England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION (ADMINISTRATIVE COURT)
Mrs Justice Elisabeth Laing
[2019] EWHC 1586 (Admin)

		Royal Courts of Justice Strand, London, WC2A 2LL
		15 March 2021

Lady Justice Simler:

Introduction

1. This appeal concerns a challenge to the application of an ex gratia scheme put in place by the respondent to support and assist third-country national refugees outside the United Kingdom who have fled the conflict in Syria. The Vulnerable Persons Resettlement Scheme ("VPRS" or "the Scheme") (and the new UK Resettlement Scheme, which replaced it ("the New Scheme")) make resettlement in the United Kingdom available to those refugees who are judged most in need of resettlement according to criteria set by the United Nations High Commissioner for Refugees ("UNHCR"). The appellants are Palestinians, who as refugees from the conflict in Syria fled from Syria to Lebanon. They wish to be considered for resettlement in the United Kingdom under the Scheme or the New Scheme and consider that they satisfy the vulnerability criteria common to both schemes. Although the claim relates to the Scheme which has reached its quota for resettlement, it is common ground that the New Scheme adopts an identical referral process to that set by the Scheme. The claim and appeal are not therefore academic.
2. For reasons explained more fully below, the appellants have little or no prospect of being considered for resettlement under either scheme. The exclusive responsibility of UNHCR to make referrals under the Scheme has prevented the appellants, as Palestinian refugees in a country covered by the exclusive mandate of the United Nations Relief and Works Agency ("UNRWA"), from being considered for resettlement under the Scheme. In areas where UNRWA operates (Lebanon, Jordan, Syria the West Bank and Gaza) UNHCR has no mandate over Palestinian refugees registered with UNRWA because they are deemed by UNHCR to be receiving assistance from another United Nations ("UN") agency, and are thereby excluded from the UNHCR's mandate. Moreover, UNRWA does not have a resettlement mandate.
3. The appellants challenged the Scheme in its original form, and as expanded subsequently. By the date of the trial their challenge was primarily directed at the respondent's decision to rely exclusively on UNHCR for referrals to the Scheme, and to fail to put in place any alternative mechanism to enable Palestinian Refugees from Syria ("PRS") like them to be referred, thereby excluding them from resettlement in the United Kingdom under the Scheme. They contended that the Scheme unlawfully discriminated against them as Palestinians in this regard and that the respondent had breached the public sector equality duty ("PSED") in implementing it.
4. Elisabeth Laing J held that the Scheme did not unlawfully, indirectly discriminate against the appellants on the grounds of race, contrary to section 29(6) of the Equality Act 2010 (the "EA 2010"), because the claim was outside the territorial scope of the legislation and, in any event, any indirect discrimination was not unlawful because it was justified. She also rejected a common law rationality challenge. The appellants challenge these three conclusions as wrong, albeit accepting that the rationality challenge adds nothing to their argument on unlawful discrimination.
5. Elisabeth Laing J also found that the PSED in section 149(1) EA 2010 applied to the decision to establish the Scheme; and that, on the footing that it applied, there had been a breach of the section 149(1)(b) duty. In holding that the PSED applied she considered herself bound to follow the decisions of the Divisional Court in Hottak v Secretary of State for the Home Department [2015] EWHC 1953 (Admin) ("Hottak DC") and R (Hoareau and another) v Secretary of State for Foreign and Commonwealth Affairs [2019] EWHC 221 (Admin), [2019] 1 WLR 4105. However, she did not agree with the reasoning in those decisions on this question. The respondent challenged these conclusions by way of cross-appeal but following Elisabeth Laing J's judgment, a further Policy Equality Statement has been conducted to cover both Schemes (completed on 3 April 2020) and accordingly the challenge to the application of section 149 on the facts is regarded by the respondent as academic and is no longer pursued. The respondent maintains however, that the Judge was right to find the reasoning of the Divisional Court in Hottak DC and Hoareau unpersuasive because the reasoning against an extraterritorial interpretation of section 29(6) applies with at least equal force in respect of section 149. The respondent supports Elisabeth Laing J's conclusion that a legislative scheme which had the effect that a public authority could not breach the substantive provisions of the EA 2010, but could still breach the procedural requirements of section 149, would be anomalous and incoherent.
6. The issues that are accordingly raised by this appeal are as follows:

i) What is the territorial reach of section 29(6) of the EA 2010?

ii) If section 29(6) applies outside the United Kingdom, is the indirect discrimination inherent in the Scheme justifiable as a proportionate means of achieving a legitimate aim?

iii) Is the Scheme irrational as a matter of common law because of the exclusive referral role assigned to UNHCR?

iv) What is the territorial reach of section 149(1)(b) of the EA 2010?

The appellants were represented by Ben Jaffey QC, Blinne Ní Ghrálaigh and Julianne Kerr Morrison. The respondent was represented by Sir James Eadie QC and David Blundell QC. I am grateful to all counsel and those instructing them for their excellent written and oral arguments.

The facts

7. The Judge dealt carefully and in some detail with the chronology of the proceedings, and the wider factual background to the judicial review claim as it changed with changes to the Scheme itself (see paragraphs 8 to 67 of her judgment). I will not repeat the detail here. It is sufficient for my purposes to set out the scope and rationale of the Scheme as originally implemented, and to identify the significant changes made to it over the period of this claim. I will also set out in a little more detail the nature of the mandates under which UNHCR and UNRWA operate in the countries surrounding Syria, and their impact on PRS wishing to access the Scheme.
8. The Scheme was launched on 29 January 2014 by a statement in Parliament by the then Home Secretary. In its original form the Scheme applied only to Syrian nationals. The Secretary of State described the "staggering" number of people affected, and the "immense" scale of the refugee crisis. She referred to a "proud tradition" of protecting those in need and said that the United Kingdom would consider cases "where there are particularly vulnerable refugees who are at grave risk"; and said that having consulted with UNHCR, the Government was launching a programme to give "emergency sanctuary" to "displaced Syrians who are particularly vulnerable". The Scheme was described as a resettlement programme for vulnerable refugees fleeing the Syrian conflict who have been "recommended" to the Government "for relocation by UNHCR".
9. The rationale for the Scheme was explained in a written ministerial statement of 25 March 2014 by the Minister for Security and Immigration (the Rt Hon James Brokenshire MP). It said the Scheme would "provide protection in this country to particularly vulnerable refugees who are at grave risk" and displaced by the conflict in Syria. It explained that the Government had "been working closely with the United Nations High Commissioner for Refugees (UNHCR) to identify those who are most vulnerable" and that those admitted to the Scheme would be granted five years' humanitarian protection with all the rights and benefits that go with that status.
10. The Minister confirmed that he had agreed a ministerial authorisation (under paragraph 17(4) of Schedule 3 EA 2010) "to allow differentiation in favour of Syrian nationals whom we want to bring to the UK under the VPR scheme". The authorisation itself, dated 13 March 2014 and signed by him, authorised:

"the grant of entry clearance outside the immigration rules to Syrian nationals under the Syrian Vulnerable Persons Relocation scheme for three months. The purpose of the entry clearance outside the immigration rules is to allow Syrian nationals to come to the UK, whereupon their applications for humanitarian protection under the immigration rules will be considered.…"

In fact it is difficult to see why it was necessary to authorise discrimination in favour of Syrian nationals. They were not being disadvantaged. If anything, the authorisation should have permitted discrimination against similarly affected non-Syrian nationals (for example, vulnerable Iraqi refugees displaced by the Syrian conflict) as the potentially disadvantaged group, but that was no doubt implicit. Although this authorisation was initially relied on by the respondent as an answer to the direct discrimination claim, no similar authorisation was sought in respect of the subsequently widened scheme; and the respondent did not seek to place any reliance on the authorisation, either below or in this court.

11. Following criticism of the limited number of United Kingdom resettlement places on the Scheme and in recognition of the worsening refugee crisis in the Syrian region and across Europe, an extension was announced on 7 September 2015: up to 20,000 Syrian refugees would be resettled by 2020. The vulnerability criteria for resettlement were set out in Home Office Guidance, "Syrian Vulnerable Person Resettlement (SVPR) Programme", 28 October 2015, as follows:

"The people coming to the UK under the [Scheme] are in desperate need of assistance and many have significant needs. It prioritises those who cannot be supported effectively in their region of origin: women and children at risk, people in severe need of medical care and survivors of torture and violence amongst others."

The Guidance explained that "the UK sets the criteria and then UNHCR identifies and submits potential cases for our consideration. Cases are screened and considered by us and we retain the right to reject on security, war crimes or other grounds."

12. On 29 June 2017 the Secretary of State produced a document headed "Annex A: Policy Equalities Statement" ("the PES"). The title referred to the Scheme and explained that from July 2017, the scope of the Scheme would be expanded to include refugees who had fled the Syrian conflict and sought refuge in the region around Syria, specifically in Turkey, Jordan, Lebanon, Iraq and Egypt. A premise of the PES was that UNHCR would be closely involved in the operation of the Scheme on the ground. It addressed the impact of widening the Scheme under each of the protected characteristics. In relation to race, it stated that the Scheme positively discriminated in favour of Syrian nationals as they were the only eligible nationality for the Scheme until July 2017 and explained that a ministerial authorisation had been sought for that discrimination which was considered reasonable, fair and justifiable. As for the change to the scope of the Scheme to include any refugees who had been resident in Syria and had fled the conflict whatever their nationality, it continued:

"This will still constitute indirect discrimination as it will favour Syrian nationals significantly more than those of other nationalities – and will by definition be limited to those nationalities who were in Syria. Indirect discrimination is lawful provided the action is a proportionate means of achieving a legitimate aim. Our legitimate aim is to support the refugee crisis triggered by the Syrian conflict. As the Syria crisis continues, civilians continue to bear the brunt of the conflict marked by unparalleled suffering, destruction and disregard for human life.… Under its Syrian resettlement operational plan, UNHCR considers that non-Syrian refugees from Syria who sought asylum in neighbouring countries should not be treated differently to Syrian nationals, as the refugees have fled the same conflict and suffered similarly as a result of the consequence of the violence in Syria. Many of these groups, who have been displaced multiple times, are unable to return in safety and dignity to their home country. Expanding the scope of the scheme is a proportionate way to meet this aim."

13. Under the heading "Advancing equality of opportunity" the PES stated that the schemes assist the most vulnerable refugees and "in doing so are helping to advance equality of opportunity. The scheme is being monitored and evaluated to track the longer term outcomes for refugees, including employment and other well-being outcomes.…" Under the heading "Fostering good relations" similar points were made by the PES, including that the schemes "help to foster good relations by supporting refugees to integrate into UK society" and reference was made to bringing communities together, helping traumatised and vulnerable families to recover and thrive in the UK and enabling refugees "to gain easier access to English language tuition and Higher Education, supporting self-sufficiency early on in their resettlement."
14. The PES made no reference to the mutual exclusivity of the mandates of UNHCR and UNRWA and their impact on vulnerable refugees needing resettlement. It appears that there was simply no awareness or appreciation of the impact of using UNHCR as the exclusive gatekeeper for the Scheme on those who fell outside its mandate, and of the difficulty caused (for PRS in particular) in this regard.
15. The expansion of the Scheme to include non-Syrian nationals was announced on 3 July 2017 in a written statement to Parliament. The Secretary of State said the Scheme's focus hitherto had been on "the most vulnerable Syrians" and had led to a "quick and efficient response to the crisis". The Government worked "closely with UNHCR to identify the individuals who are most at risk in the region and whose particular needs can only be met in countries like the UK". Up to the end of March 2017, 7,307 Syrians had been resettled in the United Kingdom, with half of those children. The Government said it was on track to meet the commitment to resettle 20,000 refugees by 2020. However, there were "additional groups in the region who have fled Syria and are also extremely vulnerable" but might not have access to one of the resettlement schemes. UNHCR's advice was that "a diversified resettlement quota" was needed to address "the needs of the most vulnerable refugees from all refugee populations in the region". In the light of that, the Secretary of State announced the expansion of the Scheme to "enable UNHCR to refer the most vulnerable refugees in the MENA region who have fled the Syrian conflict and cannot safely return to their country of origin, whatever their nationality". Reference was made to the Scheme offering resettlement to Syrian refugees in Turkey, Jordan and Lebanon.
16. In 2019, it was announced that the Scheme would be consolidated with two other resettlement schemes, namely the Vulnerable Children Resettlement Scheme and the Gateway Protection Programme, to become the UK Resettlement Scheme, the New Scheme. The respondent has confirmed that referrals to the New Scheme continue to operate in the same manner, with UNHCR remaining the sole referral agency. As the governmental guidance to local authorities regarding the New Scheme makes clear:

"3.4. The Scheme is run in partnership with the United Nations High Commission for Refugees (the 'UNHCR'). It demonstrates the UK's support for the UNHCR's global effort to relieve the humanitarian crisis through the provision of resettlement opportunities for some of the most vulnerable people into communities within the UK who:

3.4.1.  have registered with the UNHCR; and

3.4.2.  the UNHCR consider meet one of their resettlement submission categories."

17. Following the expansion of the Scheme in July 2017, the position adopted by the respondent in correspondence and in the pleadings in this litigation, reflected, as the Judge found, "a puzzling obtuseness about the consequences of the exclusive mandates of UNRWA and UNHCR. Since the widening of the Scheme, the Secretary of State has continued to maintain in the proceedings that there really is no problem, and PRS can get access to the Scheme. Indeed, on that account, she/he argued, the claim was academic." Shortly after the expansion announcement, the Government Legal Department ("the GLD") wrote on behalf of the respondent to the appellants' solicitors, Leigh Day, inviting them to withdraw the judicial review claim on the basis that it was now academic because they could approach UNHCR for referral to the United Kingdom.
18. Leigh Day responded asking, "How can our clients apply to the scheme as Palestinians, given they cannot register with UNHCR?" There was further correspondence and ultimately the respondent was ordered by the court to explain in writing how a PRS could apply effectively to and be accepted as eligible for the Scheme and if that was not possible to confirm this in writing.
19. By letter dated 22 December, the GLD responded to that order, making the following points:

i) There was no application process for any resettlement scheme, but while refugees could not apply for resettlement, vulnerable refugees who had fled the Syrian conflict could be identified and considered for referral to the United Kingdom by UNHCR.

ii) To be referred for resettlement, a person had to be registered with UNHCR or, in the case of PRS living in Lebanon or Jordan, with UNRWA, and meet "the preconditions for resettlement consideration and fall under one of UNHCR's resettlement submission categories".

iii) The preconditions were that a person had been "determined to be a refugee by UNHCR", that the person's "prospects for all durable solutions" had been assessed and that "resettlement is identified as the most appropriate solution".

iv) PRS in Lebanon and Jordan were required to be registered with UNRWA. UNRWA would consider whether it was able to provide the support needed by PRS under UNRWA's mandate in the place where the PRS was living. If that was not possible the person would be identified to one of UNRWA's partner organisations to consider the provision of services to them, including UNHCR. UNHCR would consider vulnerability and assess what support services were required. This assessment could include the possibility of resettlement, and accordingly there was every possibility that a PRS could be included in the Scheme provided they were registered. However, their individual circumstances would have to be assessed and processed by the relevant organisation. The respondent was not involved in these processes until a person or family was referred to the United Kingdom for resettlement under one of the schemes.

20. The Judge dealt with the correspondence that followed. She concluded that the scope of the mandates of UNRWA and of UNHCR were at the heart of the claim and found the two mandates to be mutually exclusive as I have foreshadowed.
21. UNRWA was established by a United Nations General Assembly Resolution (no. 302, 8 December 1949) following the 1948 Arab-Israeli conflict, to carry out direct relief and works programmes for Palestinian refugees and their descendants who met the criteria to be eligible for registration with UNRWA. The human development and humanitarian services provided by UNRWA are confined to those in the occupied Palestinian territories, Syria, Jordan and Lebanon. However, it was (and is) no part of its mandate to provide durable solutions to Palestinian refugees, including resettlement elsewhere. As the 4 April 2019 note from UNRWA's Department of Legal Affairs makes clear, in the course of carrying out its mandate UNRWA may identify Palestinian refugees (including those from Syria in Jordan or Lebanon) as high-risk cases with exceptional protection needs which it is unable to address directly. In those circumstances it may consider referral to a relevant partner organisation on a strictly humanitarian and case-by-case basis, but it is then for the partner organisation to decide what further assistance to provide, if any. However, the note makes two things clear: first, lack of access to services or insufficient coverage of needs by UNRWA services as such will not be sufficient to qualify as a high-risk case with exceptional protection needs for possible referral to UNHCR; and secondly, in view of the strict separation of mandates with regard to Palestinian refugees within UNRWA's area of operation, referrals to UNHCR are exceptional and UNRWA does not have a process for Palestinian refugees to apply for referrals to UNHCR.
22. UNHCR is also limited by the scope of its mandate. Although UNHCR has a mandate to refer refugees for resettlement in third countries, including in the United Kingdom under the Scheme, in countries where UNRWA is mandated to provide Palestinians with humanitarian assistance, Palestinian refugees registered with UNRWA are deemed by UNHCR to be receiving assistance from another UN agency, and are therefore excluded in practice from UNHCR's mandate (including its resettlement mandate) (see Article 7(c) of the Statute of the UNHCR of 1950, and Article 1D of the 1951 Convention Relating to the Status of Refugees). As a consequence, UNHCR in Lebanon has no mandate over PRS, and no general mandate to refer PRS, such as the appellants, to the Scheme or the New Scheme. That explains why the Judge doubted whether even a PRS who passed UNRWA's stringent criteria as a high-risk case with exceptional protection needs it could not address, who was referred by UNRWA to UNHCR, would be referred to the Scheme by UNHCR given the terms of its mandate, which exclude PRS. She concluded that "PRS are in practice wholly excluded from the Scheme, or very nearly so" (a conclusion which applies equally to the New Scheme).
23. Statistics (drawn from a live database for 2018, 2019 and 2020 and therefore subject to change) were provided by the respondent in the High Court proceedings and as an exhibit to the statement of Gideon Winward for this appeal. So far as concerns referrals of Palestinian refugees by UNHCR to the four different United Kingdom resettlement schemes in place until 2020, they confirm that there were no resettlements of Palestinian refugees from Lebanon to the United Kingdom. The only Palestinian refugee (unsuccessfully) referred to the Scheme from Lebanon was a non-UNRWA-registered refugee, who was not excluded from UNHCR's mandate. The only successful referral by UNHCR of a Palestinian from an UNRWA area of operation (Jordan) was of a Palestinian in a "mixed marriage" with a non-Palestinian refugee within UNHCR's mandate, referred as a member of their family group. One of the tables compiled by the appellants on the basis of the respondent's statistics (Table 4) shows the number of referrals of non-Syrian national refugees from UNHCR to the UK under the Scheme. The largest groups of non-Syrians were Iraqis (52 or 53 referrals) and stateless people (23 or 24). For other nationalities (for example, Algerian, Ethiopian, Lebanese, Jordanian, Sudanese) there were between 1 and 3 referrals. The number of Palestinian referrals (10 or 18) includes both UNRWA-registered and non UNRWA-registered refugees. Details concerning which referrals resulted in resettlement were not provided, other than for the Palestinian refugees: a total of four Palestinian refugees were resettled under the Scheme, but all were in "mixed marriages" with non-Palestinian refugees falling under UNHCR's mandate and were referred and resettled with their family group and all but one were referred from outside UNRWA's areas of operation. Finally, the respondent has confirmed that none of the final 307 refugees due to be resettled under the Scheme in 2020, but delayed due to the COVID-19 pandemic, were Palestinian.
24. As Mr Jaffey QC emphasised, the gatekeeper role of UNHCR in making all referrals to the Scheme operates despite PRS in Lebanon facing significant challenges: UNRWA reports that PRS registered in Lebanon live "an extremely fragile and precarious existence and are forced to subsist on humanitarian handouts". According to UN figures, 95% of PRS are food insecure, and experience "hunger, cold and illness in pre-existing camps already lacking resources". Access to healthcare is restricted, as is access to education. As PRS, they have no state to assist them, and are entirely dependent on international agencies and other states for assistance and support.
25. Although not assessed by UNHCR for obvious reasons, on the face of it, the appellants meet the criteria for resettlement under the Scheme: they have sought refuge in Lebanon from the Syrian conflict and meet one or more of the Scheme's vulnerability criteria. Nobody has suggested otherwise. Mr Saleh Turani is extremely unwell with chronic heart disease and lacks access to medical treatment. Ms Haifaa Marouf suffers from a serious medical condition, for which she is unable to access adequate medical treatment. She is a woman at risk, alone in Lebanon, without family: her husband and son are missing in Syria, presumed dead; her son-in-law and minor granddaughter are missing, presumed dead, having attempted to seek refuge abroad; her surviving children and minor grandchild who made it to Lebanon have left the country to attempt to seek refuge elsewhere. Both appellants lack foreseeable alternative durable solutions and for present purposes I proceed on the basis that they continue to meet the criteria for referral for resettlement.

The legal framework

26. Part 2 of the EA 2010 sets out the key concepts on which the Act is based, including the "Protected Characteristics" to which it applies. They include race, and race includes nationality or national origin: see section 9(1)(b). As the Judge observed, the EA 2010 does not create any hierarchy of protected characteristics. Rather, the language and structure suggest that each protected characteristic ranks equally with the others.
27. The EA 2010 prohibits direct and indirect discrimination. This appeal concerns indirect discrimination only. Indirect discrimination involves the application of an apparently neutral "provision, criterion or practice" (or PCP as it is generally known) that puts or would put people with a particular protected characteristic at a particular disadvantage compared with others. Nationality or national origin is a relevant protected characteristic and being Palestinian is therefore a protected characteristic. Section 19(1) and (2) define indirect discrimination as follows:

"(1) A person (A) discriminates against another (B) if A applies to B a provision, criterion or practice which is discriminatory in relation to a relevant protected characteristic of B's.

(2) For the purposes of subsection (1), a provision, criterion or practice is discriminatory in relation to a relevant protected characteristic of B's if—

(a) A applies, or would apply, it to persons with whom B does not share the characteristic,

(b) it puts, or would put, persons with whom B shares the characteristic at a particular disadvantage when compared with persons with whom B does not share it,

(c) it puts, or would put, B at that disadvantage, and

(d) A cannot show it to be a proportionate means of achieving a legitimate aim."

Section 23(1) makes clear that when comparing cases for direct and indirect discrimination purposes "there must be no material difference between the circumstances relating to each case."

28. Part 3 EA 2010 (which includes Schedules 2 and 3) makes it unlawful to discriminate against a person when providing a service (which includes the provision of goods or facilities) or when exercising a public function. As Baroness Hale explained in R (European Roma Rights Centre) v Immigration Officer at Prague Airport [2004] UKHL 55, [2005] 2 AC 1 at [76] ("Roma Rights") the prohibition against unlawful discrimination has always applied to public authority providers of employment, education and housing, (and other services, as long as they are of a similar kind to those supplied by private persons) but in R v Entry Clearance Officer, Bombay, Ex p Amin [1983] 2 AC 818, the House of Lords held that it did not apply to acts done on behalf of the Crown which were of an entirely different kind from any act that would ever be done by a private person, in that case the application of immigration controls.
29. That changed so far as race discrimination is concerned with amendments introduced by the Race Relations (Amendment) Act 2000, and more broadly with section 29 EA 2010, which so far as is relevant, provides:

"29 (1) A person (a "service-provider") concerned with the provision of a service to the public or a section of the public (for payment or not) must not discriminate against a person requiring the service by not providing the person with the service.

…

(6) A person must not, in the exercise of a public function that is not the provision of a service to the public or a section of the public do anything that constitutes discrimination, harassment or victimisation.

…

(9) In the application of this section, so far as relating to race or religion or belief, to the granting of entry clearance (within the meaning of the Immigration Act 1971), it does not matter whether an act is done within or outside the United Kingdom.

(10) Subsection (9) does not affect the application of any other provision of this Act to conduct outside England and Wales or Scotland."

30. Section 31(4) explains that a "public function" is "a function that is a function of a public nature for the purposes of the Human Rights Act 1998". Section 31(10) provides that the exceptions in Schedule 3 have effect.
31. The exceptions in Schedule 3 EA 2010 are exceptions from the prohibition on discriminating against a person when providing services or exercising a public function set out in section 29 EA 2010. Paragraph 17 is one such exception. It only applies in relation to race discrimination on nationality, ethnic or national origins grounds. It provides that section 29 does not apply to anything done by a "relevant person" in the exercise of "relevant functions" (paragraph 17(2)). "Relevant functions" are defined as "functions exercisable by virtue of the Immigration Acts, the Special Immigration Appeals Commission Act 1997 or … retained EU law … [relating] to immigration or asylum". "Relevant person" is defined as either a Minister of the Crown acting personally, or a person acting in accordance with a relevant authorisation (paragraph 17 (3)).
32. This paragraph is the source of the power to make ministerial and class authorisations the effect of which can be to exempt, for example an immigration officer acting under the Immigration Act 1971 to grant or refuse entry clearance, from the requirement not to discriminate on race (as defined) grounds. Thus in the context of this case, paragraph 17 provides for the disapplication of the prohibition against discrimination on grounds of nationality in section 29 where the otherwise-discriminatory act is done by a Minister personally (sub-paragraph 3(a)) or by a person authorised by a Minister personally or by legislation (sub-paragraph 3(b) and 4(b)).
33. Section 149(1) EA 2010 Act obliges a public authority, "in the exercise of its functions", to have "due regard" to the equality needs listed in section 149(1)(a), (b) and (c). Those are the needs to:

"(a) eliminate discrimination, harassment, victimisation and any other conduct that is prohibited by or under this Act;

(b) advance equality of opportunity between persons who share a relevant protected characteristic and persons who do not share it;

(c) foster good relations between persons who share a relevant protected characteristic and persons who do not share it."

34. Section 149(3) explains that having due regard to the need described in paragraph (b) "involves having due regard, in particular, to the need to -

"(a) remove or minimise disadvantages suffered by persons who share a relevant protected characteristic that are connected to that characteristic;

(b) take steps to meet the needs of persons who share a relevant protected characteristic that are different from the needs of persons who do not share it;

(c) encourage persons who share a relevant protected characteristic to participate in public life or in any other activity in which participation by such persons is disproportionately low."

35. Section 149(5) explains that having due regard to the need described in paragraph (c) "involves having due regard, in particular, to the need to –

"(a) tackle prejudice, and

(b) promote understanding".

36. Section 149(7) specifies the relevant protected characteristics for these purposes as age, disability, gender reassignment, pregnancy and maternity, race, religion or belief, sex and sexual orientation.
37. Section 149(9) enacts Schedule 18. Paragraph 2(1) of Schedule 18 provides that, in relation to the exercise of "immigration and nationality" functions, section 149(1)(b) has effect as if it did not apply to the protected characteristic of race (defined as meaning nationality or ethnic or national origins). "Immigration and nationality" functions are defined in paragraph 2(2). The phrase means functions which are exercisable by virtue of the Immigration Acts and other specified legislation concerned with immigration and nationality.  Section 150(5) like section 31(4) defines "public function" as a function that is a function of a public nature for the purposes of the Human Rights Act 1998.
38. Section 217(1) EA 2010 provides that the Act "forms part of the law of England and Wales." Subsections (2) and (3) make clear which provisions form part of the law of Scotland and Northern Ireland. These provisions address the extent to which relevant provisions in the Act form part of the law of different parts of the United Kingdom. They do not however, govern the geographical scope of, for example, the functions covered by the EA 2010 and in particular, the public functions referred to in both sections 29 and 149 EA 2010.
39. The EA 2010 is generally silent on extraterritorial application, though there are certain express provisions that deal with extraterritorial effect in specific cases, such as section 29(9) in relation to the grant of entry clearance. Others include sections 30 and 81 dealing with ships and hovercraft, and section 82 which deals with offshore work.
40. The appellants rely on paragraph 15 of the Explanatory Notes to the EA 2010 as explaining the approach adopted by the Government in relation to the territorial application of the Act. It provides as follows:

"As far as territorial application is concerned, in relation to Part 5 (work) and following the precedent of the Employment Rights Act 1996, the Act leaves it to tribunals to determine whether the law applies, depending for example on the connection between the employment relationship and Great Britain. … In relation to the non-work provisions, the Act is again generally silent on territorial application, leaving it to the courts to determine whether the law applies. However, in a limited number of specific cases, express provision is made for particular provisions of the Act to apply (or potentially apply) outside the United Kingdom."

The judgment below

41. Elisabeth Laing J held that the exclusive reliance by the respondent on UNHCR as the sole referral agency or "gatekeeper" to the Scheme (the impugned PCP pursuant to section 19(1) EA 2010) made it "much harder for a PRS to get near the gatekeeper than it is for people who were displaced by the same conflict, but are not PRS": see [121].
42. However, the Judge rejected the appellants' claim of unlawful indirect discrimination in the exercise of public functions contrary to section 29(6) EA 2010. She did so first on the basis of territorial reach. Section 29(9) EA 2010 provided a modest exception to the normal presumption that Acts of Parliament are not intended to extend to things which happen outside the United Kingdom; and section 29(10) demonstrated that Parliament did not intend that modest express exception to undermine the normal presumption about extraterritorial effect, other than to the extent expressly stated in the express exception. She held that section 29(9) applies only to the grant of entry clearance and not to the exercise of common law powers to implement the ex gratia Scheme for resettlement of foreign refugees. In reaching those conclusions she said:

"107. I consider that I am bound by the decision in Hottak in the Court of Appeal to hold that the territorial reach of Part 3 is to be decided in accordance with the reasoning of the House of Lords and of the Supreme Court which applies in claims under the Employment Rights Act 1996 ('the ERA'). Even if I am not bound to do so (Mr Husain submitted that the reasoning about Part 3 was obiter), that reasoning is strongly persuasive. There is no reason in principle or logic why Parliament can possibly be taken to have intended that a Part 3 claim should have a territorial reach which is different from, and wider than, that of a Part 5 claim. Where the claimant is not a British citizen, does not live or work here, and has no other link at all with the United Kingdom, other than a wish to benefit from a policy of the Secretary of State, the very exceptional connection with the United Kingdom which is required is absent."

43. That conclusion meant that the unlawful indirect discrimination claim could not succeed. However, the second basis on which it failed, as the Judge found, was on the issue of justification. The Judge had no doubt that the PCP had the disadvantageous effect described in section 19(2), but concluded that the Scheme had a broader purpose to:

"123…. help vulnerable refugees in Syria as candidates for resettlement in the United Kingdom as quickly and effectively as possible. The partnership with UNHCR, the body which is to assess and chose the candidates in the region, is essential to that purpose. UNHCR has been chosen because of the Government's long relationship with it, because it is present in the relevant regions and because it has the necessary expertise. Mr Husain accepted that UNHCR was a good choice. It was clear, right from the start, that UNHCR was integral to the aims and operation of the Scheme; it is mentioned in each of the three principles on which, when the Scheme was announced, the Scheme was said to be based. The use of UNHCR is closely and rationally connected to the achievement of the Scheme's purpose.

124. Nor do I accept that other less intrusive means were available. It is suggested that NGOs could be used, or that PRS could refer themselves to the Home Office or to British embassies in the region. None of these methods could have achieved the security, reliability, speed, and consistency which flow from using UNHCR as a gatekeeper. There was little, if any, evidence, that any specific NGO had the necessary capacity.

125. I accept Mr Hall's submissions on justification. The severity of the impact of the PCP on the Claimants is mitigated, to some extent, by the very thing which means that in practice they have such limited access to the Scheme. That is, that they are within the scope of UNRWA's mandate. As Mr Husain submitted, the PRS are part of a larger, unique group. The UN has only two refugee organisations; UNHCR and UNRWA. UNRWA's mandate, and its material help, are solely focused on refugees from Palestine. Unlike other people displaced by the Syrian conflict, PRS already have their own relief organisation; and for historical reasons over which the Secretary of State has no control, UNRWA's mandate excludes resettlement and UNHCR's mandate excludes those who are subject to UNRWA's mandate. The fact that UNRWA's resources, and its capacity to help PRS are limited, is nothing to the point. I note, nevertheless, that when UNRWA assessed the Claimants, it concluded that the difficulties faced by the Claimants were common to many PRS (see paragraph 62, above).

126. I take into account that the Secretary of State put forward no justification in July 2017 for the disparate impact of the PCP on PRS. I also take into account that PRS are in practice wholly excluded from the Scheme, or very nearly so. But PRS are not the only vulnerable refugees in the region. It cannot be assumed that they are in fact the most vulnerable, contrary to Mr Husain's submissions. It is impossible to make a judgment about the comparative vulnerability of groups of people, or of individuals, from the comfort of a London court room. On the figures in the Secretary of State's skeleton argument, more than ten times more Syrian people were displaced from Syria by the Syrian conflict than were Palestinians. There are millions of refugees. I consider that, given the aim of the Scheme, to which UNHCR was integral (see above), and given the fact that UNRWA is responsible for PRS, the impact of the PCP on the Claimants was proportionate. Proportionality is for the court, but the design of a scheme such as this is not."

44. The Judge rejected the appellants' common law rationality claim for essentially the same reasons.
45. So far as the PSED claim was concerned, the Judge held (but only because she regarded herself as bound to do so) that in relation to section 149(1)(b) only, it had the extra-territorial reach contended for by the appellants. She set out her reasons for disagreeing with Hottak DC and Hoareau as to the territorial reach of section 149(1) at [113] and [114] (to which I shall return below when dealing with this ground of challenge).
46. Having reached that conclusion, the Judge held that it was unlikely that the Secretary of State did not always know that UNHCR and UNRWA have exclusive mandates; but if she did not, she could have found out that was so. The timing of the widening of the Scheme suggested that the widening, was, in part, at least, prompted by the direct discrimination claim brought by the appellants as PRS. In the circumstances, she held that section 149(1)(b) required the Secretary of State to confront the way in which the Scheme would, despite its widening, still not increase (or materially increase) equality of opportunity for PRS. There was simply "no evidence that the PES was revised, either on its advertised revision date, or later, to show that the Secretary of State had made reasonable inquiries about this issue and had confronted it." (See [129]).
47. Accordingly she concluded:

"132. In that situation, in my judgment, the Secretary of State has not had due regard to the equality need listed in section 149(1)(b). The question is not whether the Scheme in its current form is justified. The questions, rather, are whether it ever occurred to the Secretary of State that the widening of the Scheme, as respects PRS, was theoretical rather than real, and whether it crossed his mind that he should consider whether or not to widen the equality of opportunity for PRS by changing the Scheme so as to enable another gatekeeper to refer their cases to him, and whether he faced up to the fact that if he did widen the Scheme in the way which he did, PRS would be excluded, or virtually excluded from it."

The appeal

Issue 1: Territorial scope of section 29(6) EA 2010

48. As just indicated, Elisabeth Laing J held that the Scheme did not fall within the territorial scope of the EA 2010: section 29(9) extends only to the specific act of granting entry clearance and not to the exercise of common law powers under which the Scheme was created. Section 29(10) and other similar provisions in the EA 2010 demonstrate that, save where express statutory provision is made to the contrary, Parliament adopted the normal presumption that Acts of Parliament are not intended to extend to things that happen outside the United Kingdom. Further, the Judge considered herself bound by Hottak v Secretary of State for the Home Department [2016] WLR 3791 ("Hottak CA") and if the reasoning in Hottak CA was obiter, she regarded that reasoning as strongly persuasive in any event.
49. The appellants challenged this analysis as incorrect. I summarise the main arguments advanced by Mr Jaffey on this part of the appeal as follows:

i) The Judge misconstrued section 29(9) EA 2010 in holding that it applies to the grant of entry clearance but not to the exercise of common law powers to make a policy governing how such decisions are made. Mr Jaffey accepted that the case was put differently below, but in this court made clear that the appellants challenge the decision in their individual cases as to whether they qualify for resettlement and can enter the UK. Sensibly construed section 29(9) is not limited to the specific act of granting or refusing entry clearance. To offer effective protection against discrimination, it must also apply to related acts, including the refusal of entry clearance, and the decision-making processes that lead up to the decision on entry clearance. He accepted that the words used in paragraph 15 of the Explanatory Notes "in connection with" may import too wide a causal connection, but submitted that a policy which determines when entry clearance will be granted or refused must be covered. Indeed, it would make little sense for Parliament to extend the territorial reach of section 29(9) to the formal decision, but not the decision-making process and policies or procedures that govern that process.

ii) If he is wrong about that, the wider question of the territorial reach of section 29 arises. Elisabeth Laing J misinterpreted section 29(10) EA 2010. Parliament deliberately left open the extent to which section 29(6) of the Act would apply extraterritorially. This is made clear in the text of the scheme: section 29(9) does not limit or constrain the other circumstances in which section 29 would apply extraterritorially. To the extent there is any doubt on the point, paragraph 15 of the Explanatory Notes resolves the doubt.

iii) It is for the courts to determine whether the provisions of different parts of the Act apply to the particular facts. The ordinary presumption as regards the territorial reach of statutes does not apply to section 29 and a more nuanced approach to the presumption is required in any event by KBR Inc (see below). Parliament recognised that the proper territorial scope of the legislation was a difficult issue; that it was better approached by judicial decision making than prescriptive statutory rules as with the predecessor legislation; and was content for the law to continue to develop that approach by judicial decision.

iv) Thus, the EA 2010 does have extraterritorial effect in cases falling outside the express exceptions set out in the EA 2010 itself (contrary to the apparent premise of the judgment at [104]). The same case-by-case approach as adopted for Part 5 should be adopted in the context of the exercise of public functions. This may lead to different principles being developed than those applied in the employer-employee relationship (contrary to the apparent assumption made in the judgment at [107]) because very different circumstances apply to the exercise of public functions. A public function exercised by a public authority in the United Kingdom in relation to the admission of a person to the United Kingdom necessarily has a close link to the United Kingdom.

v) Elisabeth Laing J erred in holding herself bound by the judgment in Hottak CA to conclude that section 29(6) has no extraterritorial effect and/or that any such effect should mirror the extraterritorial reach of Part 5 EA 2010.

50. The respondent resists those arguments. The main arguments advanced by Sir James Eadie QC on this part of the appeal can be summarised as follows:

i) The starting point is that, unless the contrary intention appears, Parliament is taken to intend an Act to extend to each territory of the UK, and to apply only to all persons and matters within the territory to which it extends: R (XH and AI) v SSHD [2018] QB 355, (Sir Terence Etherton MR at [97]). See to the same effect: Lawson v Serco Ltd [2006] UKHL 3, [2006] ICR 250. With that general principle in mind, the interpretive exercise requires "an inquiry to be made as to the person with respect to whom Parliament is presumed, in the particular case, to be legislating. Who, it is to be asked, is within the legislative grasp, or intendment, of the statute under consideration?": Clark v Oceanic Contractors Ltd [1983] 2 AC 130, (Lord Wilberforce at [152]) (affirmed in Lawson at [6]).

ii) Sir James Eadie also developed an argument by analogy with MN v Belgium App no. 3599/18 (unreported decision of 5 March 2020) which dealt with the territorial scope of Convention rights in a similar context: the applicants fled the Syrian conflict and travelled to the Belgian Embassy in Beirut to submit visa applications to enter Belgium. The applications were refused and they brought proceedings against Belgium under articles 3 and 6 of the Convention. The ECHR rejected their claims. In doing so, it reiterated that article 1 of the Convention limits its scope to persons within the jurisdiction of the particular state but that jurisdiction can extend extraterritorially in certain limited and well established circumstances. The exercise of a public power in relation to the application for entry to Belgium was not sufficient to bring the applicants under Belgium's territorial jurisdiction. There would have to be exceptional circumstances that could lead to a conclusion that Belgium was exercising extraterritorial jurisdiction here. None of the well-established exceptions applied. Moreover, the mere fact that an applicant brings proceedings in a state with which he has no connecting tie cannot suffice to establish that state's jurisdiction over him: see [123]. Sir James Eadie submitted that the same points can be made here.

iii) The structure and terms of section 29 EA 2010 indicate that it should apply to people and matters in the United Kingdom only. There is nothing to suggest that section 29(6) is a worldwide provision save to the limited extent to which section 29(9) extends its operation to conduct outside the United Kingdom.

iv) Parliament has expressly delimited the circumstances in which the territorial application of section 29(6) EA 2010 is extended to conduct outside the United Kingdom. Section 29(9) extends the application of the EA 2010 in one specific and limited respect, namely as regards the protected characteristics of race, religion or belief, and in respect of the public function of the granting of entry clearance. If Parliament had intended to include acts beyond the grant of entry clearance within the scope of section 29(9) it could easily have achieved that aim by either specifying particular acts or using the appellants' language of "related acts". Its intention is clear from its choice not to do so.

v) Section 29(10) must be read in context with section 29(9) as the Judge found. The combined effect of these provisions is that Parliament made a modest extension to the usual territorial application of a domestic statute (by section 29(9)) and made clear its intention that the extension was not intended to undermine the ordinary presumption about territorial effect. If anything, section 29(10) positively supports the respondent's position. Nor do the explanatory notes help the appellants' case. On the contrary, the Notes clearly intend that the courts should determine this question in the ordinary way – i.e. applying the presumption and taking into account the carefully crafted provisions of section 29(9).

vi) The appellants have no connection at all with the United Kingdom. Such a connection is not established (still less is an intention to bring them within the legislative grasp demonstrated) by the fact that the decision to set up the Scheme was an act done in the United Kingdom. That fact is of no assistance in interpreting the legislation in question. The appellants are outside the territory of the United Kingdom. They are not British citizens. They have never worked or lived in the United Kingdom. They have no links with the United Kingdom other than a wish to benefit from the Scheme. In reality, they have no connection at all with the country. There is nothing to indicate that they fell within the legislative grasp or intendment of the statute.

vii) Elisabeth Laing J was right to find that she was bound by Hottak CA. Paragraph 69 was not obiter. The second reason given by the CA for its conclusion on this point is ratio. In any event, Elisabeth Laing J was right to follow it for all the reasons set out above.

Discussion and analysis

51. I start with the general approach to extraterritorial jurisdiction but address first the judgment in Hottak CA on the question of the territorial reach of Part 3 EA 2010. Hottak concerned a judicial review challenge by a group of Afghan nationals, employed by the British government to work as interpreters with the British Forces in Afghanistan, in relation to a scheme of assistance offering financial benefits and relocation opportunities (including, in limited circumstances, to the United Kingdom). They alleged (among other things) unlawful discrimination because of nationality. By the time the case reached the hearing in the Divisional Court it appears to have been conceded that the challenge to differential relocation opportunities under section 29(6) was excluded by paragraph 17 Schedule 3 because, I infer, there was an authorisation for discrimination on grounds of nationality in relation to decisions under the Immigration Acts and rules under this particular scheme (see Hottak DC at [46] and Hottak CA [66]). The primary case on unlawful discrimination was that the scheme unlawfully discriminated in the provision of employee benefits under section 39(2), Part 5 EA 2010 but, in the alternative, the claimants argued there was unlawful discrimination in the provision of financial benefits under section 29(6), Part 3 EA 2010.
52. Having rejected the primary case under Part 5 at [46] to [57], Sir Colin Rimer dealt with the claim under Part 3 at [68] to [70]. At [69] he held:

"69. … Section 28 provides that Part 3 of the 2010 Act does not apply to discrimination that is prohibited by Part 5. That means that a work-related discrimination claim can be brought only under Part 5 and not under Part 3. If (as I would hold) the claimants are not entitled to invoke Part 5 in pursuit of their work-related discrimination claim (because Parliament did not intend Part 5 to extend to their employment in Afghanistan) then, in agreement with the Divisional Court, I would regard it as surprising if Parliament must nevertheless have imputed to it an intention that the claimants can instead invoke the provisions of Part 3. Quite apart from the point that Burnett LJ made in [49] as to many of the schedule 9 exceptions (relating to Part 5) having no counterpart in the schedule 3 exceptions (relating to Part 3), my intuitive sense as to Parliament's presumed intention is that (a) if a work-related discrimination claim can be brought, it can only be brought under Part 5; and (b) if, for jurisdictional reasons, a work-related discrimination claim cannot be brought under Part 5, it cannot be brought at all. But if that is to put the matter too broadly, I would anyway accept Mr Swift's submission that section 29(6) should not be interpreted as extending to claims other than in respect of the exercise of public functions in Great Britain. I would not accept that there is any warrant for imputing to Parliament an intention to extend it to claims based on the extra-territorial effect of exercise of public functions."

53. Accordingly, two reasons were given for rejecting the interpreters' financial claims. The first, that the claims were work-related discrimination claims and so were required to be brought under Part 5, obviously has no application here. The second reason, that section 29(6) does not have "extra-territorial effect" in relation to claims based on the exercise of public functions, concerned the claim for provision of financial benefits. Read in the context of the earlier paragraphs, I do not think that Sir Colin Rimer can have intended by the second reason to hold that Part 3 EA 2010 can never apply extraterritorially to the exercise of public functions. Such a conclusion would be inconsistent with his implied statement that the second reason was narrower than the first, and his earlier conclusion that the extraterritorial effect of Part 3 EA 2010 should not be interpreted as going further than Part 5, where the case-law has established extraterritorial effects in certain cases. More importantly, the court in Hottak was only concerned with territorial effect relating to the provision of financial benefits, and not with any claim for relocation (or entry clearance) where, at the very least, the role of section 29(9) would require consideration. In these circumstances I do not consider that the ratio is binding on this court in relation to the territorial reach of Part 3 EA 2010 in this case.
54. It is therefore necessary to approach the question of territorial reach from first principles.  As Lord Hoffmann observed in Lawson at [6], the United Kingdom rarely purports to legislate for the whole world and accordingly there is a general presumption that our domestic legislation does not extend beyond the United Kingdom's territory. There are exceptions, and in some cases the question of territorial scope is not straightforward, but again as Lord Hoffmann held:

"6. ... In principle, however, the question is always one of … construction …. As Lord Wilberforce said in Clark v Oceanic Contractors Inc [1983] 2 AC 130, 152, it

'requires an inquiry to be made as to the person with respect to whom Parliament is presumed, in the particular case, to be legislating. Who, it is to be asked, is within the legislative grasp or intendment, of the statute under consideration'."

55. A more nuanced approach to the general presumption was adopted recently in R (on the application of KBR Inc) v Director of the Serious Fraud Office [2021] UKSC 2, which concerned the extraterritorial effect of section 2(3) of the Criminal Justice Act 1987, at [21] to [32]. The Supreme Court held that where the principle of comity that underlies the presumption has less force (perhaps because the state is legislating for the conduct of its own nationals abroad or seeking to hold state officials to certain standards wherever they exercise public functions) the strength of the presumption may itself be reduced. However, ultimately the question in each case is a question of construction to be decided according to established principles of statutory interpretation, seeking to give effect to what Parliament can reasonably be taken to have intended, on the premise that Parliament intended a rational and coherent scheme.
56. I do not find the approach set out in MN v Belgium helpful in resolving the question of construction in this case. Section 29 is part of a scheme for the fair regulation without discrimination of public services and functions and Parliament has plainly considered the question of its extraterritorial effect, albeit the extent of that is in issue. There is no direct analogy. Nor do I accept the suggestion floated in argument by Sir James Eadie that section 31(4) EA 2010 assists in identifying the scope of territorial jurisdiction here. That provision merely defines public functions and has nothing to do with jurisdiction.
57. So I start by considering the scheme of section 29 EA 2010. Sir James Eadie made a number of general points in relation to its broad structure, which I accept. First, it concerns the provision of a service (section 29(1)) and the exercise of a public function not to do with the provision of services (section 29(6)), to the public or a section of the public. It is reasonable to infer that in both cases "the public" is the same; and that the public referred to in section 29(1) in particular, given its scope, is the public in Great Britain, rather than the rest of the world. Secondly, given that the language of section 29(6) EA 2010 tracks the language of section 29(1) to include public functions, on the face of it these provisions have the same territorial reach. Thirdly, section 29(9) and (10) do not distinguish between the provision of services to the public and the exercise of public functions; they apply equally to both.
58. Whatever its scope, section 29(9) EA 2010 is a specific and limited provision that extends the territorial reach of section 29, to conduct outside the United Kingdom, but only in its application "to the granting of entry clearance (within the meaning of the Immigration Act 1971)" so far as "relating to race or religion or belief". The words "relating to" (which are used to refer to the protected characteristics in question) or other similar words (such as "in connection with the grant of entry clearance") could have been but were not used in referring to the grant of entry clearance. Instead, the language demonstrates an express intention by Parliament to extend the territorial effect of section 29 to things done outside the United Kingdom, but only in specified, limited circumstances. It follows from this subsection that a natural reading of section 29(1) and (6) is that they do not ordinarily extend to things done outside the United Kingdom because otherwise section 29(9) would not be necessary. That is consistent with Parliament's intention to legislate for conduct relating to its own citizens – the public – in the United Kingdom. I do not consider that section 29(10) answers this point, as Mr Jaffey suggested. In my judgment section 29(10) is a neutral provision that simply makes clear that section 29(9) does not affect the application of other provisions in the EA 2010 to conduct outside the United Kingdom either positively or negatively. The general approach applies. This subsection does not advance either side's case. Nor does paragraph 15 of the Explanatory Notes lead to a different conclusion. Leaving aside the question whether such Notes are admissible as an aid to construction, this paragraph merely confirms the general approach.
59. However, it also follows from section 29(9) that section 29(6) does in principle apply to functions exercisable under the Immigration Acts, otherwise paragraph 17 of Schedule 3 (and indeed 29(9)) would be redundant.
60. Prima facie, therefore, section 29(6) applies to decisions to grant entry clearance in individual cases, and as Sir James Eadie accepted, that must inevitably cover refusals as well, and section 29(9) makes clear that it makes no difference whether the grant (or refusal) is done within or outside the United Kingdom.  (An oddity identified by Underhill LJ in the course of argument is the restriction of this provision to the protected characteristics of race, religion or belief. Given, at least, the potential for discrimination on grounds of sex in the grant of entry clearance by entry clearance officers abroad, it is difficult to see why the protected characteristic of sex is not also provided for. However that does not affect the reasoning or outcome in this case.)
61. The scheme of the EA 2010 allows for limited discrimination in the exercise of immigration functions on nationality grounds by virtue of paragraph 17 Schedule 3. However in authorising such discrimination, the Minister must act personally in securing the authorisation and so be accountable for it. Although an authorisation was made in relation to the Scheme as originally implemented, no authorisation was made in relation to the Scheme as expanded to apply to non-Syrian nationals. No explanation has been given as to why those different courses were taken, though it was suggested on behalf of the respondent that the authorisation given for the Scheme as originally promulgated may have been misconceived. However it is interesting to note here, as the Judge recorded at [15], that at least initially the "Secretary of State submitted that the Authorisation was an answer to the claim under the 2010 Act. The Scheme was set up under common law powers, but was operated by the grant of entry clearance. It was artificial to separate the establishment of the Scheme from the decision to grant entry clearance." The appellants make the same point in relation to section 29(9), and I shall return to this point below.
62. I agree with the Judge that section 29(9) applies to the grant and refusal of entry clearance and that the making of the Scheme (albeit providing an alternative route to the grant of entry clearance for a particular cohort of people referred by UNHCR) is not itself the grant of entry clearance under section 29(9) or the exercise of a function under the Immigration Acts within the meaning of paragraph 17.  Rather it is the exercise of prerogative powers to make a policy about how statutory immigration powers will be exercised. That is what I understand the Judge to have said when she described the Scheme as something done under common law powers. 
63. However, as Mr Jaffey accepted, the case now presented is put differently to the case presented below. In this court, Mr Jaffey contended that the making of the policy (or Scheme), which he accepted does not fall within section 29(9), is different to the application of such a policy to an individual's case. Sensibly construed, and in order to offer effective protection against discrimination, section 29(9) must also apply to the decision-making process that governs the decision on entry clearance. This includes a policy which determines when entry clearance will be granted or refused. Indeed, it would make little sense for Parliament to extend the territorial reach of section 29(9) to the ultimate decision, but not the decision-making process and the policies and procedures that govern that process. Here, he submitted, the challenge is to the application of the policy to the appellants' individual cases that means they cannot be considered for entry clearance under the scheme.
64. In resisting that argument, Sir James Eadie emphasised that neither a decision by UNHCR nor a decision by UNRWA not to refer an individual for resettlement is or can reasonably be described as a refusal of entry clearance; and at no point does the individual apply for entry clearance. The referral stage is the aspect of the scheme under challenge in this case. If a referral is made the process that follows involves, first, a resettlement decision by the Home Office, based on questions of vulnerability, need and biographic and other security checks; and secondly an allocation decision involving liaison with the relevant local authority as to whether there is the ability and capacity to resettle the individual within that local authority. Only if those decisions are positive is an entry clearance decision made at the final stage, and again at that stage there may be further or different biometric and other security checks. The grant of entry clearance is made if at all by an entry clearance officer located in Amman, in Jordan, under section 3 of the Immigration Act 1971, the entry clearance officer exercising the statutory functions of an immigration officer under the 1971 Act.
65. Accepting as I do, of course, this description of the way in which the Scheme works, I have concluded nonetheless that it is artificial on the facts of this case, to separate the decision-making into different stages, thereby divorcing the earliest stage from the final entry clearance decision. My reasons follow. There is no dispute, at least for present purposes, that the appellants are highly vulnerable refugees who have fled the conflict in Syria and qualify for resettlement under the Scheme. Again, for present purposes at least, the case has proceeded on the basis that the only reason for the failure to refer them for resettlement is the operation of the exclusive mandates of the two agencies resulting in a situation where they cannot in practice be referred. This is, accordingly, not a case where there is any challenge to any independent decision-making or exercise of discretion by UNHCR at the referral stage (or indeed at any of the later stages of the process). Rather, as a consequence of the rules of the Scheme set by the respondent, the appellants are simply outside the mandate of UNHCR and cannot access the Scheme at all.
66. It is accepted that section 29(9) EA 2010 cannot be read literally and extends to refusals as well as grants, and must cover the substance of the decision and not merely the formal decision itself. Accordingly, it must also extend to matters that are integral to the grant (or refusal) decision. For example, as Sir James Eadie accepted, it would extend to cover the questioning by an entry clearance officer leading to the entry clearance decision (as occurred in Roma Rights, albeit the point was not disputed in that case) as integral to the decision. If conditions on the grant are imposed, in my judgement it would plainly extend to cover those.
67. The whole purpose of the Scheme is to resettle vulnerable refugees fleeing the Syrian conflict in the United Kingdom, by granting them entry clearance. The Scheme sets out special entry control rules outside the immigration rules for a class of vulnerable refugees. Whether or not a particular vulnerable refugee can come into the United Kingdom is determined by the rules set by the Scheme. The impugned rule applied to the appellants is the requirement of a referral by UNHCR for resettlement. Although this is not the grant of entry clearance itself, it is an absolute precondition to obtaining a grant of entry clearance under this special route.
68. If the process described by Sir James Eadie were reduced to a written application form in three or four parts (and I appreciate that no application can be or is actually made) requiring a stamp from UNHCR to signify that the refugee applicant is within UNHCR's mandate at the first stage, without the stamp the application could not proceed. If there was a refusal of entry clearance at that point, it seems to me that it would clearly be covered by section 29(9). The need for the stamp would properly be regarded as integral to the entry clearance decision. It is artificial in those circumstances to treat the failure to meet the precondition, which means that the stage of a refusal of entry clearance is never reached, as not also covered by section 29(9). The Scheme does not operate outside and independently of entry clearance as the respondent contended. Although set up under common law powers, it operates to control entry to the United Kingdom. Its rules are applied to individual cases and determine whether entry clearance is granted. The Scheme cannot be divorced from the decision to grant entry clearance. It follows that, by virtue of section 29(9), section 29(6) applies to the "decision" by application of the rules of the Scheme to their cases, to exclude the appellants from consideration for resettlement with the grant of entry clearance under the Scheme despite meeting the vulnerability and other criteria set out in the Scheme.
69. In my judgement it also follows that, although the promulgation of the Scheme does not involve the exercise of functions under the Immigration Acts, each time the rules of the Scheme are applied to an individual's case they bite on the exercise of functions under the Immigration Acts and are capable of being exempted by paragraph 17 accordingly.
70. In his judgment Underhill LJ has identified a possible alternative route to the same result if it were thought that the policy itself might not fall within the terms of section 29(9) EA 2010. I respectfully agree with him that even if that were the case, given the terms of section 29(9), it might be thought reasonable to infer that Parliament intended the same territorial scope to apply to the conditions of a policy governing how entry clearance decisions are made in an individual's case for the reasons he gives at paragraph 109.
71. For these reasons, on the facts of this particular case, unlike the Judge who was dealing with a quite different argument on this issue, I have concluded that the prohibition on unlawful discrimination in section 29(6) EA 2010 applies to the exclusion of the appellants from the Scheme, notwithstanding that they are outside the United Kingdom.

Issue 2: Justification

72. In those circumstances it is necessary to consider the challenge to the Judge's conclusion that the indirect discrimination involved in applying the PCP (that UNHCR is the exclusive gatekeeper to the Scheme) to the appellants' cases as PRS, was justified. Central to their challenge is the contention that there were two erroneous premises to the finding of justification: first an erroneous identification of the purpose of the Scheme; and secondly, the erroneous reliance on the existence of UNRWA as mitigating the impact of the PCP.
73. Developing those grounds, Mr Jaffey submitted that the Judge was wrong to see the Scheme as requiring speed and to determine that the purpose of the Scheme was "to help [the most] vulnerable refugees in Syria as candidates for resettlement in the United Kingdom as quickly and effectively as possible" (see judgment at [125]). On that reframed basis and having regard to the purported objective of speedy resettlement, she determined that UNHCR "was integral to the aims and operation of the Scheme" and that reliance on it was "closely and rationally connected to the achievement of the Scheme's purpose". That ignored her earlier finding that exclusive reliance on UNHCR necessarily excluded the vast majority of PRS from resettlement under the Scheme. Contrary to her finding, the statistics did resolve the question whether PRS had no chance of resettlement and fully supported the case advanced by the appellants that no UNRWA registered PRS has been able to overcome the barriers to obtaining access to UNHCR's mandate. That is relevant to justification given the more severe impact of the PCP. What had to be justified was the respondent's failure to introduce an alternative, additional mechanism to enable PRS who cannot be referred through UNHCR, to access the Scheme. The appellants discharged the evidential burden on them by producing evidence (as reflected in the UNHCR Resettlement Handbook) that Norway, Canada and the United States have adopted arrangements designed to address the problem of the UNRWA mandate: these countries permit referrals for resettlement to be made outside UNHCR's mandate by NGOs, while retaining UNHCR as the primary referral agency. To the extent that the Judge appeared to have expected the appellants to produce evidence concerning a specific NGO capable of fulfilling this role, that was to reverse the burden of proof and was itself an error.
74. Secondly Mr Jaffey submitted that Elisabeth Laing J made a material error in concluding that the severity of the impact of the PCP on the appellants was "mitigated, to some extent, by the very thing which means that in practice they have such limited access to the Scheme", in other words, by the fact that they "already have their own relief organisation", namely UNRWA. This was wrong and formed no part of the respondent's case. As the judge accepted, UNRWA has no resettlement mandate and there was no evidence that the limited humanitarian assistance to Palestinian refugees living in difficult conditions in, for example, Lebanon, in any way mitigated the denial of the opportunity to be considered for resettlement (a durable solution for the most vulnerable refugees) under the Scheme. Although the Judge held that the fact that "UNRWA's resources, and its capacity to help PRS are limited, is nothing to the point" the reverse is true. It is that lack of resources, coupled with the differential treatment of PRS as compared with Syrian refugees that contributes directly to the particular vulnerability of PRS. Had the availability of a different relief agency been a separate but equal resource that could have formed the basis of a justification argument. However, on any view the mandate of UNRWA is a lesser mandate and not arguably equal.
75. I do not accept these submissions and have concluded that the Judge was entitled (as a matter of fact and law) to make the findings she did on the issue of justification for the reasons that follow.
76. There is no dispute about the proper approach to justification in this context; nor is any criticism made of the Judge's adoption of the four-stage test described by Lord Reed in Bank Mellat v HM Treasury [2013] UKSC 38, [2014] AC 700 at [74]. Further, it is common ground that the burden of establishing justification is on a respondent decision-maker and not a claimant. It is open to respondents to advance an ex post facto justification but courts may approach this with more scepticism than would otherwise be the case if the issue of justification had been considered in advance. It is also clear and not disputed that the fact that the Scheme being challenged is an ex gratia scheme is irrelevant to the proportionality assessment.
77. Against that background, I have concluded that the Judge correctly identified the purpose of the Scheme, which was supported by the King-Fisher witness statements and Parliamentary Statements and debates. There was undoubtedly a very serious humanitarian emergency that was treated as such from the outset. When the Scheme was first announced in Parliament on 29 January 2014, its launch was described as a "new programme to provide emergency sanctuary in the UK for displaced Syrians who are particularly vulnerable". The emergency nature of the scheme and the need for speed in addressing the crisis is both explicit and implicit in the statements and debates. The fact that the Scheme was contemplated to last five years does not in any way undermine that. I am satisfied that urgency was an important element in the decision-making in what was undoubtedly a difficult situation.
78. Moreover, the situation was complex: the Government wished to provide help as soon as possible to a necessarily limited number of particularly vulnerable refugees and had to make use of the best available tools in so doing. To that end, there were good and clear reasons for working with UNHCR as the referral agency under the Scheme, again as the King-Fisher statements explained and the Judge was entitled to find. These reasons included, first, the significant expertise of UNHCR (described by the Judge as the world's foremost resettlement expert with capabilities second to none); and secondly, the significant degree of trust placed by the United Kingdom Government in UNHCR as uniquely trusted to perform functions that would otherwise have fallen on Government, including in relation to collecting bio-data, and carrying out checks and assessments, including to identify former combatants and those guilty of war crimes. As the Judge held:

"122(ii) … A comparison is inherent in the Scheme, which is designed to help the neediest and most vulnerable displaced people, from a large cohort of people all of whom are needy and vulnerable. The comparison requires a close understanding of the position of applicants and of those in a similar position. A consistent approach is desirable, which can be achieved if the same entity choses the applicants. UNHCR has the staff in the relevant regions to do this. It was untenable to suggest that the United Kingdom Government should send officials out to the Middle East to do this. UNHCR's extensive capacity on the ground also made it untenable to suggest that the Home Office should do the necessary checks and assessments from the United Kingdom. UNHCR exists, and is present; it could administer the Scheme immediately."

79. As for the statistical evidence, although it demonstrates that the absolute number of PRS accepted for resettlement under the scheme is vanishingly small, it does not indicate the total number of PRS registered with UNRWA, nor the total number of such PRS to whom the Scheme might apply. There are also no statistics for the numbers of non-PRS who met the vulnerability and other criteria. I can understand in those circumstances, the observation made by the Judge that the statistics did not resolve the question. Nonetheless she proceeded on the basis that although the exclusive mandates did not amount to an absolute exclusion, in practice PRS were "wholly excluded or very nearly so" given those mandates. In other words, she properly recognised the severity of the impact of this aspect of the Scheme on the affected individuals, and weighed that in the balance when assessing justification.
80. That brings me to the challenge to the Judge's finding that there were no less intrusive means that could have been adopted without unacceptably compromising the objective of the Scheme. Before addressing the substance of the arguments advanced by Mr Jaffey on this point, it is important to understand the way in which the issue arose. The claim as originally pleaded was one of direct race discrimination focused on the complete exclusion of the appellants as non-Syrian nationals from the Scheme. Following the widening of the Scheme in July 2017 to include non-Syrian nationals, and the correspondence that followed about how PRS could access the scheme, the appellants produced a "Reply/Update to the court" in October 2017 in response to the "widening decision". In the Update section they summarised their continued denial of access to the Scheme by reference to the exclusive mandate of UNHCR to make referrals, and identified a contrast with "resettlement schemes for refugees from the conflict in Syria run by other countries, which allow refugees to register for resettlement with other agencies and/or directly with their embassies in country." They did not, however, explain how this related to the lawfulness or otherwise of the respondent's decision-making.
81. In the Reply section of the same document, the appellants characterised their exclusion from the expanded scheme as indirect discrimination contrary to section 19 EA 2010 that was not justified. They did not plead less intrusive means that could have been adopted as part of this argument. Furthermore, they did not at any stage seek to amend their original grounds of claim.
82. The respondent's pleading and evidence in response were directed at demonstrating that PRS could in practice access the Scheme. It remained the GLD's understanding at least, as at 3 April 2019, that the real point in issue was the appellants' ability to access the Scheme in practical terms given UNRWA's involvement; and the appellants' suggestion that the United Kingdom ought to amend the Scheme to enable them to access it. The author of the letter dated 3 April 2019 made clear GLD's intention to file a further witness statement from Ms King-Fisher that would, among other things, explain "why the UK does not accept direct applications for resettlement."
83. A further witness statement from Ms King-Fisher dated 25 April 2019 was served. She said:

"3. The Government does not accept direct applications for resettlement under the VPRS but relies upon identification of suitable cases by the UNHCR in accordance with established criteria.

4. The United Kingdom has a well-established relationship with the UNHCR in identifying those living in formal refugee camps, informal settlements and host communities who would benefit most from resettlement to the United Kingdom. The nature of the work that the UNHCR is able to undertake in relation to the VPRS is set out in a briefing paper dated 7 March 2016….

5. The UNHCR is better able to carry out this work in identifying suitable cases for possible resettlement under the VPRS. The UNHCR has a very large number of locally engaged and international staff throughout the regions where it operates and is able to engage with a very large number of individuals, in a way that is simply not open to United Kingdom officials. The UNHCR deploys its operational expertise not only in support of United Kingdom resettlement schemes but also those of other resettlement states.

6. For this reason the Government does not accept direct applications for resettlement."

84. The appellants' skeleton (served at 6:43pm on 25 April 2019) dealt with the issue at paragraph 77 where it was said that there were other less intrusive means available to operate the Scheme in order to permit equal access, for example, the Scheme "could permit self-referrals or referrals from NGOs or similar bodies to be made to the British Embassy in the relevant State or to the government in the UK". A footnote referred to the UNHCR Resettlement Handbook confirming that some "NGOs make resettlement referrals directly to resettlement states".
85. On 29 April 2019 Leigh Day referred again to the Resettlement Handbook (and extracts were attached to this letter) and queried why the respondent's evidence did not address the question of accepting NGO referrals. The GLD response dated 1 May 2019 noted that Leigh Day were seeking to make additional points and said that this was being considered.
86. On 2 May 2019, the respondent served her skeleton argument expressing surprise at the "new and detailed legal arguments, not previously articulated in any form". There was an objection to the admission of expert evidence (that does not concern this court) but the respondent indicated that she would respond to the case now advanced. She said, though, that it was inevitable that her "evidence [did] not deal with all matters as fully as would have been the case if the … case had been properly pleaded".
87. The hearing below was on 8 and 9 May 2019. The Judge required the appellants to set out in writing the precise PCP relied upon and the possible less intrusive means on which they relied. They did so in a document dated 8 May 2019. So far as less intrusive means are concerned, they identified these as including self-referral to the local British Embassy, or Home Office directly; and referral by one or more NGOs active in the region. They referred to the Resettlement Handbook extracts appended to the letter of 29 April 2019.
88. The chronology demonstrates that there was a lack of clarity until very shortly before the substantive hearing about the precise nature of the issues to be determined in the judicial review. In consequence the evidence did not deal as fully and clearly with the issues as it would otherwise have done. In particular, while Ms King-Fisher addressed expressly the problem with self-referral as an alternative referral mechanism, she did not address the suitability or otherwise of NGOs that could fill this role. The criticism made by Mr Jaffey of the evidence and the Judge's finding on less intrusive means must be seen in this context.
89. I have concluded against that background that the Judge did not reverse the statutory burden of proof. She was entitled to be critical of the evidence on alternative means. It was for the respondent to demonstrate that there was no mechanism equivalent to UNHCR that could properly be established without unacceptably compromising the objective of the Scheme. But because of the way that the case developed, it is clear that the evidence did not deal directly with referral by an NGO. The Judge's complaint that there was little if any evidence that any specific NGO had the necessary capacity was a statement of fact and did not involve a reversal of the burden of proof.
90. I have also concluded that there was (just) a sufficient evidential basis to support the finding made by the Judge in this regard The qualities she identified in UNHCR including as the world's foremost resettlement expert and the extent of the functional delegation and trust put in UNHCR by the United Kingdom government in relation to this and other resettlement schemes (that enabled UNHCR uniquely to identify those who had come from Syria, conduct checks and assessments on the ground, and make a comparison based on a close understanding of the position of particular individuals in a large cohort of vulnerable displaced people to identify the neediest and most vulnerable) were inevitably qualities that could not be replicated, or replicated immediately in an NGO. For example, an NGO operating in Lebanon would not have the same comparative data points necessary for identifying the neediest and most vulnerable displaced people from the larger cohort. Furthermore, unlike the other countries referred to in the Resettlement Handbook which appeared to have existing referral relationships with NGOs in the region, the United Kingdom used UNHCR as the sole referral agency for the Scheme and this was consistent with the way it had previously operated its other resettlement schemes. In other words, it did not have the pre-existing referral relationships with NGOs that Norway, Canada and the United States may have done. In these circumstances, and given the urgency and complexity of the process of this very large resettlement scheme from start to finish, as described by Ms King-Fisher, the Judge was entitled to conclude that to operate the Scheme in a different way (through an NGO) would have had serious resource and timing implications. The Judge was entitled to conclude that as with self-referral, referral by an NGO could not have achieved the "security, reliability, speed and consistency which flow from using UNHCR as a gatekeeper".
91. Furthermore, I do not accept that the Judge made a material error in stating that the severity of the impact of the PCP on the appellants was mitigated to some extent by the very thing which limited their access to the Scheme, namely UNRWA's mandate which was specifically designed to assist them on a humanitarian basis as Palestinian refugees in Lebanon. There is nothing in her judgment to suggest that she regarded UNRWA as a separate but equal alternative arrangement to UNHCR, or as a "safety valve" (as discussed in R(TW) v Hillingdon LBC [2019] EWCA Civ 692, [2019] PTSR 1783 at [82]-90]) that operated to eliminate the actual disadvantage complained of by reference to the PCP. The contrary is clearly the case. The judgment read as a whole demonstrates clearly that the Judge was fully alive to UNHCR's different and more extensive mandate, and to the fact that resettlement is the avenue that provides a durable solution for the most vulnerable refugees.
92. Nonetheless, she was entitled to note that some other forms of humanitarian assistance (not including resettlement) were available to Palestinian refugees living in difficult conditions in, for example, Lebanon through this UN refugee agency and that they were not therefore totally without assistance. This point, though it had limited weight, was not irrelevant to the proportionality analysis. What the Judge was really saying was that in a complex crisis involving millions of displaced people, a high percentage of whom were vulnerable, and with ten times more displaced Syrians than Palestinians, it was proportionate to structure a scheme which only had capacity for 20,000 people by targeting it at the biggest part of the pool in the knowledge that vulnerable PRS were not totally without assistance. In the result I accept, as Sir James Eadie submitted, that it was a factor she was entitled to consider as providing some limited support for the United Kingdom's decision that using UNHCR exclusively to make referrals for resettlement was proportionate in the circumstances.
93. Finally, there was nothing inconsistent in the Judge's acceptance that the PCP was a proportionate and objectively justified mechanism to adopt in the circumstances of this case, while at the same time, identifying inadequacies in the respondent's PES. These issues are conceptually distinct and the Judge was well aware of the absence of any justification for the disparate impact on PRS set out in the PES.
94. Accordingly, and for all these reasons, I would dismiss this ground of appeal.

Issue 3: Common law rationality

95. The appellants contended that the respondent's choice of the UNHCR as the exclusive conduit for referrals to the Scheme is irrational at common law for the reasons already identified, and that the direct consequence of that choice is that a class of particularly vulnerable refugees, including the appellants, who do not fall under UNHCR's mandate, are in effect excluded from consideration for resettlement under the Scheme. The failure to establish unlawful indirect discrimination in relation to the Scheme makes it difficult if not impossible for the appellants to succeed on a common law rationality basis. In light of my conclusions set out above, and for the reasons given by the Judge, I would hold that this ground also fails and should be dismissed.

Issue 4: the extra-territorial reach of the section 149(1)(b) duty

96. The complaint made by the appellants in relation to the PSED was that in introducing and in amending the Scheme the respondent failed to comply with the "due regard" duties, as demonstrated by their failure to appreciate the effect of the exclusive referral mechanism on UNRWA registered PRS in countries where UNRWA operates. Elisabeth Laing J held that section 149 EA 2010 has extra-territorial effect but that section 149(1)(a) EA 2010 was not engaged on the facts because no conduct prohibited by the EA 2010 could be committed in this context. Although I have reached a different conclusion in relation to territorial effect, that conclusion is limited to the application of the scheme rules to the appellants' individual cases and does not extend to the making of the Scheme. The Judge's conclusion in relation to that still holds, as does her conclusion in relation to section 149(1)(a) EA 2010 in that connection. So far as the due regard duty identified by section 149(1)(c) EA 2010 is concerned, the Judge concluded that it was not engaged on the facts of this case as regards the appellants or PRS more generally, any more than it would be in the case of others who might benefit from the Scheme. I agree and there is no challenge to this conclusion.
97. That leaves only the question whether section 149(1)(b) EA 2010 (due regard for the advancement of equality of opportunity between persons who share a relevant protected characteristic and persons who do not share it) which the Judge concluded was engaged on the facts, has extraterritorial effect in relation to the making of the Scheme. On this question, Elisabeth Laing J reasoned as follows:

"113. I respectfully disagree with the reasoning of the Divisional Court in Hottak and Hoareau about the territorial reach of section 149. Section 29 of the 2010 Act adopts, if anything, a more powerfully 'functional' approach than that which applies to section 149. Section 29(6) applies to any person when he exercises a public function, while section 149 applies only to those public authorities which are listed in Schedule 19 to the 2010 Act. Section 149 applies to some listed authorities when they exercise all of their functions (section 150(3)), unless they are bodies in respect of which only certain functions are specified in Schedule 19 (section 150(4)).

114. Even if I am wrong about that, and both provisions apply a functional approach (whatever that means), that tells us nothing about Parliament's intention in relation to territoriality. The functions to which a provision applies and its intended territorial reach are conceptually distinct. All the reasons which suggest that Parliament did not intend Part 3 to have other than very exceptional extra-territorial effect apply with as much force to section 149. The approach of the Divisional Court means that even though a public authority cannot breach the substantive provisions of the 2010 Act in the exercise of a particular public function which has only extra-territorial effects, it is nonetheless required, when exercising it, to have due regard to the listed equality needs as respects people who are outside the jurisdiction and whose equality of opportunity, and whose good relations with others, it will, necessarily, have a limited, if any, scope, to influence. A legislative scheme with that effect is incoherent.

115. In R v Manchester Coroner ex p Tal [1985] QB 67 the Divisional Court held, at pages 79E-81D that, on an application for judicial review, a Divisional Court is free to depart from an earlier decision of a Divisional Court, but that it will only do so 'in rare cases.' The Divisional Court added '…we find it difficult to imagine that a single judge exercising this jurisdiction would ever depart from a decision of a divisional court.' If the question of such a departure should arise, it should be listed before a Divisional Court. In the circumstances of this case, therefore, I must follow the approach of the Divisional Court in Hottak and Hoareau and hold that section 149 has extra-territorial effect."

98. Sir James Eadie submitted that these criticisms of this aspect of the decisions in Hottak DC and Hoareau, and the Judge's reasoning, have real force. In Hoareau at [159] the court proceeded on the basis that, because the functions in question in that case did not fall within the exceptions in Schedule 18 EA 2010, section 149(1)(b) EA 2010 must have extraterritorial effect in respect of those functions. That was wrong. Having established that the alleged breach related to persons outside the United Kingdom, the right question is whether there is anything in the EA 2010 to suggest that Parliament intended this duty to have extraterritorial effect. In the absence of any such evidenced intention, the default presumption should have applied. Likewise, the Divisional Court in Hottak DC erred at [60] in its reliance on the fact that "the territorial limitations implicit in section 149(1)(a) follow the application of the substantive part of the Act but otherwise there are no territorial limitations." Here too there was a failure to recognise the important presumption of territorial effect. Sir James Eadie submitted that the starting point must be the presumption of territorial effect and the absence of any provision in section 149 (or elsewhere in the EA 2010) purporting to extend the territorial reach of section 149(1)(b). Where Parliament intended such an extension to apply, other provisions of the EA 2010 explicitly set out the scope of the intended extension: see, for example, section 29(9)-(10).
99. Mr Jaffey resisted those submissions. So far as section 149(1)(b) is concerned, Elisabeth Laing J rightly considered herself bound to apply Hottak DC and Hoareau in holding that the PSED does have extra-territorial effect. He adopted the analysis in those cases, which he submitted were both correctly decided. The requirement to have due regard is directed at achieving better standards of decision-making. This is a procedural and not a substantive duty and there are good reasons in those circumstances for giving this provision a wider rather than a narrower reach than the substantive obligations in the EA 2010, where decisions of public bodies can have effects both here and abroad – see for example R (Elias) v Secretary of State for Defence [2006] 1 WLR 3212 (a compensation scheme for civilians interned during the Japanese occupation of Hong Kong between 1941 and 1945) and Hoareau (a refusal to support resettlement of Chagossians to the British Indian Ocean Territory) – and a decision to give foreign aid or to set up a resettlement scheme should promote equality of opportunity and not the reverse. It is unsurprising that Parliament would require public bodies to have appropriate due regard in accordance with the PSED in such cases; and in practice this is done.
100. Despite the force of these submissions, I am persuaded (albeit with some hesitation) that the criticisms made by the Judge and adopted by Sir James Eadie on behalf of the respondent, are correct. The fact that this provision applies to public authorities in the exercise of their functions does not answer the question of territorial effect any more than it answered it in relation to section 29 EA 2010. The two are conceptually distinct. That the PSED applies to a function exercised in the United Kingdom (for example the making of a policy) does not lead inevitably to the conclusion that the duty is to have regard to persons or matters outside the United Kingdom. The Divisional Court in Hottak DC concluded that in the formulation of policy "it does not matter… that the policy may have an impact wholly or partly outside Great Britain." That is true so far as it goes. The duty will apply to the formulation of the policy whatever its impacts, but the real question is whether the duty extends to having due regard for promoting equality (or fostering good relations) in respect of persons or matters outside the United Kingdom when formulating that policy. The Divisional Court also said that "the territorial limitations implicit in section 149(1)(a) follow the application of the substantive parts of the act, but otherwise there are no territorial limitations". In reaching that conclusion, the normal presumption against extraterritoriality does not appear to have been drawn to the attention of the Divisional Court; and that may explain why the approach adopted appears to reverse the normal presumption without any basis for doing so having been identified. Nor is the question answered by determining whether the exception in Schedule 18 EA 2010 applies to disapply the section 149(1)(b) duty in circumstances where section 29(9) EA 2010 makes clear that the exercise of entry control functions is within the scope of the Act, no matter where the conduct takes place.
101. The starting point, as before, must be the presumption that Parliament legislated for territorial effect only, unless Parliament can reasonably be taken to have intended to legislate for extraterritorial effect. There is nothing in the express words of section 149(1)(b) EA 2010 to suggest that Parliament intended to extend the territorial reach of subsection 1(b) to persons or matters outside the United Kingdom. The points made earlier which suggest that Parliament did not intend Part 3 to have extraterritorial effect save in limited circumstances apply with equal force to this provision.
102. The duty in subsection (1)(b) is explained by section 149(3) and (5) EA 2010 as including the need to remove or minimise disadvantages suffered by people who share a protected characteristic that are connected to that characteristic; the need to take steps to meet the needs of people who share a protected characteristic that are different from the needs of people who do not; and the need to encourage people who share a relevant protected characteristic to "participate in public life or in any other activity in which participation by such persons is disproportionately low". The latter involves tackling prejudice and promoting understanding. I find it difficult to see why or how Parliament could have expected public authorities to take these steps in relation to people outside the United Kingdom in a place where the authority is unlikely to have any real sphere of operation, or in a place or country where different views may be taken on questions of equality and non-discrimination as reflected in local laws, customs and traditions. Certain characteristics that are protected characteristics in Great Britain are far from protected elsewhere and there may be great sensitivity in this regard. It cannot be for a public authority in this country to determine how best to advance equality of opportunity between people subject to foreign law, traditions and customs. These points reinforce the force of the normal presumption in this case.
103. Although, as Mr Jaffey submitted, the duty is a procedural one, it seems to me that the purpose of this legislation can properly and effectively be achieved without implying extraterritorial effect. To require a public authority to have regard to the equality needs of people who are outside the jurisdiction and whose equality of opportunity and good relations with others it will necessarily have limited if any scope to influence is not implicit in the statutory scheme and as Elisabeth Laing J noted, would make it "incoherent". It would mean that section 149(1)(b) requires a public authority to have due regard to the need to advance equality for anyone anywhere in the world, regardless of the public authority's actual capacity to advance that person's equality of opportunity. That seems to me to be a meaningless duty. There is nothing to suggest that was Parliament's intention.
104. For a claimant who is not a British citizen, does not live or work here, and has no other link at all with the United Kingdom other than a wish to benefit from a policy of the Secretary of State, the exceptional connection with the United Kingdom which would otherwise be required to establish a territorial extension to the EA 2010 is absent. The fact that government practice may extend to conducting equality impact assessments where the effects are felt outside the United Kingdom is not a guide to the interpretation of this provision, and does not lead me to a different conclusion.
105. For these reasons accordingly I would allow the cross-appeal on this point.

Conclusion

106. Accordingly and if my Lords agree, the appeal should be dismissed for the reasons given above; and the cross-appeal allowed.

Warby LJ:

107. I agree with the proposed disposal of the appeal and cross-appeal, and with the conclusions and reasoning of Simler LJ. Like her, I agree with the reasoning of Underhill LJ in paragraph 109 below.

Underhill LJ:

108. I agree with Simler LJ's proposed disposal of the appeal and the cross-appeal but I wish to say something of my own about issues 1, 2 and 4.
109. As regards issue 1, I agree with Simler LJ's conclusion and, with some hesitation, also with her reasoning.  The hesitation is because, while I have no difficulty in reading the reference in section 29(9) to the grant of entry clearance as applying also to matters forming part of the decision-making process in an individual case, I am less comfortable about treating it as applying to the adoption of a general policy about the grant of entry clearance in a class of cases.  In the end I am persuaded by the points which she makes at paragraphs 65-69.  But I think I should say that, even if I were not, there may be another route to the same result.  Even if the policy itself may not fall within the terms of section 29(9), it need not follow that the fact that it applies only to persons outside the United Kingdom takes it outside the scope of section 29(6).  On Lawson v Serco principles the question is one of presumed Parliamentary intention.  If the decision whether to grant entry clearance in the cases of individuals who are outside the United Kingdom is, by virtue of section 29(9), within "the territorial grasp" of section 29(6), it might be thought reasonable to infer that Parliament intended that the same should be the case as regards the conditions of a policy governing how such decisions are to be made.  However, it is unnecessary to reach a concluded view about this. 
110. As regards issue 2, at first blush there are obviously very good reasons why the Secretary of State should have wished to use UNHCR as the sole gatekeeper for access to the Scheme, both in its original form and as expanded, as Simler LJ explains at paragraph 78. The difficulty, however, is the fact that for particular historical reasons UNHCR has no mandate covering PRS in Lebanon, Jordan, Syria, the West Bank and Gaza ("the UNRWA territories"). If it had been clear from the start that the real issue as regards justification was whether, that being so, the Secretary of State should have put in place other arrangements which would have allowed PRS in those territories access to the Scheme, I have no doubt that the Judge would have been presented with full evidence addressing the practicability and desirability of such arrangements. But, as Simler LJ has explained at paragraphs 80-88, the issue only clearly emerged at a very late stage and the evidential material before the Judge was exiguous: the story is indeed a good illustration of the perils of "rolling judicial review". Neither party having sought an adjournment, the Judge had to do her best on the basis of such evidence as there was. In those circumstances I agree that for the reasons given by Simler LJ at paragraphs 90-92 the conclusion that she reached was open to her.
111. It is a surprising feature of the case that at the time that the expanded Scheme was promulgated the Secretary of State was unaware of the fact that UNHCR would be unable to act as the gatekeeper in the case of the majority of the PRS whom the Scheme was intended to assist; and that it took her some time to appreciate, or at least acknowledge, that there was a real problem. It follows that any attempt to justify the discriminatory impact of that feature of the Scheme cannot reflect judgments which she made at the time, and Mr Jaffey understandably drew our attention to the authorities which urge caution in such a case. But it is clear that ex post facto justification is not inherently inadmissible, and I do not believe that in the circumstances of this case that feature is sufficient to undermine the Judge's reasoning.
112. I turn to issue 4. Like Simler LJ, I have not found this entirely easy, and I naturally attach weight to the views of the Divisional Court in Hottak and Hoareau (comprising Burnett LJ and Irwin J in the former and Singh LJ and Carr J in the latter). But in the end I have come to the same conclusion as her. The starting-point is to identify exactly what the issue is. As Simler LJ says at paragraph 100, the real question is whether section 149(1)(b) imposes a duty on a public authority to have regard to the need to "advance equality of opportunity" (in the sense defined in subsection (3)) between persons sharing a relevant protected characteristic and those who do not, where the persons in question do not live in the United Kingdom. For the reasons that she gives at paragraph 102, I do not believe that Parliament can have intended to impose such a duty. That conclusion is not inconsistent with my view that section 29(6) has extraterritorial scope in the circumstances of this case. The duty under section 149(1)(b) is entirely different in character from the substantive duty not to discriminate provided for by Parts 3-7 of the Act.
_____________________________________

Appeal Refs: C4/2019/1685 and 1707

IN THE COURT OF APPEAL

CIVIL DIVISION

BEFORE: The Vice President, Simler and Warby LJJ
BETWEEN: THE QUEEN On the application of (1) SALEH MOHAMMAD TURANI (2) HAIFAA MAROUF

Appellants

-and- SECRETARY OF STATE FOR THE HOME DEPARTMENT

Respondent

_____________________________________ ORDER _____________________________________

UPON HEARING Mr B Jaffey QC for the Appellants and Sir James Eadie QC for the Respondent

IT IS ORDERED THAT:

113. The Appellants' appeal be dismissed;
114. The Respondent's cross-appeal be allowed;
115. The claim for judicial review be dismissed;
116. The Appellants do pay 75% of the Respondent's costs of the appeal, with such costs to be subject to detailed assessment if not agreed, and subject to section 26 of the Legal Aid, Sentencing and Punishment of Offenders Act 2012;
117. There shall be a detailed assessment of the Appellants' publicly funded costs; and
118. Any application for permission to appeal must be made within 14 days of this Order.