Congress and Universal Injunctions

Introduction

Lower federal courts continue to issue universal (or non-particularized) injunctions1E.g., HIAS, Inc. v. Trump, 985 F.3d 309, 326–27 (4th Cir. 2021); New York v. U.S. Dep’t of Homeland Sec., 969 F.3d 42, 87–88 (2d Cir. 2020), cert. dismissed, 141 S. Ct. 1292 (2021); City of Chicago v. Barr, 961 F.3d 882, 911–931 (7th Cir. 2020); E. Bay Sanctuary Covenant v. Trump, 950 F.3d 1242, 1260 (9th Cir. 2020), amended and superseded on denial of reh’g en banc, 993 F.3d 640 (9th Cir. 2021); Rodgers v. Bryant, 942 F.3d 451, 457–58 (8th Cir. 2019); Regents of the Univ. of Cal. v. U.S. Dep’t of Homeland Sec., 908 F.3d 476, 511–12 (9th Cir. 2018), rev’d in part, vacated in part, 140 S. Ct. 1891 (2020); Texas v. United States, 809 F.3d 134 (5th Cir. 2015), aff’d by an equally divided Court, 136 S. Ct. 2271 (2016); Wynn v. Vilsack, No. 21-cv-514-MMH-JRK, 2021 WL 2580678, at *17 (M.D. Fla. June 23, 2021); Louisiana v. Biden, No. 21-CV-00778, 2021 WL 2446010 (W.D. La. June 15, 2021); Alabama Ass’n of Realtors v. U.S. Dep’t of Health & Hum. Servs., No. 20-cv-3377, 2021 WL 1779282, at *9 (D.D.C. May 5, 2021), stay granted pending appeal, 2021 WL 1946376 (D.D.C. May 14, 2021); Texas v. United States, No. 21-cv-00003, 2021 WL 247877, at *6–7 (S.D. Tex. Jan. 26, 2021); Danville Christian Acad. v. Beshear, 503 F. Supp. 3d 516, 531 (E.D. Ky.), stay granted on other grounds, 981 F.3d 505 (6th Cir. 2020), denying motion to vacate stay, 141 S. Ct. 527 (2020); Pennsylvania v. Trump, 351 F. Supp. 3d 791, 830–35 (E.D. Pa. 2019), rev’d sub nom. Little Sisters of the Poor Saints Peter and Paul Home v. Pennsylvania, 140 S. Ct. 2367 (2020); Batalla Vidal v. Nielsen, 279 F. Supp. 3d 401, 437–38 (E.D.N.Y. 2018), aff’d in part, rev’d in part, vacated in part sub nom. Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891 (2020); Cnty. of Santa Clara v. Trump, 250 F. Supp. 3d 497, 539–40 (N.D. Cal. 2017).—injunctions prohibiting government defendants from enforcing challenged laws against all rights-holders who might be enforcement targets, whether parties or non-parties to the litigation. Many scholars support these efforts2E.g., Rachel Bayefsky, Remedies and Respect: Rethinking the Role of Federal Judicial Relief, Geo. L.J. (forthcoming); Zachary D. Clopton, National Injunctions and Preclusion, 118 Mich. L. Rev. 1, 5–7 (2019); Amanda Frost, In Defense of Nationwide Injunctions, 93 N.Y.U. L. Rev. 1065, 1069 (2018); Suzette M. Malveaux, Class Actions, Civil Rights, and the National Injunction, 131 Harv. L. Rev. F. 56 (2017); Bradford Mank & Michael E. Solimine, State Standing and National Injunctions, 94 Notre Dame L. Rev. 1955, 1973 (2019); Portia Pedro, Toward Establishing a Pre-Extinction Definition of “Nationwide Injunctions,” 91 U. Colo. L. Rev. 847 (2020); James E. Pfander & Jacob P. Wentzel, The Common Law Origins of Ex Parte Young, 72 Stan. L. Rev. 1269, 1353–54 (2020); Mila Sohoni, The Power to Vacate a Rule, 88 Geo. Wash. L. Rev. 1121, 1124–27 (2020) [hereinafter Sohoni, Vacate]; Mila Sohoni, The Lost History of the “Universal” Injunction, 133 Harv. L. Rev. 920, 924 (2020) [hereinafter Sohoni, History]; Alan M. Trammell, Demystifying Nationwide Injunctions, 98 Tex. L. Rev. 67, 73–74 (2019). in the face of scholarly3Douglas Laycock, Modern American Remedies 436 (4th ed. 2010); Samuel L. Bray, Multiple Chancellors: Reforming the National Injunction, 131 Harv. L. Rev. 417, 469 (2017); Ronald A. Cass, Nationwide Injunctions’ Governance Problems: Forum Shopping, Politicizing Courts, and Eroding Constitutional Structure, 27 Geo. Mason L. Rev. 29 (2019); John Harrison, Section 706 of the Administrative Procedure Act Does Not Call for Universal Injunctions or Other Universal Remedies, 37 Yale J. Regul. Bull. 37 (2020); Michael T. Morley, Disaggregating Nationwide Injunctions, 71 Ala. L. Rev. 1, 7–8 (2019) [hereinafter Morley, Disaggregating]; Michael T. Morley, Nationwide Injunctions, Rule 23(b)(2), and the Remedial Powers of the Lower Courts, 97 B.U. L. Rev. 615, 620 (2017) [hereinafter Morley, Nationwide]; Jonathan Remy Nash, State Standing for Nationwide Injunctions Against the Federal Government, 94 Notre Dame L. Rev. 1985, 2012 (2019); Howard M. Wasserman, Precedent, Non-Universal Injunctions, and Judicial Departmentalism: A Model of Constitutional Adjudication, 23 Lewis & Clark L. Rev. 1077, 1080–81, 1093–94 (2020) [hereinafter Wasserman, Departmentalism]; Howard M. Wasserman, Concepts, Not Nomenclature: Universal Injunctions, Declaratory Judgments, Opinions, and Precedent, 91 U. Colo. L. Rev. 999, 1000 (2020) [hereinafter Wasserman, Concepts]; Howard M. Wasserman, “Nationwide” Injunctions Are Really “Universal” Injunctions and They Are Never Appropriate, 22 Lewis & Clark L. Rev. 335, 353 (2018) [hereinafter Wasserman, “Nationwide”]. and judicial4Dep’t of Homeland Sec. v. New York, 140 S. Ct. 599, 600 (2020) (Gorsuch, J., joined by Thomas, J., concurring in the grant of stay); Trump v. Hawaii, 138 S. Ct. 2392, 2424–25 (2018) (Thomas, J., concurring); City of Chicago, 961 F.3d at 936–38 (Manion, J., concurring in the judgment); Rodgers, 942 F.3d at 460 (Stras, J., concurring in part and dissenting in part); E. Bay Sanctuary Covenant v. Barr, 934 F.3d 1026, 1028–30 (9th Cir. 2019); California v. Health & Hum. Servs., 351 F. Supp. 3d 1267, 1300 (N.D. Cal. 2019), vacated by 977 F.3d 801 (9th Cir. 2020). criticism.

I have argued in multiple articles that federal courts lack the power or the need to issue universal/non-particularized injunctions and never should do so.5Wasserman, Departmentalism, supra note 3, at 1080–81, 1093–94; Wasserman, Concepts, supra note 3, at 1000; Wasserman, “Nationwide,” supra note 3, at 353; see sources cited supra note 3. But controlling precedent in many circuits accepts them, at least in identified circumstances.6See, e.g., E. Bay Sanctuary Covenant v. Biden, 993 F.3d 640, 680–81 (9th Cir. 2021); HIAS, 985 F.3d at 309, 326–27; City of Chicago, 961 F.3d at 911–931; Texas, 2021 WL 247877, at *6–7. The Supreme Court has avoided the issue,7Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891, 1916 n.7 (2020). apart from scattered views of individual Justices.8Compare New York, 140 S. Ct. at 600 (2020) (Gorsuch J., joined by Thomas, J., concurring in grant of stay), and Hawaii, 138 S. Ct. at 2425 n.1 (2018) (Thomas, J., concurring), with Little Sisters of the Poor Saint Peter and Paul Home v. Pennsylvania, 140 S. Ct. 2367, 2412 n.28 (2020) (Ginsburg, J., dissenting), and Hawaii, 138 S. Ct. at 2446 (Sotomayor, J., dissenting). The Court’s October Term 2020 calendar included several cases in which lower courts had issued broad injunctions against Trump Administration policies,9See, e.g., Sierra Club v. Trump, 977 F.3d 853 (9th Cir. 2020); Innovation L. Lab v. Wolf, 951 F.3d 1073 (9th Cir.), vacated and remanded with instructions to dismiss as moot by Mayorkas v. Innovation L. Lab, No. 19-1212, 2021 WL 2520313 (U.S. June 21, 2021) (mem.). but the new Biden Administration rescinded the challenged policies, mooting the appeals.10Biden v. Sierra Club, No. 20-138, 2021 WL 2742775 (U.S. July 2, 2021) (mem.), vacating and remanding Sierra Club v. Trump, 963 F.3d 874 (9th Cir. 2020); Mayorkas, 2021 WL 2520313; Amy Howe, Biden administration asks justices to take immigration cases off February calendar, SCOTUSBlog (Feb. 1, 2021), https://www.scotusblog.com/‌2021/‌02/‌biden-administration-asks-justices-to-take-immigration-cases-off-february-calendar [https://perma.cc/‌4LFH-ESED]..

Halting the judicial practice of issuing universal/non-particularized injunctions thus requires congressional action. As a default, courts control their equitable powers to grant injunctive relief.11Armstrong v. Exceptional Child Ctr., Inc., 575 U.S. 320, 327–28 (2015). But Congress may limit or alter the equitable power, so long as it provides a clear statement of its intent to change or alter courts’ remedial authority.12Id. at 327–28; Weinberger v. Romero-Barcelo, 456 U.S. 305, 313 (1982); Hecht Co. v. Bowles, 321 U.S. 321, 330 (1944); William N. Eskridge, Jr., Public Values in Statutory Interpretation, 137 U. Pa. L. Rev. 1007, 1023 (1989).

The political valence of universal injunctions has shifted—from conservative plaintiffs seeking to universally enjoin enforcement of Obama Administration policies13See, e.g., Nevada v. U.S. Dep’t of Lab., 275 F. Supp. 3d 795 (E.D. Tex. 2017) (Department of Labor overtime regulations); Nevada v. U.S. Dep’t of Lab., 218 F. Supp. 3d 520 (E.D. Tex. 2016) (same); Texas v. United States, 809 F.3d 134 (5th Cir. 2015), aff’d by evenly divided Court, 136 S. Ct. 2271 (2016). to liberal plaintiffs seeking to universally enjoin enforcement of Trump Administration policies14See, e.g., HIAS, Inc. v. Trump, 985 F.3d 309, 326–27 (4th Cir. 2021); New York v. U.S. Dep’t of Homeland Sec., 969 F.3d 42, 87–88 (2d Cir. 2020), cert. dismissed, 141 S. Ct. 1292 (2021); City of Chicago v. Barr, 961 F.3d 882, 911–931 (7th Cir. 2020); E. Bay Sanctuary Covenant v. Trump, 950 F.3d 1242, 1260 (9th Cir. 2020), amended and superseded on denial of reh’g en banc, 993 F.3d 640 (9th Cir. 2021). back to conservative plaintiffs seeking to universally enjoin enforcement of Biden Administration policies.15 See, e.g., Wynn v. Vilsack, No. 21-cv-514-MMH-JRK, 2021 WL 2580678, at *17 (M.D. Fla. June 23, 2021); Louisiana v. Biden, No. 21-CV-00778, 2021 WL 2446010, at *22 (W.D. La. June 15, 2021); Alabama Ass’n of Realtors v. U.S. Dep’t of Health & Hum. Servs., No. 20-cv-3377, 2021 WL 1779282, at *9 (D.D.C. May 5, 2021), stay granted pending appeal, 2021 WL 1946376 (D.D.C. May 14, 2021); Texas v. United States, No. 21-cv-00003, 2021 WL 247877, at *6–7 (S.D. Tex. Jan. 26, 2021).. Republican-controlled congressional committees convened two hearings to criticize the practice during the Trump Administration,16 Rule by District Judge: The Challenges of Universal Injunctions: Hearing Before the S. Comm. on the Judiciary, 116th Cong. (2020); The Role and Impact of Nationwide Injunctions by District Courts: Hearing Before the Subcomm. on Cts., Intell. Prop., and the Internet of the H. Comm. on the Judiciary, 115th Cong. (2017). but conservative activists discovered their benefits with the coming of the Biden presidency.17 See generally Scott A. Keller, Nationwide Injunctions Will Be a Vital Check if Biden Overreaches, Wall St. J. (Nov. 22, 2020), https://www.wsj.com/‌articles/‌nationwide-injunctions-will-be-a-vital-check-if-biden-overreaches-11606081224 [https://perma.cc/‌J837-NQJJ]. Democrats appeared more receptive to universal injunctions when the injunctions targeted objectionable Republican policies and less inclined to strip courts of all power to issue them.18Rule by District Judge: The Challenges of Universal Injunctions: Hearing Before the S. Comm. on the Judiciary, 116th Cong. (2020) (opening statement of Sen. Feinstein, Ranking Member); The Role and Impact of Nationwide Injunctions by District Courts: Hearing Before the Subcomm. on Cts., Intell. Prop., and the Internet of the H. Comm. on the Judiciary, 115th Cong. (2017) (statement of Rep. Nadler, Ranking Member).

Members of Congress and commentators have offered numerous concrete legislative approaches to stopping or limiting universal/non-particularized injunctions. One proposed approach is appropriate, a valid exercise of legislative power, and an actual solution to the actual problem—a flat and unequivocal prohibition on injunctions protecting or binding anyone other than parties to litigation.19 H.R. 77, 116th Cong. (2019); H.R. 6730, 115th Cong. (2018); infra Part I. Other proposals fail by misunderstanding or ignoring the real problems, repeating the mistakes that have plagued courts and scholars in this debate.20 Infra Parts II–V. Because of that misunderstanding, these proffered approaches fail—as a matter of substance and as a matter of nomenclature—to resolve the scope-of-remedy debate.

Three preliminary points set up the discussion that follows. First, if Article III’s case-or-controversy requirement prohibits universal/non-particularized injunctions,21Bray, supra note 3, at 421, 471; Cass, supra note 3, at 58–59; Harrison, supra note 3; Morley, Disaggregating, supra note 3, at 14; Michael T. Morley, De Facto Class Actions? Plaintiff- and Defendant-Oriented Injunctions in Voting Rights, Election Law, and Other Constitutional Cases, 39 Harv. J.L. & Pub. Pol’y 487, 516 (2016) [hereinafter Morley, De Facto Class Actions?]. these legislative efforts are superfluous, announcing that courts cannot do something they are not constitutionally empowered to do. A legislative pronouncement that courts lack this power perhaps offers an expressive benefit22See Cass R. Sunstein, On the Expressive Function of Law, 144 U. Pa. L. Rev. 2021, 2022 (1996). but no practical consequence. If the limitation on universal/non-particularized injunctions derives from sub-constitutional equitable and remedial principles subject to congressional adjustment,23Wasserman, Departmentalism, supra note 3, at 1094; Wasserman, “Nationwide,” supra note 3, at 354–55. these laws serve an important clarifying and limiting purpose.

Second, an injunction must accord “complete relief” to the plaintiff and should be commensurate with and match the constitutional violation, while being no more burdensome to the defendant than necessary. It must ensure that the plaintiff is not subject to or threatened with present or future enforcement of the constitutionally invalid law.24Lewis v. Casey, 518 U.S. 343, 357 (1996); Califano v. Yamasaki, 442 U.S. 682, 702 (1979); Hills v. Gautreaux, 425 U.S. 284, 293–94 (1976); Wasserman, Departmentalism, supra note 3, at 1094.

The third issue concerns nomenclature—what to call the injunctions that Congress is attempting to regulate or eliminate. The scope of an injunction entails two features—its “who” (what persons are bound or protected by the injunction) and its “where” (the geographic locales in which they enjoy that protection).25 Wasserman, “Nationwide,” supra note 3, at 349. The prevailing term “nationwide injunction”26Texas v. United States, No. 21-cv-00003, 2021 WL 247877, at *6 n.5 (S.D. Tex. Jan. 26, 2021). describes the “where” of geography, rather than the “who” of protected parties. In fact, to accord the plaintiff the required complete relief, all injunctions should be nationwide—they should protect the plaintiff throughout the nation by prohibiting enforcement of the challenged law against her anywhere in the nation she might go.

The scholarly, judicial, and legislative debate is over injunctions with an overbroad “who.” “Universal injunction” better describes an order protecting the universe of potential targets of enforcement of the challenged law, whether parties to the litigation or not.27Wasserman, Concepts, supra note 3, at 1006; Wasserman, “Nationwide,” supra note 3, at 350; see also Dep’t of Homeland Sec. v. New York, 140 S. Ct. 599, 600 (2020) (Gorsuch J., joined by Thomas, J., concurring in the grant of stay); Trump v. Hawaii, 138 S. Ct. 2392, 2425 n.1 (2018) (Thomas, J., concurring); City of Chicago v. Barr, 961 F.3d 882, 913–14 n.7 (7th Cir. 2020). A related term is “non-particularized,” which captures injunctions protecting beyond the parties—injunctions not “particularized” to the parties—whether covering the universe of enforcement targets or a set of targets smaller than the universe but larger than the plaintiff.28Wasserman, Concepts, supra note 3, at 1007.

With those points, this essay explores five legislative proposals for eliminating or limiting universal/non-particularized injunctions—one of which resolves the problem and four of which fail for various reasons. This analysis can inform Congress about the true problem and can guide future legislative efforts. With the political valence of the debate and of arguments for and against these injunctions having flipped with the 2021 change in presidential administrations and congressional control,29 See supra notes 14–16. the time for legislative action has arrived.

I. Injunctions Limited to Parties

Two House bills from 2018 and 2019, titled the Injunctive Authority Clarification Act,30H.R. 77, 116th Cong. (2019); H.R. 6730, 115th Cong. (2018). present the single appropriate way to halt overbroad injunctions. Both prohibit any federal court from issuing any order “that purports to restrain the enforcement against a non-party of any statute, regulation, order, or similar authority.”31H.R. 77, 116th Cong. (2019); H.R. 6730, 115th Cong. (2018). Both include an exception where the “non-party is represented by a party acting in a representative capacity pursuant to the Federal Rules of Civil Procedure.”32H.R. 77, 116th Cong. (2019); H.R. 6730, 115th Cong. (2018).

These bills identify and address the real problem of universal/non-particularized injunctions—that they purport to prohibit government officials from enforcing laws not only against the parties to the action, but against non-parties who should not otherwise be protected by the court’s order. And they offer the appropriate solution—a flat prohibition on universal/non-particularized injunctions and on courts attempting to protect strangers to the litigation or to limit future government enforcement against those strangers to the litigation. These bills establish by statute the state of affairs that judicial and scholarly critics of  universal injunctions have urged—injunctions protecting the plaintiffs against future enforcement of the challenged law, but protecting no one else.33See supra notes 3–4.

The exception for class actions preserves the essential role of civil rights injunctive class actions under Rule 23(b)(2).34Fed. R. Civ. P. 23(b)(2); see Maureen Carroll, Class Action Myopia, 65 Duke L.J. 843, 857–58 (2016); David Marcus, Flawed But Noble: Desegregation Litigation and Its Implications for the Modern Class Action, 63 Fla. L. Rev. 657, 680–81 (2011). A class-wide injunction remains particularized and non-universal—it protects and benefits only the plaintiff. But the plaintiff is the class of rights-holders, the class having assumed an identity and legal status independent of the representative individual plaintiff.35United States v. Sanchez-Gomez, 138 S. Ct. 1532, 1538 (2018); Sosna v. Iowa, 419 U.S. 393, 399 (1975). Rule 23 expands the scope of the injunction by expanding who is a party before the court and therefore who is and may be protected by an appropriate non-universal/particularized injunction.36Sosna, 419 U.S. at 399; Rodgers v. Bryant, 942 F.3d 451, 464 (8th Cir. 2019) (Stras, J., concurring in part and dissenting in part); Morley, De Facto Class Actions?, supra note 21, at 541; Morley, Disaggregating, supra note 3, at 17–19.

II. Injunctions Limited to Parties and District

Three bills, two bearing the not-subtle title of “Nationwide Injunction Abuse Prevention Act,” offer a second solution.37H.R. 4927, 115th Cong. (2018); H.R. 4292, 116th Cong. (2019); S. 2464, 116th Cong. (2019). They employ identical language to prohibit district courts from issuing any order for injunctive relief except when limited to the parties to the action or to the federal judicial district in which the order was issued.38H.R. 4927, 115th Cong. (2018); H.R. 4292, 116th Cong. (2019); S. 2464, 116th Cong. (2019).

This approach gets it half-right but half-wrong. Like the appropriate approach described in Part I, these bills recognize that the injunction’s “who” must protect the parties to the action, but should not extend to strangers to the litigation.

But they fail by conflating the injunction’s “who” with its “where”—conflating universality in who is protected with nationwide as to where they are protected. A geographically cramped injunction means a plaintiff who convinces the court that enforcement of the challenged law against her violates her rights remains subject to enforcement of that law in a different place. An injunction prohibiting enforcement of a federal law issued by a district court in California would not protect the plaintiff from enforcement of the same federal law in Arizona. An injunction prohibiting enforcement of an Arkansas anti-loitering law issued by a court in the Eastern District of Arkansas would not protect the plaintiff from enforcement of that Arkansas law in a part of the state falling within the Western District of Arkansas. Alternatively, a plaintiff must file suit and obtain a unique injunction in every federal district within which she acts and in which she might be subject to enforcement of the challenged federal or state law.

Either way, these bills strip federal courts of too much remedial authority. They deprive courts of power to accord geographically complete relief that is commensurate with and matching of the constitutional violation, at least where the plaintiff operates in multiple places and outside “neat geographic boundaries.”39See E. Bay Sanctuary Covenant v. Trump, 950 F.3d 1242, 1282–83 (9th Cir. 2020) (quoting E. Bay Sanctuary Covenant v. Trump, 354 F. Supp. 3d 1094, 1120 (N.D. Cal. 2018)), amended and superseded on denial of reh’g en banc, 993 F.3d 640 (9th Cir. 2021).

Litigation over Trump Administration asylum regulations demonstrates the mischief of the second piece of these proposals and the potential conflation of their parts. Four advocacy organizations sued to stop enforcement of regulations stripping asylum eligibility from anyone who entered the United States other than at a point of entry.40E. Bay Sanctuary, 950 F.3d at 1260. The Northern District of California enjoined all enforcement of the regulations, in and out of the Ninth Circuit; the Ninth Circuit stayed the injunction’s extra-circuit reach pending appeal, while leaving the injunction unstayed within the circuit as to all potential enforcement targets, beyond the four plaintiff entities.41E. Bay Sanctuary Covenant v. Barr, 934 F.3d 1026, 1028–29 (9th Cir. 2019). The Supreme Court stayed the injunction in full pending review. Barr v. E. Bay Sanctuary Covenant, 140 S. Ct. 3 (2019).. The temporary relief was over- and under-protective. It was over-protective by protecting the non-particularized universe of every potential enforcement target within the Ninth Circuit. It was under-protective by failing to protect the named plaintiffs outside the Ninth Circuit, where they operated and remained subject to enforcement of the challenged regulations.

The Ninth Circuit corrected the remedial problem when it considered the injunction on the merits later in the litigation. Affirming a preliminary injunction protecting the plaintiff organizations (but not non-parties) outside their home states, the court explained that the plaintiffs’ operations were not limited to “neat geographic boundaries,” so an injunction limited to California or to the Ninth Circuit would not protect them from losing clients in Texas as a result of the regulation.42E. Bay Sanctuary, 950 F.3d at 1282–83 (quoting E. Bay Sanctuary, 354 F. Supp. 3d at 1120). The Supreme Court granted, then dismissed, certiorari. Wilkinson v. City & Cnty. of San Francisco, 141 S. Ct. 1292 (2021) (mem.).

Unfortunately, these bills compel the problematic initial outcome and prevent the appropriate later outcome in East Bay. In fact, they compound the problem by requiring the court to limit the injunction not to the eleven jurisdictions comprising the Ninth Circuit or to the State of California, but only to the Northern District of California.

The bills also contain a linguistic glitch that perhaps confuses their effects. They prohibit injunctions except where the order applies only to (1) the parties “or” (2) in the federal district that issued the injunction.43H.R. 4927, 115th Cong. (2018); H.R. 4292, 116th Cong. (2019); S. 2464, 116th Cong. (2019). “Or” suggests that the clauses are disjunctive—that only one statutory condition must be satisfied.

Thus, an injunction limited to the parties (satisfying (1)) need not be limited to the districts (not satisfying (2)). If so, the bills leave us where we should be—prohibiting universal/non-particularized injunctions that protect strangers to the litigation but permitting nationwide particularized injunctions that protect the named plaintiff wherever she is or goes.

But the disjunctive reading permits the converse—an injunction limited to the federal district (satisfying (2)) need not be limited to the parties (not satisfying (1)). That is, the proposals permit injunctions that are overly non-particularized but insufficiently nationwide. That result gets everything backwards—insufficient relief to the plaintiff, excessive relief to strangers to the litigation. A universal-but-not-nationwide injunction provides improper in-district relief to non-parties who should not be within the injunction’s umbrella, while failing to accord necessary out-of-district relief to the parties whom the injunction is supposed to protect.

III. Immediate Supreme Court Review

A third approach leaves untouched whatever power federal courts might possess to issue what the bill calls nationwide injunctions but provides that “appeal from the order granting such injunction shall lie to the Supreme Court.”44H.R. 4219, 116th Cong. (2019). The word “appeal” indicates that review is mandatory, rather than discretionary.45See 28 U.S.C. § 1253. Although it uses the term nationwide injunction, the statutory definition matches what I label a universal/non-particularized injunction that protects non-parties.

This proposal leaves district judges free to issue broad injunctions but limits their time in existence through mandatory and speedy reversal or affirmance by the Court of last resort. It also addresses the criticism that plaintiffs forum shop in seeking universal relief,46Bray, supra note 3, at 457–60; Cass, supra note 3, at 42–44; Gregg Costa, An Old Solution to the Nationwide Injunction Problem, Harv. L. Rev. Blog (Jan. 25, 2018), https://blog.harvardlawreview.org/‌an-old-solution-to-the-nationwide-injunction-problem [https://perma.cc/‌Q47P-5XNF]; Wasserman, “Nationwide,” supra note 3, at 363–64. by running to a presumptively favorable court (conservative plaintiffs to the Fifth Circuit, liberal plaintiffs to the Ninth Circuit). Immediate and mandatory Supreme Court review neutralizes any geographic or ideological gamesmanship in the plaintiff’s initial choice of forum.

But the proposal ignores the foundational point that no federal court should issue universal/non-particularized injunctions.47See supra notes 3, 5, 19–26. The problem is not district courts issuing overbroad relief; the problem is any federal court at any point in the judicial hierarchy issuing relief that expands and exceeds its remedial powers. The Supreme Court possesses no more power to affirm a universal/non-particularized injunction than the district court possesses to issue that universal/non-particularized injunction. Quick and mandatory review produces a speedier Article III final judgment from the Supreme Court, but that Article III final judgment must remain particularized to the parties.

This bill fails to address the true universality/non-particularity problem in three further respects. First, it conflates judgments that resolve a specific legal action between adverse parties and opinions that explain the reasons for the judgment.48Wasserman, Departmentalism, supra note 3, at 1104–05, 1107–08; Wasserman, Concepts, supra note 3, at 1017, 1021–22. Direct and mandatory review speeds the issuance of a Supreme Court opinion, establishing binding precedent and binding resolution of constitutional questions regardless of the scope of the judgment the district court issues or the Supreme Court affirms. But the creation of Supreme Court precedent through the opinion obviates the need for universality/non-particularity in the judgment. If the Supreme Court must take the case and must resolve the constitutional question, every case will “finally” resolve the issue for all rights-holders, party and non-party, through binding judicial precedent. If binding precedent affecting future enforcement by all government officials as to all rights-holders emerges from every case, there is no need to make the judgment and injunction universal, even for a limited period.

Second, limiting speedier and mandatory Supreme Court review to cases of universal injunctions creates unintended perverse incentives. Plaintiffs (especially ideologically motivated parties and counsel) will seek, and district courts will grant, more universal/non-particularized injunctions, recognizing that it guarantees Supreme Court review and resolution. Given a choice between ordinary three-step particularized litigation and a certain, immediate answer through universal litigation, litigants and lower courts will choose the latter. Litigants want final resolution of the question; lower courts want to flex the power of issuing one all-controlling order, if for a limited period.

Finally, this proposal may not achieve what advocates hope. The bill presumes a particular partisan posture—a conservative administration enacting conservative policies challenged by liberal plaintiffs, universally enjoined by liberal lower-court judges appointed by past presidents. The solution is a conservative Supreme Court reviewing and (presumably) reversing those judgments, smacking down over-reaching lower-court judges, and enabling enforcement of the conservative policies. And that is, in the barest sense, the 2019 milieu in which the bill’s Republican sponsors operated. Notwithstanding President Trump’s success in filling the federal courts,49Howard M. Wasserman, Academic Feeder Judges, 105 Judicature 60, 71 (2021). however, that milieu is not permanent. At some point, a Democratic administration and Democratic-controlled Congress will create and enforce liberal policies that conservative plaintiffs may wish to universally enjoin; they may not want mandatory review by an ideologically different Supreme Court.

IV. Three-Judge District Courts with Immediate Review

Fifth Circuit Judge Gregg Costa imports the entire three-judge district court system—three-judge courts5028 U.S.C. § 2284. with mandatory direct Supreme Court review5128 U.S.C. § 1253.—for all requests for universal/non-particularized injunctions.52Costa, supra note 46; Mank & Solimine, supra note 2, at 1980–81. He adds a further wrinkle—any action seeking a universal/non-particularized injunction would be randomly assigned to a regional circuit whose chief judge would select the panel from circuit and district judges within the circuit.53Costa, supra note 46. Under the current system, a request for a three-judge court is referred to the chief judge of the circuit embracing the district in which an action is filed. 28 U.S.C. § 2284(b)(1). Although never reduced to legislation, codifying Costa’s proposal would be easy, as the current three-judge statute anticipates applying that process whenever required by a new act of Congress.5428 U.S.C. § 2284(a).

Costa’s proposal employs the logic that motivated Congress to create three-judge district courts for challenges to state legislation in 1910 and to extend them to challenges to federal legislation in 1937.55See 28 U.S.C. § 2281 (repealed 1976); 28 U.S.C. § 2282 (repealed 1976); Costa, supra note 46; Mank & Solimine, supra note 2, at 1980; Michael T. Morley, Vertical Stare Decisis and Three-Judge District Courts, 108 Geo. L.J. 699, 728–31, 734–36 (2020); Michael E. Solimine, Congress, Ex Parte Young, and the Fate of the Three-Judge District Court, 70 U. Pitt. L. Rev. 101, 116–18 (2006). One district judge should not prevent a state or the United States from enforcing its laws, and a panel of judges is more likely to arrive at the correct constitutional answer than a single trial judge.56 Costa, supra note 46. A universal/non-particularized decision from that single district judge—preventing all enforcement of the challenged law by all government officials against all rights-holders and all potential enforcement targets in all places—magnifies those concerns.

Like the proposal for immediate Supreme Court review discussed in Part III,57Supra Part III. Costa’s proposal emphasizes and resolves the forum-shopping problem.58Bray, supra note 3, at 457–60; Cass, supra note 3, at 42–44; Costa, supra note 46; Wasserman, “Nationwide,” supra note 3, at 363–64; supra note 46 and accompanying text. Plaintiffs cannot forum-shop to an ideologically sympathetic single-division district seeking universal relief from the lone active judge.59Costa, supra note 46. Plaintiffs cannot shop for a presumptively favorable circuit because the place of filing does not dictate the place of adjudication. Randomizing the circuit, the appointing chief judge, and the roster of judges from which the three-judge panel can be drawn eliminates any influence the plaintiff might exercise in choosing any judge on the panel.

But Costa’s proposal fails for the same reason that immediate review fails: no federal court can protect non-party strangers to litigation or enjoin defendants as to non-party strangers to the litigation, regardless of number of judges, composition, location, or position in the judicial hierarchy. A three-judge district court’s injunctive power is no broader than a single-judge district court’s injunctive power. The Supreme Court possesses no greater power to affirm universal/non-particularized injunctions from a three-judge district court than from a single-judge district court.60 Supra note 47 and accompanying text.

Costa’s approach also produces the same perverse incentives to pursue a universal remedy in all cases. Every plaintiff will request a universal/non-particularized injunction, guaranteeing a three-judge panel (which, as during the Civil Rights Era, might benefit plaintiffs61Solimine, supra note 55, at 126–29.) and speedy and mandatory Supreme Court review. Three-judge district courts will return to being routine in constitutional litigation, requiring more three-judge panels each year and increasing the burdens on lower-court judges and Supreme Court dockets. This recreates the docket burdens that compelled the Justices to resist62Morley, supra note 55, at 744; Solimine, supra note 55, at 140. prior three-judge statutes and Congress to eliminate routine three-judge courts in 1976.63Act of Aug. 12, 1976, Pub. L. No. 94-381, 90 Stat. 1119 (1976); Morley, supra note 55, at 744; Solimine, supra note 55, at 134.

V. Mandatory Litigation in the District of Columbia

A final option—discussed but never drafted—places all actions seeking universal/non-particularized injunctions in the District of the District of Columbia, with review to the D.C. Circuit.64 Mank & Solimine, supra note 2, at 1978–80; The Role and Impact of Nationwide Injunctions by District Courts: Hearing Before the Subcomm. on Cts., Intell. Prop., and the Internet, H. Comm. on the Judiciary, 115th Cong. (2017) (statement of Hans von Spakovsky, Heritage Found.). The United States is “in” D.C., the logic goes, so cases seeking this extraordinary relief should go exclusively to the United States’ home district.65Mank & Solimine, supra note 2, at 1978–80; The Role and Impact of Nationwide Injunctions by District Courts: Hearing Before the Subcomm. on Cts., Intell. Prop., and the Internet, H. Comm. on the Judiciary, 115th Cong. (2017) (statement of Hans von Spakovsky, Heritage Found.). Like Costa’s three-judge proposal and the immediate-review proposal,66Costa, supra note 46; supra Parts III–IV. this addresses the forum-shopping criticism by funneling remedially identical cases to one district and one regional circuit. It strips plaintiffs of the opportunity and temptation to shop for a favorable district with a favorable judge within a favorable circuit.

It also should be obvious that identical objections apply to this proposal as to the prior proposals. Federal judges in the District of Columbia have no greater power to issue or affirm universal/non-particularized injunctions than federal judges in any district anywhere in the United States.

This proposal introduces three new problems. First, Bradford Mank and Michael Solimine argue that it injects a dose of partisanship. Routing cases to the District of Columbia may increase the partisan tensions of constitutional litigation, given the political nature of cases in which states challenge the validity of federal law67Mank & Solimine, supra note 2, at 1975–76. and the increasing polarization over appointments to the D.C. Circuit as incubator for Supreme Court Justices.68Id. at 1979. Depending on the political valence of a case,69Id. at 1975–76. D.C.-based judges become more likely to issue universal/non-particularized injunctions, the opposite of the plan’s intent.

Second, this proposal presumes that universal/non-particularized injunctions arise only in challenges to federal law. But injunctions barring enforcement of a state law are non-particularized/universal if they protect beyond the plaintiff and prohibit enforcement of that state law against all potential enforcement targets, including non-parties.70Supra notes 21–28 and accompanying text. The “universe” of enforcement targets is smaller as to state law, but the problem of a non-particularized order protecting strangers to the litigation remains.71Wasserman, Concepts, supra note 3, at 1008. Courts and judges have discussed the problem of universality in actions challenging the validity of state laws.72In re Abbott, 954 F.3d 772, 786 n.19 (5th Cir. 2020), vacated, Planned Parenthood Ctr. for Choice v. Abbott, 141 S. Ct. 1261 (2021) (mem.); Jones v. Governor of Fla., 950 F.3d 795, 800 (11th Cir. 2020); Rodgers v. Bryant, 942 F.3d 451, 460 (8th Cir. 2019) (Stras, J., concurring in part and dissenting in part).

That the federal government is “home” in D.C. is irrelevant to challenges to the constitutional validity of state laws. States are not “home” in D.C., and federal judges in D.C. have no experience or expertise with the laws of those states. Congress might limit the proposal to universal injunctions against enforcement of federal law, while leaving challenges to enforcement of state laws, even where a universal/non-particularized injunction is sought, to ordinary forum-selection rules. But familiarity with the specifics of a state and its laws may be important in resolving challenges to federal law, especially where a state brings the action.73Mank & Solimine, supra note 2, at 1968–69; Nash, supra note 3, at 1990–91; see, e.g., Trump v. Hawaii, 138 S. Ct. 2392 (2018); New York v. U.S. Dep’t of Homeland Sec., 969 F.3d 42, 87–88 (2d Cir. 2020), cert. dismissed, 141 S. Ct. 1292 (2021); Regents of the Univ. of Cal. v. U.S. Dep’t of Homeland Sec., 908 F.3d 476, 511–12 (9th Cir. 2018), rev’d in part, vacated in part, 140 S. Ct. 1891 (2020); Texas v. United States, 809 F.3d 134 (5th Cir. 2015), aff’d by evenly divided Court, 136 S. Ct. 2271 (2016); Louisiana v. Biden, No. 21-CV-00778, 2021 WL 2446010 (W.D. La. June 25, 2021) (mem.); Texas v. United States, No. 21-cv-00003, 2021 WL 247877, at *6–7 (S.D. Tex. Jan. 26, 2021). This proposal forces D.C. judges to understand state law and the unique circumstances of many states that underlie and affect the constitutional analysis.74Mank & Solimine, supra note 2, at 1979. Federal judges in D.C. possess no such expertise.

Finally, the proposal does not address what happens if the D.C. court decides that universal/non-particularized relief is inappropriate, but some non-universal/particularized relief is appropriate. That is, the court decides that the challenged law is constitutionally invalid and an injunction prohibiting its enforcement should issue, but the injunction should be particularized to the plaintiffs. The rationale for placing the case in a D.C. court has evaporated, so there is no reason for that court to hear and decide the case. Having recognized that a universal injunction is inappropriate, the D.C. court must dismiss or transfer the action to a federal court for the state where the plaintiffs originate, a wasteful process. And a case may have multiple plaintiffs hailing from multiple states, leaving no obvious transferee district.

Alternatively, the D.C. court might issue the universal/non-particularized injunction, even where unnecessary, to justify exercising jurisdiction. As with the three-judge and immediate-review proposals, a plan to limit universal/non-particularized injunctions perversely incentivizes plaintiffs to seek and courts to issue more of them.

Conclusion

Any legislative response to the scope-of-remedy dispute must address the real problem—court orders protecting non-party strangers to the litigation. Legislators must not be distracted by side issues and misnomers, which lead to different legislation that addresses different problems, that fails to resolve the real problem, and that perhaps worsens the remedial situation.

Congress could, and should, resolve the judicial and scholarly debate through simple legislation imposing a flat and unequivocal prohibition on universal/non-particularized injunctions that protect anyone other than the plaintiffs, wherever those plaintiffs go in the nation. That approach—the simplest one—represents the only solution.