The Safe Country of Origin Concept and the New Pact on Migration and Asylum: Streamlining or Sidelining?
By Camila Nogueira Crispim
Edited by Veronica Corcodel and Dimitra Fragkou
Abstract: Within EU asylum law, the safe country of origin concept has sparked controversy due to its procedural impact and potential for discrimination. This blog post considers the discriminatory risks inherent in this concept and critically assesses whether the recently adopted EU New Pact on Migration and Asylum exacerbates or mitigates those risks.
Keywords: Common European Asylum System (CEAS), Safe Country of Origin (SCO), Country of Origin Information (COI), Recognition Rates, New Pact on Migration and Asylum.
I. Introduction
Within the Common European Asylum System (CEAS), the safe country of origin (SCO) concept constitutes a ground for accelerated asylum procedures, having evolved from an optional to a mandatory mechanism with the 2024 New Pact on Migration and Asylum (the New Pact). Such an accelerated process often sacrifices fairness, prioritising fast-tracking over the accurate assessment of asylum applications.[1] While the designation of a country as ‘safe’ is not legally based on low recognition rates, in practice, these might be used as a proxy, thereby perpetuating a cycle that further reduces the likelihood of applicants from listed countries having their claims approved.[2]
This concept can also obscure another underlying motive and tendency: the implementation of increasingly restrictive migration and asylum policies within the European Union (EU). The New Pact confirms this tendency and has already given rise to numerous critiques. This blogpost assesses some of the changes introduced by the New Pact in relation to the concept of safe country of origin but also takes stock with the concept’s overall potentially discriminatory nature.
II. SCO: A Concept Amounting to Discrimination?
According to Annex I of the recast Asylum Procedures Directive (APD),[3] a safe country of origin is a country where it can be demonstrated that “(…) there is generally and consistently no persecution (…), no torture or inhuman or degrading treatment or punishment and no threat by reason of indiscriminate violence in situations of international or internal armed conflict.”
When a country is broadly listed as safe by Member States (MS) pursuant to Article 37(1),[4] a presumption of safety is established for all applications of international protection from that country, including stateless individuals who were residents there.[5] This produces two effects:[6] procedurally, it can trigger the accelerated or border procedure under Articles 31(8)(b)[7] and 43(1)(b) of the recast APD; substantively, applications are expected to be considered unfounded,[8] unless such presumption of safety is rebutted based on the applicant’s specific circumstances.[9]
An important concern in relation to this concept is that generalised assumptions may be self-perpetuating, reinforcing a country’s presumed safety while concealing the rejection of valid claims and potentially resulting in refoulement.[10] Such a broad approach compromises the required individualised assessment, risking the rejection of genuine claims and the misrepresentation of complex human rights realities.[11]
This raises the question of whether this practice is compatible with international refugee law and anti-discrimination standards. Costello argues that the way the SCO concept is implemented potentially violates Article 3 of the Refugee Convention,[12] which guarantees access to asylum procedures “without discrimination as to race, religion or country of origin.”[13] Although the EU as such is not a party to the Refugee Convention, the obligation for the EU to comply with international refugee law is recognized in article 78(1) of the Treaty on the Functioning of the EU. Thus, not only does the collective logic of the SCO concept contradict the individualized nature of the refugee status determination, but it also raises concerns of discrimination based on the country of origin, both being in contradiction with international refugee law.[14]
Discrimination is also prohibited in the Charter of Fundamental Rights of the EU[15], although scholars generally agree that article 21(2), specifically dedicated to discrimination on the ground of nationality, does not apply to third country nationals.[16] Moreover, in the H.I.D. case, while the CJEU acknowledged the importance of procedural safeguards in asylum procedures, it ultimately granted Member States considerable discretion to apply accelerated procedures based on the applicant’s country of origin, without elaborating on discrimination challenges.[17] A more promising argument would perhaps rely on a combination of articles 3 and 14 of the European Convention on Human Rights before the European Court of Human Rights. While the ECtHR hasn’t had yet the opportunity to decide on such a claim, the concept’s exclusive focus on the country of origin precludes a priori an entire group of asylum-seekers, typically sharing the same nationality, from being considered for refugee status with the same individualized assessment characterizing the regular asylum procedure, creating a substantial risk of refoulement to which courts should not be indifferent.
The challenges produced by this concept are exacerbated by the fact that the standard for rebutting the presumption of safety is excessively burdensome on the applicant, who has limited access to quality legal assistance in many MS.[18] Despite long-standing recommendations to lower the standard of proof from “serious grounds for considering the country of origin not to be safe” to “reasonable grounds”,[19] the high standard of proof persists in the New Pact for the rebuttal of the presumption of safety.
Moreover, while the Pact introduces the possibility to exempt certain groups or regions within a country from the safe designation, this change might inadvertently entrench ‘new’ forms of discrimination.[20] Selective recognition of safety may result in discriminatory effects based on ethnic origin or other social group characteristics. Moreover, while such move may seem to aim at refining protection, it also risks localizing the risk of serious human rights violations in ways that conceal the rejection of valid claims.
The following part will provide some national examples of safe country of origin implementation, highlighting the issue of inconsistencies between countries that exacerbates the discriminatory potential of the concept.
III. The Issue of Divergent National Practices
On September 2023, the Ireland High Court rejected the appeal of a Georgian national (MZ) for international protection based on COI compiled by Irish authorities, which designated Georgia as a SCO. For the High Court, the Tribunal’s decision was based on a correct application of law to the facts.[21]
The Tribunal found that MZ had a well-founded fear of persecution and would face a genuine risk of torture or inhuman treatment by criminals if he returned to Georgia, but that state protection was available in Georgia.[22] The Tribunal also noted that Ireland deemed Georgia a SCO under the International Protection Act 2015[23] (Safe Countries of Origin) by order of 2018.[24] It also referred to the COI by the United States State Department Country Report of 2021 and a Human Rights Watch report[25] to conclude that state protection is generally available in Georgia and thus to the applicant as well.[26]
This case illustrates the issue of outdated SCO lists and the inconsistency in MS’ national designations. Ireland’s list has not been updated since 2018,[27] disregarding ECRE’s repeated warnings to states that the protection needs of asylum seekers must be based on an updated assessment of their country of origin.[28] Costello points out that human rights situations can change rapidly, undermining the credibility of SCO lists that are considered as ‘accurate’.[29]
Also, this echoes the question, ‘Safe Country? Says Who?’[30] if one considers that Belgium removed Georgia from its SCO list in March 2023.[31] How can it be objectively considered a safe country by one member state but not by another? National lists differ significantly, highlighting the likely politicisation of a practice that undermines the objectivity of the assessment.[32] In this sense, they might be included or excluded based on foreign policy objectives, exacerbating the discriminatory potential of the concept.[33]
In contrast to the stance taken by the High Court of Ireland, the Tribunal of Florence, also on September 2023, disapplied the Ministerial Decree of March 17, 2023, determining that Tunisia can no longer be regarded as SCO based on updated COI.[34] This decision came from a Tunisian national’s appeal against the rejection of his international protection application, arguing that Tunisia could not be qualified as safe.[35]
Despite not providing evidence regarding his specific situation, the tribunal emphasised that judicial authorities have a duty to verify the correctness of a country’s inclusion on the SCO list, even if the administrative body’s decision meets the criteria set by law.[36] It found that the administrative body only partially considered some of the socio-political crises mentioned by the applicant and was wrong in not considering updated COI sources, in light of which the Ministerial Decree on the security situation in Tunisia had to be disapplied.[37]
Hence, the tribunal reconsidered Tunisia’s presumption of safety based on up-to-date COI sources and a report from the Commissioner for Human Rights of the Council of Europe on a Memorandum of Understanding between the EU and Tunisia. It highlights the worsening of crises in Tunisia and the significant changes in the human rights situation.[38] Consequently, the tribunal concluded that the Ministerial Decree must be disapplied for failing to meet the criteria of the recast APD and suspended the previous negative decision issued to the applicant.[39]
Both cases highlight the inconsistency of SCO designations, not only across the EU, but also within national systems. The lack of uniformity in safety assessments under the CEAS, as highlighted by the Irish case, makes it clear that what constitutes a safe country might not be objective, thus exacerbating the risks of discrimination. This undermines the credibility and fairness of the SCO mechanism as a whole. Hence, the approach taken by the Florence Tribunal – which directly challenged an executive stance – illustrates the critical importance of thoroughly considering updated COI. The Florence Tribunal decision, however, also illustrates how administrative bodies might in practice rush into decisions based on existing SCO lists, leaving it ultimately to the applicant, who is in a weaker position, to challenge the designation of a country of origin as safe.
Even without an official SCO list, MS like Portugal have reportedly applied the concept informally, with applicants from specific countries such as Gambia and Senegal being informed that their nationality negatively impacts their chances of receiving a positive asylum decision.[40] In this sense, the risk of discrimination might exist even without SCO lists, although the concept as such facilitates a discriminatory reasoning.
The New Pact introduces some steps toward harmonisation through the prospect of adopting common EU lists for the SCO concept, prompting the question of whether it might remedy the issue of inconsistencies between member states. The Pact builds on previous (failed) efforts made in that direction and allows such lists to be adopted under the ordinary legislative procedure, subject to compliance with safety criteria and regular review by the Commission.[41] The EU Asylum Agency provides relevant country information, and Member States can propose new countries for inclusion, a fast-track mechanism being foreseen for sudden changes in country conditions. National lists can continue, however, to exist alongside EU lists, but Member States may not list a country suspended by the EU. If a country is removed from the EU list, the Commission can block its reintroduction on national lists for two years. While these changes promise some improvements for the issue of inconsistencies between national lists, whether this system truly ensures harmonisation remains debatable.[42] Moreover, the concept as such and the controversies around potential discrimination would remain in place. In addition, a new ground of accelerated procedure discussed below, formally distinct from the SCO concept but nevertheless assuming ‘safety’ based on recognition rates, will likely exacerbate the overall issue of discriminatory practices in asylum procedures.
IV. Asylum Recognition Rates: A Disguised ‘Safety’ Ground?
The New Pact does not seem to offer significant solutions to the challenges outlined in the previous sections. Instead, it exacerbates these challenges through the manadatory use of accelerated procedures for applicants from countries with an EU-wide recognition rate of 20% or lower, a newly added ground for such procedures.[43] As Steve Peers notes, this integrates hard-wiring statistical assumptions into legal procedures, further prioritizing control and fast-tracking over individualized justice.[44] As noted above, this new ground for accelerated procedures is formally distinct from the SCO concept, but the assumption behind it does rejoin a general idea of presumed safety, although not as defined in article 61 of the Asylum Procedure Regulation. A similar understanding of safety, based on quantitative thresholds of low acceptance rates or high rates of withdrawal of claims, used to be part of the Canadian legal system until it was found unconstitutional, particularly on equality and non-discrimination grounds.[45]
In theory, exceptions are made if the country of origin has significantly changed since the data was published or if the applicant belongs to a category for whom such a rate is not representative of their protection needs.[46] However, it remains to be seen how effectively these exceptions will be implemented, as past practices indicate that COI is not consistently updated in a timely or adequate manner to reflect evolving conditions.
The EU’s attempt to harmonize asylum procedures under the New Pact risks codifying discriminatory practices under a veneer of expediency. The SCO concept, already problematic, is now being complemented by recognition rate thresholds that reinforce nationality-based procedural shortcuts. It is highly probable that this will allow for a high number of rapid removals of migrants in a discriminatory manner.[47] This hierarchical categorisation of migrants based on their nationality and the associated recognition rates, denies them the opportunity to engage with the system as human beings in need of international protection.[48]
V. Conclusion
This blog post has discussed some of the major challenges and controversies surrounding the SCO concept within the CEAS. It has been argued that this concept can lead to discrimination, since it excludes a priori a group of asylum seekers, based on their country of origin, from proper individualized assessment.
It has also been argued that the application of SCO lists might be inconsistent, with political motivations possibly influencing safety designations or the lack of removals from such lists. This exacerbates the risk of discriminatory practices. Although the New Pact reintroduces the prospect of adopting common EU lists for the SCO concept, whether this system would truly ensure harmonization remains debatable.
This blog post has also contended that the New Pact exacerbates the risk of discriminatory practices in asylum procedures through the addition of a new ground of accelerated procedures, formally distinct from the SCO concept but carrying with it an assumption of safety, for applicants from countries with an EU-wide recognition rate of 20% or lower.
In matters of asylum, where human lives and dignity are at stake, expediency should never replace justice. A truly fair asylum system must ensure that protection is determined not by where a person comes from, but by what they have endured and what they fear, through a proper individualized assessment and in line with international refugee law, as well as international and European human rights law.
HOW TO CITE THIS BLOG POST: Crispim, Camila Nogueira. “The Safe Country of Origin Concept and the New Pact on Migration and Asylum: Streamlining or Sidelining?” NOVA Refugee and Migration Clinic Blog, April 2025, available at: https://novarefugeelegalclinic.novalaw.unl.pt/?blog_post=the-safe-country-of-origin-concept-and-the-new-pact-on-migration-and-asylum-streamlining-or-sidelining
[1] Michael John-Hopkins, ‘The Emperor’s New Safe Country Concepts: A UK Perspective on Sacrificing Fairness on the Altar of Efficiency’ (2009) 21 International Journal of Refugee Law 218.
[2] Matthew Hunt, ‘The Safe Country of Origin Concept in European Asylum Law: Past, Present and Future’ (2014) 26 International Journal of Refugee Law 533.
[3] Directive 2013/32/EU of the European Parliament and of the Council of 26 June 2013 on common procedures for granting and withdrawing international protection (recast) https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32013L0032
[4] ibid, Article 37(1) states: “Member States may retain or introduce legislation that allows, in accordance with Annex I, for the national designation of safe countries of origin for the purposes of examining applications for international protection.”
[5] ibid, Article 36 (1)(a) and (b).
[6] European Union Agency for Asylum (EUAA), ‘EUAA Expert Panel: The Use of Safe Country Concepts in the Jurisprudence of European and National Courts and Tribunals’ (2023) 1-2.
[7] Directive 2013/32/EU, Article 31(8)(b) states: “Member States may provide that an examination procedure in accordance with the basic principles and guarantees of Chapter II be accelerated and/or conducted at the border or in transit zones in accordance with Article 43 if: the applicant is from a safe country of origin within the meaning of this Directive.”
[8] UNHCR, ‘Improving Asylum Procedures – Comparative Analysis and Recommendations for Law and Practice: Key Findings and Recommendations’ (2010) 65.
[9] EUAA, ‘EUAA Expert Panel’. See also Directive 2013/32/EU, Article 36(1).
[10] Cathryn Costello, ‘Safe Country? Says Who?’ (2016) 28 International Journal of Refugee Law 2. The principle of non-refoulement is enshrined in Article 33 of the Refugee Convention. Under international human rights law, this prohibition applies to any form of removal or transfer of individuals, irrespective of their legal status, when there are substantial reasons to believe that the person being returned would face a risk of severe harm, such as torture, ill-treatment, or other serious violations of human rights obligations. See more at https://www.ohchr.org/sites/default/files/Documents/Issues/Migration/GlobalCompactMigration/ThePrincipleNon-RefoulementUnderInternationalHumanRightsLaw.pdf
[11] Henry Martenson and John McCarthy, ‘“In General, No Serious Risk of Persecution”: Safe Country of Origin Practices in Nine European States’ (1998) 11 Journal of Refugee Studies 307.
[12] Cathryn Costello, ‘The Asylum Procedures Directive and the Proliferation of Safe Country Practices: Deterrence, Deflection and the Dismantling of International Protection?’ (2005) 7 European Journal of Migration and Law 50.
[13] United Nations, ‘Convention Relating to the Status of Refugees’ (28 July 1951) <https://www.ohchr.org/en/instruments-mechanisms/instruments/convention-relating-status-refugees> accessed 21 April 2025, Article 3.
[14] Matthew Hunt, ‘The Safe Country of Origin Concept in European Asylum Law’, 533.
[15] Charter of Fundamental Rights of the European Union, Article 21.
[16] Claire Kilpatrick and Hanna Eklund, ‘Article 21 – Non-discrimination’ in Steve Peers and others (eds), The EU Charter of Fundamental Rights (2nd edn, Nomos 2022).
[17] Cathryn Costello, ‘Safe Country? Says Who?’ (2016) 28 International Journal of Refugee Law 15. See Case C-175/11 HID and BA v Refugee Applications Commissioner and Others [2013] ECLI:EU:C:2013:45.
[18] European Council on Refugees and Exiles (ECRE), ‘ECRE Comments on the Commission Proposal for an Asylum Procedures Regulation COM(2016) 467’ (2016) 58.
[19] International Commission of Jurists (ICJ), ‘Compromising Rights and Procedures: ICJ Observations on the 2011 Recast Proposal of the Procedure Directive’ (2011) <https://www.icj.org/wp-content/uploads/2012/05/ICJ-asylum-procedure-directive-legal-submission-2011.pdf> accessed 21 April 2025, para 75.
[20] Steve Peers, ‘The New EU Asylum Laws: Taking Rights Half-Seriously’ (2024) Yearbook of European Law yeae003.
[21] EUAA Case Law Database, ‘MZ v International Protection Appeals Tribunal & Ors’ (September 2023) <https://caselaw.euaa.europa.eu/pages/viewcaselaw.aspx?CaseLawID=3812> accessed 21 April 2025.
[22] MZ v International Protection Appeals Tribunal & Ors (Ireland High Court 2023), paras 18-19.
[23] It must be noted that Ireland is not taking part in the adoption of the recast APD and is not bound by it or subject to its application for what a national legal framework of SCO applies.
[24] MZ v International Protection Appeals Tribunal & Ors, para 22. See<http://www.ipo.gov.ie/en/IPO/SI%20No%20121%20of%202018.pdf/Files/SI%20No%20121%20of%202018.pdf> accessed 21 April 2025.
[25] MZ v International Protection Appeals Tribunal & Ors, para 21. It is to be noted that the report’s date is not mentioned in the judgment under analysis.
[26] ibid,para 24.
[27] EUAA, ‘Who Is Who in International Protection in the EU+: Countries Applying the Concept of Safe Countries in the Asylum Procedure’ (2023) <https://whoiswho.euaa.europa.eu/Pages/safe-country-concept.aspx> accessed 21 April 2025.
[28] ECRE, ‘Information Note on Directive 2013/32/EU of the European Parliament and of the Council of 26 June 2013 on Common Procedures for Granting and Withdrawing International Protection (Recast)’ (December 2014) <https://www.refworld.org/reference/research/ecre/2014/en/103320> accessed 21 April 2025, 42-43.
[29] Cathryn Costello, ‘The Asylum Procedures Directive and the Proliferation of Safe Country Practices’, 66.
[30] Cathryn Costello, ‘Safe Country? Says Who?’ (2016) 28 International Journal of Refugee Law 2.
[31] EUAA, ‘4.3.2.1. Recent Changes in National Lists of Safe Countries’ (2023) <https://euaa.europa.eu/asylum-report-2023/4321-recent-changes-national-lists-safe-countries#footnote360_o54a249> accessed 21 April 2025. See also <https://news.belgium.be/fr/determination-des-pays-dorigine-surs> accessed 21 April 2025.
[32] Cathryn Costello, ‘The Asylum Procedures Directive and the Proliferation of Safe Country Practices’, 66.
[33] Henry Martenson and John McCarthy, ‘Safe Country of Origin Practices in Nine European States’, 307.
[34] EUAA Case Law Database, ‘Applicant v Ministry of the Interior’ (2023) <https://caselaw.euaa.europa.eu/pages/viewcaselaw.aspx?CaseLawID=3740> accessed 21 April 2025.
[35] Applicant v Ministry of Interior (Tribunal of Florence, 2023) 2-3.
[36] ibid,5-6 and 9.
[37] ibid, 10-11.
[38] ibid, 11-15.
[39] ibid, 15-16.
[40] AIDA, ECRE, ‘Safe Country of Origin – Asylum Information Database (5 July 2024) <https://asylumineurope.org/reports/country/portugal/asylum-procedure/the-safe-country-concepts/safe-country-origin/> accessed 21 April 2025.
[41] Steve Peers, ‘The New EU Asylum Laws’, yeae003.
[42] ibid.
[43] Regulation of the European Parliament and of the Council establishing a common procedure for international protection in the Union and repealing Directive 2013/32/EU (26 April 2024), Article 42(1)(j). See also EUAA, ‘Asylum Report 2024’ (14 June 2024) <https://euaa.europa.eu/publications/asylum-report-2024> accessed 21 April 2025, 186.
[44] Steve Peers, ‘The New EU Asylum Laws’, yeae003.
[45] YZ and the Canadian Association of Refugee Lawyers v Minister for Citizenship and Immigration 2015 FC 892 (CanLII). Cathryn Costello, ‘Safe Country? Says Who?’ (2016) 28 International Journal of Refugee Law.
[46] EU Regulation on common procedures for international protection (2024), Article 42(1)(j).
[47] InfoMigrants, ‘EU Approves New Migration Pact’ (12 April 2024) <https://www.infomigrants.net/en/post/56394/eu-approves-new-migration-pact> accessed 21 April 2025.
[48] ibid.
