For the Balkinization Symposium on The Chevron Doctrine through the Lens of Comparative Law
Susan Rose-Ackerman & Oren Tamir
Over
the past two weeks, the Balkinization blog hosted a wide-ranging symposium
about judicial deference to the statutory interpretations of administrative
agencies (or the executive more broadly) around the world. As we explained in
our opening essay, the immediate trigger for this
symposium was the forthcoming Loper case that has put Chevron on the Court’s agenda in the
current term, and Justice Gorsuch’s dissent from the denial of cert in Buffington where he invoked comparative law to critique Chevron
deference. However, our goal in organizing the symposium stretches beyond the fate
of Chevron deference in U.S. public law. We believe there is merit in
more broadly exploring what the United States can learn from comparative
administrative law in general. Moreover, we want to ask questions about
the responsibilities of those who wish to engage in comparative public law to
shed light on their own systems of law (as we both think they should).
***
The first message, reflected powerfully
in all the contributions, is that every modern democracy requires some
combination of administrative bodies and courts to resolve complex questions of
statutory interpretation. But which institution has the last word? If administrative
interpretations are subject to judicial review, several countries’ judiciaries
assert that their courts do not (and should not) explicitly defer to the
executive. Thus, Carols Ari Sundfeld and Yasser Gabriel show that in Brazil courts do not
recognize a place for general deference to the administration. Liz Fisher confirms the same about England, by
saying that its judges “do not talk of deference in the statutory construction
context” but rather announce that “questions of law are questions for the
court.” In Australia, Janina Boughey claims that there is no explicit
judicial deference to the executive. In fact, Australia’s High Court stated
that deference on interpretive questions was nothing less than an “abdication
of judicial responsibility” (in a case considering Chevron in obiter). As
Boughey further mentions, Australian judges would even get seriously angry if
one asked them straightforwardly to defer. In India, explicit talk of judicial
deference to executive interpretation is similarly absent and would be
considered an “absurdity,” as Raeesa Vakeel explains. And this is the status quo as
well in France as Duncan Fairgrieve and François Lichère show. In Germany, although some
statutes provide a “margin of appreciation” (or an “assessment prerogative”), Jasper
Kamradt and Matthias Ruffert claim that this leeway is available only
in a few “pockets” of German administrative law. They also suggest that the
level of judicial control over the interpretation of delegations of power to
administrative agencies in Germany is exceptional even by European standards.
Even
where the contributions highlight that formal deference to the executive’s
interpretations of law is explicitly recognized as a doctrinal matter, they
also show that this form of deference operates very differently than it does under
Chevron. For instance, in Canada, as Paul Daly powerfully shows, deference is recognized in many
areas of the law and was entrenched and affirmed in a recent canonical judgment
of the Canadian Supreme Court (called Vavilov). But, the Canadian
approach depends not so much on the ambiguity or vagueness of statutory terms.
Rather, it has to do with the overall quality of the reasoning agencies use
to justify their actions (as well as the quality of the processes that led them
to their decisions). Chevron deference, of course, does depend upon the
“reasonableness” of the agency’s rule but, as interpreted by courts today, it is
not necessarily or even primarily geared toward that the quality of the
reasoning (on the unique features of review based on the quality of
reason-giving, see especially the work of Jerry Mashaw). Similarly, in South Africa, as
Geo Quinot details in his own interesting post, there is also wide and explicit
recognition of the need for judicial deference to the executive’s
interpretation. But, as Quinot is emphasizes, that recognition is not tied to vagueness
or ambiguity in statutory language as in Chevron. Rather, it is linked
to South Africa’s aspirations to “transformative constitutionalism”—that is, it depends on how much
the executive’s proposed interpretation substantively advances the project of attaining
a more just and equal post-apartheid society that is at the core of South
African public law. Everywhere, but especially in South Africa, the protection
of constitutional rights necessarily frames the meaning of statutes.
With
this background, it is perhaps understandable why one who looks abroad would
start questioning whether Chevron makes sense. If we stand alone in
deferring to agencies’ interpretations under Chevron, are we really right
to do so? And, indeed, this is exactly the kind of comparative intuition that
Justice Gorsuch relied on in his opinion in Buffington that invoked
global practices. He said that the lack of comparative precedent weakens Chevron’s
foundations in and of itself.
But,
the contributions to this symposium clearly illustrate that the asymmetry that
Justice Gorsuch has identified dissipates very quickly once one takes a more
serious look. And, as these contributions suggest, if one wants to engage seriously
in comparative law, there is simply no substitute for taking a deeper look.
Part
of our critique of the shallowness of the reasoning underlying Justice
Gorsuch’s invocation of comparative law should already be evident from our
references to the posts concerning Canada and South Africa. In the former, the
Supreme Court largely insists on the quality of the government’s reason-giving
as a condition for deference rather than on ambiguity or vagueness per se. In the latter, deference is
buttressed by a keen judicial emphasis on assuring that interpretive decisions align
with the country’s (“transformative”) constitutional principles. Chevron
can’t really be equated with the deferential regimes in polities transitioning
from an authoritarian past.
But
the fact that the asymmetry isn’t really all that it appears to be at first
sight also applies with similar force to jurisdictions where deference isn’t
explicitly recognized. This is so because those systems differ from the United
States along six crucial dimensions:
- The
way systems apply or even understand the distinction between “law” and “policy
or politics,” and the range of justiciable issues that come before the courts,
- Nations’
divergent constitutional structures, including the parliamentary/presidential
divide and nations’ distinctive ways of organizing their judiciaries,
- The
existence of differences in both the scope and substance of constitutional
protections that nations provide for positive socioeconomic rights,
- Constitutional
norms or statutory provisions that mandate the procedures for promulgating
administrative rules with legal force,
- Judicial
protocols for interpreting statutory language that differs across
jurisdictions, and finally
- To
differences in countries’ politics and culture, broadly understood.
Start
with the first dimension. As the contributions to the symposium show, several
countries’ judiciaries categorically assign legal interpretation to themselves
because, in contrast to the United States, they draw a much sharper distinction
between law, on the one hand, and “policy” or “politics,” on the other.
Interpreting the former is the exclusive responsibility of the courts; the latter
is outside of their purview. Yet, many administrative law issues, in practice, are
a mixture of law, fact, and policy. Thus, courts can insist both that they do
not defer to administrative interpretation of statutory terms and at the same
time accept the way the government implements the law. In other words, the
courts defer to the legislature’s decision to use statutory terms that are open
to interpretation by government officials.
Second, the government’s constitutional
structure provides a set of key background conditions. Many of the countries
represented in this symposium (England, Canada, Germany, South Africa, India,
and Australia) are parliamentary systems. But, as Boughey reminds us in
her contribution, in parliamentary systems the executive and legislature are “fused”
rather than tightly separated. And this matters greatly for the question of
deference and why it doesn’t appear to the same extent in these other systems.
For one thing, this fusion makes it easier for the government to respond to
erroneous judicial interpretations (reached without deference) by legislative
fixes. In a presidential system, such as the U.S., such fixes are much more
difficult, making judicial deference more justified. For another, judicial
review of administration is potentially more important in parliamentary systems
than it is in presidential ones in another key way. It “compensates” for the relative
lack of institutional contestation in the legislative process (a point that
Boughey emphasizes in her contribution when she mentions that in parliamentary
systems the government controls much of the legislative agenda and writes most
of the statutes).
France and Brazil are, of course, presidential systems.
But they differ from the United States in important respects, which also help
explain why something like Chevron has not emerged there. France has
both an elected president and a prime minister appointed by the president and
confirmed by the parliament. Unlike parliamentary systems, the strong president
stays in office even if his party or coalition is subject to a vote of no
confidence. In Brazil, the president has independent power similar to the U.S.
president. However, unlike the U.S., the legislatures do not provide a powerful
counterweight in either country; thus, the courts serve as important additional
constraints. As in the parliamentary cases, neither Brazil or France has an APA
that requires notice and comment procedures for executive rulemaking. Here, the
explanation is not the unity of executive and the legislature, but the weakness
of parliament and its internal divisions. Brazil, as Ari Sundfeld and Gabriel
note, has been influenced by French public law, but it is now increasingly
copying developments in U.S. public law, including the creation of independent
agencies, including deference to their policy choices. Nevertheless, the fact
remains that the Brazilian legislature is divided into a multitude of parties
and is not a strong check on the president.
The
presidential v. parliamentary divide is not the only structural constitutional
difference of interest, however. Some of the countries represented have a judicial
system to adjudicate administrative law disputes that differs from the
unitary U.S. system. In France, as Fairgrieve and Lichère mention, and as Vincent
Martenet moreover highlights, there is strong separation
between administrative law adjudication and adjudication in other fields of
law. In France and other polities worldwide, administrative law disputes come
before administrative courts headed by a Conseil d’Etat that applies principles
of administrative law. In Germany, as Kamradt and Ruffert emphasize,
administrative law disputes also are heard within a separate system of administrative
courts. This difference in judicial organization and perception can also help to
explain distinctive features of public law in France or in Germany, including
why deference Chevron style hasn’t emerged there. Such courts have a
much higher level of systematic specialization in adjudicating administrative
law disputes than generalist courts in the common law tradition have—making the
need for deference in such systems to be much less called for.
The third axis of
difference is a
constitutional text that provides for the protection and enforcement of positive
socioeconomic rights. These rights are very clearly a feature of the Brazilian
and South African constitutions, promulgated as part of a democratic transition
from an authoritarian system (military rule in Brazil and apartheid in South
Africa). These documents include provisions protecting a right to public
health, education, and social services (among other things). Germany fits into
this category, as well. Its constitution, the Grundgesetz, has been in place since soon after the end of World
War II and begins with a long catalogue of rights some of which are positive or
protective in nature and thus require substantial public spending to bring to
life. The reality of strong positive constitutional rights, which the U.S.
currently lacks (certainly at the federal level), also provides important context
that can explain why the emergence of interpretive deference has not emerged. In
countries committed to positive rights, legal cases seeking to assure these
rights have been a major source of judicial activism. Lacking other routes into
the courts (or to influence the government more broadly), the plaintiffs are
mostly individuals or specific communities seeking benefits. These cases inject
the judiciary into national budgetary debates. In some cases, the courts have
shown little deference to government spending decisions on social services,
seeing themselves as primarily in charge of enforcing, sometimes strongly,
these positive elements of constitutionalism.
This
recruitment of courts to the cause of positive rights and positive
constitutionalism more broadly also suggests that the lack of Chevron deference
would have quite different consequences in other countries compared with the
US. As we have suggested, in other countries, the lack of Chevron allows
courts to intervene to encourage more governmental protection and activity. It
creates the possibility for more governmental regulation and interventionism—which
is exactly what socioeconomic rights are in large part about; how government’s
are not doing enough. So, for example, as Ari and Gabriel argue, Brazilian
courts have rejected deference in large part to further the protection of
socio-economic rights to social benefits or health. As Quinot assesses the South
African case, courts there have sidestepped deference especially in domains
where courts have felt that the administration could do more to advance South
Africa’s transformative constitutionalist goals—such as eradicating social and
material inequality. And as Vakil mentions, Indian courts have eschewed
deference in large part to engage in a positive project of “good governance.”
In the United States, getting rid of Chevron deference is not likely to
have the same effect. Rather than leading courts to engage in more proactive
enforcement of positive obligations on government to act and provide services for
better health, protection of the environment, and other rights for material
equality, the elimination of Chevron would likely trigger judicial efforts
to restrict government regulation.
Fourth, and
to return to an issue we alluded to earlier, the jurisdictions represented here diverge in the procedures
used by the public administration when it makes policy under delegated
authority. Indeed, although there appears to be a global trend toward the
enactment of framework statutes such as the APA, not all countries have jumped
on the bandwagon. As Vakil illustrates, India is an important example of a
country that lacks a framework APA-style statute, though there are others,
including countries represented in this symposium (e.g., England and
Australia). But, importantly, even those countries that did enact framework administrative
law statutes, have not necessarily endorsed the same procedures relied on in
the US: namely, notice-and-comment rulemaking. For example, as Quinot shows
with respect to South Africa (a country that has a framework administrative law
statute known as PAJA), and as Martenet discusses broadly with respect to both Europe
and Canada, the U.S. notice-and-comment procedure is not systematically
required or widely used elsewhere. Most general policies are developed through
political processes, which are largely internal to
the executive branch and sometimes subject to legislative scrutiny. When other
countries outside the U.S. draw on something like a public notice-and-comment
process, they do so voluntarily or when legally required by particular,
domain-specific, legislation. We think that the absence of wide
notice-and-comment processes clearly influences the debate over Chevron:
namely, it seems to make much more sense to reject Chevron in a system
that doesn’t rely on notice-and-comment procedures compared to a system that
does. If notice-and-comment does not exist, more robust, non-deferential
judicial review on questions of statutory interpretation compensates somewhat for
the lack of public participation and reason-giving in the policymaking process.
Fifth, another difference that the various
contributions to the symposium highlight concerns the methods that judiciaries use
to interpret statutes. Chevron works quite well with the more textualist
tendencies that have been increasingly embraced by U.S. courts, given its
two-step framework and the way that deference is keyed to questions about whether
or not statutory ambiguity exists in a way that gives interpretive leeway to
the administration. But, other countries do not embrace anything similar to our
own increasingly invoked textualism. Instead of proudly announcing that “we’re all textualists today,” other jurisdictions are much
more eclectic in the way they interpret statutes. Fisher, for example,
forcefully shows how courts in both England, with no written constitution, and
Australia, with one, are not woodenly textualist but take a much more
contextual approach that defies crisp theorization or doctrinal formulation.
And Daly in his discussion of Canadian law also mentions that the judiciary
looks beyond the text and refers quite openly to context and purpose. This also
helps explain why we have Chevron and others do not.
Finally, and most broadly, some of the
contributions emphasize differences between the U.S. and the other
jurisdictions anchored in overall culture or politics. For instance, in his
discussion of Canada, Daly highlights how American hostility to public administration
is utterly foreign in Canada. As he says, in Canada there are “few who question
the legitimacy of the administrative state.” And arguments that may appear bizarre
(“off the wall,” as Jack Balkin calls them) to our own ears according to which a
robust public administration is constitutionally mandatory are quite forceful in Canada
(and likely also in the countries of most of the other contributors). A similar
political and cultural embrace of administrative governance through law is also
implicit in Boughey’s contribution when she says that the lack of deference in
Australia has not “rendered the Australian administrative state incapable of
performing its function.” Both France and Germany, similarly, have strong civil
service traditions that differ in important ways but unite in giving
considerable respect to the professional public administration.
To
summarize our argument so far, it is a mistake to conclude that the reality of
no Chevron abroad implies something meaningful about the relationship
between administrative rulemaking and statutory interpretation in the U.S. Although
the U.S. is in some sense an outlier, the contributions vividly show that many factors
help to account for the divergence, as well as to justify it. These factors include
constitutional structure, the organization of the judiciary, differences in
administrative procedures, differences in interpretive methodologies, the reality
of positive constitutional rights abroad, and, most broadly, differences in
history, culture, and politics that exist between jurisdictions.
***
But
we actually think that the symposium helps show that the flaw in invoking
comparative law to undermine Chevron is meaningful in yet another way. As
the contributions demonstrate, even if Chevron does not have a precise
analogy elsewhere, all judiciaries inevitably defer to the administration on
some of their exercises of discretion. This is, of course, most clear in Canada
and South Africa, which, as Daly and Quinot show, accept the idea of
interpretive deference. But, as other contributions illustrate, this is also
true for the countries that reject deference officially. For example, in some
jurisdictions, deference enters administrative law under a different heading, as
in Australia under the label of “judicial restraint”. In France, Fairgrieve and
Lichère address the reality that, at least in technically complex matters,
deference is recognized by the administrative courts. In Germany, Kamradt and Ruffert
show that the Federal Constitutional Court has accepted deference beyond the
traditional areas where the doctrine of “margin of appreciation”
applies—namely, in cases of what they describe as “knowledge gaps.”
Occasionally,
the symposium contributors argue that deference exists in their legal system, but
more sub rosa. Thus, Vakil shows that, in spite of formal claims that
judges in India have full control over interpretive matters, in practice, they
provide strong de facto deference in technically complex matters. They do this
through the flexible use of doctrinal standards that conventionally imply more
rigorous judicial scrutiny in some cases and various evasive strategies (and
sometimes political ones) in others to avoid deciding contentious cases that
may be beyond the competence of judges. Similarly, in Brazil, Ari Sundfeld and
Gabriel suggest that outside of the domain of socioeconomic rights, Brazilian
courts, in practice, are quite deferential notwithstanding rhetoric to the
contrary.
If
this is not enough, the contributions to the symposium also emphasize that some
countries may be considering expanding their deference regime, coming
closer to a more explicit understanding that administrative agencies’ actions
deserve greater respect than they currently receive. Sometimes this deferential
expansion is expressed judicially, as in the case of Germany as Kamradt and
Ruffert’s discussion of a recent Federal Constitutional Court case powerfully demonstrates.
And Canada’s recent pronouncement of deference, in Vavilov, importantly
strengthened the presumption of deference that existed before. However, sometimes
the need for expanded deference is expressed in legal scholarship, not in
judicial judgments. So, for instance, Boughey mentions that there are influential
academic calls for Australian courts to accept more deference (or judicial
restraint) in some types of cases (mainly those that involve strong reliance
interests). Martenet’s contribution in this symposium could be understood in
itself as a standalone academic intervention which calls for more
deference by judiciaries in countries that have so far been reluctant to accept
it, such as France. Indeed, Martenet defends in these systems what he calls a
“contextual” or “nuanced” approach that more directly and openly appreciates
the place of the executive’s superior expertise (possibly tailored to the
processes used by agencies, including by distinguishing between rulemaking and
adjudication).
In
short, by invoking comparative law to undermine Chevron, Justice Gorsuch
got it doubly wrong: first, by ignoring the fact that significant cross-national
differences can account for the lack of Chevron abroad and help to explain
why we have it here. And, second, by ignoring the fact that the subject of judicial
deference to statutory interpretations is very much alive in judicial systems that
are working to expand it in important ways rather than shrinking it. These
developments indicate that deference has strong reasons supporting it, and
illustrate powerfully how, given the limited capacity of courts, it is, in
practice, unavoidable. Indeed, as we tried to summarize here, the judiciaries
in all eight countries ultimately give considerable weight to administrative
decisions that interpret statutory language to make policy, within “reasonable”
bounds, even if they do so much more implicitly.
***
Our
conclusion is that, at a minimum, it is irresponsible to use comparative law
and the lack of global analogies to Chevron to simply undermine it. This
is the negative lesson that comparative law can teach us. At the same time, we also
want to make clear that comparative law has a more positive implications; that
is, we believe that it can teach us more concrete lessons about the future of Chevron
and the relationship between courts and executive policymaking. These lessons
might be of use to the Court as it reconsiders Chevron deference in Loper.
Or, if not for the Court itself, then at least it might be of use to the
broader administrative law community.
We
see two possibilities. One, contrary to Justice Gorsuch’s suggestion in Buffington,
is that comparative law can support efforts to retain Chevron,
not to get rid of it or to cut it back. Indeed, as the discussion up to this
point suggests, comparative law implies that Chevron has firm
justifications in the U.S. constitutional structure (presidentialism); in the
reality of a generalist court system that adjudicates administrative law
disputes in an enormous array of domains of governmental regulation; in the
existence of the APA and the importance of notice-and-comment rulemaking as a central
vehicle for executive branch rulemaking; in the formalistic (and textualist)
tendencies in our law, that the Court itself breeds and maintains today; and in
the lack of firm constitutional protections for positive rights. Furthermore,
the judiciary itself needs to recognize its own limited ability and legitimacy
as the reviewer of administrative policymaking choices. This is especially so
given that SCOTUS is changing other tenets of administrative law to
further shrink the space for administration—including especially by the rise of
an expansive “major questions” doctrine and the possible revival of the
non-delegation doctrine.
We
do not think, however, that a defense of the status quo is the only possible future
for U.S administrative law. Even from a point of view such as ours which is
highly sympathetic to the administrative state and to a positive role for
governmental regulation, Chevron itself, however interpreted, is not necessarily
ideal. Indeed, we think there is something valuable in the approach to
deference seen, for example, in Canada with its emphasis on reason-giving by
the government rather than a more formalistic search for ambiguity. In other
words, U.S. courts could go beyond the reason-giving required for APA
rulemaking and examine the reasons given by other sorts of executive
policymaking, and get rid of the often opaque divide between law and politics
that Chevron deference now is understood to depend on. Here, the U.S.
precedent that most clearly supports this move is, surprisingly perhaps, the
Supreme Court case preventing President Trump from
canceling the Deferred Action for Childhood Arrivals (DACA). The opinion cited
the beneficiaries’ reliance interest as a reason to strike down DACA’s repeal that
occurred without procedures to took account of that reliance. Similarly, Canadian
law recognizes the importance of deference, but, at the same time, also permits
courts to evaluate the quality of reasons or processes behind agencies’ interpretive
choices, and to send agencies back to the drawing board if these reasons are
unconvincing in the context at hand. In that way, the Canadian approach avoids
the formalistic two-step process or an excessive focus on whether or not
statutes are ambiguous or vague. That kind of regime, expressed in the DACA
case and more evident in Canadian practice, strikes us as a worthy alternative to
the current one created by Chevron.
***
We
end our concluding reflection on the fate of Chevron and comparative
administrative law with thanks to Jack Balkin for agreeing to host this
symposium on Balkinization. We could not think of a better platform to amplify
the importance of comparative administrative law and to correct incidents of its
misuse. And, of course, we thank as well our wonderful contributors, whose
contributions here—and whose work more broadly—illustrate both how challenging
it is to do good and responsible comparative law, but also how much we can
learn from taking that challenge seriously.