Nationality and Borders Bill




1. Established in 2010, the Refugee Law Initiative is a specialised independent academic centre, based at the School of Advanced Study, University of London, which serves as the foremost platform for bringing together leading refugee law researchers and practitioners from across the world. Our evidence shares with the Committee our viewpoint, as subject matter specialists in refugee law, on selected provisions of the Nationality and Borders Bill that directly concern the law and protection of refugees and asylum-seekers.

2. We focus our evidence on the provisions of most acute concern to us and where our particular expertise may prove of greatest value, in particular: Clause 10 on the differential treatment of refugees; and Clauses 27-35 on the interpretation of the refugee definition and related provisions of the 1951 Convention relating to the Status of Refugees (hereafter ‘the Refugee Convention’). Our view is that clause 10 should be removed in its entirety from the Bill and that several problematic provisions of clauses 27-35 require amendment.


3. Clause 10 allows the UK to treat some refugees (Group 2) less favourably than others (Group 1). In essence, this provision creates a two-tier system of refugee status in which Group 2 refugees are disadvantaged in material ways as a sanction for the manner in which they sought asylum in the UK. However, as our analysis shows, the proposed clause raises a range of serious legal and policy concerns. Clause 10 should be removed in its entirety from the Bill.

Purpose of the clause

4. According to the Explanatory Notes to the Bill, the purpose of this clause ‘is to discourage asylum-seekers from travelling to the UK other than via safe and legal routes’ (para. 145). If so, then the clause rests on a fallacy. No ‘safe and legal routes’ exist by which asylum-seekers can travel to the UK, since the UK does not presently allow applications for asylum to be lodged from outside the UK. [1] Thus, clause 10 merely unjustly penalises asylum-seekers for a decision by the UK not to create such routes. This injustice is particularly acute for sur place refugees, [2] who can never fulfil the Group 1 criteria.

5. The Explanatory Notes add that clause 10 is aimed at ‘encouraging individuals to seek asylum in the first safe country they reach’ (para. 145). However, the data suggests that these kinds of ‘hostile environment’ measures in the UK have little to no impact on decisions about where to seek asylum. [3] As such, clause 10 is not likely to influence decisions about where to claim asylum. It serves merely as a ‘sting in the tail’ to pointlessly penalise asylum-seekers after arrival in the UK.

6. Moreover, the attempt to push persons fleeing persecution to claim asylum only in frontline States runs contrary to the refugee law principle of cooperation. [4] Thus, in the 2018 Global Compact on Refugees, the UK and other UN States agreed to the objectives of ‘easing the pressure’ on frontline host States and ‘expanding access to third country solutions’ (para. 7). Clause 10 not only casts doubt on the UK’s good faith in shouldering its fair share of refugees. It also seriously undermines the already fragile basis for cooperation for those countries in a global refugee regime in which many of them shelter a greater proportion of the world’s refugees than does the UK. [5]

Structure of the clause

7. The Explanatory Notes state clause 10 is ‘based on the criteria set out in Article 31(1) of the Refugee Convention’ (para. 142). In fact, clause 10 inverts the logic of that provision, creating a perverse outcome for some refugees, contrary to the rationale of Article 31(1).

8. In the Refugee Convention, Article 31(1) establishes an obligation not to impose penalties for ‘illegal entry or presence’ on refugees who meet its terms. Thus, those individual refugees must be exempt from the usual penalties that might otherwise be applied to aliens for illegal entry or presence. The provision functions as a limited exception to immigration law penalties applicable to aliens generally.

9. Clause 10 of the Bill, by contrast, draws on the Article 31(1) criteria to create a two-tier system of entitlements within the refugee class only. As such, the penalties imposed by clause 10 on Group 2 refugees have no parallel in, and are additional to, other penalties that may be imposed on aliens for illegal entry or presence. As a result of clause 10, Group 2 refugees will thus face penalties for illegal entry/presence additional to those imposed on aliens generally, merely because they are refugees.

Content of the clause

10. Clause 10 also confers excessively broad discretionary powers. Firstly, although clauses 10(5)-(6) do not specify any limit on how the UK ‘may treat Group 1 and Group 2 refugees differently’, [6] important general legal limits to such powers do exist as a result of binding standards contained in treaties to which the UK is party. [7] Moreover, the modes of differentiation listed in clauses 10(5)-(6) also raise questions of compliance with specific obligations in those treaties, including the non-penalisation provisions in Article 31(1) and 31(2) of the Refugee Convention, the employment and welfare rights in Articles 17-24 of the Refugee Convention, and equality and non-discrimination obligations in domestic and international human rights law.

11. Secondly, by allowing the discretionary powers to be exercised in particular cases by the ‘Secretary of State or an immigration officer’, clauses 10(5)-(6) create the risk that other factors, including discrimination on the basis of protected characteristics, may influence decisions about whether to treat an individual as a Group 1 or Group 2 refugee. The courts have had to intervene previously where UK immigration authorities have abused such broad discretionary powers to disadvantage asylum-seekers on discriminatory racial grounds. [8] Article 3 of the Refugee Convention forbids discrimination on grounds of ‘race, religion or country of origin’, as do other domestic and international legal standards.

Implications of the clause

12. Recent data confirm that the UK asylum system remains beset by inefficiency, delays and backlogs in decision-making, which have only increased during the COVID-19 pandemic. [9] The two-tier refugee status envisaged by clause 10 will not achieve its purported purpose, nor does it serve any practical function. The only guaranteed impact on the UK asylum system will be to make it even more complex, bureaucratic, costly and unworkable.

13. Few refugee situations globally resolve themselves promptly, such that refugees usually wait years if not decades before it is safe to return to their countries of origin. [10] Against this backdrop, clause 10 will have the effect of exponentially increasing the number of decisions that the Home Office has to take each year (due to shorter periods of leave and increased requirements for indefinite leave) and increasing the number of legal challenges to decisions likely to come before UK tribunals and courts (due to the lack of access to public funds and the denial of family reunification). At the same time, clause 10 will curtail the prospects of integration for Group 2 refugees, bringing with it the consequent tangible costs to UK society of their exclusion, marginalisation and lost productivity.

14. Meanwhile, on the international stage, the UK will firmly position itself as an outlier or pariah in the collective endeavour of refugee protection, rather than a global leader post-Brexit. The Clause 10 intention to push refugees to seek asylum elsewhere implicitly banks on other countries maintaining a more equitable reception policy for refugees. But other countries will not easily reconcile the UK playing this ‘beggar-my-neighbour’ brand of international relations with our aspiration to be a leader on the world stage. Moreover, those same countries may decide to follow our lead and also row back from their commitments, unravelling the global refugee regime. A world without a viable global refugee regime is one in which the UK receives many more refugees entering illegally.


15. Clauses 27-35 interpret key provisions of the Refugee Convention. The UK played a major part in drafting the Convention, including the refugee definition at Article 1A(2). [11] Since then, courts in the UK have had a notably distinguished role in developing interpretations of the Convention refugee definition that carry substantial weight internationally. [12] However, whilst the courts have been careful not to go beyond legitimate interpretation of the Convention, [13] some of the legal approaches proposed in clauses 27-35 are not easily reconciled with correct interpretation of the Convention.

16. The Explanatory Notes also state that ‘The UK’s departure from the EU provides an opportunity to clearly define, in a unified source, some of the key elements of the Refugee Convention in UK domestic law’ (para. 315). Certainly, in the UK, the Convention is well-understood without reference to the Common European Asylum System (‘CEAS’), not least as the Convention and its application in the UK predate the CEAS and its 2004 EU Qualification Directive. However, the EU Qualification Directive contains several unclear provisions, which some clauses of the Bill reproduce. ‘Moving on’ from the CEAS need not mean reproducing its flaws.

17. Selected examples of particular provisions in clauses 27-35 that we consider problematic, and which we consider are in need of amendment, appear below.

Clause 28

18. Clause 28(2) largely reproduces Article 9(1) of the EU Qualification Directive. However, it reproduces a flaw in the EU provision by incorrectly specifying in sub-paragraph (a) that a ‘basic human right’ is to be equated with ‘a right from which derogation cannot be made under Article 15 of the [ECHR]’. If the latter phrase is to be retained, then the preceding words ‘in particular’ should be amended to ‘including’ or ‘such as’. This is because the Court of Justice of the EU has made clear that, in Article 9(1) of the Qualification Directive, basic rights are not confined to non-derogable rights under Article 15 ECHR. [14]

Clause 29

19. The Explanatory Notes suggest a provision on the concept ‘well-founded fear’ will achieve clarity and restrain broad interpretation. But a ‘well-founded fear’ provision is absent from the EU Qualification Directive precisely because the concept is extremely ill-suited to codification. Rather, over the past thirty years, consistent practice by the UK courts and tribunals has gradually developed a clear and workable single-stage risk assessment. [15] This must be conducted to a ‘reasonable likelihood’ standard and takes account of all past, present and future evidence in a holistic manner.

20. By contrast, clause 29 seeks to apply a three-stage test, [16] combining two different standards of proof. The first precondition that clause 29(2)(a) adds to the existing risk assessment is a ‘characteristic’ test that, if interpreted in line with Clause 30(2)(c) and (5), could prevent the holistic assessment of relevant circumstances. [17] The second precondition that clause 29(2)(b) adds is a ‘fear’ test. But, if understood as ‘subjective’ (rather than objective) fear, the test opens the door to endless re-litigation of the (settled) question of whether refugees must possess subjective fear (for example, a young child or mentally disabled person may not have subjective fear). Moreover, being cast in the present tense, the test is of open-ended temporal scope and cannot be operated without overlapping heavily and confusingly with the risk assessment enjoined by Clause 29(4). If, however, the Clause 29(2)(b) ‘fear assessment’ includes a forward-looking assessment on the ‘balance of probabilities’ standard, then it too could prevent Clause 29(4) from functioning as a holistic risk assessment to the lower ‘reasonable likelihood’ standard, contrary to the Refugee Convention.

21. Presumably, the Bill’s drafters envision that, by requiring some parts of the refugee inquiry to apply a ‘balance of probabilities’ standard, they would tighten existing law and practice. However, the Court of Appeal has consistently held that, even in classic civil litigation, application of the ‘balance of probabilities’ standard cannot preclude a decision-maker from taking account of uncertainties. [18] Imposing a civil standard legislatively will not prevent that. In any event, refugee law litigation is public law litigation not easily shoehorned into a civil litigation model.

22. In short, clause 29 wholly fails to achieve a more stringent approach or legal certainty. Its opacity will inevitably set judges and other decision-makers aghast and generate protracted litigation. Its effect will be the opposite of what is intended: it will neither unify nor clarify; and risks breaching the UK’s obligations under the Refugee Convention.

Clause 30

23. Clause 30 largely transcribes Article 10(1) of the EU Qualification Directive. Clause 30(2) thus requires that both conditions in clause 30(3) and 30(4) must be met before a group comprises a ‘particular social group’ under the Refugee Convention. Yet the UK House of Lords has held that this aspect of Article 10(1) goes beyond what is justified by interpretation of the Convention. [19] More recently, the UK Upper Tribunal has likewise held that the separate requirements are disjunctive, not conjunctive. [20] This is in accordance with UNHCR guidance. [21] In Australia, an attempt in 2014 to introduce a conjunctive approach into legislation was abandoned and replaced by a disjunctive approach. [22] As such, the word ‘both’ in Clause 30(2) should be replaced by the word ‘either’.

Clause 32

24. Clause 32(2)(b) stipulates that, in considering internal relocation, the decision-maker ‘must disregard any technical obstacles relating to return to that part of that country’. This formulation replicates Article 8 of the EU Qualification Directive (and 339O of the Immigration Rules). However, both have been treated as out of line with international standards. This is reflected in the 2014 recast of the Qualification Directive. Courts in the UK have also not accepted that ‘technical obstacles’ can be interpreted in such a way as to cause a claimant who cannot access an internal relocation alternative due to such obstacles to fail in their claim for protection. [23] Clause 32(2)(b) should be deleted.

Clause 33

25. Clause 33(3) of the Bill addresses the phrase at Article 1(F)(b) of the Refugee Convention - ‘prior to his admission to that country as a refugee’. However, clause 33(3) elaborates the point in time of such admission in a very prescriptive manner, as ‘the day on which they are issued with a relevant biometric immigration document by the Secretary of State’. To the best of our knowledge, the current practice of the Secretary of State is to issue such a document after the decision that someone is entitled to protection as a refugee. But clause 33(3) makes the application of Article 1(F) of the Convention depend upon the timely performance by the Secretary of State of current procedure.

26. That seems open to the objection that the point in time of ‘admission… as a refugee’ for the purposes of Article 1F(b) must be no later than that of the actual decision notified to the individual. [24] It cannot be the date of a later administrative action to issue a document the purpose of which is declaratory, not constitutive, of refugee status. Moreover, it should be noted that UNHCR interprets ‘admission as a refugee’ as referring to the physical arrival of the asylum-seeker in the host country, explaining that ‘[a]cts committed by the refugee during his stay in the host country, prior to grant of any residence permit, should be dealt with through criminal procedures and, where applicable, in the context of the exception to the non-refoulement principle’. [25]

Clause 34

27. Clause 34(1)-(3) imposes a narrower interpretation of Article 31(1) of the Refugee Convention than appears permissible on an ordinary reading of that important provision. It is well-recognised, including by courts in the UK, that a proper reading of Article 31 requires States to ‘put in place procedures to ensure that those entitled to its protection are not prosecuted, at any rate to conviction, for offences committed in their quest for refugee status.’ [26]

Clause 35

28. Clause 35 addresses Article 33(2) of the Refugee Convention. The Explanatory Notes state that the aim is to ‘reduce the threshold at which a refugee is considered to have committed a particularly serious crime’ from ‘a period of imprisonment of at least two years to [one] of at least 12 months’ (para. 364). In this context, it is noted that past legislative prescription that ‘particularly serious crime’ should include all crimes enumerated in a relevant instrument by the Secretary of State and/or exceeding two years’ imprisonment was found by the Court of Appeal to be unlawful in part. [27] The Court accepted the creation of rebuttable presumptions in relation to both the seriousness of the crime and danger to the community, but held that the power to specify offences conferred by section 72(4)(a) Nationality Immigration and Asylum Act 2002 was restricted to offences which the Secretary of State could reasonably consider as giving rise to the statutory presumptions. A statutory instrument listing a number of criminal offences which would be presumed, irrespective of the sentence imposed, to be consistent with the definition of ‘particularly serious crime’ at Article 33(2) of the Convention , was ultra vires and unlawful. Read according to its ordinary language, clause 35 is thus incompatible with the established legal meaning of Article 33(2) of the Convention.

15 September 2021

[1] In general, refugee resettlement schemes are open only to refugees and not asylum-seekers. For instance, a UK scheme to resettle a set number of Syrian refugees in the MENA region is not open to asylum-seekers at all, nor is it open to refugees from other countries and regions (UK Visas and Immigration, Vulnerable Persons and Vulnerable Children's Resettlement Schemes Factsheet, 18 March 2021,

[2] A person becomes a refugee "sur place" due to circumstances arising in his country of origin during his absence’ (United Nations High Commissioner for Refugees (UNHCR), Handbook on Procedures and Criteria for Determining Refugee Status, Geneva, UNHCR, para. 95).

[3] There is a longstanding body of research on this point. See, for example, H. Crawley & J. Hagen-Zanker, "Deciding Where to Go: Policies, People and Perceptions Shaping Destination Preferences", International Migration, 57(1), 2019, 20-35; A. Gilbert & K. Koser, "Coming to the UK: What do Asylum-seekers Know about the UK before Arrival?", Journal of Ethnic and Migration Studies, 32(7), 2006, 1209-1225; E. Thielemann, Does Policy Matter? On Governments' Attempts to Control Unwanted Migration. Center for Comparative Immigration Studies, Working Paper 112, December 2004,

[4] See, for example, the preamble to the Refugee Convention: ‘Considering that the grant of asylum may place unduly heavy burdens on certain countries, and that a satisfactory solution… cannot therefore be achieved without international co-operation’.

[5] See the most recent absolute numbers of refugees per country, World Bank, Refugee population by country or territory of asylum,

[6] The modes of differentiation listed in clause 10(5) are clearly illustrative rather than exhaustive, as signalled by the words ‘for example in respect of’.

[7] Pertinent standards from treaties to which the UK is party derive from the 1951 Refugee Convention, 1950 European Convention on Human Rights, 1966 International Covenant on Civil and Political Rights and 1966 International Covenant on Economic, Social and Cultural Rights.

[8] See, for example, the facts and findings in Regina v. Immigration Officer at Prague Airport and Another, Ex parte European Roma Rights Centre and Others [2004] UKHL 55.

[9] See, for instance, A. Hewett, Living in Limbo: A decade of delays in the UK asylum system, London, Refugee Council, July 2021,

[10] UNHCR estimates that, at the end of 2020, 15.7 million refugees (76 per cent) were in a ‘protracted’ situation (i.e. one in which 25,000 or more refugees from the same nationality have been in exile for at least five consecutive years in a given host country). See UNHCR, Global Trends in Forced Displacement in 2020, Geneva, UNHCR, 2021,, 20.

[11] See for instance, the British proposal in drafting, Ad Hoc Committee on Statelessness and related problems, UN Doc A/AC.32/L.2 (1950), presaging all major aspects of the final definition in the Convention.

[12] See, for example, Islam v SSHD; R (ex p Shah) v SSHD [1999] UKHL 20; [1999] 2 AC 629 and SSHD v Fornah; K v SSHD [2006] UKHL 46; [2007] 1 AC 412; HJ (Iran) and HT (Cameroon) v SSHD [2010] UKSC 31; [2011] 1 AC 596, RT (Zimbabwe) and Ors [2012] UKSC 38; [2013] 1 AC 152, WA (Pakistan) v SSHD [2019] EWCA Civ 302; [2019] All ER (D) 124 (Mar); EB (Ethiopia) v SSHD [2007] EWCA Civ 809; [2009] QB 1.

[13] ST (Eritrea) v SSHD [2012] UKSC 12; [2012] 2 AC 135, para 31.

[14] C-71/11 and C-99/11, Y and Z (2013), Judgment of the CJEU, para. 57.

[15] See, for example, R (Sivakumaran) v SSHD [1987] UKHL 1; [1988] 1 AC 958; and Karanakaran v SSHD [2000] EWCA Civ 11; [2000] 3 All ER 449

[16] Paragraphs 330 and 331 of the Explanatory Notes to the Bill portray Clause 27 as a ‘two-limb’ test. That is misleading. It is in fact a three-limb test. What the Explanatory Notes call the ‘first limb’ is two-pronged. Only if an applicant has a ‘characteristic’ as defined by 27(2)a) can the decision-maker go on to 27(2)(b).

[17] By excluding circumstances relevant to risk (e.g. the fact that the applicant left her country illegally). This is a circumstance that is presently seen by Tribunal country guidance ( MST and Others (national service – risk categories ) Eritrea CG [2016] UKUT 00443 (IAC) (7 October 2016) and the Home Office ( as sufficient (for those of military age) to give rise to a real risk of persecution.

[18] Karanakaran v. Secretary of State for the Home Department , [2000] EWCA Civ. 11, [16].

[19] SSHD v Fornah; K v SSHD [2006] UKHL 46, para. 16; [2007] 1 AC 412.

[20] DH (Particular social group: Mental health) Afghanistan [2020] UKUT 223 (IAC).

[21] UNHCR, Guidelines on International Protection No. 2: "Membership of a Particular Social Group", Geneva, UNHCR, 7 May 2002, paras. 10-13.

[22] See Australian Migration Act (as amended), section 5L.

[23] HH (Somalia) [2010] EWCA Civ 426; AA (Article 15(c)) [2015] UKUT 544 (IAC) §116.

[24] A decision has effect only from the date of notification: R (Anufrijeva) v SSHD [2003] UKHL 36; [2004] 1 AC 604 at 621, at para 26 per Lord Steyn (with whom Lords Hoffman, Millett and Scott of Foscote agreed), ‘Notice of a decision is required before it can have the character of a determination with legal effect because the individual concerned must be in a position to challenge the decision in the courts if he or she wishes to do so. This is not a technical rule, it is simply an application of the right of access to justice…’

[25] UNHCR, Comments on the European Commission’s proposal, Geneva, UNHCR, July 2010, 12,,UNHCR,COMMENTARY,,4c503db52,0.html.

[26] R v Uxbridge Magistrates Court ex p Adimi [1999] EWHC Admin 765; [2001] QB 667, para 47 per Simon Brown LJ. See also the examination of Article 31 and its interpretation in R v Asfaw [2008] UKHL 31; [2008] 1 AC 1061.

[27] EN (Serbia) v SSHD; KC (South Africa) v Same [2009] EWCA Civ 630; [2010] QB 633.


Prepared 21st September 2021