Unsafe from Any Angle: Vulnerability-Generation on the US–Canada Border

Abstract

This article provides a review of the functioning and legality of the Safe Third Country Agreement between Canada and the United States, placing it in the broader context of systemic factors that generate and exacerbate the vulnerability of protection seekers. It offers a critical evaluation of what the legal challenges against the STCA reveal about the promises and limitations of safe-country-related litigation and the future of the Agreement.

1. Introduction

The ‘Safe Third Country Agreement’ (STCA) between Canada and the United States requires refugee claimants to request refugee protection in the first safe country they arrive in, unless they qualify for a narrow set of exemptions. In practice, it prohibits third-country nationals who first set foot in the US from making a refugee claim in Canada (and vice versa) at the official land ports of entry (POE)1. In its nearly two decades of functioning, the Agreement has made the US–Canada border less secure by diverting migratory movements away from official ports of entry (POE), and made asylum seekers more vulnerable by returning them to the United States despite significant discrepancies between the quality of protection provided by the two asylum systems. The Agreement has been repeatedly found unconstitutional by Canadian federal trial courts, only to narrowly escape suspension and revocation on appeal. Many of the foundational concerns about the legitimacy and efficacy of such agreements, and indeed about the concept of ‘safe country’, remain open questions.

In order for the right to access asylum to be fully realized, it must be matched with a corresponding allocation of responsibility for determining asylum claims. As Itamar Mann aptly put it, ’to identify someone’s right, we must be able to determine a person or authority with the corresponding duty’ (Mann 2020). This requires the effective connection of protection seekers and their protection claims with corresponding State duties. It is in the making and unmaking of this necessary connection that the vulnerability of migrants and their potential erasure is perhaps best captured. It is both inherent to the process of protection seeking and systemically generated and exacerbated. Writing on the matter of statelessness in relation to the Holocaust, Arendt remarked that ’Not the loss of specific rights, then, but the loss of a community willing and able to guarantee any rights whatsoever, has been the calamity that has befallen ever-increasing numbers of people’ (Arendt 2001, p. 297). Safe country agreements represent a loss of the will to protect, with an increasing number of wealthy countries avoiding and externalizing their obligations under international refugee law. Court challenges brought against such agreements are an exercise in internalizing the consequences of harms perpetuated when countries use asylum cooperation as a form of responsibility deflection.

Building on legal and policy analysis as well as relevant case law, this article argues that, given its general shortcomings, the STCA represents a form of ‘systemically generated vulnerability’, exacerbating the challenges embedded in the protection-seeking process along the US–Canada border. It further demonstrates how the operation of the Agreement exacerbates the ‘inherent vulnerabilities’ associated with certain subgroups of migrants who are traditionally recognized as vulnerable based on gender and age. The US–Canada safe third country agreement exhibits several symptoms of politically and pragmatically motivated manipulation of the concept of ‘safety’, ultimately at the expense of the actual safety and legitimate rights and interests of asylum seekers, and in violation of Canada’s obligations under international law and its own constitution.

After the Introduction, Section 2 describes the basic content and functions of the STCA, while Section 3 provides a brief overview of the illusive concept of vulnerability and its relevance to migration governance. Section 4 focuses on how the Agreement generates systemic vulnerability for migrants by incentivizing unsafe border crossing, and has a weak review process and a flawed ‘safe country’ determination process that too often ignore the substantive differences between Canadian and US asylum systems. Section 5 examines how the Agreement exacerbates the situation for certain subgroups of protection seekers found to be ‘inherently vulnerable’ in migration and asylum governance: unaccompanied minors and women fleeing gender-based violence. The Federal Court and advocates seem to be playing an elaborate game of ‘whack-a-mole’ with the Court of Appeals, wherein the Court of Appeals continuously moves the target without actually dismissing findings of genuine rights violations and harms suffered by refugee claimants subject to STCA enforcement. Given these realities, the Agreement is sure to face future challenges on both the political and judicial fronts.

2. What Is the Safe Third Country Agreement?

Much like the definition of vulnerability (see more in Section 3), the safe third country concept is applied broadly yet inconsistently, based on ambiguous criteria across the world. Whereas legally and administratively recognized ‘vulnerability’ tends to trigger states into granting special protections and accommodations to migrants, the ‘safe third country’ concept seeks to remove such state responsibility under the blanket assumption that protection can and should be found outside of its borders.

The concepts of ’safe country’, ‘safe third country’ or ‘first country of asylum’ are designed to create a legal ‘buffer zone’ between the protection seekers and states they are seeking protection from. They fit into the tapestry of externalized migration controls and ‘non-entrée’ policies, well-documented across Western and Northern countries (Ghezelbash 2018; Gil-Bazo 2015; FitzGerald 2019; Hathaway and Gammeltoft-Hansen 2014, p. 102; ECRE 2017). Perhaps the most sophisticated mechanism developed by states to embody the ‘safe third country’ notion is contained in the so-called Dublin III Regulation,2 designed to give member states the right to transfer migrants to the ‘first country of entry’ to the EU.3 The aim is to prevent asylum seekers (with some exceptions to the rule) from applying for asylum if they have transited through a ‘safe country’, meaning EU Member States in particular.4 Australia’s ‘Pacific Solution’ is another well-known and widely criticized third-country immigration processing scheme(Lacertosa 2014). These approaches have come to represent the dark side of cooperation in asylum matters. Far from merely reassigning and redistributing asylum procedures, they have resulted in an ever narrowing and precarious space for asylum seekers to make a refugee claim (Hyndman and Mountz 2008, p. 268; Frelick et al. 2016; Mann 2018).

The ‘safe third country’ provision was first introduced into Canada’s legal system in 1987, during a tumultuous time during which there were concerns over the arrival of ‘boatloads’ of South Asian refugees (The Canadian Encyclopedia 2017). Bill C-55.5 It eventually came into effect in 1989, resulting in the creation of the Immigration and Refugee Board (IRB) and introducing a policy allowing for the return of some refugee claimants who had traveled through another country.6

Negotiations on a safe-country agreement between the US and Canadian governments began in the early 1990s, against the backdrop of consistent criticism from the refugee advocacy community.7 The initiative, however, lacked political momentum, particularly on the American side, for fear of having to re-intake large numbers of asylum claimants (Bolter and Meissner 2018). The real catalyst for the STCA turned out to be the terrorist attacks on September 11, 2001, and their aftermath. US demands for closer intelligence and security collaboration at the border provided Canada with a better bargaining position to reopen the issue of concluding a safe-country agreement between the two countries.8

The STCA was signed in December 2002 and came into effect in December 2004. To date, the US is the only country that is designated as a safe third country by Canada.9

The stated purpose of this bilateral Agreement is to help ‘both governments better manage access to the refugee system in each country for people crossing the Canada–US land border’.10 It prohibits third-country nationals who first set foot in the US from making a refugee claim in Canada (and vice versa) at the official land ports of entry (POE). The Agreement does not apply to asylum seekers who make inland refugee claims from within Canada or the US.11 Under the Agreement, refugee claimants are required to request refugee protection in the first safe country they arrive in, unless they qualify for an exemption under the Agreement. Some of these exceptions correspond to commonly recognized ‘vulnerable groups’ within asylum management, such as unaccompanied minors. However, as subsequent sections of this article demonstrate, these exceptions are rather arbitrary.

Article 2 of the Agreement declares that it does not apply to US citizens or habitual residents of the US who are not citizens of any country (‘stateless persons’).

According to Article 4(2), a claimant will be admitted into the territory of the ‘receiving Party’ for the purposes of making a refugee claim if she meets one of the four following exceptions:

(a)

Has at least one family member in the territory of the receiving Party who has been accepted as a refugee or has lawful status, other than as a visitor;

(b)

Has at least one family member in the territory of the receiving Party who is over 18, and has an eligible refugee claim pending;

(c)

Is an unaccompanied minor, meaning that she is unmarried, under 18, and has no parent or legal guardian in either Canada or the US; or

(d)

Arrived in the territory of the receiving Party with a validly issued visa (other than a transit visa), or without a visa because none is required to enter the receiving Party.

In addition, Article 6 of the Agreement grants either party the discretion to consider a refugee claim ‘where it determines that it is in its public interest to do so’. For example, according to section 159 (6) of the Immigration and Refugee Protection Regulations SOR/2002-227 (IRPR), the ‘public interest exception’ applies to asylum seekers who are facing the death penalty in the US. An earlier comprehensive review of practice found that exceptions are used in an inconsistent manner, and that family exceptions are the most common, while public-interest exceptions seem to be the rarest (Bordering on Failure 2013, p. 91).

Even if they qualify for one of these exceptions, refugee claimants must still meet all other eligibility criteria of Canada’s immigration legislation. For example, if a person seeking refugee protection has been found inadmissible in Canada on the grounds of security, for violating human or international rights, or for serious criminality, that person will not be eligible to make a refugee claim (pursuant to IRPA 101(1)).

An additional safeguard is the rule that no refugee claimant who is subject to this Agreement may be removed to a country outside Canada or the US until the refugee claim has been determined by one of the parties (STCA Article 3). This provision is intended to preclude two related phenomena: chain refoulement,12 and the ’refugee in orbit’ problem.13

In the past two decades, numerous court cases have been brought by refugees and their advocates (Applicants) to challenge the validity and the constitutionality of the legislation implementing the STCA.14 The specific provisions in question are s. 101(1)(e) of the Immigration and Refugee Protection Act (IRPA) and s. 159.3 of the Immigration and Refugee Protection Regulations (IRPR or the Regulations), which designate the US to be a ‘safe third country’. The Applicants argued that the STCA violates section 7 of the Canadian Charter of Rights and Freedoms (which protects the life, liberty, and security of the person) by returning ‘ineligible’ refugee claimants to the United States where they face onerous filing deadlines, and abusive and punitive detention measures. They also argued a violation of section 15 of the Charter, claiming that the policy of returning refugee claimants disproportionately impacted women fleeing gender-based violence. In addition, the Applicants argue that the STCA violates Canada’s obligations under international law by designating the US as a safe country for refugees, which they consider unreasonable given the US’s ongoing institutional violations of the 1951 Convention Relating to the Status of Refugees (UN General Assembly 1951). According to the Applicants, the STCA also contravenes Canada’s international obligations under the United Nations Convention Against Torture and Other Cruel, Inhumane or Degrading Treatment or Punishment (CAT), as it exposes refugee claimants to unsafe conditions in the US and introduces the risk of their refoulement from the United States.

3. Vulnerability in the Migration Context—A Conceptual Excursus

Vulnerability is an elusive and contested theoretical concept with increasing yet ambiguous operational relevance in migration governance (Welfens and Bekyol 2021). The International Organization for Migration (IOM) defines vulnerability as ’the diminished capacity of an individual or group to have their rights respected, or to cope with, resist or recover from exploitation, or abuse’ (IOM 2017).

Generally, in immigration and refugee law, vulnerability is applied narrowly to various special sub-categories of migrants (i.e., women, minors, persons with disabilities, victims of torture and other forms of violence, etc.). Humanitarian assistance and asylum has become increasingly linked to such ‘inherent’ vulnerability criteria (Welfens and Bekyol 2021). In such cases, the concept of vulnerability is operationalized to provide substantive and/procedural accommodations for persons with distinct needs.

An alternate approach, in the context of migration, is to use vulnerability to describe the intrinsically precarious position of all people seeking protection, due to their legal status and potential lack of access to state protection (AIDA 2017). The European Court of Human Rights in M.S.S. v Belgium and Greece reiterated that the recognition of the inherent vulnerability and special protection needs of asylum seekers rests on broad international consensus referencing the Geneva Convention, the UNHCR and the European Reception Directive.15 Objective 7 of the UN’s Global Compact on Safe, Orderly, and Regular Migration aims to ’address and reduce vulnerabilities in migration’ (UNGA 2018, p. 5) requiring contracting states to review and eliminate policies and laws that exacerbate migrant vulnerability (Atak et al. 2018). These perspectives reiterate a focus on systemically generated vulnerability related to displacement, rather than vulnerability stemming from other shared characteristics, such as age or gender.

Critics of either approach to vulnerability point out that the legal ’vulnerabilization’ of applicants for international protection has its pitfalls (AIDA 2017, p 12). Considering all protection seekers as vulnerable can fuel negative stereotypes of asylum seekers’ lack of agency and dependency on the welfare state (Peroni et al. 2013). In addition, if we take refugee vulnerability for granted, this can lead to a rather high threshold of harm that needs to be met in order to merit state protection against further risks. This was most recently demonstrated by the Canadian Federal Court of Appeal’s (FCA) rather limited and generalized view on the issue of refugee ‘vulnerability’.16 In spite of Kreishan v Canada (Citizenship and Immigration), 2019 FCA 223, 438 D.L.R. (4th), which determined that ‘psychological suffering is inherent in the plight of refugees’ (at para. 100), the FCA in the STCA appeal case remained unconvinced that returning claimants to the United States ’actually increased psychological suffering above this inherent level’.17

There are also consequences to moving away from generalized understandings of vulnerability. State officials and advocates interpret ‘inherent vulnerabilities’ of certain sub-groups in arbitrary and narrow ways, which may exclude other vulnerable migrants. In addition, this approach obscures the root causes of people’s vulnerabilities, which are frequently generated and/or exacerbated by administrative structures and processes governing mobility and legal status (Kaga et al. 2021).

Canadian law and policy do not meaningfully engage with or define the concept of ‘vulnerability’. Similarly, most relevant cases include no more than a single reference to ‘vulnerability’ and suggest very little recognition of structural or ‘systemic vulnerability’, generated and exacerbated by the immigration and asylum system (Purkey 2022). This is problematic because misrecognition of vulnerability may lead to lack of access to special protections, procedural safeguards, and other life-changing opportunities for protection seekers. In addition, any reforms aimed at addressing and reducing migrants’ vulnerabilities require a clear understanding of what vulnerability means and how it is generated and experienced by various stakeholders.

In the following, this article analyzes the Safe Third Country Agreement between Canada and the US in this broader context of ‘vulnerability-generation’, to contribute to the evolving understanding of systemic and inherent vulnerability in asylum management.

4. Generating ‘Systemic Vulnerability’—The General Shortcomings of the STCA

There are several interconnected problems with the substance and functioning of the STCA that individually and together generate systemic vulnerabilities along the border. With its narrowly tailored exceptions, the STCA has increased incentives for irregular and unsafe entry into Canada, making it neither a just nor an efficient mechanism for asylum management.

4.1. Increasing Vulnerability by Driving People towards Unsafe Journeys

The troubles began directly after the STCA was implemented on 29 December 2004, in the middle of winter and the holiday season (Krauss and Pear 2004). Regarding the Agreement’s stated goal of enhancing refugee protection and promoting orderly handling of asylum claims, the STCA has been found to achieve just the opposite. The agreement’s most immediate effect was the reduction in the numbers of asylum seekers being able to claim protection at ports of entry on the U.S–Canada border by up to fifty percent.18 However, there is no conclusive or persuasive evidence that the Agreement has reduced the overall asylum-claim ‘burden’ on Canada. In fact, the overall number of asylum claims have either held or increased in the years since the STCA, save for a three-year period dip between 2013 and 2015.19 The Agreement, however, ‘succeeded’ in restricting safe and orderly access to refugee protection in a country that applies broader criteria, imposes a lower standard of proof and has more extensive procedural guarantees than the US (Bordering on Failure 2006, 2013). Since asylum seekers may only be turned away under the STCA if they enter at the official ports of entry, this has provided a set of perverse incentives for asylum seekers to enter Canada informally, avoiding detection until they are inland.

The summer of 2017 saw a 284% increase (of 3000 people) in irregular border crossings in Quebec alone from June to July ((Global News 2017); See also: (Immigration, Refugees and Citizenship Canada 2017)). This occurred directly after President Trump introduced new policies to crack down on undocumented immigrants, including reversing the Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA) program in June 2017 (Kopan 2017). The trend continued into the winter of 2017, with asylum seekers risking freezing conditions and injury; two Ghanaian men lost fingers to frostbite and another was nearly killed in Manitoba (Kassam 2017; Sanders 2017). In fact, close to 50,000 asylum seekers have entered Canada at the unofficial border crossing point between 2017 and 2019 (Ormiston 2019).

In practice, therefore, the STCA encourages asylum seekers to risk their physical health, utilize smuggling networks, and possibly fall victim to trafficking and other exploitation, in order to be able to file a claim from within Canada, rather than enter Canada at an official border crossing. As the following two sections demonstrate, given substantive differences between US and Canadian asylum systems, those seeking protection frequently had legitimate reasons to do so.

4.2. Weak Review Process—How Safe Is a Safe Country?

As mentioned before, there are inconsistencies and ambiguities about the criteria and concept of ‘safe countries’. The presumption of ‘safe country’ must remain rebuttable,20 which requires a sufficiently individualized process for each applicant, as well as adequate monitoring between states entering such arrangements.

At a minimum, no country can be considered safe if it practices ‘refoulement’, that is, the return of asylum-seekers to a country where they would be persecuted, in contravention of Art. 33 of the 1951 Convention Relating to the Status of Refugees (Refugee Convention).21 The UNHCR judges as inadequate those agreements that provide for readmission without ensuring sufficient procedural safeguards for the asylum seeker (UNHCR 2005). In addition, a country’s asylum system must provide so called ‘effective protection’.22 This concept, too, lacks a uniform definition, but the UNHCR has adopted the ‘Legomsky criteria’ to certify the basic conditions that must be met by a state deemed ‘safe’.23 UNHCR guidance points out critical factors intrinsic to effective protection, such as fair and efficient procedures, adequate and dignified means of subsistence, availability of durable solutions, and special care given to specific protection needs connected to vulnerabilities, including those arising from age and gender (UNHCR 2003b).

When compared to critical elements of ‘effective protection’ suggested by the UNHCR above, Canadian rules on the relevant factors in a ‘safe country’ review are relatively simple. The legislation requires that the review of a designated country be based on the following four factors:

• whether it is party to the 1951 Refugee Convention and the 1984 Convention Against Torture (The United Nations 1984);
• its policies and practices with respect to claims under the 1951 Refugee Convention, and its obligations under the 1984 Convention Against Torture;
• its human rights record; and
• whether it is party to an agreement with the Government of Canada for the purpose of sharing responsibility with respect to claims for refugee protection.

Putting in place credible mechanisms of ongoing monitoring of a potentially ‘safe’ receiving state’s compliance is another key benchmark for states to meet.24 Canadian practice, for example, requires a consideration of the human rights record of the potential receiving state.25 Section 102 of the IRPA requires the continual review of all countries designated as safe third countries. The purpose of the review process is to ensure that the conditions that led to its designation as a safe third country continue to be met. The parties also agree to cooperate with the United Nations High Commissioner for Refugees (UNHCR) in monitoring the implementation, and will also seek input from non-governmental organizations.

In its first-year review, however, the Canadian Council for Refugees found little indication that either government had monitored the Agreement’s impact on the rights of refuges.26 Critics have pointed out that what ‘monitoring’ means in practice is questionable. The burden of proof, regarding whether a state is ’safe’ for an individual refugee, is also unclear, and appellate decisions reflecting state practice regarding the burden of proof are rare.27 This criticism was later reinforced by the findings of the Federal Court in Canadian Council for Refugees v. R (2007).28 The Court found that the federal Cabinet failed to comply with its obligation under the law to ensure continuing review of the status of the US as a safe third country [2007 FC 1262 §338]. The government could not and would not refute or provide evidence of a systematic, continuing review of US policies and practices under the Refuge Convention and CAT [2007 FC 1262 §269]. Though the STCA does not provide specific instructions on what is to be done following a review, the court found it implied and reasonable to conclude that when evidence becomes available, the initial conclusion of compliance of the third country can no longer stand, and the Government should either suspend or terminate the STCA [2007 FC 1262 §272-273].

This guidance on the substance of the review process was short lived and overturned on appeal by the Federal Court of Appeal (FCA) in Canadian Council for Refugees v Canada, 2008 FCA 229. The FCA applied a formalist approach and found that it did not need to consider any evidence of US law and practice. On the administrative law grounds, the FCA held that a plain reading of the IRPA’s statutory requirement for the Cabinet to ‘consider’ a country’s conformity with the Refugee Convention and the CAT prior to designating it a ‘safe third country’, simply means an obligation to ‘consider’. As long as Cabinet ‘considered’ and was satisfied of US conformity with the treaties, the vires of the Agreement was unaffected by whether the US actually complied with them. In 2021, following yet another case against the STCA brought by refugee advocates in the previous year,29 the Federal Court of Appeal once again returned to the issue of ‘continuing reviews’ as it struck down the lower court’s finding of the STCA’s unconstitutionality. In Citizenship and Immigration v Canadian Council for Refugees, 2021 FCA 72, the Court of Appeal chastised the refugee advocates for not focusing their original efforts on the ‘safety valves’ built onto the STCA, in subsection 102(3) ‘continuing reviews’ and related administrative conduct and their effect.30 FCA declared that subsection 102(3) reviews and related administrative conduct would be the real causes of any infringement the government’s actions in enforcing the STCA may or may not represent.

The FCA’s vehement criticism and insistence on this line of reasoning is surprising. It disregards compelling reasons claimants had for not pursuing this line of advocacy in 2020 FC 770, namely the evidence presented in earlier litigation on the Government’s refusal to provide substantive evidence of a systematic review of US policies and its reliance on a right to confidentiality.31

In other words, both in practice and case law, hope for a substantive and transparent ongoing review process challenging the safe third country determination seems faint.

4.3. Generating Vulnerability through Returns—The Weaknesses of the US

Most of the STCA-related critique and legal challenges focus on the substantive and procedural deficiencies in the US asylum system and how the implementation of this Agreement will impose on Canada a share of indirect responsibility for the excesses, the harms and the rights violations inflicted in law and practice in the United States. To that end, this section will briefly address the most relevant general weaknesses of the US asylum system that negatively distinguish it from the Canadian system (including expedited removal, criminalization of migration, mandatory detention, and the one-year limit on filing asylum claims).

Since its introduction in 1996, the practice of ‘expedited removal’ of non-citizens lacking proper documentation has been expanded from ports of entry to cover the entire territory of the United States, with well-documented problems related to insufficient safeguards and lack of further hearing or review in its ‘inadmissibility’ and ‘credible fear’ determinations. It has since expanded the practice, making it enforceable throughout the US, despite evidence that its safeguards provide inconsistent protection against refoulement.32

In addition, by the mid-2000s, US immigration authorities had begun charging newly arrived asylum seekers with the criminal offence of entering the country with false documents. This practice is in direct contravention of Article 31.1 of the Refugee Convention, which states that:

’The Contracting States shall not impose penalties, on account of their illegal entry or presence, on refugees … provided they present themselves without delay to the authorities and show good cause for their illegal entry or presence’.

The categorization of asylum seekers as ’illegal’ or ’irregular’ migrants carries significant legal consequences. Whereas asylum seekers at the border are entitled to make a refugee claim and cannot easily be deported, undocumented migrants may be summarily excluded at the border and remain easily deportable while within the country. The political focus on ’illegality’ of entry led to public perceptions of refugees being ’frauds’ whose protection claims will more likely be rejected (Macklin 2005, p. 367). In fairness, at least on the discursive level, such misleading narratives have taken hold in Canada as well, especially after the increased number of irregular border crossings following the implementation of the STCA. Data shows that it is not the mode of entry, but rather the country of origin and other contextual and individual factors that constitute the determining factors in asylum claim adjudications.33

Access to legal aid makes a significant difference in the outcome of asylum determination processes (Macklin 2004). Whereas most refugee claimants in Canada are eligible for legal aid, there is no systematic provision for legal aid for asylum seekers in the US A one-year filing deadline for making an asylum claim on those already present in the US (with a few exceptions)34 has further complicated and denied protection to tens of thousands of claimants, most with otherwise legitimate claims.35 It also runs contrary to the UNHCR Executive Committee’s conclusions on not excluding claims based on failure to fulfill formal requirements (UNHCR 2009).

Experts have argued that it is not Canada’s place to police or dictate another country’s refugee determination system. On the other hand, as Macklin states, ’(t)aken on their own, each of the aforementioned features of the US system is worrisome. Taken together, they cast into serious doubt whether the US is able and willing to provide a ’full and fair’ refugee status determination procedure’ (Macklin 2004, p. 16). Canada’s basic responsibility here is that a safe third country agreement should not allow the potential sending country to enable practices by the proposed ‘safe country’ that would be prohibited in the sending country.

4.4. Canadian Courts’ Opinions on Constitutionality and Rights Violations

In Canadian Council for Refugees v R (2007), Justice Michael Phelan of the Federal Court found that it was unreasonable to conclude that the US complied with its non-refoulement obligations under the Refugee Convention and the CAT, and that the application of the safe third country rule violated refugees’ rights to life, liberty, and security (section 7) and to non-discrimination (section 15) under the Canadian Charter of Rights and Freedoms (’Charter’). As for the substantive differences between Canada’s and the United States’ approach to asylum filing delays, the principal distinction was that delay is never determinative of an asylum claim in Canada [2007 FC 1262 §155]. Given the evidence, the Court determined that the one-year filing deadline was incompatible with the CAT. [2007 FC 1262 §165]. The findings of this court remain relevant to the present analysis, despite the decision having been overturned by the FCA in 2008 FCA 229, because the FCA did not dispute or even consider the lower court’s factual findings on substantive Charter violations and conclusions of US noncompliance.

According to UNHCR guidance, international refugee and human rights law and standards require that the detention of protection-seekers should only exceptionally be detained for immigration-related reasons. In addition, detention must not be discriminatory, should include independent monitoring and procedural safeguards, conditions must be humane and dignified, and special circumstances and needs of asylum-seekers must be taken into account (UNHCR 2012).

In 2020, the Federal Court revisited the question of Charter violations and declared that US detention conditions often fall short in ensuring basic human dignity, medical care, and food. The term of detention may last for months without any review [2020 FC 770 §82]. Furthermore, the Court found that immigration detention routinely impedes access to legal counsel and increases the risk of refoulement, implicating a violation the security of the person. The Court also found that asylum claimants who were found ineligible for any of the exceptions under the STCA were immediately returned and handed over to US authorities by Canadian officials where they were automatically detained, resulting in a de facto form of punishment without charges or a trial—merely for making a refugee claim in Canada—in contravention of the Refugee Convention.

Justice McDonald cited Suresh v Canada (Minister of Citizenship and Immigration) (2002), to send an important message in the era of border and immigration control externalization: Canada ’does not avoid the guarantee of fundamental justice merely because the deprivation in question would be affected by someone else’s hand’ [2020 FC 770 §54]. Accordingly, the fact that STCA returnees are imprisoned by US authorities does not absolve the actions of Canadian officials. Evidence makes it clear that Canadian officials not only inform US officials that STCA claimants are being returned, but they are involved in the physical handover of ineligible claimants where they are immediately and automatically imprisoned by US authorities [2020 FC 770 §101–§103].

Again, victory for the advocates and asylum seekers was short lived. In overturning Justice McDonald’s decision, the Court of Appeal in 2021 FCA 72 seems to have largely disregarded the lived experiences and exceptional difficulties faced by refugee claimants returned by Canadian officials. The FCA diminished the relevance of detention experienced by returned asylum seekers in the US. It found the evidence of ten formerly detained individuals, as well as expert testimony showing that most refugee claimants are detained upon returning, insufficient to suggest anything more than the risk of ‘discretionary detention’ with sporadically available counsel and release [2021 FCA 72 §135–142]. This seems to pose a rather high threshold of harm, implying that so long as there are some exceptions to the norm of immigration detention, the application of the STCA is not in violation of section 7 of the Charter. Additionally, it suggests that if some applicants found remedy and stay of removal by Canadian officials, the ‘safety valves’ of the STCA cannot be found to be ‘illusory’ as suggested by the Federal Court [2021 FCA 72 §135–142].

The fate of the Agreement remains in question as refugee claimants and their advocates are poised to appeal the most recent decision at the Supreme Court (CTV News 2021).

5. Exacerbating ‘Inherent Vulnerabilities’ Based on Age and Gender

Beyond pushing asylum seekers towards dangerous journeys and disregarding substantive rights violations at the hands of US authorities, there are some other specific ways in which the STCA can be found to have exacerbated the inherent vulnerability of certain groups of asylum seekers already considered uniquely vulnerable.

5.1. Unaccompanied Minors

Compared to the US, Canada adopts more protective approaches to minors in immigration and asylum law and policy, pursuant to the best-interest principle defined in Article 3 of the United Nations Convention on the Rights of the Child (UN CRC 1990).36 US law contains no explicit references to the best interest of the child as a guiding principle in immigration law matters. In contrast, Canada’s Immigration and Refugee Protection Act identifies the best interest of the child as a relevant and primary consideration in several immigration matters, including detention, appeal, removal, and humanitarian and compassionate considerations.

Due to the lack of adult guardianship, unaccompanied minors37 represent a particularly vulnerable group of child migrants, deserving of special protections and assistance. Article 4(c) of the STCA provides an exemption and thereby a protection from returns for unaccompanied minors. The Agreement has come under criticism for excluding children who have a parent in either the US or Canada and for not providing any provisions for reuniting an unaccompanied child claimant already in Canada with a parent who is applying at the border, within the Article 4(2) exemption rules (Canadian Council for Refugees 2005, p. 13). This results in a limited and arbitrary application of the ‘best interests of the child’ principle in the case of separated children. It violates children’s and parents’ right to family life and family reunification under Article 9 and 10 of the UN CRC. To illustrate the stakes, Macklin brings up the example of a young refugee claimant awaiting a decision in Canada as an unaccompanied minor, while being unable to reunite with her mother who is seeking entry at the US–Canada border, because the only family member the mother has in Canada is a minor child, who cannot serve as a grounds for an exemption request under the STCA (Macklin 2004, pp. 8–9). Even when they are granted protection and permanent residence in Canada, unaccompanied minors face significant limitations on their ability to reunite with family members under IRPA’s sponsorship program until they reach the age of majority and can demonstrate sufficient financial capacity as a sponsor.38 Such narrowly tailored exemptions not only run against international obligations, but also weaken family reunification opportunities for unaccompanied minors in Canada, despite it being a guiding principle in Canadian immigration strategy (Luke 2007).

5.2. Female Asylum Seekers and Gender Based Violence

The experiences of female asylum seekers,39 especially in the context of gender-based violence, are also problematically divergent in the Canadian and US systems. In fact, repeated legal challenges to the STCA argued that it violates the non-discrimination rules of Section 15 of the Charter because the practice of returning asylum-seekers has a disproportionately negative impact on women. Women facing sexual violence in Central America have a particularly hard time proving they merit protection as a ’social group’ in the US, despite systematic, gender-based violence and inadequate state protections in their countries of origin. This restrictive attitude by US asylum officials is inconsistent with the Refugee Convention and leads to an increased risk of refoulement, according to the Applicants’ argument in Canadian Council for Refugees v Canada (Immigration, Refugees, and Citizenship), 2020 FC 770.40 US policies, such as expedited removal, mandatory detention41 (especially in the case of pregnant women) (Center for Reproductive Rights 2020), and the one-year filing deadline have been found to be particularly detrimental to women claimants who are potentially returned by Canada (Asthana 2011). Exceptions to the one-year filing-deadline rule do not take into account trauma and circumstances unique to women’s experiences that may hinder their ability to flee persecution and seek protection in a timely manner (National Immigrant Justice Center 2011).

Of particular concern is the fate of women fleeing domestic violence. Canadian law and case law has a consistent marginal advantage over the US in establishing the possibility for domestic violence victims to constitute a ‘particular social group within the meaning of the Convention’s definition of ’refugee’.42 According to the IRB’s Guideline on Gender-Related Persecution and Refugee Status (IRB 2003), women who flee private violence in circumstances where their home state is unable or unwilling to protect them may be recognized as refugees. The Guideline lists domestic violence as one of the circumstances giving rise to women’s fear of persecution that are unique to women, but it falls short of including gender as a ground of persecution, and also fails to include identity as a woman as constituting membership in a particular social group.43

Following Canada’s example, the US also developed gender guidelines as issued by the Immigration and Naturalization Service, though these are even less comprehensive and more tentative than the Canadian ones.44 In addition, they are not binding on all decision makers and are often ignored.45 Canadian case law has confirmed that the discrepancy between the two systems place women subject to domestic violence with gender-based asylum claims in the US at a real risk of refoulement.46 In Canadian Council for Refugees v R (2007), Justice Phelan noted that it is ’entirely foreseeable’ to Canadian officials that genuine claimants turned away under the STCA would be refouled, engaging sections 7 and 15 of the Charter.47 The same conclusion was reached in Canadian Council for Refugees v Canada (Immigration, Refugees, and Citizenship), 2020 FC 770. The Court found that detained asylum seekers face insurmountable systemic barriers in the US, especially in accessing legal representation, access to translators and assistance with legal forms and weaker asylum protections for gender-based violence. One of the Applicants, ‘ABC’, had already faced violence at the hand of the MS-13 gang, who threatened her and her daughters’ lives if she were to return to El Salvador [2020 FC 770 §105]. Justice McDonald concluded that had ABC been detained in the US, there would have been a real risk of refoulement [2020 FC 770 §105–§108]. Appellants provided compelling evidence that recent case law on gender-based asylum claims in the US reframed the non-state actor doctrine (relevant in domestic violence cases) in a manner that makes it more restrictive, imposing a disproportionate impact on women. [2020 FC 770 §151–152].

Unfortunately, Justice McDonald refused to address the section 15 violation in her decision, finding it sufficient to conclude that the STCA infringes on the right to life, liberty, and security of the person under section 7 of the Charter. This was arguably a missed opportunity to assert the unique vulnerability and special protection needs of people fleeing gender-based violence.

6. Conclusions

The number of asylum seekers sent back under the STCA in recent years constitutes a very small percentage of overall refugee claims in Canada (Harris 2020), indicating that it would have the resources to process claims at the border. Both the fear of return as well as the impact of return to a sub-standard asylum system carry dire consequences for the life of a refugee claimant. This article highlighted that in practice the ’safe country’ concept and related agreements represent the dark side of otherwise necessary inter-state cooperation in the field of migration management. The STCA also contributes to the generation of systemic vulnerabilities and the exacerbation of inherent vulnerabilities experienced by protection seekers at the US-Canada border. The result is the de facto and de jure erasure of many asylum seekers with credible claims and the obstruction of their opportunities to reach safety. The court decisions examined in this article, however, demonstrate that despite repeated reversals, the debates and legal challenges surrounding the legality of the STCA will continue in the future. In the meantime, they remain useful sources to illustrate the vulnerability-generating potentials of safe country agreements.