Date: 20150724

Docket: IMM-798-14

Citation: 2015 FC 909

Ottawa, Ontario, July 24, 2015

PRESENT:    The Honourable Mr. Justice Boswell

JUDGMENT AND REASONS

I.                   Introduction and Background

[1]               This is an application for judicial review of a decision by the Refugee Appeal Division of the Immigration and Refugee Board.

[2]               The Applicants are a 41-year-old woman [Principal Applicant] and her 11-year-old daughter from Pakistan. They allegedly fear that they will be persecuted in that country for being Shia Muslims. The Principal Applicant stated in her narrative that she had always experienced discrimination, but the harassment became more severe after she and her husband began hosting religious functions at their home. This culminated in an incident on March 19, 2012, when several masked men attacked her as she was returning from the market. The Principal Applicant said that she went into hiding after the police did nothing, and she remained in hiding until she was able to secure passage to Canada for herself and her daughter. They arrived in this country on December 28, 2012, and made refugee claims some two months later.

[3]               Their claims were dismissed by the Refugee Protection Division [RPD] of the Immigration and Refugee Board [IRB] on October 11, 2013. The RPD found that the Principal Applicant was a poor witness. The Principal Applicant originally testified that the men who attacked her did not say they belonged to any organization; but then later she evasively testified that they belonged to the Tehreek e Tahafuz e Islam, which is the same as the Tehreek e Sipa e Muhammed. Neither of these organizations was identified in the Principal Applicant’s basis of claim form, and the RPD doubted that she would omit such a central detail of her claim simply because she was frightened. The RPD thus decided that the Principal Applicant had not established that the agents of persecution were affiliated with any organization at all.

[4]               As there was no organization pursuing her, the Principal Applicant was unable to convince the RPD that she had no internal flight alternative [IFA]. The RPD noted that Karachi was a large, populous city far away from where the Applicants had lived, and the random men who attacked the Principal Applicant would not likely be motivated or equipped to seek her out in Karachi. While there was some general sectarian violence everywhere in Pakistan, the RPD observed that there are between 9 and 38 million Shia Muslims in the country. According to the RPD, the documentary evidence did not support a finding that the scale of the violence was such that every Shia Muslim in Karachi faced more than a mere chance that they would be harmed. As well, there was insufficient evidence to persuade the RPD that the Applicants would be unable to practice their religion openly or that the Principal Applicant could not find employment. Thus, the RPD rejected the Applicants' claims.

[5]               The Refugee Appeal Division [RAD] dismissed the Applicants' appeal and confirmed that they were not entitled to protection under either section 96 or subsection 97(1) of the Immigration and Refugee Protection Act, SC 2001, c 27 [Act]. The Applicants now seek judicial review of the RAD's decision pursuant to subsection 72(1) of the Act, requesting that the RAD's decision be set aside and the matter returned to a different panel of the RAD for redetermination.

II.                Issues

[6]               This application raises the following issues:

1.                  What standard of review should this Court apply to the RAD's decision?

2.                  Did the RAD apply an appropriate standard of review to the RPD's decision?

3.                  Were the RAD's determinations on credibility and IFA unreasonable?

III.             The RAD’s Decision

[7]               The RAD dismissed the Applicants' appeal on January 20, 2014. After reviewing relevant provisions of the Act and the Alberta Court of Appeal's decision in Newton v Criminal Trial Lawyers' Association, 2010 ABCA 399, 38 Alta LR (5th) 63 [Newton], the RAD decided that a refugee appeal was not de novo and, therefore, it should defer to the RPD's findings of fact and mixed fact and law. Both the RPD's credibility assessment and its determination that Karachi was an IFA fell into that category, so the RAD applied the reasonableness standard of review.

[8]               The RAD found that the Applicants' complaints about the RPD's credibility findings mischaracterized the RPD's decision. The RPD never made a global finding that the Principal Applicant was not credible, nor did it find that there were no extremist groups in Pakistan. Rather, the RPD impugned only the Principal Applicant's testimony that the men who attacked her belonged to an extremist group. The RAD endorsed that determination by the RPD, stating that:

[36]      …If the Appellants are being targeted by an extremist group, the identity of that group is certainly very significant to their refugee claims. However, their BoC [Basis of Claim] forms do not name any extremist groups, nor did the principal Appellant do so in her initial testimony to the RPD. When pressed, she named a group that does not appear to be mentioned in the objective evidence, and then she gave a different name for the group, which is also conspicuous by its absence from country condition documents. The letters she provided do not name any group. In this context, it was reasonable for the RPD to make a negative credibility finding and to find that the Appellants had not established that the agents of persecution were affiliated with any extremist group.

[9]               The RAD also found no fault with the RPD's IFA analysis. While there was some evidence that there was no IFA for people targeted by specific extremist groups, the RAD noted that there was no reason for the RPD to comment on that given the RPD’s credibility findings. Rather, the RAD stated that only the general risks were relevant, and the RPD had reviewed the country conditions and made specific reference to the sectarian violence affecting Shia Muslims. The RAD opined (at paragraph 42 of its decision) that, while it “might have considered the same evidence and reached a different conclusion, the RPD's decision is to be reviewed on a standard of reasonableness.” The RAD found that the RPD's decision regarding the situation of Shia Muslims in Karachi and the Applicants' ability to openly practice their religion satisfied the criteria set out in Dunsmuir v New Brunswick, 2008 SCC 9 at paragraph 47, [2008] 1 SCR 190 [Dunsmuir].

[10]           Finally, the RAD was satisfied that the RPD had considered all the relevant factors set out in the Chairperson Guidelines 4: Women Refugee Claimants Fearing Gender-related Persecution (13 November 1996), even though the RPD had not referred to them specifically in its reasons. Thus, the RAD confirmed that the Applicants were neither Convention refugees under section 96 of the Act nor persons in need of protection under subsection 97(1).

IV.             Analysis

A.                What standard of review should this Court apply to the RAD’s decision?

[11]           The standard for this Court's review of the RAD's determination as to the scope of its review of RPD decisions is open to some question. As noted by Mr. Justice Simon Noël in Yin v Canada (Citizenship and Immigration), 2014 FC 1209 at paragraph 32 [Yin], the case law diverges on this issue.

[12]           Some decisions state that the correctness standard applies, either because the issue is one of central importance to the legal system and outside of the RAD's expertise, or because it affects the jurisdictional lines between the RPD and the RAD (e.g. Huruglica v Canada (Citizenship and Immigration), 2014 FC 799 at paragraphs 25-34, [2014] 4 FCR 811 [Huruglica]; Spasoja v Canada (Citizenship and Immigration), 2014 FC 913 at paragraphs 7-8 [Spasoja]; Dunsmuir at paragraphs 60-61). Other cases disagree and say that it is no more than a question of interpreting the RAD's home statute, which presumptively attracts the reasonableness standard (Akuffo v Canada (Citizenship and Immigration), 2014 FC 1063 at paragraphs 16-26, 31 Imm LR (4th) 301 [Akuffo]; Djossou v Canada (Citizenship and Immigration), 2014 FC 1080 at paragraphs 13-37 [Djossou]; Brodrick v Canada (Citizenship and Immigration), 2015 FC 491 at paragraphs 20-29 [Brodrick]; Dunsmuir at paragraph 54; McLean v British Columbia (Securities Commission), 2013 SCC 67 at paragraphs 26-33, [2013] 3 SCR 895). Questions on this issue have been certified in several of these cases and the appeal from Huruglica has now been scheduled for later this year, so this division in the case law will soon be considered by the Federal Court of Appeal.

[13]           In the meantime, I agree with Mr. Justice Luc Martineau's pragmatic approach to the issue in Alyafi v Canada (Citizenship and Immigration), 2014 FC 952 at paragraphs 46-52 [Alyafi]. As he notes and as I summarize below, the case law is divided on the scope of review that the RAD should apply. This creates a problem similar to one this Court once faced with respect to the residency test for citizenship (Huang v Canada (Citizenship and Immigration), 2013 FC 576 at paragraphs 1, 24-25, [2014] 4 FCR 436). If the correctness standard is applied by every judge of this Court, the RAD could diligently follow one line of cases only to see its decisions set aside whenever they are reviewed by a judge who favours the other line of cases. The law requires more certainty than that. As the Federal Court of Appeal observed in Wilson v Atomic Energy of Canada Limited, 2015 FCA 17 at paragraph 52, 467 NR 201 [Wilson], “the meaning of a law should not differ according to the identity of the decision-maker.” The Federal Court cannot fix the problem at the RAD level so long as the judges of this Court disagree on the solution, and applying the correctness standard in this situation would undermine the rule of law even more than if the dispute was just at the RAD level (Wilson at paragraph 52). For similar reasons, Justice Martineau concluded in Alyafi that decisions of the RAD should be upheld so long as either of the two approaches currently accepted by the Court, or any other reasonable approach, is applied by the RAD (Djossou at paragraph 91).

[14]           Adopting such an approach in this case, the RAD's decision in this case should be reviewed on a standard of reasonableness. This standard also applies to the RAD's factual findings, and its assessment of the evidence before it is entitled to deference (see: Dunsmuir at paragraph 53, [2008] 1 SCR 190; Siliya v Canada (Citizenship and Immigration), 2015 FC 120 at paragraph 20 [Siliya]; Yin at paragraph 34; Akuffo at paragraph 27; Mohamed v. Canada (Citizenship and Immigration), 2015 FC 758 at paragraph 19). The RAD's decision should therefore not be disturbed so long as it is justifiable, intelligible, transparent and defensible in respect of the facts and the law (Dunsmuir at paragraph 47). Those criteria are met if “the reasons allow the reviewing court to understand why the tribunal made its decision and permit it to determine whether the conclusion is within the range of acceptable outcomes” (Newfoundland and Labrador Nurses' Union v Newfoundland and Labrador (Treasury Board), 2011 SCC 62 at paragraph 16, [2011] 3 SCR 708 [Newfoundland Nurses]).

B.                 Did the RAD apply an appropriate standard of review to the RPD’s decision?

[15]           The Applicants claim that the RAD was wrong to apply the reasonableness standard of review. The Applicants say the RAD can be compared to the Immigration Appeal Division, and they criticize the RAD for relying on a single case, Newton, which involved a very different type of tribunal. Specifically, Newton dealt with a police disciplinary process where the relevant events occur in the past, while refugee protection is primarily concerned about the future and the assessment needs to happen at the time of inquiry (citing e.g. Fernandopulle v Canada (Minister of Citizenship and Immigration), 2005 FCA 91 at paragraph 25, 253 DLR (4th) 425).

[16]           The Applicants also attack the RAD's reasoning that its role was constrained by the limited circumstances in which it could hold an oral hearing or accept new evidence. While the proceeding before the RAD is shaped by the RPD's decision, in the sense that those are the findings being challenged by an appellant, the Applicants submit it does not follow that the RAD must completely defer to those findings. The Applicants argue that the RAD could always convoke a hearing pursuant to subsection 110(6) of the Act if it is necessary to assess credibility. In this case, the IFA was determinative and that finding was based primarily on a review of the documentary evidence, so the Applicants contend that the RPD was in no better position to make that finding than the RAD would be.

[17]           The Applicants further claim that the RAD's concern that the RPD is better placed to make credibility findings does not justify a deferential standard of review by the RAD. The Applicants say it is inappropriate and wasteful for the RAD to adopt administrative law standards of review and simply duplicate the function of the Federal Court. The standards of correctness and reasonableness were developed to govern the relationship between the executive and judicial branches of government, and the Applicants argue that different considerations apply when both the lower and higher tribunal belong to the executive branch (British Columbia Society for the Prevention of Cruelty to Animals v British Columbia (Farm Industry Review Board), 2013 BCSC 2331, 67 Admin LR (5th) 152 [BC SPCA]). The Applicants point out that many decisions of this Court have set aside decisions of the RAD which have applied the reasonableness standard for similar reasons (citing e.g. Huruglica at paragraphs 43-45).

[18]           The Respondent submits that the RAD should review RPD decisions for palpable and overriding error (citing Spasoja at paragraph 40). In its view, that conclusion is supported by the statutory scheme, the legislative history of the RAD, and the central role of the RPD in determining refugee claims and finding the facts. The Respondent points out that the RPD is the only tribunal to which all eligible refugee claims are referred, and it makes decisions on cessation and vacation of claims that will never go to the RAD.

[19]           Furthermore, the Respondent argues that the Act expressly requires the RPD to conduct an oral hearing, and the RPD is actively involved in creating the record by questioning witnesses, distributing country conditions documents, requesting information from the IRB Research Directorate, and inquiring into “any matter that it considers relevant” (Act, s 170(a)). These differences make the RPD much better suited than the RAD to make findings of fact. Because it hears and elicits evidence, the Respondent says that the RPD is more familiar with a refugee's claim as a whole, and this is historically one of the reasons that appellate courts defer to trial courts on factual questions (citing Housen v Nikolaisen, 2002 SCC 33 at paragraph 14, [2002] 2 SCR 235 [Housen]). In the Respondent's view, the RAD should only substitute its own opinion for that of the RPD on questions of law.

[20]           The Respondent also states that, while the RAD can substitute its disposition for that of the RPD's in some circumstances, the Federal Court of Appeal can do the same with respect to decisions of the Federal Court (Federal Courts Act, RSC 1985, c F-7, s 52(b)(i) [FCA]); and yet, that power alone does not demand an appeal de novo. Neither does the forward-looking assessment of risk in the refugee process require a completely new assessment of the evidence by the RAD when it has not had the benefit of hearing from the claimants directly. Thus, in situations where there is no new evidence, the Respondent submits that the RAD is intended to serve a true appellate function and search only for palpable and overriding error when reviewing findings of fact. As that is functionally equivalent to the reasonableness standard, the Respondent argues that the RAD did not err by referring to Newton or Dunsmuir (citing Housen; and HL v Canada (AG), 2005 SCC 25 at paragraph 110, [2005] 1 SCR 401).

[21]           As mentioned above, the judges of this Court disagree on the correct interpretation of the RAD's standard of review in respect of the RPD's findings of fact and mixed fact and law. One line of cases concludes that the RAD should review the RPD's findings of fact for palpable and overriding errors (see e.g.: Eng v Canada (Citizenship and Immigration), 2014 FC 711 at paragraphs 26-34; Spasoja at paragraphs 14-46; and Triastcin v Canada (Citizenship and Immigration), 2014 FC 975 (CanLII) at paragraphs 27-28). Another line of cases concludes that the RAD must independently come to a decision and is not limited to intervening on the standard of palpable and overriding error, although it can “recognize and respect the conclusion of the RPD on such issues as credibility and/or where the RPD enjoys a particular advantage in reaching such a conclusion” (Huruglica at paragraph 55; Yetna v Canada (Citizenship and Immigration), 2014 FC 858 at paragraphs 16-20; and Akuffo at paragraph 39; Ozdemir v Canada (Citizenship and Immigration), 2015 FC 621 at paragraph 3).

[22]           In this case, the RAD clearly adopted the factors found in Newton in determining the appropriate standard of review in the appeal. Indeed, citing Newfoundland Nurses and Dunsmuir, the RAD adopted a position of deference and applied a reasonableness standard of review to the RPD's decision, stating as follows:

[29]      For these reasons, the RAD concludes that, in considering these appeals, it must show deference to the factual and credibility findings of the RPD. The notion of deference to administrative tribunal decision-making requires a respectful attention to the reasons offered or which could be offered in support of the decision made. Even if the reasons given do not seem wholly adequate to support the decision, the RAD must first seek to supplement them before it substitutes its own decision.

[30]      The appropriate standard of review in these appeals is one of reasonableness. Reasonableness is concerned mostly with the existence of justification, transparency, and intelligibility within the RPD’s decision-making process, but also with whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and the law.

[23]           Both lines of cases noted above have condemned this approach by the RAD (see e.g. Alyafi at paragraphs 17-18, 39 and 46; Huruglica at paragraphs 45 and 54; Spasoja at paragraphs 12-13, 19-25 and 32-38; Djossou at paragraph 37; Brodrick at paragraphs 32-34; and Ching v Canada (Citizenship and Immigration), 2015 FC 725 at paragraphs 48, 50). The RAD has an appellate function, and it cannot limit its analysis merely to whether the RPD acted reasonably and reached a decision that fell within the range of acceptable outcomes which are defensible in respect of the facts and the law. Applying the reasonableness standard, as the RAD did in this case, is typically an error, since it denies to appellants the appeal to which they were entitled (Alyafi at paragraph 46; Siliya at paragraph 23; Djossou at paragraphs 38-44). Accordingly, the Applicants have established that they have grounds for relief under paragraph 18.1(4)(c) of the FCA.

[24]           This conclusion, however, need not always end the matter (see e.g. Pataraia v Canada (Citizenship and Immigration), 2015 FC 465 at paragraphs 13-14; Ali v Canada (Citizenship and Immigration), 2015 FC 500 at paragraphs 8-9). The issue becomes whether relief should be withheld in the face of such an error (see: Lemus v Canada (Citizenship and Immigration), 2014 FCA 114 at paragraph 38, 372 DLR (4th) 567), which in turn reduces to this question: might the RAD have reached a different result had it selected an appropriate standard of review?

C.                 Were the RAD’s determinations on credibility and IFA unreasonable?

[25]           The Applicants criticize the RPD's credibility finding, and they say the RAD should have considered whether it could grant an oral hearing. In the Applicants' view, the RPD's finding was not made in clear and unmistakable terms and no deference is owed to it by the RAD.

[26]           The Applicants also challenge the RAD's decision regarding the IFA. The documentary evidence described widespread violence against Shia Muslims in Karachi and stated that hundreds are killed every year, but the RPD implied that the Applicants had to prove that every Shia Muslim in Pakistan was being persecuted. The RAD never explained why it upheld that finding by the RPD, apart from saying that it might not have reached the same conclusion and could not reweigh the evidence. However, the Applicants argue that is precisely what the RAD should have done.

[27]           The Respondent argues that the Applicants cannot succeed unless they can show it was unreasonable for the RAD to decide that neither the RPD's credibility finding nor its IFA determination were palpable and overriding errors. The Respondent contends that the RAD properly recognized that the RPD did not ignore any evidence and explained that decision sufficiently enough to satisfy the criteria set out in Newfoundland Nurses (at paragraphs 14 and 16). The Respondent also says that the RPD never applied an inappropriate standard of proof; the Principal Applicant simply failed to establish any link between her attackers and any extremist organization with credible evidence. The Respondent further argues that the RAD was not required to reassess the Applicants' credibility. It was sufficient that the RAD noted that the RPD based its adverse decision on a number of detailed factors, and gave reasons for rejecting each of the Applicants' objections.

[28]           While the RPD's credibility finding might survive no matter what standard of review was applied by the RAD in this case, the same cannot be said for the IFA finding. On the contrary, when it considered whether the violence against Shia Muslims in Pakistan was widespread enough that the Applicants would not be safe in Karachi, the RAD stated (at paragraph 42 of its decision) that “it is possible that the RAD might have considered the same evidence and reached a different conclusion.” The RAD refused to conduct its own analysis of the evidence in this regard expressly because it was applying the reasonableness standard. Under the approach set out in Huruglica, that is an error since that particular finding depended only on documentary evidence which the RAD could assess just as well as the RPD could (see Kurtzmalaj v Canada (Citizenship and Immigration), 2014 FC 1072 at paragraphs 28-29 and 34).

[29]           The Respondent argues that the RAD’s approach could still be appropriate if the palpable and overriding error standard is applied, at least insofar as there are similarities between that standard and the reasonableness standard (see Akuffo at para 38). This argument, however, presumes that the RAD member, informed that it was an error to apply the reasonableness standard, would have followed Spasoja and not Huruglica. Nothing supports that presumption.

[30]           Furthermore, I also reject the Respondent's argument that the Applicants need to prove that both the credibility finding and the IFA finding were erroneous in order to attract relief. The credibility finding related only to the RPD's determination that the attackers did not belong to an extremist organization, and thus influenced the IFA decision only insofar as it permitted the RPD to disregard evidence about the reach and influence of such organizations. The other part of the IFA finding was independent of that and based solely on the Applicants' profiles as Shia Muslims and the risk they would be targeted by random sectarian violence in Karachi. It was that part of the finding about which the RAD acknowledged that it might have reached a different conclusion had it re-weighed the evidence.

[31]           Thus, it is possible that the RAD would have reached a different result had it not erroneously applied the reasonableness standard of review, and it would be inappropriate to withhold relief from the Applicants.

V.                Conclusion

[32]           In view of the foregoing reasons, the RAD's decision cannot be justified and is not defensible in respect of the facts and the law. It must be set aside and the matter returned to the RAD for a new determination.

[33]           In the result, therefore, the Applicants' application for judicial review is granted. No question of general importance is certified.

JUDGMENT

THIS COURT’S JUDGMENT is that: the application for judicial review is allowed; the matter is returned to the Refugee Appeal Division for redetermination by a different panel member; and there is no question of general importance for certification.

"Keith M. Boswell"

FEDERAL COURT

SOLICITORS OF RECORD

APPEARANCES:

SOLICITORS OF RECORD: