Expulsions and Article 31 of the Convention Relating to the Status of Refugees, 1951

Many complex issues surrounding the international regime of refugee protection were considered by the UK Court of Appeal (CA) in its judgment in R (AAA) v SSHD, which was handed down on 29 June 2023. Among these was the question of the compatibility with article 31 of the Convention Relating to the Status of Refugees, 1951 (Refugee Convention) of decisions taken by the competent authorities of a state to remove a refugee suspected of illegal entry to another state, which the expelling state deems to be ‘safe’ for refugees, without the expelling state first making a substantive assessment of the merits of the claim for refugee status. This is a question that reflects well-documented concerns over how the rights of refugees and other victims of human rights abuses are being compromised by measures taken by states to curb illegal migration.

The judgment of the majority of the CA was exemplary in several respects - not least in its finding that - primarily due to the deficiencies in its asylum assessment regime - Rwanda is not a safe place to which the UK could send refugees in the expectation that their claims to refugee status would be properly assessed there. However, the same cannot be said of the CA’s treatment of the question of whether article 31 (which prohibits the the imposition of penalties on refugees who arrive illegally on the territories of states) is breached when states remove refugees back to countries through which they have travelled en route, or, as in the case of the UK Rwanda scheme, when such refugees are removed to a third state with which they have no prior connection. On this point, the judgment is deeply unsatisfactory - both in terms of the space devoted to the question and the conclusions reached on it.

The essence of the CA’s position on the relation between expulsion and article 31 is summarised by Underhill LJ between paragraphs 304-321 (pp. 94-102). Having reviewed key judicial and academic authorities, Underhill LJ concludes that article 31’s conception of a penalty very likely excludes expulsions entirely (see para. 318, p. 101) and would certainly not extend to expulsions to states that are deemed to be safe. For these purposes, a safe third state is one in relation to which there are insufficient grounds for supposing that it would expose a refugee - directly or indirectly - to inhuman or degrading treatment or punishment or other serious human rights violations. In short, according to Underhill LJ, if an expulsion is ever capable of being construed as a penalty, it would only be so if the consequences of the expulsion were that the refugee would be exposed to harm of a type that is prohibited under article 33 of the Refugee Convention, which establishes the principle of non-refoulement. Specifically at para. 329 (p. 104), Underhill LJ states:

[I]t is in my view inconsistent with the well-recognised scheme of the Convention that the expulsion of a migrant to a safe third country should be treated as a penalty within the meaning of article 31 (1), whatever the reasons for taking that course may be and however unwelcome it may be to the migrant in question.

The text of article 31 reads as follows:

1. The Contracting States shall not impose penalties, on account of their illegal entry or presence, on refugees who, coming directly from a territory where their life or freedom was threatened in the sense of Article 1, enter or are present in their territory without authorization, provided they present themselves without delay to the authorities and show good cause for their illegal entry or presence.

2. The Contracting States shall not apply to the movements of such refugees restrictions other than those which are necessary and such restrictions shall only be applied until their status in the country is regularized or they obtain admission into another country. The Contracting States shall allow such refugees a reasonable period and all the necessary facilities to obtain admission into another country.

A regrettable feature of the immigration policies of many states, including the UK, is that indirect entry to the country in which the asylum claim is first registered in and of itself constitutes illegal conduct. Consequently, refugees who have not arrived ‘directly from a territory where their life or freedom was threatened’ could be said to fall outside the ambit of article 31 - although even this is a disputed question in academic, policy and practice circles. Questions over the precise scope of that particular element of article 31 aside, there are clearly other forms of illegality that, in today’s culture of draconian immigration controls, would render refugees liable to expulsion to a third state. Their return would be facilitated by agreements between the third state and the first destination state - the UK Rwanda scheme being a flawed example of such agreements. It follows that even refugees who have unequivocally arrived ‘directly’ in the place at which they first seek asylum may be liable to being sent on to a third state. Thus, the question of whether expulsions are compatible with the terms of article 31 is a very live one.

It is clearly consistent with both the letter of article 31, and with the spirit of the Convention as a whole, to construe a penalty for the purposes of article 31 as any imposition by a state that aims to punish or deter illegal entry. In most instances the imposition will fall short of action that directly or indirectly results in the illegal entrant suffering inhuman or degrading treatment or other serious human rights violation. But the fact that an imposition does not result in such injurious treatment does not make it any less of a penalty and, therefore, prima facie a breach of article 31. Contrary to Underhill LJ’s position, the operative article 31 question is not whether the illegal entrant who is subject to the imposition is safe from serious harm but whether the imposition in question is intended solely or primarily to deter illegal entry and, if so, whether the individual being expelled meets the definition of refugee under article 1. A (2).

It is ironic that a judgment that takes the Divisional Court (DC) gently to task for deploying the wrong framework of analysis when considering whether there was a risk that refugees would be exposed either to serious ill-treatment in Rwanda, or to the risk of being returned from Rwanda to the states from which they sought asylum (see paras. 128-129, pp. 43-44), so evidently overstates the relation between article 31 and article 33 of the Refugee Convention whilst at the same time understating the relation between article 31 and article 1.A (2). The prohibition against expelling a refugee to an unsafe state is undoubtedly the core obligation of state signatories. Indeed, it is now accepted by the international community that article 33 has matured into a rule of customary international law - binding even states that are not signatories to the Refugee Convention. However, to accept that article 33 is the preeminent clause within the Refugee Convention is not to say that other articles, such as article 31, must be interpreted solely in light of it.

The judicial and academic authorities that Underhill LJ deploys simply attest to the fact that expulsion per se does not amount to unlawful refoulement in breach of article 33 of the Refugee Convention. This is because expulsion would very often be to a third state that, unlike Rwanda, is rightly considered safe for refugees. However, to conclude that a policy of routine expulsions of illegal entrants, including refugees, is not, without more, incompatible with article 33 is not the same as saying that expulsions of that kind are incapable of being construed as penalties, contrary to article 31.

It follows from the above that neither the CA nor the DC - whose reasoning on the question of where expulsions sit within the terms of article 31 are for all intents and purposes identical (see para. 330, pp. 104-105) - gave proper consideration to the submissions made on behalf of the claimants to the effect that the motive behind the uses to which expulsion decisions are being made by states, like the UK, are germane to the question of whether the practice contravenes article 31 (see paras. 315 & 324, pp. 99 & 103).

Today, expulsion is routinely used as a means of general immigration control. This was not the practice at the time that cases like Bugdaycay [1987] were decided, and, therefore, it is doubtful whether they can bear the weight that Underhill LJ and other members of the senior judiciary continue to place on them. It is also important to push back against the idea that refugees should be allowed to bear any kind of penalty - provided that if falls short of arbitrary incarceration, death, torture or inhuman and degrading treatment