Dunsmuir – Plus ça change Redux (Lorne Sossin)
Lorne Sossin is the Dean of the Osgoode Hall Law School
A few months after the Supreme Court released Dunsmuir v. New Brunswick (2008 SCC 9), a number of administrative law colleagues, led by David Dyzenhaus, convened a symposium at the University of Toronto Faculty of Law to explore its implications. My contribution to that collection of perspectives at that time, entitled “Dunsmuir: Plus ça change” highlighted the largely cosmetic revisions to the standard of review evident in the Supreme Court’s majority reasons in the Dunsmuir. Those reasons suggested the Court was embarking on a dramatic step forward, but the resulting doctrinal tweaks seemed likelier to produce sideways movement. Below, I return to this question a decade on and revisit the critique of the standard of review in Canadian administrative law on the grounds that it fails to engage with the lived experience of administrative decision-making.
In Dunsmuir, in my view, the Supreme Court of Canada rebottled an abstract, one-size fits all formula for determining how courts would review administrative decision-makers. New and improved packaging enhanced the marketing (at least for a while) but the flaws in the content were left untouched. The constitutional backdrop for Dunsmuir – and the standard of review generally – is fidelity to the twin pillars of the rule of law and democracy. Fidelity to democracy is represented in the idea that where the legislature, representing the will of the majority, has authorized administrative decision-makers to make important decisions based on their expertise and particular role in advancing policy objectives of legislation, courts should defer to that legislative choice, and hence to the decisions of these decision-makers. This deference is particularly important where the legislature has expressly provided for it in a privative clause. Fidelity to the rule of the law requires intervention by independent courts in the actions of the executive branch to safeguard the boundaries of legitimate statutory authority for executive action – for example, by ensuring decision-makers do not stray from the jurisdiction provided to them by their empowering statutes, or otherwise engage in unreasonable exercises or abuses of that statutory authority. In almost all cases, these twin pillars mandate that courts defer to some extent when executive action is challenged, but can never defer completely.
The problem with this constitutional narrative is that those who actually make the decisions and those who actually are affected by the decisions are completely missing from the analysis. For example, where a tribunal provides rigorous education and training for its adjudicators, or whether a ministerial discretion is elaborated through detailed guidelines and soft law are critical to the lived experience of decision-makers but irrelevant under Dunsmuir to the standard of review. Similarly, whether the parties affected by a statutory decision are vulnerable or powerful, whether repeat players or a one-time participant, play no role whatsoever in determining or applying the standard of review. The abstraction of the standard of review analysis from the facts and circumstances of actual cases lies at the root of the Court’s struggles for consistency and coherence. It is this exercise in abstraction that led Justice Binnie in Dunsmuir to decry the “law office metaphysics” (at para. 122).
Lebel and Bastarache JJ., introduced their majority reasons in Dunsmuir by throwing down a gauntlet that it was time to fundamentally overhaul the standard of review analysis. They indicate their aspiration is to an “holistic” approach. On this, however, their reasons over-promise and under-deliver. The majority doubles down on the old pragmatic and functional criteria of the existence of a privative clause, statutory context, the nature of the question and expertise. While the removal of the intermediate reasonableness simpliciter standard from the framework was a significant cosmetic change, the key premises of the analysis remained. Indeed, to the extent this was in doubt, the Court did not suggest that any previous standard of review decision by the Court (or any other court) was wrongly decided or needed to be reversed in light of this cosmetic refinement.
After a brief respite of consensus on the standard of review after Dunsmuir, fissures in how the remaining two standards were to be determined and applied returned. By the time of the Court’s 5-4 judgment in Edmonton (City) v. Edmonton East (Capilano) Shopping Centres Ltd., 2016 SCC 47, deep divisions had resurfaced challenging the coherence of the standard of review analysis and the meaning of deference itself.
So, what would actual change look like in the standard of review analysis? I believe the only way to achieve forward movement in the standard of review is to alter the premise of this analysis. Rather than focus exclusively on an abstract examination of statutory intent, statutory context and category or question, the standard of review analysis needs to bring decision-makers and parties into the analysis. This shift means acknowledging that the same decision under the same legislation may lead the court to different standards of review depending on the vulnerability of the affected party and the intensity of the decision’s impact on that party. For example, removing the license of a driver who enjoys recreational trips to visit friends is a categorically different decision than removing the license of a trucker whose livelihood and ability to find employment depends on that license. The impact of ministerial discretion in settings of vulnerability is not the same as the impact of discretion in settings of economic regulation, and so forth. In other words, such a shift involves viewing the standard of review the way we view the framework for procedural fairness, where two decisions may result in different procedural guarantees depending on the importance of the decision to the parties affected by it.
The logic of including the impact on the party and the context of the decision-maker in the analytic framework for procedural fairness is that the accountability of executive action under administrative law in a constitutional democracy is best understood as holistic. This exercise cannot be complete just by analyzing statutes and classifying questions. The people involved, and the way in which people vary the other dynamics (for example, the difference between a question of fact, law or discretion, or whether a statute intended courts to defer) ought to be an essential element of the standard of review.
This proposed contextual turn for the standard of review is not new – it is a vision laid out in more detail in “The Contextual Turn: Iacobucci’s Legacy and the Standard of Review in Administrative Law” (2007), 57 U.T.L.J. 581, an article Colleen Flood and I co-authored for a 2007 symposium in honour of Frank Iacobucci (ironically, the jurist whose vision of administrative law and the reasonableness simpliciter standard in particular Dunsmuir consigned to a history footnote).
This more truly holistic approach also has been adopted in other peer jurisdictions. For example, writing for the New Zealand High Court in Wolf v. Minister of Immigration  NZAR 414 (H.C.), Justice Wild summed up this framework as follows (at para. 47):
Whether a reviewing Court considers a decision reasonable and therefore lawful, or unreasonable and therefore unlawful and invalid, depends on the nature of the decision: upon who made it; by what process; what the decision involves (ie its subject matter and the level of policy content in it) and the importance of the decision to those affected by it, in terms of its potential impact upon, or consequences for, them.” (Emphasis added.)
The final argument for focusing on who makes decisions and who is affected by them in the standard of review analysis is transparency. While the Dunsmuir Standard of Review analysis glosses over these lived realities of administrative decision-making, it is clear that in the realities of judicial review, facts actually do matter. I should add that facts matter in ways that are not one-dimensional. Vulnerability will sometimes militate for additional deference; while in other contexts, this factor will justify intervention. My point is not that facts matter to the determination of the standard of review in the same way in every case, but rather that, in every case, facts matter. Without the ability to talk about how those facts might legitimately affect the rationale for deference, courts will simply bend existing doctrines to fit the necessities and equities of particular cases. In my view, recognizing the complexity of de facto considerations in a more authentic de jure doctrinal analysis would be a good thing for the administration of justice and ultimately enhance public confidence in the justice system.
I concluded my original analysis in “Plus ça change” with a call for a “spectrum” of deference akin to the “spectrum” of intensity of fairness guarantees mapped out in Baker v. Canada (Minister of Citizenship and Immigration),  2 S.C.R. 817 -and its progeny – for procedural fairness. I continue to view this as another inevitable de facto reality of deference. Even if every case were resolved on a deferential standard of reasonableness, as Justice Abella mused might be preferable in Wilson v. Atomic Energy of Canada Ltd., 2016 SCC 29, the intensity of reasonableness review would nonetheless shift according to the circumstances. We pursue doctrinal simplicity for all the right reasons (consistency, efficiency, etc) but as the spectrum of fairness has demonstrated so vividly in Canadian administrative law, one size – or binary distinctions between two sizes – really will never fit all.
This content has been updated on March 4, 2018 at 15:10.