Whitfield v. Mullin

FOR PUBLICATION FILED UNITED STATES COURT OF APPEALS OCT 16 2026 MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS FOR THE NINTH CIRCUIT DARYL WHITFIELD, No. 25-5220 D.C. No. Plaintiff - Appellant, 2:25-cv-04499-HDV-PD Central District of California, v. Los Angeles MARKWAYNE MULLIN, in his official ORDER capacity as Secretary of the Department of Homeland Security, Defendant - Appellee. Before: M. Margaret McKeown, N. Randy Smith, and Holly A. Thomas, Circuit Judges. Dissent by Judge Bress; Dissent by Judge VanDyke. Judge H.A. Thomas voted to deny the petition for rehearing en banc, and Judges McKeown and N.R. Smith have so recommended. The full court was advised of the petition for rehearing en banc. A judge requested a vote on whether to rehear the matter en banc. The matter failed to receive a majority of the votes of the non-recused active judges in favor of en banc consideration. See Fed. R. App. P. 40. The petition for rehearing en banc is DENIED.  FILED OCT 16 2026 Whitfield v. Mullin, 25-5220 MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS BRESS, Circuit Judge, joined by CALLAHAN, BENNETT, R. NELSON, BADE, COLLINS, LEE, BUMATAY, VANDYKE, and TUNG, Circuit Judges, dissenting from the denial of rehearing en banc: Under Supreme Court precedent, “[a] plaintiff seeking a preliminary injunction must establish [1] that he is likely to succeed on the merits, [2] that he is likely to suffer irreparable harm in the absence of preliminary relief, [3] that the balance of equities tips in his favor, and [4] that an injunction is in the public interest.” Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008). But Ninth Circuit precedent permits plaintiffs to obtain preliminary injunctions through an alternative showing, known as the “sliding scale test,” “which allows a strong showing on the balance of the hardships to compensate for a lesser showing of likelihood of success.” Where Do We Go Berkeley v. Cal. Dep’t of Transp., 32 F.4th 852, 859 (9th Cir. 2022) (citing All. for the Wild Rockies v. Cottrell, 632 F.3d 1127, 1134–35 (9th Cir. 2011)). Under that alternative test, “when plaintiffs establish that the balance of hardships tips sharply in their favor, there is a likelihood of irreparable injury, and the injunction is in the public interest, they need only show ‘serious questions’ on the merits.” Id. (quoting All. for the Wild Rockies, 632 F.3d at 1135). Our “serious questions” test for preliminary injunctions is inconsistent with Supreme Court precedent, especially after Starbucks Corp. v. McKinney, 602 U.S. 339 (2024). There is also a clear split of authority in the lower courts on the permissibility of this alternative test, and we are on the wrong side of that split. The issue is important, as the “serious questions” test affords district courts too much latitude to grant preliminary injunctions. This case presented a good opportunity to revisit our “serious questions” test and bring circuit precedent in line with Supreme Court precedent. I respectfully dissent from the decision to deny rehearing en banc. I I draw the relevant facts from the panel’s decision. See Whitfield v. Mullin, 2026 WL 1230550 (9th Cir. May 5, 2026). The plaintiff, Daryl Whitfield, is an import specialist at U.S. Customs and Border Protection (CBP). Whitfield suffers from various disabilities related to his prior military service. Before 2020, Whitfield worked in the office one day each week. From March 2020 to April 2022, CBP suspended in-office work for import specialists, in light of the COVID-19 pandemic. In April 2022, CBP resumed requiring import specialists to be in-office for one day each week. In March 2025, the agency required import specialists to return to in- person work full time. In March 2022, before CBP began requiring import specialists to be physically present in the office one day each week, Whitfield submitted requests seeking an accommodation to work remotely full time. He later filed additional accommodation requests and administrative complaints. In May 2025, Whitfield, acting pro se, filed this lawsuit alleging, inter alia, that CBP failed to provide him 2 reasonable accommodations or engage in the required interactive process, in violation of the Rehabilitation Act. Whitfield also filed a motion for a preliminary injunction seeking to prevent CBP from requiring him to return to in-person work. The district court denied the motion for a preliminary injunction. It applied the usual four-factor Winter preliminary injunction test, but it also regarded Whitfield’s request as seeking a “mandatory injunction,” so that Whitfield was required to show that the law and facts “clearly favor” his position. The district court found that Whitfield did not make the required showing of likelihood of success on the merits given CBP’s efforts to accommodate Whitfield’s disabilities, which included offering to return him to his previous schedule of working one day in the office per week and granting him other accommodations. Whitfield appealed pro se, and the panel reversed. It held that the district court erred in “appl[ying] an incorrect preliminary injunction standard.” Whitfield, 2026 WL 1230550, at *1. According to the panel, the district court should not have applied the standard for mandatory injunctions, which concerns requests for the defendant to take action and go beyond maintaining the status quo. Id. at *2. This was so because in March 2022, when Whitfield began challenging CBP’s in-office policies, the status quo was fully remote work. Id. Accordingly, “Whitfield’s request is one for an ordinary prohibitory injunction.” Id. 3  The panel instructed the district court on the proper standard to apply on remand, invoking our alternative “sliding scale” approach. The panel explained that “[w]hen evaluating the Winter factors, we employ a ‘sliding scale test,’” which “means that ‘when plaintiffs establish that the balance of hardships tips sharply in their favor,’ and the other factors are met, plaintiffs ‘need only show serious questions on the merits.’” Id. (first quoting All. for the Rockies, 632 F.3d at 1134– 35, and then quoting Where Do We Go Berkeley, 32 F.4th at 859). The panel thus advised that “under our ordinary sliding-scale approach for prohibitory injunctions, Whitfield needed only establish that a ‘serious question’ existed as to whether he was denied a reasonable accommodation under the Rehabilitation Act if the balance of hardships tips ‘sharply’ in his favor.” Id. (quoting Where Do We Go Berkeley, 32 F.4th at 859). The panel did “not decide whether the balance of hardships tips sharply in Whitfield’s favor or if Whitfield has successfully established a ‘serious question’ as to the merits of his reasonable accommodation claim.” Id. It instead remanded for further consideration under what the panel described as “the Winter sliding-scale prohibitory injunction standard.” Id. The United States filed a petition for rehearing en banc. The petition did not take issue with the panel’s determination that the district court improperly applied the heightened standard for mandatory injunctions. Instead, the United States asked us to reevaluate circuit precedent on the “serious questions” sliding scale test, as set 4 forth in the panel’s instructions on remand. The panel appointed pro bono counsel to represent Whitfield in connection with the government’s petition for rehearing en banc. II Our alternative “serious questions” test for preliminary injunctions is inconsistent with Supreme Court precedent, conflicts with the precedent of other circuits, and grants district courts too much leeway to issue preliminary injunctions. The proper standard for obtaining preliminary injunctive relief—an extraordinary remedy—is a recurring and important issue that merited en banc review. Prior to Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7 (2008), Ninth Circuit precedent offered “two alternatives” for obtaining a preliminary injunction, which were said to “represent extremes of a single continuum.” Clear Channel Outdoor Inc. v. City of Los Angeles, 340 F.3d 810, 813 (9th Cir. 2003) (quoting Walczak v. EPL Prolong, Inc., 198 F.3d 725, 731 (9th Cir. 1999)). A plaintiff could demonstrate “either: (1) a likelihood of success on the merits and the possibility of irreparable injury; or (2) that serious questions going to the merits were raised and the balance of hardships tips sharply in its favor.” Id. (quoting Walczak, 198 F.3d at 731). In Winter, the Supreme Court rejected the first of these formulations. It held that “the Ninth Circuit’s ‘possibility’ standard is too lenient,” because the Supreme 5 Court’s “frequently reiterated standard requires plaintiffs seeking preliminary relief to demonstrate that irreparable injury is likely in the absence of an injunction.” 555 U.S. at 22. A “possibility” of irreparable harm standard, the Supreme Court went on, was further “inconsistent with [its] characterization of injunctive relief as an extraordinary remedy that may only be awarded upon a clear showing that the plaintiff is entitled to such relief.” Id. Winter thus reaffirmed that “[a] plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.” Id. at 20. After Winter, the question arose whether our “serious questions” on the merits test was still good law. We held that it was. See All. for the Wild Rockies, 632 F.3d at 1131–32 (“[W]e hold that the ‘serious questions’ approach survives Winter when applied as part of the four-element Winter test.”). That determination was incorrect even then given Winter’s plain rejection of a modified preliminary injunction test, as well as its reaffirmation that “[a] plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits.” Winter, 555 U.S. at 20. Although some circuits agreed with us that Winter did not abrogate the “serious questions” test, see Citigroup Global Mkts., Inc. v. VCG Special Opportunities Master Fund, Ltd., 598 F.3d 30, 35–38 (2d Cir. 2010); Hoosier 6 Energy Rural Elec. Co-op., Inc. v. John Hancock Life Ins. Co., 582 F.3d 721, 725 (7th Cir. 2009), others have rejected “serious questions” in light of Winter, see The Real Truth About Obama, Inc. v. F.E.C., 575 F.3d 342, 346–47 (4th Cir. 2009), vacated on other grounds, 559 U.S. 1089 (2010) (“The Winter requirement that the plaintiff clearly demonstrate that it will likely succeed on the merits is far stricter than the . . . requirement that the plaintiff demonstrate only a grave or serious question for litigation.”); Diné Citizens Against Ruining Our Env’t v. Jewell, 839 F.3d 1276, 1282 (10th Cir. 2016) (“Although the Ninth Circuit standard overruled in Winter dealt with a different prong of the preliminary injunction test than the prong at issue in this case, Winter’s rationale seems to apply with equal force here. . . . Under Winter’s rationale, any modified test which relaxes one of the prongs for preliminary relief and thus deviates from the standard test is impermissible.”); see also Clevinger v. Advocacy Holdings, Inc., 134 F.4th 1230, 1235–36 (D.C. Cir. 2025) (“It is questionable that the sliding scale approach remains good law after 2008, when the Supreme Court decided Winter . . . .”); Davis v. Pension Ben. Guar. Corp., 571 F.3d 1288, 1296 (D.C. Cir. 2009) (Kavanaugh, J., concurring) (“[U]nder the Supreme Court’s precedents, a movant cannot obtain a preliminary injunction without showing both a likelihood of success and a likelihood of irreparable harm.”); Sierra Club v. U.S. Army Corps of Eng’rs, 645 F.3d 978, 993–94 (8th Cir. 2011) (noting that a district court that employed a 7 “serious questions” test had used “imprecise” language following Winter). The split of authority post-Winter is clear. But if there were any doubt about Winter’s import, the Supreme Court’s later decision in Starbucks Corp. v. McKinney, 602 U.S. 339 (2024), erased it. Starbucks concerned a Sixth Circuit decision affirming the entry of a preliminary injunction under Section 10(j) of the National Labor Relations Act. Under Sixth Circuit precedent, the NLRB could secure such an injunction if “‘there is reasonable cause to believe that unfair labor practices have occurred,’” and “injunctive relief is ‘just and proper.’” Id. at 344 (quoting McKinney v. Ozburn-Hessey Logistics, LLC, 875 F.3d 333, 339 (6th Cir. 2017)). Further, “[t]he Board could establish reasonable cause by simply showing that its ‘legal theory [was] substantial and not frivolous.’” Id. (quoting McKinney, 875 F.3d at 339). The Supreme Court held that the Sixth Circuit’s standard for § 10(j) preliminary injunctions was contrary to Winter and therefore improper. As the Supreme Court explained, “[w]hen Congress empowers courts to grant equitable relief, there is a strong presumption that courts will exercise that authority in a manner consistent with traditional principles of equity,” and “[f]or preliminary injunctions, the four criteria identified in Winter encompass the relevant equitable principles.” Id. at 345. It followed that the Sixth Circuit’s test was incorrect, because 8 a “reasonable-cause” standard “substantively lowers the bar for securing a preliminary injunction.” Id. at 349. Our “serious questions” test also substantively lowers the bar for securing a preliminary injunction. Indeed, we have specifically described the “serious questions” test as involving “a lesser showing of likelihood of success.” Where Do We Go Berkeley, 32 F.4th at 859; see also, e.g., Assurance Wireless USA, L.P. v. Reynolds, 100 F.4th 1024, 1031 (9th Cir. 2024) (describing “serious questions” as a “‘less demanding’ merits standard”) (quoting All. for the Wild Rockies, 68 F.4th at 497). Starbucks thus confirms that our “serious questions” test is invalid. See Starbucks, 602 U.S. at 350 (“There is an obvious difference between having the Board show that it is ‘likely’ to succeed on the merits and having it show only that its theory of the case is ‘substantial and not frivolous,’ without having to convince a court that its theory is likely meritorious.”); see also Nken v. Holder, 556 U.S. 418, 438 (2009) (Kennedy, J., concurring) (observing in the related context of stay motions that “[w]hen considering success on the merits and irreparable harm, courts cannot dispense with the required showing of one simply because there is a strong likelihood of the other”). Our “serious questions” test is not only at odds with Supreme Court precedent, it also impedes the even-handed application of the law. A matter so important as whether to grant a preliminary injunction, an “extraordinary remedy,” Winter, 555 9 U.S. at 24, should be guided by one clear standard, not two different ones. That is especially so when “serious questions” is a more open-ended standard that lacks the heritage of the traditional “likelihood of success” test. See Starbucks, 602 U.S. at 348 (explaining that the Winter framework embodies “traditional” equitable principles). Because what counts as a “serious question” is less determinate and more open to interpretation, the “serious questions” test broadens the circumstances in which district courts may issue preliminary injunctions. This case demonstrates the point. The district court determined that Whitfield could not show a likelihood of success on the merits because CBP had offered him various accommodations, including returning him to his same pre-pandemic schedule of working one day per week in the office. The panel’s remand implies that a “serious questions” analysis may be a difference-maker. But it is unclear how that analysis should proceed, except that the district court, by whatever metric it evaluates the “seriousness” of a question, can now grant preliminary injunctive relief in the absence of a likelihood of success on the merits. And although this case does not arise in an emergency posture, the “serious questions” test is of particular concern given the influx of emergency appeals we are seeing from the grant of preliminary injunctions. Whitfield’s counseled response to the government’s petition did not attempt to defend our “serious questions” test on the merits. Instead, Whitfield argues that 10 this case is not a good vehicle for addressing the issue. Contrary to Whitfield’s opposition, this case presented a perfectly suitable vehicle for considering the conflict between Ninth Circuit precedent and the Supreme Court’s decisions in Winter and Starbucks. Although the panel concluded that the district court erred in applying a heightened standard for mandatory injunctions, the “serious questions” test is fully implicated here because the panel’s decision discussed it at length in explaining what test the district court should have applied, which explains why the panel directed the district court to apply the “serious questions” test on remand. If our sliding scale approach is overruled as inconsistent with Supreme Court precedent, the instructions to the district court on remand would need to be different, i.e., limited to the traditional Winter factors. Indeed, given the district court’s conclusion that Whitfield “cannot show a likelihood of success on the merits,” if the “serious questions” test were overruled, it may be that the panel should simply affirm the district court by treating its reference to mandatory injunctions as harmless error. Either way, the “serious questions” test is front and center in this case, given the panel decision. Further, although the propriety of the “serious questions” test did not arise until the petition for rehearing en banc, that is because the district court did not invoke that test, the government was defending the district court’s injunction on 11 other grounds, and because it was the panel’s decision that first focused on the “serious questions” test. The government thus properly raised the issue once it became a feature of this case, namely, once the panel affirmatively invoked it and directed the district court to apply it on remand. Finally, that the panel decided this case in an unpublished disposition did not make the case unworthy of en banc review, as the panel’s decision was premised on long-established circuit precedent. * * * Our “serious questions” test for preliminary injunctions is inconsistent with Supreme Court precedent and the precedent of other circuits. We should have reevaluated this important issue. I respectfully dissent from the court’s decision to deny rehearing en banc. 12  FILED OCT 16 2026 Whitfield v. Mullin, 25-5220 MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS VANDYKE, Circuit Judge, joined by BUMATAY and TUNG, Circuit Judges, dissenting from the denial of rehearing en banc: I agree with, and am pleased to join, Judge Bress’s excellent dissental. I write separately to underscore that this case illustrates (yet again) a more fundamental problem within our circuit: Miller v. Gammie’s improper demand that our own circuit’s precedent be privileged over the Supreme Court’s more recent instructions unless our precedent is “clearly irreconcilable” with what the Supreme Court has subsequently told us. 335 F.3d 889, 893 (9th Cir. 2003). This prevents three-judge panels of our court from applying more recent Supreme Court authority whenever it is merely inconsistent with our own. See In re Manion, No. 25-2507, 2026 WL 2641687, at *3 (9th Cir. Sept. 8, 2026) (VanDyke, J., concurring in part); see also Arizona All. for Retired Ams. v. Mayes, 186 F.4th 622, 652–56 (9th Cir. 2026) (VanDyke, J., concurring in part). The effect of this rule is both obvious and obviously wrong—we systematically hold the Supreme Court’s instructions about the right answer to any number of legal questions in lower regard than our own wrong precedents. As Judge Bress correctly explains, our circuit’s “serious questions” test is inconsistent with Starbucks Corp. v. McKinney, 602 U.S. 339 (2024), which recently disavowed the Sixth Circuit’s fabricated standard for obtaining a preliminary injunction under Section 10(j) of the National Labor Relations Act because it 1 substantively lowered the bar for obtaining relief. In theory, a three-judge panel of our court might have fixed our similarly errant test to conform to the Supreme Court’s more recent instruction. Yet, because it is not “clearly irreconcilable,” Miller would prevent a three-judge panel from harmonizing our precedent with Starbucks here—or indeed, intervening Supreme Court authority in nearly any case, unless the Supreme Court issues a decision that is basically a dead-ringer for our own controlling precedent. Arizona All., 186 F.4th at 653 (VanDyke, J., concurring in part). And predictably, a majority of our court refuses to fix it. So once again, the Supreme Court must do what a properly functioning inferior-court judicial system should have done itself: correct one of our precedents that has been undermined by subsequent Supreme Court authority. And make no mistake, this problem extends far beyond the “serious questions” test. As I’ve explained elsewhere, Miller ossifies erroneous Ninth Circuit precedent across doctrinal areas, forcing litigants to seek en banc or Supreme Court review to obtain the benefit of intervening Supreme Court decisions. See id. at 653– 56. The reality is that our en banc court is often unable or unwilling to fix things at odds with Supreme Court precedent. Sometimes, the obstacle is simply institutional inertia; the costs of en banc review make correction burdensome enough that erroneous precedent persists by default. See In re Manion, 2026 WL 2641687, 2 at *4. We are a very busy court, and we can’t be expected to take every case en banc that presents a legal issue where subsequent Supreme Court authority is inconsistent with, but not “clearly irreconcilable” with, our own. But that is not the only problem with Miller’s rule. The effect of that inverted rule is especially insidious (and no doubt intentional) when our precedent involves one of the many contested legal questions where our circuit’s ideological commitments are at odds with what the Supreme Court has recently told us. In those instances, it is very unlikely a majority of our court would ever vote to take a case en banc to abandon the majority’s own preferred doctrinal approach, even after the Supreme Court has undermined its foundation. See, e.g., Knife Rts., Inc. v. Bonta, 182 F.4th 797, 802–10 (9th Cir. 2026) (VanDyke, J., dissenting) (explaining how the Ninth Circuit ignores the Supreme Court’s Second Amendment precedent). Judge Reinhardt famously announced the Ninth Circuit’s motto: the Supreme Court can’t “catch ’em all.” See, e.g., Linda Greenhouse, Dissenting Against the Supreme Court's Rightward Shift, N.Y. Times (Apr. 12, 2018), https://www.nytimes.com/2018/04/12/opinion/supreme-court-right-shift.html [https://perma.cc/H473-3MRD]. Miller’s strictures ensure that the Supreme Court’s guidance will, for the most part, be limited in this circuit to only those instances where the Court actually “catches” a specific case—that is, where it actually grants review and issues a binding opinion in that case. And even in that rare instance, 3 Miller ensures that whatever guidance the Supreme Court provides will have little spillover effect beyond the very narrow issue actually decided by the Court in that case. See, e.g., Wolford v. Lopez, 116 F.4th 959, 970 (9th Cir. 2024), rev’d and remanded, 609 U.S. 185 (2026); Wolford v. Lopez, 184 F.4th 1167, 1170–71 (9th Cir. 2026) (accepting, reluctantly, the Supreme Court’s direct holding that a Hawaii law “prohibit[ing] the carry of firearms onto private property unless the owner allows it” violates the Second Amendment, but refusing to re-evaluate any other aspect of the panel’s previous analysis upholding other Hawaii arms restrictions in light of Wolford). Everything else is not “clearly irreconcilable” (judges and lawyers alike are, after all, pretty good at distinguishing), so three-judge panels of this court that would very much like to follow the Supreme Court’s recent guidance can’t. And those who would very much like to avoid the Supreme Court’s guidance can do so simply by not voting to take a case en banc. The consequence is perverse. To correct our circuit’s many erroneous precedents and ensure litigants can benefit from intervening Supreme Court decisions, the Supreme Court will have to take case after case after case and reverse them all, slowly, over the course of decades, one by one by one, as it has done so many times before. See, e.g., United States v. Hansen, 25 F.4th 1103, 1105 (9th Cir. 2022) (misconstruing a federal law prohibiting “encourag[ing] or induc[ing]” illegal immigration in order to find it unconstitutionally overbroad under the First 4 Amendment), rev’d, 599 U.S. 762 (2023); United States v. Sineneng-Smith, 910 F.3d 461 (9th Cir. 2018) (injecting sua sponte into an appeal an overbreadth challenge to this same statute), vacated and remanded 590 U.S. 371, 375 (2020) (holding that the Sineneng-Smith panel had “departed so drastically from the principle of party presentation as to constitute an abuse of discretion” and remanding “for an adjudication of the appeal attuned to the case shaped by the parties rather than the case designed by the appeals panel”); Martin v. City of Boise, 920 F.3d 584, 603 (9th Cir. 2019) (holding that the Eighth Amendment’s Cruel and Unusual Punishments Clause barred a city “from prosecuting people ... for sleeping outside on public property when those people have no home or other shelter to go to”), abrogated by City of Grants Pass v. Johnson, 603 U.S. 520 (2024); Martinez v. City of Oxnard, 270 F.3d 852, 857–58 (9th Cir. 2001) (holding that mere “coercive questioning” violated the Fifth and Fourteenth Amendments even when the defendant’s statements were not used against him in a criminal proceeding), rev’d sub nom. Chavez v. Martinez, 538 U.S. 760 (2003); Compassion in Dying v. Washington, 79 F.3d 790, 816 (9th Cir. 1996) (en banc) (holding that the “Constitution encompasses a due process liberty interest in controlling the time and manner of one’s death—that there is, in short, a constitutionally recognized ‘right to die’”), rev’d sub nom. Washington v. Glucksberg, 521 U.S. 702, 772 (1997). 5  The Supreme Court should not have to keep trimming the same poison ivy one tiny branch at a time. There are many judges on our circuit who would be glad to do the right thing and apply the Supreme Court’s guidance in three-judge panel decisions whenever that guidance is inconsistent with our circuit’s own earlier precedents. By overturning Miller in this or some future case, the Supreme Court would remove a perverse rule that entrenches erroneous precedent in this circuit, enabling three-judge panels to help hack away at the overgrowth. Help us help you. The Supreme Court should take up this case to correct our circuit’s erroneous “serious questions” rule. But while it’s at it, why not also correct Miller’s “clearly irreconcilable” standard? Addressing the former will eliminate one of many legal doctrines in our court that are inconsistent with subsequent Supreme Court authority. But also addressing the latter would do much more to conform the Ninth Circuit’s law to the Supreme Court’s—without the Supreme Court’s needing to exhaustively address every single error seriatim. 6