“Safe” Third Country Agreements and Judicial Review in the United States and Canada

The subject of safe third country agreements, or as the U.S. government has begun calling them “Asylum Cooperation Agreements”, has been in the news lately in both the United States and Canada.  The U.S. and Canada have had such an agreement with one another since 2002, implemented pursuant to section 208(a)(2)(A) of the Immigration and Nationality Act (INA), 8 U.S.C. 1158(a)(2)(A), and section 208.30(e)(6) of Title 8 of the Code of Federal Regulations (CFR) in the U.S., and pursuant to section 101(1)(e) of the Immigration and Refugee Protection Act (IRPA) and sections 159.1-159.7 of the Immigration and Refugee Protection Regulations (IRPR) in Canada.

The idea of a safe third country agreement (STCA) as it has traditionally existed, as between the U.S. and Canada, is that the two countries agree that many applicants for asylum or refugee status, who have first come to one of the two countries should apply in that country where they first arrived, rather than going to the other country to apply.  If both countries have “generous systems of refugee protection,” as the preamble to the U.S.-Canada STCA puts it, then there may arguably be efficiency in relegating people to apply for asylum in the first of the two countries that they reach.

The U.S.-Canada STCA covers only people who make a claim at a land border port of entry or are being removed from one of the two countries through the territory of the other.  It does not apply to citizens of either country or stateless persons last habitually residing in either country.  It also has other exceptions for applicants who have certain family members with lawful status in the country where they wish to make a claim, or whose family members already have ongoing asylum/refugee claims in that country, or who are unaccompanied minors, or who either have a valid visa for the country where they wish to apply or did not need a visa for that country but did for the other.  In addition, each country may exempt additional applicants whom it determines to process itself on the basis that it is in its public interest to do so. These exceptions are laid out at Articles 2, 4, and 6 of the agreement itself, and are also detailed at 8 C.F.R.§ 208.30(e)(6) and at IRPR sections 159.2 and 159.4-159.6.

As many refugee claimants have come to have less faith in the U.S. asylum system than the Canadian refugee system, and due to the restriction of the U.S.-Canada STCA to entrants at land ports of entry (or instances of removal through one country by the other), an increased number of refugee claimants have entered Canada at unauthorized crossing points outside a port of entry in order to make a claim, most notoriously at Roxham Road along the New York-Quebec border.  (The idea is not to evade immigration officers, but simply to avoid the application of the STCA; news articles describe an oft-repeated formal warning to the applicants that entry at that specific place will result in arrest, which does not generally dissuade people since being arrested after entry into Canada, and making a refugee claim, is precisely their goal.)  There has been discussion of modifying, suspending, or terminating the agreement, and one Conservative Member of Parliament suggested that the entire U.S.-Canada border be designated as a port of entry, although in practice that would have very peculiar consequences for the immigration system as a whole (since it would mean applicants for admission in general could show up at any point along the border to be processed).  There has also been a legal challenge, discussed further below, to the notion that the U.S. can currently qualify as a safe third country consistent with Canadian constitutional law and international obligations.  The Canadian government also recently, as part of a budget bill, added to IRPA a provision separate from the STCA but also evidently designed to discourage claimants from the United States, section 101(1)(c.1), which bars refugee claims by those who have previously claimed in the United States or another country with which Canada has an information-sharing agreement—a provision I have criticized in a prior blog post.

The U.S., meanwhile, has been entering and seeking to enter into “Asylum Cooperation Agreements” with various Central American countries, Guatemala being the first followed by Honduras and El Salvador.  An interim final rule was published last week to implement such agreements, and removal from the United States under the Guatemala agreement is said to have begun just a few days ago.

The signing and implementation of these “Asylum Cooperation Agreements” has attracted a great deal of criticism, because describing Guatemala, Honduras and El Salvador as safe third countries, or safe countries in any sense, would be highly dubious, to put it mildly.  A piece in Foreign Policy described the Agreement with Guatemala as “a lie”, given Guatemala’s high level of crime – the U.S. State Department’s Overseas Security Advisory Council (OSAC) having written of the country as “among the most dangerous countries in the world” with “an alarming high murder rate” – and lack of resources to process asylum cases – the country having apparently received only 257 asylum claims in 2018 and adjudicated only 17.  Indeed, even the U.S. asylum officers implementing the plan to remove asylum-seekers to Guatemala under the agreement have reportedly been given “materials . . . detailing the dangers faced by those in the country, including gangs, violence, and killings with “high levels of impunity.””  Honduras, where the U.S. Government is said to intend to implement a similar agreement by January, is also problematic to describe as a “safe” country, with OSAC relaying a Department of State Travel Advisory “indicating travelers should reconsider travel to the country due to crime.”  Similarly El Salvador, with which the U.S. has also signed an agreement despite its having one of the highest homicide rates in the world.  And given the radically underdeveloped asylum systems in Guatemala and likely the other countries as well, the requirement under INA 208(a)(2)(A) that “the alien would have access to a full and fair procedure for determining a claim to asylum or equivalent temporary protection” if removed seems unlike to be met.  The point of threatening removal to these countries seems to be more to discourage asylum claims entirely.

Moreover, the screening process to determine whether asylum-seekers can be exempted from removal to Guatemala, based on a fear of persecution there, is already being implemented in a way that has been described as “a sham process, designed to generate removals at any cost.”  Under the interim final rule, those subject to potential removal under an “Asylum Cooperation Agreement” will not be allowed to consult with attorneys or others during the screening process, or present evidence.  As the interim final rule puts it at new 8 C.F.R. § 208.30(e)(7), “In conducting this threshold screening interview, the asylum officer shall apply all relevant interview procedures outlined in paragraph (d) of this section, except that paragraphs (d)(2) and (4) of this section shall not apply to aliens described in this paragraph (e)(7)”—8 C.F.R. § 208.30(d)(4) being the provision that gives the right in ordinary credible-fear proceedings to “consult with a person or persons of the alien’s choosing prior to the interview or any review thereof, and . . . present other evidence, if available.” This is different than the U.S.-Canada STCA, which specifically provides in its Statement of Principles that “Provided no undue delay results and it does not unduly interfere with the process, each Party will provide an opportunity for the applicant to have a person of his or her own choosing present at appropriate points during proceedings related to the Agreement.”  Unless vulnerable people with no legal representation, no opportunity to consult with an attorney or anyone else, and no opportunity to present evidence can prove to an asylum officer that they are more likely than not to be persecuted on a protected ground or tortured in Guatemala, they will be given the option of being returned to their home country of feared persecution or being removed to Guatemala.  One might reasonably describe this as outrageous, and a question that naturally comes to mind is whether it would survive review by a court.

The question of judicial review of the propriety and application of safe third country agreements and “Asylum Cooperation Agreements” has indeed arisen in both the United States and Canada, with different initial indications regarding the result.  In the U.S., the suggestion has been made by at least one commentator that the recent U.S. decision to send certain asylum applicants to Guatemala pursuant to an agreement may not be judicially reviewable (or at least “any lawsuits challenging the new rule will face significant obstacles”.)  In Canada, on the other hand, a challenge to the Canada-United States STCA, asserting that given current conditions in the U.S. the STCA violates the Canadian Charter of Rights and Freedoms, was argued before the Federal Court of Canada earlier in November and is awaiting decision.  A previous challenge to the Safe Third Country Agreement and the regulations implementing it had some success before being rejected by the Federal Court of Appeal of Canada in part on the basis that the organizations which had brought the challenge did not have standing to do so in the abstract; the current challenge includes a family of rejected refugee applicants seeking judicial review.

The basis for the potential lack of judicial review in the United States regarding safe third country agreements and their implementation is section 208(a)(3) of the INA, 8 U.S.C. § 1182(a)(3), which provides that “No court shall have jurisdiction to review a determination of the Attorney General under paragraph (2).”  (Recall that safe third country agreements, as a bar to asylum, are authorized by section 208(a)(2)(A) of the INA, which is part of the referenced paragraph 2.)  There are potential exceptions to this rule, such as the exception at 8 U.S.C. § 1252(a)(2)(D) for review of constitutional claims or questions of law on petition for review of a removal order, and the provision at 8 U.S.C. § 1252(e)(3) for review of written policies regarding expedited removal procedures within 60 days of their implementation, as in Grace v. Barr (formerly known as Grace v. Sessions and Grace v. Whitaker).  It may well be that the deeply problematic agreements with Guatemala, Honduras, and El Salvador, and the details of their implementation, will be subject to judicial review under one of these exceptions or otherwise.  There is certainly at least some basis, however, for the existing conventional wisdom that judicial review will be quite difficult.

Under Canadian law, on the other hand, judicial review of administrative action cannot be precluded in this way.  As I explained in my above-mentioned prior blog post, even aspects of the refugee determination as to which administrative decision-makers are given deference by the courts will be reviewed for reasonableness, because as explained by the Supreme Court of Canada in Dunsmuir v. New Brunswick, 2008 SCC 9, “The rule of law requires that the constitutional role of superior courts be preserved and, as indicated above, neither Parliament nor any legislature can completely remove the courts’ power to review the actions and decisions of administrative bodies.  This power is constitutionally protected.”

Admittedly, even under Canadian law judicial review can be procedurally limited by legislation, although not eliminated.  Under section 72(1) of IRPA, for example, seeking review in the Federal Court of Canada of any decision under IRPA generally requires “making an application for leave to the Court.”  Moreover, under section 72(2)(d) of IRPA, the leave application is adjudicated by a single judge of the Federal Court “without delay and in a summary way and, unless a judge of the Court directs otherwise, without personal appearance”, and under section 72(2)(e), there is no appeal with regard to an application for leave.  Even if leave is granted, the single level of judicial review at the Federal Court may be all there is: under section 74(d) of IRPA, an appeal to the Federal Court of Appeal of Canada in an immigration matter is generally possible “only if, in rendering judgment, the judge certifies that a serious question of general importance is involved and states the question.”  (Under certain very limited circumstances, an appeal can be taken to the Federal Court of Appeal even absent a certified question, but the threshold for that is quite high, as clarified recently in the citizenship context by Canada (Citizenship and Immigration) v. Fisher-Tennant, where the Court of Appeal quashed a government appeal against what was at most an ordinary error by the Federal Court in finding Andrew James Fisher-Tennant to be a citizen of Canada.)  So there is not an unlimited amount of judicial review, but there is necessarily some.

To the extent that the existing exceptions under the INA do not prove adequate to allow for full judicial review of decisions under safe third country agreements or “Asylum Cooperation Agreements”, Congress should give serious consideration to revising INA § 208(a)(3) to provide for judicial review of such decisions to exist in the United States, as it exists in Canada.  It was one thing, although still deeply problematic from a rule-of-law perspective, when the statute in practice only attempted to guard from judicial review decisions to return asylum-seekers to a plausibly safe country such as Canada.  If agreements are to be made with other, much more dangerous countries such as Honduras, Guatemala, and El Salvador, however, then judicial review of these agreements and how they are applied in practice becomes significantly more urgent.  Given the conditions in these countries, the question of whether people can be sent to such countries under safe-third-country agreements without any judicial review could literally be a life-or-death issue.