By Vadita Agarwal
A report released in February of this year by the US Senate Committee on Foreign Relations revealed that in 2026 alone, the Trump administration had spent an estimated $40 million on deportation of migrants to third countries, often those with a consistent record of human rights abuses (pg. 6). With Sierra Leone being the latest State to accept deportees, US migration policy is but one example of a larger global shift towards externalisation of the movement of migrants and asylum-seekers. Broadly, externalisation refers to the process of shifting border control functions that are normally undertaken by a State within its own territory so that they take place, in part or in whole, outside its territory. While externalisation is complex and varied, its underlying objective is often to prevent migrants or refugees from entering a jurisdiction, allowing the State to sidestep human rights responsibilities. In an agreement involving two States, the relevant entities here are ones promulgating the externalisation policy (externalising State) and the one facilitating the externalisation (receiving State).
Because externalisation involves delegating State functions outside the territory of the State, extraterritoriality of human rights obligations is an important and frequently discussed concept. Extraterritoriality discusses whether a State owes obligations under human rights treaties to an individual located outside its territory. While the existing literature on this subject generally focuses on the extraterritorial obligations of the externalising State, the lacuna that this piece aims to explore is the obligations of the receiving State.
The aim of this piece is not to minimise the obligations of the externalising State but to acknowledge that implementation of its policies is only being carried out by willing receiving States. Framing receiving States in this context as neutral third parties is dangerous because it gives them impunity to hide behind the equally egregious but more publicly discussed violations of the externalising State.
Contextualising Externalisation and its Rationale
Externalisation of border control can take many forms and any discussion on externalisation must accommodate its unclear definitional boundaries. While externalisation is most often discussed in the context of States avoiding human rights obligations, scholars also acknowledge the possibility of a ‘rights-based’ externalisation which intends to protect refugee rights.
The United Nations High Commissioner for Refugees (UNHCR) distinguishes between externalisation of international protection and “lawful practices involving transfer of the responsibility for international protection.”
Among the latter, a relatively regulated method of externalisation is the ‘safe third country’ (STC) method which authorises refugees to be sent to designated ‘safe States’ (pg. 295). While STC rules are not unlawful under international law per se, the receiving State in an STC arrangement needs to, at minimum, provide the refugee with all the rights they have already obtained while under the jurisdiction of a State party to the 1951 Refugee Convention, particularly by not exposing the refugee to the risk of refoulement (pg.333).
Externalisation of international protection, which is the focus of this piece, comprises measures taken by a State, usually in the form of externalisation agreements with other States, aiming to transfer asylum-seekers and refugees from one State to another, often in the absence of adequate safeguards and resulting in their inability to reach their desired ‘destination’ and claim protection there.
There are various ways in which border control procedures can be externalised, each involving the delegation of a Specific state function undertaken with respect to migrants or asylum-seekers. A common tactic of externalisation is pushbacks; described by the United Nations Human Rights Council as various measures taken by States, either unilaterally or in collaboration with third States which result in migrants, including asylum-seekers, being forced away from the border of the State without an individual assessment of their status or which territory they are legally entitled to enter. The way externalisation manifests here is that States delegate pushbacks to third States to prevent upstream movement of people to their own territories. In practice, despite the dubiousness of their legality, pushbacks are ubiquitous. Notably, pushbacks have been carried out in Europe, along the Balkan route and by the U.S along the U.S.-Mexico border. The practice of pushbacks often exacerbates the vulnerability of migrants and asylum-seekers by denying human-rights based and child-sensitive treatment as well as denying entry on the basis of their gender, religion, nationality, etc. (para. 49).
States can also opt to externalise their asylum systems. States may transfer asylum-seekers who have arrived in their jurisdiction to another State for the purpose of determining their refugee status and, in some cases, providing them with territorial asylum there. David Cantor et al explain that third country processing comprises externalisation of post-arrival asylum obligations and is thus distinct from more protective forms of externalised procedures, including pre-arrival processing of an asylum claim by the authorities of a destination State in another State (pg. 141).
An important part of understanding externalisation of border control is understanding the incentive that a State has to engage in it.
Gammeltoft-Hansen explains that, apart from the basic right of non-refoulement which kicks in as soon as the asylum-seeker arrives at the border of a specific State, the rights contained in the 1951 Convention follow a “incremental structure” wherein the catalogue of human rights granted to the refugee expand with their attachment to the host state (pg. 101). Thus, refugees who display a higher level of attachment to the host State through, for example, habitual residence or being admitted in the host State’s asylum procedure, are entitled to broader rights under the 1951 Convention. The fundamentally territorial nature of refugee law which increases the obligations of a contracting State based on the level of attachment that the refugee has in its territory makes externalisation an attractive route.
Despite their consistent association with rampant human rights violations, I argue that externalisation policies are often deceptively framed as “either or both a security imperative and a life-saving humanitarian endeavour rather than simply a strategy of migration containment and control.”
Analysing the Role of the Receiving State
The role of the receiving State in an externalisation agreement is not homogeneous.
Liberalising entry into the externalising State’s borders and the accompanying monetary benefits is often a strong incentive for receiving States to get into externalisation agreements. For instance, in the 1980s, the US persuaded Mexican authorities to limit Central American migration through Mexico into the US, in exchange for liberalising US policy on entry of Mexican migrants into the country. It is argued that limiting this migration was, for Mexico, “a small price to pay in order to achieve their larger goal of ensuring that Mexican workers could continue to enter the U.S.” As Mexico’s economy became largely dependent on emigration to the US, Mexico effectively prioritised US’s immigration needs over its own.
The willingness of the receiving State to use resources of existing (often already overburdened) asylum and allied facilities towards housing refugees and migrants either voluntarily headed to or coercively displaced to another destination State is another significant variable in understanding the role of the receiving State.
For instance, the now cancelled UK Rwanda Migration and Economic Development Partnership (the Plan) foresaw “relocation to Rwanda of individuals who arrived in the UK through an illegal and dangerous route…” In return, as noted by a public spending watchdog, Rwanda would receive an initial £370 million and then further amounts based on the number of people relocated. The Plan also involved the UK “resettling a portion of Rwanda’s most vulnerable refugees in the UK.” In addition to short-term financial gains, vaguely worded promises of relocation make the Plan attractive to Rwanda which is already struggling with overcrowding due to an influx of refugees.
Despite seeming reciprocal, the impact of this agreement is far from equitable. The UNHCR noted that the transfer of an undetermined number of refugees was likely to add additional pressure on Rwanda’s “nascent and already overstretched” asylum system. In reaching an agreement with States like the UK which possess disproportionately more capital than itself, States from the Global South thus run the risk of overburdening their asylum systems even further.
While I argue that receiving States must also share the burden of human rights violations in exploitative externalisation agreements, there is undoubtedly some merit to the argument that externalisation facilitated by the Global North needs to be understood in the context of being negotiated between States with asymmetrical bargaining power.
Moreno Lax describes these exploitative externalisation measures as a means to ensure “irresponsibilisation”, wherein an externalising State can exploit geographical distance to “create and legitimate ethical and legal detachment from its own wrongdoing.”
However, while acknowledging the inherent power dynamic between States in reaching externalisation agreements is imperative, unbiased inquiries into this subject must not fall into the trap of infantilising States from the Global South and stripping them of their agency. In its arbitration against the UK before the Permanent Court of Arbitration, Rwanda submitted that the Plan was a major milestone in securing the objectives of the Rwanda-UNHCR Joint Strategy which focused on moving away from ‘warehousing’ of refugees to integrating them in the economy and making them self-reliant.
Unlawful externalisation policies, albeit dominated by States in the Global North, are no longer their exclusive domain. As far back as 2015, Thailand, Malaysia and Indonesia’s rampant pushbacks at sea of boats containing Bangladeshi and Rohingya migrants and asylum-seekers raised human rights concerns.
Thus, any discussion on the obligations of receiving States needs to be cognizant of the complex and varied circumstances surrounding each externalisation agreement.
Extraterritoriality vis a vis the Receiving State-Revisiting Case Law
Because the crux of externalisation policies is subjecting migrants to control outside the territory of the state, the question of how specialised regimes of international law apply outside this territorial context arises. Extraterritoriality then becomes a logical starting point because it extends the jurisdiction; and subsequently, the human rights obligations of States to acts beyond those carried out on its territory.
The scope and tests for applying extraterritoriality have evolved significantly over time, specifically through an expansion of the spatial and individual control criteria to meet contemporary scenarios.
One example is the 2021 case of A.S., D.I., O.I. and G.D v. Italy (A.S. v. Italy) heard by the Human Rights Committee (HRC). The complainants in the case were relatives of the asylum-seekers who died in 2013 on board a vessel which was shipwrecked in the Mediterranean Sea, between Italy and Malta (A.S. v. Italy, para. 1.1). A few hours after setting off, the vessel had been shot at by another boat, causing large amounts of water to enter it, prompting the passengers onboard to call the Italian number for emergencies on sea to report the boats’ imminent sinking (A.S. v. Malta, para. 2.1). The passengers were informed by the Italian officials that they were in the Maltese Search and Rescue (SAR) Zone and redirected to the Maltese rescue center. Despite multiple calls to both Maltese and Italian authorities, no rescue arrived until much later when the vessel had already capsized (A.S. v. Malta, para. 2.3). Alleging that Italy and Malta had failed to take appropriate measures to render assistance to the victims of the shipwreck, the authors argued violation of inter alia, Articles 6 and 2(3) of the International Covenant on Civil and Political Rights (ICCPR).
While the case against Malta was held to be inadmissible due to failure to exhaust local remedies (A.S. v. Malta, para. 6.9), the observations of the HRC with respect to Malta’s extraterritorial obligations are noteworthy. The HRC recalled that the vessel was under distress and subsequently shipwrecked in the Maltese SAR area, a place where Malta exercised effective control, “potentially resulting in a direct and reasonably foreseeable causal relationship between the States parties’ acts and omissions and the outcome of the operation” (A.S. v. Malta, para. 6.7). Against Italy, the HRC held that in light of the “particular circumstances of the case”, “a special relationship of dependency had been established between the individuals on the vessel in distress and Italy” (A.S. v. Italy, para. 7.8). Resultantly, the HRC considered that this relationship of dependency made the deceased subject to Italy’s jurisdiction, “notwithstanding the fact that they were within the Maltese search and rescue region and thus concurrently subject to the jurisdiction of Malta” (A.S. v. Italy, para 7.8). The concurring individual opinions of Gentian Zyberi and José Manuel Santos Pais both recognised that this was a case of “shared responsibility” (A.S. v. Italy, pg. 15, 16). Despite this clear holding, the apportionment of responsibility between Italy and Malta could not be detailed due to the complaint being inadmissible against Malta.
As correctly cited in A.S. v. Italy, it was unequivocally clarified in the HRC’s 2019 General Comment No. 36 on the Right to Life that the jurisdiction requirement as seen in Article 2(1) ICCPR, when it comes to protection of life under Article 6, extends to “persons located outside any territory effectively controlled by the State whose right to life is nonetheless affected by its military or other activities in a direct and reasonably foreseeable manner.”
General Comment No. 36, coupled with case law such as A.S. v. Italy and Malta make some progress when it comes to discerning the obligations of the externalising State (which might claim lack of exterritoriality due to absence of effective spatial or individual control). However, it does little to ascertain where this leaves the receiving State.
It effectively creates a scenario where both Italy and Malta’s actions can foreseeably impact an individual’s right to life, giving rise to extraterritorial obligations for both States. In externalisation agreements, while the externalising State sets the broad aims and goals of the agreement, their implementation is dependent on the receiving State. In the event where under the direction of the externalising State, an act or omission by the receiving State violates human rights obligations, strict delineation of each State’s obligations is tricky. Then, the shared responsibility framework of A.S. v. Italy is a helpful starting point. The question then becomes, how can international law govern division of obligations between the two States?
Relevant Factors in Assessing Receiving State Obligations – The Road Ahead
The opacity of externalisation agreements is the clearest roadblock in determining the discretion that a receiving State has in the implementation of the terms of the agreement and thus on the apportionment of obligations. Any third country processing agreement must be publicly available and clearly delineate the role of the various States involved in the arrangement. Closed door, vaguely worded agreements between States not only affect the rights of potential litigants who might suffer human rights abuses in lieu of the externalisation but also the ability of the States involved to withstand judicial scrutiny.
While moving away from a rigid spatial or individual effective control standard is generally helpful when it comes to ascertaining extraterritorial obligations of the externalising State, it complicates the scenario for the receiving State. This is in light of the inherent imbalance of negotiating powers wherein the receiving State is likely to accept any terms put forward by the externalising State and the lack of transparency surrounding externalisation agreements. This makes it exceedingly difficult to practically apply the standard of ‘relationship of dependency’ as seen in A.S. v. Italy and Malta.
Issues of delegation of duties between States have been historically problematic in the context of extraterritoriality. A good example is the 2014 European Court of Human Rights (ECtHR) case of Jaloud v. Netherlands which concerned the fatal shooting of Mr Azhar Sabah Jaloud at a checkpoint in Iraq by Dutch troops. On the question of its extraterritorial obligations, the primary argument made by Netherlands concerned how in Iraq, pursuant to UN Security Council Resolution 1483, US and UK were the only “occupying powers” (Jaloud, para. 113). Additionally, Netherlands also argued that its “contingent had at all times been under the operational control of the commander of MND (SE), an officer from the United Kingdom” (Jaloud, para 115). Holding Netherlands responsible for violation of its extraterritorial obligations, the ECtHR held that “the fact of executing a decision or an order given by an authority of a foreign State is not in itself sufficient to relieve a Contracting State of the obligations which it has taken upon itself under the Convention” (Jaloud, para. 143). Thus, a State is not divested of jurisdiction “solely by dint of having accepted operational control” of a commander-in charge. While the complex facts of Jaloud cannot be directly transplanted in an externalisation context; the Court’s observations on execution of orders by a foreign state shed some light on where the rationale behind imposing extraterritorial obligations on the receiving State may stem from. For instance, it is not unusual to expect States like the US or UK to frame their respective externalisation agreements in a deliberately vague manner to sidestep legal obligations. In light of this lack of clarity, can a specific act by the receiving State which it is only broadly authorised by the externalising State to carry out, impose extraterritorial obligations on the Receiving State?
Because there is no clear case law on extraterritoriality which answers this, the status of the Receiving State is currently undecided. While a precise allocation of responsibility is naturally determined on a case-to-case basis, a complete lack of guidelines on the obligations of the receiving State vis a vis the externalising State presents the risk of letting receiving States who willingly participate in en masse human rights violations go scot-free.
Conclusion
Receiving States occupy a precarious position in the discussion of externalisation of migration policies. While they are often at the mercy of States from the Global North, they are arguably often consenting participants in a larger scheme of violation of human rights. I argue that extending the logic of A.S. v. Italy & Malta, both the externalising and receiving State should share responsibility for human rights violations resulting from externalisation agreements. Specific violations by the receiving State must not be imputed solely to the externalising State merely because the former was acting in furtherance of the latter’s goals.
This piece only explores the first level of the inquiry: the extraterritorial obligations of a receiving State in a simplistic bilateral externalisation agreement. Real life scenarios involving multiple states, non-state actors and even international organisations further problematize apportionment of obligations. Cases like A.S v. Italy and Malta where the shared responsibility of a State could not be discussed due to admissibility issues are merely an unhappy coincidence and cannot be considered precedent for ignoring this issue indefinitely.
Vadita Agarwal is a graduate of the West Bengal National University of Juridical Sciences. Vadita takes an avid interest in Human Rights Law, particularly from a gendered and TWAIL perspective.
The views expressed in this article belong to the author/s and do not necessarily reflect those of the Refugee Law Initiative. We welcome comments and contributions to this blog – please comment below and see here for contribution guidelines.