude that a motion to intervene, brought promptly after the intervenor discovered its interests were no longer adequately represented, is untimely just because the party waited until the appeal stage to abruptly change its litigation posture. Instead of applying the majority’s wooden, inflexible standard that apparently accounts for nothing more than the stage of the proceeding, we would do well to remember that the “most important circumstance relating to timeliness is that the [States] sought to intervene as soon as it became clear that [their] interests would no longer be protected by the parties in the case.” Cameron v. EMW Women’s Surgical Ctr., P.S.C., 595 U.S. 267, 279–80 (2022) (cleaned up and emphasis added). That “most important” consideration strongly weighs in favor of intervention here. Not only does the majority dramatically overread the supposed rule from its lead case, Bates v. Jones, but the result in that case actually supports granting the States’ request for intervention. 127 F.3d 870 (9th Cir. 1997). In Bates, the state raised for the first time after oral argument before this court a position that would imperil the intervenors’ interests—as our court noted, “[u]ntil that assertion, the proposed intervenors believed that they would be able to benefit from a favorable ruling by this court.” Id. at 873. Similarly, the federal government here did not raise the prospect of a settlement that could affect the States’ interest in the Rule until after we had already heard oral argument. Until then, the government’s vigorous defense of the Rule led the States to reasonably believe that their interests were adequately represented. The Bates court ultimately considered the motion to intervene timely even on appeal because the intervenors were not aware until then that their interests would no longer be protected. Id. The States’ motion here is timely for the same reason: the States were understandably unaware of any possibility that their interests were not adequately being protected until the government changed its position on defending the Rule by seeking an abeyance from this court after oral argument was heard. The other cases cited by the majority do not support the majority’s rule because none of them involve a case where a party vigorously defended the intervenor’s interests until changing course at the last minute during the appeal. For example, in Landreth Timber Co. v. Landreth, the intervenors sought to intervene at the appellate stage “without stating any reason for failure to intervene in the district court” beyond wanting to avail themselves of a legal doctrine that was already implicated in the defendants’ litigation position. 731 F.2d 1348, 1353 (9th Cir. 1984), rev’d on other grounds, 471 U.S. 681 (1985). The Landreth court found it important that “intervention would raise new issues of fact and law not before the district court.” Id. In this case, by contrast, the States have provided a compelling reason for not intervening earlier: their interests were adequately protected by the government’s then-ongoing defense of the Rule. And because the States merely seek to continue the same defense that the government might now abandon, their intervention introduces no new issues of fact or law. 2

2 The remaining cases cited by the majority are similarly distinguishable. For example, in Public Service Company of New Mexico v. Barboan, the party with whom the intervenor’s interests aligned continued to “have the same legal objective” throughout the case and the intervenor “had ample opportunity to be heard at the district court and declined to do so.” 857 F.3d 1101, 1113–14 (10th Cir. 2017). And in Amalgamated Transit Union International, AFL-CIO v. Donovan, “the alleged conflict of interest between [the intervenor] and the [party] existed at the

EAST BAY SANCTUARY COVENANT V. BIDEN 17

The Fourth Circuit case cited by the majority is even less relevant, and the majority’s reliance on it is especially puzzling. In Association for Education Fairness v. Montgomery County Board of Education, the intervenors actually sought intervention in the district court, 88 F.4th 495, 499 (4th Cir. 2023), which already suggests that it is nothing like the present case. And when the district court finally denied their motion to intervene, it did so not “because it was unwarranted under” Rule 24, but because “the intervention question had been rendered academic by its decision to enter a final judgment for the parties on whose side the organizations sought to intervene.” Id. After the adverse party appealed that judgment, the intervenors again sought to intervene. Because the intervention issue was unnecessary until the appeal was initiated, the Fourth Circuit “reject[ed] the argument that the organizations delayed unduly in seeking to bring the intervention issue before th[e] Court.” Id. at 500. That court then analyzed the other intervention factors and, although it determined the question of “whether the existing parties [would] adequately protect the organizations’ interests” was a “close one,” it ultimately found this factor unsatisfied for the straightforward reason that the party “unambiguously represented to th[e] Court” that it intended to defend the position advanced by the intervenors. Id. at 500–01. Because that issue was close, the Fourth Circuit denied the motion to intervene without prejudice so that the

commencement of th[e] litigation,” so the reasons provided for intervention on appeal “would have been equally applicable at much earlier stages of th[e] case.” 771 F.2d 1551, 1554 (D.C. Cir. 1985). In none of the cases cited by the majority did the party that ostensibly protected the intervenor’s interests abandon its defense of those interests at the appellate stage. intervenors could renew their request if the party failed to fulfill its promise to advance that position. Id. at 502. As the Fourth Circuit denied intervention on the sole basis that the intervenors’ interests were adequately represented—a prong of the intervention analysis entirely unrelated to the majority’s rationale here—Education Fairness cannot support the majority’s supposed rule that a motion to intervene presented on appeal is untimely. To reiterate, none of the cases cited by the majority where a court denied intervention at the appeal stage as untimely involve anything even close to this situation, where the federal government vigorously defended the Rule but then appears poised to abandon its position on appeal. The majority brusquely notes that it is “unaware of any instance where a court has permitted intervention” under these unique circumstances, but that observation cuts both ways: the lack of precedent going either way is hardly surprising given that it is a rare thing indeed for the government to pull this kind of surprising switcheroo after litigating a case all the way past oral argument on appeal. The government’s highly unusual behavior not only explains why intervention is justified; it makes one wonder if the parties are engaging in “rulemaking-by-collective-acquiescence” or some other collusive strategy designed to lessen the political impact of this litigation during an election year. Arizona v. City & County of San Francisco, 596 U.S. 763, 766 (2022) (Roberts, C.J., concurring) (quoting City & County of San Francisco v. USCIS, 992 F.3d 742, 744 (9th Cir. 2021) (VanDyke, J., dissenting)). Regardless, the bottom line in a case like this is ultimately pretty simple: where the States “sought to intervene as soon as it became clear that [their] interests would no longer be protected by the parties in the case,” their

EAST BAY SANCTUARY COVENANT V. BIDEN 19

motion is not untimely simply because it was filed at the appeal stage. Cameron, 595 U.S. at 279–80 (cleaned up). 3 In contrast to the majority’s string cite of dissimilar cases, a more analogous situation to the present circumstance is the Supreme Court’s practice of appointing an amicus to argue a position before the Supreme Court where a party is no longer willing to do so. See, e.g., Mata v. Lynch, 576 U.S. 143, 147 (2015) (“And because the Federal Government agrees with Mata that the Fifth Circuit had jurisdiction over his appeal, we appointed an amicus curiae to defend the judgment below.”). Like many situations in which the Supreme Court is faced with substantial, unexpected agreement between adverse parties, here, the government has now indicated it may take a joint position with the plaintiffs, leaving no one to argue for the validity of the Rule. The States don’t seek to make any new arguments; they just want to continue to advance the same position that may be abandoned by the government. If the practice of allowing a third party to argue

3 The majority emphasizes that intervention at the appellate stage is not “impossible,” citing two cases, one of which is Peruta, in which California was permitted to intervene in order to argue against the Second Amendment. Aside from furthering the ugly perception that our intervention standard may implicitly differ based on a putative intervenor’s position on the merits, cases like Peruta actually support intervention here. In Peruta, our court granted California’s motion to intervene at the appellate stage because, even though “California sought to intervene at a relatively late stage in the proceeding,” it “had no strong incentive to seek intervention … at an earlier stage, for it had little reason to anticipate” that its interests would not be protected. 824 F.3d at 940. Sound familiar? Similarly, in Day, we granted Hawaii’s motion to intervene at the appellate stage and recognized that “[a] would-be intervenor’s delay in joining the proceedings is excusable when the intervenor does not know or have reason to know that his interests might be adversely affected by the outcome of litigation.” 505 F.3d at 965 (cleaned up). a neglected position is good enough for an appeal before the Supreme Court, it should be good enough for this appellate court. We are particularly likely to conclude a motion to intervene on appeal is timely where the parties have not alleged any prejudice. See Bates, 127 F.3d at 874 (“Like other cases in which intervention has been allowed on appeal, the state does not assert that it will be prejudiced by intervention, and we find no reason to think otherwise.”); Peruta, 824 F.3d at 940 (granting California’s motion to intervene at the appellate stage because even though the state “sought to intervene at a relatively late stage in the proceeding,” “the timing of California’s motion to intervene did not prejudice Plaintiffs”). Here, neither the plaintiffs nor the government have alleged they would be prejudiced by granting the States’ motion to intervene. The second factor therefore also supports the timeliness of the States’ motion for intervention. As discussed above, “the most important circumstance relating to timeliness,” Cameron, 595 U.S. at 279–80, involves the third prong: if the States delayed, and their reasons for doing so. The critical date for assessing delay is “when proposed intervenors should have been aware that their interests would not be adequately protected by the existing parties.” W. Watersheds Project v. Haaland, 22 F.4th 828, 836 (9th Cir. 2022) (citation omitted). The States argue that they were unaware that their interests were not being protected by the government until the filing of the joint motion to hold the appeal in abeyance. And for good reason: the government actually defended the Rule until that time. The plaintiffs argue that in spite of the government’s defense of the Rule, the States should have known that their

EAST BAY SANCTUARY COVENANT V. BIDEN 21

interests would not be protected because of the administration’s hostility to the States’ general policy preferences. But just because the States may have general policy preferences different than the administration’s does not mean that they have opposing interests in every case. The very case the plaintiffs rely on highlights this issue. In United States v. Washington, we denied intervention because of years-long, ongoing hostility regarding the actual litigation in that case. 86 F.3d 1499, 1503–04 (9th Cir. 1996). Here, the government initially defended the Rule, a litigation position that aligned with the States’ interests. So even assuming the States may disagree with many of the government’s immigration positions, their interests were aligned in this litigation by the government’s continued defense of the Rule. Their interests only ceased to be adequately protected once the government gave up its defense of the Rule and entered into settlement negotiations. Furthermore, endorsing the plaintiffs’ position would have wide-reaching implications for future litigation. Their argument is essentially that the States should have moved to intervene at the very beginning of this litigation because their policy positions made them generally distrustful of this administration’s commitment to defending our nation’s immigration laws. If we applied this reasoning, then any state with different policy positions than a current administration would have to intervene at the beginning of any suit against the government it could have an interest in— or essentially waive its intervention rights entirely. Given the diversity of policy opinions in this country, there would conceivably always be some states generally at odds with an administration, and those states would be saddled with the burden of seeking to intervene in every case against the federal government, even where the government has acted consistently with their interests in that particular litigation. Part of the reason we are generally generous in allowing intervention, Wilderness Soc’y, 630 F.3d at 1179, is precisely to avoid requiring parties to seek merely prophylactic intervention where it’s unnecessary. Because the government adequately represented the States’ interests in this litigation until it didn’t, the States could not have known that their interests would be abandoned until the parties filed their joint motion. The States moved swiftly to intervene after the joint motion was filed, and so the motion to intervene was timely. See Cameron, 595 U.S. at 279–80.

II. The States’ Interests Could be Impaired by a

Disposition in This Action. The two factors associated with the States’ interests are related and can be addressed together. As an initial matter, the States do not need Article III standing to intervene. The Supreme Court has told us that, in the intervention context, “[f]or all relief sought, there must be a litigant with standing.” Town of Chester v. Laroe Estates, Inc., 581 U.S. 433, 439 (2017). So “an intervenor of right must have Article III standing in order to pursue relief that is different from that which is sought by a party with standing.” Id. at 440 (emphasis added). Because the States seek to intervene as defendants and seek only the same outcome as the government—to uphold the Rule—standing is not required. While intervention does not necessarily require standing, it does require a “significantly protectable interest” that could be impaired by a disposition of the action. Wilderness Soc’y, 630 F.3d at 1177. In making this determination, we are guided by “practical and equitable considerations and construe [this requirement] broadly in favor of proposed

EAST BAY SANCTUARY COVENANT V. BIDEN 23

intervenors.” Id. at 1179 (cleaned up). The States easily satisfy this permissive standard. The States have put forward a number of interests that would be impaired by an abandonment of the Rule. When immigrants enter the country illegally and settle in a state, that state has mandatory duties to provide certain government resources to them. See, e.g., Plyler v. Doe, 457 U.S. 202, 230 (1982) (public education); Gideon v. Wainwright, 372 U.S. 335, 345 (1963) (publicly funded counsel). And even where the States cannot provide certain resources to illegal immigrants because of state or federal law, they still incur increased administrative costs to determine if unlawfully present applicants are eligible. See, e.g., Kan. Stat. Ann. § 8-237(i); 8 U.S.C. § 1621(a), (c)(1). The States also assert a political interest in maintaining the Rule because an increase in other states’ populations from illegal immigration would give those states additional representatives, electoral votes, and federal funding, and necessarily take those from the states seeking intervention. See Dep’t of Commerce v. New York, 139 S. Ct. 2551, 2565 (2019). According to the government’s own data, there were 1.7 million border encounters between October 2023 and March 2024 alone. 4 Abandoning the Rule and increasing the influx of illegal immigrants therefore would have a sizeable and concrete impact on the States’ asserted interests. In spite of the permissive intervention standard, the majority concludes that these interests are not protectable. In doing so, the majority relies in part on United States v. Texas, 599 U.S. 670 (2023). But Texas stands for the

4 Customs and Border Protection, Nationwide Encounters, https://www.cbp.gov/newsroom/stats/nationwide-encounters. proposition that states do not have standing to sue to compel the Executive to “alter its arrest policies so as to make more arrests.” Id. at 686. In coming to this conclusion, the Supreme Court drew from a long line of cases concluding judicial review does not extend to refusals of executive enforcement. Id. at 680 (collecting sources for the proposition that “federal courts are generally not the proper forum for resolving claims that the Executive Branch should make more arrests or bring more prosecutions”). There is a clear difference between the executive’s nonenforcement of a rule and its effective complete removal of the rule altogether through collusive settlement.5 Consider a simplistic example involving speed limits. A police force could decide that it will no longer enforce the speed limit on motorists in certain circumstances. Under Texas, it seems a party might not have standing to judicially compel the police force to enforce the speed limit. On the other hand, this case is more akin to a city attorney colluding with a plaintiff to get rid of a lawfully enacted speed limit altogether through settlement, rather than doing so through

5 The Supreme Court in the Texas case itself recognized that states could have an interest in situations other than the exact enforcement issues presented in that case. For example, the Court suggested that standing might exist where “the Executive Branch wholly abandoned its statutory responsibilities to make arrests or bring prosecutions,” where “a challenge to an Executive Branch policy … involves both the Executive Branch’s arrest or prosecution priorities and the Executive Branch’s provision of legal benefits or legal status,” and where a challenged policy governs “the continued detention of noncitizens who have already been arrested.” 599 U.S. at 682–83. The situation here is, if anything, even more distinct than those examples provided in Texas, since those examples were all various twists on enforcement and arrests. Here, the States don’t assert any interest in enforcement—merely in the underlying validity of the Rule itself.

EAST BAY SANCTUARY COVENANT V. BIDEN 25

proper legislative or regulatory channels. This case is not about refusal to enforce a rule; this case is about whether the rule itself is valid or not. The plaintiffs in this case need to be careful what they ask for. Reading Texas to say that affected parties have no protectable interest in the continuing validity (or invalidity) of a rule would expand it so far that even the organizational plaintiffs in this case would not have standing. The majority concludes that the States’ increased expenditures related to the abandonment of the Rule do not impair a protectable interest. But increased expenditures due to the Rule is exactly the injury that supposedly gives the organizational plaintiffs standing to sue under our precedent. See E. Bay Sanctuary Covenant v. Biden, 993 F.3d 640, 664 (9th Cir. 2021). The majority also concludes that the effects that vacating the Rule might have on the States’ expenditures and political representation are “attenuated and speculative.” But the States have identified a number of concrete ways that the Rule protects their economic and political interests, as well as ways those interests would necessarily be impaired if the Rule were abandoned. See id. at 662–64 (relying on similar reasoning to explain the organizational plaintiffs’ stake in this litigation). As the government itself has recognized in this litigation, “increases in the number of migrants unlawfully present in the country” directly “strains [] government operations and resources.” That is no less true of the States than it is of the federal government. At the end of the day, the standard for intervention is a permissive one. “Constru[ing] the [intervention] Rule broadly in favor of [the] proposed intervenors,” Wilderness Soc’y, 630 F.3d at 1179 (cleaned up), the increased expenditures and political harms that the States will incur if the Rule is abandoned represent a sufficient impairment of the States’ protectable interests.

III. The States’ Interests Are Inadequately Represented

by the Government. Like the rest of the intervention rule, this factor too is supposed to be permissive. Indeed, “[t]he burden of showing inadequacy of representation is ‘minimal’ and satisfied if the applicant can demonstrate that representation of its interests ‘may be’ inadequate.” W. Watersheds, 22 F.4th at 840–41 (cleaned up). And like the rest of the factors, this lenient standard is easily satisfied here. The government argues that it adequately represents the States’ interest in maintaining the Rule because it “has repeatedly defended the Rule’s legality.” But that was before the government abruptly changed course and asked to put this case on ice while it considers settlement. From the beginning of this litigation until submitting the joint motion, the government has made clear that the Rule is critical to the ongoing administration of our border. 6 Consistent with this representation, the government vigorously defended the Rule in the district court and on appeal and has indicated an intent to petition the Supreme Court if we vacated the Rule. On the other hand, the organizational plaintiffs insist that the

6 As the government explained, “[w]ithout the Rule, the expected increase in border encounters threatened to overwhelm the Departments’ ‘ability to effectively process, detain, and remove, as appropriate, the migrants encountered,’ with attendant increases in the number of migrants unlawfully present in the country, strains on government operations and resources, health and safety concerns for migrants at overcrowded processing facilities, and impacts on local communities along the southwest border.”

EAST BAY SANCTUARY COVENANT V. BIDEN 27

Rule is contrary to law, and harms themselves and the immigrants they help. As I have previously observed, the plaintiffs’ position makes it difficult to imagine how they could accept any settlement less than rescission of the rule. See E. Bay Sanctuary Covenant v. Biden, 93 F.4th 1130, 1134 (9th Cir. 2024) (VanDyke, J., dissenting). Therefore, the very fact that the government is now considering such a settlement suggests that its position regarding this litigation has significantly changed. The States simply want to step in and continue the government’s previous litigation position. They have made clear that they oppose settlement for a number of compelling reasons, and so actively defending the Rule is the only way to continue protecting their interests. First, the States argue that the Rule is either lawful or it is not, and since this Court was poised to decide the issue before the parties filed a joint ruling, settlement now works only to unsettle the law. Second, since the factual basis for the plaintiffs’ standing is contested, “it would be absurd for the federal government to change federal policy based on litigation brought by parties who have no legally enforceable interest in the first place.” Finally, the government has represented elsewhere that the settlement discussions include “related policies,” and so a settlement could affect the interests of the States in other unforeseen ways. The government’s decision to enter into settlement negotiations rather than continue to actively defend the rule therefore represents a break with the States’ interests. And “[u]nless the [States] [are] made [parties] to these proceedings,” if the government settles this case no one will seek a final decision from this panel, “no petition for rehearing can be filed in this Court, and there will be no opportunity for the Supreme Court to consider whether to grant certiorari.” Day, 505 F.3d at 966. We have previously considered these circumstances “determinative.” Id. Given that it is no longer clear that the government will continue to defend the Rule, the government’s representation of the States’ interests—at the very least—“may be” inadequate. W. Watersheds, 22 F.4th at 840. That is all that is required for this factor to be satisfied. Id. *** The standard for intervention as of right is permissive and should be “construe[d] … broadly in favor of proposed intervenors.” Wilderness Soc’y, 630 F.3d at 1179 (cleaned up). Under this lenient standard, the States’ motion is timely, their significantly protectable interests may be impaired by a disposition of this case, and those interests are inadequately represented by the government. I therefore would grant their motion to intervene. Because the States are entitled to intervene as of right, I do not address the States’ request for permissive intervention.