For the Balkinization Symposium on The Chevron Doctrine through the Lens of Comparative Law
Paul Daly
I am a committed comparativist — in my 2021 monograph on judicial review of
administrative action, I developed and applied a unified interpretive framework
to the administrative law of Australia, Canada, England, Ireland and New
Zealand; a decade ago I wrote a book on deference that drew on American,
Canadian and English jurisprudence. And yet today, as far as judicial review of
administrative action is concerned, I have the distinct impression that the
United States and Canada are two countries divided by a common concept —
deference. Although there is much in the Supreme Court of Canada’s recent root-and-branch reform of
Canadian administrative law in the 2019 Vavilov decision that could illuminate the contemporary
challenges for American administrative law, it seems doubtful to me that the
SCOTUS will reach across the divide in its (re)consideration of Chevron deference
in Loper Bright.
There are two aspects to this divide, which map onto steps one and two of Chevron, relating respectively to statutory interpretation and the methodology of deferential judicial review.
At Chevron step one, statutory language
plays an important role in determining whether there is space for an
administrative agency to act. Ambiguity is a pre-requisite. But Americans and
Canadians think very differently about statutory interpretation. For
a generation now, it has been accepted that statutory
interpretation in Canada involves the examination of statutory text, purpose
and context. The use of purpose and context are
debated, as one would expect, but it is
nonetheless uncontroversial that statutory context can include relevant international law
instruments and soft law
and may extend to the values
underpinning the Charter of Rights and Freedoms
and the constitutional
imperative of reconciliation with Indigenous peoples.
Notably, resort
to contextual considerations in statutory interpretation in Canada is not
premised on the existence of statutory ambiguity. The dominant Canadian view is
that there is no
such thing as ‘clear’ statutory text which can be established by consideration
of a limited range of interpretive tools. There are no
distinct ‘steps’ in statutory interpretation but rather a holistic analysis of
statutory language in light of the purpose of the relevant provisions, their
place in the statutory scheme and other contextual considerations (such as
common law principles, relevant international law commitments and so on).
In previous iterations of the Supreme Court
of Canada’s approach to judicial review of administrative action, this holistic
treatment of statutory language was a central feature. Treatment, that is, of
the legislative intent revealed by the entirety of a statutory scheme rather
than one provision read in isolation (a feature that puzzled Professor Solum). The role of
legislative intent in the reformulated approach in Vavilov is difficult to summarize but is similarly
holistic. A delegation of authority triggers a presumption of deferential review and in conducting
this review, the relative breadth of the relevant
statutory provisions setting out the decision-maker’s powers are a
contextual consideration in assessing the reasonableness of the decision, to be
taken into account alongside a wide range of other legal and factual
constraints.
At Chevron step two, the question
becomes whether the agency has offered a permissible construction of the
statute. Consistent with the American commitment to “reasoned administration”, administrative
agencies generally give extensive reasons in support of their preferred
interpretation — the “reasoned explanation” that the court below
relied on to uphold the rule at issue in Loper Bright ran to some 17 pages with three
columns of text each. But the treatment of reasons north and south
of the 49th parallel is diametrically opposed. Unlike their American
counterparts, Canadian courts are not to conduct their own independent analysis of the statute
and use that as a yardstick against which to measure the decision-maker’s
interpretation. Whereas an American judge begins by looking at statutory text, a
judge in Canada must first look to the reasons offered by the agency. As the
Supreme Court of Canada put it in Vavilov, “A principled approach
to reasonableness review is one which puts … reasons first”.
For
this reason, the Court could plausibly say in Vavilov that “reasoned
decision-making is the lynchpin of institutional legitimacy”. This might be true in
a general sense of American administrative law: Jerry Mashaw has explained how “reasoned administration” is a
significant feature on the American legal landscape. Nonetheless, in the specific area of judicial
review, the Canadian “reasons-first” approach is markedly different to the
“statute-first” methodology in the United States. It is almost as if the two
steps of Chevron are reversed.
Furthermore, in Vavilov, the Supreme Court
of Canada offered a thick conception of reasonableness review. The
dominant principle is that of “responsive
justification”: administrative decisions must be
justified in view of the legal
and factual constraints on
decision-makers,
who must demonstrate their expertise through contemporaneous reasons that
grapple with the fundamental issues and arguments raised by the parties. It is not
enough for a decision to be justifiable, it must actually be justified. This is a far cry from
the deferential version of the court’s
review responsibility in Chevron step two that Justice Kennedy and some others have
perceived in recent American jurisprudence.
It is theoretically possible
that the SCOTUS could draw nourishment from Ottawa to fatten up Chevron step
two, in the same way that Justice Kagan rescued Auer deference from the
hangman’s noose by
giving it a highly contextual and reasoned structure
in Kisor
v Wilkie (see also the
discussion of Chevron’s footnote 11 in the amicus
brief of Professors
Barnett and Walker). The Canadian experience certainly attests to the value of
reasoned decision-making in generating legitimacy for public administration and
might prompt some rethinking about the relationship between Chevron’s two
steps. But none of the briefs submitted in Loper Bright seek to draw the
judges’ attention to developments elsewhere (despite Justice Gorsuch’s brief
(and somewhat selective) reference to comparative law in his dissent from
denial of cert in Buffington).
And as I have tried to show, there is now such a significant divide between
America and Canada on the central issues in Loper Bright as to make borrowing
or transplantation hazardous.
Indeed, Chevron
has never had the force of law in Canada. Prior to Vavilov, there
were occasional
suggestions that Canada should import the Chevron
two-step. But these never garnered the support of the Supreme Court of
Canada and Vavilov forecloses its importation now. In fact, in Vavilov,
the Supreme Court of Canada stated that in considering statutory text, purpose
and context, an administrative decision-maker is “not required to engage in a
formalistic statutory interpretation exercise in every case”. Even in applying
its specialized expertise and experience, the decision-maker might take an
approach different to that of a court, relying “on considerations that
a court would not have thought to employ but that actually enrich and elevate
the interpretive exercise”. In Canada,
non-courts have in
many ways become equal partners to the courts:
Canadian administrative decision-makers can consider constitutional issues (and
disapply unconstitutional statutes or grant individualized remedies for
unconstitutional state action), may assist in implementing Canada’s
international obligations, and can adopt interpretations of statutory
provisions that bind the courts. Even where there is a statutory right of
appeal – which, under Vavilov, leads to de novo review
– a court might
legitimately adopt the decision-maker’s interpretation as its
own. Canadian administrative
decision-makers and Canadian courts are both seen to have a legitimate role in
saying what the law is. One could not confidently say the same of the United
States.
Ultimately, it may be that the US/Canada divide runs deeper than doctrine. In Canada, there are few who question the legitimacy of the administrative state: if Canadian scholars are found debating this topic, they are more likely to be considering whether the administrative state is constitutionally required than constitutionally prohibited. As I am sure this symposium will demonstrate, the same is true elsewhere in the Commonwealth. Meanwhile, in the United States, the proposition that administrative law is unlawful because inconsistent with the constitutional vision of a group of men many generations past is now taken seriously by some commentators. I agree with Paul Craig that in the form it has taken in the last decade, this proposition is based on normative commitments that — to say the least — do not have an obvious basis in the historical record. When I look around the common law world, however, I see abundant evidence that there has long and largely uncontroversially been an administrative state, often armed with significant powers. Reading the tables of contents of the amicus briefs in Loper Bright — replete with references to “bureaucrats” “stifling” “freedoms” — makes me feel like a visitor in a strange world, far removed from my own. Hence my doubts about the likelihood or feasibility of the SCOTUS reaching across the US/Canada divide in its (re)consideration of Chevron.
Professor Paul Daly holds the University Research
Chair in Administrative Law & Governance at the University of Ottawa,
paul.daly@uottawa.ca