Skip to content
Conceptualising externalisation: still fit for purpose?

Since its emergence in the early 2000s, the term ‘externalisation’ seems to have developed into an umbrella concept encompassing any migration control measure affecting refugees undertaken either unilaterally or multilaterally, either extraterritorially or with extraterritorial effects. Despite its ubiquitous use, however, the term has rarely been defined, and related and often overlapping concepts have emerged.

Externalisation is linked to concepts of remote control, non-entrée, deterrence, offshoring, extraterritorialisation and protection elsewhere. Moreover, externalisation and these related terms all place States in the Global North at the centre, notwithstanding the frequent involvement of other States. This has the effect of sidelining the significant role, responsibility and accountability structures of ‘external’ States, who are increasingly taking on migration controls at the behest of destination States.

Definition and scope

Confusion as to the definition of the term externalisation centres on a number of questions relating to its scope:

First, the geographic scope of externalisation practices is not always clear. For example, is externalisation limited to a State’s conduct beyond its borders? A narrow geographic scope focuses on extraterritorial migration control, such as pushbacks and extraterritorial asylum processing. Crisp has recently defined externalisation as “measures taken by states in locations beyond their territorial borders to obstruct, deter or otherwise avert the arrival of refugees”.[1] However, a broader definition could include practices implemented after an asylum seeker’s arrival in the destination State that have the effect of externalising protection, thus including primarily territorial mechanisms such as the ‘safe third country’ and ‘first country of asylum’ concepts. Indeed, in its recent Note on the “Externalization” of International Protection, UNHCR defined externalisation as “measures preventing asylum-seekers from entering safe territory and claiming international protection, or transfers of asylum-seekers and refugees to other countries without sufficient safeguards”.[2]

Second, the practical scope of externalisation is necessarily broad to accommodate the wide spectrum of State policies and practices deterring and diverting asylum seekers. These include boat pull and pushbacks, extraterritorial processing and protection, visa controls, carrier sanctions, the posting of immigration officers internationally, and the funding, equipping and training of migration management in third countries. However, some practices sit at the conceptual edges. For instance, do extraterritorial procedures for resettlement and ‘complementary pathways’ fall under the concept of externalisation? They are, after all, asylum and refugee policy measures taken outside a State’s borders. While resettlement procedures are not expressly aimed at deterring asylum seekers, they are at the least highly selective in determining who receives protection and who does not. Resettlement allows destination States to maintain a commitment to the international refugee regime while restricting access to territorial asylum. For example, the EU–Turkey Statement of 2016 includes a built-in resettlement element, with one Syrian refugee resettled for every one returned, while Australia’s externalisation efforts are often justified in terms of a relatively generous resettlement programme. Most recently, the temporary hosting of Afghan refugees in third States, at the request of the US, raises questions of whether we are seeing a new form of externalisation.[3]

Finally, on a related point, there is the question of the extent to which externalisation is a normative term. Do externalisation practices by definition undermine the spirit of destination States’ obligations under international human rights and refugee law or do they in fact violate these obligations outright? In many cases, externalisation practices are the result of governments’ strategic avoidance of their obligations under the 1951 Refugee Convention or other international or regional human rights instruments. On the other hand, do some externalisation practices that are compliant with international law have the potential to protect refugee rights – and could be termed ‘rights-based’ externalisation? Notably, UNHCR has specifically defined the concept as unlawful, contrasting externalisation with “lawful practices involving the transfer of the responsibility for international protection, undertaken in accordance with international standards”.[4]

Still fit for purpose?

Recent years have seen the emergence of what I have termed ‘complex externalisation’, or the embedding of certain protective practices in a broader externalisation or containment agenda. This has also recently been referred to as ‘contained mobility’.[5] For example, while the EU’s role in supporting the Libyan Coast Guard in preventing departures for Europe is a classic example of externalisation, related mobility practices have emerged. The EU is the primary funder of UNHCR’s Emergency Transit Mechanism, which evacuates highly vulnerable asylum seekers and refugees from Libya – where they have been detained as a direct result of EU-supported pullback practices by the Libyan Coast Guard – to Niger and Rwanda. In turn, France, for example, is currently engaged in extraterritorial asylum processing in Niger for the express purpose of resettlement.

Given how embedded some policies of externalisation have become in the Global North, a clear-eyed concept of which practices fall under this label is vital, for at least three reasons. First, as the term ‘externalisation’ does not appear in international law, clarity is needed as to which externalisation practices are in compliance with or violate international law. Second, definitional clarity matters because including or excluding certain practices from the broader concept of externalisation has an impact on how we measure the effects of such policies. Third, the emergence of ‘complex externalisation’ calls for caution in further stretching the conceptual boundaries of externalisation and instead invites an analysis of both the intention of States and the impacts on refugees themselves. To remain categorically useful, future work on externalisation will need to grapple with these definitional questions.


Nikolas Feith Tan @ndftan

Senior Researcher, Danish Institute for Human Rights


[1] Crisp J (2019) ‘Externalization and the erosion of refugee protection’

[2] UNHCR Note on the “Externalization” of International Protection, 28 May 2021

[3] Tan N F (2021) ‘The Temporary Hosting of Evacuated Afghans in Third Countries: Responsibility Sharing or Externalisation?’

[4] See endnote 2.

[5] Carrera S and Cortinovis R (2019) The EU’s Role in Implementing the UN Global Compact on Refugees: Contained mobility vs International Protection, CEPS Paper in Liberty and Security in Europe

This site is registered on as a development site. Switch to a production site key to remove this banner.