The Appeal of Appeals: Canadian National Railway Company v. Canada (Transportation Agency), 2025 FCA 184

In a previous post, I discussed a recent decision of the Federal Court of Appeal about the ground covered by an appeal on a question of law or jurisdiction, on the one hand, and an application for judicial review, on the other hand. Now the Federal Court of Appeal has returned to the issue, this time in the rate-setting context, in Canadian National Railway Company v. Canada (Transportation Agency), 2025 FCA 184, doubling down on one of its core propositions, namely that inadequately reasoned decisions constitute an error of law for the purposes of a statutory appeal.

At issue here was a railroad rate-setting decision under s. 127.1 of the Canada Transportation Act, SC 1996, c. 10. The railroad complained that the rate set was not “commercially fair and reasonable to all parties” as required by s. 112 of the Act, because the Agency had failed to take commercial market factors into account. In a series of prior decisions, the Agency had set out a rate-setting methodology that did not rely on or refer to market factors. Indeed, as Stratas JA explained, the basis for the Agency’s methodology was opaque:

Of note—and we will return to this at the end of these reasons—the Agency has never conducted and presented, with supporting reasons, a full analysis of the text, context and purpose of the sections in the Canada Transportation Act that bear on the issue in this case. Here, at least judging by the Agency’s reasons, it did not do that analysis, nor did it cite to any decision that did that analysis. Instead, over many years, in case after case, it seems that the Agency has applied standards that may or may not have come from the Act—we simply do not know (at para. 9).

Stratas JA analyzed the text, context and purpose of the relevant provisions of the Act and concluded that the Agency’s interpretation was wrong. For instance, as to text and context:

The Agency has regarded commercial market factors as irrelevant to its determination of interswitching rates. The plain meaning of “commercially” and the legislative history concerning and surrounding section 112 show that this is wrong. Effectively, the Agency has acted as if “commercially” were read out of section 112. This it cannot do.

Overall, the three elements of section 112—“fair and reasonable”, “to all parties”, and “commercially”—all lead to the conclusion that when setting interswitching rates, the Agency must receive evidence relevant to the commercial market factors and consider those factors (at paras. 27-28).

Statutory purpose, too, weighed in favour of the relevance of commercial market factors, given an express legislative commitment to competitiveness and economic growth in s. 5 of the Act (see e.g. at paras. 32, 38).

Now to the (even more) interesting part. In general, Stratas JA observed, a failure to conduct a rigorous statutory interpretation analysis means that the decision under appeal is unlawful:

Cutting corners and conclusory statements, without more, are not how the Agency should roll: see Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653 at paras. 115-124. In Vavilov, the Supreme Court instructed administrative decision-makers to show in their reasons that they are alive to the issues of text, context and purpose in the statutory interpretation process. For a major administrative decision-maker like this, one that is dealing with an issue like this, only explicit and rigorous analysis will do: Mason v. Canada (Citizenship and Immigration), 2023 SCC 21. The same must be said for applying the statutory standards to the evidence in a case like this (at para. 44).

The requirements of reasoned decision-making from Vavilov and Mason were set out in the context of the application of the reasonableness standard of judicial review, not statutory rights of appeal. Stratas JA acknowledged this but insisted that adequate reasons are required in all cases — whether the decision comes before a court by way of appeal or judicial review, the culture of justification applies and adequate reasons must be given (at paras. 45, 47). In particular, Stratas JA noted, the underlying rationale is not limited to either appeals or judicial reviews:

There are at least three rationales. First, adequate reasons, especially those that analyze text, context and purpose, require careful and rigorous work that often exposes faulty reasoning before the decision is released. Second, in high stakes determinations like this, adequate reasons tell the parties that their key arguments were taken on board and considered, something resting at the core of procedural fairness. Third, adequate reasons further the transparency, legitimacy and accountability of administrative decision-makers to the parties before them, other regulatees, reviewing courts, and the wider public—something needed more than ever in these days of widespread skepticism, cynicism, and mistrust of government. See Vancouver International Airport Authority v. Public Service Alliance of Canada, 2010 FCA 158, [2011] 4 F.C.R. 425 at para. 16 and Canada (Citizenship and Immigration) v. Tennant, 2018 FCA 132 at paras. 23-24 (at para. 46).

As to the vanishing distinction between judicial review and appeal, benediction from the Supreme Court may yet be required (as I suggested previously). But there is also a hint of the logic animating the Federal Court of Appeal’s approach in the majority reasons in Northback Holdings Corporation v Alberta Energy Regulator, 2025 ABCA 186. This was a case featuring a right of appeal, with leave, on a question of law or jurisdiction to the Alberta Court of Appeal (Responsible Energy Development Act, SA 2012, c R-17.3, s. 45) coupled with a privative clause purporting to eliminate judicial review (ibid., s. 56). The applicants sought leave to appeal the Regulator’s decisions to decline to approve an open-pit coal mining project. Leave was refused. The applicants then sought judicial review in superior court. The application was struck out. On appeal, Khullar CJ would have permitted the judicial review to proceed, on the basis that the constitutional core minimum of judicial oversight of the administrative process “must include review on questions of fact and mixed fact and law” (at para. 225).

But her colleagues in the majority took a different view, essentially that the application for judicial review was misconceived because the right of appeal on a question of “law” or “jurisdiction” was potentially wide enough to encompass grounds of judicial review. As the applicants had not raised the issue of the scope of “law” or “jurisdiction” on the leave application, the first-instance judge was in an impossible position:

The parties debate whether the term “jurisdiction” is broad enough to encompass questions of fact. As noted, however, they did not raise the question before this Court on the permission application. In the absence of that argument being made and decided in the context of the s 45 process, the chambers justice was left in the realm of hypotheticals. She could not embark on an analysis or make any assumptions as to how this Court would decide the issue if it was raised on a s 45 appeal. As a justice of the Court of King’s Bench sitting on a judicial review matter (and not an appeal of the Permission Decision), she could not comment on whether the single judge of this Court properly interpreted its jurisdiction under s 45 in light of the constitutional argument, nor did she purport to do so. She simply observed that s 56 barred the appellants’ applications for judicial review (at para. 44).

Without necessarily committing themselves definitively to the view that “jurisdiction” must encompass some or all of the grounds that can be raised by way of judicial review, the majority held that the issue would have to be raised and adjudicated in the context of an application for leave to appeal:

Vavilov established that institutional design choices are to be respected. This Court gives meaning to the terms “question of jurisdiction” and “question of law” in the context of a statutory appeal under s 45. If Vavilov has resulted in a constitutional minimum of review based on an expansive approach to Crevier’s “jurisdiction”, it would seem the logical starting point to address this would be in the interpretation of “question of jurisdiction” in s 45 in the context of the statutory appeal regime established by the legislature for review of administrative decisions (at para. 53).

It might be objected that if an appeal on a question of “jurisdiction” encompasses some or all of the grounds of judicial review, it is impossible to give effect to the institutional design choice made by the legislature to create a narrow route to the Court of Appeal on carefully targeted issues.

As against that, the alternative is to have a limited right of appeal to the Court of Appeal with (if Khullar CJ is right and, of course, I think she is) a judicial review running along a parallel track in the Court of King’s Bench. Moreover, if one takes the view that judicial review is always available (as befits a “cornerstone” of our constitutional order: Immeubles Port Louis Ltée v. Lafontaine (Village), [1991] 1 SCR 326), then a right of appeal, even a narrow one, can be seen as an ‘addition’ to what the common law already guarantees: the same grounds could be raised, but in a proceeding subject to different procedural rules (including, if leave is granted and the appellant is successful on the merits, a right to relief rather than the inherently discretionary remedy available on judicial review). On this view, the right of appeal enhances, rather than restricts, judicial oversight of the administrative process. That would be the institutional design choice to respect.

Ultimately, though, the reasons why legislatures granted rights of appeal on questions of “law” or “jurisdiction” are lost in the mists of time, which is a large part of the reason for the current uncertainty. For my part, I very much welcome the addition of Vavilovian reasonableness review to the context of statutory appeals: this is territory that the culture of justification ought to occupy, in my view, correcting the anomaly of less intrusive judicial review where the legislature has created a right of appeal that presumably was chosen for ensuring more intrusive judicial review.

CN is an especially significant decision in the transportation context. There is no right to seek judicial review of the Agency’s decisions, by operation of s. 18.5 of the Federal Courts Act, because a person aggrieved by an Agency decision can petition the Governor in Council to intervene (as is also true under the Telecommunications Act). The Federal Court of Appeal recently confirmed that s. 18.5 is constitutionally valid, despite the argument that reasonableness review or an equivalent form of judicial oversight is now constitutionally entrenched (Canadian National Railway Company v. Alberta Pacific Forest Industries Inc., 2025 FCA 160). Hence, therefore, Stratas JA’s observation that the “assessment of weight is entirely for the Agency to decide on the evidence in each case, relying upon its industry appreciation, regulatory experience, and transportation expertise” (at para. 40): absent an extricable legal error or inadequate reasons, an aggrieved party’s recourse is to the Governor in Council on considerations of policy or fact. Policing the reasons requirement in this context fills an important gap, as the Governor in Council’s role will not include reviewing a regulator’s reasons for compliance with Vavilov’s reasonableness standard.

One last important point about this important case. Does it stand for a general proposition that, in rate-setting matters, a regulator should look to commercial market factors? Stratas JA was clear that this is not so. Commercial market factors are relevant here because of the nature of the industry and the explicit reference to competitiveness in s. 5 of the Act:

For completeness, I have considered two other Supreme Court authorities on rates-setting: Ontario (Energy Board) v. Ontario Power Generation Inc., 2015 SCC 44, [2015] 3 S.C.R. 147 and Northwestern Utilities Ltd. v. City of Edmonton, 1929 CanLII 39 (SCC), [1929] S.C.R. 186. The parties did not refer us to these. I consider them distinguishable.

These cases concerned “just and reasonable” rates or “fair and reasonable rates” for the services provided by certain utilities. But the utilities were sole, monopolistic suppliers in all their activities. This differs from CN, which competes with other railways and modes of transportation in many of its operations.

As well, the statutory regimes in those two Supreme Court cases did not have the robust purpose clause we have in section 5 of the Canada Transportation Act, with its mention of “competition and market forces” to further a “competitive, economic and efficient national transportation system”, and section 112’s mention of “commercial[ity]”, with all the legislative history surrounding it. For example, by contrast, in Ontario Power Generation, the purpose clause of the relevant statute referred to “the maintenance of a financially viable electricity industry” and it did not refer to “competition and market forces” (at paras. 36-38).

Put another way, this ticket is good for railway rates only. Nonetheless, it is a useful reminder to regulators and regulated alike that individual statutes can have a significant bearing on the task of identifying just and reasonable rates.

This content has been updated on October 28, 2025 at 12:05.