Judicial Review - SOR - Reasonableness - Not Own Analysis. Public Service Alliance of Canada v. Canada (Senate)
In Public Service Alliance of Canada v. Canada (Senate) (Fed CA, 2023) the Federal Court of Appeal stresses the point that, under Vavilov, the reviewing court should not approach their JR task 'from the ground up', but should rather adhere strictly to the 'reasonableness' analysis:
 Under the judicial review framework set out in Vavilov, a reasonable decision is "“one that is based on an internally coherent and rational chain of analysis and that is justified in relation to the facts and law that constrain the decision maker”" (Vavilov at para. 85). The burden is on the party challenging a decision to show that it is unreasonable, a conclusion that requires showing that the decision contains a serious flaw. In addition, the reviewing court must refrain from reweighing and reassessing the evidence considered by the decision maker. Reviewing courts must also ordinarily refrain from deciding the issue that was before the decision maker and must respect the decision maker’s role and expertise (Vavilov at paras. 75, 83, 100, 125). Public Service Alliance of Canada v. Canada (House of Commons)
In Public Service Alliance of Canada v. Canada (House of Commons) (Fed CA, 2023) the Federal Court of Appeal stresses the point that, under Vavilov, the reviewing court should not approach their JR task 'from the ground up', but should rather adhere strictly to the 'reasonableness' analysis:
 Under the judicial review framework set out in Vavilov, a reasonable decision is "“one that is based on an internally coherent and rational chain of analysis and that is justified in relation to the facts and law that constrain the decision maker”" (Vavilov at para. 85). The burden is on the party challenging a decision to show that it is unreasonable, a conclusion that requires showing that the decision contains a serious flaw. In addition, the reviewing court must refrain from reweighing and reassessing the evidence considered by the decision maker. Reviewing courts must also ordinarily refrain from deciding the issue that was before the decision maker and must respect the decision maker’s role and expertise (Vavilov at paras. 75, 83, 100, 125).. Imperial Oil Limited v. Haseeb
In Imperial Oil Limited v. Haseeb (Ont CA, 2023) the Court of Appeal elaborated the SOR on an appeal from a JR application ["stepping in the shoes"] - and (mostly) criticizes the Divisional Court below on their own SOR analysis, where the CA accuses them of 'doing their own analysis' rather than a proper arm's-length 'reasonableness' assessment:
 In an appeal to this court from a decision of the Divisional Court on an application for judicial review, this court must determine whether the Divisional Court identified the appropriate standard of review and applied that standard correctly in reviewing the tribunal’s decision. The latter step requires this court to “step into the shoes” of the Divisional Court and focus on the tribunal’s decision, applying the applicable standard of review. The appeal is, in effect, a de novo review of the tribunal’s decision. This court is not restricted to asking whether the Divisional Court committed a palpable and overriding error in its application of the appropriate standard: Ontario (Health) v. Association of Ontario Midwives, 2022 ONCA 458, 161 O.R. (3d) 561, at para. 42; Turkiewicz (Tomasz Turkiewicz Custom Masonry Homes) v. Bricklayers, Masons Independent Union of Canada, Local 1, 2022 ONCA 780, 476 D.L.R. (4th) 421, at para. 49, leave to appeal to S.C.C. requested, 40564 (January 16, 2023); Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36,  2 S.C.R. 559, at paras. 45-46; Canadian Federation of Students v. Ontario (Colleges and Universities), 2021 ONCA 553, 157 O.R. (3d) 753, at para. 20; Briggs v. Durham (Police Services Board), 2022 ONCA 823, at para. 36, leave to appeal to S.C.C. requested, 40587 (January 27, 2023).
 As outlined above, the standard of review in this case requires this court to step into the shoes of the Divisional Court and assess the reasonableness of the tribunal’s decision. For this reason, rather than structure the analysis as a point-by-point assessment of the appellant’s grounds of appeal and then the respondent’s additional issues, I structure the analysis around whether the tribunal’s decision was reasonable and focus on the areas where the reasonableness of the decision is challenged. The ultimate question this court must answer is whether the decision of the tribunal “is based on an internally coherent and rational chain of analysis and … is justified in relation to the facts and the law that constrain the decision maker:” Vavilov, at paras. 85 and 99.
(f) The Divisional Court majority did not correctly apply the standard of review
 As the standard of review directs this court to step into the shoes of the Divisional Court on an appeal from an application for judicial review, I need not examine in detail the errors made by the Divisional Court majority. But I will flag them briefly.
 First, the majority reasons did not take as a starting point respectful attention to the reasons of the tribunal. Rather, the majority started by assessing the question of whether Imperial discriminated against the appellant by re-doing its own analysis from scratch. This is evident from, for example, the hypothetical fact situations involving non-citizens without any legal right to work in Canada that appear to have driven the majority’s analysis. These hypothetical situations do not arise from the tribunal’s analysis, which was focused squarely on the record before it.
 Second, the majority mischaracterized the reasons of the tribunal. Respectful attention to a tribunal’s reasons, which is the starting point of reasonableness review, cannot be accorded if the reviewing court mischaracterizes the tribunal’s reasons. The reasons of Lederer J. characterize the tribunal as having effectively created a new ground of discrimination not found in the Code based on permanent resident status. As I have outlined above, that was not the basis on which the tribunal found that Imperial had discriminated against the appellant. Nor did the tribunal fail to examine the “plain and ordinary” meaning of “citizenship” and “permanent residence”. The tribunal’s reasons show that it understood the difference between these statuses as a matter of federal immigration law. The tribunal’s finding of discrimination on the basis of citizenship was not based on subsuming the concept of “permanent residence” into citizenship.
 Third, the majority of the Divisional Court failed to follow well-established analytical principles for human rights claims. I will highlight two examples of this.
 The first is in the reasons of Lederer J. His reasons fail to consider the principle from Brooks and subsequent decisions that partial discrimination on a prohibited ground is still discrimination. Indeed, his reasons proceed on a basis that is contrary to this well-established principle. In substance, he found that Imperial did not discriminate on the basis of citizenship because it permitted hiring of one sub-group of non-citizens – permanent residents of Canada.
 The second failure to follow well-established principles of analysis for human rights claims is the focus in the reasons of Mew J. on the hardship for employers if Imperial’s policy constitutes discrimination on the basis of citizenship. This approach failed to follow the established structure to analysis of human rights claims. Potential hardship to an employer is not considered when assessing whether a prima facie claim of discrimination on a prohibited ground has been established. Rather, it is a matter to be considered after a prima facie case of discrimination has been established, as part of a BFOR defence and consideration of an employer’s duty to accommodate to the point of undue hardship: see e.g., Stewart, at paras. 22-23; Pieters, at paras. 64-66.
 I note as well two other problems with the focus on employer hardship in the reasons of Mew J. First, as I have noted, Imperial did not pursue the BFOR defence in the Divisional Court. With the issue not having been raised by Imperial, it was unfair to resurrect it as a ground to overturn the tribunal decision.
 Second, the record did not support the conclusion that Imperial would suffer undue hardship if it was required to change its policy and consider PGWP-holders for employment. Indeed, the tribunal made specific findings of fact to the contrary. The tribunal considered whether there would be hardship to Imperial if it were required to consider PGWP-holders for employment in its consideration of whether Imperial had established a BFOR defence. Imperial had asserted that being required to hire PGWP-holders would pose a risk to succession planning and to Imperial losing its investment in training of new hires. The tribunal found that this assertion was not borne out on the evidence led by Imperial.
 In particular, the tribunal found that, to the extent Imperial had in some cases waived its policy for more senior positions and hired PGWP-holders, there was “no empirical data” to demonstrate that these PGWP-holders left employment with Imperial at higher rates than Canadian citizens or permanent residents within the two to three years it generally took to obtain permanent resident status, or that those who left within that timeframe did so because of a failure to obtain permanent resident status.
 Because Imperial did not challenge the reasonableness of the tribunal’s alternate finding that a BFOR defence was not established, it is not necessary for this court to consider whether the tribunal’s conclusion in that regard was reasonable. However, I flag these issues in relation to the record because they demonstrate that Mew J. did not start his analysis from a posture of respectful attention to the reasons of the tribunal. Rather, he raised concerns about hardship to employers which, as related to the employer before the tribunal, Imperial, the tribunal had found were not supported in the record before it.
 What appears to have driven the reasoning of the majority in the Divisional Court was a floodgates concern that if Imperial’s treatment of the appellant constituted discrimination on the basis of citizenship, then, in the view of the majority, any non-Canadian citizen anywhere in the world could claim discrimination in employment on the basis of citizenship if they were not considered for a job in Canada. Both judges in the majority referred to the situation of a hypothetical American who lives in Detroit (just over the Canadian border). The majority was of the view that the tribunal’s analysis of the appellant’s claim would have the result that this hypothetical American, with no status in Canada at all, and no legal right to work in Canada, could allege discrimination on the basis of citizenship if they were refused employment by Imperial.
 The analogy posited by the majority of the Divisional Court is flawed. It fails to recognize the fundamental distinction between PGWP-holders and other non-Canadian citizens – PGWP-holders have a legal right under federal immigration law to work full-time, for any employer, anywhere in Canada. This distinguishes PGWP-holders from other non-Canadian citizens in terms of their right to work in Canada. The tribunal was clear in its reasons that the central fact that drove its analysis was that when the appellant received his PGWP on graduation (of which there was no doubt based on the tribunal’s findings), he would have the right to work full-time, anywhere in Canada, for any employer. The tribunal’s reasoning would not extend to a person without an unrestricted right to work in Canada.
 This is particularly so given s. 16(1) clarifies that the right to non-discrimination based on citizenship is not infringed “where Canadian citizenship is a requirement, qualification or consideration imposed or authorized by law.” Because federal immigration law does not give most non-citizens a right to work in Canada, the hypothetical American living in Detroit without any status in Canada would not have a viable citizenship discrimination claim, as they would not be authorized by law to work in Canada. Accordingly, the concern of the majority in the Divisional Court that the tribunal’s analysis would allow any non-Canadian citizen to allege discrimination in employment is unfounded.
 In sum, the tribunal’s finding that Imperial discriminated against the appellant on the basis of citizenship was reasonable. The majority of the Divisional Court applied the standard of review incorrectly. When the standard of review is correctly applied, there is no basis for the reviewing court to intervene.