Solutions to Nationwide Injunctions

WHO, WHAT, AND WHERE: A CASE FOR A MULTIFACTOR BALANCING TEST AS A SOLUTION TO ABUSE OF NATIONWIDE INJUNCTIONS

I. History of Equitable Remedies

The very ability of courts to issue nationwide injunctions is due to the nature of equitable remedies themselves. Therefore, to make an argument about how the judiciary’s equitable powers ought to change, it is important to understand something of their origin and how they have evolved over time. The fact that the discretion accorded to judges in equity is a fundamental feature has strong implications for what type of solutions are available to modern problems in equity.

A.     In England

Equity has long afforded substantial discretion to judges. The first reason for this is historical. The Court of Chancery in England started out as a court of conscience, to which parties could turn to when the common law courts failed to give them justice: “[t]he Office of the Chancellor is . . . to soften and mollify the Extremity of the Law.”[1] The king was seen as the fount of all justice, and the office of chancellor evolved to assist the king in administering the “king’s justice.”[2] The fact that early chancellors were religious figures cemented the role of conscience over form, and caused ecclesiastical law to have a strong influence on equity.[3] The freezing of additional common law writs with the Provisions of Oxford in 1258 exacerbated law’s more formalistic tendencies, which made common law courts useless in solving certain new legal problems, particularly cases surrounding use of real property.[4]  Accordingly, the chancellor was accorded a great deal of discretion in addressing these novel problems, and in ordering the appropriate remedy.

Indeed,
the chancellor’s discretion was so absolute that it prompted harsh criticism,
perhaps none more well-known than John Selden’s witty remark that:

Equity is a Roguish thing, for Law we have a measure . . . . Equity is according to the conscience of him that is Chancellor, and as that is larger or narrower so is equity. “Tis all one, as if they should make his foot the standard for the measure we call a Chancellor’s foot; what an uncertain measure would this be! One Chancellor has a long foot, another a short foot, a third an indifferent foot.”[5]

The view that equity existed to make up for the common law courts’ rigidity and formalism existed not only in the 17th century, but was echoed by Blackstone over one hundred years later. Paraphrasing Aristotle, Blackstone wrote:

For since in laws all cases cannot be foreseen or expressed, it is necessary, that when the general decrees of the law come to be applied to particular cases, there should be somewhere a power vested of excepting those circumstances, which (had they been foreseen) the legislator himself would have excepted.[6]

Even as the chancellor ceased to be seen as a religious figure and took on a more secular role, equity’s role in correcting injustices that resulted from the inadequacies of common law courts continued.[7] In sum, regardless of the reason¾philosophy, history, or simple necessity¾discretion became an intrinsic feature of equity.

B.     The Founding and the Role of the Judiciary

Over
the years, questions about the role of the judiciary in controlling legislation
and enforcing the constitutional order have abounded.[8] The
accusation that Justice Marshall invented judicial review has been widely
circulated.[9]
Notwithstanding the widespread traction this idea has gained, it is false.
Professor Randy Barnett convincingly made this argument from an originalist
perspective, showing that both proponents and opponents of the Constitution
understood it to give the courts this power.[10]
Most important for our purposes, this power included the power of judicial
nullification of legislation.[11]
While I mention this part of the history only in passing, it is important to
the question of nationwide injunctions because in constitutional cases, it
shows that the courts are to play a role in being intermediaries between the
people and the other two branches of the government, not merely the servants of
Congress.

II. Comparing the Benefits and Harms of Nationwide Injunctions

Part of what makes the debate over nationwide injunctions so complex is that there are powerful values on both sides of the debate, any of which could be outcome determinative in a given case. This section examines both the benefits and harms of nationwide injunctions.

A.     Benefits

• Rule of Law and Uniformity

Perhaps
the strongest reason for the proliferation of nationwide injunctions is a
simple, moral intuition that it is wrong to allow illegal, especially
unconstitutional, conduct from continuing unchecked. While nationwide
injunctions are a relatively new phenomenon,[12]
the beginning of their logic stems all the way to Marbury[13]
v. Madison.  At least since Marbury, it has been firmly established
that the courts have the power to strike down laws. It is only a small step
from the concept of judicial review to the nationwide injunction, because if it
is the job of the courts to declare what the law is, any view which contradicts
a judicial opinion is, in a sense, not the law. This is particularly true in
constitutional matters because the constitution is the supreme law of the land,
rendering all statutes that conflict with it moot.[14]
To allow an agency to continue enforcing a “moot” law violates this basic principal.

Indeed,
in the first judicial decision to directly address concerns about the propriety
of nationwide injunctions, Judge Leinenweber identified precisely this concern:
“The rule of law is undermined where a court holds that the Attorney General is
likely engaging in legally unauthorized conduct, but nevertheless allows that
conduct in other jurisdictions across the country.”[15]
This persuasive argument in favor of nationwide injunctions also justifies
their increasingly widespread use. Since the federal government generally
enacts policies and statutes uniformly around the country, if one of those
policies or statues is unconstitutional, the harm is likewise nationwide. If
the rule of law is harmed by allowing any similarly situated citizen to be
harmed, nationwide injunctions will become the norm.[16]

• Egalitarian Concerns

By
protecting affected individuals not party to the litigation, nationwide
injunctions also promote equality. In fact, Wirtz
v. Baldor Electric Company, the case that first issued a nationwide injunction,
advanced the idea that all similarly situated plaintiffs deserve the same
outcome from the law counseled toward issuing a broad injunction.[17]
“[Where] a lower court … has spoken, that court would ordinarily give the
same relief to any individual who comes to it with an essentially similar cause
of action.”[18]
These concerns about equality are exacerbated when similarly situated
plaintiffs have different access to legal resources.[19]

• Judicial Economy

Lastly,
nationwide injunctions help preserve judicial resources. While a slower, more
incremental approach may allow for “percolation” of the best ideas on the
subject,[20]
nationwide injunctions decide the issue immediately for the whole nation,
preventing duplicative litigation and reducing the courts’ dockets.[21]
Given widespread concern about the sometimes exorbitant costs of litigation and
costs to society of paying for a judicial system, it may be most beneficial to
have some legal issues decided once and for all.

B.     Harms

• Forum Shopping

Accusations
of forum shopping for judges who are ideologically inclined to the plaintiff or
plaintiffs have, with the increased use of broad equitable remedies, become
commonplace and well publicized. They also cross political boundaries. Most
recently, many conservatives cried foul when judges issued nationwide
injunctions against the Trump administration’s so-called “Muslim ban.”[22] It
probably did not help that two of these cases were issued by judges in the 9^th
Circuit, which has a reputation for being one of the more liberal circuits in
the country.[23]
Similarly, multiple pieces of notable environmental litigation against the Bush
Administration was filed in the Ninth Circuit.[24]

Possibly
the most obvious case of forum shopping occurred in United States v. Texas, where Texas and other states sued the Obama
Administration over its immigration program, “Deferred Action for Parents of
Americans and Lawful Permanent Residents” (DAPA), which granted legal status to
millions of undocumented immigrants.[25] The
case was brought in the Southern District of Texas in the Brownsville division
which had only two active judges.[26]
One of those judges was Andrew Hanen who was known to be conservative and had
publicly criticized the Obama Administration’s immigration policies.[27]
Judge Andrew Hanen ended up getting the case.[28] Forum
shopping runs the risk of painting the judicial process as just another
instance of political gamesmanship, endangering respect for the judicial
process as a whole.

• Asymmetric Effects of Injunctions

An
issue similar to yet distinct from forum shopping is asymmetric issue
preclusion. If a challenge is brought to a statute and it is upheld, that
decision is directly binding only on the parties to that case. If instead the
statute is found unconstitutional and a nationwide injunction is issued, the
federal government is enjoined from enforcing that statute anywhere.[29] A
sobering example of this asymmetry is illustrated in United States v. AMC Entertainment, which involved interpretation
of the Americans with Disabilities Act.[30]
While the Fifth Circuit held for AMC, a district court in California ruled
against AMC, issuing an injunction that was nationwide in scope and conflicted
with the Fifth Circuit’s decision.  While
the 9th Circuit later narrowed the injunction to exempt the 5th Circuit because
comity concerns, the results asymmetrically benefitted the plaintiffs. Even
though AMC won once and lost once, it had to comply with the 9th Circuit’s
injunction in eleven out of twelve circuits. While that case involved a private
party defendant, it illustrates the same problems the federal government faces
as a defendant.[31]
This asymmetry exacerbates the potential for forum shopping because even if the
initial plaintiff does not forum shop and the statute is upheld, future
plaintiffs” need only find a single ideologically aligned judge to defeat the
government.

• Conflicting Injunctions

The
possibility that judges will issue conflicting injunctions is a doomsday
scenario which has never taken place, but the previously mentioned case United States v. Texas offers some
valuable lessons in the dangers of nationwide injunctions. After 5th Circuit
affirmed the district court, the death of Justice Scalia caused the Supreme
Court to split 4-4, affirming the 5^th Circuit’s decision. Many
plaintiffs attempted to circumvent the District of Texas’ injunction by
challenging its scope, while one district judge in New York even signaled his
willingness to disregard the district court in Texas’ injunction.[32]  While Scalia’s seat has since been filled,
this case showed a possible doomsday scenario: two circuits issue conflicting
decisions, while the Supreme Court deadlocks 4-4 on the constitutionality of
the statute.[33]
This could leave a plaintiff in a bind, mandated by one judge to follow a statute,
but prohibited by another from doing the same.

No
such doomsday scenario has occurred recently,[34]
since most conflicts are resolved by one court backing down,[35]
the case being reversed by the circuit court,[36] or
one court’s decision to exclude from injunctions circuits which have upheld
challenged policies.[37]
However, it is just common sense conflicting injunctions are much more likely
if judges issue more injunctions with broad scopes. Since nationwide
injunctions are a relatively recent and rapidly proliferating phenomenon,[38]
it is likely that more conflicts will occur.

• Damages the Supreme Court’s Supervisory Position

One
widely hailed benefit of the circuit court system is that it allows for what
Judge Leventhal famously called a “value in percolation among the circuits.”[39]
Under this account, the ability of multiple circuits to review novel issues of
law and fashion different solutions to them provides at least two advantages.
First, it helps filter out the truly difficult cases, which would benefit most
from Supreme Court review. If an issue seems difficult at first but every
circuit to review it comes to the same conclusion, the likelihood that it was
decided correctly are high. If not, than the Supreme Court can and should
exercise appellate review to resolve the circuit split. Second, percolation
allows for more judges to lend their voices to the discussion, increasing the
diversity of viewpoints presented, which may cause the final version of
arguments on both sides to be presented in their most compelling forms. In
Justice Ginsburg’s words, “[W]hen frontier legal problems are presented,
periods of ‘percolation’ in, and diverse opinions from, state and federal
appellate courts may yield a better informed and more enduring final pronouncement
by this Court.”[40]

Nationwide
injunctions undercut both of these features of our system. First, they prevent
the system from screening meritorious issues as effectively. A nationwide
injunction may increase or decrease the chance of certiori being granted. If the case is relatively mundane, the
chances of cert. being granted are
likely decreased since no circuit split is able to develop, keeping the case
off the radar.[41]
If the case is high-profile, as was the case with DAPA litigation, a nationwide
injunction may have the opposite effect, forcing the Supreme Court to grant cert. as soon as possible to resolve the
resulting crisis. In both cases, however, the filtering process fails, forcing
the Supreme Court to take cases it might not have needed to take, or preventing
important legal issues from garnering the attention they deserve. When cases
are forced through the system more quickly with no other circuits having a
chance to review the issue, the Supreme Court looses out on some of the insight
it might have gained from those circuit judges and legal scholarship that would
otherwise have time to be written.

• Lack of Remedial Uniformity

One of the primary articulated advantages
of nationwide injunctions is that they promote uniformity among the courts.[42]
Ironically, however, the use of nationwide injunctions has left the law of
equity itself in a state of disarray. Contrast, for example, two quotes from
the same circuit. Judge Posner wrote in 2011 “[w]hen the court believes the
underlying right to be highly significant, it may write injunctive relief as
broad as the right itself.”[43] However,
another 7th Circuit decision that Posner himself signed onto stated the
opposite conclusion: “A wrong done to plaintiff in the past does not authorize
prospective, classwide relief unless a class has been certified. Why else
bother with class actions?”[44]
Nor is this confusion limited to the relationship between class actions and
broad injunctive relief. Courts are split on whether the Administrative
Procedure Act calls for nationwide injunctions,[45]
and whether o not facial challenges are more deserving of nationwide
injunctions.[46]
As a result, judges have virtually complete discretion whether or not to issue
an injunction, and can cite authority supporting any decision about the scope
of equitable relief.

• Conflicts with Other Doctrine

Part
of the appeal of nationwide injunctions is it appears egalitarian¾a
narrow injunction protects only the plaintiff who happened to initiate the
litigation and had the resources to do so. It may seem unfair to allow the
government to enforce an unconstitutional in other jurisdictions when, with the
stroke of a pen, the district court judge can right all the nation’s wrongs.
However reasonable, this conflicts with two widely accepted legal rules: the
rule that non-mutual issue preclusion does not apply to the federal government,
and intercircuit agency non-acquiescence.

The
first of these rules was endorsed unanimously by the Supreme Court in United States v. Mendoza, which held
that the federal government was not subject to non-mutual issue preclusion.[47]
This means that the federal government is free to re-litigate issues that it
previously lost.[48]
The doctrine of issue preclusion (also known as collateral estoppel) holds that
parties are estopped from relitigating issues that they had already litigated
and lost.[49]
The common-law rule was that collateral estoppel required mutuality, meaning
that it only applied if the party asserting collateral estoppel and the party
it was being asserted against were parties (or privies) to the original
litigation.[50]
The Supreme Court relaxed the requirement of mutuality in the interest of
preserving judicial resources, but also to promote finality and uniformity.[51] In
Mendoza, however, the Supreme Court
ruled that nonmutual offensive collateral estoppel does not apply to the
government because doing so would “substantially thwart the development of
important questions of law by freezing the first final decision rendered on a particular
legal issue.”[52]

Nationwide
injunctions, however, constitute an end-around Mendoza because they effectively prevent the federal government
from relitigating issues in other courts, subjecting the federal government to
a form of de-facto claim preclusion. As noted above, this poses the same harms
to the judicial system that non-mutual claim preclusion against the government
would have: thwarting percolation among the circuits and putting pressure on
the Supreme Court’s certiori process.[53] Even
though Mendoza signaled the
importance of the federal government having discretion in relitigating
previously decided issues, courts have frequently prevented exactly that by
issuing nationwide injunctions.

Intercircuit
agency nonacquiescence is a second doctrinal inconsistency with the use of nationwide
injunctions is.[54]
While the Supreme Court has never specifically endorsed any form of agency nonacquiescence,
it flows logically from Mendoza and
the rule against intercircuit stare decisis: if court of appeals decisions are
only precedential within their circuit, and if according to Mendoza the federal government should be
free relitigate issues in multiple circuits, then agencies should not be bound
by a judicial decision nationwide until the Supreme Court resolves the issue.[55]
Leaving to one side the argument over whether intracircuit nonacquiescence is constitutional,[56]
intercircuit nonacquiescence has been widely accepted.[57] However,
just as nationwide injunctions constitute an end-around Mendoza, they effectively nullify agency nonacquiescence.[58]
By enjoining agencies from enforcing policies anywhere in the country,
nationwide injunctions essentially require agencies to acquiesce to the
injunction-issuing district court.

III. Current Scholarship on Nationwide Injunctions

A.     Professor Bray’s Plaintiff-limited injunction

One
of the first and widely cited articles to argue how nationwide injunctions
could be limited was written by Professor Bray.[59]
He argues that, because the role of the courts is solely to do justice to the
parties before the court, courts do not and should not have the power to enjoin
a defendant vis-à-vis anyone but named plaintiffs.[60]
Professor Bray finds evidence for this proposition in two sources: Article III
and traditional principles of equity. Bray argues that because, at equity,
there were no injunctions against the crown, since as noted above the
chancellor spoke on behalf of the king, there can be no support in equity for
nationwide injunctions.[61]
Second, Professor Bay argues that because Article III gave the courts the
“judicial power” and this power was only a power to do justice between the
parties before the court, any equitable remedy that reaches beyond the named
plaintiffs violates Article III.[62]
Lastly, Professor Bray argues that limits on traditional equity were not
necessary because there was one chancellor, while because of the circuit court
system in the United States there are “multiple chancellors.”[63]

While
Professor Bray’s plaintiff-limited injunction is clear and easy to apply, it is
to be disfavored for multiple reasons. First, as noted above, while it is true that
traditional principles of equity did not often allow for courts to reach
non-parties,[64]
equity constantly evolved to meet new issues of the day that the formalistic
and static common law could not.[65]
Even if this is true, it represents an unfortunate misunderstanding of the role
of equity. Because the chancellor stood in the place of “God and King,”[66]
equity’s ability to fashion new remedies to meet new problems is clearly
established. Had there been a widespread failure of the common law that
required injunctions to other parties, it seems clear that equity would have
had the authority to do exactly that.

Professor
Bray’s Article III argument is distinct, but also vulnerable. Under this
account, the “older” (and correct) conception of the judiciary is one in which
the judiciary merely refused to apply statutes that conflicted with the
constitution because there was a higher law, while the “new” (and false)
conception of judicial review is that judges “strike down” unconstitutional
statutes.[67]
While a thorough argument about the proper conception of the judiciary is
beyond the purview of this note, Professor Bray’s article endorses a narrow
view of judicial review that, while endorsed off and on by the Court at various
times,[68]
is inconsistent with the founder’s understanding of the judiciary. As noted
above in the section on the founding, Professor Barnett convincingly argues
that the original meaning of the “judicial power” included the judicial
nullification.[69]
While Barnett cites may convincing sources of the original meaning of the
“judicial power,” one quote from Alexander Hamilton is particularly
instructive: “the courts were designed to be an intermediate body between the
people and the legislature in order, among other things, to keep the latter
within the limits assigned to their authority.”[70] If
one accepts an originalist framework, as Professor Bray appears to do, it is
clear that the courts have a role in supervising the constitutional order, and
are not relegated to declining to enforce statutes out of protest.

Even if
Professor Bray were correct about the proper role of the judiciary, his
proposition is clearly inconsistent with the Supreme Court’s current view of
the propriety of injunctions reaching beyond class members. The Supreme Court
has recognized that district courts enjoy substantial discretion in fashioning
equitable relief, just as in England since equity was “transplanted” here in
1789.[71] While
no doubt Professor Bray would gladly acknowledge what a revolutionary effect
his theory would have if it were implemented, a rigid rule that a court could
never reach beyond named plaintiffs is not only contrary to the idea that
equity involves discretion, but would lead to a plethora of unintended
consequences.

Professor
Bray freely acknowledges one of the most striking¾plaintiff
detection.[72]
The example he provides is a good one, which was a set of fourteen plaintiffs
who challenged¾and won¾ a
California Highway Patrol policy of aggressive enforcement of helmet law, which
the court found to violate Fourth Amendment.[73] The
Ninth Circuit affirmed an injunction against the Highway Patrol enforcing the
policy against anyone, not just the named plaintiffs, because it would be
incredibly impractical for officers to be able to tell whether a motorcyclist
on the road was one of the fourteen named plaintiffs that had won an injunction
against the state.[74]
Professor Bray responds that this does not matter because the burden is on the
State to find a way to comply with the injunction, and state could embrace a
more “creative option, such as distributing decals to the [plaintiffs]” to
assist officers with identification.[75]
Of course, there is a simpler option: allowing a statewide injunction.

Similar
but even more difficult cases cannot be solved by issuing decals to plaintiffs,
even if that were an acceptable option. Take, for example, school
desegregation. It is perhaps unsurprising that injunctions against governmental
entities changed during this time as Bray points out[76]
since the harm felt by discrimination was felt by an entire race of citizens.
As a result, the Supreme Court affirmed complex and far-reaching remedies,
desegregating entire school districts, setting up bussing, which was an
extremely complex remedial scheme.[77]
Another possible example would be nuisance law, where it would make little
sense to enjoin Party A from polluting the stream of Party B if the pollution,
in fact, affects a class of people not represented. Systemic harms require
systemic remedies, but Professor Bray’s theory would atomize the legal system,
handicapping it from addressing these types of problems.[78]

B.     Siddique’s “Complete Relief” Principal

A
second, more modest proposal is Zayn Siddique’s “complete relief” principle,[79]
which more or less restates the status quo approach to issuing injunctions.[80] This
rule would direct courts that “injunctive relief should be no more burdensome
to the defendant than necessary to provide complete relief to the plaintiffs.”[81]
This principle is sound, but is insufficient for a few reasons. The most
fundamental is that since this rule is already widely accepted by the courts,[82]
if it was going to work in constraining judges, it would have done so already.
Second, the “complete relief” principle works for and against the cause of
reducing judicial discretion with nationwide injunctions depending on which
half of the rule is emphasized. Judges who decide to issue an injunction
emphasize affording “complete relief,” to plaintiffs, while those who decide
that a nationwide injunction is inappropriate emphasize fashioning remedies to
be “no more burdensome to the defendant than necessary.” Therefore this
principle does nothing to change the status quo, and does not present an
obstacle to a judge who feels strongly that an injunction is necessary.

C.     Berger’s Circuit-Border Rule

A
third approach to limiting the nationwide injunction is that nationwide
injunction should be geographically limited to the circuit in which the issuing
court resides.[83]
Berger argues that this approach is preferable because sometimes injunctions
broader than just the named parties is necessary, but that nationwide
injunctions are too extreme of a remedy and should not be used.[84]
Berger advocates this proposal for two reasons: the congressional policy choice
to divide the circuits into geographic units, and the policies underlying the
Supreme Court’s decision Califano,
which held that the federal government is not subject to nonmutual issue
preclusion.[85]
With the first warrant, Berger writes that the circuit court system “embody a
congressional policy judgment that they are large enough to create regional
uniformity yet distinct enough to foster genuine percolation and intercircuit
dialogue, thus striking an equilibrium between these competing values. The
circuit-border rule does the same.”[86]
As to the second warrant¾the
precedent of Califano¾Berger
argues that because the Supreme Court communicated that the ability for the
circuits to communicate with each other was important, and the circuit-border
rule promotes the same value, it is supported by precedent.[87]

While
Berger’s thesis is an admirable attempt to solve a complex problem, it has a couple
of shortcomings. The first of these is that it is supposed rest on implied
congressional intent to limit the jurisdiction of courts to their circuits. The
problem with this argument is that traditional rules of equity have long held
that equity could enjoin acts committed outside of the court’s territorial
jurisdiction.[88]
Indeed, language in Califano, itself
a source Berger identifies as support for the circuit-border rule, rules out
the idea that equity is limited by geography.[89]
Given the long, unbroken history of equity acting “in personam” regardless of
geographical lines, one would expect Congress to be more explicit if it
intended to restrict courts’ authority in this way.

Second,
the circuit-border rule is arbitrary. While the circuit borders are a useful
way of breaking up the nation, one could instead use the borders of the 94
district courts. While such a system would doubtless be difficult to administer
(though still more flexible than the principle advocated by Bray), the choice
of where to draw the lines is a policy choice, and has no equitable “pedigree.”[90]
Therefore, if the nationwide injunction is to be constrained, a different
approach is necessary.

D.    Other Approaches

Other
recommendations for constraining the national injunction have been advanced which
deserve mention here. Professor Morley has suggested that courts apply an equal
protection and severability analysis before issuing a nationwide injunction,
requiring plaintiffs properly seeking broad remedies to certify as a 23(b)(2)
class action.[91]
Maureen Carroll recommends a similar but softer approach: a set of changes to
make class action lawsuits more appealing.[92] Daniel
J. Walker proposes a list of nine factors which courts should consider,
suggestive of a multifactor balancing test.[93] Lastly,
Michelle R. Slack proposes a rule that courts should employ presumption against
certifying a class action lawsuit when the government is a party.[94]
Each of these theories make good points but are not mutually exclusive with my
proposal,[95]
but are included here for completeness.

IV. Triangulating Equitable Remedies

The
existence of discretion and ability to evolve is a feature, not a bug in the
system of equity. However, it is not without drawbacks, namely the
indeterminacy and inconsistency inherent with any system of standards and
rules. This paper suggests that the solution to the problem is the obvious and
time-tested one: equitable balancing. The concept of “balancing the equities”
is hardly new, and courts already apply it in some form. In this section I
argue that equitable balancing is consistent with equity’s history, and that
balancing tests are beneficial because they force judges to justify their
choice of remedy, an important feature of the common law method. The biggest
roadblock to use of a balancing test in this arena are the sheer number of
possible factors that could be used. This paper’s first contribution to the
doctrinal landscape is that it proposes a three-factor framework to simplify
the multiplicity of possible factors into a more manageable balancing test,
providing both a means for evaluating different factual scenarios and for
critiquing them.

Ultimately,
it seems inevitable that courts will have a great deal of discretion when
fashioning equitable remedies as this has been a feature of equity for
centuries. Therefore, the only solution absent Congressional attention will
return to the roots of equity, engaging in a multifactor balancing test to
shape the appropriate remedy.[96]
As noted above, equity has a long history of providing courts with discretion.[97]
Therefore, absent a statute or evidence of congressional intent to the
contrary, this discretion must be a feature of any system of injunctions. A
system of multi-factor balancing is the only system that meets this
description. However, to effectively construct a multifactor balancing test, it
is necessary to examine as many possible factors as possible. Unfortunately,
the sheer number of possible factors has not been appreciated by previous
scholars, and a thorough examination shows that there are far too many to
construct a simple balancing test.

A.     The number of factors makes balancing difficult to impossible

While
there has been some attempt to evaluate the factors courts have and should
consider when deciding whether to issue a nationwide injunction,[98]
any list will almost certainly be under inclusive since it is likely impossible
to consider, ex ante, every possible factor. Nevertheless, this note attempts
to collect the most important factors¾and
quite a few miscellaneous factors¾that
have been and should be considered by courts when fashioning an injunction.

• State Courts, “Our Federalism,” and Abstention Issues

The
doctrine of “Our Federalism” holds that federal courts should not enjoin an
ongoing state prosecution out of concern for the interest of the states in
carrying out prosecutions.[99]
“Our Federalism” is conceptually similar to anti-commandeering cases
established by the Rehnquist court in New
York v. United States[100]
and Printz v. United States.[101] There
are a number of abstention doctrines, including Pullman,[102] Colorado River,[103] Burford, and Rooker-Feldman[104]
abstention, which are similar in the sense that they often consider issues of
comity between the courts, but since they are doctrines of abstention they
leave discretion to the judge applying them that the 10th Amendment cases do
not. While the details of injunctions will vary from doctrine to doctrine, the
existence of a state as a party, especially as a defendant, may counsel towards
a narrower injunction or even no injunction at all.[105]

• Federal Agencies, Intercircuit Nonacquiescence, and Mendoza

Because
of the holding in United States v.
Mendoza,[106]
the presence of an agency as a defendant is a significant factor in weighing
whether to issue a nationwide injunction or not. Because a nationwide
injunction does, to some extent, subject an agency to de-facto issue preclusion
and renders intercircuit nonacquiescence moot,[107]
if an agency is the defendant against whom an injunction is sought, a narrowing
presumption is to be favored.

• Type of Agency being Enjoined

The
type of agency is also material in the likelihood that a court will show
difference when balancing the equities.[108]
Courts may be more likely to give deference to law enforcement agencies a
greater deal of deference than they would other types of agencies.[109]
For example, while this may be partially due to the nature of an inmate’s
rights rather than the type of agency itself, the Supreme Court has approved
some restrictions on inmate behavior by prison administrators, like limits on
contact with the outside world, that would not be permissible in another
context.[110]
Similar in principle to issues of comity, federalism, and separation of powers,
courts must be sure to not unduly pressure law enforcement’s ability to carry
out their legal duties, while simultaneously protecting the rights of those law
enforcement interacts with.

• Substantive Area of Law and Scope of the Injury

One
of the most important factors courts must consider is the substantive area of
law being applied.[111]
As noted above, civil rights cases are more likely to require complex, systemic
remedies than a breach of contract case (even one involving the federal government)
would.[112]
First Amendment may likewise be a candidate for broader injunctions because the
harm experienced, through a chilling of speech, is diffuse and difficult to
trace completely to one particular plaintiff or group of plaintiffs.[113] Professor
Morley’s recommendation that courts conduct a kind of severability analysis
concerns exactly this issue.[114]
If the harm to particular plaintiffs can be addressed by eliminating just
certain portions of the statute, then a broad injunction overturning the entire
statute is not necessary. By examining how localized the harm is, geographical,
statutory, or otherwise, courts have an indication of how broad the remedy may
need to be to afford complete relief.

• Facial vs. As Applied Challenges

While
some courts have arguably gone too far in treating the existence of a facial
challenge as prima facie support for a nationwide injunction against an act,[115]
the nature of the challenge is nonetheless important. An influential opinion in
this regard is Justice Blackmun’s dissent in Lujan v. National Wildlife Federation,[116]
where he stated that as-applied challenges generally justify narrow remedies,
while facial challenges justify remedies that benefit non-parties.[117]
These broad remedies are not necessarily mandated in such cases and courts
should consider other factors, but they are permissible.

• The Type of Injunction Sought

The type of injunction sought is
material to the scope of that injunction. For example, mandatory injunctions
are generally more onerous on the defendant than a prohibitory injunction
because they require the defendant to actively carry out some action rather
than simply refraining from action.[118]
In addition, the procedural facts of the case are important, as preliminary
injunctions are generally easier to get than permanent injunctions, and should
therefore be more narrowly tailored until more discovery can be done into the
nature of the violation (assuming that a legal wrong has even been committed).[119]

• Judicial Resources and Venue Considerations

In at least one case that discussed
the propriety of class-wide relief in the absence of a class-action lawsuit,
conservation of judicial resources was an important factor that weighed in
favor of issuing an injunction.[120]
While the court also argued that the APA mandated broad injunctions, a concern
that a plaintiff-focus injunction would “merely to generate a flood of
duplicative litigation.”[121]
The fact that much of this litigation centered in the D.C. Circuit factored
into the court’s prediction.[122]
The pressures each circuit faces are different and the chances for duplicate
litigation are unique to each case, so likely no abstract principle can be
stated, but conservation of judicial resources may sometimes weigh towards
issuing a broad injunction.

• Boundaries of the Potential Class and Asymmetric Effects of Res Judicata

The boundary of potentially
affected non-parties influences whether a broad remedy is appropriate absent a
class action suit for two reasons. The first of these is that if the affected
class of non-parties is very small and cohesive, an injunction that reaches
non-parties even more clearly falls within a court’s equitable powers, as it
approximates the English concept of a “bill of peace” which even Professor Bray
acknowledges may be an appropriate use of the judicial power.[123]

More fundamentally, however, is the
problem that allowing for a single class member to win an injunction that
affects a large class puts defendants at systemic disadvantage.[124]
With a class action, a defendant benefits from getting res judicata against
members of the class, but faces an equally large liability towards each member
of the class. In contrast, if a single plaintiff can win a broad injunction
that benefits an entire class, the liability to the defendant approximates that
of a class action lawsuit, but if the defendant wins the res judicata effect of
that decision applies only against one plaintiff, leaving an almost infinite
number of other plaintiffs to bring the same lawsuit.[125]

• Miscellaneous Factors

In
addition to the factors listed above, there are five potential issues with
nationwide injunctions which I list here because they have been discussed at
length elsewhere. Uniformity,[126]
plaintiff detection,[127] egalitarian
concerns[128],
forum shopping,[129]
comity concerns surrounding conflicting injunctions,[130]
and percolation,[131]
are all factors that affect the proper scope of an injunction. For discussion
of these topics in greater detail, see section II.

In
sum, no fewer than fourteen factors are relevant in discussing the proper scope
of a nationwide injunction, and there are certainly more, and some of these
listed could doubtlessly be broken up in even more detail. However, it seems
clear that Daniel Walker’s complex balancing test is even more complex than it
first appears, and a purely ad hoc balancing of so many factors will leave
judges with almost infinite discretion.

B.     A “triangulation” of equitable remedies simplifies balancing

This
paper’s first, and perhaps most significant innovation is a new framework for
categorizing and evaluating the factors that affect nationwide injunctions.
While these factors are deserving of individual consideration, paying attention
to the key characteristics of each reveals that there are really three
categories of factors: the nature of the parties before the court, the
substantive law at issue, and the effect of the decision on the court system as
a whole. These factors can be thought of as asking “who,” “what,” and “where”
respectively.

• “Who:” the parties before the court

The
first of these categories is the nature of the parties before the court. This category
includes:

• The type of party involved
  (state, federal, and private parties)
• Federalism and abstention
  issues associated with state defendants
• The precedent of Mendoza and intercircuit agency
  nonacquiescence
• The type of agency being
  sued (including whether is a law enforcement agency)
• The presence of a
  certified class (or conversely the size of the non-party beneficiaries)
• Plaintiff-detection
  issues.

• “What:” The substantive issue being litigated

This category includes:

• The Substantive Area of
  Law/Scope of the Injury
• Facial vs. As Applied
  Challenges
• Type of Injunction Sought

1. “Where:” The court
   deciding the case and the macro effects on the court system as a whole
2. Boundaries of the Class
   and Asymmetric Effects of Res Judicata
3. Conservation of Judicial
   Resources / Venue Considerations
4. Uniformity in Application
   of the Law
5. Comity and Conflicting
   Injunctions
6. Forum Shopping
7. Percolation

V. “Hot and Cold”: Easy and Difficult Cases for Equitable Balancing and a New Path Forward

This
note’s second innovation is in how courts can use this factor category
framework to effectively apply a multifactor balancing test in determining when
a nationwide injunction is appropriate.

A.     Abuse of discretion in “hot” and “cold” cases

Because
the main obstacle towards reigning in the unfettered use of nationwide injunctions
is in the ability to reverse at the court of appeals level, and a multifactor
balancing test is a kind of standard, district courts will be reviewed for
abuse of discretion. Professor Bray argues that this is a fundamental flaw with
using a standard.[132]
While I acknowledge this difficulty, the danger is overstated. At the very
least, there exists a category of cases which are “hot:” when all three
categories point towards issuing a broad injunction, then a court is firmly
within its power to do so. In contrast, when all three factors point against
issuing a broad injunction, then it is abuse of a discretion for a court to do
so. While this is imperfect since often the factors will be split,[133]
this at least affords a way to curb the greatest errors of the judiciary
without resorting to the extreme position of never allowing remedies to reach
non-parties.

B.     The common law method in “warm” cases

Even
in those cases where the categories are split, resorting to a balancing test
still affords a benefit over the status quo: it forces judges to describe and
justify their reasoning. The good news for those concerned about nationwide
injunctions is that, perhaps, a dialogue about the proper scope of injunctive
relief is beginning which can iron out the details of these difficult cases. This
is the ordinary conception of how the common law works.[134]
As Oliver Wendell Holmes’ eloquently stated “We must think things not words, or
at least we must constantly translate our words into the facts for which they
stand . . . .”[135]
Legal rules are made to apply to real circumstances, and by testing out
different approaches, over time more effective rules are developed. Injunctions
should be no exception.

C.     Application #1: Trump’s Travel Ban

While
nationwide injunctions in general are controversial, none has gotten as much
widespread attention as Executive Order 13769[136]
and its successor Executive Order 13780,[137] popularly
referred to as the travel ban.[138]
While the first[139]
and second[140]
executive orders differed in important ways from a liability perspective, from
a remedies perspective they were nearly identical and will be analyzed
together.

With
the first category, the identity and nature of the parties before the court,
the defendant was (obviously) the federal government and the specific agencies
involved included the Department of Homeland Security, the Secretary of State,
and the President himself.[141] The
dangers of enjoining the President at least are strong, analogous to the
dangers of enjoining a law enforcement agency[142]
since the President is made responsible for overseeing enforcement of the laws
by the Take Care Clause.[143]
Indeed, the Fourth Circuit and Ninth Circuits both reversed the district
courts’ injunctions against the President personally for these exact prudential
reasons, though it kept the rest of the injunction intact.[144]
Issues surrounding intercircuit nonacquiescence were somewhat ameliorated
because two district courts[145]
and two courts of appeals[146]
had the chance to review the revised order and another district[147]
and circuit[148]
court had already reviewed the first order. The last relevant sub-factor, the
breadth of the affected class, weighs strongly against issuing an injunction.
While it is difficult to say how many people the first or second travel bans
would have affected, the revised travel ban limited the number of refugees per
year to 50,000 while President Obama had planned to admit around 100,000,[149]
so the number of unnamed plaintiffs affected is likely in the tens of
thousands. Putting the pieces together, every relevant sub-factor counsels
against issuing an injunction, so this category counsels against issuing an
injunction.

The
second category, the substantive area of law, is more balanced but also
counsels against issuing an injunction. Immigration law does not have the same
spill-over effects associated with nuisance law for example. The Fourth Circuit
did reason that Establishment Clause violations justify broader injunctions
because allowing the policy to be enforced against others would send a message
that the plaintiffs were “outsiders” and “not full members of the political
community” citing Santa Fe Independent
School District v. Doe.[150]
Yet the Fourth Circuit’s reliance on Santa
Fe seems misplaced as the quote in question went to liability, not the
appropriate scope of the remedy. More importantly, it was possible to make
distinctions between the injury felt by the individual plaintiffs seeking entry
to the United States, while prayer at school football games (which was at issue
in Santa Fe) is binary¾either
it happens or it does not). Because the injury is severable from the whole,
this sub-element councils against issuing an injunction. The challenge was a
facial challenge however, which cuts the other direction. The type of
injunction sought was also prohibitory (against the executive order taking
effect) which is less onerous than a mandatory injunction. On balance this
category counsels against issuing an injunction because if the mere fact that a
challenge were facial and prohibitory¾which
almost all constitutional suits against enforcement of a law are¾would
be sufficient to justify an injunction. Therefore the type of injury involved
does not justify a broad injunction.

The
third and final category, the effect on the court system, likewise points away
from a nationwide injunction. The boundaries of the class size are very large,
so the asymmetric effects of the Res Judicata are large. Only one plaintiff must
prevail for every class member to benefit, while the government gets the
benefit of Res Judicata. There does not appear to be a court crowding issue
specifically at play like in National
Mining Association[151]as none of the cases had to be filed in
a specific circuit. Forum shopping is certainly a potential problem since an
immigrant could plan to use any city as port of entry (that received
international flights) to receive standing to sue in the district of their
choosing.[152]
Similar to the issue of nonmutual issue preclusion, the chance for percolation
among the courts was hurt by these injunctions, though multiple courts still
issued opinions[153]
and the issue was so well publicized that there was little chance that the
Supreme Court would need a circuit split to be convinced to take the case.
Uniformity in application of the law, as with any case, also played a role and
was specifically mentioned by both courts of appeals.[154]
Ultimately the direction this factor ought to lean depends on the weight one
puts on the doctrinal problem with unjust Res Judicata results against the
government, forum shopping, and percolation, compared with the benefit of uniformity
and efficiency. However, as Professor Bray eloquently argues, Congress’s choice
to break up the courts into circuit was itself a policy decision to endure a
loss of uniformity and efficiency for more incremental, but less unstable
change.[155]
“Each legal system can pick its poison, tending toward the vices of immediate,
final resolution or the vices of slow, provisional resolution.”[156]
In close cases, because Congress has already made this policy decision, courts
should apply that same policy preference in cases like this one.

D.    Application #2: Deferred Action for Parents of Americans (DAPA)

On
June 15, 2012, the Department of Homeland Security implemented a policy known
as the Deferred Action for Childhood Arrivals program (DACA), by which DHS
would no longer prosecute those who were brought to the United States as
children but had not immigrated legally.[157]
That policy was supplemented on November 20, 2014 through the Deferred Action
for Parents of Americans (DAPA) program which extended similar protections to
parents of American citizens or lawful permanent residents.[158]
Seventeen states jointly sued the United States under the Take Care Clause and
Administrative Procedure Act.[159]
The district court issued a preliminary injunction which was nationwide in scope[160]
which was upheld by the Fifth Circuit.[161]
The Supreme Court however, due to the death of Justice Scalia, split 4-4,
upholding district court’s opinion.[162]

In reviewing the propriety of this injunction, the first factor of the identity of the parties before the court, favors issuance of a nationwide injunction. While the defendant in this case was the federal government, the plaintiffs were also governments. More importantly, because the size of the “class” of states is fifty, seventeen out of the fifty in that class were represented, and apparently all of the states who wished to sue had joined the lawsuit, most of the sub-factors in this category were not a problem. The government had no interest in continuing to litigate the issue in other circuits because there would be no more litigation on this subject. The non-party beneficiary class was very low. Therefore, this factor could only have more strongly favored a nationwide injunction if every single state joined the lawsuit, which effectively means that one could hardly imagine a better set of facts for purposes of this category.

The second category, the substantive nature of the claim, likewise supports an injunction. This was a facial challenge,[163] the injunction was prohibitory (against the memo having legal effect), and most importantly the plaintiffs could not be made whole without a nationwide injunction because, much like with constitutionality of school prayer in Santa Fe,[164] legal status is binary¾either immigrants have the right to be in the United States or they do not. Perhaps a circuit border rule could be implemented based on where an immigrant was apprehended, but this would likely be a logistical nightmare. Moreover, this would make it harder for states who won to keep undocumented immigrants out of their state since every border with a state that did not join the lawsuit¾not just the Mexico and Canadian borders¾would become a possible entry point, which the Border Patrol was almost certainly unable to handle. Therefore this injunction also strongly favored a nationwide injunction.

Lastly
the third factor, the effect on the courts, also mostly favors issuing an
injunction. Once again, because there would only be one lawsuit of this kind
since 17 states all joined the lawsuit, there were no asymmetric disadvantages
to the federal government, and neither are there comity or percolation issues.
Uniformity also favors a nationwide injunction (as it almost always does).
Forum shopping is something of a concern, perhaps in this case more than any
other because there were such strong indications of careful venue selection.[165]
This makes this case interesting because single other sub-factor discussed was
either moot or favored a broad remedy, but this one factor is extremely clear
and negative. While this is unfortunate, the effect on the court system of
issuing this injunction were still generally positive in this case because the
case was likely to head to the Supreme Court anyways (it took about a year for
the Supreme Court to issue a decision, and at the time no-one could have
predicted Scalia’s death), and had the policy gone into effect the courts would
have had to expend considerably more resources unwinding the policy. While this
category is the closest of the three, it still pointed towards issuing an
injunction.

In
summary, while there are lessons about the dangers of forum shopping to be
learned from Texas, all three
categories pointed towards issuing an injunction. While a careful look at the
disadvantages of nationwide injunctions should, at least compared to the status
quo, lead to fewer nationwide injunctions, this is not always the case as Texas v. United States shows.

---

[1] Earl of Oxford’s Case,
21 Eng. Rep 485, 486 (1615).

[2] See Howard L. Oleck, Historical
Nature of Equity Jurisprudence, 20
Fordham L. Rev. 23, 35 (1951).

[3] See id. at 33-34 (“When one recalls
that the first Chancellors of the king were churchmen, and that this situation
continued for a long time, it is hardly surprising that Roman and
ecclesiastical law and equity had so profound an effect on English law.”).

[4] See id. at 37-38.

[5] Selden, Table Talk 43 (Pollock ed. 1927).

[6] William Blackstone, Commentaries on the Laws Of England 61
(Oxford, Clarendon Press 2009) (1765).

[7] See id. at 39 (discussing King James’ affirmation of equity’s
ability to enjoin enforcement of common law judgements found to be unjust).

[8] See, e.g., Leonard Levy, Original Intent and the Framers’
Constitution100 (MacMillan, 1988).

[9] See, e.g., William W. Van Alstyne, A Critical Guide to Marbury v. Madison, 1969 Duke L.J. 1, 1 (1969) (“[T]he concept of
judicial review of the constitutionality of state and federal statutes by the
Supreme Court is generally rested upon the epic decision in Marbury v.
Madison.”).

[10] See Randy E. Barnett, The
Original Meaning of the Judicial Power, 12
Sup. Ct. Econ. Rev. 115, 117 (2004).

[11] See id. at 116-117.

[12] Professor Bray
identifies the 1960’s as the beginning of the nationwide injunction. See Samuel L. Bray, Multiple
Chancellors: Reforming the National Injunction, Harvard Law Review (forthcoming
2017) (manuscript at 32-34),
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2864175.

[13] 5 U.S. 137, 177 (1803)
(“It is emphatically the province and duty of the judicial department to say
what the law is.”).

[14] See Marbury, 5 U.S. at 180 (“It is also not entirely unworthy of
observation, that in declaring what shall be the supreme law of the land, the
constitution itself is first mentioned; and not the laws of the United States
generally, but those only which shall be made in pursuance of the constitution,
have that rank.”).

[15] City of Chicago v.
Sessions, Case No. 17 C 5720, 2017 WL 4572208, at * 4 (N.D. Ill. Oct. 13th,
2017).

[16] See Sam Bray, Finally, a
court defends the national injunction, The
Volokh Conspiracy: Washington Post, (Oct. 14, 2017),
https://www.washingtonpost.com/news/volokh-conspiracy/wp/2017/10/14/finally-a-court-defends-the-national-injunction/?utm_term=.e7d4d97888e9
(“Once that proposition is accepted, the national injunction will have become
the norm for all challenges to the validity of a federal statute, regulation,
or order.”).

[17] 337 F.2d 518, 534-35 (D.C. Cir. 1963).

[18] Id. at 534. The court in City
of Chicago v. Sessions likewise endorsed this reasoning in deciding to
issue a nationwide injunction. See City
of Chicago v. Sessions, 2017 WL 4572208, at *4 (“All similarly-situated
persons are entitled to similar outcomes under the law, and as a corollary, an
injunction that results in unequal treatment of litigants appears arbitrary.”)

[19] See Kate Huddleston, Nationwide
Injunctions: Venue Considerations, 127 Yale
L.J. F. 242 (2017), http://www.yalelawjournal.org/forum/nationwide-injunctions-venue-considerations
(arguing that eliminating nationwide injunctions would lead to “differential
access to favorable judgments based on litigant resources.”).

[20] See infra Section II.b.iv.

[21] See National Mining Ass’n v. US. Army Corps of Engineers, 145 F.3d
1399, 1410-1411 (D.C. Cir. 1998) (granting nationwide injunction on the basis
that a narrower injunction would lead to a “flood of duplicative litigation”
and could overburden the D.C. Circuit).

[22] See Washington v. Trump, No. C17-0141JLR, 2017 WL 462040, at *2
(W.D. Wash. Feb. 3, 2017), stay denied,
847 F.3d 1151 (9th Cir. 2017) (issuing a nationwide injunction against the
United States from enforcing President Trump’s executive order); Int’l Refugee
Assistance Project v. Trump, No. TDC-17-0361, 2017 WL 1018235, at *17-18 (D.
Md. Mar. 16, 2017) (injunction against the revised version of the travel ban);
Hawai’i v. Trump, CV. NO. 17–00050 DKW–KSC, 2017 WL 1167383, at *9 (D. Haw.
Mar. 29, 2017) (injunction against other parts of the travel ban).

[23] See, e.g., John Schwartz, ‘Liberal’
Reputation Precedes Ninth Circuit Court, N.Y.
Times, (Apr. 24, 2010), http://www.nytimes.com/2010/04/25/us/25sfninth.html
(conservatives attacking the 9th Circuit is “a familiar refrain.”). But see Erwin Chemerinsky, The Myth of the Liberal Ninth Circuit,
37 Loy. of Los Angeles L. Rev. 1,
1 (2003) (“The popular image of the Ninth Circuit . . . is that it is a far
left court that is reversed more often than any other circuit in the country. This
is simply wrong.).

[24] See, e.g. Earth Island Inst. v. Ruthenbeck, 490 F.3d 687 (9th Cir.
2007); Earth Island Inst. v. Pengilly, 376 F. Supp. 2d 994 (E.D. Cal. 2005);
California ex rel. Lockyer v. U.S. Dep’t of Agric., 468 F. Supp. 2d 1140 (N.D.
Cal. 2006).

[25] Texas v. United States,
787 F.3d 733, 769 (5th Cir. 2015).

[26] See Andrew Kent, Nationwide
Injunctions and the Lower Federal Courts, Lawfare,
(Feb. 3, 2017, 3:02 PM)
https://www.lawfareblog.com/nationwide-injunctions-and-lower-federal-courts.

[27] Id.

[28] Id.

[29] Other scholars have
widely noted this asymmetry. See Maureen
Carroll, Aggregation for Me, but Not for Thee: The Rise of Common Claims in
Non-Class Litigation, 36 Cardozo
L. Rev. 2017, 2020-2021 (2015).; Michael T. Morley, De Facto Class Actions?
Plaintiff- and Defendant-Oriented Injunctions in Voting Rights, Election Law,
and Other Constitutional Cases, 39 Harv.
L. & Pub. Pol’y 487, 494 (2016); Bray, supra note 11 at 8-12.

[30] 549 F.3d 760 (9th Cir.
2008).

[31] This type of asymmetry
also played out in the Trump travel ban litigation when a judge in in Boston
refused to enjoin the travel ban.  See
Louhghalam v. Trump, No. CV 17-10154,
2017 WL 479779 at *8 (D. Mass. Feb. 3,

2017)
(denying injunctive relief on the basis that plaintiffs failed to show
likelihood of success on

the
merits). That victory was essentially rendered meaningless when it was
effectively overruled by the nationwide injunctions issued by the other
district courts.

[32] See Alan Feuer, Brooklyn
Lawsuit Could Affect the Fate of Millions of Immigrants Nationwide, N.Y. Times (Oct. 9, 2016),
https://nyti.ms/2dMnV2P. During the hearing, Judge Garaufis stated “. . . I
have absolutely no intention of simply marching behind in the parade that’s
going on out there in Texas, if this person has rights here.” Id.

[33] See Bray, supra note 8 at
14.

[34] However, Professor Bray
rightly points to the Erie Railroad legal battles in the nineteenth century,
which involved conflicting inunctions between state judges, as historical
examples of conflicting injunctions. See Bray,
supra note 8, at 13-14.

[35] See, e.g., California ex rel. Lockyer v. USDA, 710 F. Supp.
2d 916, 920 (N.D. Cal. 2008) (narrowing the scope of an injunction against the
USDA to avoid conflict with the District of Wyoming’s injunction, but lamenting
“the unfortunate appearance of a lack of judicial comity that has arisen in the
wake of the Wyoming court’s decision and the awkward position in which the
United States Department of Agriculture finds itself.”)

[36] Feller v. Brock, 802
F.2d 722, 727-728 (4th Cir. 1986) (reversing a preliminary injunction for
improperly balancing the equities).

[37] This final remedy,
however, still leads to the asymmetries illustrated by United States v. AMC noted above. See, supra text accompanying notes 13-17.

[38] See Bray, supra note 8 at
20-39.

[39] Harold Leventhal,
Eleventh Annual Mooers Lecture, A Modest Proposal for a Multi-Circuit Court of
Appeals, 24 Am. U. L. Rev. 881,
907 (1975).

[40] Arizona v. Evans, 514
U.S. 1, 23 n.1 (1995).

[41] See GetzelBerger, Note,
92 N.Y.U. L. Rev. 19 n. 99 (“[M]ost nationwide injunctions receive far less
attention, and, therefore, likely have a lower chance of receiving certiorari
absent a circuit split.”).

[42] See supra Section II.a.i.

[43] Zamecnik v. Indian
Prairie Sch. Dist. No. 204, 636 F.3d 874, 879 (7th Cir. 2011) (quoting 1 Dan B.
Dobbs, Law of Remedies 113, § 2.4(6) (2d ed.1993)).

[44] McKenzie v. Chicago,
118 F.3d 552, 555 (7th Cir. 1997).

[45] Compare Earth
Island Inst. v. Ruthenbeck, 490 F.3d 687, 699 (9th Cir. 2007) (holding the
language “set aside” in the APA compelled a nationwide injunction), with Virginia
Soc’y for Human Life, Inc. v. FEC, 263 F.3d 379, 394 (4th Cir. 2001) (stating
that “[n]othing in the language of the APA” compels nationwide injunctions).

[46] Compare, e.g., Franciscan All., Inc. v. Burwell, No.
7:16-CV-00108-O, 2016 WL 7638311, at *22 (N.D. Tex. Dec. 31, 2016) (“[A]
nationwide injunction is appropriate when a party brings a facial challenge to
agency action.”), with Los Angeles
Haven Hospice, Inc. v. Sebelius, 638 F.3d 644, 665 (9th Cir. 2011) (declining
to issue a nationwide injunction in spite of a successful facial challenge to a
regulation).

[47] 464 U.S. 154, 162–63
(1984).

[48] See id.

[49] See, e.g., Brockman v. Wyoming Dept. of Family Services, 342 F.3d
1159, 1165-66 (10th Cir. 2003)
(“‘The collateral estoppel doctrine prevents relitigation of issues which were
involved actually and necessarily in a prior action between the same
parties.’”) (quoting Kahrs v. Bd. of Trs. for Platte County Sch. Dist. No. 1,
901 P.2d 404, 406 (Wyo.1995)).

[50] See Gary R. Cunningham,
Collateral Estoppel: The Changing Role of the Rule of Mutuality,41 Mo.
L. Rev. 521, 522 (1976) (“Because only parties or their privies could be
bound by a prior adjudication, mutuality requires that only they may benefit
from one.”) (citations omitted).

[51]See Parklane Hosiery Co. v.
Shore, 439 U.S. 322, 331 (1979) (non-mutual offensive issue preclusion);
Blonder-Tongue Labs., Inc. v. Univ. of Ill. Found., 402 U.S. 313, 333–34 (1971)
(non-mutual defensive issue preclusion). Defensive issue preclusion refers to a
defendant’s use of issue preclusion, while offensive issue preclusion refers to
its use by a plaintiff. See Linda J.
Soldo, Parklane Hoisery: Offensive Use of Nonmutual Collateral Estoppel in
Federal Courts, 29 Cath. U. L. Rev.
509, 510 (1980) (discussing the differences between offensive and defensive
issue preclusion). Judge Traynor initiated the relaxation of the rule of
mutuality in Bernhard v. Bank of America National

Trust
& Savings Association in 1942, which has since been adopted by the
federal courts. 122 P.2d 892 (1942).

[52] Mendoza, 464 U.S. at 160.

[53] Getzel Berger
insightfully recognized that both the benefits and harms of nonmutual issue
preclusion against the federal government mirror those of nationwide
injunctions. See Berger, supra note 20 at 28 (“The systemic
policy considerations weighed in Mendoza mirror the key policy
considerations on nationwide injunctions. The Court’s analysis framed the issue
as pitting uniformity and efficiency against percolation and intercircuit
dialogue.”).

[54] There are two primary
forms of agency nonacquiescence, both of which flow from Mendoza: intercircuit and intracircuit nonacquiescence.
Intercircuit nonacquiescence refers to an agency’s decision to not be bound by
a court’s decision in other jurisdictions, while intracircuit nonacquiescence
is an agency’s decision to not be bound by a decision in the same jurisdiction that issued the decision. See Kevin Haskins, A Delicate Balance: How Agency

Nonacquiescence and the
EPA’s Water Transfer Rule Dilute the Clean Water Act after Catskill Mountains
Chapter of Trout Unlimited, Inc. v. City of New York, 60 Me. L. Rev. 173, 175 (2008).

[55] For a summary of the
arguments for and against different forms of agency nonacquiescence, see
Haskins, supra note 28 at 176-183.

[56] See Samuel Estreicher & Richard L. Revesz, Nonacquiescence
by Federal Administrative Agencies, 98 Yale
L.J. 679, 743 (1989) (arguing that intracircuit nonacquiescence should
only be allowed in limited circumstances).

[57] See, e.g.,
Indep. Petroleum Ass’n of Am. v. Babbitt, 92 F.3d 1248, 1261 (D.C. Cir. 1996)
(“[I]ntercircuit nonacquiescence is permissible, especially when the law is
unsettled.”). See also Berger, supra note 20 at 30 n.160 (compiling
sources recognizing intercircuit nonacquiescence).

[58] See Berger, supra note 20
at 30 (“Nationwide injunctions flatly prohibit intercircuit nonacquiescence.).

[59] See Samuel L. Bray, Multiple Chancellors: Reforming the National
Injunction, Harvard Law Review (forthcoming
2017) (manuscript at 32-34),
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2864175.

[60] See id. at 3 (“A federal court should give what might be called a
“plaintiff-protective injunction,” enjoining the defendant’s conduct only with
respect to the plaintiff. No matter how important the question and no matter
how important the value of uniformity, a federal court should not award a
national injunction.”).

[61] See, id. at 18 (“In English equity before the Founding of the
United States, there were no injunctions against the Crown. No doubt part of
the explanation was the identification of the chancellor with the king . . .
.).

[62] See id. at 4.

[63] See id. at 4, 41-43.

[64]Though Professor Bray
himself acknowledges that this was not universally true, as the principle of a
“bill of peace” allowed the chancellor to resolve multiple claims of a cohesive
group all at once, a type of “proto-class action.” See id. at 22.

[65] See supra Section I.a.

[66] See Bray, supra note 10
at 63 (“In
this regard, there is a sharp contrast between the English Chancery and the
federal courts. A medieval chancellor spoke on behalf of God and king. An early
modern chancellor spoke on behalf of conscience and king . . . .”).

[67] See id. at 45.

[68] See, e.g., Carter v. Carter Coal Co., 298 U.S. 238, 296–97 (1936)
(stating that courts are “required to ascertain and apply the law to the facts
in every case or proceeding properly brought for adjudication, [and] must apply
the supreme law and reject the inferior statute whenever the two conflict.”).

[69] See Barnett, supra note 1
at

[70] Clinton Rossiter ed, The
Federalist No. 78 at 467 (Alexander Hamilton) (Penguin Books, 1961).

[71] See Guarantee Trust Co. of N.Y. v. York, 326 U.S. 99, 104 (1945)
(“The suits in equity of which the federal courts have had ‘cognizance’ ever
since 1789 constituted the body of law which had been transplanted to this
country from the English Court of Chancery.”).

[72] See Bray, supra note 10
at 58-59.

[73] See Easyriders Freedom F.I.G.H.T. v. Hannigan, 92 F.3d 1486, 1502-04 (9th Cir. 1996).

[74] See id. at 1502 (“[Since] it is unlikely that law enforcement
officials . . . would inquire before citation into whether a motorcyclist was
among the named plaintiffs or a member of Easyriders, the plaintiffs would not
receive the complete relief to which they are entitled without statewide
application of the injunction.”).

[75] See Bray, supra note 10
at 60.

[76] See id. at 48-50.

[77] See Brown vs. Board of Educ. of Topeka Kan., 349 U.S. 294, 301
(1955) (“The judgments below . . . are accordingly reversed and the cases are
remanded to the District Courts to take such proceedings and enter such orders
and decrees consistent with this opinion as are necessary and proper to admit
to public schools on a racially nondiscriminatory basis with all deliberate
speed the parties to these cases.”). See
also Daniel J. Walker, Note, Administrative
Injunctions: Assessing the Propriety of Non-Class Collective Relief, 90 Cornell L. Rev. 1119, 1132 (2005) (“The
extraordinary nature of [school desegregation] required the courts’ to stretch
their injunctive powers well beyond their historical limits.”).

[78] To his credit,
Professor recognizes that the plaintiff-only injunctions may be unworkable, but
does not believe that this matters. See Bray,
supra note 10 at 60 n. 259 (“In some
cases, if a plaintiff-protective injunction really is hopelessly impractical,
that is a reason not to give an injunction.”).

[79] See Zayn Siddique, Note, Nationwide
Injunctions, 118 Colum. L. Rev.,
(forthcoming 2018).

[80] See Bray, supra note 10
at 17 (“The [approach] most commonly raised by courts and commentators is the
principle of ‘complete relief’.”). See
also Califano v. Yamasaki, 442 U.S. 682, 702 (1979) (“[I]njunctive relief
should be no more burdensome to the defendant than necessary to provide
complete relief to the plaintiffs.”); Lewis v. Casey, 518 U.S. 343, 360 (1996)
(reversing broad injunction on the basis that it was not necessary to address
the plaintiff’s injury.”).

[81] See Siddique, supra note
76 at 8 (quoting Califano, 442 U.S.
at 702).

[82] The exception to where
this principal is not consistently applied is in APA challenges to agency
regulations. Siddique and other have identified a split in courts, some of
which hold that the APA allows for nationwide injunctions for all offending
regulations, while other courts do not. See
Siddique, supra note 76 at 23-28. See also supra note 42.

[83] See Berger, supra note 38
at 31-36.

[84] See id. at 31.

[85] See id. at 32.

[86] See id.

[87] See id. (“The nature of the regional circuits features prominently
in Mendoza, which focused on the ability of the regional courts of
appeals to disagree with each other.”).

[88] See Bray, supra note 10
at 31 n.143 (“Equity was willing to enjoin acts committed outside of the
chancellor’s territorial jurisdiction. Geographical lines were not the stopping
point.”).

[89] See Califano v. Yamasaki, 442 U.S. 682, 702 (“Nor is a nationwide
class inconsistent with principles of equity jurisprudence, since the scope of
injunctive relief is dictated by the extent of the violation established, not
by the geographical extent of the plaintiff class.”). It could be argued, as
Berger does, that the court in Califano was
speaking about the propriety of injunctions in class actions, not equity in
general. See Berger, supra note 38 at 33. However, the
context of the quote seems to be “principles of equity jurisprudence,” not the
geographical limitations associated with Rule 23.

[90] There is a third
problem with Berger’s proposal, though it is a small issue compared with the
others. In a few, rare cases the circuit-border rule could actually be more
restrictive than Bray’s proposal because some cases would necessarily involve
issues that cross circuit lines. For example, environmental litigation
concerning Great Smoky Mountains National Park would have to be conducted in
two parallel proceedings, because the park is partially in Tennessee and North
Carolina, which are in the 6th and 4th Circuits, respectively. See Great
Smoky Mountains National Park, https://www.nps.gov/grsm/index.htm (last
visited Jan. 2nd, 2018).

[91] See Morley, supra note 26
at 549-550.

[92] See Maureen Carroll, Aggregation
for Me, but Not for Thee: The Rise of Common Claims in Non-Class Litigation,
36 Cardozo L. Rev. 2017, 2017
(2015).

[93] See Walker, supra note 74
at 1144-49.

[94] See Michelle R. Slack, Separation
of Powers and Second Opinions: Protecting the Government’s Role in Developing
the Law by Limiting Nationwide Class Actions against the Federal Government,
31 Rev. Litig. 943, 947

(2012).

[95] However, Slack’s idea
appears to conflict with Morley and Carrols’, since a court cannot
simultaneously encourage or require plaintiffs to certify as a class and employ
a presumption against certifying classes where the government is a defendant.

[96] This note leaves to one
side the discussion of whether balancing of the equities or tailoring the
remedy are more appropriate approaches. While there may be some instances where
Congress has clearly stated its intent about the proper scope of a remedy, see supra note 42 (comaring cases
discussing whether the APA requires a nationwide injunction against illegal
agency actions), with constitutional remedies and many statutory schemes, a
statute is largely silent on the issue of remedies. For a comparison of balancing
of the equities with tailoring the remedy, see David Schoenbrod, The Measure of an Injunction: A Principle to
Replace Balancing the Equities and Tailoring the Remedy, 72 Minn. L. Rev. 627, 637-47 (1988).

[97] See supra Section I.A.-B.

[98] See Walker, supra note 60
at 1144-1152.

[99] See Younger v. Harris, 401 U.S. 37, 44-45 (1971) (“[Our Federalism]
represent[s] is a system in which there is sensitivity to the legitimate
interests of both State and National Governments, and in which the National
Government, anxious though it may be to vindicate and protect federal rights
and federal interests, always endeavors to do so in ways that will not unduly
interfere with the legitimate activities of the States. It should never be
forgotten that this slogan, ‘Our Federalism,’ born in the early struggling days
of our Union of States, occupies a highly important place in our Nation’s
history and its future.”).

[100] 505 U.S. 144 (1992)
(holding that a portion of the Low-Level Radioactive Waste Policy Amendments
Act of 1985 was invalid on the grounds that violated principals of federalism
by unfairly coercing states into taking title to radioactive waste).

[101] 521 U.S. 898 (1997)
(holding that provisions of the Brady Act violated principles of federalism by
compelling state officers to enforce federal law).

[102] See Railroad Commission v. Pullman Co., 312 U.S. 496 (1941)
(holding that federal courts may stay a claim until a state’s supreme court has
a chance to review the constitutionality of the act itself).

[103] See Colorado River Water Conservation District v. United States,
424 U.S. 800 (1976) (holding that abstention is appropriate in certain
circumstances where there is parallel litigation in the state and federal
courts).

[104] See Rooker v. Fidelity Trust Co., 263 U.S. 413 (1923) (holding that
the federal courts are not, absent direction from Congress, to sit in review of
state court decisions); District of Columbia Court of Appeals v. Feldman, 460
U.S. 462 (1983) (same).

[105] See Walker, supra note 74
at 1145 (first factor).

[106] 464 U.S. 154, 162–63
(1984). See supra notes 44-55 and
accompanying text.

[107] See supra Section II.b.vi.

[108] See Walker, supra note 74
at 1148.

[109] See id. (“Courts tend to show a high degree of deference to law
enforcement agencies, and this deference seems to influence the courts’
decisions to narrow the scope of injunctions that might otherwise be
acceptable.”).

[110] See O’Lone v. Estate
of Shabazz, 482 U.S. 342, 349 (1987) (“To ensure that courts afford appropriate
deference to prison officials, we have determined that prison regulations
alleged to infringe constitutional rights are judged under a “reasonableness”
test less restrictive than that ordinarily applied to alleged infringements of
fundamental constitutional rights.”).

[111] See Walker, supra note 74
at 1146 (“The fourth consideration that courts should take into account is

the
nature of the right being vindicated.”).

[112] See supra Section III.a.

[113] See Walker, supra note 74
at 1146.

[114] See Morley, supra note 26
at 551 (“Generally, a court severs the invalid . . . unless: (i) the remaining
sections cannot operate coherently as a law, or (ii) the court concludes that
the entity that enacted the statute or regulation would not have intended for
its remaining sections to

be
enforced without the invalidated portions.”).

[115] See, e.g., City of Chicago v. Sessions, Case No. 17 C 5720, 2017 WL
4572208, at * 4 (N.D. Ill. Oct. 13th, 2017). See also supra text accompanying notes 1417.

[116] 497 U.S. 871 (1990).

[117] See id. at 913 (“In some cases the ‘agency action’ will consist of
a rule of broad applicability; and if the plaintiff prevails . . . . he . .
.  may obtain “programmatic” relief that
affects the rights of parties not before the court.”).

[118] See Walker, supra note 94
at 1147.

[119] See id.

[120] See National Mining Ass’n v. US. Army Corps of Engineers, 145 F.3d
1399, 1410-1411 (D.C. Cir. 1998).

[121] See id. at 1409.

[122] See id. (“Moreover, if persons adversely affected by an agency rule
can seek review in the district court for the District of Columbia, as they
often may, see 28 U.S.C. § 1391(e), our refusal to sustain a broad injunction
is likely merely to generate a flood of duplicative litigation. Even though our
jurisdiction is not exclusive, an injunction issued here only as to the
plaintiff organizations and their members would cause all others affected by
the Tulloch Rule . . . to file separate actions for declaratory relief in this
circuit.”).

[123] See Bray, supra note 10
at 22 (justifying the use of equity in the 19th century to enjoin collection of
illegal municipal taxes on the basis that the relevant classes were small,
representing the type of “micro-polity” characteristic of a bill of peace.).

[124] See Walker, supra note 74
at 1149-51.

[125] See id. at 1150.         

[126]See
supra Section
II.a.i.

[127] See supra text accompanying notes 69-72.

[128] See supra Section II.a.ii.

[129] See supra Section II.b.i.

[130] See supra Section II.b.iii.

[131] See supra II.b.iv.

[132] See Bray, supra note 10
at 61 (“A district court selected through forum-shopping will apply a
relatively indeterminate standard, which will then be leniently reviewed by a
court of appeals . . . .”).

[133] And it is no answer to
say that the factors would still point 2-1 in one direction since they likely
vary based on the circumstances in importance.

[134] There are many
different justifications and accounts for the common law method. See generally R. L. Brilmayer, Judicial Review, Justiciability and the Limits
of the Common Law Method, 57 B.U. L.
Rev. 807 (1977) (discussing the value of justiciability requirements in
preserving a common law method of legal interpretation).

[135] Oliver Wendell Holmes, The Essential Holmes 198 (Richard
Posner ed. 1992).

[136]Protecting the Nation
From Foreign Terrorist Entry Into the United States, Exec. Order No. 13769, 82
Fed. Reg. 8977 (2017).

[137] Protecting the Nation
From Foreign Terrorist Entry Into the United States, Exec. Order No. 13780, 82
Fed. Reg. 13209 (2017).

[138] See Melanie Zanona, Timeline:
Trump travel ban’s road to the Supreme Court, The Hill (Sept. 17, 2017, 8:30 AM),
http://thehill.com/homenews/administration/350932-timeline-trump-travel-bans-road-to-the-supreme-court. 

[139] See Washington v. Trump, No. C17-0141JLR, 2017 WL 462040 (W.D.
Wash. Feb. 3, 2017) (enjoining potions of the travel ban).

[140] See Int’l Refugee Assistance Project v. Trump, 265 F. Supp. 3d 570
(D. Mass. 2017) (issuing nationwide injunction against the president’s revised
travel ban); Hawai’i v. Trump, 245 F.Supp. 3d 1227 (D. Haw. 2017) (enjoining
other parts of the revised travel ban).

[141] See Washington v. Trump, 847 F.3d 1151 (9th Cir. 2017).

[142] See supra Section IV.a.iii.

[143] U.S. Const. art. II, § 3, cl. 5.

[144] See Int’l Refugee Assistance Project v. Trump, 857 F.3d 554, 605
(4th Cir. 2017) (“We recognize that ‘in general, this court has no jurisdiction
of a bill to enjoin the President in the performance of his official duties . .
. .’”) (quoting Franklin v. Massachusetts, 505 U.S. 788, 802–03 (1992)); Washington, 847 F.3d at 788.

[145] See supra note 137.

[146] See Int’l Refugee Assistance Project, 857 F.3d 554; Hawaii v.
Trump, 859 F.3d 741 (9th Cir. 2017).

[147] See Washington, 2017 WL 462040.

[148] See Washington, 847 F.3d 1151.

[149] See Alexander Burns, 2
Federal Judges Rule Against Trump’s Latest Travel Ban, The New York Times (Mar. 15, 2017),
https://nyti.ms/2mJEIb1 (“[The travel ban] would have also . . . limited
refugee admissions to 50,000 people in the current fiscal year. Mr. Obama had
set in motion plans to admit more than twice that number.).

[150] See Int’l Refugee Assistance Project, 857 F.3d at 605 (quoting Santa Fe Indep. Sch. Dist. v.
Doe, 530 U.S. 290, 309 (2000)).

[151] See supra text accompanying note 57.

[152] President Trump took to
Twitter to specifically decry “judge shopping” after the 9th Circuit’s
decision. See Donald J. Trump,
(@realDonaldTrump), Twitter (Apr.
26, 2017, 4:38 PM),
https://twitter.com/realDonaldTrump/status/857182179469774848.

[153] See supra text accompanying notes 142-44.

[154] See Int’l Refugee Assistance Project, 857 F.3d at 605; Hawaii v.
Trump, 859 F.3d 741, 789 (9th Cir. 2017).

[155] See Bray, supra note 10
at 62-63.

[156] Id. at 62.

[157] Memorandum from Janet
Napolitano, Sec’y, Dep’t of Homeland Sec., to David Aguilar, Acting Comm’r,
U.S. Customs and Border Prot., et al. 1 (June 15, 2012) (the “DACA Memo”),
https://www.dhs.gov/xlibrary/assets/s1-exercising-prosecutorial-discretion-individuals-who-came-to-us-as-children.pdf.

[158] Memorandum from Jeh
Johnson, Sec’y, Dep’t of Homeland Sec., to Leon Rodriguez, Dir., USCIS, et al.
3–4 (Nov. 20, 2014),
https://www.dhs.gov/sites/default/files/publications/14_1120_memo_deferred_action.pdf.

[159] See Texas v. United States, 809 F.3d 134, 146 (5th Cir. 2015).

[160] See Texas v. United States, 86 F. Supp. 3d 591, 677-78 (S.D. Tex.
2015).

[161] See Texas, 809 F.3d at 146.

[162] United States v. Texas,
136 S.Ct. 2271 (2016) (mem.).

[163] See Texas, 86 F. Supp. 3d at 638-39, 666-67.

[164] See Santa Fe Indep. Sch. Dist. v. Doe, 530 U.S. 290, 309 (2000).

[165] See supra text accompanying notes 22-25.