Externalization of migration governance: new trends and specific aspects

In his report to the General Assembly in 2025,¹ the Special Rapporteur on the human rights of migrants warned of the growing practice of outsourcing asylum and migration processes, which entails serious risks of human rights violations and abuses. These practices form part of a broader deterrence-based approach that restricts access to protection, increases the risk of violations and reinforces existing gender inequalities and discrimination. In light of the scale and complexity of externalization measures and the proliferation of new agreements, the Special Rapporteur decided to continue working on this subject.

1. A/80/302. See also A/80/632, para. 24; A/HRC/61/31, para. 15; and joint general recommendationNo. 39 (2025) of the Committee on the Elimination of Racial Discrimination and general comment No. 8 (2025) of the Committee on the Protection of the Rights of All Migrant Workers and Members of Their Families on thematic guidelines for eradicating xenophobia towards migrants and others perceived as such, paras. 55–57. ↩

Externalization is the process of shifting functions that are normally carried out by a State within its own territory so that they take place, in part or in whole, outside its territory.² Member States that externalize migration processes are referred to as “externalizing States”, while those cooperating with them are referred to as “third States”. Externalization measures, often framed as forms of cooperation or partnership, should be distinguished from migration cooperation aimed at facilitating safe and regular migration, ensuring access to protection, and strengthening collective responses to migration and asylum governance, including through genuine responsibility-sharing.³ In practice, externalization agreements aim to prevent migrants and refugees from reaching externalizing States or to remove them to third States – practices that may be difficult to reconcile with the principle of good faith. Accordingly, externalization may be understood as forms of migration cooperation that aim at, or result in, shifting responsibility for migration and asylum processes away from externalizing States.⁴

2. Refugee Law Initiative, “Refugee law initiative declaration on externalisation and asylum”, International Journal of Refugee Law, vol. 34, No. 1 (March 2022), pp. 114–119. ↩
3. Submissions from Brazil and Montenegro. ↩
4. The Office of the United Nations High Commissioner for Refugees (UNHCR) relies on a narrower understanding of externalization. See UNHCR, “International agreements for the transfer of refugees and asylum-seekers”, 7 August 2025, available at https://www.refworld.org/policy/legalguidance/unhcr/2025/en/150357, para. 5. ↩

The Special Rapporteur has addressed specific externalization measures in his communications⁵ and submissions to courts.⁶ In the present report, he draws on these recommendations and offers a more holistic assessment. The Special Rapporteur welcomes the written submissions received from Member States, national human rights institutions, international organizations, including United Nations agencies, civil society organizations and academia.⁷ Additional publicly available sources were also consulted in preparing the report.

5. See https://spcommreports.ohchr.org/TmSearch/Mandates?m=33. ↩
6. See https://www.ohchr.org/en/special-procedures/sr-migrants/submissions-courts-and-other-bodies. ↩
7. Submissions received in response to the call for inputs are available at https://www.ohchr.org/en/calls-for-input/2026/call-input-externalization-migration-governance-new-trends-and-specific. ↩

Building on the Special Rapporteur’s report to the General Assembly on the externalization of migration governance and its effect on the human rights of migrants,⁸ the present report discusses recent externalization measures and the specific features of this trend, including human rights risks and impacts on specific groups. The Special Rapporteur also expands on human rights responsibilities and on measures to foster transparency and accountability. The final section provides conclusions and recommendations. 1 A/80/302. See also A/80/632, para. 24; A/HRC/61/31, para. 15; and joint general recommendation No. 39 (2025) of the Committee on the Elimination of Racial Discrimination and general comment No. 8 (2025) of the Committee on the Protection of the Rights of All Migrant Workers and Members of Their Families on thematic guidelines for eradicating xenophobia towards migrants and others perceived as such, paras. 55–57. 2 Refugee Law Initiative, “Refugee law initiative declaration on externalisation and asylum”, International Journal of Refugee Law, vol. 34, No. 1 (March 2022), pp. 114–119. 3 Submissions from Brazil and Montenegro. 4 The Office of the United Nations High Commissioner for Refugees (UNHCR) relies on a narrower understanding of externalization. See UNHCR, “International agreements for the transfer of refugees and asylum-seekers”, 7 August 2025, available at https://www.refworld.org/policy/legalguidance/unhcr/2025/en/150357, para. 5. 5 See https://spcommreports.ohchr.org/TmSearch/Mandates?m=33. 6 See https://www.ohchr.org/en/special-procedures/sr-migrants/submissions-courts-and-other-bodies. 7 Submissions received in response to the call for inputs are available at https://www.ohchr.org/en/calls-for-input/2026/call-input-externalization-migration-governance-newtrends-and-specific. 8 A/80/302. GE.26-06322

8. A/80/302. ↩

Between October 2025 and May 2026, the Special Rapporteur requested visits to Azerbaijan, Ireland, Malaysia, Mauritius, Thailand and the United States of America and to the European Union. He also reiterated his interest in visiting Canada, New Zealand, Portugal, South Africa and Spain. The Special Rapporteur conducted an official visit to Brazil from 16 to 27 March 2026;⁹ his report will be presented in June 2027.

9. See https://www.ohchr.org/en/press-releases/2026/03/brazil-un-expert-welcomes-strong-migration-legal-framework-calls-more. ↩

Since the presentation of his report to the General Assembly in October 2025,¹⁰ the Special Rapporteur has continued to address externalization through communications to Governments and regional organizations¹¹ as well as through his participation in public conferences. In March 2026, he took part in the regional hearing of the Inter-American Commission on Human Rights on the situation of persons deported, expelled or transferred in the context of human mobility in the region.¹² He addressed the Council of Europe Steering Committee for Human Rights on the Safe Third Country Concept and participated in the annual meeting of the Global Alliance of National Human Rights Institutions. In addition, as a member of the Platform of Independent Experts on Refugee Rights, in November 2025 he participated in the fourteenth World Conference of the International Association of Refugee and Migration Judges, in Nairobi.

10. A/80/302. ↩
11. See https://spcommreports.ohchr.org/Tmsearch/TMDocuments. ↩
12. See https://www.ohchr.org/sites/default/files/documents/issues/migration/unsr-migrants-iachr-regional-hearing-11-03-26.pdf. ↩

The Special Rapporteur participated in the second International Migration Review Forum, which took place from 5 to 8 May 2026 in New York.

Since the finalization of his report to the General Assembly on the externalization of migration governance and its effect on the human rights of migrants,¹³ new externalization agreements and arrangements have emerged and the implementation of recently concluded agreements has provided further detail. The Special Rapporteur groups the externalization measures into three categories: (a) prevention of arrival; (b) extraterritorial processing of asylum claims; and (c) removal to third countries.¹⁴

13. A/80/302. ↩
14. Ibid., para. 6. ↩

This form of externalization seeks to prevent migrants from reaching the territory of the externalizing State.¹⁵ Within such frameworks, third States intercept, divert or otherwise impede onward movement, including by preventing departure, restricting transit, or returning migrants to their countries of transit or origin. Such measures raise serious risks of refoulement, including chain refoulement, and of collective expulsion and the absence of individualized assessment, and may engage the international human rights obligations of externalizing States, including through complicity or where they exercise jurisdiction or effective control extraterritorially. To support these measures, externalizing States provide financial assistance, training, equipment and broader capacity-building to migration and border authorities in third States.¹⁶ Increasingly, this includes the deployment of surveillance technologies, such as biometric systems, drones and border-monitoring tools, facilitating the 9 See https://www.ohchr.org/en/press-releases/2026/03/brazil-un-expert-welcomes-strong-migrationlegal-framework-calls-more. 10 A/80/302. 11 See https://spcommreports.ohchr.org/Tmsearch/TMDocuments. 12 See https://www.ohchr.org/sites/default/files/documents/issues/migration/unsr-migrants-iachrregional-hearing-11-03-26.pdf. 13 A/80/302. 14 Ibid., para. 6. 15 Ibid., para. 7. See also the submission from Ombudsman’s Office of Colombia. 16 The United States of America has provided large amounts of funding to Mexico to improve Mexican border control with Guatemala and Belize; see https://www.congress.gov/crs_external_products/IF/PDF/IF10215/IF10215.30.pdf. GE.26-06322 tracking and interception of migrants.¹⁷ For example, the migration database of Türkiye (Göç Net) has been expanded with support from the European Union with the aim of preventing arrivals to the European Union,¹⁸ while in November 2025 the European Border and Coast Guard Agency (Frontex) and Cabo Verde launched cooperation for aerial surveillance over routes towards the Canary Islands.¹⁹

15. Ibid., para. 7. See also the submission from Ombudsman’s Office of Colombia. ↩
16. The United States of America has provided large amounts of funding to Mexico to improve Mexican border control with Guatemala and Belize; see https://www.congress.gov/crs_external_products/IF/PDF/IF10215/IF10215.30.pdf. ↩
17. Antonella Napolitano, “Outsourcing surveillance: the technological side of border externalisation”, 30 September 2025. ↩
18. Submission by E. Kuşkonmaz and E. Küçük. ↩
19. See https://www.frontex.europa.eu/media-centre/news/news-release/frontex-and-cabo-verde-launch-aerial-surveillance-to-strengthen-maritime-awareness-and-save-lives-at-sea-c0uEB1?utm. ↩

Such practices are embedded in broader cooperation frameworks, including European Union arrangements with countries such as Egypt, Jordan, Lebanon, Mauritania, Morocco and Tunisia.²⁰ As a key recent development, in November 2025, the Memorandum of Understanding between Italy and Libya was automatically renewed for a further three-year period.²¹ It is aimed at strengthening cooperation in addressing irregular migration, including through enhanced border control and surveillance. The Memorandum of Understanding was accompanied by training, funding and equipping of the Libyan Coast Guard by Italy and the European Union, the establishment of a rescue coordination centre in Tripoli, and the declaration by Libya of a search and rescue region over a large area of the central Mediterranean. These measures contributed to a shift in search and rescue coordination in the central Mediterranean towards the Libyan Coast Guard, and coincided with increasing efforts to restrict the operations of humanitarian non-governmental organizations (NGOs), including the firing of live ammunition.²² The Libyan Coast Guard has intercepted numerous boats at sea and returned individuals to Libya, where they face a high risk of serious and widespread human rights violations.²³

20. A/80/302, paras. 10 and 11. ↩
21. See https://www.asgi.it/wp-content/uploads/2017/02/ITALY-LIBYA-MEMORANDUM-02.02.2017.pdf. ↩
22. See communications OTH 131/2025, LBY 2/2025 and ITA 6/2025 and the replies thereto. All communications mentioned in the present report are available from https://spcommreports.ohchr.org/Tmsearch/TMDocuments. ↩
23. United Nations Support Mission in Libya (UNSMIL) and Office of the United Nations High Commissioner for Human Rights (OHCHR), Business as Usual: Human Rights Violations and Abuses against Migrants, Asylum-Seekers and Refugees in Libya, February 2026. Similar cooperation at sea occurs between Malta and Libyan authorities, while Greece has reportedly supported the Libyan National Army since summer 2025 through training and technical assistance, including maritime operations; see https://www.reuters.com/world/africa/libyan-coast-guards-train-greece-under-plan-stem-migrant-flows-2025-07-30/ and the submission by the University of Bristol Human Rights Implementation Centre. ↩

Under externalization involving extraterritorial processing of asylum claims, States relocate elements of asylum procedures to a third State,²⁴ including registration, processing or determination. Individuals transferred under such arrangements have not had their claims assessed on the merits by the externalizing State, raising concerns about access to procedures, safeguards, oversight and effective remedies. Examples of such arrangements include the Memorandum of Understanding between Australia and Nauru²⁵ and the original iteration of the Italy–Albania protocol.²⁶ Recent developments in the United States of America and in the European Union provide further examples of this form of externalization.

24. A/80/302, para. 12. ↩
25. Ibid., para. 13. ↩
26. Ibid., para. 15. ↩

European Union: reform of the “safe third country” concept

Legislative amendments to be implemented from June 2026 expand the ability of European Union member States to designate non-European Union countries as “safe third countries” and reject asylum claims as inadmissible. Firstly, the current requirement that a third country be able to grant refugee status is replaced with a more flexible notion of “effective protection”, and a country may be considered safe with the exception of certain regions or for specific categories of persons.²⁷ Secondly, the current requirement of a genuine connection between the individual and the safe third country is weakened, and mere transit through the country, or presence at the border or a transit zone, may be deemed sufficient. Transit may not be required where an agreement or arrangement with the third country provides for examination of the merits of a claim for protection.²⁸ Asylum-seekers may be transferred to countries with which they have no connection, raising risks of refoulement and chain refoulement, which are increased by weakened individual assessments and the limited suspensive effect of appeals. Overreliance on the “safe third country” concept may shift responsibility to third States and undermine access to asylum in the European Union. In contrast, the Council of Europe relies on stronger safeguards in its use of the “safe third country” concept.²⁹ A parallel can be drawn between the European Union safe third country mechanism and the Dublin system, which allocates responsibility primarily to the country of first entry into the Schengen area based on a presumption of safety. In practice, however, this system has exposed applicants to risks due to deficiencies in some asylum systems³⁰ and has effectively shifted responsibility between States. The system will be replaced in June 2026 by the Asylum and Migration Management Regulation, which largely maintains this presumption.

27. Asylum Procedure Regulation (2024/1348), arts. 57 and 59. ↩
28. Safe Third Country Regulation (2026/463), art. 1. ↩
29. Submission from Council of Europe, Directorate General Human Rights and Rule of Law. ↩
30. Bayush Alemseged Araya v. Denmark (CCPR/C/123/D/2575/2015); and K.J. v. Switzerland (CEDAW/C/91/D/169/2021). ↩

United States: asylum cooperative agreements

The United States has reportedly concluded or reactivated so-called asylum cooperative agreements – a form of “safe third country” agreement – with Belize, Cabo Verde, Ecuador, Guatemala, Honduras, Liberia, Paraguay and Uganda.³¹ The United States may consider a person ineligible to apply for asylum in the United States due to the application of a safe third country agreement and, pursuant to the asylum cooperative agreements, remove asylum seekers – including those with cases already pending before its immigration courts – without a full examination of the merits of their claims. In the countries concerned, these individuals are expected to have access to asylum procedures; however, serious concerns arise regarding the accessibility, effectiveness and fairness of these procedures for persons removed from the United States. Between November 2025 and February 2026, nearly 9,500 asylum applications were reportedly declared inadmissible, with a view to transferring applicants to asylum cooperative agreement partner countries – approximately half to Ecuador and one third to Honduras.³² Among other concerns, practices under the asylum cooperative agreements risk undermining the right to seek asylum and may result in refoulement, arbitrary detention and the denial of due process.

31. See https://www.thirdcountrydeportationwatch.org; and submissions from Fundación Scalabriniana and others and from the Centro de Derechos Humanos de la Universidad Católica Andrés Bello and others. ↩
32. See https://www.thirdcountrydeportationwatch.org/. ↩

In externalization involving readmission or expulsion to a third State, the externalizing State removes an individual to a country other than their country of nationality.³³ Such transfers have traditionally involved transit countries. Recent developments envisage sending individuals to countries with which they have no prior connection (e.g. “return hubs”). In contrast to externalization involving extraterritorial asylum processing, these arrangements are often presented as applying only to migrants in an irregular situation, including persons whose asylum applications have been examined and rejected. In practice, however, they may well apply to individuals whose asylum claims have not been assessed on the merits. These measures raise serious risks of refoulement, including chain refoulement, of lack of individual assessment and effective remedy, and of arbitrary detention and ill-treatment.

33. A/80/302, para. 17. ↩

Australia: third-country reception arrangement with Nauru

Following the ruling by the High Court of Australia in the NZYQ case that indefinite immigration detention where removal is not reasonably practicable is unconstitutional,³⁴ at least 300 individuals, many under non-refoulement protection, were released. On 29 August 2025, Australia concluded a new, non-public Memorandum of Understanding with Nauru to facilitate their removal to Nauru, alongside adopting legislation limiting procedural fairness in removal decisions.³⁵ The law allows transfer to third countries without proper notice, an opportunity to be heard or the assessment of risks, and may permit removal before appeals have been exhausted. These measures revive long-standing concerns about the treatment of individuals sent to Nauru.³⁶

34. NZYQ v. Minister for Immigration, Citizenship and Multicultural Affairs (2023). ↩
35. Submissions from the Australian Human Rights Commission, the Asylum Seeker Resource Centre and the Refugee Council of Australia. ↩
36. A/80/302, para. 13. ↩

United States: recent third-country transfer agreements

Since the finalization of the report of the Special Rapporteur to the General Assembly on the externalization of migration governance and its effect on the human rights of migrants,³⁷ the United States has reportedly concluded so-called third-country removal agreements with multiple countries,³⁸ including Antigua and Barbuda, Burundi, Cabo Verde, Cameroon, Central African Republic, Dominica, Equatorial Guinea, Eswatini, Ghana, Guyana, Palau, Poland, Rwanda, Saint Kitts and Nevis, Sierra Leone, South Sudan and Uzbekistan. Removals under these agreements have reportedly taken place to – among other countries – Cameroon, Equatorial Guinea,³⁹ Eswatini, Ghana, Republic of Moldova, Rwanda, South Sudan and Uzbekistan.⁴⁰ Discussions with additional countries have been ongoing, and more recently, transfer agreements have been concluded with Costa Rica and the Democratic Republic of the Congo, with the first deportation flights to these countries reportedly carried out in April 2026.⁴¹

37. A/80/302. ↩
38. Ibid., para. 22. ↩
39. See https://www.ohchr.org/en/press-releases/2026/05/experts-urge-equatorial-guinea-uphold-international-obligations-and-halt. ↩
40. See https://www.thirdcountrydeportationwatch.org/. ↩
41. See https://www.thirdcountrydeportationwatch.org/democratic-republic-of-the-congo,and https://www.refugeesinternational.org/statements-and-news/refugees-international-concerned-by-u-s-costa-rica-third-country-national-deportation-deal (this is a distinct agreement from the one concluded in 2025 noted in A/80/302, para. 22). ↩

European Union: introduction of “return hubs” into European Union legislation

The current reform of the European Union return rules, set to be adopted in the first half of 2026, introduces so-called return hubs into the European Union legal framework. Under the future Return Regulation, member States of the European Union may be able to remove a person, apart from in cases where a narrow range of exceptions applies, to a third country with which they have an agreement or arrangement. Although such an agreement or arrangement may be concluded only with a third country that is assessed as respecting international human rights standards and principles, including the principle of non-refoulement, it is unclear how these safeguards will be implemented in practice. Crucially, no connection between the country hosting the return hub and the person will be required. In their joint letter to the European Union of January 2026, 16 special procedures of the Human Rights Council expressed their concern about return hubs agreements and/or arrangements not being conditional on a prior human rights impact assessment and independent oversight, and about the lack of an individual assessment of protection needs prior to the transfer. The return hubs raise serious human rights concerns, in particular regarding deprivation of liberty, conditions of stay, access to remedies, and the risk of onward return or chain refoulement, as well as the absence of safeguards to ensure full respect for other international human rights obligations.⁴²

42. See OTH 166/2025, p. 15; submission from Brazil; https://europe.ohchr.org/sites/default/files/2026-03/EU Return Regulation - Key Takeaways_final.pdf; and Council of Europe Commissioner for Human Rights, Observations on the new European Union returns framework, March 2026. ↩

As described in detail in his previous report,⁴³ the arrangements between Italy and Albania, based on the 2023 protocol and subsequent implementing measures, were originally designed to enable Italy to transfer to Albania certain asylum applicants intercepted in international waters, coming from “safe countries of origin”, for the purpose of conducting accelerated asylum procedures and – where applicable – return procedures, under Italian law.⁴⁴ In practice, however, the implementation of this model has repeatedly been suspended by domestic courts, and since March 2025 the arrangements have been adapted to allow the transfer to Albania of persons already present in Italy who are subject to return decisions. Individuals transferred are processed in facilities established by Italy on Albanian territory, notably in Shëngjin for initial processing and in Gjadër for detention pending removal, under the responsibility of Italian authorities.⁴⁵ These arrangements raise serious concerns regarding the adequacy of procedural safeguards in accelerated procedures, the use and conditions of detention, the risk of onward removal, and respect for the right to health, as well as persistent shortcomings in transparency and independent oversight.

43. A/80/302, para. 21. ↩
44. See ITA 3/2024 and ALB 1/2024, and the government reply dated 23 August 2024. ↩
45. Klodiana Millona and Kristina Millona, “Europe expands its carceral edge into Albania”, 23 February 2026. ↩

Recently, comparable arrangements have also been considered or pursued by other European countries. In September 2025, Netherlands (Kingdom of the) and Uganda signed a letter of intent to establish a “transit hub” in Uganda, where migrants from the region would be held awaiting onward return to their countries of origin.⁴⁶ In October 2025, media reports indicated that the United Kingdom of Great Britain and Northern Ireland had been in discussions with Kosovo⁴⁷ regarding the establishment of “return hubs”.⁴⁸

46. See https://www.government.nl/latest/news/2025/09/26/the-netherlands-and-uganda-make-arrangements-on-returning-migrants-via-uganda. ↩
47. References to Kosovo shall be understood to be in the context of Security Council resolution 1244 (1999). ↩
48. See https://www.theguardian.com/world/2025/oct/21/kosovo-offers-to-become-first-country-to-accept-britains-unsuccessful-asylum-seekers. ↩

Externalization agreements and arrangements entail a high risk of human rights violations, arising from a range of interrelated factors.⁴⁹ Firstly, they involve measures that are inherently sensitive from a human rights perspective, such as removal and immigration detention. Secondly, the relationships between the States concerned may facilitate violations, as third States – some of which face instability or conflict – may lack the capacity or willingness to ensure adequate protection and respect for the rights and well-being of migrants and refugees. At the same time, efforts by externalizing States to transfer migrants to or contain them in third States may reflect a deprioritization of their rights and their dignity, as evidenced by instances where allegations of human rights violations have not triggered effective responses. Thirdly, limited transparency surrounding such arrangements, coupled with the absence of effective and independent oversight mechanisms, further heightens the risk of violations and abuse. Fourthly, implementing partners, including international and intergovernmental organizations as well as private sector actors, may contribute to the implementation of activities carrying heightened human rights risks, including where funding structures create adverse incentives.⁵⁰

49. A/80/302, paras. 23–41. ↩
50. Submissions from K. Allinson, Human Rights Advocates and the Kennedy Human Rights Center. ↩

In his report to the General Assembly on the externalization of migration governance and its effect on the human rights of migrants,⁵¹ the Special Rapporteur identified 10 key rights at risk of violation, namely: the right to leave any country, including one’s own; the prohibition of refoulement, collective expulsion, arbitrary detention, torture and ill-treatment, enforced disappearances, and racial discrimination; the right to life, due process and an effective remedy; and economic and social rights.⁵²

51. A/80/302. ↩
52. See also the submission from Algeria. ↩

The Special Rapporteur is deeply concerned by recent reports of violations of the principle of non-refoulement. Reports indicate that migrants removed from the United States to third countries pursuant to recent agreements have been subjected to arbitrary detention, torture and ill-treatment.⁵³ Some individuals transferred to El Salvador have reportedly been subjected to enforced disappearance and have been denied access to family members and legal counsel.⁵⁴ Reports also indicate violations of the prohibition of indirect (chain) refoulement. Receiving States subsequently returned individuals to their countries of origin, exposing them to a risk of irreparable harm.⁵⁵ Third States reportedly provide assurances against persecution, torture or onward refoulement, but such assurances do not absolve States of their non-refoulement obligations and have proven ineffective in practice. Blanket diplomatic assurances are particularly deficient, as they do not permit the required individualized risk assessment and are difficult to monitor in practice. Ongoing violations of migrants’ rights in Libya⁵⁶ and Tunisia⁵⁷ further suggest that migration cooperation aimed at preventing arrivals to the European Union (see paras. 9 and 10 above) risks resulting in violations of the principle of non-refoulement.

53. See USA 25/2025, VEN 3/2025 and SLV 6/2025 and the replies thereto; the amicus brief available at https://www.ohchr.org/sites/default/files/documents/issues/migration/2025-07-31-amicus-brief-srs-no-mc-578-25-el-salvador.pdf, paras. 9.2 and 15; and the submission from Global Detention Project. ↩
54. Amicus brief, paras. 15 and 22.5; and submission from Human Rights First. ↩
55. See https://www.hrw.org/news/2025/09/23/us/africa-expulsion-deals-flout-rights; and Cristiano d’Orsi, “The United States’ attempt to deport third country nationals to Africa: is this a mutually beneficial situation?”, Externalizing Asylum (15 April 2026). See also USA 42/2025 and HND 7/2025. ↩
56. UNSMIL and OHCHR, Business as Usual: Human Rights Violations and Abuses against Migrants, Asylum-Seekers and Refugees in Libya. ↩
57. See TUN 2/2026, https://www.amnesty.org/en/documents/mde30/0180/2025/en/ and https://www.hrw.org/news/2025/09/26/shadow-report-on-the-implementation-of-the-african-charter-on-human-and-peoples. ↩

Beyond removals to third States and extraterritorial asylum processing, detention may also result from externalization measures aimed at preventing arrivals (see para. 9 above). For instance, in the framework of the 2024 arrangement between Mauritania and the European Union,⁵⁸ two European Union-supported transit centres in Nouakchott and Nouadhibou were transferred to the Mauritanian authorities in October 2025, potentially entrenching detention practices at the domestic level.⁵⁹ Evidence from other contexts highlights the serious risks inherent in the use of detention to prevent migrants from reaching externalizing States. In Libya, migrants, asylum-seekers and refugees are routinely subjected to arbitrary detention in inhuman conditions, where they face torture, ill-treatment, sexual violence, forced labour and extortion.⁶⁰

58. A/80/302, para. 11. ↩
59. See A/HRC/62/35/Add.1. ↩
60. UNSMIL and OHCHR, Business as Usual: Human Rights Violations and Abuses against Migrants, Asylum-Seekers and Refugees in Libya. ↩

In addition to the 10 human rights at heightened risk of violation in the context of externalization (see para. 21 above), the Special Rapporteur highlights three further rights. First, externalization practices may facilitate forced labour and practices amounting to 50 Submissions from K. Allinson, Human Rights Advocates and the Kennedy Human Rights Center. 51 A/80/302. 52 See also the submission from Algeria. 53 See USA 25/2025, VEN 3/2025 and SLV 6/2025 and the replies thereto; the amicus brief available at https://www.ohchr.org/sites/default/files/documents/issues/migration/2025-07-31-amicus-brief-srsno-mc-578-25-el-salvador.pdf, paras. 9.2 and 15; and the submission from Global Detention Project. 54 Amicus brief, paras. 15 and 22.5; and submission from Human Rights First. 55 See https://www.hrw.org/news/2025/09/23/us/africa-expulsion-deals-flout-rights; and Cristiano d’Orsi, “The United States’ attempt to deport third country nationals to Africa: is this a mutually beneficial situation?”, Externalizing Asylum (15 April 2026). See also USA 42/2025 and HND 7/2025. 56 UNSMIL and OHCHR, Business as Usual: Human Rights Violations and Abuses against Migrants, Asylum-Seekers and Refugees in Libya. 57 See TUN 2/2026, https://www.amnesty.org/en/documents/mde30/0180/2025/en/ and https://www.hrw.org/news/2025/09/26/shadow-report-on-the-implementation-of-the-african-charteron-human-and-peoples. 58 A/80/302, para. 11. 59 See A/HRC/62/35/Add.1. 60 UNSMIL and OHCHR, Business as Usual: Human Rights Violations and Abuses against Migrants, Asylum-Seekers and Refugees in Libya. GE.26-06322 slavery, prohibited under article 8 of the International Covenant on Civil and Political Rights. In Libya, migrants, asylum-seekers and refugees are frequently trafficked, sold or transferred between actors and compelled to perform labour under coercion, including in detention facilities and private households and at informal worksites.⁶¹ Also, in other countries with which the European Union has established migration control cooperation, migrants are at heightened risk of exploitation, including trafficking for sexual exploitation, forced labour and domestic servitude.⁶²

61. Ibid. ↩
62. See https://www.ohchr.org/en/press-releases/2026/03/egypt-un-experts-raise-alarm-over-violations-against-refugees-and-migrants. ↩

Second, externalization measures risk separating families, potentially violating the obligation to protect the family, and the prohibition of arbitrary interference with one’s privacy, family or home, under articles 23 and 17 of the International Covenant on Civil and Political Rights respectively. Cases have been reported of migrants being deported from the United States to third countries, resulting in separation from their children and spouses residing there.⁶³ Concerns also arise regarding the recent Memorandum of Understanding between Australia and Nauru, which risks separating individuals subject to its arrangements from their family members in Australia, particularly where they have lived in the Australian community for extended periods.⁶⁴

63. Submission from Human Rights First. ↩
64. Submission from Australian Human Rights Commission. ↩

Third, the use of digital technologies in the context of externalization (see para. 9 above) poses additional human rights risks. For instance, surveillance technologies, including drones and border monitoring systems, may facilitate violations of several rights, including the right to leave any country and the prohibition of refoulement.⁶⁵ The collection of personal data for border management purposes may give rise to violations of the right to privacy and of data protection standards. Under the right to privacy, protected by article 17 of the International Covenant on Civil and Political Rights, the collection, storage and processing of personal data must be regulated by law and accompanied by safeguards against arbitrary or unlawful interference, including protection against misuse. Personal data should not be used for purposes incompatible with the Covenant.⁶⁶ More detailed data protection standards – such as collection limitation, data quality, purpose specification, use limitation, security safeguards, openness, individual participation and accountability – are reflected in international standards.⁶⁷

65. OHCHR and University of Essex, “Digital border governance: a human rights based approach”, September 2023. ↩
66. Human Rights Committee, general comment No. 16 (1988), para. 10. ↩
67. Organisation for Economic Co-operation and Development, Guidelines on the Protection of Privacy and Transborder Flows of Personal Data, 2002. ↩

Finally, and more broadly, the Special Rapporteur is concerned that externalization frameworks are prone to human rights violations due to their arbitrary and deterrent nature. In some cases, individuals with protection status or long residence have been removed to third States, creating protection gaps. Those affected may face legal and social limbo, with limited access to livelihoods and heightened vulnerability to exploitation. The threat of transfer and precarious conditions may also pressure individuals to leave despite their protection needs.

Externalization measures have particularly severe impacts on persons in vulnerable situations and entail heightened risks for groups exposed to intersectional discrimination.

Externalization measures render migrant and asylum-seeking children particularly vulnerable to violence, abuse and violations of their rights, thereby significantly affecting 61 Ibid. 62 See https://www.ohchr.org/en/press-releases/2026/03/egypt-un-experts-raise-alarm-over-violationsagainst-refugees-and-migrants. 63 Submission from Human Rights First. 64 Submission from Australian Human Rights Commission. 65 OHCHR and University of Essex, “Digital border governance: a human rights based approach”, September 2023. 66 Human Rights Committee, general comment No. 16 (1988), para. 10. 67 Organisation for Economic Co-operation and Development, Guidelines on the Protection of Privacy and Transborder Flows of Personal Data, 2002. GE.26-06322 their safety, well-being and best interests. Reports indicate that, at various borders, children are exposed to complex forms of violence, including physical harm, sexual violence, pushbacks and other forms of forced removal, detention, psychological abuse, racism, neglect, and deprivation of basic needs.⁶⁸ 68 Accelerated procedures commonly associated with externalization frameworks may not allow for proper age assessment, for an adequate determination of the best interests of the child or for the effective realization of the child’s right to be heard.⁶⁹ Removals to third countries or extraterritorial asylum processing may also result in the separation of children from their parents, in violation of article 9 of the Convention on the Rights of the Child.⁷⁰

68. Submission from I. Derluyn. ↩
69. Submissions from Associazione Comunità Papa Giovanni XXIII and from AsyLex. ↩
70. See USA 37/2025; and submissions from Human Rights First and from A. Talbot. ↩

Externalization measures may give rise to violations of the principle of non-refoulement in respect of children. Under this principle, States must not return a child where there are substantial grounds for believing that he or she would face a real risk of irreparable harm. In the case of children, the scope of such harm is broader and may include, for example, the serious consequences of inadequate access to food, healthcare or other essential services.⁷¹ In A.M. v. Switzerland, the Committee on the Rights of the Child found that removing a child with his mother to a transit country would violate several provisions of the Convention on the Rights of the Child, including the rights to survival and development, to an adequate standard of living, to education, and to freedom from torture and ill-treatment.⁷² Removal to a third State will rarely be in the best interests of the child, particularly where there are concerns as to the ability or willingness of that State to ensure adequate protection and well-being. For instance, in Nauru, children transferred from Australia have reported abuse, including sexual abuse.⁷³ In Tunisia, children in transit reportedly often lack access to public healthcare.⁷⁴

71. Z.S. and A.S v. Switzerland (CRC/C/89/D/74/2019), para. 7.3. ↩
72. A.M. v. Switzerland (CRC/C/88/D/95/2019), paras. 10.7 and 11. ↩
73. Submissions from A. Talbot and from the Asylum Seeker Resource Centre. ↩
74. World Organisation Against Torture, The right to health for children on the move in Tunisia, 2025. ↩

Immigration detention, which frequently results from externalization measures (see paras. 22 and 23 above), also affects children. Recent reports indicate that in the United States, children have been included in deportations to third countries, often following periods of detention pending removal. The Special Rapporteur is concerned by recent accounts of children being systematically detained in the United States for prolonged periods in substandard conditions and, in many cases, without effective access to legal representation.⁷⁵ In Libya, children are detained in deplorable conditions, where they are exposed to, or witness, physical abuse.⁷⁶ Immigration detention is inherently harmful to children’s physical and mental health and development, and is never in their best interests. Accordingly, the possibility under international human rights law of detaining children as a measure of last resort, which may apply in other contexts such as juvenile justice, does not apply in immigration proceedings. Detention of children on the basis of their own or their parents’ migration status is therefore impermissible.⁷⁷ In M.I. et al. v. Australia, the Human Rights Committee found that Australia had violated article 9 (1) of the International Covenant on Civil and Political Rights by detaining unaccompanied children in the offshore processing centre in Nauru, in overcrowded and inadequate conditions.⁷⁸

75. See https://www.ohchr.org/en/press-releases/2026/01/un-experts-alarmed-child-rights-violations-usa-immigration-procedures; USA 39/2025; and Human Rights First and RAICES, “A new era of ICE family prisons”, April 2026. ↩
76. Submission from I. Derluyn. ↩
77. Joint general comment No. 4 of the Committee on the Protection of the Rights of All Migrant Workers and Members of Their Families and No. 23 of the Committee on the Rights of the Child (2017) on State obligations regarding the human rights of children in the context of international migration in countries of origin, transit, destination and return, paras. 9–12. ↩
78. M.I. et al. v. Australia (CCPR/C/142/D/2749/2016), para. 10.11. ↩

Externalization measures can have gendered impacts,⁷⁹ affecting women, men and persons of diverse sexual orientation, gender identity, gender expression and sex characteristics in different ways. Pre-existing gender inequalities render risks to human rights, safety and well-being particularly acute for women, girls and persons of diverse sexual orientation, gender identity, gender expression and sex characteristics, and may be exacerbated by externalization, increasing their exposure to gender-based violence, exploitation and abuse. Despite these risks, externalization agreements often lack mechanisms to identify and prevent gender-specific harm, including by ensuring access to gender-responsive and culturally sensitive healthcare, including sexual and reproductive health services, which may be unavailable, inadequate or even criminalized in destination States.

79. A/80/302, paras. 10, 13, 33 and 63 (a). ↩

The externalization of migration control has coincided with intensified border controls, particularly in destination countries, contributing to increasingly dangerous and violent migration routes. These dynamics give rise to gender-specific risks and forms of harm: while men more frequently report physical violence, women and girls are disproportionately exposed to sexual harassment, violence and exploitation. During interception and at borders, migrants may face sexual harassment, extortion and coercion by State and non-State actors, including in the context of pushbacks that leave them stranded in unsafe areas with heightened risks of rape, kidnapping and trafficking. Protection gaps persist, including the absence of gender-responsive screening, of confidential reporting mechanisms for gender-based violence, and of effective access to asylum procedures. By restricting the access to safe and regular pathways, border militarization increases reliance on smuggling networks, where women and girls are particularly exposed to power imbalances, gender-based violence and exploitation.⁸⁰

80. Submissions from the United Nations Entity for Gender Equality and the Empowerment of Women (UN-Women) and from Women in Migration Network; and see EuroMed Rights, “A gendered analysis of EU migration externalisations policies”, February 2024. ↩

Removal within externalization frameworks may expose migrants to a risk of gender-related irreparable harm, both in the receiving third State and in any subsequent State in the context of onward removal. Under the Convention on the Elimination of All Forms of Discrimination against Women, States must not remove a person to any country where they would face a real, personal and foreseeable risk of serious forms of discrimination, including gender-based violence.⁸¹ In K.E. v. Switzerland, the Committee on the Elimination of Discrimination against Women found that returning a survivor of gender-based violence to a third State would violate the Convention, and emphasized that any removal must be preceded by an individualized, trauma-informed and gender-sensitive assessment of the risk of irreparable harm.⁸² States cannot thus rely on a general designation of a country as “safe” without assessing the specific risks faced by women and girls and others in need of protection, including in the light of intersecting forms of discrimination.⁸³ Externalization measures may also result in migrants with diverse sexual orientation, gender identity, gender expression or sex characteristics being removed to contexts where same-sex relations are criminalized or where State authorities lack the capacity or willingness to ensure gender-responsive protection.⁸⁴ In some documented cases, migrants with diverse sexual orientation, gender identity, gender expression and sex characteristics removed from the United States to third States were subsequently returned to their countries of origin, exposing them to a risk of persecution or torture.⁸⁵

81. A. v. Denmark (CEDAW/C/62/D/53/2013), para. 9.8; and R.S.A.A. et al. v. Denmark (CEDAW/C/73/D/86/2015), para. 8.9. ↩
82. K.J. v. Switzerland (CEDAW/C/91/D/169/2021), paras. 7.6, 7.7 and 9. ↩
83. Submissions from M. Gleeson and from AsyLex. ↩
84. Submissions from Asylum Seeker Resource Centre and from Refugee Council of Australia. ↩
85. Human Rights Watch, US/Africa: expulsion deals flout rights; and Cristiano d’Orsi, “The United States’ attempt to deport third country nationals to Africa: is this a mutually beneficial situation?”. ↩

Immigration detention also has gendered impacts. Migrant women face heightened risks of sexual and gender-based violence in detention, while access to gender-responsive healthcare, including sexual and reproductive health services, is often lacking. Poor conditions, including overcrowding, inadequate sanitation and limited access to medical, legal and social support, disproportionately affect women and girls.⁸⁶ In Libya, gender-based violence against women and girls is reported to be systematic. Men are also subjected to sexual violence, including rape and sexual exploitation, although such violations are likely to be underreported due to stigma and trauma.⁸⁷

86. Submissions from UN-Women; International Detention Coalition; and Women in Migration Network; and see EuroMed Rights, “A gendered analysis of EU migration externalisations policies”. ↩
87. UNSMIL and OHCHR, Business as Usual: Human Rights Violations and Abuses against Migrants, Asylum-Seekers and Refugees in Libya. ↩

Externalization should be examined in the light of systemic discrimination embedded in migration governance, including racialized mobility hierarchies and discriminatory enforcement practices. In practice, externalization agreements and their implementation result in disproportionate adverse impacts on nationals of global South countries, particularly those belonging to racialized and ethnic minority groups, restricting access to asylum and protection. Externalization measures frequently rely on securitized approaches, including the use of digital technologies at borders, that may result in racial profiling and disproportionate targeting of migrants from the global South for interception, detention and deportation.⁸⁸ Externalization thus tends to perpetuate patterns of exclusion and unequal treatment affecting racialized migrants and to weaken protection.

88. Joint general recommendation No. 38 (2025) of the Committee on the Elimination of Racial Discrimination and general comment No. 7 (2025) of the Committee on the Protection of the Rights of All Migrant Workers and Members of Their Families on general guidelines for eradicating xenophobia towards migrants and others perceived as such, paras. 13–19 and 48;A/75/590, paras. 22 and 43; and submission from Women in Migration Network. ↩

Externalization is further structurally intertwined with intersectional forms of discrimination, whereby multiple and intersecting grounds – including race, ethnic origin, nationality, gender, sexual orientation, gender identity, migration status and socioeconomic position – interact to compound vulnerability and heighten exposure to human rights violations, discrimination and harm.⁸⁹ From an intersectional perspective, racial discrimination and gender-based violence are interconnected. For example, women and persons with diverse sexual orientation, gender identity, gender expression or sex characteristics may face heightened risks of discrimination, gender-based violence, trafficking and labour exploitation.⁹⁰ Reports indicate that in Nauru, migrants with intersectional vulnerabilities may not receive adequate protection or targeted support, including women who lack access to safe housing and to gender-responsive healthcare.⁹¹

89. A/HRC/59/62, para. 10; joint general recommendation No. 38 (2025) of the Committee on the Elimination of Racial Discrimination and general comment No. 7 (2025) of the Committee on the Protection of the Rights of All Migrant Workers and Members of Their Families, para. 20;and submission from Montenegro. ↩
90. Joint general recommendation No. 38 (2025) of the Committee on the Elimination of Racial Discrimination and general comment No. 7 (2025) of the Committee on the Protection of the Rights of All Migrant Workers and Members of Their Families, paras. 48 and 49;and submission from Women in Migration Network. ↩
91. Submissions from A. Talbot and from the Asylum Seeker Resource Centre. ↩

Externalization measures pose a significant risk of violating multiple human rights of migrants, yet establishing responsibility and securing redress for victims remains challenging. In addition to the often informal nature of such arrangements, two factors are particularly significant: their extraterritorial character, and the involvement of multiple actors, which may diffuse responsibility. Alongside the States concerned, a range of non-State actors are involved in the implementation of externalization measures, including international organizations, such as the United Nations agencies,⁹² and NGOs, as well as private entities such as technology firms, detention facility operators, security companies and private carriers.⁹³ For example, deportation flights from the United States are frequently carried out through contracted private aviation services, often coordinated by companies acting as brokers for multiple carriers on behalf of the Government.⁹⁴ Building on the foregoing analysis,⁹⁵ the present section examines potential avenues for redress at both the domestic and international levels.

92. Submissions from Brazil and from K. Allinson. ↩
93. Submission from Ecuador. ↩
94. Submissions from Human Rights Advocates and from the Kennedy Human Rights Center. ↩
95. A/80/302, paras. 49–59. ↩

States are required, in line with their positive obligations, to conduct prompt, independent, impartial and effective investigations, to hold those responsible accountable, and to provide redress for harm, whether caused by their own organs or agents or by private actors.⁹⁶ In many legal systems, private entities may incur civil, administrative, or in serious cases, criminal liability, including where they perform public functions. Domestic courts have increasingly been called upon to assess externalization arrangements and related measures. In Costa Rica, the Constitutional Chamber found that migrants removed from the United States had been subjected to violations of their rights to liberty, information and access to legal counsel, and ordered their release.⁹⁷ In Italy, courts have on several occasions refused to validate detention orders issued in respect of migrants removed to centres in Albania, finding that their countries of origin could not be considered “safe”, and ordering their return to Italy.⁹⁸ In Australia, the Federal Court held that the authorities owed a duty of care to an asylum-seeker removed to Nauru and subsequently to Papua New Guinea for medical treatment, and granted injunctive relief.⁹⁹ In Eswatini, the High Court held that migrants deported from the United States were entitled to access legal counsel.¹⁰⁰ At the time of writing, several cases challenging externalization measures remain pending before domestic courts, including in Ghana and the United States.¹⁰¹

96. International Covenant on Civil and Political Rights, art. 2 (1) and (3); and Human Rights Committee, general comment No. 31 (2004), para. 8. ↩
97. Sala Constitucional Protege Derechos Fundamentales de Personas Migrantes Deportadas a Costa Rica, 24 June 2025. ↩
98. A. Natale and F. Filice, Nota ai provvedimenti di rigetto delle richieste di convalida dei trattenimenti disposti dalla Questura di Roma ai sensi del Protocollo Italia-Albania, emessi dal Tribunale di Roma, sezione specializzata nella protezione internazionale, Questione Giustizia, 18 October 2024. ↩
99. Federal Court of Australia, Plaintiff S99/2016 v. Minister for Immigration and Border Protection, FCA 483, File No. VID 305 of 2016, judgment of 6 May 2016. ↩
100. See https://www.reuters.com/world/africa/eswatini-court-rules-first-trump-deportees-jail-have-right-lawyer-2026-04-10/. ↩
101. U.T. v. Bondi, United States District Court for the District of Columbia, No. 1:20-cv-00116;D.A. et al. v. Noem, United States District Court for the District of Columbia, No. 1:25-cv-03135;D.V.D. v. United States Department of Homeland Security, 25-1393, (1st Cir.); and see https://www.jurist.org/features/2025/11/11/explainer-does-ghanas-deportation-deal-with-the-us-breach-constitutional-procedure/#. ↩

Given the multiplicity of actors involved in externalization measures, questions arise as to the attribution of conduct and the allocation of responsibility where human rights violations occur. As discussed in the Special Rapporteur’s previous report,¹⁰² these matters are governed by the articles on responsibility of States for internationally wrongful acts, in respect of States, and by the articles on the responsibility of international organizations, in 92 Submissions from Brazil and from K. Allinson. 93 Submission from Ecuador. 94 Submissions from Human Rights Advocates and from the Kennedy Human Rights Center. 95 A/80/302, paras. 49–59. 96 International Covenant on Civil and Political Rights, art. 2 (1) and (3); and Human Rights Committee, general comment No. 31 (2004), para. 8. 97 Sala Constitucional Protege Derechos Fundamentales de Personas Migrantes Deportadas a Costa Rica, 24 June 2025. 98 A. Natale and F. Filice, Nota ai provvedimenti di rigetto delle richieste di convalida dei trattenimenti disposti dalla Questura di Roma ai sensi del Protocollo Italia-Albania, emessi dal Tribunale di Roma, sezione specializzata nella protezione internazionale, Questione Giustizia, 18 October 2024. 99 Federal Court of Australia, Plaintiff S99/2016 v. Minister for Immigration and Border Protection, FCA 483, File No. VID 305 of 2016, judgment of 6 May 2016. 100 See https://www.reuters.com/world/africa/eswatini-court-rules-first-trump-deportees-jail-have-rightlawyer-2026-04-10/. 101 U.T. v. Bondi, United States District Court for the District of Columbia, No. 1:20-cv-00116; D.A. et al. v. Noem, United States District Court for the District of Columbia, No. 1:25-cv-03135; D.V.D. v. United States Department of Homeland Security, 25-1393, (1st Cir.); and see https://www.jurist.org/features/2025/11/11/explainer-does-ghanas-deportation-deal-with-the-usbreach-constitutional-procedure/#. 102 A/80/302, paras. 54–59. GE.26-06322 respect of international organizations. Under the primary rule of attribution under the articles on responsibility of States for internationally wrongful acts (art. 4), conduct is attributable to a State where it is carried out by its organs. Among other bases for attribution, conduct may also be attributable to a State where it is carried out by entities empowered by law to exercise elements of governmental authority (art. 5), or by persons or groups acting on the State’s instructions or under its direction or control (art. 8). This may include private actors, such as security companies or air carriers, where they are entrusted with functions of a public character normally exercised by State organs.¹⁰³ Parallel rules on attribution apply in respect of international organizations, under the articles on the responsibility of international organizations.

102. A/80/302, paras. 54–59. ↩
103. At the same time, private companies themselves have independent responsibilities to respect human rights, in line with the Guiding Principles on Business and Human Rights, including avoiding infringing on the human rights of others (principle 11) and avoiding causing or contributing to adverse human rights impacts (principle 13). ↩

Of particular relevance in the context of externalization, State responsibility is not necessarily exclusive. Depending on the nature and degree of their involvement, one or more States may incur responsibility for violations of international law arising in the course of such arrangements. States may be responsible for their own conduct, including where it occurs alongside, but independently of, that of other actors. Where cooperation is more integrated, multiple States or international organizations may incur joint responsibility¹⁰⁴ or shared responsibility¹⁰⁵ for the same internationally wrongful act. Even where direct responsibility cannot be established, a State may incur responsibility for its contribution to a wrongful act attributable to another State. This is particularly relevant in externalization arrangements aimed at preventing arrivals (see para. 9 above), where the involvement may take the form of financing, training or the provision of equipment, and where extraterritorial jurisdiction is not established. In such circumstances, a State may incur responsibility for aiding or assisting, directing or controlling, or coercing another State or an international organization in the commission of an internationally wrongful act, where it does so with knowledge of the circumstances and is bound by the relevant international obligation.¹⁰⁶

104. Articles on responsibility of states for internationally wrongful acts, art. 47; and articles on the responsibility of international organizations, art. 48. ↩
105. André Nollkaemper et al., “Guiding principles on shared responsibility in international law”, European Journal of International Law, vol. 31, issue 1 (February 2020), available at https://academic.oup.com/ejil/article/31/1/15/5882075, pp. 15–72. ↩
106. Submission from K. Allinson. ↩

A key challenge in establishing responsibility in the context of externalization measures lies in establishing the jurisdiction of the States concerned so as to engage their human rights obligations. Under international human rights law, States are required to respect and ensure the rights of all persons within their territory or subject to their jurisdiction.¹⁰⁷ In the context of externalization, third States act within their territorial jurisdiction, and it is generally uncontroversial that they are bound by their human rights obligations. The African Commission on Human and Peoples’ Rights has further stressed that States Parties to the African Charter on Human and Peoples’ Rights must take all necessary measures to protect the life, dignity, security and fundamental rights of all persons within their jurisdiction, including migrants transferred pursuant to bilateral or multilateral agreements.¹⁰⁸ This underscores that international cooperation arrangements do not dilute States’ human rights obligations.

107. Human Rights Committee, general comment No. 31 (2004), para. 10. ↩
108. African Commission on Human and Peoples’ Rights, Resolution on the obligations of African States in the context of the externalization of migration governance and the extraregional transfer of migrants to Africa, ACHPR/Res.645 (LXXXV) 2025, 30 October 2025. ↩

For externalizing States, however, it may be necessary to demonstrate that individuals outside their territory are nevertheless within their power or effective control, in order to establish extraterritorial jurisdiction and engage their human rights obligations.¹⁰⁹ Concerning extraterritorial asylum processing (see para. 11 above) and transfers to third States (see para. 14), the jurisprudence of United Nations treaty bodies regarding Australia’s arrangements with Nauru and, previously, with Papua New Guinea, is instructive. In a series of cases, including M.I. et al. v. Australia, Nabrahi v. Australia and A.A. v. Australia, the Human Rights Committee and the Committee against Torture rejected the arguments made by Australia that it did not exercise effective control over detention facilities in Nauru and Papua New Guinea. Both Committees found that Australia’s funding, joint management, including monitoring, and contracting of service providers for the construction and operation of detention centres amounted to a sufficient degree of control and influence to establish jurisdiction under the International Covenant on Civil and Political Rights and the Convention against Torture.¹¹⁰

109. Human Rights Committee, general comment No. 31 (2004), para. 10. ↩
110. M.I. et al. v. Australia (CCPR/C/142/D/2749/2016), paras. 9.5–9.9; Nabhari v. Australia (CCPR/C/142/D/3663/2019), paras. 7.6–7.15; and A.A. v. Australia (CAT/C/83/D/1079/2021),paras. 7.6–7.8. ↩

With regard to measures aimed at preventing entry (see para. 9 above), the degree of involvement of externalizing States varies along a spectrum; however, even where such involvement is less direct, responsibility may arise where the conduct or support provided has a foreseeable impact. In relation to pushback measures, the European Court of Human Rights held in the Hirsi case that Italy exercised jurisdiction by intercepting migrants on the high seas and transferring them to Libya because it exercised continuous and exclusive control over the individuals, notwithstanding that they were outside its territory.¹¹¹ Following Hirsi, Italy shifted from direct pushbacks to cooperation with the Libyan Coast Guard, which intercepts migrants at sea and returns them to Libya, where they face a real risk of serious human rights violations, with reduced Italian involvement (so-called “pullbacks”) (see para. 10). This approach was examined by the European Court of Human Rights in the case of S.S., where the Court found that the role of Italy in coordinating and providing financial and logistical support to the Libyan Coast Guard was insufficient to establish extraterritorial jurisdiction under the European Convention on Human Rights.¹¹² This finding does not, however, preclude the possibility that different forms or degrees of involvement may give rise to such jurisdiction. In this regard, the Human Rights Committee, in the case of A.S., concerning a shipwreck resulting in approximately 200 deaths, found that Italy exercised extraterritorial jurisdiction, as the individuals were within its power or effective control due to a special relationship of dependency.¹¹³

111. European Court of Human Rights, Hirsi Jamaa and Others v. Italy (application No. 27765/09), judgment of 23 February 2012, paras. 70, 74 and 81. ↩
112. European Court of Human Rights, S.S. and Others v. Italy (application No. 21660/18), decision on admissibility of 20 May 2025, paras. 100–108. ↩
113. A.S. et al. v. Italy (CCPR/C/130/D/3042/2017), para. 7.8. ↩

Limited transparency is a recurring feature of externalization cooperation,¹¹⁴ due to its extraterritorial nature and the involvement of multiple actors. Arrangements often rely on informal and opaque instruments that escape parliamentary scrutiny and public debate, sometimes supplemented by non-public technical agreements. Complex funding mechanisms and multi-actor partnerships further obscure responsibility and weaken accountability, while increased use of surveillance technologies may hinder oversight. Strengthening transparency is therefore essential to reduce human rights risks, ensure accountability, and uphold the rule of law.

114. A/80/302, paras. 42–45. ↩

Human rights risk assessments can strengthen transparency and accountability in migration cooperation, particularly when their findings are made public or shared with oversight bodies.¹¹⁵ A tool to operationalize States’ positive obligation to prevent human rights violations, human rights risk assessments are understood as systematic processes for identifying and evaluating the potential and actual impacts of policies or projects on the enjoyment of human rights, with a view to preventing or mitigating adverse effects and informing decision-making in line with States’ human rights obligations.¹¹⁶ Special procedures of the Human Rights Council and treaty bodies regularly call upon States to conduct human rights impact assessments in the context of international trade and investment agreements.¹¹⁷ In the same vein, migration cooperation agreements, given their inherent risks to migrants’ human rights, should be subject to ex ante and periodic human rights risk assessments. The need to conduct an impact assessment is particularly acute where non-State actors, including business enterprises, are involved in the implementation of projects under such arrangements, as their involvement may further diffuse responsibility and exacerbate existing accountability gaps. In line with their human rights due diligence responsibilities as set out in the Guiding Principles on Business and Human Rights, business enterprises should likewise carry out their own risk assessments.¹¹⁸

115. A/80/302, para. 46; and European Ombudsman, Decision on how the European Commission intends to guarantee respect for human rights in the context of the European Union-Tunisia Memorandum of Understanding (OI/2/2024/MHZ), 21 October 2024, para. 37. ↩
116. Guiding Principles on Business and Human Rights, principle 17; and see A/HRC/19/59/Add.5. ↩
117. A/HRC/19/59/Add.5, para. 2. ↩
118. Guiding Principles on Business and Human Rights, principle 17. ↩

Human rights risk assessments should include clear, publicly available criteria for suspending contracts and funding in the event of violations.¹¹⁹ Where such assessments identify a real risk of breaching human rights obligations, migration cooperation agreements should incorporate safeguards to ensure that these risks are mitigated effectively. For example, European Union financial instruments supporting migration management in third countries, including the Neighbourhood, Development and International Cooperation Instrument, are formally required to comply with human rights standards, the rule of law and due diligence standards. However, available information suggests that certain European Union-funded migration activities in Libya and Tunisia have not been subject to comprehensive human rights risk assessments, nor have they identified adequate mitigation measures.¹²⁰

119. European Ombudsman, Decision on how the European Commission intends to guarantee respect for human rights in the context of the European Union-Tunisia Memorandum of Understanding (OI/2/2024/MHZ), para. 38. ↩
120. Ibid., conclusion. ↩

Human rights monitoring should be integral to any migration and border governance, in particular when elements of it are externalized. Human rights monitoring is a central tool in identifying and documenting human rights violations and abuses, understanding their patterns and causes, developing possible solutions, contributing to transparency and accountability and helping to prevent further harm. To do so, monitors should be institutionally, financially and personally independent of State authorities and be endowed with an adequate budget, resources, and a mandate.¹²¹ It is essential to ensure that all migration cooperation agreements and arrangements are subject to genuinely independent monitoring mechanisms. Such mechanisms should draw on a combination of national human rights institutions and oversight bodies in the States concerned, as well as civil society organizations with relevant expertise. Clear provisions should be included to guarantee access to relevant locations and affected populations, including through the facilitation of entry visas for monitors. Adequate funding for such monitoring should be ensured, including through support provided by externalizing States.¹²²

121. A/HRC/60/67, paras. 4, 5, 19–23 and 47–50. ↩
122. Submission from Quaker United Nations Office; and see European Union Agency for Fundamental Rights, “Planned return hubs in third countries: EU fundamental rights law issues”, 6 February 2025, para. 124. ↩

Monitoring border areas where pushbacks and interceptions are likely to occur can significantly enhance transparency and accountability. To be effective, border monitoring mechanisms should be granted unhindered access to border areas, including checkpoints and places of detention, and be able to observe operations, including border surveillance and search-and-rescue activities.¹²³ Where individuals are transferred to third States, arrangements should ensure continuity of monitoring. As an example of monitoring activities within third States, the European Union introduced a third-party monitoring framework for European Union-funded migration cooperation projects in 2019, intended to strengthen situational awareness in complex operational contexts where direct access may be limited.¹²⁴ Third-party monitoring framework initiatives have since been implemented in Libya and Tunisia, however their effectiveness remains in question. Available information suggests that European Union monitoring systems have not consistently provided reliable information on human rights risks or ensured adequate follow-up.¹²⁵ Concerns also persist regarding the limited scope and transparency of monitoring, including its focus on project compliance, and the lack of public reporting.

123. European Union Agency for Fundamental Rights, “Monitoring fundamental rights during screening and the asylum border procedure – a guide on national independent mechanisms”, 19 September 2024. ↩
124. See https://spcommreports.ohchr.org/TMResultsBase/DownLoadFile?gId=39484. ↩
125. European Court of Auditors, Special report 17/2024, “The EU trust fund for Africa – Despite new approaches, support remained unfocused”, 25 September 2024, para. 55. ↩

Complaint mechanisms are essential for enhancing transparency and ensuring accountability, complementing human rights impact assessments and monitoring.¹²⁶

126. A/80/302, para. 48. ↩

They may be integrated into existing frameworks or established as stand-alone procedures, and should be accessible and effective. Mechanisms should be open to anyone alleging a violation, regardless of location, and individuals must be informed of their availability. Procedures should be simple and user-friendly, with safeguards against retaliation and the possibility of third-party complaints where patterns of violations exist. Measures should address practical barriers, such as geographic and linguistic obstacles. Complaints must be examined promptly and impartially by independent bodies, leading to effective remedies and, where appropriate, sanctions.¹²⁷

127. See https://cdn.ceps.eu/wp-content/uploads/2018/03/Complaint Mechanisms_A4.pdf, pp. 10–13; and OHCHR and Global Migration Group, Principles and Guidelines, Supported by Practical Guidance, on the Human Rights Protection of Migrants in Vulnerable Situations, principle 17, para. 4. ↩

At the European Union level, several mechanisms offer useful practices for the establishment of complaint mechanisms for persons subject to externalization measures. The Frontex complaints mechanism allows individuals who consider that their fundamental rights have been violated during its operational activities, or their representatives, to submit complaints directly to the agency.¹²⁸ However, a key limitation is that complaints are handled through an internal process rather than by an independent body. More recently, the European Union has introduced an internal procedure to address human rights complaints in European Union-funded projects. This procedure provides for the verification of complaints and, where justified, the activation of suspension or termination clauses.¹²⁹ While this constitutes a positive development, its effectiveness in ensuring accountability and redress would be enhanced by the involvement of independent bodies and the establishment of clear criteria for the suspension of funding. Concerns have also been raised regarding its accessibility and transparency.

128. See https://www.frontex.europa.eu/fundamental-rights/frontex-complaints-mechanism/. ↩
129. See https://www.europarl.europa.eu/doceo/document/E-10-2025-004302-ASW_EN.html. ↩

The Special Rapporteur notes with concern the increasing reliance on the externalization of migration control. Despite concerns raised in the Special Rapporteur’s previous report,¹³⁰ new agreements and arrangements continue to be 123 European Union Agency for Fundamental Rights, “Monitoring fundamental rights during screening and the asylum border procedure – a guide on national independent mechanisms”, 19 September 2024. 124 See https://spcommreports.ohchr.org/TMResultsBase/DownLoadFile?gId=39484. 125 European Court of Auditors, Special report 17/2024, “The EU trust fund for Africa – Despite new approaches, support remained unfocused”, 25 September 2024, para. 55. 126 A/80/302, para. 48. 127 See https://cdn.ceps.eu/wp-content/uploads/2018/03/Complaint Mechanisms_A4.pdf, pp. 10–13; and OHCHR and Global Migration Group, Principles and Guidelines, Supported by Practical Guidance, on the Human Rights Protection of Migrants in Vulnerable Situations, principle 17, para. 4. 128 See https://www.frontex.europa.eu/fundamental-rights/frontex-complaints-mechanism/. 129 See https://www.europarl.europa.eu/doceo/document/E-10-2025-004302-ASW_EN.html. 130 A/80/302. GE.26-06322 concluded and operations continue to expand. Externalization agreements and arrangements carry a high risk of human rights violations and have severe impacts on persons in vulnerable situations, including children, entail gendered risks, and pose heightened risks for groups exposed to intersectional discrimination.

130. A/80/302. ↩

Externalization also raises broader ethical and rule-of-law concerns. Differences in bargaining power between the States involved may enable the use of pressure in the conclusion of such arrangements. The increasing use of development funding to support migration cooperation may be at variance with the objective of promoting sustainable development and may prove counterproductive in the longer term. Similar effects may result from restrictions on regional free-movement regimes essential to sustaining livelihoods. Civil society actors who support migrants subject to externalization measures may also face reprisals, particularly where migration cooperation is framed in political or security terms. At the same time, the persistent lack of transparency, including reliance on informal, opaque arrangements, risks undermining democratic oversight and eroding the rule of law.¹³¹

131. Submissions from Brot für die Welt; EuroMed Rights and others; and European Council on Refugees and Exiles. ↩

The Special Rapporteur recalls that States remain bound by their obligations under international human rights law when designing, financing and implementing migration cooperation, including where measures are carried out extraterritorially or through third States. States cannot avoid responsibility by contracting out their obligations to other States, international organizations or private entities. Rather than seeking to outsource responsibilities through arrangements that create human rights risks and reputational and financial costs, States should prioritize rights-based, fair and sustainable asylum and migration systems that fully respect human rights and related procedural safeguards. This includes moving away from broad reliance on “safe country” concepts, ensuring access to international protection in line with international human rights and refugee law, and creating genuine responsibility-sharing mechanisms. As removal is not the only measure to address the situation of persons in an irregular situation, greater emphasis should be placed on developing and implementing rights-based and sustainable alternatives, such as safe and regular pathways for admission and stay, including regularization.¹³² In doing so, States should recognize migrants’ sociocultural, civic-political and economic contributions to society.¹³³

132. OHCHR, Leveraging regular migration pathways for human rights, 2025. ↩
133. See A/HRC/56/54. ↩

While the recommendations set out in the previous report remain valid,¹³⁴ the Special Rapporteur seeks to further clarify and expand upon them. In order to uphold their human rights obligations in the context of migration cooperation, States should: (a) Ensure that migration cooperation fully respects the international human rights obligations of all States involved and that all cooperation instruments include explicit, binding and enforceable human rights and non-discrimination clauses; (b) Respect, protect and fulfil the human rights of all migrants and refugees under their jurisdiction, including when exercised extraterritorially, and prevent and stop any activity resulting in a violation of the right to leave any country, of the prohibition of refoulement, collective expulsion, arbitrary detention, torture, ill-treatment, forced labour and enforced disappearance, in racial discrimination, in violation of the right to life, to due process and to an effective remedy, in violation of the right to family life, or of economic and social rights, or in violation of the rights to privacy and data protection; (c) Refrain from any measure leading to direct or indirect violations of the principle of non-refoulement, ensure that the “safe country” concept is applied only on the basis of an individual and thorough assessment of safety and a meaningful 131 Submissions from Brot für die Welt; EuroMed Rights and others; and European Council on Refugees and Exiles. 132 OHCHR, Leveraging regular migration pathways for human rights, 2025. 133 See A/HRC/56/54. 134 A/80/302, para. 63. GE.26-06322 connection, and guarantee access to effective remedies with automatic suspensive effect, including where the safe third country concept is applied; (d) Guarantee that the use, transfer and deployment of digital and surveillance technologies comply with international human rights law, including the right to privacy and data protection standards and principles, ensuring transparency and independent oversight, and suspending the use of technologies where risks cannot be effectively mitigated, and consider establishing standards for the use of technology in migration governance, including the implementation of United Nations principles and guidelines on human rights-based digital border governance; (e) Ensure that all measures respect the rights of the child, with the best interests of the child as a primary consideration, by explicitly exempting children from externalization measures, prohibiting immigration detention, ensuring individualized assessments with full procedural safeguards, stopping and preventing any violence against children, reinforcing child protection systems, avoiding any family separation, and refraining from removing children where there is a real risk of irreparable harm; (f) Ensure that all measures are gender-responsive, by systematically identifying and preventing gender-based risks, including violence, exploitation and discrimination, guaranteeing access to gender-sensitive and trauma-informed procedures and services, refraining from removal where there is a real risk of gender-based violence or irreparable harm, and ensuring that detention and reception conditions respect the dignity, safety and specific needs of women, girls and persons with diverse sexual orientation, gender identity, gender expression and sex characteristics; (g) Prohibit racial profiling and discriminatory enforcement practices and ensure that all measures comply with human rights norms, and adopt an intersectional approach that prevents and addresses compounded discrimination against racialized groups; (h) Ensure transparency in migration cooperation, including by making agreements and their implementation frameworks publicly available and subject to appropriate parliamentary and public oversight; (i) As part of their human rights due diligence obligations, conduct comprehensive gender-sensitive human rights impact assessments, on an ex ante and an ongoing basis, of any cooperation, with specific attention to intersecting forms of exclusion affecting migrants, identifying human rights risks and risk mitigation strategies, and make the results publicly available and act upon them; (j) Ensure that private companies involved in migration cooperation exercise human rights due diligence in line with the Guiding Principles on Business and Human Rights, including by identifying, preventing and mitigating adverse impacts, establish accountability frameworks, enable access to effective remedy for victims, and suspend or terminate cooperation where companies contribute to or are linked to abuses; (k) Ensure independent, effective, gender-responsive and child-sensitive human rights monitoring of all measures preventing arrivals, outsourcing asylum processing and removing migrants to third States, ensuring an adequate mandate, access and resources, and meaningful follow-up to the findings of the monitoring; (l) Guarantee access to independent, accessible and effective complaint mechanisms to safely and confidentially report violations and abuse in the context of migration cooperation measures, guaranteeing access from abroad, protecting complainants, and promptly and effectively investigating well-founded allegations and ensuring that those responsible are brought to justice, and that victims and their relatives have access to effective remedies, including reparation; GE.26-06322 (m) Ensure that the use of development funding for migration-related objectives is consistent with its primary development purpose and supports sustainable development, and ensure that migration cooperation supports regional stability and mobility arrangements, and economic livelihoods, and genuinely benefits all countries involved and their populations; (n) Protect human rights defenders and civil society actors assisting migrants subject to externalization from reprisals and support their activities; (o) Create and expand pathways for admission and stay, including regularization programmes and mechanisms, that are based on humanitarian, human rights or other grounds, recognizing that these pathways are an effective tool for preventing and addressing situations of vulnerability for migrants.

134. A/80/302, para. 63. ↩

The Special Rapporteur urges international organizations and NGOs implementing projects in externalization contexts to exercise human rights due diligence, in line with their commitments and the applicable codes of conduct, ensure that their activities do not contribute to or benefit from human rights violations, and take appropriate action, including disengagement, where risks cannot be mitigated effectively.

Finally, the Special Rapporteur urges States to support, through the Human Rights Council, the establishment of an independent international human rights monitoring mechanism mandated to monitor, investigate and report on human rights violations and abuses arising from migration governance at international borders and in transit, including those linked to externalization practices, with a view to contributing to prevention, accountability, and access to remedy.