Printer Friendly



Canada's refugee status determination process is in a continual state of flux. Since 1976, there have been four major modifications and many more minor ones. (1) The first came in 1989 when the initial, backlog-prone system of determination based solely on a transcript of the claimant's testimony established in 1976 was completely overhauled. The main features of the change wete an oral hearing before a quasi-judicial administrative tribunal staffed by political appointees, a triage system designed to weed out fraudulent claims, and a more effective removals process. In 1993, important changes were made to the appeals process for claimants, putting judicial review of asylum decisions in the hands of trial-level as opposed to appellate-level judges. In 2002, stringent security measures were introduced, ostensibly to prevent the asylum system from being infiltrated by terrorists or criminals. Although the quasi-judicial hearing was maintained and complementary forms of protection were enhanced, the system again became subject to crippling backlogs in the late 2000s. (2)

The period between 2010 and 2012 saw another round of sweeping changes. (3) This reform project transformed initial decision makers from political appointees to members of the public service, and instituted an internal administrative appeal mechanism, a truncated, inferior process for claimants from so-called "safe" countries of origin, faster timelines, mandatory detention of maritime and group arrivals, restrictions on access to complementary forms of protection, and restrictions on access to Canada's public healthcare system. Several of these initiatives were overturned by the courts, (4) and there has been doubt cast as to whether these reforms were effective. (5) In 2017, mostly in response to increasing delays and backlogs, the Canadian government announced that it will be reviewing the refugee system yet again. (6)

As this chronology of events suggests, the impulse for reform persists, but the tesults nevet seem to be satisfactory. It is high time to stop tinkering and think seriously about finding a model for refugee status detetmination in Canada that is sustainable, and this article attempts to do that. The proposals put forward here reflect the viewpoint of someone with 30 yeats' experience on the front lines of refugee status determination both as a practitioner and decision maker. These proposals are informed by first-hand observation of hundreds of refugee hearings and interviews, extensive discussions and communications with refugee advocates and decision makers, and in-depth knowledge of refugee jurisprudence. Advancing ideas for reform is necessarily a hypothetical exercise, at least to some extent. Nevertheless, every effort has been made to bolster all assertions with empirical evidence and to address any unfounded assumptions and intrinsic biases that may be inherent in a practitioners worldview. The goal is to initiate a conversation around sustainable reform of Canada's asylum system and to prompt further research. In this regard, a front-line perspective such as the one set out here is an important part of the mix and should be evaluated on its own terms.

We will begin by briefly reviewing the various dimensions of refugee status determination in Canada, including the current structure and the criticisms which have been levelled against it. We will then put forward proposals for structural reform that attempt to address these criticisms. The imagined end result is a refugee status determination model that may seem uncomfottably unfamiliar. Nonetheless, such discomfort is to be embraced, as disrupting the established conventions around asylum may be the only thing that will serve as a catalyst for sustainable solutions.


Refugee status determination is one of Canada's international obligations. According to the 1951 Convention Relating to the Status of Refugees (7) (the "Convention"), when an asylum seeker arrives on our shores, Canada is required to determine whether he or she:
owing to well-founded fear of being persecuted for reasons of race,
religion, nationality, membership of a particular social group or
political opinion, is outside the country of his nationality and is
unable ot, owing to such fear, is unwilling to avail himself of the
protection of that countty... . (8)

Some signatories to the Convention left the application of this definition to the United Nations High Commissioner for Refugees (UNHCR). Most others, Canada included, establish a refugee status determination system within their existing legal systems. Needless to say, the essential function of any refugee status determination system is to identify those genuinely in need of asylum.

In Canada, initial determination is done by one division--the Refugee Protection Division (RPD)--of a quasi-judicial administrative tribunal called the Immigration and Refugee Board (IRB). RPD members are civil servants. (9) The claimant fills out a written information form and then tells his or her story face-to-face with an RDP member in what resembles a mini 'trial' in which he or she is assisted by a lawyer and an interpreter. Error-correction mechanisms (10) consist of an appeal to an internal administrative appeal body within the IRB known as the Refugee Appeal Division (RAD), introduced in 2012, and/ot an application for judicial review (with leave) to the Federal Court. RAD members are political appointees and 10% of them are required to be lawyers. (11) Access to the Federal Court of Appeal is limited to serious questions of general importance. (12) Two forms of what we might categorize as complementaty post-determination protection mechanisms, the Pre-Removal Risk Assessment (PRRA) (13) and the humanitarian and compassionate application, are available in some circumstances. (14) Both of these protection mechanisms are normally paper-based applications conducted by an immigration official, and can themselves be challenged by way of judicial review at the Federal Court, but again, only with leave.

The features of the asylum process are essentially the same across all states that have signed the Convention: some sort of face-to-face encounter (an oral hearing or an interview) where an initial determination of the asylum claim is made, followed by one or more error correction mechanisms and the possibility of accessing complementary protection regimes. (15) In the vast majority of signatory states, the initial determination proceeding takes the form of an interview, Canada is quite unique in that it affords claimants a full-blown hearing before an administrative tribunal right at the outset of the claim process.

Qualitatively speaking, determining who is and is not entitled to asylum is a daunting task. A United States Court of Appeals judge once noted that, "[a]sylum petitions of aliens seeking refuge from alleged persecution are among the hardest cases faced by our courts." (16) The difficulty arises not only from the complex legal definition that must be applied, but also from the fact that assessment of credibility in the asylum context is especially hard. It is well documented that fear, trauma, mistrust, culture shock, and even mourning can frequently undermine asylum seekers' ability to express themselves. (1)" For this reason, decision makers require a broad range of knowledge in order to negotiate these obstacles. As has been noted,
Les habiletes exigees par ce contexte constituent un tout qui ne
s'apprend pas "sur le tas" ou par une formation continue, meme de
qualite. Une competence specifique, acquise par une experience
professionnelle ou benevole prealable, est essentielle, sans laquelle
la valeur de l'ensemble du processus decisionnel est en cause. (18)

To illustrate the difficulty, imagine being faced with a claimant who alleges she was taped during an ethnically motivated communal riot. She has arrived in Canada with a baby suffering from a complex medical condition. She has no documents to prove her identity, has very little education and has great difficulty testifying in a cohetent manner. Her counsel is aggressively raising several procedural objections, including problems with the interpreter and a request that testimony be dispensed with because of possible te-ttaumatization. In this scenario, the initial decision maker is faced with a number of judgment calls. First, what should be done with counsel's procedural requests and how should counsel's aggressive attitude be dealt with? Second, given the paucity of solid evidence as to the facts of the case, should the claimant be given the benefit of the doubt ? Could any incoherent testimony given be the result of trauma? Is her account of the surrounding circumstances of the rape consistent with known information on her country of origin? Third, assuming her story is true, does she meet the definition of a Convention refugee ? Did the rape occur by reason of a Convention ground? Does rape by a private individual during a communal riot constitution "persecution" given the fact that no state agent was directly involved? Does she need to establish that she faces further violence, or is the past instance enough to make her fear well-founded? Could her ethnicity alone give rise to a well-founded fear of persecution or would she simply face discrimination? Would she be able to move to another area within her country where she would be threat-free, such as an area where her ethnic group was in the majority?

Practically speaking, these types of determinations require in-depth knowledge of administrative law, country conditions worldwide, case law, and UN guidance around the refugee definition. They also require knowledge of the legal hallmarks of a "credible" story and knowledge of how trauma and cultural differences can affect the telling of a story. Our hypothetical claimant's life, liberty, and security turn on the quality of such knowledge, as well as the quality of the decision maker's integrity, disposition, and judgment. If the initial decision is negative, it seems only fair that it be scrutinized on its merits. If the refusal is maintained, it seems only fair that there be an adequate forum to address the baby's medical situation, the possibility of ethnic discrimination, or any new developments in the country of origin prior to being removed from Canada. And it seems only fair that this young woman and her child not be left in limbo for years waiting for a final decision on whether they can make Canada their new home. Looking through the eyes of our hypothetical claimant, we can understand why the quality and nature of the various intersections with the asylum system become so pivotal. With little or no margin for error, it is crucial that the structure of the asylum system be designed to meet all the challenges inherent in refugee cases.


Canada's refugee status determination system has quite frequently been the object of criticism in both the public and private space. (19) Emotions run high when claimants, like our hypothetical rape victim, find themselves refused asylum and facing removal. Most of the ire has focused on the quality of first-level decision makers, insufficient error correction mechanisms, stractural complexity, and delay. (20) These criticisms have come from both governmental and non-govemmental organizations (NGOs), and frequently take the form of assertions that the system is "broken". (21) While that characterization is not entirely accurate, the specific criticisms mentioned above do need to be addressed, and we will briefly do so now.


As the Canadian government recognized in 1986 when it was attempting to reframe the refugee status determination system to conform with constitutional guarantees of due process "[i]n refugee matters, a wrong decision can mean death." (22) The question of who decides is crucial. It is therefore not surprising that the locus of many of the complaints around Canada's refugee status determination system is the IRB itself--specifically the quality of IRB decision makers. Although the large majority of board members are generally described by the refugee bar as polite and competent, (23) given the "life and death" nature of refugee status determination, the margin of error is small and the minority who do not do their jobs well can wreak havoc on many individual lives.

Traditionally, IRB members have been political "patronage" appointees, without legal training or expertise in refugee law. (24) This has led to skepticism as to the validity of the IRB's determinations. The low point in terms of public criticism came shortly after the IRB's inception, when the media referred to it as a "Kangaroo court". (25) A review by Professor James Hathaway determined that "far too many" (26) IRB members were "unfit to exercise their responsibilities by reason either of simple incompetence or a negative attitudinal predisposition", (27) that the board's administration displayed a "lack of a juristic culture", (28) and that a shift to a meritocratic appointment and re-appointment process needed to begin immediately. (29)

Some improvements were made in the 2000s,'" but by 2009 Audrey Macklin noted that, "there can be little doubt that the overall level of competence is lower than what might be achieved from a purely merit-based system." (31) Pursuant to the 2012 reforms, RPD members became civil servants, not political appointees, and a more transparent process has reduced the risk of patronage. Nonetheless, there is still no absolute requirement that IRB members at the first-instance RPD level have either a legal background or subject matter expertise.3 (2) Accordingly, despite a significantly higher acceptance rate since the 2012 changes," concerns about the quality of RPD decision making in individual cases have not yet been tesolved. For example, in August 2015, the executive ditector of the Canadian Council for Refugees noted that: "Thete's perhaps a slightly higher level of confidence in the board members." (34) In 2017, the Refugee Lawyers' Association expressed concern that the government was opting for an "appointment process that is political and discretionary instead of going for a transparent process to appoint the most suitable candidates who are competent, judicious, fair-minded and efficient". (35) First-hand observation of current RPD decisions and commentary from the refugee bar show that serious issues still exist in some cases--including inappropriate focus on marginal or irrelevant factors, unfair discounting of evidence and failure to correctly apply the appropriate legal tests or standards.

In terms of academic thought on the question, Kirmayer believes that first-instance IRB decision makers have tended to suffer from "failures of imagination". (36) Rousseau et al more concretely identify legal, psychological and cultural factors as being at the heart of IRB dysfunction. (37) In their evaluation of 40 refugee cases, they found that members had difficulty administering and assessing the evidence, understanding the political and social conditions in other countries, interpreting administrative and international law and understanding the rules of politeness and decorum. (38) They also found "massive avoidance of traumatic content", (39) cynicism and secondary trauma symptoms, as well as significant rates of cultural misunderstanding, prejudice and stereotyping. (40) A further study of the IRB conducted by Crepeau and Nakache (41) argues that the IRB never truly succeeded in becoming a so-called "critical space" (42) impeding the development of institutional professionalism. (43) While these studies are obviously quite old and do not deal with the new crop of board members appointed in 2012, complaints from the refugee bar in the same vein persist to this day, as mentioned earlier. We have seen the high level of skill, knowledge, and sensitivity necessary to determine the claim of our hypothetical rape victim; it is alarming to think that some decision makers do not measure up.

It is also concerning that the IRB has failed to develop a body of regularly cited, persuasive refugee law jurisprudence in its over 25 years of existence. To date, there has been no specific research on this issue, so one can only hypothesize as to why this might be the case. The phenomenon might indicate that the IRB is not conceived of as having a "law-making" function. This would be a sharp contrast to both its predecessor, the Immigration Appeal Board (IAB), (44) and the Board of Immigration Appeals (BIA) in the United States, (45) which have both produced a body of case law that is referenced by both counsel and the courts. The lack of regular ciration might also suggest that IRB decisions are so divergent as to be impossible to synthesize into a coherent body of opinion, (46) or that there is a lack of respect for the institution on the part of counsel. While both the RPD and the RAD have the power to convene three-member panels (47) (ostensibly for cases involving a serious question of law) this is not frequently done."4 (8) Again, specific research is needed in this area, but these observations could arguably suggest a suboptimal level of engagement with the subject matter--and may be a reflection of the "lack of a juristic culture" encountered by Hathaway during his IRB review of 1993. (49) Indeed, the IRB's failure to require its members to possess either subject-matter expertise or a legal background makes it an anomaly among comparable federal administrative tribunals in Canada. (50) Suboptimal initial decision making continues to be one of the most serious criticisms levelled at Canada's asylum system.


Another element that has been traditionally raised in critiques of the refugee status determination system is the lack of an effective error-correction mechanism. As we have seen with our hypothetical claimant, merits review is a life-line in cases of refusal. Significant progress was made in 2012 on this front. The implementation of the RAD introduced a fact and merits-based nunc pro tunc appeal into Canada's asylum process for the very first time. Nonetheless, this remedy is only as effective as the quality of the decision makers. As previously indicated, the RAD is staffed by political appointees, and there have been some allegations of patronage." More significantly, the RAD selection criteria do not necessarily require either legal training or subject-matter expertise. (52) The quality of decision making on appeal is therefore a cause for concern, although initial signs are positive: the RAD overturned 33% of negative RPD decisions in 2016. (53)

For those claimants who do not have access to the RAD, (54) or who are not successful there, judicial review with leave of the Federal Court is the next available recourse. Accordingly, the remedy at the Federal Court remains vitally important, but overall success rate at the Federal Court for refused protection seekers has been notoriously low--around 7%. (55) The leave requirement is considered by many practitioners to be largely to blame for this, and has come under intense scrutiny over the years. Refugee lawyers despise it. It has been referred to as a "black hole". (56) Audrey Macklin has noted that it "directly and incontrovertibly breaches" the "guarantee of access to an independent and impartial court", which is "[o]ne of the foundational tenets of the rule of law". (57) Studies of Canada's leave requirement show widely divergent grant rates among individual judges (58) and grant rates potentially based on factors unrelated to legal issues, such as the gender and country of origin of the applicant. (59) Even where leave is granted, the Federal Court does not conduct a merits review of the asylum determination: the remedy is merely judicial review on a "reasonableness" standard. This can be cold comfort for many claimants--and even a source of frustration for many judges. (60) Error correction should be squarely within the sights of any reform initiative.


Yet another criticism levelled against Canada's asylum process is that it has traditionally incorporated a multiplicity of proceedings and is therefore too slow and too complex. This criticism is made to some extent by both refugee advocates as well as government. Scheduling initial determination hearings promptly has proven extremely difficult. Yet one study found that claimants find it very difficult to wait long periods of time (sometimes several yeats) for hearings in part because of intense feelings of liminality: "You're always nervous, you're always stressed, you are not comfortable because you'te still not part of anything. You're not part of your country anymore because you cannot go back thete, and you're not part of where you are." (61) Some reported stress so severe, it impaired their ability to sleep of prompted them to seek medical attention. (62)

But, as we have seen, the claim process does not end with the IRB hearing: complementary forms of protection (PRRA and humanitarian relief) may be available as well and are in turn subject to review by the courts. In this way, a single asylum claim can entail three separate proceedings before the federal courts--not to mention ancillary applications for stays of removal. The availability of these multiple recourses and reviews has the potential to lengthen the process considerably--sometimes for over five years, as mentioned earlier. (61) From the perspective of the asylum seeker, the multiple proceedings are commenced in an attempt to rectify an unreasonable and incorrect first instance denial of a safe haven. From the government and community perspective, multiple avenues of recourse allow the refugee claimant to indefinitely delay his or her deportation--sometimes at the taxpayer's expense, if legal aid, health benefits, or welfare benefits are needed.

While streamlining was attempted in the 2010-12 legislative changes (RPD hearings within 60 days, RAD decisions within 90 days, restricted access to complementary protection mechanisms), (64) it has not truly succeeded because no provision was made for realistic numbers of claims. (65) The 2012 introduction of an additional step in the process, namely the internal appeal to the RAD available to most claimants, has the potential to neutralize stteamlining gains made by limiting so-called complementary forms of protection once numbers rise to historical levels.

Both delay and complexity take their toll on claimants. To someone like our hypothetical rape victim, the process would appear to be an incomprehensible legal maze and a formidable obstacle to navigate in a context where her primary concerns would be recovering from sexual violence, adapting to a foreign cultute, and caring for a sick child. She would most certainly need a lawyer to successfully forestall removal, but might not be able to afford one. (66) Long periods of time spent fighting for the resolution of an asylum claim has been found to have a profound psychological impact. (67) Concerns around these issues are well-founded and must be addressed if reform is to succeed.


It is now time to look at how the pitfalls of the past can be avoided in the future. Inspired by academic thought, other asylum systems, and other successful Canadian adjudicative structures, we will make the case for radical reform of refugee status determination in Canada. An alternative model for refugee status determination in Canada must be explored. This new model must respond to the challenges inhetent in refugee status determination--even if in so doing, it questions some of the assumptions underlying reform proposals of the past.

Indeed, as noted at the outset, Canadians have not been shy about proposing changes to the asylum system. One of the most notable proposals was that of Rabbi Gunther Plaut in 1985 which--following the Supreme Court of Canada's decision in Singh etalv Canada (Minister of Employment and Immigration) (68)--advocated a single oral hearing before a quasi-judicial, community-based refugee tribunal, followed by a right of appeal to the federal judiciary with leave. (69) Plauts recommendations were largely implemented in 1989 and the current IRB-centered system was created. Only four years later, as mentioned earlier, came the Hathaway report, which contained 40 recommendations for reform of refugee status determination in Canada. (70) The Law Reform Commission of Canada also issued a report in 1992, (71) and Davis, Kunin, and Trempe did likewise in 1998. (72) In 2009, Peter Showier produced Fast, Fair and Final, (73) which recommended speeding up processing times and adding an internal appeal layer to the process, as well as dramatically increasing the number of board members with legal training. In 2017, yet another review of IRB functioning will begin, as noted at the outset. (74)

Plauts proposals were arguably the most transformative. His scheme made the leap from an executive/administrative model for initial determinations (a Ministerial committee recommending asylum based on a written transcript of a claimant's testimony) (75) to a quasi-judicial adjudication model (initial determinations by an administrative tribunal). Subsequent reform proposals never really questioned this basic quasi-judicial model for refugee status determination, or looked at other facets of asylum comprehensively. This is not intended to be a criticism ; the mandate of many of these studies was quite restricted and limited to the issues and conditions prevailing at the time. In contrast, the proposals that will be put forward here do question underlying assumptions and entrenched practices, and look at the process much as a real claimant would--i.e., as a unified whole.


We will explore, on the structural level, the possibility of establishing a new initial asylum determination body, a new error correction mechanism, a simplified overall concept and a system of reasonable timelines. We will describe what any new structures could look like, explore their feasibility, and explain why they might represent an improvement over both the status quo and previous reform proposals. Although the argument in favour of these proposals will be supported by evidence wherever possible, it will be hard to avoid entering the realm of the purely hypothetical at times. Nonetheless, any opinion put forward is done in good faith and informed by three decades of observation and practical experience. This endeavour is intended to stimulate discussion and spur further research--ideally by academics and policy makers.

In brief, the suggestion is to 1) replace the RPD with either an Asylum Court or an Asylum Tribunal that has comprehensive protection jurisdiction (including cettain humanitarian considerations and non-Convention related risks); 2) institute a system of reasonable timelines; and 3) abolish the Federal Court leave requirement and institute an appeal in writing to the Fed et al Court of Appeal as of right. The proposed structure is represented in Figure 1 :



Because of its inherent complexity, it stands to reason that refugee status determination should take place in a high-knowledge legal forum. Accordingly, it is time to consider moving away from an administrative or quasi-judicial model for refugee status determination and towards a wholly judicial one. This is a radical idea, but it can be supported. It is not controversial that refugees and host community alike strive for fair, appropriate, and reasonable decisions, and judges are ostensibly the most qualified decision makers in the legal sphere. Hathaway, in his 1992 report, advocated for a more "juristic" culture at the tribunal of first instance at least eight times. (76) Theoretically, population by jurists could be one way to achieve this. In the context of the Canadian asylum system, a concept along these lines would replace the Refugee Protection Division of the IRB with a specialized wholly judicial body. (77) We could call it the Asylum Court of Canada.

This is not a new idea; after an extensive study, the American Bar Association (ABA) made essentially the same proposal in 2010 as a substitute for non-independent Office of the Immigration Judge. (78) The concept of a wholly judicial determination system was also recommended for further study by the Australian Senate over ten years ago. (79) The ABAs decision to recommend moving towards a wholly judicial model was based on the following considerations:
Immigration judges already make binding decisions in cases or
controversies involving individual rights and liberty--a job that is
normally assigned to a formal court father than an administrative
Courts are experienced with providing judgments affecting unpopulat of
isolated minorities....
The existence of an Article I court [such as the US Tax Coutt, the
Bankruptcy Court, etc.] may encourage a more uniform development and
application of law....
The Article I model is likely to be viewed as more independent than an
agency because it would be a wholly judicial body, is likely, as such,
to engender the greatest level of confidence in its results, can use
its greater prestige to attract the best candidates for judgeships, and
offers the best balance between independence and accountability to the
political branches of the federal government. Given these advantages,
along with the othets discussed . . . the Article I coutt model has
been selected as the preferred option. (80)

The Australian Senate's position on the issue was as follows:
[W]ork should also be undertaken on the feasibility of a wholly
judicial detetmination process. An objective of this would be to assess
if such a process could be mote open and transparent than the current
multi-tiered system, which the majority of the Committee considers has
been highly criticized. (81)

The idea has also been previously floated in the Canadian context, but not in academic literature or an official report. (82)

Although a non-starter for at least one well-respected academic, (83) specialized judicial involvement in first-instance asylum decisions has enough credible support to be worth exploring. In practical terms, this vision of a wholly judicial site for initial refugee status determination might include the same sorts of structures forjudge selection that exist in courts of law, plus screening for specialization. Refugee status determination hearings could proceed before one specialized judge and continue to be informal and non-adversarial.

But is the idea of replacing the IRB with an Asylum Court truly viable ? The answer could turn on the fact that Canada has done something similar before. In 1983, the Tax Court of Canada was created to replace the Tax Review Board, an unsatisfactory administrative model for de novo taxation appeals, and this could be a relevant model for our purposes. The Tax Court is a specialized wholly judicial tribunal that sits in appeal de novo of decisions of the Canada Revenue Agency. Proceedings before the Tax Court are similar to trials and many taxpayets are unrepresented. The Court has instituted the Informal Procedure which applies to less sophisticated litigants. (84) The Tax Court is, accordingly, regularly required to assess the credibility of witnesses at least to some extent. Decisions of the Tax Coutt may be appealed as of right to the Federal Court of Appeal. Qualifications for Tax Court judges are governed by section 4 of the Tax Court of Canada Act, RSC 1985, c T-2, and require either the status of a superior court judge or a lawyer with 10 years' experience. In practice, every single Tax Courtjudge is a former tax lawyer--from either a private law firm or the Depattment of Justice. (85) Thus, the Tax Court is inhabited by specialist judges and has developed a substantial body of jurisprudence which is regularly cited by litigants.

As we have indicated, the Tax Court replaced the Tax Review Board because the lattet tended to be seen by the general public as insufficiently independent of government. (86) Thus, the Tax Court was in some measute the cute for a deficient administrative tribunal, and represented a shift from a quasi-judicial model to a wholly judicial one. The creation of the Tax Coutt has turned out to be an excellent idea. A recent article describes the Tax Court in glowing terms:
The court has garnered respect from both the general public and the tax
From the perspective of the general public, the accessible and flexible
procedures available under the informal procedure allow for relatively
straightforward appeals to be heard regardless of location or the
amount at issue. At the same time, the genetal procedure permits more
complex tax appeals to be determined in accordance with a formal
process better suited to complex cases involving sophisticated
taxpayers where the amounts at issue of the principles involved are
Over time, the court has consistently demonstrated that it is a fait,
neutral arbiter The court's judges admirably navigate complex and often
daunting provisions of various taxing statutes to ensute that justice
is done....
Within the tax community, the court has been recognized for the
superior quality of its judges and the decisions that they tender.
Judicial appointments to the court reflect its status as a prestigious
institution and include private practitionets at the height of their
In respect of its judgments, the court has been acknowledged for its
role as the court of first instance that must deal with novel,
difficult, and potentially far-reaching cases, and its ability to
artive at the correct result in very challenging circumstances. In
particular the court has played a significant role in developing tax
The respect that the Tax Court commands in the community is
deserved... . The court is well positioned for a bright future. (87)

Can the same be said of the IRB?

Unfortunately, as we have seen, the IRB does not always garner respect within the community and is not populated with "private practitioners at the height of their practice." (88) It is therefore worth exploring whether an Asylum Court modeled on the Tax Court would represent an improvement not only on the status quo, but also on other reform proposals that leave the quasi-judicial model of lay decision making in place.

In this regard, it could be argued that judges are professional decision makers selected for their knowledge and temperament, and should in theory make better decisions than quasi-judicial lay decision makers. Certainly, that was the thinking behind the creation of the Tax Court. As noted above, refugee status determination is the one of the most complex adjudicative processes in the Canadian legal system and it follows that it should be petformed by the highest-level decision makers society has to offer. Consequently, there is some logical force to an argument that "real" judges would be well-suited to tackle the inhetent difficulty and high stakes which asylum determinations present. And, as noted earlier, if we require these "real" judges to be specialized judges, the case becomes even stronger. More to the point, if the switch from a quasi-judicial model to a wholly judicial one was an improvement for de novo tax appeals, why should it be questioned for first-instance asylum determinations which are so much weightier?

A move to a wholly-judicial model seems justified based on our earlier examination of the criticisms levelled against IRB decision making. At root, these criticisms seem to be based on frustration around a lack of judiciality--i.e. around the failure of IRB members to act more like judges. This failure is borne out by observation and practical experience, which suggests that areas of friction between individual IRB members and the refugee bar tend to center on a lack of sensitivity to legal rights and doctrines. For example, a recent media account of an IRB detention hearing shows the lawyer straggling to defend his client's disclosure rights to the presiding member, who instead characterized counsel's disclosure request as a "fishing expedition". (89) The incident presents a good example of the tensions between counsel's desire for the rules of natural justice to be followed (as befits a judge) and the member's desire to proceed quickly and informally (as befits an administrative agent). (Another even more striking illustration was the hearing during which a lawyer told an IRB member: "If you were a real judge you would understand my argument.") (90) Given the fairly commonplace frustrations felt by counsel appearing before lay asylum decision makers, it is not surprising that the model selected by the ABA (a group of practitioners) was a wholly judicial model; the quasi-judicial model was rejected.

As noted earlier, the view that a lack of judiciality at the IRB is a lingering concern of stakeholders is also supported by academic research. This research has highlighted complaints that IRB members lack an open mind; that they have difficulty administering and assessing the evidence and interpreting administrative and international law; that they engage in stereotyping." Interestingly, the qualities required of professional judges are quite the opposite; judges
should be knowledgeable about the law, willing to undertake in-depth
legal research, and able to write decisions that are clear and cogent.
Their judgment should be sound and they should be able to make informed
decisions that will stand up to close scrutiny. Judges should be fair
and open-minded, and should appear to be fair and open-minded. They
should be good listenets but should be able, when required, to ask
questions that get to the heatt of the issue before the court. (92)

If a deficiency identified is that the IRB is not acting judicially enough, replacing it with something like the Tax Court would seem to be the common-sense solution. Academics have warned that judges are not perfect and that specialization is not a panacea, (93) but the fact remains that one real-life shift from a quasi-judicial (Tax Review Boatd) to a wholly judicial model (the Tax Court of Canada) has been a success.

There are, to be sure, valid concerns associated with such a major paradigm shift. One such concern is rooted in the conviction that thete can be "no going back" from administrative tribunals replacing generalist courts. In this regard, it is important to note, first of all, that the proposal for an Asylum Court is not modeled on a generalist court, but on the Tax Court, which employs specialized judges. Nonetheless, the "administrative tribunal vs. generalist court" issue needs to be addressed.

While many academics may tend to favour administrative tribunals over generalist courts for their flexibility and expertise, (94) it is important to consider that what may be true in most other contexts is not necessarily true in the context of the asylum system. That particular context involves initial decision makets interpreting and applying provisions of an international treaty with life-or-death consequences (the Convention), not merely establishing administtative policies or interpreting a domestic statute in light of such policies. Because it is treaty based, the definition of a "refugee" must be applied consistently throughout the world and legal accuracy is of paramount importance. The idea that administrative tribunals make access to justice "less cumbetsome, less expensive, less formal and less delayed," (95) may hold sway in the realm of trade, business, labout and social entitlements; however, the practical reality on the ground is that an asylum seeket (much like a criminal defendant) is primarily in need of a full and fair hearing and a correct determination of the facts and the law--things squately in the wheelhouse of courts. Observation and practical experience suggest that, for refugees, it is the content of justice that is paramount--not so much the delivery system. (96)

Nonetheless, the asylum context undeniably calls for flexibility in adjudication--an area where administrative tribunals excel. It is generally accepted by counsel and courts alike that asylum seekers have great difficulty marshalling evidence to support the veracity of their narratives, and it could be disastrous for formal rules of evidence to apply to them. In the current quasi-judicial system, asylum seekers benefit from looser evidentiary rules that allow them to submit newspaper articles or medical reports without being required to call the author as a live witness, as would be required in a court setting. The loss of the status quo informality could be a major disadvantage of shifting to a wholly judicial paradigm.

In contrast, the argument that administrative tribunals are preferable because they bring specialization and expertise does not necessarily serve as a defense of the status quo in the context of Canada's current asylum process. For an indeterminate reason, the IRB is and was conceived as a largely lay tribunal--with the vast majority of its members taken from the broad spectrum of Canadian society. (97) The IRB has never been the subject of a privative clause, and the government appears to acknowledge that it holds no special expertise as currently constituted: "[e]ndingthe practice of appointing individuals without subject matter expertise to the Immigration and Refugee Board of Canada" (98) was a campaign promise of the winning side in Canada's 2015 federal election. Even the IRB does not regard itself as an expert tribunal. (99) Thus, the argument that the quasi-judicial model pools expertise and knowledge for the benefit of users does not seem persuasive when it comes to the IRB; not enough was done to truly create a body with such assets.

Some may be wary of generalist courts opetating in a specialized administrative context such as asylum, and it is certainly ttue that the federal judiciary in Canada has been criticized for their decision making in refugee cases. (There have been gross disparities in the application of the test for leave, for example). (100) On the other hand, it is often ovetlooked that the so-called generalist judiciary have made some remarkably refugee-friendly decisions--both in Canada and elsewhere. (101) These judicial actions included ensuring gender sensitivity within the asylum system, censuring irrational and arbitrary credibility assessment, restoring publically funded health care coverage, and outlawing mass expedited determination. On the plus side as well, a recent study of migration courts in Sweden found professional judges to be more impartial than lay judges. (102)

In fact, the foregoing discussion around the issue of "administrative tribunals vs. generalist courts" has shown that the potential loss of informality and flexibility is the only clear advantage of the status quo quasi-judicial model. Yet, as noted above, bodies like the Tax Court (which, after all, is the model here) have demonstrated the ability to retain a considerable degree of the flexibility and informality of an administrative tribunal (103)--and if these attributes are needed in asylum adjudication, then this could be achieved through rules and regulations. In any event, , regardless of the theoretical debate on this issue, the fact remains that the current model of quasi-judicial lay decision making has not proven entirely satisfactory in practice; there is little to be lost by exploring alternatives.

Another concern about a paradigm shift of this nature would be to question how well-suited a courtroom atmosphere would be to refugee status determination and how well judges could relate to individual refugee claimants. However, as the ABA pointed out, judges are accustomed to making the types of decisions required by refugee status determination: They regularly deal with marginalized populations. (104) They also are routinely required to assess the credibility of witness testimony--particulatly in the ctiminal context, and more and more frequently in a cross-cultural context. Judges have shown adaptability to culturally appropriate forms of dispute resolution. (105) They undergo sensitivity training. Regarding courtroom atmosphere, procedures can be adapted to context, such as in specialized youth and family courts. (106) Canada's Tax Court itself has a proven ability to relate to individual litigants, as alluded to earlier: it has reduced formality, (10)" is not bound by the rules of evidence, (108) adapts to unrepresented litigants (109) and regularly hears the viva voce testimony of individual taxpayers--a process which necessarily involves assessment of credibility (110) And in actual fact, how different would it really feel for our hypothetical refugee claimant to appear before a "real" judge as opposed to a Board member? Likely, both would feel equally intimidating to her (both IRB members and judges sit on raised daises), but either type of decision maker is capable of reducing formality to the point where she would feel comfortable enough to express herself and have enough leeway to present evidence in whatever form was accessible to her.

Although there are, as we have noted, potential concerns and drawbacks, there are certainly positive facets to a wholly judicial model for refugee determination. As the ABA pointed out, it is reasonable to expect that placing asylum decisions in the hands of judges would result in improved interpretation and application of refugee law, and stricter adherence to constitutional norms and the common law rules around assessment of credibility and evidence. (111) Already in possession of legal acumen and an open mind, "real" judges would seem well-placed to complement these attributes with knowledge around the particular psycho-social components of "refugeehood". Moreover, the judiciary is an institution that is fully independent from political interference and generally respected. (112) Independence in decision making is particularly crucial in the asylum context: it serves to counter-balance the politically and emotionally charged atmosphere in which asylum decision making takes place, and ensures that concerns over political and cultural sovereignty do not distort the decision-makingprocess. Independence was touted as the most important factor in the creation of the IRB in the Plaut model. (113)

At the end of the day, it is perhaps impossible to provide proof positive that any individual judge would make better decisions than any individual lay administrative tribunal member. And perhaps the positive view of judges underlying the Asylum Court proposal is informed by assumptions and professional biases that are unfounded. But the fact temains that Canadian society operates under similar assumptions nonetheless: we allow neither rape cases nor even de novo tax appeals to be heard by lay administrative tribunal members. It is at least reasonable to question why we allow it with respect to asylum claims.

There is also some support for the notion that adopting a wholly judicial model for refugee status determination could also represent a more efficient use of resources. Currently, Canada has a system of successive filtering mechanisms (RPD, RAD, judicial review, PRRA, humanitarian application, as noted above). A single wholly judicial body of first instance could potentially obviate the need for multi-tiered administrative tribunals (such the RPD/RAD divisions of the IRB, the FTTIAC/UTIAC) in the UK), or an agency/administrative tribunal tandem as in the US, Australia, New Zealand, and many European countries. (114) Indeed, one has to wonder why most systems bring out the high-powered decision-making talent only on appeal after the damage has (presumably) already been done, instead of using it at the first tier to make the initial determination. Furthermore, why spend time and money to train administrative tribunal members to perform difficult adjudicative tasks more "judicially" when we could just assign the task to judges (often the products of publicly subsidized legal education) and be done with it? A significant pool of talent already exists: in Canada, for instance, Asylum Court judges could be appointed from among the refugee bar, the Department of Justice, legal academia, or even other courts. (115) Again, the Tax Court model is instructive in this regard.

Could an Asylum Court deliver the timely decisions refugee claimants desire? It could, if it had a sufficient number of judges (116) and was able to set aside the same amount of hearing time as the RPD currently does (roughly three hours per case). In fact, there is some indication that judges can actually be more productive than civil servants in terms of the number of cases handled. For instance, statistics show that in 2013-14 Canada's Tax Court judges averaged 207 dispositions each, whereas the RPD members finalized about 60 claims each over the same time period. (117) Although tax appeals and refugee hearings might not be identical in nature, this disparity in productivity does provide food for thought.

But what about the cost ? There is no doubt that creation of an Asylum Court would be an expensive proposition in the short term. For instance, in Canada, Tax Court judges earn about $300,000 per year, as opposed to approximately $ 100,000 per year for IRB members. (118) But in the long term, the cost differential with the current system might not be that great, if multiple tiers and recourses were rendered unnecessary and the length of stay of unsuccessful claimants was reduced. (119) As the ABA pointed out in its proposal, expenditures in one area could be offset by savings in others. (120)

In summary, a wholly judicial model for first-instance refugee determination may not be something we are entirely comfortable with, but to the extent that it has the potential to improve the quality and efficiency of decision-making, it is at least wotthy of further research.


A far less radical alternative to a wholly judicial model would be a "parajudicial" one. (121) This change would see the IRB teplaced by a new administrative tribunal (we could call it the Asylum Tribunal of Canada) whose members would be required to be lawyers with appropriate expertise. Many other asylum systems make use of parajudicial models--although usually for appeal and not for initial determination. (122) In addition, the parajudicial model, since it retains the flexibility of an administrative tribunal, could also have space for an intetdisciplinary element.

An appropriate real-life model in the Canadian context is the Tribunal Administratif du Quebec (TAQ). It is a large tribunal which plays a decisive role in administrative justice in the province of Quebec. (123) It is parajudicial in the sense that every panel must always include at least one member with professional legal training, (124) a fact which is viewed as crucial by the Quebec bar. (125) Both the TAQ's de novo and de novo appellate jurisdiction touch on core liberty and security of the person issues--everything from mental disorder, to health care entitlement, to educational rights, to denial of social assistance benefits, to municipal assessments.

The TAQ is required by law to be populated by lawyers, psychiatrists, social workers, psychologists, and chartered appraisers. (126) The panel hearing a particular case is required by law to have an appropriate composition and, as indicated earlier, must always include a lawyer or a notary. (127) For instance, cases relating to confinement of mentally-ill persons must be heard by a three-member panel composed of a lawyer or notary, a psychiatrist, and a social worker or psychologist. (128) Cases involving municipal taxation must be heard by a lawyer, notary, or chartered appraiser. (129) Thus, transferring such a model to the Canadian asylum system, first-instance decision-making responsibility could be shared between lawyers and other appropriate professionals, such as psychologists or anthropologists. (130)

Like a wholly judicial model, a parajudicial model would have certain benefits in terms of improving the quality of decision making, and several previous reform proposals have advocated for a version of such a model. (131) Retaining first-instance decision making in an administrative tribunal, but staffing it exclusively with lawyers would preserve the flexible and informal nature of the current system while at the same time adding the benefits of higher-knowledge decision making. Although those admitted to the bar are not screened specifically for judgment, sensitivity, and temperament in the same way that professional judges are, the legal mind is in some sense well-suited to asylum decision making. This is the case because--as illustrated by our hypothetical claimant--statutory interpretation, weighing and assessing of competing factots, and application of the rales of credibility assessment all come into play. While these tasks could also be performed by nonlawyets, years of legal training would seem to give lawyers an advantage over lay persons in this regard.

The parajudicial model may have one distinct advantage over the wholly judicial model--namely, the aforementioned possibility of interdisciplinarity, as advocated by Rousseau, et al. (132) Not being subject to the constraints of the judicial appointment process, a parajudicial model could still bring legal expertise to refugee status determination, but augment it with other appropriate forms of expertise as well (as is the case with the TAQ). Since refugee status determination (particulatly with respect to credibility assessment) has cultural and psychological dimensions, expertise in these areas might well enhance decision making as much as involving the judiciary would. The personal qualifications necessary for Asylum Tribunal members should ideally be enshrined in legislation--as is the case with the TAQ and New Zealand's Immigration and Protection Tribunal (IPT), which are both de novo appeal tribunals. (133)

Nonetheless, retaining an administrative tribunal structure does not help avoid the pitfalls around questions of independence. As we have seen, the independence of the quasi-judicial IRB has been called into question in the past, largely because of the lack of security of tenute of its members. (134) Although these issues seem to have been resolved for the moment, (135) vigilance is still required. Independence is crucial to level-headed application of refugee law--i.e., the ability to tune out the political or social pressures that abound in the asylum context. "Real" judges are notionally independent since they enjoy security of tenure, but that would not necessarily be the case for decision makers on a parajudicial board. The key would be to impose judge-like rules about the appointment and reappointment process for tribunal members. In order to ensure compliance, these rules should also, ideally, be legislated and not left to the government to implement through hiring policies.

In terms of the timeliness of decision making, an Asylum Tribunal would likely function at no worse a level than the current RPD, if all members were lawyers and all cases were heard by one-member panels. If interdisciplinarity were added to the decision-making landscape, however, panels would necessarily be composed of more than one member--i.e., a lawyer and at least one other professional, as is the case at the TAQ. Cases might then take longer to come for hearing--unless the size of the current RPD were doubled or tripled. Intetdisciplinarity could therefore entail either increased delays or increased costs, and thus more research would be needed to determine viability.

In terms of efficiency, it could be argued that creating a parajudicial board would, if subject to enforceable rules as to the quality of decision makers, be making good use of resources already expended on the training of lawyers and other professionals. As well, there would be adherence to the principle of using the best and brightest to make the most complex decisions, although perhaps not to the same extent as a wholly judicial model. As well, a parajudicial model would likely be much less expensive to set up in the short term than an asylum court, while still theoretically retaining the same long-term cost advantages associated with better decision making. In order to create the Asylum Tribunal, the RPD could essentially be reconfigured and repopulated, pursuant to a few relatively simple legislative changes. (136)

In short, the model could certainly be viable, but only with a commitment to true "parajudiciality"; otherwise, there would probably not be enough of a discernable difference from the current quasi-judicial model to justify a shift. Unfortunately, in the quest to establish a new site for initial detetminations, there will always be some unknowns. It is certainly true that there have been many excellent "lay" IRB members and plenty of members with legal and subject-matter expertise who have been disappointments. (137) But refugee claimants deserve more than hit-or-miss policies; they deserve a coherent working theory that represents society's best efforts to ensure accurate and efficient decision making over the long run. In our view, that working theory should be the same as adopted by the Tax Court and most other Canadian tribunals: tax cases should be decided by tax lawyers, human rights cases should be decided by human rights lawyers, and refugee cases should be decided by refugee lawyers. Our hypothetical refugee claimant would likely expect and hope that anyone deciding on her asylum claim would be exceptionally well versed in asylum law. The working model for the current IRB (i.e., that of a lay administrative tribunal) has not been as successful as intended and it may be time to rethink it.


In order to address concerns around complexity and delay, it might be advantageous to consider implementing an "all-in-one" approach to refugee status determination. Under this approach, the first-instance decisionmaking body (be it either a specialized court or administrative tribunal) would deal not only with the question of Convention refugee status, but also with as many other forms of complementary protection as possible. This is currently the approach in Canada, but perhaps not to a sufficient extent. It is true that the RPD currently deals with both Convention refugee status under section 96 of the IRPA as well as "protected person status" under section 97 (which incotporates protections that stem from other international conventions). (138) However, humanitarian and compassionate relief requires a separate and costly application process and provides no automatic protection against removal. Accordingly, it could be beneficial to add to the initial decision makers jurisdiction a new humanitarian protection category designed to cover difficulty with return to country of origin that falls short of the other protection grounds, such as concerns around discrimination, armed conflict, oppression, or inadequate medical treatment (as exemplified with our hypothetical claimant and her baby). It might then no longer be necessary to make the humanitarian and compassionate application process available to most asylum claimants post-determination, thereby avoiding a multiplicity of proceedings. In addition, the initial decision makers should have some limired residual reopening jurisdiction to determine new risk developments and/or subsequent applications. Currently, these sorts of concerns are dealt with in the entirely separate PRRA process (available to those still in the country one year post-determination), increasing the systems complexity and cost. (139) It is reasonable to believe that having the same decision maker render a single decision on all possible grounds for protection on the same evidence would be an improvement on multiple decision makers rendering multiple decisions in succession. The chances of accurate and consistent decisions may well be higher where there is a thorough presentation of the facts, as they may relate to all possible grounds of protection before the same, high-quality decision maker. Of course, an all-in-one model would deny claimants additional "kicks at the can", so to speak. However, if the initial asylum determination becomes fairer and more accurate (and there is an effective correction mechanism for those errors that do occur, as we will discuss shortly), then the absence of additional and separate routes of redress might become considerably less significant. As well, consolidating as many protection grounds as possible in one proceeding might discourage false claims. If claimants arrived knowing that there were several grounds for asylum, they might be less inclined to try to fabricate stories designed to fit into a particular mould. This could be particularly true where one of the protection grounds was the proposed new category of "humanitarian protection", which might assume a sort of safety-net function.

As for the question of improvements in efficiency, it would seem simpler, quicker, and more cost effective to resolve all issues with respect to an asylum claim at one time and in one place--rather than allowing for multiple successive proceedings as is now the case. This is a sort of "one question, one answer" model: our hypothetical claimant wants to know if she can stay, and wants a yes of no answer as soon as possible. By extension, consolidation of protection grounds in one proceeding might do the same thing for an error correction mechanism: if all protection grounds are canvassed at the initial hearing, all grounds will be verified by the courts in a sort of "super-review". The cost savings could arguably be substantial.

There may be, nonetheless, some serious questions about whether expanded protection grounds would represent an improvement on the current system. As was raised in one previous reform proposal, it is possible that having to address not two, but three, grounds of protection in one hearing would lengthen the hearings considerably. (140) This outcome is not inevitable, however. The amount of extra time spent exercising humanitarian jurisdiction would simply depend on the scope of this recourse--i.e., whether humanitarian protection could be granted on essentially the same set of facts as the other protection grounds, and whethet the experiences of some European countries in this regard could be instructive.1 (41) In our hypothetical example, the claimant could testify about her rape, problems related to her ethnicity, and then explain why her baby would not receive medical treatment in her home country. The decision maker could look at all the facts and decide which category of protection was more appropriate.

Another problem, alluded to earlier, is the fact that eliminating the separate post-determination PRRA and/or humanitarian processes leaves no satisfactory way of dealing with new facts or circumstances that may arise after the initial refusal of protected status, but before removal takes place. That is why the possibility of some sort of reopening mechanism based on a residual jurisdiction may be advisable--perhaps similar to what exists in the United States. (142) One advantage that reopening would have over a separate PRRA procedure is that the decision maker would have already met the claimant and would not have to study the case from scratch. Needless to say, however, this reopening mechanism must be designed in such a way as to avoid the very real risk of creating a new morass of legal procedures. Perhaps reopening could be limited to changes of conditions in the country of origin or determinative and ctedible newly discovered evidence. Processing claims rapidly from intake to removal could also reduce the possibility of reopening becoming necessary--as would ensuring fair and accurate determinations in the first place. (143)


In otder to improve the ability of Canada's refugee status detetmination system to correct errors in the initial, considetation should be given to allowing direct access to the courts without a leave requirement: instead of judicial review, an appeal on the merits that would lie directly to the Federal Court of Appeal as of right. Such an approach would obviate the need for an internal appeal step in the process. The Federal Court of Appeal, as opposed to the trialdevel Federal Court, is suggested as the more appropriate forum, in part because of the elevated status of the initial decision maket undet our proposal. Indeed, this is the current appeal trajectory for Tax Court decisions. It was also the trajectory (albeit with leave) in respect of the former IAB until 1978 and in respect of the IRB between 1989 and 1993. (144) One ttade-off for the elimination of the leave requirement, however, would be a written appeal instead of oral argument. This trade-off is necessary because of the large numbers of refugee appeals filed each year. (145) This proposal would constitute, essentially, adoption of the US model : instead of the customary oral argument, appeals would, as a rule, (146) be decided solely on the basis of written materials and by a panel of three appellate-level judges. Under such a scheme, fairness would require that the appeal have a suspensive effect on removal and that the Court of Appeal issue written reasons for decision, particularly where the appeal is denied. And, as noted above, if all grounds for protection are consolidated at first instance, so would they be on appeal. Further, as is currently the case for appeals under section 52(c) of the Federal Courts Act, (147) the Court would have the power to render the decision that the Asylum Court or Asylum Tribunal should have rendered in the first place--i.e., to grant or deny asylum.

Would direct access to the federal appellate judiciary really be an improvement on the current internal administrative appeal model (i.e., the RAD)? The RAD has similar remedial powers (it can declare the claimant to be a Convention refugee), but has one very significant advantage: it has the power to hear a claim de novo and also to accept new evidence, (148) which is something a court of appeal cannot do. Yet, unfortunately, there is a disconnect between theory and practice with the RAD. As we have seen, the RAD is staffed for the most part with members who are not required to be lawyers, who are not required to have subject matter expertise, and who really occupy no "higher" juridical position than RPD members. It is a situation that risks the blind leading the blind, so to speak. In contrast, it is relatively safe to assume that appellate courts generally have an advantage with respect to legal reasoning skills over lay decision makers (although there could always be exceptions). Thus, as the RAD is currently constituted, it is by no means clear that the RAD's broader de novo powers would necessarily outweigh the Federal Court of Appeal's superior legal capacities. If the qualifications of the RAD members were changed, the picture would be significantly different; but until that happens, it is worthwhile continuing to look at the federal appellate option. We should bear in mind as well that the proposed appellate option is predicated on the existence of a higher-quality initial assessment of the facts and evidence--which would make de novo powers on appeal less crucial.

But is the proposed error correction mechanism feasible? In order to answet that question, one consideration should be whether it would be logistically possible for the Federal Court of Appeal to handle several thousand written refugee appeals annually. Again, there is considerable guesswork involved, but there is evidence that it did so fairly successfully between 1989 and 1993. (149) In addition, the Federal Court does so now, as do the Circuit Courts of Appeals in the United States. (150) Nonetheless, it is very likely that the number of appellate judges will need to be greatly increased--or possibly a special division of the Federal Court of Appeal created--in order for the proposed appeal mechanism to function smoothly and without generating significant backlogs Would it be feasible for the Federal Court of Appeal to review several thousand refugee appeals sitting in panels of three and to also provide written reasons? Again, the Circuit Courts in the United States do--and one might envision a system whereby a lead judge is assigned to each appeal who, after receiving input from the test of the panel, is then charged with producing written reasons that are adequate, but brief. (151)

In essence, what is being proposed here is a shift in the use of judicial resources: from triage of several thousand cases per year (leave applications) to merits reviews of the same cases (written appeals). While these might appear to be two quite different tasks, practical experience shows they are actually more similar than meets the eye, thus enhancing the proposal's feasibility. A leave application is a complex document consisting of the IRB's reasons, an affidavit and/or a transcript, the evidentiary and documentary record before the IRB, and memoranda of fact and law. (152) A paper appeal would involve essentially the same type of material (tribunal record and memoranda). In terms of the decisional process involved, instead of making a decision to grant or deny leave to proceed to an oral hearing on the merits as the Federal Court does, the Federal Court of Appeal would use these written materials to make a final determination instead. Furthermore, although the Federal Court judges are currently supposed to be performing a sort of triage to weed out frivolous or marginal cases that do not warrant an oral hearing, recent research suggests that they are actually already doing a merits review on paper at the leave stage in any event. (153) Accordingly, there is some reason to be optimistic about the feasibility of having the Federal Court of Appeal perform error correction in the manner outlined above. Nonetheless, as noted above, the number of appellate judges will very likely need to be substantially increased--or possibly a specialized division of the Federal Court of Appeal would need to be created. (154)

Would consecrating error correction to the Federal Court of Appeal yield better outcomes for claimants than the current system? We must recall that the complaints around error correction under the current system center on a lack of effectiveness, which implies that incorrect asylum determinations are not always being rectified. Certainly, elimination of the leave requirement would remove a major bone of contention in this regard. Perhaps more significantly, a remedy at the Federal Court of Appeal could be quite effective because it would not be merely judicial review, but a true appeal governed by section 52(c) of the Federal Courts Act. (155) That is, the Federal Court of Appeal would be making a determination of the correctness of an asylum decision below and could therefore reverse it. As noted earlier, this is the same error correction trajectory that exists currently in respect of the Tax Court. (156) Although the RAD has the same attributes, as it stands now, RAD members do not have the legal acumen of appellate judges--and they do make errors. (157)

Of course, there is another trade-off for the elimination of the leave requirement: the elimination of the oral judicial review hearing that rejected claimants now enjoy at the Federal Court level if they have been granted leave, and which every refused refugee claimant enjoyed as of right at the Federal Court of Appeal level prior to 1989. (158) As indicated earlier, there are simply too many (159) negative asylum decisions rendered annually for an in-person appeal as of right to be realistically attainable. Thus, the suggestion is that, instead of being able to triage, the courts would entertain all appeals but dispose of them (normally) entirely in writing As mentioned earlier, this would be substantially the same form of redress which exists ostensibly without objection in the US as between the BIA and the Circuit Courts of Appeals. (160) The trade-offs could be worthwhile to the extent that the proposed error correction mechanism would be more effective overall, and there is an argument to be made that it would be. In contrast to the current system, what is proposed would be 1) an appeal on the merits (not just the more restrained judicial review); 2) an appeal as of right (without the "black hole" of the leave requirement); and, 3) an appeal conducted by appellate-level judges (instead of RAD members or trial-level judges). The improvement over the system currently in place lies in the fact that it provides all claimants with an unqualified right to a merits review of their refusals by three appellate judges. Again, this system seems to correct errors well in the US, although one must be cautious when drawing comparisons between jurisdictions.

The involvement of the Federal Court of Appeal in the refugee process could also improve error correction by encouraging more consistent development of the law. Under the current system, the Federal Court of Appeal rarely pronounces itself on issues of refugee law, because the vast majority of Federal Court decisions in refugee cases are barred from appeal. As a result, the judges of the Federal Court, despite having to deal with thousands of cases every year, are provided with relatively little legal guidance, resulting in a plethora of contradictoty decisions, as well as continual relitigation of the same issues. It is probably not a coincidence that the state of refugee law seemed much clearer--and actually evolved--during the 13-year period when the Federal Court of Appeal was in charge of error correction in the asylum process. (161) Furthermore, an appellate court sitting in a panel of three tends to generate more stable precedent than trial judges sitting alone, and when the law is clearer, refugees should become easier to identify.

Overall efficiency may be improved as well. For one thing, having error correction done at the appellate level would simplify things by reducing one or two layer s of process: currently the trajectory of a claim is RPD to RAD to the Federal Court to (in some cases) the Federal Court of Appeal. Under the proposed mechanism, the RAD would be bypassed and the Federal Court replaced. (162) The cost of the current 25 RAD members' $ 100,000 per year salaries would be saved and the size of the Federal Court could possibly be reduced. Giving more cases to the Federal Court of Appeal might ease the burden on the Federal Court. Despite having 15 judges, the Federal Court of Appeal disposed of only 541 cases in 2013-14. (163) In contrast, the 44 Federal Court (trial level) judges disposed of 11,460 cases over the same time period--the majority of them concerning asylum seekers. (164) It would seem more rational to spread out the wotkload more evenly between the two levels of courts--although it must be acknowledged that there are important distinctions in the types of decision making each level engages in.

Another efficiency advantage of using the Federal Court of Appeal for error correction would be its aforementioned power, as an appellate-level court, to render the decision that should have been rendered by the first-level tribunal in the first place (although admittedly this power also exists in the RAD). (165) It is interestingto note that, between 1980 and 1993 (before judicial supervision of refugee status determination was transferred away from the Federal Court of Appeal to the Federal Court) the Federal Court of Appeal fairly regularly either declared applicants to be Convention refugees or mandated this result on rehearing, even in credibility cases. (166) There does not seem to be any readily-apparent reason why this practice could not be revived--thus, eliminating the costs involved in rehearing the same claim twice. So, for instance, if the initial decision maker encountered by our hypothetical refugee claimant erred in the interpretation of whether rape during a communal riot constituted "persecution", the appeal court could reverse that determination and grant her asylum.

Nonetheless, we must take care to avoid a long and drawn out error-correction process that leaves claimants in limbo for years. The question of whether the proposed error-correction mechanism would be a timely one is difficult to answer. It might be instructive to look at past and current judicial performance in similar contexts. Generally, leave applications at the Federal Court are disposed of in writing within four months before a single judge. (167) For those cases which do receive leave and proceed to an oral judicial review hearing, the time elapsed between the filing of the originating document and the issuance of a reasoned decision on the merits of the judicial review is about eight to nine months. (168) For Tax Court appeals to the Federal Court of Appeal, the total time elapsed appears to be around 9-10 months. (169) Looking at the United States Circuit Courts of Appeals (by which the proposed error-correction mechanism was inspired), however, we see significantly longer processing times. For example, the median interval from filing to disposition in the Circuit courts for administrative cases (including immigration) from September 2014 to September 2015 was 15.3 months for all the Circuits. (170) For the Ninth Circuit (which handles the greatest number of immigration cases), the interval was 28.9 months. (171) Nonetheless, it should be noted that the Circuit Courts of Appeals have been dealing with significant backlogs caused by a huge spike in immigration appeals after 2002, (172) and this might account for some of the delays.

What we might conclude on timeliness is that, although there may be a reason to be hopeful that our proposed error-correction mechanism could adhere to reasonable timelines (4-10 months), there is a realistic possibility that the new mechanism could become a significant locus of delay within the system, thereby compromising efficiency. The risk of a major slowdown at the error-correction stage might be mitigated if better initial determinations result in higher asylum grant rates and therefore fewer appeal files, and/or if a reduction in judicial workload follows a reduction in court proceedings filed in relation to applications for complementary forms of protection. Also, increasing the number of judges from the current levels, (173) as alluded to earlier, could go along way towards heading off delays as well. But, in the final analysis, even if there would be significantly longer processing times for error correction, it is at least arguable that this would be offset by time saved in the asylum system as a whole through the elimination of separate proceedings related to complementary protection. In other words, it would be the net result in terms of time savings that would be significant. As long as claims could be finalized faster than the current multiple recourse system at its worst (three to six years), a case could be made that the proposed reform to error correction would improve the flow of the system.

But what about concerns over the Federal Court of Appeals attitude towards refugees? As alluded to earlier, in the past the Federal Court of Appeal has made quite a few refugee-friendly decisions and has displayed significant sensitivity by showing awareness of the particular difficulties faced by refugees within a Western legal system and by denouncing conjugal violence, sexism, and the culture of disbelief at the IRB and former IAB. (174) Since 1993, however, the Federal Court of Appeal has had limited contact with refugees due to legislative restrictions on the right of appeal. What few refugee law decisions it has made in recent years have not necessarily drawn rave reviews from counsel, but there is nothing inherently suspect about appellate-level judges doing error correction in the asylum system. In fact, the asylum decisions of the Circuit Courts of Appeals in the US are almost always well informed and well-reasoned--and often exhibit an awareness that asylum claims operate in a unique context. More contact with refugee cases might spark a positive change in decision making. If not, as indicated earlier, a specialized section of the Federal Court of Appeal might need to be cteated just for asylum cases.


A lengthy tefugee status determination process benefits no one. Ideally, then, there needs to be maximum processing times established for both initial determination and error correction, and these times should be enshrined in legislation to help ensure enforceability. (175) But how long is too long? And how short is too shott?

A refugee claim is a complex, high-stakes procedure, and accuracy and soundness of judgment might be compromised by attempting to conduct proceedings too quickly. Reasonable processing times might be six to nine months for the initial determination and six to nine months for error correction--twelve to eighteen months in total. These timelines are in keeping with other reform proposals. (176)

A six to nine month maximum processing time would seem feasible for initial determinations based on past experience and on the IRB's own estimates. (177) The timelines imposed on the RPD in 2012 were too short, (178) and statistics for 2017 show that the RPD is drifting towards six-month processing times in any event. (179) Our suggested six-month interval would also be fair to claimants, as it has been recognized in other reform proposals as being enough time to allow for the collection of relevant evidence and for the claimant to prepare to testify. (180)

Although they appear efficient and economical, compressed timelines are actually neither: they are likely to engender multiple requests for extensions of time and result in incomplete evidentiary records and ensuing appeals and reviews. (181) Compressed timelines are also not responsive to the realities of refugee status detetmination--to the plight of a trauma victim from a different country required to produce a narrative that is understandable in an alien cultural and discursive context. As noted earlier, one study of refugee claimants concluded that claimants feel considerable stress in connection with expedited time frames, particulatly with regard to gathering sufficient documentation. (182) On the other hand, long delays can be devastating in other ways, as alluded to earlier. One recent study has explored the psycho-social effects of delay within the asylum system and notes that "[t]he state of existential limbo mediated asylum seekers' everyday lives, manifesting as a form of social defeat, a disruption of emplotment and emplacement, and as a kind of trauma or violence." (183) Thus, reasonable timelines for initial determination are timelines that are long enough to ensure that all information is gathered and allow for the effects of culture shock, trauma, and displacement to soften, but not so long that memory will be distorted and the asylum seeker will suffer the ill effects of a seemingly endless wait.

Reasonable timelines for error correction are equally important. The RAD currently has a 90-day timeline for rendering decisions on appeals, which it is not currently meeting by reason of a large influx of claims. (184) Under a model that uses the Federal Court of Appeal for error correction, it is certainly possible that the Federal Coutt of Appeal could adhere to a maximum processing time (counted from the filing of the originating appeal document) as long as it was reasonable vis-a-vis historical claim levels. The Federal Court is currently subject to legislated scheduling times for judicial review hearings on the merits (60 days), with which it has complied without obvious difficulty. (185) Based on our earlier analysis, it seems feasible that error correction could be done within six to nine months, and that could be considered as a maximum processing time frame.

The cost efficiencies around promptly resolving asylum claims are fairly obvious: there will be less money that will need to be expended on social benefits for those asylum seekers who are unable to work or are in poor health. These types of expenditures can arouse social tensions at times when there are large numbers of claimants. In addition, excessive time between arrival and removal is generally thought to encourage fraudulent claims and act as a "pull-factor". (186)

Whatever timelines are ultimately selected, and in respect of whatever forum (initial determination or appeal), it would seem critically important that these timelines be enshrined in legislation in order to provide an incentive for the government to keep the relevant bodies properly staffed and accountable for their performance. Excessive processing times can cripple even the best-designed asylum system.


The new model for Canada's asylum system proposed here seeks to judicialize initial determination, enhance the availability and quality of error correction, and simplify and streamline the process. It represents a paradigm shift--from a multiplex system to an essentially binary one. Instead of resources being expended on several recourses in succession (RPD hearing, RAD appeal, judicial review application, PRRA, humanitarian application, etc.), we propose focusing our efforts on only two: a comprehensive initial hearing before a specialized court (or court-equivalent), followed by an appeal on the merits to the appellate judiciary. It is an overall approach loosely described by the aphorism "put all your eggs in one basket and watch the basket". (187) Getting it right the first time is key.

Although this paradigm shift is proposed sincerely and with conviction, it is not intended to be definitive. As noted earlier, it is simply intended to serve as a springboard for further discussion around sustainable, long-lasting, and generally acceptable reform. The specific reform proposals presented here could be adopted in whole or in part. Where more than one alternative has been presented, thete is no intention that these alternatives be seen as opposing each other, but simply as different manifestations of the kinds of institutional changes that could help avoid the pitfalls prevalent in many asylum systems, including in Canada. Some aspects of this proposed reform project are already being tried--either in Canada or elsewhere, either in the asylum context or in some other context.

The exercise of thinking about a new refugee-status-determination model is of value in and of itself: it not only exposes the flaws in the cutrent system, but also deepens our understanding of what is at play. For instance, floating the idea of a wholly judicial model for initial-determination making brings up the issue of the lack of a jurist culture at the IRB, and the value of expert and judicious decision making to stakeholders. Exploring ways of simplifying the process prompted us to assess the value and purpose of complementary protection mechanisms: when we considered rolling them into the initial hearing to avoid a confusing multiplicity of proceedings, we better understood that they function not merely as a safety net, but too often as a lifeline, and this reinforced the importance of improving initial decision making. Consideting a new error-correction mechanism brought home the profound unfairness of the leave requirement and the inadequacy of judicial review as a remedy for denials of asylum. And reviewing the issue of appropriate timelines sensitized us to the societal impact and human cost of systemic delay in the asylum context. Contemplating change is difficult, but asylum seekers deserve honest efforts to grapple with these longstanding shortcomings in our current model.

Continuing to merely tinker with the asylum system without addressing its underlying flaws is an exercise in futility. Policy makers must examine what is causing the sustained dissatisfaction by undertaking a comprehensive analysis. Who are the stakeholders in Canada's asylum system and what are their needs ? How can these needs best be met? What design would make the system most easily adaptable to changing global conditions ? Exploring such questions is the first step on the path towards a stable and sustainable model for refugee status determination in Canada. Continual upheaval without real change does a disservice to refugees.

PIA ZAMBELLI ([dagger])

([dagger]) The author is a practising immigration lawyer and former member of the Immigration and Refugee Board of Canada. Thanks are due to the peer reviewers, as well as Professors Evan Fox-Decent, Payam Akhavan, and the late Roderick Macdonald of the McGill Faculty of Law for their help and inspiration. This article is a reworked and updated version of part of the author's LL.M. thesis. See Pia Zambelli, Refugee Status Determination in Canada and the Path to Radical Reform (LLM Thesis, McGill University Faculty of Law, 2012) [unpublished]. Any views expressed in this article are those of the author and should not be attributed to any organization with which she is or has been associated.

(1) The system established by the Immigration Act I976, SCI 976, c 52, s 1 was modified in 1989 and again in 1993. In 2002, the Immigration and Refugee Protection Act, SC 2001, c 27 [IRPA] made significant changes to the system. The IRPA was amended again in 2010 and 2012 by the Balanced Refugee Reform Act, SC 2010, c 8 [BRRA] and the Protecting Canada's Immigration System Act, SC 2012, c 17 [PCISA] (Bill C-31), respectively.

(2) For instance in the 2009-2010 fiscal year, the average processing time for a refugee claim ar the inirial srage was 19.2 months, with a backlog of some 59,000 claims. See Immigration and Refugee Board, Performance Report for the Period Ending March 31, 2010, online: <> at 16.

(3) See BRRA, supra note l,andPCISA,supranote Lpresenrly enacted as ss 169.1(2), 110, 109.1,111.1(2), 55(3.1), 112(2)(b.l),and25(l)1.2)of the IRPA,supra note l;see also Order Respecting the Interim FederalHealth Program, 2012, SI/2012-26, (2012) C Gaz II, 1135.

(4) In Canadian Doctors for Refugee Care v Canada (Attorney General), 2014 FC 651, 28 Imm LR (4 (!h)) 1 [Canadian Doctors for Refugee Care], the Federal Court ruled rhar curs to a publicly funded health insurance scheme for refugees constituted were unconstitutional. In Z(Y) v Canada (Minister of Citizenship and Immigration), 2015 FC 892, 387 DLR (4th) 676 [Z(Y)\ the Federal Courr ruled that inferior refugee procedures based on country of origin violated principles of equality.

(5) See Idil Atak, Graham Hudson & Delphine Nakache, "'Making Canada's Refugee System Faster and Fairer': Reviewing the Stated Goals and Unintended Consequences of the 2012 Reform" (2017) Canadian Association for Refugee and Forced Migration Studies Working Paper No 2017/3.

(6) Government of Canada, "Immigration and Refugee Board Review--Terms of Reference", online: <> [IRB, Review 2017].

(7) 28 July 1951, 189 UNTS 137 (entered into force 22 April 1954).

(8) Ibid, art 1A(2).

(9) See IRPA, supra note l,s 169.1(2). Previously, RDP members were Governor in Council appointees and were similar to conrract employees. Being subject to the Public Service Employment Act means that RDP members benefit from grievance procedures and more stringent labour standards norms.

(10) An "error correction mechanism" is known in policy circles as any legal mechanism that allows for the correction of a granr or denial of asylum on the merirs. It is not necessarily the same rhing as judicial review.

(11) See IRPA, supra note 1, ss 153(l)(a), 153(4).

(12) See ibid, s 74(d).

(13) The PRRA occurs after appeals and judicial reviews have been exhausted. In most cases, the decision maker is free to apply the Convention refugee definition. See ibid, s 112.

(14) PRRAs may not be done until one year after the claim is rejected, and three years in the case of claimants from so-called "safe" countries of origin. See ibid, s 112(2)(b.l). Humanitarian relief is also not available for one year after the claim is rejected, subject to certain exceptions. See ibid, s 25(1.2).

(15) For instance, in the United States, initial examination is performed by Asylum Officers (who are employees of the United States Citizenship and Immigration Service) or Immigration Judges (who work for the Justice Department). In terms of error correction, the US has the Board of Immigration Appeals (BIA) and then the circuit Courts of Appeals for the Federal Circuit on judicial review. See Regina Germain, AILA's Asylum Primer: A Practical Guide to US Asylum Law and Pratice, 5th ed, (American Immigration Lawyers Association, 2007), 227,242; American Immigration Council, Asylum in the United States, August 2016, online: < www.americanimmigration>. In the UK, Europe, Australia, and New Zealand, initial examination is performed by agencies. For the UK and Europe, see Asylum Information Database, Country Reports, online: <>; for Australia, see Ausrralia, Adminisrrative Appeals Tribunal, Migrarion and Refugee Division, Guide to Refugee Law, online: <>. For New Zealand, see New Zealand Immigration, Informationfir Asylum Seekers, online : <>. In the UK, error correction is consecrated first to the First-Tier Tribunal, Asylum and Immigration Chamber (FTTIAC) and then to the Upper Tribunal, Asylum and Immigration Chamber (UTIAC), and then to the courts (see Asylum Information Database, Country Reports UK, online: <>. In New Zealand, error correcrion is done by the legally trained members of the Immigration and Protection Tribunal (IPT) and then the courrs (see New Zealand, Immigration Act2009, ss 217, 219,245); and in Ausrralia error correction is assigned to the Migration and Refugee Division (MFD) of the Administrative Appeals Tribunal (formerly the Refugee Review Tribunal (RRT)) and then to the courts (see Australia, Administrative Appeals Tribunal Act 1975, s 44).

(16) Xue v Board of Immigration Appeals, 439 F3d 111 ar 1 (2d Cir 2006).

(17) See e.g. Alan D DeSantis, "Caught Between Two Worlds: Balduins Dialogism in the Exile Experience" (2001) 14:lJRefugee Studies 1 (refugees speak and rhink in dialogic terms); Jane Herlihy & Stuart W Turner, "The Psychology of Seeking Protection" (2009) 21:2 Intl J Refugee L 171 (review of psychological facrors that affecr refugee testimony); Walrer Kalin, "Troubled Communication: Cross-Culrural Misundersrandings in the Asylum-Hearing" (1986) 20:2 Inrl Migration Rev 230 at 232 (classic study of the effect of culture shock on refugee testimony); John Chr Knudsen, "When Trusr is on Trial: Negotiating Refugee Narratives" in E Valentine Daniel & John Chr Kndusen, eds, Mistrusting Refugees (Berkeley: Uni versity of California Press, 1995) 13 (refugees may lack the ability to trust authorities) ; Cecile Rousseau & Patricia Foxen, "Look Me in the Eye: Empathy and the Transmission of Trauma in the Refugee Determination Process" (2010) 47:1 Transcultural Psychiatry 70 (effect of trauma on testimony); Cecile Rousseau et al, "The Complexity of Determining Refugeehood: A Multidisciplinary Analysis of the Decision-Making Process of the Canadian Immigration and Refugee Board" (2002) 15:1J Refugee Studies 43 (the effect of trauma on testimony) [Rousseau et al, "The Complexity of Determining Refugeehood"]; Jeanette L Schroeder, "The Vulnerability of Asylum Adjudications to Subconscious Cultural Biases: Demanding American Narrative Norms", (2017) 97:1 BULR 315 (cultural norms of narrative can affect testimony); Vamik D Volkan, Immigrants and Refugees: Trauma, Perennial Mourning, Prejudice and Border Psychology (London: Karnac, 2017) (cultural bereavement impacts testimony).

(18) Francois Crepeau & France Houle, "Competance et independance: Clefs de la credibilite de lAgence de protection", Memoire depose aupres de la Ministre de la Citoyennete et de l'Immigration lors des consultations sur le Rapport Trempe-Davis-Kunin, 6 March 1998, at 4. See also Francois Crepeau & France Houle, "The Security of Refugees and the Abilities of IRB Members: The Inappropriate Silence of Bill C-l 1" (2001) [unpublished, archived at The Montreal Centre for International Studies].

(19) For the purposes of this article, the source of the public criticism referred to is public comments made by academics, refugee advocates (lawyers and NGOs), government officials, or sometimes refugees themselves. These comments have been made in the context of written reports, scholarly articles or books, Parliamentary testimony, and media coverage. The source of private criticism consists of comments made largely by refugee lawyers and advocates in private conversations or on restricted email lists.

(20) There has also been the suggestion that there are high numbers of fraudulent asylum claims, but discussion of this topic is beyond the scope of this paper.

(21) It is the government and some NGOs that tend to make these sorts of statements. See e.g. "Tories Aim to Fix 'Broken' Immigration System", CTVNews (16 February 2012), online: <>. See also Martin Collacott, "Canada's Inadequate Response to Terrorism: The Need for Policy Reform" (February 2006), online: <>, which refers to "our highly dysfunctional refugee determination system" (at 2).

(22) Canada, Library of Parliament, "The Refugee Status Determination Process: Pressure Points", by Margaret Young, Law and Government Division (Ottawa: LOP, 29 May 1986).

(23) This assessment is based on personal observation and discussions with members of the refugee bar.

(24) See Canada, Immigration and Refugee Board, Rebuilding Trust: Report of the Review of FundamentalJustice in Information GatheringandDissemination at the Immigration and Refugee Board of Canada, by James C Hathaway (Toronto: York University, 1993) at 38, n 64 [Hathaway, Rebuilding Trust].

(25) See ibid at 1. For a detailed discussion of the attacks on the IRB's legitimacy, see Mary C Hurley, "Principles, Pracrices, Fragile Promises: Judicial Review of Refugee Determination Decisions Before the Federal Courr of Canada" (1996) 41:2 McGill LJ 317 ar 372-75.

(26) Harhaway, Rebuilding Trust, supra note 24 at 37. (r) See ibid at 37.

(28) See ibid at 3.

(29) See ibid at 39.

(30) A descriprion of the selection process berween 2007 and 2012 can be found in Government of Canada, Governor in Council Appointment Process, Immigration and Refugee Board (January 2007), online: <>. It employs objective criteria and is more transparent. When members became civil servants in 2012, the criteria were restated. See e.g. Canada, GC Jobs, "Member, Refugee Protection Division--Inventory 'Amended Essential Qualifications", online: <>.

(31)" Audrey Macklin, "Asylum and the Rule of Law in Canada: Hearingthe Other (Side)" in Susan Kneebone, ed, Refugees, Asylum Seekers and the Rule of Law: Comparative Perspectives (New York: Cambridge University Press, 2009) 78 at 96.

(32) Qualifications for the advertised position focus on previous decision-making experience or experience presenting cases before courts or tribunals. Legal training or subject-matter expertise are considered only "assets", not essential qualifications. See Immigration and Refugee Board of Canada, "Member, Refugee Protection Division", GC Jobs (12 October 2017), online: < 1800?poster=840021 >. In contrast, the BIA requires members to have a law degree and at least seven years of legal experience, as well as comprehensive knowledge of immigration law or the proven ability to acquire such knowledge. See US, Department of Justice "Board Member", online: <>.

(33) The RPD recognition rate was 63.1% for the years 2013 and 2014, 58% in 2015, and 66% in 2016. See Angus Grant & Sean Rehaag, "Unappealing: An Assessment of the Limits on Appeal Rights in Canada's New Refugee Determination System" (2016) 49:1 UBC L Rev 203 at 264, Table 1; Canadian Council for Refugees [CCR], "2015 Refugee Claim Data and IRB Member Recognition Rates", (6 March 2017), online: <>.

(34) Douglas Quan, "Canada Accepting More Roma Asylum-seekers Amid Growing Evidence of Persecution in Hungary", National Post (23 August 2015), online: <>.

(35) Nicholas Keung, "Liberals Accused of'Housecleaning' of Tory Appointees at Refugee Board", Toronto Star (23 April 2017), online: <>.

(36) Laurence J Kirmayer, "Failures of Imagination: The Refugee's Narrative in Psychiarry" (2003) 10:2 Anrhropology & Medicine 167 ar 168.

(37) Rousseau er al, "The Complexity of Determining Refugeehood", supra note 17.

(38) Sec ibid at 55-57,62.

(39) See ibid at 64.

(40) See ibid at 62, 64.

(41) Francois Crepeau & Delphine Nakache, "Critical Spaces in the Canadian Refugee Determination System: 1989-2002" (2008) 20:1 Intl J RefugeeL 50. This research was done prior to the creation of the RAD.

(42) See ibid at 120-21.

(43) See ibid at 52-53.

(44) The IAB rendered several seminal decisions that are followed to this day. See e.g. Ribic v Canada (Minister of Employment and Immigration), [1985] IABDNo4(QL) [1985] DSAINo4, cited in Ivanov v Canada (Minister of Employment and Immigration), 2007 FCA315, [2008]2FCR502.

(45) For seminal BIA decisions, see e.g. Matter of Acosta, 19 I & N Dec 211 (BIA 1985); Matter of Chen 10 I & N Dec 671 (BIA 1964). Matter of Acosta was recently cited in Riosv Lynch, 807 F (3d) 1123 (9th Cir 2015).

(46) For a study of the variation in acceptance rates among first-instance refugee decision makers prior to the 2012 changes, see Sean Rehaag, "Troubling Patterns in Canadian Refugee Adjudication" (2008) 39:2 Ottawa L Rev 335 [Rehaag, Troubling Patterns]. For divergences in the post-2012 system, see Grant & Rehaag, supra note 33.

(47) Sec IRPA, supra note 1, s 163.

(48) A three-member panel was convened by the RAD in May 2017 in order to resolve an imporranr legal quesrion. Several inrerveners participated as well. Unfortunately, no clear resolurion was achieved because a split decision resulted and the matter is before the courts. See RAD Decision TB6-03419/20/21/22, online: <>. The US BIA invites interveners to participate in important cases regularly, for instance on the issue ofwhether family can constitute a "particular social group" under the Convention: US, Board of Immigration Appeals, "Amicus Invitation No.16-01-11" at 1, online: <>.

(49) Hathaway, Rebuilding Trust, supra note 24 at 3-37,43, 80.

(50) For instance, the Canadian Human Rights Tribunal requires at least four out of its fifteen members to have a law degree, and also requires experrise in human righrs issues, and knowledge of human rights law, public law and constirurional law, ere. See Canadian Human Rights Act, RSC 1985, c H-6, s 48.1(1), 48.1(3). The Competition Tribunal is staffed by Federal Court judges and lay members. Lay members are required to have a post-graduare degree in industrial organization economics, business, commerce, or finance; a professional qualificarion in business, accounting, of law; or extensive business-relared experience as a senior corporate executive, senior academic, or senior public servant; as well as an understanding of economics, markers, and financial statemenrs. See Canada, Competition Tribunal, "Member Appoinrmenrs", online: <>. The Parole Board of Canada requires members to have a degree in law, social work, psychology, or criminology, and knowledge of the criminal justice system and relevant societal issues.

(51) See IRPA, supra note 1, s 153(l)(a).Concernsaboutthecontinuedpoliticizationofthe RAD appointment process have been expressed by lawyers and former board members. The assertions are that "housecleaning" at the RAD occurred by reason of a 2015 change in government. A competency test administered post-election was described as "flawed" and "rigged": Keung, supra note 35.

(52) The criteria were explained in a recent recruitment campaign. See "Notice of Vacancies" (Immigtation and Refugee Board of Canada), (2015) C Gaz I, 1274 [IRB Notice of Vacancies, 2015]. Requirements are a degree from a recognized university or an acceptable combination of relevant education, job-related training and/or experience, a minimum of five years of experience as a decision maker in an adjudicative tribunal and/or in the interpretation or application of legislation. Knowledge of refugee law is considered an asset but not a requirement. RAD requires competency tests that are difficult and do not test subject-matter knowledge. There is anecdotal evidence that several experienced refugee lawyers and RPD members have not been able to pass them.

(53) See CCR, "2016 Refugee Claim Data and IRB Member Recognition Rates", (8 March 2017), online: <> [CCR,"2016Refugee Claim Data"].

(54) This number is almost 3000 claimants annually, as there are several categories of claimant who are "RAD-barred" under IRPA, s 110(2). See Grant &C Rehaag,supra note 33, Table 8.

(55) "This is the average success rate in perfected Federal Courr applicarions for judicial review brought by unsuccessful refugee claimants from 2005 to 2010": Granr & Rehaag, supra note 33 at 222, n 96. See also Grant & Rehaag, supra note 33 at 212; Rehaag, "Troubling Patterns", supra note 46. There has been no evidence of a significant change since then.

(56) Hurley, supra note 25 at 363.

(57) Macklin, supra, note 31 at 104-05.

(58) See Ian Greene & Paul Shaffer, "Leave to Appeal and Leave to Commence Judicial Review in Canada's Refugee-Derermination System: Is the Process Fair ?" (1992) 4:1 Inrl J Refugee L 71 ; Sean Rehaag, "The Luck of the Draw ?" (2012) 38:1 Queen's LJ 1 at 30 [Rehaag, "Luck of the Draw"].

(59) See Jon B Gould, Colleen Sheppard & Johannes Wheeldon, "A Refugee from Justice ? Disparare Treatment in the Federal Courr of Canada" (2010) 32:4 Law & Pol'y 454.

(60) For instance, in refusing to ser aside a negarive IRB decision involving a Roma family from Hungary who had been physically attacked several rimes, the judge stared: "The Hungarian situation is very difficult to gauge. Much will depend upon the facts and evidence adduced in each case, and on whether the RPD goes about the analysis in a reasonable way. Where it does, it is my view that it is not for this Courr to inrerfere even if I mighr come to a differenr conclusion myself": Molnar v Canada (Minister of Citizenship and Immigration), 2012 FC 530 at para 105, 214 ACWS (3d) 554. In another case, the judge stated: "I may well have come to different conclusions from the RPD, bur if I were to intervene I would merely be substituting my own assessment of the situation for that of the RPD, and this I cannot do": Trinidad Reyes v Canada (Minister of Citizenship and Immigration), 2012 FC 926 at para 32, 416 FTR 42. For more cases with similar laments, see e.g. Mohammed v Canada (Minister of Citizenship and Immigration, 2011 FC 1473, [2011] FCJ No 1869 (QL); Dinok v Canada (Minister of Citizenship and Immigration), 2014 FC 1199, [2014] FCJ No 1246 (QL); Paradi v Canada (Minister of Citizenship and Immigration), 2013 FC 996, [2013] FCJ No 1095 (QL); Horvath v Canada (Minister of Citizenship and Immigration), 2014 FC 313, [2014] FCJ No 330 (QL); Vargas vMCI, 2008 FC 1347, [2008] FCJ No 1706 (QL); Asadnejadv Canada (Minister of Citizenship and Immigration), 2014 FC 379, [2014] FCJ No 405; Jaivad v Canada (Minister ofCitizenship and ImmigrationJ, 2012 FC 232, [2012] FCJ No 232 (QL); Multani v Canada (Minister of Citizenship and Immigration), 2012 FC 734,218 ACWS (3d) 630; Majlatv Canada (Citizenship and Immigration), 2014 FC 965,246 ACWS (3d) 664.

(61) CCR, "The Experience of Refugee Claimants at Refugee Hearings at the Immigration and Refugee Board" (7 January 2012) at 11, online: <> [CCR, "Experience, 2012"].

(62) See ibid.

(63) See Peter Showier, Fast, Fair and Final: Reforming Canada's Refugee System (Ottawa: The Maytree Foundation, 2009) at 8 [Showier, Fast, Fair and Final].

(64) Immigration and Refugee Protection Regulations, SOR/2002-227, ss 159.9(l)(b), 159.92. Access to PRRA and humanitarian relief has been significantly curtailed--it is normally only available on certain restricred grounds and/or after a significant period of time has elapsed since the claim was refused. See IRPA, supra note 1, ss 112(2)(b.l), 25(1.2)(c).

(65) Canada has traditionally seen some 20,000-30,000 claims annually. See e.g. Statistical Unit, United Nations High Commissioner For Refugees, "Refugees and Others of Concern to UNHCR: 1997 Statistical Overview", UNHCR (July 1998) at 37, online: <>; Immigration and Refugee Board of Canada, "Refugee Protection Claim Sraristics", online: <>. There have been highs and lows--approximately44,000 claims in 2001 and only approximately 10,000 in 2013. See Immigration and Refugee Board of Canada, "Refugee Protection Claims (New System) by Country of Alleged Persecurion- 2013", online: <>; "Asylum and Refugee Admission in Industrialised Counrries", UNHCR (8 November 2002) at 64, online: < There is already evidence that processing targets in the 2012 reforms are not being met. See Kelly Malone & Karen Pauls, '"Justice Delayed is Justice Denied': Fewer Than Half of Refugee Claims Being Heard on Time", CBCNews (7 June 2017), online: <>.

(66) See Peter Showier, Alexandra Belluz & GregEraux, "Legal Aid for Refugee Claimants in Canada", Refugee Forum & University of Ottawa Refugee Assistance Project (September 2012), online: < September_2012.pdf>. The funding levels have not changed significantly since this report was drafted.

(67) See Bridget M Haas, "Citizens-in-Waiting, Deportees-in-Waiting: Power, Temporality, and Suffering in the U.S. Asylum System", (2017) 45:1 Ethos 75 at 93.

(68) [1985] 1 SCR 177,17 DLR (4th) 422.

(69) Employment and Immigration Canada, Refugee Determination in Canada: Proposals for a New System; Report to the Honourable Flora MacDonald, Minister of Employment and Immigration, by W Gunther Plaut (Ottawa: Minister of Supply and Services Canada, 1985) at 107-18 [Plaut Report]. For a description of the system from 1976to 1989see Christopher J Wydrzynski, "Refugees and the Immigration Act" (1979) 25:1 McGillLJ 154 at 160-66.

(70) Hathaway, Rebuilding Trust, supra note 24.

(71) See Law Reform Commission of Canada, The Determination of Refugee Status in Canada: A Review of the Procedure, Draft Final Report, March 1992.

(72) See Citizenship and Immigration Canada, NotJust Numbers : A Canadian Framework for Future Immigration, by Susan Davis, Roslyn Kunin & Roberr Trempe, Catalogue No Q63-21/1998E (Otrawa: Minister of Public Works and Government Services Canada, 1998).

(73) Showier, Fast, Fair and Final, supra nore 63 at 14.

(74) See IRB, Review 2017, supra note 6.

(75) This was called the Refugee Status Advisory Committee (RSAC). There was a right to a rederermination of the claim by the IAB based on the transcript of the proceedings only.

(76) Hathaway, Rebuilding Trust,supra note 24 at 3, 33-37,43, 80.

(77) The appeal and adjudicative functions of the IRB in non-refugee immigration matters could remain within an administrative tribunal.

(78) American Bar Association (ABA), Commission on Immigration, Reforming the Immigration System: Proposals to Promote Independence, Fairness, Efficiency, and Professionalism in the Adjudication of Removal Cases (Washington, DC: ABA, 2010) [ABA, Reforming the Immigration System].

(79) Senate of Australia, A Sanctuary under Review: An Examination of Australia's Refugee and Humanitarian Processes (Canberra: Senate of Australia, 2000).

(80) ABA, Reforming the Immigration System, supra, note 78 at 6-35 to 6-36. The ABA favoured the tax court model over the idea of an independent administrative tribunal.

(81) Senate of Australia, supra note 79 at 201.

(82) Mary Hurley notes that "[r]ecent upheavals at the Board prompred some observers to call for the replacement of lay Board members as firsr-insrance decision makers by immigration judges or the transfer of Boardjurisdiction to a refugee division of the Federal Court": supra note 25 at 328, n 318 [emphasis added], citingFC Leclerc, "FaultyJustice: Immigration Tribunals Have Been a Disaster" The [Montreal] Gazette (14 November 1994) B3; D Stoffman, "Canada's Farcical Refugee System", The [SaintJohn] Telegraph Journal (6 September 1995) A9.

(83) In a comprehensive reform proposal pur forward in the Unired States, author David Martin had this to say: "Almost no one believes feasible a system in which the courts make de novo determinations of asylum": David Martin, "Reforming Asylum Adjudication: On Navigating the Coast of Bohemia" (1990) 138:5 U Pa LR 1247 at 1362.

(81) See Tax Court of Canada Rules (Informal Procedure), SOR/90-688b [Tax Court Informal Procedure].

(85) See Tax Court of Canada, "Judges" (9 September 2015), online: <>.

(86) See Ronald Appleby & A Lome Greenspoon, "The Tax Court of Canada and the Tax Appeal Process" (1984) 5:3 Adv Q331 at 331-32.

(87) Ian MacGregor et al, "The Developmenr of the Tax Courr of Canada: Srarus, Jurisdiction, and Stature" (2010) 58:4 Can Tax J 87 ar 98-100 [foornores omitted]. This assessment was echoed by Patrick Monahan: "The Tax Court of Canada has established a strong repuration in the tax community as an efficient, fair and independenr adjudicator": Patrick Monahan & Elie S Rorh, Federal Regulation of Charities: A Critical Assessment of Recent Proposals for Legislative and Regulatory Reform (Toronro: York University Press, 2000) ar 105.

(88) MacGregor er al, supra note 87 at 99.

(89) Brendan Kennedy, "Heated Exchange over Legal Rights as Lawyers Battle to Have Refugee Claimant Let Out of Jail", Toronto Star (16 June 2017), online: <>. The lawyer wanted the evidence relied on by the government in its submissions to be disclosed; the board member said that "the government's oral submissions are regularly accepted as evidence in detention reviews, which are intended to be less formal than court proceedings" (ibid). The lawyer stated that his client was detained and was "in a maximum-security jail and... [therefore had] a heightened right to complete disclosure of any and all information" (ibid). The board member responded that "she saw no basis for the hearing to be used as a 'fishing expedition'" (ibid). Based on my personal experience and observation, many refugee lawyers have had these types of encounters.

(90) As recounred to the aurhor by the IRB member involved.

(91) See Kirmayer, supra note 36; Rousseau er al, "The Complexity of Determining Refugehood", supra nore 17; Crepeau & Nakache, supra note 41.

(92) Canadian Association of Superior Courr Judges, "The Qualities Required of a Judge" (2006), online: <>.

(93) See Lawrence Baum, "Judicial Specialization and the Adjudication of Immigration Cases" (2010) 59:8 Duke LJ, 1501 at 1561: "Decisions about judicial specialization in any field should be made with careful consideration of the potential effects, including the uncertainties that exist in predicting those effects."

(94) For a discussion, see Lome Sossin, "Designing Administrative Justice: Draft" (Paper delivered at the British Columbia Continuing Legal Education Administrative Law Conference, 18 November 2016), (2017) 13:6 Osgoode Legal Studies Research Paper Series 26.

(95) See Ibid at 8, quoting Rasanen v Rosemount Instruments Ltd (1994), 17 OR (3d) 267 at para 35, 112 DLR (4th) 683 (CA).

(96) Interviews and hearings with hundreds of claimants show them to be primarily concerned with obtaining permission to remain, as opposed to what they have to go rhrough to achieve this resulr--alrhough some occasionally express extreme frusrration with the length of the process or acute distress ar being derained, for example.

(97) The backgrounds of members at the IRB in Montreal berween 1993 and 1998 included real esrare, business, missionary work, priesrhood, municipal government, community organizing, UNHCR field work, refugee law, civil lirigarion, and academia. Many had been acrive in local polirics.

(98) Liberal Party of Canada, "A New Plan for Canadian Immigration and Economic Opportunity", online: <>. So far this goal has not been met.

(99) See Rousseau et al, "The Complexity of Determining Refugeehood", supra note 17 at 117.

(100) See Rehaag, "Luck of the Draw", supra note 58.

(101) See e.g. Kaur v Canada (Minister of Employment & Immigration) (1989), 104 NR 50 (FCA) [Kaur v MEI] (the Charter allows a battered woman to claim defence of duress in order to reopen inquiry and claim refugee status); Yusuf v Canada (Minister of Employment & Immigration) (1991), [1992] 1 FC 629, 133 NR 391(FCA) (IRB member's sexist comments have no place in the justice system); Cheung v Canada (Minister of Employment & Immigration) (1993), [1993] FCR314,102DLR(4th) 214 [Cheung v MEI] (forced sterilization recognized as a form of persecution). In the so-called "trilogy" (Attakora v Canada (Minister of Employment &Immigration) (1989), 99 NR 168, FCJ No 444 (FCA) [Attakora vMEI], Owusu-Ansah v Canada (Ministerof Employment & Immigration) (1989), 98 NR 312, 8 Imm LR (2d) 106 (FCA) [Owusu-Ansah v MEI], and Armson v Canada (Minister of Employment & Immigration) (1989), 101 NR 372,9 Imm LR (2d) 150 (FCA) [Armson vMEI]), the Federal Court of Appeal attempted to put a stop to irrational and arbitrary credibility assessment in refugee cases with lasting effect. In Canadian Doctorsfor Refugee Care, supra note 4, the Federal Court ruled that cuts to a publicly funded health insurance scheme for refugees were unconstitutional, andin Z(Y), supra note 4, ruled that inferior refugee procedures based on country of origin violated principles of equality. In the United States, the federal judiciaty prohibited the government from instituting a mass expedited asylum determination system and encouraging mass arrivals to withdraw their claims (see Haitian Refugee Center v Smith, 676 F (2d) 1023 (5th Cir 1982) and Orantes-Hernandez v Meese, 685 F Supp 1488 (CD Cal 1988), respectively). More recently, scathing opinions by the Federal Circuit Courts of Appeal on the poor quality of initial asylum determinations have been credited with creating impetus for reform (see Stacy Caplow, "After the Flood: The Legacy of the 'Surge' of Federal Immigration Appeals" (2012) 7:1 Norrhwesrern JL & Social Policy 1). In the United Kingdom, the High Court and the Court of Appeal forced the government to remove particular countries from irs so-called "white list" of safe countries of origin (see R (on the application ofjaved) v Secretary of Statefor the Home Department, [2001 ] EWCA Civ 789 and A (on the application ofHusan) v Secretary of Statefor the Home Department, [2005] EWHC 189 (Admin)) and the House of Lords blocked the government from withdrawing state support from asylum seekers who had fded their claims late (see R(on the application ofLimbuela) v Secretary of State for the Home Department, [2007] 1 All ER 951). In Australia, the federal courts initially attempted to restrain the government over the MV Tampa incident and, after a ten-year srruggle, dealr a dearh blow to the "Pacific Solution" once and for all. See Victorian Council for Civil Liberties Inc v Minister for Immigration & Multicultural Affairs (Tampa) (2001), 182 ALR617, FCR452 and PlaintiffM69 of2010 v Commonwealth of Australia (2010), 274 ALR 14 (HC), 243 CLR 319, respectively.

(102) Interestingly, panels of Sweden's migrarion courrs hearing asylum appeals can be composed of eirher one professional judge or one professional judge and a committee of lay persons (usually represenrarives of various polirical parties). A recenr srudy found rhar rhese lay judges tended to be more influenced by their own personal partisan beliefs rhan the professional judges. See Linna Marten, "Political Bias in Court? Lay Judges and Asylum Appeals" (2015) Deparrment of Economics, Uppsala University Working Paper 2015:2, online: < /824067258.pdf>.

(103) See e.g. Tax Court Informal Procedure, supra note 84.

(104) See ABA, Reforming the Immigration System, supra note 78 at 6-35.

(105) See R v Moses (1992), 71 CCC (3d) 347, [1992] 3 CNLR116, where a sentencing circle was employed by a Yukon judge for the first time.

(106) For instance, the Court of Quebec has a special Youth Division and Ontario has special family divisions within the court system. The Ontario family court judges take an active role in the establishment of special procedures for family law cases. See Ontario, The Superior Court ofJustice: Realizing Our Vision--Report for 2015 and 2016, online: <> at 11.

(107) See Tax Court Informal Procedure, supra note 84.

(108) See Tax Court of Canada Act, RSC 1985, c T-2, s 18.15(3).

(109) The Court anticipates unrepresented litigants, as evidenced by its website: Tax Court of Canada, "Self-Represented Litigants and Counsel", online: <>.

(110) The Tax Court anticipates hearing individual taxpayers tell their stories; the court does not hold itself out to be the exclusive domain of corporations and high-priced tax lawyers, but makes an effort to encourage ordinary citizens to access the Court, even without a lawyer. See for example Tax Court of Canada, "Your Day in Court" (9 September 2015), online: <>.

(111) See ABA, Reforming the Immigration System, supra note 78 at 6-6.

(112) See Joseph F Fletcher & Paul Howe, Public Opinion and the Courts (Montreal: Institute for Research on Public Policy, 2000). See also Mary Srrarron, Public Perceptions of the Role of the Judiciary (Edmonron: Canadian Forum on Civil Justice, 2005). The Canadian Judicial Council has launched 14 inquiries since 1990. See Canadian Judicial Council, "Inquiries and Investigations", online: <>.

(113) Plaut Reporr, supra note 69 at 20.

(114) For details, see supra note 15.

(115) If enough judges could not be found, a system of part-time judges could be implemented--meaning that lawyers could sit part time as judges from time to time where no conflict of interest would arise. This is allowed in some jurisdictions.

(116) An Asylum Coutt would likely have to be quite large. The pre-2012 RPD members finalized an average of around 110 claims each per year (see CCR, "2010 Refugee Claim Data and IRB Member Grant Rates" (1 March 2011), online: <>). [CCR, "2010 Refugee Claim Data"]. Current RPD members finalized an average of 69 claims each per year in 2016 (see CCR "2016 Refugee Claim Data, supra note 53). For comparison purposes, the 27 judges of the Tax Court made some 5,600 dispositions in 2013-2014, or an average of207 each per year (see Canada, Courrs Adminisrrarion Service, 2013-2014 Annual Report: Supporting the Delivery of Justicefor All Canadians, online: <> at 10 [Courrs Adminisrrarion Service, 2013-2014 Annual Report}). These figures suggest that 125 cases per decision maker per year would be possible, requiring an asylum court comprised of some 80 judges in order to finalize 10,000 claims per year.

(117) See Courts Administration Service, 2013-2014 Annual Report, supra note 117 at 10; CCR, "2014 Refugee Claim Data and Member Recognirion Rares" (8 May 2015), online : < http : //ccr>.

(118) See IRB, Norice of Vacancies, 2015, supra note 52. For Tax Court judges' salaries see Judges Act, RSC 1985, cj-l, s 11. Currently, 135 RPD and 25 RAD members cost the government approximately 16 million dollars annually in salary. Having 70 Asylum Courr judges with salaries ar the same level as Tax Courr judges would mean expending some 21 million dollars in salary annually. If 70 judges were not enough, the costs would increase proportionally.

(119) Citizenship and Immigration Canada "estimates some [5.6] million dollars annually is spenr on PRRA and humanitarian with risk applications": Citizenship and Immigration Canada, Formative Evaluation of the Pre-Removal Risk Assessment Programme (2008), online: <>. "The Minister of Immigration has estimated that 165 billion is spent on social benefits for extended claimant stays" (ibid). See Tobi Cohen, "Refugee Claimants Crackdown Draws Fire", Montreal Gazette (17 February 2012).

(120) See ABA, Reforming the Immigration System, supra note 80 at 6-34.

(121) "Parajudicial" is distinct from "quasi-judicial" and suggests decision making exclusively by lawyers.

(122) See supra note 15 for a description of some major asylum systems.

(123) The Bastarache Report states that "[l]e TAQpossede plusieurs des attributs d'une cour de justice et ses fonctions sont importantes pour la societe quebecoise": Michel Bastarache, Commission d'enquete sur le processus de nomination des juges du Quebec: synthese (Quebec: Les Publications du Quebec, 2011) at 20.

(124) See Administrative Justice Act, CQLR c J-3, ss 18-40.

(125) See Lise I Beaudoin, "Projet de loi 4 sur la justice administrative: La presence de juristes sur les banc doit etre maintenue" (2003) 35:16 J Barreau.

(126) See Administrative Justice Act, supra note 124, s 33.

(127) See ibid, ss 18-40.

(128) See ibid, s 22.1. Notaries in Quebec arrend law school.

(129) See ibid, s 33.

(130) See also Rousseau et al, "The Complexity of Determining Refugeehood", supra nore 17.

(131) See comments in Showier, Fast, Fair and Final, supra note 63 at 14. See also Martin, supra note 83 at 1358.

(132) Rousseau et al, "The Complexity of Determining Refugeehood", supra note 17.

(133) The qualifications of the members of this de novo appeal tribunal are statutorily mandated in New Zealand--namely, a chair who is a District Court judge, other members appointed by the Governor General who are lawyers having held a practicing certificate for at least five years, and representatives of the UNHCR. Former refugee and protection officers, immigration officers, or visa officers are not allowed to be members of the tribunal. See Immigration Act 2009 (NZ), 2009/51, s 219.

(134) See Hathaway, Rebuilding Trust, supra note 25; Hurley, supra note 25.

(135) RPD members are employed civil servants who do not need to continually seek reappointment. RAD members are still political appointees, but the hiring process has been largely de-politicized.

(136) I.e., legislating rules as to member qualifications and/or panel composition; see e.g. Quebec's rules about the TAQ: Administrative Justice Act, supra note 124, ss 18-40.

(137) Based on personal experience and discussions with counsel, some of the best members have been lay persons--one had been a homemaker, one a priest, one a missionary, one a municipal functionary, one a businessperson and hockey coach. Conversely, one intensely negative member had been a refugee lawyer in BC, and another had been a Jesuit refugee aid worker.

(138) IRPA, supra note 1, ss 96-97.

(139) Although 2012 reforms called for all PRRA applications to be determined by the IRB instead of by the immigrarion department, this has not yet been implemented and, in any event, determinarions would likely be made by funcrionaries wirhin the IRB, not IRB members themselves.

(140) The idea of humanitarian jurisdiction was considered by Showier in Fast, Fair and Final, supra note 63 at 17, but no recommendation was made. In contrast, Joseph Rose advocated in favour of such a jurisdiction in the American context. See Joseph Rose, "The Asylum Seeker: The Proverbial Rat in the Evidentiary Maze of Asylum Law? Some Suggestions for Reform" (1994) 24:2 Southwestern UL Rev 473 at 499.

(141) A consolidated grounds approach which includes nonpersecutory humanitarian factors (such as health, environmental disaster, domestic abuse, family problems, family unity, and ties to the host country) has been employed in various European countries in the past. See European Council on Refugees and Exiles (ECRE), Complimentary Protection in Europe (July 2009), online: <>. For example, Finland has a consolidared, front-end prorecrion model whereby residence permits can be based on international prorection (refugee status), subsidiary protection (capital punishment, torture, cruel and unusual treatment), and protection on compassionate grounds (healrh or ties established in Finland or for some other individual, humane reason) or other grounds (family ties, work, and studying). See Finnish Immigration Service, "Residence Permit on Other Grounds", online: < applying_for_asylum/decision/residence_permir_on_other_grounds>. Humanirarian prorecrion for civil srrife was abolished in 2016: see Finnish Immigrarion Service, "Humanirarian Protection No Longer Granred; New Guidelines Issued for Afghanisran, Iraq and Somalia" (17 April 2016), online: <hrrp://>. Belgium has humanitarian protection based on medical grounds (see Loi sur l'acces au territoire, le sejour, l'etablissement et l'eloignement des etrangeres (Aliens Acr), 15 December 1980, art 9ter) and Germany currently has a "ban on deporrarion" category that could include rhreats from medical conditions: Federal Office for Migrarion and Refugees, "National Ban on Deportation", online: <>.

(142) In the United States there is a right to apply to reopen a BIA appeal in limited circumstances. See Aliens and Nationality Act, 8 CFR[section]1003.2(a), [section]1003.23(b)(l).

(143) As a last resort, claimants whose circumstances have suddenly changed dramatically could always apply to the Federal Court for some form of injunctive relief.

(144) See Federal Courts Act, RSC 1985, c F-7, s 27. The Court's appeal jurisdiction is more limited in respect of Tax Court cases that follow the informal procedure (ibid, s 28). The Court also conducts judicial review of several federal administrative tribunals under section 28 of the Federal Courts Act. Until 1978, the former IAB's decisions were appealable (with leave) to the Federal Court of Appeal under the Immigration Appeal BoardAct, RSC 1970, c 1-3, s 23(1). After 1978, IAB decisions were subject to judicial review under section 28 only. Prior to the creation of the Federal Court in 1971, IAB decisions were appealable directly to the Supreme Court of Canada (with leave). See e.g. Gana v. Minister of Manpower and Immigration, [1970] SCR 699,13 DLR (3d) 699. Upon the creation of the IRB in 1989, a negative decision could be appealed (with leave) to the Federal Court of Appeal under section 82.3 of the Immigration Act, 1976, supra note 1. In 1993, the remedy was downgraded to judicial review with leave, and the forum was changed to the Federal Court Trial Division (now known as the Federal Court).

(145) For insrance, between 2005-10, inclusive, there were some 23,000 leave applications filed against negarive refugee determinations. See Rehaag, "Luck of the Draw", supra note 58. This averages out to about 4,000 cases per year.

(146) There is the possibility of an oral hearing before the US Circuir Courts of Appeals, and such a possibility could also be allowed under our proposals.

(147) See supra note 144, s 52.

(148) See IRPA, supra note 1, ss 110(3), (4).

(149) See Greene & Shaffer, supra note 58. In the course of conducting this study, the authors found that the 12 judges of the Federal Court of Appeal had disposed of 2000 leave applications in 1990 and the first quatter of 1991.

(150) Using statistics from before the 2012 reforms, 25 judges of the Federal Court disposed of over 9,000 cases in 2010-11 (see Courts Administration Setvice, Annual Report 2010-2011, at 22). According to Rehaag's research, some 4000 of these would have been refugee cases. (See Rehaag, "Luck of the Draw", supra note 58 at 2). In the United States, over roughly the same time period (March 2010 to March 2011), the 44 judges of the Ninth Circuit Court of Appeals disposed of some 4,000 administrative appeals on the merits See Administrative Office of the United States Courts, 2011 Annual Report, Washington DC, 2012 at 59 (online: < l.pdf, accessed 11 Dec 2017>). Most of these would have been from the BIA (ibid chart on page 60) and disposed of without an oral hearing.

(151) For instance, on 14 June 2012, the Second Circuit disposed of a BIA appeal from a Mauritanian asylum applicant dealing with issues of credibility assessment and ineffective assistance of counsel in a two-page, unpublished opinion. See Ba v Holder, 474 Fed Appx 837,2012 WL 2146323 (2d Cir 2012).

(152) See Federal Courts Citizenship, Immigration and Refugee Protection Rules, SOR/93-22, r 10-13.

(153) See Rehaag, "The Luck of the Draw", supra nore 58 at 15.

(154) For an in-depth discussion of the viabiliry of specialized judges in the asylum context, see Baum, supra note 93.

(155) See Federal Courts Act, supra note 144, s 52(c).

(156) See ibid, ss 27(1.1)-(1.4).

(157) A database search indicares rhar, since implementarion of the RAD in 2012, the Federal Court has reviewed 425 RAD decisions and ser aside 161 of them. See CanLIl, online: <>.

(158) Negative decisions of the IAB were reviewable by the Federal Courr of Appeal under section 28 of the previous Federal Court Act, RSC 1970 (2nd Supp), c 10. A direct appeal as of right to the Federal Courr of Appeal from a refugee board was part of one of three models (Model A) presented in the 1985 Plaut Report, supra note 69, but was not adopted.

(159) For example, there were over 13,000 negative decisions rendered by the IRB in 2010. See CCR, "2010 Refugee Claim Data", supra, note 117.

(160) In its extensive study of reform of the immigration system, the ABA did not identify the written appeal as a problem. See ABA, Reforming the Immigration System supra note 78. Current RAD appeals are normally done in writing, as were the redetermination applications at the former IAB. It is a familiar form.

(161) A number of landmark decisions were rendered by the Federal Court of Appeal during this period. See e.g. Attakora v MEI, supra note 101; Adjei v Canada (Minister of Employment & Immigration) (1989), 57 DLR (4th) 153, 132 NR 24; Salibian v Canada (Minister of Employment & Immigration) (1990), 73 DLR (4th) 551,113 NR 123; Ramirez v Canada (Minister of Employment & Immigration) (1992), 89 DLR (4th) 173, 135 NR 390; Cheung v MEI, supra note 101; Rasaratnam v Canada (Minister of Employment & Immigration), [1991] FCJ No 1256, 140 NR 138. See also Hurley, supra note 25 at 378-80.

(162) Our suggested reforms would make no changes to the Federal Court's power to judicially review decisions of immigration officers and visa officers, or to issue injunctions or other prerogative writs.

(163) See Courts Administration Service, 2013-2014 Annual Report, supra note 117 at 7.

(164) See ibid at 8.

(165) Federal Courts Act, supra note 144,s. 52(c). The US Circuit Courts of Appeal can do this with respect to the BIA as well--either remand or reverse.

(166) Between 1984 and 1993 the Federal Court of Appeal declared the applicant to be a Convention refugee in the following cases: Dhillon v Canada (Minister of Employment & Immigration), [1990] FCJ No 1040, 131 NR 62; Mahathmasseelan v Canada (Minister of Employment & Immigration), [1991] FCJ No 1110, 137 NR 1; Hilo v Canada (Minister of Employment & Immigration), [1991] FCJ No 228, 130 NR 236; Sabaratnam v Canada (Minister of Employment & Immigration), [ 1992] FCJ No 901, 1992 CarswellNar 1182,; Ali v Canada (Minister of Employment & Immigration), [1992] AC F No 892, [1992] FCJ No 892; Veeravagu v Canada (Minister of Employment & Immigration), [1992] FCJ No 468, 1992 CarswellNat 1270; Wu v Canada (Minister of Employment & Immigration), [1992] FCJ No 458, 1992 CarswellNat 1269; Addo v Canada (Minister of Employment & Immigration),(1992), 142 NR 170, 1992 CarswellNat 553; Chen v Canada (Minister of Employment & Immigration), [1993] FCJ No 1436, 1993 CarswellNat 1980; Cheung v MEI, supra, note 102. In Surujpal v MEI (1985) 60 NR 73, Chaudri v Canada (Minister of Employment & Immigration) [ 1986] FCJ No 363,69 NR 114, Attakora v MEI, supra note 102, and Jocelyn v Canada (Minister oj"Employment& Immigration) (1991), 15 Imm LR (2d) 272, 1991 CarswellNat 100, the IRB was directed to find the applicant a Convention refugee on rehearing. Attakora v MEI was a credibility case.

(167) See Showier, Fast, Fair and Final, supra note 63 at 8.

(168) See e.g. Federal Court file IMM-3132-11: proceeding commenced 11 May 2011, leave granted 19 Sept 2011, oral hearing 15 Dec 2011, and decision rendered with reasons 6 Jan 2012. More recently, see file IMM-2048-16: proceedingcommenced 15 May 2016, leave granted 16 August 2016, oral hearing 27 September 2016, decision rendered with reasons 9 February 2017.

(169) For instance in Tax Court file 2009-782 (a case of average complexity involving a taxi driver who was audited), the notice of appeal was filed on 25 October 2011 (Federal Court of Appeal file A-398-11) and the appeal was disposed of on 26 July 2012. In Tax Court file 2015-123 (involving an individual's charitable donations), the notice of appeal was filed on 4 January 2016 and the appeal was disposed of on 29 November 2016.

(170) See United States Courts, "Table B-4C--U.S. Courts of Appeals Judicial Business (30 September 2015), online: <>.

(171) Sec ibid.

(172) See Caplow, supra note 101.

(173) There are currently 16 judges of the Federal Court of Appeal, and 42 judges of the Federal Court. See Federal Court of Appeal, "Judges of the Courr", online: < fca-caf_eng/judges-juges_eng> and Federal Court, "Judges and Prothonotaries" online: <hrrp://>.

(174) See Kaur v MEI, supra note 101; Yusuf v Canada (Minister of Immigration & Employment), [1991] FCJ No 1049, 133 NR 391; Cheung v MEI, supra note 101; Attakora v MEI, supra note 101, Owusu-Ansah v MEI, supra note 102; Armson v MEI, supra note 101.

(175) In this respect, the Canadian government's 2010-2012 reforms were positive. Processing times were established in regulations. See Immigration and Refugee Protection Regulations, supra note 64, ss 159.9(l)(b), 159.92.

(176) Peter Showier advocates for a total processing time of 13 months. See Showier, Fast, Fair and Final, supra note 63 at 20.

(177) The IRB was able to process claims wirhin 7 months in 1993-1994 and finalized some 24,000 claims in that period. See Immigrarion and Refugee Board of Canada, Performance Reportfor the Period Ending March 31,1996 (Ottawa: Minister of Public Works and Government Services, 1996) ar 9-10. In 2014 the currenr RPD was able to hear only 49% of claims wirhin the srarurorily mandated 60-day time frame. See Immigrarion and Refugee Board of Canada, Performance Report for the Period Ending March 31, 2014 (Ottawa: Minister of Citizenship and Immigration, 2014) at 14. In April 2017 only 41% of hearings were held on rime. See Malone & Pauls, supra note 65.

(178) Sixty days for an RPD hearing under the Immigration and Refugee Protection Regulations, supra note 64, s 159.9(l)(b).

(179) Average wait times, as of June 2017, were 5.6 monrhs. See Nick Wells, "Legal Limbo: Massive Backlog Leaves Asylum Seekers in for a Long Wait", CTVNews (27 June 2017), online: <>. According to IRB statistics, the average rime for finalizinga refugee claim ar the RPD was 2.9 months. However, it is not clear from what point this time period runs. The RPD is currenrly receiving abour 2,000 more claims than it finalizes. See Immigration and Refugee Board of Canada, Performance Report for the Period Ending March 31, 2016 (Ortawa: Minister of immigrarion, Refugees and Citizenship, 2014) at 21 [IRB, Performance Report 2016].

(180) Showier supports this time frame for refugee determinarions. See Showier, Fast, Fair and Final, supra note 63 at 20.

(181) PCISA, supra note 1, the Canadian government's 2012 asylum reform initiative, envisioned a total processing time of 216 days--with initial hearings taking place within 60 days and internal appeals complered within 105 days. See Government of Canada, "Archived: Backgrounder: Summary of Changes to Canada's Refugee System in the Protecting Canada's Immigration System Act" (16 February 2012), online: < 16f.asP>.

(182) See CCR, "Experience 2012", supra note 61 at 25.

(183) Haas, supra note 67 at 93.

(184) See IRB, Performance Report 2016, supra note 179 at 22. The 90-day deadline was met only 53 percent of the time in 2015-2016.

(185) See IRPA, supra note 1, s 74(b).

(186) See Peter Showier, "The Bogus Refugee Myth", Ottawa Citizen (13 Aug 2009): "The real problem is delays in the claim process that are a principle facror in arrracting fraudulenr claimants who count on years of employment in a wealthy economy before being deporred."

(187) This saying has been arrribured to Andrew Carnegie.
COPYRIGHT 2018 University of British Columbia Law Review Society (Canada)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Zambelli, Pia
Publication:University of British Columbia Law Review
Date:Jan 1, 2018

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |