The past decade has not been kind to migrants. The events of 9/11 added extra hurdles to immigration process as immigration authorities’ paranoia grew. Western countries’ hospitality is much colder as xenophobia increased fuelled by right wing political groups such as the Tea Party in the USA, the Front National in France or the Conservatives in Canada and the UK. In Canada, since the election of the conservative government of Prime Minister Harper, we have adopted a series of reforms in immigration and refugee law. The focused is now on so called “desirable” migrants, usually people with education and language proficiencies in English or French.[1] Inadmissibility rules continue to prohibit family from reuniting, disabled and sick migrants from settling, and otherwise qualified migrants with tenuous link to alleged terrorist groups from immigrating.[2] Parliament adopted the so called Balanced Refugee Reform Act in 2010 making it harder for refugees to seek protection in Canada.[3] Additionally, Refugees’ access to health care was cut by the federal government.[4] Many of those changes are being or will be challenged in court. Whether these challenges will succeed is hard to predict as the Supreme Court has tended to be deferential towards the government on immigration and refugee issues. Two cases handed down in the last two months may shed some light on what the future holds for refugee and immigration law.

A Step Forward: Refugees and War Crimes

The first case, Ezokola v Canada (Citizenship and Immigration),[5] was a small step forward. Mr. Ezokola was a civil servant for the government of the Democratic Republic of Congo (DRC). He left his job, claiming he couldn’t work for a corrupt government anymore, and fled to Canada claiming that the DRC viewed him as a traitor and had threatened him. In Canada, the IRB found that there were serious reasons for considering that Mr. Ezokola committed a crime against peace, a war crime, or a crime against humanity and that he was thus precluded from claiming refugee status.[6] The finding was based on the fact that Mr. Ezokola previously worked as an economic advisor and a diplomat for the DRC and that it was known that the DRC committed war crimes and crimes against humanity in the recent pass. The IRB found that through his position, Mr. Ezokola was complicit in the war crimes. The IRB and the government’s broad interpretation of complicity as resulted in the exclusion of many refugee claimants who would never be held as complicit in criminal activities in a court of law. Any relatively senior public servants of a government who was once involved in crimes against humanity or war crimes was thus at risk of being denied refugee status in Canada even though their claim might be legitimate. The Federal Court quashed the decision based on the lack of nexus between the employment of Mr. Ezokola and the war crimes.[7] The Federal Court of Appeal rejected the Federal Court’s approach, but found that the IRB had applied the wrong test.[8] It found that Mr. Ezokola and others in similar situation could be found complicit if they were aware that crimes were being committed by their government.[9]

Defining War Criminal

The Supreme Court unanimously rejected the view of the IRB and the Federal Court of Appeal. It stated that claimants should not be excluded from the definition of refugee based on mere association with war criminals or passive acquiescence of war crimes. The Court determined that Article 1 F (a) of the Refugee Convention requires “serious reasons for considering that the claimant has voluntarily made a significant and knowing contribution to the organization’s crime or criminal purpose.”[10] As an exception to the rule, Article 1 F (a) has to be read strictly. Such an interpretation was supported by basic principles of criminal law (only culpable conduct should engage criminal liability and omissions are punishable only when there is a duty to act). While the IRB does not determine guilt or innocence, it still needs to follow a definition of complicity that is compatible with international criminal law. The international criminal jurisprudence has adopted a fairly broad definition of complicity, but even this broad definition does not encompass mere association or passive acquiescence: it requires some form of knowing and significant participation in the criminal activities.

In making an Article 1 F (a) determination, the IRB now has to consider:

(i) the size and nature of the organization;

(ii) the part of the organization with which the refugee claimant was most directly concerned;

(iii) the refugee claimant’s duties and activities within the organization;

(iv) the refugee claimant’s position or rank in the organization;

(v) the length of time the refugee claimant was in the organization, particularly after acquiring knowledge of the group’s crime or criminal purpose; and

(vi) the method by which the refugee claimant was recruited and the refugee claimant’s opportunity to leave the organization.[11]

This Supreme Court decision also brings Canada in line with the international community. The government’s position was simply untenable. The difference between the burden of proof in a refugee setting and the one in a criminal setting does open the door for different definitions of the applicable crime. The Canadian government was advocating for an unfair and even cruel interpretation of refugee law (probably for political aims, i.e. their “bogus” refugee narrative). It is good that refugee claimants in a situation similar to Mr. Ezokola will now have a fair chance. Mr. Ezokola’s case isn’t over as the IRB still needs to rehear his case with the new criteria in mind. I do not see, however, how it could reach the same conclusion with the new criteria.

A Question of International Law

While the substantive question at play was well canvassed by the Supreme Court, there is one point that is conspicuously missing from the decision: the standard of review analysis. Ezokola was an administrative law decision, a judicial review of the findings of an administrative tribunal. The Supreme Court sometimes, in administrative law case, summarily deals with the issue of standard of review to focus on the substantive issue. However, the Court didn’t even mention the words standard of review. It didn’t even use the language of reasonableness or correctness. I find this strange because there were at least two legitimate administrative law questions in this case.

Firstly, the court seems to apply the standard of correctness (substituting its own reasoning for that of the IRB). This was the standard applied by the Federal Court of Appeal.[12] The standard was applied because the appeal court was faced with a question of law. However, since Dunsmuir and especially since Alberta Teacher’s Association, questions of law pertaining to the tribunal’s expertise or home statute are usually reviewable for reasonableness.[13] In fact, this dictum was applied only one month before by the Supreme Court in Agraira.[14] In this case, the IRB was interpreting a provision of its home statute (albeit one that is referring to another instrument). It also technically has considerable expertise in refugee law being to main adjudicative body on refugee issues in Canada. The only explanation is that the question fell within one of the exceptions: constitutional questions, questions of law that are of central importance to the legal system as a whole and that are outside the tribunal’s expertise, questions on the jurisdictional lines between two or more competing tribunals, and true questions of jurisdiction. We are not in the presence of a constitutional question and we are not dealing with jurisdiction. The question is important for refugee law, but is it of central importance to the judicial system? And even if it is, it is clearly within the expertise of the tribunal. We are left to wonder why the Court applied what seems to be correctness.

One explanation could be, and this brings us to my second point, that questions of international law are subject to the standard of correctness. This would make sense since like constitutional law, international law goes beyond the discreet application of a statutory regime. It requires consistent interpretation across Canada and arguable between countries. Furthermore, the rules applicable to international law, like the Vienna Convention of the Law of Treaties used in Ezokola to interpret the Refugee Convention, are arguably beyond the expertise of national decision-makers. A consistent interpretation of international law is required as the same treaty provision should not mean different things depending on who is the decision-maker. In the case at bar, it would explain why, even though refugee law is within the expertise of the IRB, the Court had to apply correctness. Sadly the Supreme Court did not take the opportunity to clarify the role of international law in judicial review of administrative actions. However, in the end, it is clear that the Supreme Court could not and did not defer to the IRB on the interpretation of the Refugee Convention. The courts have the last word in this field it seems, at least for now.

Two Steps Back: Migrants, National Security and … Magic?

The appellant in Agraira v Canada (Public Safety and Emergency Preparedness) was not as lucky as the Court took two steps back.[15] Mr. Agraira came from Libya to seek refugee status, first in Germany, then in Canada. He was denied status in Canada for lack of credibility. After marrying a Canadian, he tried to get sponsored but Citizenship and Immigration Canada found that he was inadmissible to Canada because they had serious reason to believe he had been a member of a terrorist group in Libya, the LNSF who fought Kaddafi when he was still in power. Ironically, one might say, the LNSF is now part of the post-revolution government in Libya. While the finding of inadmissibility can be interpreted as problematic, Mr. Agraira did not seek judicial review of that finding. Instead, he applied pursuant to s 34(2) of the Immigration and Refugee Protection Act (the IRPA) for ministerial relief of inadmissibility. The provision gives the power to the Minister of Public Safety to basically void the finding of inadmissibility if he or she is satisfied that the applicant’s presence in Canada would not be detrimental to the “national interest”. The Canada Border Services Agency recommended granting the application considering the non-violent nature of Agraira’s involvement with the LNSF and his minimal implication with the organisation. The Minister disregarded the recommendation and denied relief. He appeared to have been primarily concerned with national security. Mr. Agraira sought judicial review. He was successful at the Federal Court – where Justice Mosley found that the Minister did not consider all the relevant factors – but lost at the Court of Appeal – where the Court found that “national interest” was synonymous with national security and public safety.[16]

The Meaning of National Interest

Given the discretionary nature of ministerial relief, the Supreme Court applied the reasonableness standard. The Appellant argued that the Minister’s decision was unreasonable because he only considered national security interest and did not balance other factors found in the administrative Guidelines.[17] The Court determined that the central issue was the meaning of “national interest” found in s 34(2) of the IRPA. Justice LeBel, for a unanimous Court, found that the Minister was entitled to deference on his interpretation of “national interest”. He then proceeded to interpret the term himself. Since the IRPA used the terms “national security” in some provisions, LeBel J. found that national interest could not have the same meaning given the statutory interpretation rule that when Parliament uses different words in a statute, they mean different things. The plain meaning of “national interest”, the statute as a whole, the context of the provision, and the legislative history of the provision all confirmed this finding. National security concerns are part of the national interest analysis, but the concept embodies broader considerations such as the ones mentioned in the Guidelines. However, which factors are relevant for ministerial relief will depend on each case.[18] Justice LeBel then found that although the Minister focused on national security and public safety, his implied interpretation of “national interest” was a reasonable interpretation because he did not reject or exclude other factors. He subsequently found that the decision as a whole was reasonable:

“Having concluded that the Minister’s implied interpretation of the term “national interest” is reasonable, I should also confirm that the decision as a whole is valid. […]

As the Minister stated in his reasons, he had “reviewed and considered” (i.e. weighed) all the factors set out in the appellant’s application which were relevant to determining what was in the “national interest” in light of his reasonable interpretation of that term. […] Given that the Minister considered and weighed all the relevant factors as he saw fit, it is not open to the Court to set the decision aside on the basis that it is unreasonable.”[19]

Implied Interpretation: Divining Ministerial Interpretation

The Court’s approach to reasonableness review in Agraira is troubling. The implied interpretation the Court adopted seems at odd with the meaning of reasonableness. As stated in Dunsmuir, “reasonableness is concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process.”[20] The Minister’s reasons focus solely on national security issues. While the Supreme Court has indicated that a reviewing court can supplement a decision with the record when appropriate to avoid “subverting” the reasons,[21] I doubt that this was meant as a permission to conjure post-decision judicial reasons to uphold an administrative action. Justice LeBel relies on Alberta Teachers’ Association, but in that decision (which is also problematic in my opinion) the order of the Privacy Commissioner could have only been issued if the relevant statute was interpreted a certain way.[22] Since the order was issued, one could imply that the Commissioner had interpreted the statute in a way that permitted the existence of the order. In Agraira, there was no such indication. In fact, the Minister decision seems to imply that he interpreted “national interest” as something similar if not identical to national security and public safety (as evidenced by the decisions of the Federal Court and the Federal Court of Appeal).

The Court appears to interpret the IRPA on its own, to determine the correct interpretation of “national interest”, and to presume that this interpretation was adopted by the Minister even though there is no indication to that effect. On the one hand, this approach goes against deference as the Court seems to step into the shoes of the Minister to determine the correct interpretation (instead of determining if the Minister’s interpretation is reasonable). On the other hand, the Court effectively shields the Minister’s power from meaningful review by assuming that he adopted the Court’s correct interpretation and that he did consider all relevant factors, relying on the magic phrase “having reviewed and considered the material and evidence” found in the decision. On both points, the Court appears to partake in judicial gymnastic to uphold the Minister’s decision. I note that had the Court found otherwise, Mr. Agraira would have not necessarily obtained the relief he sought. The decision would have most likely been sent back to the Minister for a do over. As it stands, Agraira implies that courts can supplement administrative decisions to the detriment, in my view, of reasoned decision-making.

[1] See for example Trevor Melanson, Infographic: Who will benefit from Canada’s new immigration rules?, Canadian Business, 10 May 2013, online.

[2] See for example Kirec v Canada (Minister of Citizenship and Immigration), 2006 FC 800; Kobita v Canada (Citizenship and Immigration), 2012 FC 1479; and Krishnamoorthy v Canada (Citizenship and Immigration), 2011 FC 1342.

[3] Balanced Refugee Reform Act, SC 2010, c 8: <http://www.parl.gc.ca/legisinfo/BillDetails.aspx?billId=4383517&Language=E&Mode=1>.

[4] See for example André Picard, Seeking savings, Ottawa takes short-sighted view on refugees’ health care, The Globe and Mail, 18 July 2013, online.

[5] Ezokola v Canada (Citizenship and Immigration), 2013 SCC 40.

[6] X (Re), 2009 CanLII 89027 (IRB). The decision was made pursuant s 98 Immigration and Refugee Protection Act, SC 2001 c 27, which refers to Article 1 E & F.

[7] Ezokola v Canada (Citizenship and Immigration), 2010 FC 662.

[8] Canada (Citizenship and Immigration) v Ezokola, 2011 FCA 224.

[9] Ibid, at para 72.

[10] Ezokola, supra note 5 at para 84.

[11] Ezokola, ibid at para 91.

[12] Ezokola, supra note 8 at para 39.

[13] Dunsmuir v New Brunswick, 2008 SCC 9 at para 54; and Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association, 2011 SCC 61 at para 30.

[14] Agraira v Canada (Public Safety and Emergency Preparedness), 2013 SCC 36 at paras 57-58.

[15] Agraira, ibid.

[16] Ramadan Agraira v Canada (Public Safety and Emergency Preparedness), 2009 FC 1302; and Canada (Public Safety and Emergency Preparedness) v Agraira, 2011 FCA 103.

[17] Agraira, supra note 14 at Appendix 1.

[18] Ibid, at para 87.

[19] Ibid, at paras 89 & 91.

[20] Dunsmuir, supra note 13 at para 47.

[21] Newfoundland and Labrador Nurses’ Union v Newfoundland and Labrador (Treasury Board), 2011 SCC 62.

[22] Alberta Teachers’ Association, supra note 13.