For the Balkinization Symposium on The Chevron Doctrine through the Lens of Comparative Law
Vincent Martenet
Diversity
in Context
Questions relating to the relationship between
administrative bodies and courts in statutory interpretation cases arise in several democratic countries, regardless of their legal traditions. They invite us to reflect from a comparative
perspective, without ignoring the constitutional, legal, and judicial context
prevailing in each country, and being fully aware that generalizations
regarding judicial deference should be, if not completely avoided, at least
very carefully crafted. A nuanced approach is actually inevitable.
Judicial deference to administrative interpretation of statutes may arise where three main preconditions are met. First, the statute in question provides a margin of interpretation with respect to the specific issue raised. In other words, it allows at least two admissible or defensible interpretations in light of the applicable methods and canons of construction used by courts and, as the case may be, administrative bodies. The choice by the legislature of ambiguous, vague, or broad legal terms may, for instance, occur because the legislature recognizes the superior expertise of the executive or because ambiguous statutory language resulted from the political negotiations prior to the passage of the statute. Second, the court’s deference must fit within international, constitutional, as well as statutory or other constraints and, third, the court’s deference must somehow be willed or accepted by the legislature. In other words, the latter has control over this issue, as it can open or close the door to judicial deference to administrative statutory interpretation. Its intent or assent is, however, often at best implicit, which means that courts have to determine whether and, if so, to what extent they should or can defer to administrative bodies.
Standard features of modern democracies have a
contested relationship to judicial deference. In some countries, administrative
legitimacy and the principle of separation of powers, as well as the optimal
allocation of limited resources are cited to justify judicial deference to
administrative interpretation of statutes. Humility, honesty, and transparency
from the courts, themselves, may also explain some forms of judicial deference.
In contrast, elsewhere, the separation of powers, the concept of checks and
balances, the rule of law, the notion of État de droit, the
constitutional role of the judiciary, as well as procedural safeguards support
the case against judicial deference
to administrative statutory interpretation. Agency independence, when it
exists, and the risk of wrong incentives, as well as other considerations, may
also support the case for or against judicial deference.
Comparative studies in this area
suggest that the supreme courts of Canada and the United States have gone quite
far in terms of judicial deference to administrative interpretation of
statutes. A presumption of deference somehow exists in Canada, even though its
scope is significantly reduced, and the case law is not always straightforward.
In the United States, the Chevron
doctrine, even with its subsequent limitations, is rather far reaching. However,
the procedural requirement of the U.S. Administrative Procedure Act for
rulemaking, arguably, impose recognized limits on executive branch discretion. From
a principled perspective, at least, no other country seems to have ventured so
far down the deference path on questions of legal interpretation. Strong legal
and contextual reasons may explain and justify some landmark cases in both
countries. Nevertheless, does a presumption of deference adequately reflect the
role of the courts in an État de droit
when it applies to questions of law? Do U.S. or Canadian courts infer too much,
institutionally speaking, from the ambiguity of statutes? In short, should a
more nuanced approach be preferred? Secondly, procedural safeguards are deemed,
in several countries, to require full judicial review of the interpretation of
the law and its application to the facts. In other countries, they do not
exclude judicial deference to administrative interpretation of statutes. Do the
doctrines developed in Canada and the United States give sufficient weight to fundamental
procedural safeguards? This question is delicate and cannot probably be
answered in abstracto. Arguably, it
should instead be examined on a case-by-case basis or, at least, on a
procedure-by-procedure basis.
Expertise
All things considered, an important, and perhaps the most convincing,
justification for judicial deference to administrative interpretation of
statutes lies in the asymmetry of non-legal expertise.
In certain countries at least, such
a deference may be regarded as a pragmatic and realistic way of allocating
limited resources and of dealing with the asymmetry of non-legal expertise
among administrative bodies and courts, provided that it is framed by several conditions.
When these conditions are met, the separation of powers principle is not
endangered, and judicial deference in statutory interpretation cases may be
described as oil in the complex machinery of checks and balances. In any event,
the legislature retains the final say, at least ex post and pro futuro.
As statutes rarely deal with the issue of
judicial deference to their interpretation, courts themselves must determine
whether such deference is permitted or forbidden, generally or in certain cases
only. The constitutional, legal, and judicial context prevailing in each
country is of great significance in this regard. It may nevertheless provide
courts with little, if any, guidance on the specific issue of deference to
administrative statutory interpretation and leave them helpless. In this
respect, courts may eventually consider adopting a nuanced approach and
applying all or part of the following test: When (i) in light of the
applicable methods and canons of construction, a statute allows a margin of
interpretation, (ii) the administrative interpretation of the statute
remains within this margin, and (iii) the applicable international,
constitutional, statutory, or other constraints permit or, at least, do not exclude judicial deference either generally
or in the case at hand, then courts may or, depending on the country, must
defer to the administrative interpretation of the statute, especially when or,
depending on the country, provided that (iv) this
interpretation requires non-legal—scientific, technical, or policy—expertise,
(v) the administrative body enjoys an asymmetry of such expertise as
compared to courts, and (vi) the legislature was or should have been aware
of both this necessity and this asymmetry. Depending on how condition (iii) is
interpreted in a given country, courts could disregard condition (vi). The
level of courts’ expertise in the relevant subject matter should in principle
be determined after taking account of the measures available to judges,
including the appointment of experts.
Rulemaking vs. Adjudication
Various features in administrative law systems
may significantly affect debates about deference. The difference between review
in specialized administrative courts or tribunals versus general jurisdiction
courts may influence approaches to deference. The possibilities for a court to
appoint experts may play a role in this respect. Furthermore, the processes by
which statutes are drafted and the way—precise or vague, for instance—in which
they are written, may affect the salience of the issue. The breadth of
delegation to agencies or other administrative bodies of powers to adjudicate
disputes in specific areas with the force of law may considerably vary from one
country to another. Finally, the nature of the executive action
under review could also play a role. Deference may especially be justified
above all for policies formulated in a rulemaking process, but less so for
case-by-case adjudications without issuing general rules.
This
last distinction is usually not made in European countries where deference exists.
Three comments come to mind at this point. First, rulemaking—and not just some interpretive authority—is
delegated when an administrative agency issues rules (see, for instance,
Articles 55(3), 56(3) & 57(3) CISA in Switzerland). The reviewing courts then perform a narrower
task, notably to check whether the rules do not violate the statute at stake. They
are in principle bound by these rules, at least in the context of European
continental law. Or course, some rules are just interpretations of the relevant
statute, but many others are policy decisions. To an important extent, the
debate takes place at another level and in different terms. Second, notice-and-comments
procedures in the rulemaking context may provide agencies with additional
expertise coming, so to speak, from the outside, as emphasized by K. Hickman
& A. Nielson on page 966 of their article. Inside
expertise should however not be disregarded, and this expertise can be used in
adjudications. Third, rulemaking through notice-and-comments procedure may open
the door to lobbying and the risk of agency capture. This may justify closer
scrutiny by judges. In the final analysis, the strict limitation of judicial deference to rulemaking may prove
overly schematic.
Illustrations
The term “dangerous substances” used in a
statute needs to be interpreted. This task very likely requires having deep
theoretical and empirical knowledge of chemistry, biology, and other scientific
fields, as well as of the various factual situations in which the application
of the provision in question can be considered. The distinction between law and
fact definitely comprises a grey zone, not least because some issues raise
mixed questions of law and fact. Indeed, the prohibition on “dangerous
substances” raises countless factual issues, but the relevant administrative
body must determine criteria to evaluate a substance’s danger and set limits
beyond which a prohibition is justified. In other words, it must not only
determine whether some substances are dangerous, but also—at least to a certain
extent—define what “dangerous” means. By doing so, it interprets or constructs
the relevant statute. Inside or outside non-legal expertise seems relevant in
that case, irrespective whether adjudication or rulemaking is at stake.
The asymmetry of non-legal expertise between
administrative bodies and courts must be relevant to the disputed
interpretation of the statute, especially when a prediction, an appraisal, or
an assessment must be made. For instance, the word “carcinogenic” may allow a
margin of interpretation (when and at what level can a substance or a product
be qualified as such?) and its interpretation through rulemaking or
adjudication may require non-legal expertise as well as the making of
predictions. By contrast, a broad and open-ended word used in a statute such as
“reasonable” or “appropriate” can, but should not necessarily be considered as
ambiguous or unclear. In the context of the statute in its entirety and on the
basis of the applicable methods and canons of construction, the distinction
between reasonable and unreasonable may indeed be clear or require no
particular non-legal expertise.
This kind of illustrations could also help guide the reflection on judicial deference to administrative interpretation of statutes. Some micro-comparison on a specific issue would indeed be extremely interesting and very illuminating. At the end of the day, it is possible that courts which have developed a doctrine on deference defer less than they are supposed to, according to their own doctrine, and that courts which have explicitly rejected any idea of deference with regard to the interpretation of the law actually defer consciously or unconsciously. One of the great merits of a doctrine, such as the ones existing in Canada, Germany, and the United States, is to raise an issue and to open or frame a debate in a transparent and even democratic way, as the legislature can always reclaim the field. In short, transparency may open appropriate—and possibly fruitful—inter-branch dialogue.
Vincent
Martenet is Full Professor of
Swiss and comparative constitutional law as well as competition law at the
University of Lausanne, Switzerland where he is Dean of the Faculty of Law,
Criminal Justice and Public Administration. He is also Deputy Justice, Swiss
Federal Tribunal, the Supreme Court of Switzerland. vincent.martenet@unil.ch