HA (Iraq) v Secretary of State for the Home Department (Rev 1)

These two appeals have been listed together because they raise similar issues. The Appellants are non-British nationals – both, though this is coincidental, from Iraq – who came to this country many years ago and have lived here ever since. Both are in settled relationships with women of British nationality and have young children who are likewise British citizens. Both committed criminal offences for which they were sentenced to terms of imprisonment of sixteen and twelve months respectively. Those sentences attracted the automatic deportation provisions of section 32 of the UK Borders Act 2007. In both cases the Secretary of State made a deportation order but the Appellant appealed to the First-tier Tribunal (“the FTT”). The appeals are subject to the terms of Part 5A of the Nationality, Immigration and Asylum Act 2002 (in particular section 117C), and Part 13 of the Immigration Rules, of which I give more details below.

In both cases the appeals were successful, but the Secretary of State in her turn appealed to the Upper Tribunal (“the UT”). Both appeals were allowed and the decisions directed to be re-made by the UT at a further hearing. That hearing took place in RA’s case on 13 February 2019 before Lane J (the President of the Immigration and Asylum Chamber), UTJ Gill and UTJ Coker. The hearing in HA’s case was two days later, before Lane J and UTJ Gill. The decisions in both cases were likewise promulgated only a few days apart – RA on 4 March and HA on 8 March. The two appeals, together with two others (one of which was MS (Philippines) [2019] UKUT 122 (IAC)), were listed on successive days in order for the UT, in constitutions chaired by the President, to give authoritative guidance on various issues about section 117C arising out of the then recent decision of the Supreme Court in KO (Nigeria) v Secretary of State for the Home Department [2018] UKSC 53, [2018] 1 WLR 5273. The decision in RA was the principal vehicle for that guidance, though some points of general application are dealt with in the other decisions.

In both cases the UT allowed the Secretary of State’s appeal. In September last year I gave permission to appeal to this Court. I considered that there were arguable grounds of appeal in both cases, but the main reason why I thought that the second appeals test was satisfied was that I believed that this Court should have the opportunity to consider the guidance given by the UT on the issues of general application.

RA was represented before us by Mr Danny Bazini, leading Ms Jessica Smeaton, and HA by Mr Ramby de Mello. Mr Bazini and Mr Pilgerstorfer appeared before the UT in RA, while in HA in the UT the Appellant was unrepresented and the Secretary of State was represented by Mr Zane Malik. The Appellants helpfully lodged a consolidated skeleton argument in the names of all three counsel which dealt with the legal framework in an undifferentiated way but then made separate submissions about the two cases.

As appears above, in the UT the names of both Appellants were anonymised, although that had not been the case in the FTT. Mr Bazini told us that in RA he had not asked for anonymisation, but in HA the UT’s substantive Decision and Reasons is endorsed with a rubric headed “Anonymity” which records that an anonymity order had been made “in order to protect the identities of the appellant’s minor children”. I confess to some concern about the over-use of anonymisation in this field, and I am not entirely persuaded that the mere fact that a foreign criminal has minor children is sufficient to justify not using his or her full name. However, it seems that most of the appeals concerning deportation of foreign criminals have been anonymised, not only in the UT but in this Court and the Supreme Court; and in all the circumstances I am content not to seek to go behind the order of the UT.

I will start by setting out the background law. I will then address the question of the effect of KO (Nigeria). Finally, I will consider the individual cases.

Section 32 (4) of the 2007 Act provides that “the deportation of a foreign criminal is conducive to the public good”. Sub-section (5) requires the Secretary of State to make a deportation order in respect of a “foreign criminal”, defined (so far as relevant for our purposes) as a person who is not a British citizen and who is convicted in the UK of a criminal offence for which they are sentenced to a period of imprisonment of at least twelve months. That obligation is subject to various exceptions set out in section 33. We are concerned only with the exception in section 33 (2) (a), which applies “where removal of the foreign criminal in pursuance of the deportation order would breach … a person's Convention rights”.

The “Convention rights” referred to in section 33 (2) are of course rights under the European Convention on Human Rights (“the ECHR”). Typically, and on these appeals, the relevant Convention right is the right under article 8 to respect for private and family life. Paragraph 2 provides that there shall be no interference with that right “except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others” – for short, save where it is justified by the public interest.

Section 55 of the Borders, Citizenship and Immigration Act 2009 requires the Secretary of State to make arrangements to ensure that her functions as regards

The effect of section 55 in the immigration and extradition fields has been examined by the Supreme Court in a trio of cases – ZH (Tanzania) v Secretary of State for the Home Department [2011] UKSC 4, [2011] 2 AC 166; H (H) v Deputy Prosecutor of the Italian Republic [2012] UKSC 25, [2013] 1 AC 338; and Zoumbas v Secretary of State for the Home Department [2013] UKSC 74, [2013] 1 WLR 3690. At para. 10 of his judgment in Zoumbas Lord Hodge summarised the position in seven propositions, as follows:

“(1)

The best interests of a child are an integral part of the proportionality assessment under article 8 ECHR;

(2)

In making that assessment, the best interests of a child must be a primary consideration, although not always the only primary consideration; and the child's best interests do not of themselves have the status of the paramount consideration;

(3)

Although the best interests of a child can be outweighed by the cumulative effect of other considerations, no other consideration can be treated as inherently more significant;

(4)

While different judges might approach the question of the best interests of a child in different ways, it is important to ask oneself the right questions in an orderly manner in order to avoid the risk that the best interests of a child might be undervalued when other important considerations were in play;

(5)

It is important to have a clear idea of a child's circumstances and of what is in a child's best interests before one asks oneself whether those interests are outweighed by the force of other considerations;

(6)

To that end there is no substitute for a careful examination of all relevant factors when the interests of a child are involved in an article 8 assessment; and

(7)

A child must not be blamed for matters for which he or she is not responsible, such as the conduct of a parent.”

Point (2) is of particular importance in cases of this kind. Lady Hale stated the position slightly more fully in paras. 11 and 15 of her judgment in H (H), where she said, referring to her judgment in ZH (Tanzania):

“11.

I pointed out that ‘despite the looseness with which these terms are sometimes used, “a primary consideration” is not the same as “the primary consideration”, still less as “the paramount consideration”’.

…

15.

ZH (Tanzania) made it clear that in considering article 8 in any case in which the rights of a child are involved, the best interests of the child must be a primary consideration. They may be outweighed by countervailing factors, but they are of primary importance.”

The Appellants in their joint skeleton argument refer to a passage in the judgment of the European Court of Human Rights (“the ECtHR”) in Jeunesse v

Part 5A of the 2002 Act was introduced by the Immigration Act 2014 with effect from 28 July 2014. It is headed “Article 8 of the ECHR: public interest considerations”. It comprises four sections – 117A-117D.

Section 117A provides, so far as material for our purposes, that in considering whether an interference with a person's right to respect for their private and family life is justified under article 8 (2) (defined as “the public interest question”)

“the court or tribunal must (in particular) have regard—

(a)

in all cases, to the considerations listed in section 117B, and

(b)

in cases concerning the deportation of foreign criminals to the considerations listed in section 117C”.

The principal issues before us arise under section 117C, but it is necessary to set out parts of section 117B also. It is headed “Article 8: public interest considerations applicable in all cases”. It reads, so far as material for our purposes:

“(1)-(3) …

(4)

Little weight should be given to—

(a)

a private life, or

(b)

a relationship formed with a qualifying partner,

that is established by a person at a time when the person is in the United Kingdom unlawfully.

(5)

Little weight should be given to a private life established by a person at a time when the person’s immigration status is precarious.

(6)

In the case of a person who is not liable to deportation, the public interest does not require the person's removal where—

(a)

the person has a genuine and subsisting parental relationship with a qualifying child, and

(b)

it would not be reasonable to expect the child to leave the United Kingdom.”

Section 117C is headed “Article 8: additional considerations in cases involving foreign criminals”. It reads, so far as material:

“(1)

The deportation of foreign criminals is in the public interest.

The terms “qualifying partner” and “qualifying child” in sections 117B and 117C are defined in section 117D (1) as, respectively,

“a person who is under the age of 18 and who (a) is a British citizen, or (b) has lived in the United Kingdom for a continuous period of seven years or more”

and

“a partner who (a) is a British citizen, or (b) is settled in the United Kingdom (within the meaning of the Immigration Act 1971 …)”.

At the risk of spelling out the obvious, the effect of section 117C is to prescribe different approaches to “the public interest question” – that is, the justification issue under article 8 – by reference to the length of the sentence imposed. Specifically:

At the same time as the coming into effect of Part 5A, Part 13 of the Immigration Rules was amended correspondingly: see paras. 22-26 below.

Two difficulties about the construction of section 117C (6) were resolved by this Court in NA (Pakistan) v Secretary of State for the Home Department [2016] EWCA Civ 662, [2017] 1 WLR 207, in which judgment was handed down on 29 June 2016. I take them in turn.

First, on a literal reading sub-section (6) applies only to serious offenders and not to medium offenders, with the result that in such a case, if neither of the Exceptions applies, deportation must be treated as justified whatever “very compelling circumstances” there might be. However, the Court held that such a reading would be neither rational nor consistent with the Convention and cannot have been what Parliament intended; it would also be inconsistent with paragraph 398 of the Immigration Rules (see below). It held that it was necessary to read sub-section (6) as applying equally in the case of medium offenders governed by sub-section (3): see paras. 25-27 of its judgment given by Jackson LJ. The need for that correction was not controversial, and indeed the Secretary of State positively invited the Court to hold as it did.

Secondly, the phrase in section 117C (6) “very compelling circumstances, over and above those described in Exceptions 1 and 2” might be read as excluding circumstances which fell within the scope of those Exceptions, so that (for example) potential deportees could not rely on the effect of their deportation on a qualifying child, even if it would otherwise be material to the assessment of whether there were very compelling circumstances outweighing the public interest in deportation. The Court held that that was not what was meant. I say more about this at paras. 34-35 below.

Part 13 of the Immigration Rules deals with deportation: as noted above, it was amended in July 2014. Paragraphs A398-400 are headed “Deportation and Article 8”. The effect of section 117C is substantially reproduced in paragraphs A398-399A, though in more detail and with a different and rather clumsy structure. As this Court put it in para. 28 of its judgment in NA (Pakistan):

“Paragraphs 399 and 399A of the 2014 rules refer to the same subject matter as Exceptions 1 and 2 in section 117C, but they do so in greater detail.”

The Rules and the statute are plainly intended to have the same effect and should be construed so as to achieve that result.

For present purposes I can summarise the structure as follows:

Paragraph 399, which contains the equivalent to Exception 2, is described as applying where either the potential deportee

“(a)

… has a genuine and subsisting parental relationship with a child under the age of 18 years who is in the UK, and

(i)

the child is a British Citizen; or

(ii)

the child has lived in the UK continuously for at least the 7 years immediately preceding the date of the immigration decision;

And in either case

(a)

it would be unduly harsh for the child to live in the country to which the person is to be deported; and [emphasis supplied]

(b)

it would be unduly harsh for the child to remain in the UK without the person who is to be deported; …”

or he (Footnote: ²)

“(b)

… has a genuine and subsisting relationship with a partner who is in the UK and is a British Citizen or settled in the UK, and

(i)

the relationship was formed at a time when the person (deportee) was in the UK lawfully and their immigration status was not precarious; and

(ii)

it would be unduly harsh for that partner to live in the country to which the person is to be deported, because of compelling circumstances over and above those described in paragraph EX.2. of Appendix FM; and [emphasis supplied]

(iii)

it would be unduly harsh for that partner to remain in the UK without the person who is to be deported.”

I will refer to (a) as “the parent case” and (b) as “the partner case”.

Although there are other respects in which the formulation in paragraph 399 is fuller than that of Exception 2 in section 117C (5), there is only one which I need note for present purposes. It will be seen that, whereas section 117C (5) refers simply to “the effect of C’s deportation on the partner or child [being] unduly harsh”, the equivalent provisions of paragraph 399 (that is, points (a) and (b) under the parent case and points (ii) and (iii) under the partner case) identify two distinct scenarios – the scenario where the child/partner would go with the deportee (“the go scenario”) and the scenario where they would stay behind without him (“the stay scenario”). It will be seen (as italicised by me) that the two scenarios in both the parent and the partner case are linked by an “and”. There is an issue before us about the effect of this wording which I consider at paras. 70-76 below: in short, my conclusion is that “and” means “and”.

It will be useful to make some general observations about the purpose and effect of the scheme which now applies under Part 5A of the 2002 Act and Part 13 of the Rules. I can do so mainly by reference to the careful exposition in the judgment of this Court in NA (Pakistan). There have by now been several further decisions of this Court and the Supreme Court applying these provisions and considering some particular points, but NA (Pakistan) remains the fullest overall guide. I should emphasise that I am not attempting a comprehensive summary of the law in this difficult area. I focus only on the points relevant to the issues in these appeals.

The starting-point is that the purpose of the statutory scheme is to require decisionmakers to adopt a structured approach to the article 8 issues raised by the removal of a foreign national – that is, whether it will constitute a disproportionate interference with, and thus a breach of, their article 8 rights – and one which ensures that due weight is given to the public interest. It is no part of its purpose to prevent the proper application of article 8. This is clearly stated in para. 26 of the judgment in NA (Pakistan) and again in para. 38, quoted below. Following from that, the statutory structure is a “complete code” in the sense that the entirety of the proportionality assessment required by article 8 can and must be conducted within it: that point is clearly made in paras. 35 and 36.

It follows that the Strasbourg case-law about the application of article 8 in cases of this kind must and can be accommodated within the statutory structure. Important guidance about removals generally is given in Jeunesse, to which I have already referred, and there are three well-known cases concerning foreign criminals – Boultif v Switzerland (2001) 33 EHRR 50, Ȕner v Netherlands (2007) 45 EHRR 14, and Maslov v Austria [2009] INLR 47. At para. 38 of its judgment in NA (Pakistan) the Court said:

“… [T]he Secretary of State and the tribunals and courts will have regard to the Strasbourg jurisprudence when applying the tests set out in our domestic legislation. For example, a tribunal may be considering whether it would be ‘unduly harsh’ for a child to remain in England without the deportee; or it may be considering whether certain circumstances are sufficiently ‘compelling’ to outweigh the high public interest in deportation of foreign criminals. Anyone applying these tests (as required by our own rules and legislation) should heed the guidance contained in the Strasbourg authorities. As we have stated above, the scheme of Part 5A of the 2002 Act and paras. 398-399A of the 2014 rules is to ensure compliance with the requirements of Article 8 through a structured approach, which is intended to ensure that proper weight is given to the public interest in deportation whilst also having regard to other relevant factors as identified in the Strasbourg and domestic caselaw. The new regime is not intended to produce violations of Article 8.”

However, it is to be noted that the Court goes on at para. 39 to make the point that “assessments under Article 8 may not lead to identical results in every ECHR contracting state”, reflecting the margin of appreciation which the Convention allows in their assessment of the public interest.

Turning specifically to the case of foreign criminals, the effect of section 117C can be summarised as follows:

Logically it follows that the correct decision-making structure in the case of a medium offender is, as the Court said in NA (Pakistan), at para. 36:

“In relation to a medium offender, first see whether he falls within Exception 1 or Exception 2. If he does, then the Article 8 claim succeeds. If he does not, then the next stage is to consider whether there are ‘sufficiently compelling circumstances, over and above those described in Exceptions 1 and 2’. If there are, then the Article 8 claim succeeds. If there are not, then the Article 8 claim fails.”

It will be convenient to refer to the second stage as the exercise “required by section 117C (6)” or similar phrases, but that is arguably slightly misleading. The second stage is necessary not because of section 117C (6) but because the effect of article 8 is that a proportionality assessment is required in every case (at least where the issue is raised): what section 117C (6) does is to prescribe the weight that has to be given to the public interest in deportation when carrying out that assessment (in a case where neither Exception applies).

The effect of the phrase “very compelling circumstances over and above those described in Exceptions 1 and 2”, and the nature of the exercise required by section 117C (6) as it applies both to medium offenders and to serious offenders, are carefully discussed at paras. 28-34 of NA (Pakistan). It is unnecessary that I quote that discussion in full here, but I should note four points applicable to the case of a medium offender.

First, the discussion is underpinned by the fundamental point of principle which the Court identifies at para. 22 of its judgment, as follows:

“Section 117C (1) of the 2002 Act, as inserted by the 2014 Act, restates that the deportation of foreign criminals is in the public interest. The observations of Laws LJ in SS (Nigeria) [[2013] EWCA Civ 550,

[2014] 1 WLR 998], concerning the significance of the 2007 Act, as a particularly strong statement of public policy, are equally applicable to the new provisions inserted into the 2002 Act by the 2014 Act. Both the courts and the tribunals are obliged to respect the high level of importance which the legislature attaches to the deportation of foreign criminals.”

It is because of the high level of importance attached by Parliament to the deportation of foreign criminals that, where neither Exception 1 nor Exception 2 applies, the public interest in deportation can only be outweighed by very compelling circumstances.

Secondly, the Court’s explanation of the effect of the phrase “over and above those described in Exceptions 1 and 2”, at para. 29, reads as follows:

“The phrase used in section 117C (6), in para. 398 of [the Immigration Rules] and which we have held is to be read into section 117C (3) does not mean that a foreign criminal facing deportation is altogether disentitled from seeking to rely on matters falling within the scope of the circumstances described in Exceptions 1 and 2 when seeking to contend that ‘there are very compelling circumstances, over and above those described in Exceptions 1 and 2’. … [A] foreign criminal is entitled to rely upon such matters, but he would need to be able to point to features of his case of a kind mentioned in Exceptions 1 and 2 (and in paras. 399 or 399A of [the Rules]), or features falling outside the circumstances described in those Exceptions and those paragraphs, which made his claim based on Article 8 especially strong.”

That passage is expressed to cover the case of both serious and medium offenders. At para. 32 the Court specifically addresses the case of medium offenders, as follows:

“… [I]n the case of a medium offender, if all [the potential deportee] could advance in support of his Article 8 claim was a ‘near miss’ case in which he fell short of bringing himself within either Exception 1 or Exception 2, it would not be possible to say that he had shown that there were ‘very compelling circumstances, over and above those described in Exceptions 1 and 2’. He would need to have a far stronger case than that by reference to the interests protected by Article 8 to bring himself within that fall back protection. But again, in principle there may be cases in which such an offender can say that features of his case of a kind described in Exceptions 1 and 2 have such great force for Article 8 purposes that they do constitute such very compelling circumstances, whether taken by themselves or in conjunction with other factors relevant to Article 8 but not falling within the factors described in Exceptions 1 and 2. The decision maker, be it the Secretary of State or a tribunal, must look at all the matters relied upon collectively, in order to determine whether they are sufficiently compelling to outweigh the high public interest in deportation.”

Those two passages make clear that, in carrying out the full proportionality assessment which is necessary where the Exceptions do not apply, facts and matters that were relevant to the assessment of whether either Exception applied are not “exhausted” if the conclusion is that they do not. They remain relevant to the overall assessment, and could be sufficient to outweigh the public interest in deportation either, if specially strong, by themselves (Footnote: ³) or in combination with other factors.

Thirdly, at para. 33 the Court says:

“Although there is no ‘exceptionality’ requirement, it inexorably follows from the statutory scheme that the cases in which circumstances are sufficiently compelling to outweigh the high public interest in deportation will be rare. The commonplace incidents of family life, such as ageing parents in poor health or the natural love between parents and children, will not be sufficient.”

This passage makes a point which appears often in the case-law. But it is important to bear in mind that it is directed at the exercise under section 117C (6). The Court was not saying that it would be rare for cases to fall within section 117C (5).

Fourthly, at para. 34 the Court addresses the relevance of the best interests of any children affected by the deportation of a foreign criminal. It says:

“The best interests of children certainly carry great weight, as identified by Lord Kerr in H (H) v Deputy Prosecutor of the Italian Republic [2012] UKSC 25; [2013] 1 AC 338 at [145]. Nevertheless, it is a consequence of criminal conduct that offenders may be separated from their children for many years, contrary to the best interests of those children. The desirability of children being with both parents is a commonplace of family life. That is not usually a sufficiently compelling circumstance to outweigh the high public interest in deporting foreign criminals. …”

Again, this is a point frequently made in the case-law; but, again, it should be borne in mind that, as the reference to a “sufficiently compelling circumstance” shows, the final sentence relates only to the exercise under section 117C (6).

I have not so far referred to authorities about the regime which preceded the coming into force of Part 5A in 2014 and the associated changes to the Rules. However, as

The most authoritative exposition of the principles underlying the old regime can be found, two years after it had been superseded and even some months later than NA (Pakistan), in the decision of the Supreme Court in Hesham Ali v Secretary of State for the Home Department [2016] UKSC 60, [2016] 1 WLR 4799. It is authoritative on the points of principle underlying both regimes and was so treated in Akinyemi no. 2 (see paras. 46-50). That being so, I should say that I can see nothing in the judgments of the majority inconsistent with the approach taken by this Court in NA (Pakistan) as discussed above. At para. 26 of his judgment Lord Reed summarises the effect of the Strasbourg case-law about foreign criminals, and at para. 33, like this Court in NA (Pakistan), he makes it clear that the factors referred to in those cases need to be taken into account in the assessment of the proportionality of the deportation of foreign offenders (whether or not they are “settled migrants”).

Reference to the previous case-law is important for the purpose of a particular point made by the Appellants in these appeals. It will be seen that in para. 32 of its judgment in NA (Pakistan) this Court expresses the test under section 117C (6) as being whether the circumstances relied on by the potential deportee “are sufficiently compelling to outweigh the high public interest in deportation”; and it uses the same formulation in paras. 33 and 34 (see paras. 36-37 above). The Appellants contend that that is the only correct formulation, and that it is dangerous to refer simply to

“very compelling circumstances”. It would, to say the least, be surprising if it were wrong to use the very language of the statute; but in any event the position becomes clear when the development of the case-law is understood. This Court in NA

(Pakistan) took the language of “sufficiently compelling” from the decision in MF (Nigeria). Paragraph 398 of the pre-2014 Rules had used the phrase “exceptional circumstances”. At para. 42 of its judgment in MF the Court said that that did not mean that a test of exceptionality was to be applied (a point repeated in NA (Pakistan) – see para. 36 above) and continued:

“Rather …, in approaching the question of whether removal is a proportionate interference with an individual's Article 8 rights, the scales are heavily weighted in favour of deportation and something very compelling (which will be ‘exceptional’) is required to outweigh the public interest in removal [emphasis supplied].”

At para. 46 it expressed the same point slightly differently, referring to

“circumstances which are sufficiently compelling (and therefore exceptional) to outweigh the public interest in deportation [again, emphasis supplied]”. The effect is clear: circumstances will have to be very compelling in order to be sufficiently compelling to outweigh the strong public interest in deportation. That remains the case under section 117C (6).

Both Appellants contend that the effect of their deportation on their children would be “unduly harsh”, within the meaning of section 117C (5) – i.e. Exception 2 – and paragraph 399 (a) of the Rules. There is an issue before us about the height of the threshold which that phrase sets. The meaning of “unduly harsh” was considered in KO (Nigeria), which was one of four appeals raising issues about the interpretation of Part 5A which were heard together before the Supreme Court in April 2018. Its decision was given on 24 October 2018. The only judgment, with which the other members of the Court agreed, was given by Lord Carnwath.

It is important to bear in mind, and is perhaps rather unfortunate for our purposes, that the actual issue in KO (Nigeria) was a very specific one, namely whether the word “unduly” referred back to sub-section (2) of section 117C and thus required what

Paras. 1-11 of Lord Carnwath’s judgment are introductory. At paras. 12-15 he sets out the correct approach to the interpretation of Part 5A, and the associated provisions of the Rules. At para. 15 he says:

“I … start from the presumption, in the absence of clear language to the contrary, that the provisions are intended to be consistent with the general principles relating to the ‘best interests’ of children, including the principle that ‘a child must not be blamed for matters for which he or she is not responsible, such as the conduct of a parent’ (see Zoumbas v Secretary of State for the Home Department [2013] UKSC 74, [2013] 1 WLR 3690, para 10 per Lord Hodge).”

At paras. 20-23 Lord Carnwath considers the language of section 117C, and more particularly sub-section (5), as regards the relative seriousness issue. At para. 21 he analyses the wording of Exception 1, describing it as “self-contained”, and at para. 22 he concludes that the same is true of Exception 2. He continues, at para. 23:

“On the other hand, the expression ‘unduly harsh’ seems clearly intended to introduce a higher hurdle than that of ‘reasonableness’ under section 117B (6), taking account of the public interest in the deportation of foreign criminals. Further the word ‘unduly’ implies an element of comparison. It assumes that there is a ‘due’ level of ‘harshness’, that is a level which may be acceptable or justifiable in the relevant context. ‘Unduly’ implies something going beyond that level. The relevant context is that set by section 117C (1), that is the public interest in the deportation of foreign criminals. One is looking for a degree of harshness going beyond what would necessarily be involved for any child faced with the deportation of a parent. What it does not require in my view (and subject to the discussion of the cases in the next section) is a balancing of relative levels of severity of the parent’s offence, other than is inherent in the distinction drawn by the section itself by reference to length of sentence. Nor (contrary to the view of the Court of Appeal in IT (Jamaica) v Secretary of State for the Home Department [2016] EWCA Civ 932, [2017] 1 WLR 240, paras 55, 64) can it be equated with a requirement to show ‘very compelling reasons’. That would be in effect to replicate the additional test applied by section 117C (6) with respect to sentences of four years or more.”

That is an important passage, and it is necessary to identify exactly what Lord Carnwath is and is not saying.

The starting point is that the question to which the reasoning is directed is whether the word “unduly” imports a requirement to consider “the severity of the parent’s offence”: that, as I have said, was the actual issue in the appeal. Lord Carnwath’s conclusion is that it does not: see the sentence beginning “What it does not require …”. The reason why there is no such requirement is that the exercise required by Exception 2 is “self-contained”. I should note at this point that it follows that it is irrelevant whether the sentence was at the top or the bottom of the range between one year and four: as Lord Carnwath says, the only relevance of the length of the sentence is to establish whether the foreign criminal is a medium offender or not.

In order to establish that the word “unduly” was not directed to the relative seriousness issue it was necessary for Lord Carnwath to say to what it was in fact directed. That is what he does in the first part of the paragraph. The effect of what he says is that “unduly” is directed to the degree of harshness required: some level of harshness is to be regarded as “acceptable or justifiable” in the context of the public interest in the deportation of foreign criminals, and what “unduly” does is to provide that Exception 2 will only apply where the harshness goes beyond that level. Lord Carnwath’s focus is not primarily on how to define the “acceptable” level of harshness. It is true that he refers to a degree of harshness “going beyond what would necessarily be involved for any child faced with the deportation of a parent”, but that cannot be read entirely literally: it is hard to see how one would define the level of harshness that would “necessarily” be suffered by “any” child (indeed one can imagine unusual cases where the deportation of a parent would not be “harsh” for the child at all, even where there was a genuine and subsisting relationship). The underlying concept is clearly of an enhanced degree of harshness sufficient to outweigh the public interest in the deportation of foreign criminals in the medium offender category.

Lord Carnwath then turns more particularly to the case of KO. After summarising the facts, at paras. 27-32 he reviews the previous case-law on the relative seriousness issue. He maintains the view on that issue which he had reached by reference to the statutory language (see above). The only part that is relevant for our purposes is para. 27, where he says:

“Authoritative guidance as to the meaning of ‘unduly harsh’ in this context was given by the Upper Tribunal (McCloskey J President and UT Judge Perkins) in MK (Sierra Leone) v Secretary of State for the Home Department [2015] UKUT 223 (IAC), [2015] INLR 563, para 46, a decision given on 15 April 2015. They referred to the ‘evaluative assessment’ required of the tribunal:

‘By way of self-direction, we are mindful that “unduly harsh” does not equate with uncomfortable, inconvenient, undesirable or merely difficult. Rather, it poses a considerably more elevated threshold. “Harsh” in this context denotes something severe, or bleak. It is the antithesis of pleasant or comfortable. Furthermore, the addition of the adverb “unduly” raises an already elevated standard still higher.’”

It is clear that by describing it as “authoritative” Lord Carnwath means to endorse the UT’s self-direction in MK (Sierra Leone), which is consistent with his own explanation of the effect of “unduly” at para. 23. He goes on to note that that selfdirection was followed in the later case of MAB (USA) v Secretary of State for the Home Department [2015] UKUT 435.

Although it is not directly relevant for present purposes, I should note, because I shall have to return to it later, that immediately following that passage Lord Carnwath goes on to say:

“On the facts of that particular case [i.e. MK], the Upper Tribunal held that the test was satisfied:

‘Approached in this way, we have no hesitation in concluding that it would be unduly harsh for either of the two seven year old British citizen children concerned to be abruptly uprooted from their United Kingdom life setting and lifestyle and exiled to this struggling, impoverished and plague stricken West African state. No reasonable or right thinking person would consider this anything less than cruel.’

This view was based simply on the wording of the subsection, and did not apparently depend on any view of the relative severity of the particular offence. I do not understand the conclusion on the facts of that case to be controversial.”

That is not quite as straightforward as it appears. The UT in MK considered both the scenario where the appellant’s children accompanied him to Sierra Leone and the scenario where they stayed in the UK. The passage quoted by Lord Carnwath refers to the former scenario, and the UT’s conclusion about it was obviously right. Lord Carnwath does not quote the UT’s conclusion on the latter scenario, which it also found to be unduly harsh.

Finally, at paras. 33-36 Lord Carnwath considers the decision of the UT in KO itself. The essential facts were that KO was a Nigerian national who had entered the UK unlawfully in 1986 and had never had leave to remain. He was married to a British citizen, with whom he had four children, themselves British citizens, born between

On the basis (held by Lord Carnwath to be wrong) that the seriousness of KO’s offending should be taken into account UTJ Southern’s conclusion was that the effect of his deportation on his children would not be unduly harsh. He directed himself at para. 26 of his decision:

“The consequences for an individual will be ‘harsh’ if they are ‘severe’ or ‘bleak’ and they will be ‘unduly’ so if they are ‘inordinately’ or ‘excessively’ harsh taking into account ‘all of the circumstances of the individual’. Although I would add, of course, that ‘all of the circumstances’ includes the criminal history of the person facing deportation.” (Footnote: ⁴)

Applying that test at paras. 43-44, which are quoted in full by Lord Carnwath, UTJ Southern said:

“43.

… There is undoubtedly a close relationship between this father and his children, as one would expect in any family living together as does this one. The preserved finding of fact is that, although it would not be unduly harsh for the four younger children to move to Nigeria, the reality of the situation is that they will remain here and, as the family relationships cannot be maintained by modern means of communication, there will be a complete fracture of these family relationships. The claimant is not authorised to work and so has been unable to provide financial support for his family but his role within the household has meant that his wife has been able to work, which she would find hard or impossible if she had to care on a daily basis for the children without her husband's assistance. Thus it is said that if the claimant is removed, the main household income will be lost and the children would be subject to economic disadvantage. But, again, that is not an experience that can, in my judgment, be categorised as severe or bleak or excessively harsh as, like any other person lawfully settled in the United Kingdom, the claimant's wife and family will have access to welfare benefits should they be needed.

44.

Nor do I have any difficulty in accepting the submission that the children, who have enjoyed a close and loving relationship with their father, will find his absence distressing and difficult to accept. But it is hard to see how that would be any different from any disruption of a genuine and subsisting parental relationship arising from deportation. As was observed by Sedley LJ in AD Lee v Secretary of State for the Home Department [2011] EWCA Civ 348:

‘The tragic consequence is that this family, short-lived as it has been, will be broken up for ever, because of the appellant’s bad behaviour. That is what deportation does.’ (Footnote: ⁵)

This family relationship was not, of course, short lived but the point is the same. Nothing out of the ordinary has been identified to demonstrate that in the case of this particular family, when balanced against the powerful public interest considerations in play, although the children will find separation from their father to be harsh, it will not be, in all of the circumstances, unduly harsh for them each to remain in the United Kingdom after their father is removed to Nigeria [Lord Carnwath’s italicisation (Footnote: ⁶)].”

Lord Carnwath, while disapproving UTJ Southern’s self-direction in para. 26, observes at para. 36 of his judgment that his error did not in fact seem to have infected the actual reasoning at paras. 43-44, and says that he finds that reasoning “difficult to fault”.

On the alternative basis (held by Lord Carnwath to be correct) that the seriousness of KO’s offending was not taken into account UTJ Southern said, at para. 45 of his decision:

“… [I]f there is to be no balancing exercise requiring the public interest to be weighed and if the focus is solely upon an evaluation of the consequences and impact upon the claimant’s children, it is clear that the application of paragraph 399 (a) can deliver only one answer, that being that it would be unduly harsh for the claimant's children to remain in the United Kingdom without their father, given that there is a close parental relationship which cannot be continued should their father be deported.”

As to that, Lord Carnwath says, at para. 35:

“Miss Giovanetti for the Secretary of State takes issue with that alternative reasoning, which she criticises as applying too low a standard. I agree. The alternative seems to me to treat ‘unduly harsh’ as meaning no more than undesirable. Contrary to the stated intention it does not in fact give effect to the much stronger emphasis of the words ‘unduly harsh’ as approved and applied in both MK and MAB.”

What light do those passages shed on the meaning of “unduly harsh” (beyond the conclusion on the relative seriousness issue)?

The essential point is that the criterion of undue harshness sets a bar which is “elevated” and carries a “much stronger emphasis” than mere undesirability: see para. 27 of Lord Carnwath’s judgment, approving the UT’s self-direction in MK (Sierra Leone), and para. 35. The UT’s self-direction uses a battery of synonyms and antonyms: although these should not be allowed to become a substitute for the statutory language, tribunals may find them of some assistance as a reminder of the elevated nature of the test. The reason why some degree of harshness is acceptable is that there is a strong public interest in the deportation of foreign criminals (including medium offenders): see para. 23. The underlying question for tribunals is whether the harshness which the deportation will cause for the partner and/or child is of a sufficiently elevated degree to outweigh that public interest.

However, while recognising the “elevated” nature of the statutory test, it is important not to lose sight of the fact that the hurdle which it sets is not as high as that set by the test of “very compelling circumstances” in section 117C (6). As Lord Carnwath points out in the second part of para. 23 of his judgment, disapproving IT (Jamaica), if that were so the position of medium offenders and their families would be no better than that of serious offenders. It follows that the observations in the case-law to the effect that it will be rare for the test of “very compelling circumstances” to be satisfied have no application in this context (I have already made this point – see para. 34 above). The statutory intention is evidently that the hurdle representing the unacceptable impact on a partner or child should be set somewhere between the (low) level applying in the case of persons who are liable to ordinary immigration removal

Observations of that kind are, I hope, helpful, but they cannot identify an objectively measurable standard. It is inherent in the nature of an exercise of the kind required by section 117C (5) that Parliament intended that tribunals should in each case make an informed evaluative assessment of whether the effect of the deportation of the parent or partner on their child or partner would be “unduly harsh” in the context of the strong public interest in the deportation of foreign criminals; and further exposition of that phrase will never be of more than limited value.

The Appellants of course accept that Lord Carnwath said what he said in the passages to which I have referred. But they contend that it is not a complete statement of the relevant law and/or that it is capable of being misunderstood. In their joint skeleton argument they refer to the statement in para. 23 of Lord Carnwath’s judgment that “one is looking for a degree of harshness going beyond what would necessarily be involved for any child faced with the deportation of a parent” and continue:

“This statement, taken in isolation, creates the opportunity for a court or tribunal to reach a conclusion on undue harshness without due regard to the section 55 duty or the best interests of the child and without careful analysis of all relevant factors specific to the child in any particular case. Instead, such considerations risk being ‘swept up’ under the general conclusion that the emotional and psychological impact on the child would not be anything other than that which is ordinarily expected by the deportation of a parent ... that cannot have been the intention of the Supreme Court in KO (Nigeria), which would otherwise create an unreasonably high threshold.”

Mr de Mello and Mr Bazini developed that submission in their oral arguments. In fact it comprises two distinct, though possibly related, points. I take them in turn.

The first is that what Lord Carnwath says in the relevant parts of his judgment in KO makes no reference to the requirements of section 55 of the 2009 Act and is likely to lead tribunals to fail to treat the best interests of any affected child as a primary consideration. As to that, it is plainly not the case that Lord Carnwath was unaware of the relevance of section 55: see para. 15 of his judgment, quoted at para. 41 above. The reason why it was unnecessary for him to refer explicitly to section 55 specifically in the context of his discussion of Exception 2 is that the very purpose of the Exception, to the extent that it is concerned with the effect of deportation on a child, is to ensure that the best interests of that child are treated as a primary consideration. It does so by providing that those interests should, in the case of a medium offender, prevail over the public interest in deportation where the effect on the child would be unduly harsh. In other words, consideration of the best interests of the child is built into the statutory test. It was not necessary for Lord Carnwath to spell out that in the application of Exception 2 in any particular case there will need to be “a careful analysis of all relevant factors specific to the child”; but I am happy to confirm that that is so, as Lord Hodge makes clear in his sixth proposition in Zoumbas.

The second point focuses on what are said to be the risks of treating KO as establishing a touchstone of whether the degree of harshness goes beyond “that which is ordinarily expected by the deportation of a parent”. Lord Carnwath does not in fact use that phrase, but a reference to “nothing out of the ordinary” appears in UTJ Southern’s decision. I see rather more force in this submission. As explained above, the test under section 117C (5) does indeed require an appellant to establish a degree of harshness going beyond a threshold “acceptable” level. It is not necessarily wrong to describe that as an “ordinary” level of harshness, and I note that Lord Carnwath did not jib at UTJ Southern’s use of that term. However, I think the Appellants are right to point out that it may be misleading if used incautiously. There seem to me to be two (related) risks. First, “ordinary” is capable of being understood as meaning anything which is not exceptional, or in any event rare. That is not the correct approach: see para. 52 above. There is no reason in principle why cases of “undue” harshness may not occur quite commonly. Secondly, if tribunals treat the essential question as being “is this level of harshness out of the ordinary?” they may be tempted to find that Exception 2 does not apply simply on the basis that the situation fits into some commonly-encountered pattern. That would be dangerous. How a child will be affected by a parent’s deportation will depend on an almost infinitely variable range of circumstances and it is not possible to identify a baseline of “ordinariness”. Simply by way of example, the degree of harshness of the impact may be affected by the child’s age; by whether the parent lives with them (NB that a divorced or separated father may still have a genuine and subsisting relationship with a child who lives with the mother); by the degree of the child’s emotional dependence on the parent; by the financial consequences of his deportation; by the availability of emotional and financial support from a remaining parent and other family members; by the practicability of maintaining a relationship with the deported parent; and of course by all the individual characteristics of the child.

I make those points in response to the Appellants’ submissions. But I am anxious to avoid setting off a further chain of exposition. Tribunals considering the parent case under Exception 2 should not err in law if in each case they carefully evaluate the likely effect of the parent’s deportation on the particular child and then decide whether that effect is not merely harsh but unduly harsh applying KO (Nigeria) in accordance with the guidance at paras. 50-53 above.

As I noted at the beginning of this judgment, the President listed these appeals together as a vehicle to give general guidance on the law post-KO. As regards the meaning of “unduly harsh”, that exercise is carried out at Section B of the judgment in RA (paras. 9-17). Most of that section is devoted to a submission from Mr Bazini based on paragraphs in the decision of the UT in MK (Sierra Leone) which are not considered in KO. I consider the equivalent submissions later in this judgment (see paras. 127-9 below): I need only say at this point that they raise no issue of general significance about the meaning of “unduly harsh”. Beyond that, the UT says only that Lord Carnwath’s approval at para. 27 of his judgment of the self-direction by the UT in MK must be followed. I agree with that as far as it goes, though tribunals will wish to have regard to the further points made in this judgment.

The Supreme Court in KO (Nigeria) was concerned only with what was entailed in the assessment of undue harshness for the purpose of section 117C (5) (and paragraph 399 (a)). The appellant relied only on section 117C (5) and did not contend that there were in his case very compelling circumstances over and above Exception 2 which outweighed the public interest in his deportation. It is unsurprising therefore that there is in Lord Carnwath’s judgment no discussion of section 117C (6) and no reference to NA (Pakistan); but it is also slightly unfortunate. There is a risk that, in cases involving a medium offender, tribunals who are directed only to KO may think that if a potential deportee cannot bring himself within either Exception that is the end of the story. As will be clear from my discussion of NA (Pakistan) – see in particular paras. 29-30 above – that is not the case: it remains necessary in principle to conduct a full article 8 proportionality assessment, albeit one in which the public interest in deportation will only be outweighed in very compelling circumstances. That was one of the points which the UT was evidently anxious to make in this group of guideline

Although the two-stage exercise described in NA (Pakistan) is conceptually clear, it may occasionally make the analysis unnecessarily elaborate. There may be cases where a tribunal is satisfied that there is a combination of circumstances, including but not limited to the harsh effect of the appellant’s deportation on his family, which together constitute very compelling reasons sufficient to outweigh the strong public interest in deportation, but where it may be debatable whether the effect on the family taken on its own (as section 117C (5) requires) is unduly harsh. (An equivalent situation could arise in relation to Exception 1: there might, say, be significant obstacles to the appellant’s integration in the country to which it is proposed to deport him, but it might be questionable whether they were very significant.) In such a case, although the tribunal will inevitably have considered whether the relevant Exception has been satisfied, it is unnecessary for it to cudgel its brains into making a definitive finding. The Exceptions are, as I have said, designed to provide a shortcut for appellants in particular cases, and it is not compulsory to take that shortcut if proceeding directly to the proportionality assessment required by article 8 produces a clear answer in the appellant’s favour.

I should say, finally, that Mr Pilgerstorfer referred us to a number of decisions of this Court in which KO has been applied – Secretary of State for the Home Department v JG (Jamaica) [2019] EWCA Civ 982; Secretary of State for the Home Department v PF (Nigeria) [2019] EWCA Civ 1139; Secretary of State for the Home Department v

HA is an Iraqi national. He came to this country clandestinely in July 2000, when he was aged 20. His claim for asylum was refused in August 2003, and his appeal rights became exhausted in February 2004.

In 2006 HA began a relationship with a British woman, to whom I will refer as “NT”. They are not officially married, although they went through a religious ceremony in 2009. NT already had a daughter, born in 2004. They have since had three children together, born respectively in October 2011, May 2014 and November 2016: thus at the date of the hearing in the UT they were aged seven, four and two. The children are British citizens. HA, NT and their three children live together; NT’s daughter lives nearby, with her maternal grandmother, but is very much involved with the life of the family. HA does not work but NT has a full-time job.

On 15 March 2010 HA was convicted of assisting unlawful immigration and possessing an unlawfully obtained immigration card, and also of an offence of failing to surrender to custody. He was sentenced to sixteen months’ imprisonment. The circumstances of the primary offences were that he was trying to arrange the illegal entry of his mother and his brother into the UK.

On 31 May 2017 the Secretary of State made a deportation order against HA, in accordance with the foreign criminal provisions. Remarkably, that was six years after his release from custody; but the reasons for the delay were examined by the UT and were held not to reflect any culpable delay by the Secretary of State, and that conclusion is not challenged before us. The details of the procedural history thereafter are not quite clear, but on 26 February 2018 the Secretary of State refused a human rights application by HA and an application to revoke the deportation order.

HA appealed to the FTT. His appeal was heard by FTTJ Gurung-Thapa on 13 August 2018. He was represented by counsel and both he and his partner gave oral evidence and submitted a number of documents. The Judge reserved her decision. By a decision and Reasons promulgated on 1 October she allowed his appeal. Because this appeal is against the re-made decision of the UT I need not set out her reasoning.

The Secretary of State appealed. At a hearing in January 2019 the UT set aside the decision of the FTT. The UT decided to re-make the decision at a further hearing. At that hearing, on 15 February, HA represented himself and gave evidence. The UT also had a witness statement from his partner, who could not attend the hearing because her employers would not give her the day off. It also had the documents which had been before the FTT.

I need not summarise paras. 1-33 of the UT’s decision. At paras. 27-45 it summarised HA’s oral evidence. I need not refer to most of it, but he made clear that – as one would expect, given that his partner had a full-time job – he was responsible for taking the two elder children to school (which was close to where they lived) and the youngest to nursery and for picking them up afterwards: the Tribunal noted later that that was confirmed by letters from the school and nursery. He said that it would be difficult for his wife to go on working if he were deported: she might be able to work part-time or she might have to give up work altogether. He gave reasons why she was unlikely to receive any substantial assistance from her mother or other members of her family. At paras. 46-58 the UT summarises the parties’ submissions; there is nothing to which I need to refer.

The UT gives its reasons for dismissing HA’s appeal at paras. 59-111 of its decision. Proceeding in the way recommended in NA (Pakistan) in the case of a medium offender, it considers first whether he could rely on either of the Exceptions under section 117C (4) and (5). At paras. 60-65 it concludes that he cannot rely on Exception 1. That conclusion is not challenged, and I need say no more about it. It then considers Exception 2 at paras. 66-79 and “Very Compelling Circumstances” at paras. 80-110. I take them in turn.

As regards Exception 2, the Secretary of State accepted that HA has a genuine and subsisting relationship with his partner and his children. She also accepted that it

At paras. 80-110 the UT conducts the overall proportionality exercise that is required in cases where neither Exception applies. It directs itself by reference to NA (Pakistan). From para. 86 onwards it goes through a series of specific considerations relied on by HA. These were, in addition to the effect of separation on his family, the length of time that he had been in the country and his consequent social and cultural integration; a contention (which the Tribunal rejected) that he did not realise that he had no right to be in the country; the Home Office’s delay in making a deportation

The grounds of appeal as pleaded in the Appellant’s Notice were reformulated in the joint skeleton argument under three heads, but these refer back to more general points about the approach to the “unduly harsh” test developed elsewhere in the skeleton, and the basis on which points are allocated to one head rather than another is unclear. I will not therefore proceed by reference to the reformulated grounds but will address HA’s principal points in what seems to be the most logical order.

Was it necessary to consider the “stay scenario”?

HA’s ground 1, as pleaded, is that once it had been established (on the basis of the Secretary of State’s concession) that it would be unduly harsh for his children to relocate with him to Iraq he should have been treated as falling within the terms of Exception 2. He submits that it was wrong in principle for the UT to reach the opposite conclusion on the basis that it would not be unduly harsh for them to remain

That submission, as so formulated, is plainly ill-founded. In the first place, I can see no inconsistency between paragraph 399 (a) and section 117C (5). In my view it is clear that the separate identification of the stay and go scenarios in the Rules is an elaboration of Exception 2 and not a departure from it. The assessment required by section 117C (5) must in principle involve a consideration of any scenario in which the deportation could affect a partner or child; and paragraph 399 simply recognises that fact expressly. And in any event HA’s suggested construction would make no sense having regard to the statutory purpose. The purpose of Exception 2 is to preclude deportation where that would have an unduly harsh effect on the foreign criminal’s partner or child. HA’s reading would mean that it applied in circumstances where such an effect would not be suffered at all. It would be absurd if the fact that it would be unduly harsh for HA’s children to relocate to Iraq – where in practice noone was suggesting that they would go (no doubt for that very reason) – meant that he could not be deported, leaving them behind, even though (on the UT’s finding) the separation would not involve any unduly harsh effect. Indeed if HA’s submission were right he could not be deported even if the UT had found that separation from the parent would be positively beneficial for the child, as occasionally (though not here) it might be. The paragraphs referred to in Jeunesse and MM (Lebanon) have nothing to do with this issue, and I need not set them out here.

The submission does, however, raise a point which merits more detailed consideration. In oral argument we canvassed with Mr Pilgerstorfer (whose submissions I wish to say were throughout conspicuously clear and thoughtful) the possibility of a situation where, although if the partner and child stayed in the UK the effect on them would not be unduly harsh, the established likelihood was that they would in fact not stay but would relocate with him and suffer an unduly harsh effect in consequence (Footnote: ⁷): such a situation might occur where the parents failed to appreciate the harshness of the effect on the partner and/or her child, or where one or both did appreciate it but the potential deportee insisted nonetheless and the partner felt obliged to put her loyalty to him first. If we were only concerned with the partner, that might be unobjectionable, because as an adult she could be treated as free to make her own choices; but the child would have no such choice. In such a case a rule that Exception 2 was only satisfied if both scenarios were unduly harsh would produce an outcome that gave no weight to the actual best interests of the child: it would mean that the Exception was unavailable only because of an acceptable alternative which was purely theoretical. We mentioned the line of authorities which establishes that tribunals must make decisions in the light of the real world facts (see,

Mr Pilgerstorfer’s response was that there was no warrant for departing from the clear intention of paragraph 399 (a) that the effect of deportation must be unduly harsh in both scenarios. That was not only what the words said, it made sense as a matter of policy and principle. The public interest in deportation should not be outweighed where there was a realistically available option in which the deportation would not have an unduly harsh effect. If the parents nevertheless chose an option which did have such an effect that ought not to be treated as resulting from the Secretary of State’s decision. He added that if the law were otherwise there would be an obvious incentive on foreign criminals, and their partners, to claim, and perhaps even genuinely to intend, that the family would go with them if they were deported, however harsh the effect might be on the children: it would be very unsatisfactory for decision-makers to have to base their decisions on an assessment of whether such threats were genuine or tactical. He pointed out that the “real world” line of cases was concerned with different provisions of the Rules and directed to a different issue.

I am persuaded by those submissions. I think it is clear that paragraph 399 (a) means what it says and that it accords with principle and policy for the reasons given by Mr Pilgerstorfer. The Secretary of State cannot be said to be acting contrary to the best interests of the child in circumstances where an option was available which would not be unduly harsh for him or her, even if the parents decline to avail themselves of it. In fact parents can generally be expected to avail themselves of the option which is best for their child. I dare say there may be rare cases where it does seem likely – though of course at the point of decision it will never be knowable – that the deportee and his partner will try to take their child with them to a country where life for it will be unduly harsh. (Footnote: ⁸) In an appropriate case such a choice might attract the operation of the child protection legislation, but if the adverse consequences are not at that level then the fact has to be faced that parents do have the right to make choices for their children which may have harsh effects. (Footnote: ⁹)

It follows that if a decision-maker is satisfied that on one of the scenarios the effect of deportation on the partner or child is not unduly harsh (usually this will be the stay scenario) it need not consider the other, though other things being equal it would be good practice to do so in case the matter goes further.

I should say that the conclusion which I have reached on this point reflects the approach taken (albeit apparently without contrary argument) in a number of decisions of the UT to which we were referred by Mr Pilgerstorfer. I also note that in PG (Jamaica) Holroyde LJ stated at para. 34 of his judgment that:

“Pursuant to Rule 399, the tribunal or court must consider both whether it would be unduly harsh for the child and/or partner to live in the country to which the foreign criminal is to be deported and whether it would be unduly harsh for the child and/or partner to remain in the UK without him [emphasis supplied]”.

It was also assumed to be the case by the UT in MK (Sierra Leone): see para. 45 (d) of the decision.

Is the UT’s decision on Exception 2 sustainable?

As can be seen, the UT’s exposition of the factors that it took into account in reaching its conclusion that the effect on HA’s children of his deportation would not be unduly harsh was clear and careful. Nevertheless, not without considerable hesitation, I have come to the conclusion that it cannot be sustained.

My starting-point is that this is a case in which a finding of undue harshness, as regards HA’s children at least, might quite readily have been made. We were not shown all the materials that were before the UT, but it seems from the Tribunal’s own findings that his relationship with his children was particularly close. This was a household in which the mother worked full-time and the father (perforce) did not. It may go rather beyond the Tribunal’s findings to describe him, as Mr de Mello did, as the “primary and constant carer of the children” (a phrase taken from para. 119 of the judgment of the ECtHR in Jeunesse); but the Tribunal itself described him as “very much a hands-on father”, who undertook all the responsibilities described in para. 68 of its decision. He had been at home throughout the lives of all three children: because of the delay between his conviction and the decision under challenge, there had been no period (such as there often is in deportation cases) when he was separated from them by imprisonment. That being so, it would not be surprising if the emotional and psychological impact of his deportation on some or all of his children would go significantly beyond that which (using Lord Carnwath’s phrase, in the sense discussed above) any child would necessarily suffer from the deportation of a parent – in other words, for it to be not only harsh but unduly harsh. The impact would of course be all the greater given the finding that his children would not be able to visit him in Iraq.

What, therefore, was the reason why the UT reached the contrary conclusion? The answer appears from para. 71 of its decision, where it says that there was no evidence that the effect of HA’s deportation on NT and his children “would be anything other than that which is ordinarily to be expected by the deportation of a partner/parent”. I note also the UT’s statement at para. 75 that the difficulties that NT would face if HA were deported were “no more than the difficulties faced by many single parents working part-time or full-time”. That language is in the danger area which I have described at para. 56 above. Against the background of different facts, it might well be appropriate to read it as representing a considered judgement that the degree of harshness that HA’s family would suffer from his deportation was insufficient to outweigh the public interest in the deportation of medium offenders. But in the context of the primary facts which I have summarised I think that on balance it is more likely that the Tribunal has proceeded on the basis that it is sufficient to say that the situation in HA’s case is of a kind which is quite commonly encountered in deportation cases. For the reasons I have given, that is not the correct approach.

One reason for the hesitation that I have felt in reaching this conclusion is that the headline facts in the present case seem, as Mr Pilgerstorfer reminded us, very close to those of KO (Nigeria), in which Lord Carnwath endorsed the UT’s (main) conclusion that the effect of the father’s deportation would not be unduly harsh and held that the contrary conclusion applied “too low a standard”. In the end, however, I have reached the conclusion that it would be wrong to treat KO as a “factual precedent” such that any case with what may appear to be similar facts must be decided the same way. Cases of this kind are never truly identical, and each tribunal must make its own assessment on the basis of the case before it. As it happens, the same point arises in

For those reasons I would allow the appeal and remit HA’s case to the UT for a reconsideration of whether, applying the statutory test as discussed above, the effect of his deportation on his partner and children would be unduly harsh. That will in particular require a careful examination of the impact on the children, as emphasised in Zoumbas. I hope that HA and his advisers will ensure that the tribunal has the fullest possible information for the purpose of that exercise. As I have said, we were not shown all the material that was before it on the last occasion; but I have the impression that it was not as full as it might have been.

I should for completeness deal briefly with two other points made by Mr de Mello.

First, in his oral submissions he contended that the reason why the UT came to the conclusion that it did was that in his judgment in KO (Nigeria) Lord Carnwath had not drawn attention to the need to have regard to the best interests of the children and that the Tribunal here had accordingly not done so. I have already rejected the submission that Lord Carnwath’s formulation of the test ignores section 55 of the 2009 Act; and the submission about the approach of the Tribunal would in any event be hard to reconcile with its express reference to the best interests of the children at para. 78.

Secondly, Mr de Mello sought to argue that the decision of the UT on the Exception 2 issue was perverse because it was contrary to the decision of the FTT, which had made the opposite finding. With all respect, that is hopeless. The UT was entitled, and indeed obliged, to make its own decision. The fact that it reached a different conclusion from the FTT on essentially the same facts and evidence goes nowhere, save perhaps to reinforce the conclusion that this is a case where a finding of undue harshness could well have been made.

Is the UT’s decision on the overall proportionality assessment sustainable?

If HA eventually succeeds in bringing himself within the terms of section 117C (5) it will be unnecessary for him to rely on section 117C (6). But I should nevertheless deal with his appeal on this aspect, because the issues raised by it may arise again if he is unsuccessful on the remittal as regards Exception 2; and one at least is of some general importance.

The UT’s consideration of section 117C (6) was extensive, but HA only advances three challenges to it. I can dispose of two of them very briefly. The first is that the Tribunal applied a test of “very compelling circumstances” rather than “sufficiently compelling circumstances”. I have already rejected the argument that those represent different tests: see para. 38 above. The second is that it should have treated itself not as bound to apply the statutory scheme but only to have regard to it, relying on Akinyemi (no. 1). I have rejected that argument too: see para. 13.

The real challenge to the UT’s reasoning is that in striking the proportionality balance it failed to take into account the fact that HA’s sentence was only slightly above the minimum level for him to qualify as a foreign criminal.

I should start by identifying the relevant legal principle. At para. 108 of its decision the UT refers to MS (Philippines)), which was, as noted above, another of this quartet of post-KO cases. The potential deportee in that case had committed particularly serious offences of violence for which he had been sentenced to ten years’ imprisonment: the sentencing judge described him as a highly dangerous man. It was argued on his behalf that the seriousness of his offending was relevant only to whether he could rely on either of the Exceptions, and that it could not be put into the balance against him in conducting the necessary assessment under section 117C (6). That might seem a surprising submission, but it was rendered at least superficially plausible by some general observations made by Lord Carnwath at paras. 20-22 of his judgment in KO (Nigeria). The UT rejected that argument: see paras. 4-20 of its decision. It held that Lord Carnwath’s observations were directed only to the application of the Exceptions in the case of medium offenders: for the reasons which he gave in KO the exercise involved under those Exceptions is wholly self-contained. It said that it would be extraordinary if, for the purposes of the ultimate proportionality exercise which has to be performed where the Exceptions do not apply, no distinction fell to be made, between, say, an offender who had committed offences attracting a term of four years’ imprisonment and a multiple murderer.

There was no issue before us as to the correctness of this part of the decision in MS (Philippines), and we accordingly heard no argument about it, but in my view the UT was plainly right. The question then arises whether it cuts both ways – that is, whether in principle a potential deportee can rely, as part of the overall proportionality assessment, on the fact that his offence was at or near the bottom of the scale of seriousness. It seems to me that that must indeed be the case. There can be no principled reason for treating the two arguments differently. That conclusion is also in accordance with the Strasbourg jurisprudence, to which, as is confirmed both by NA (Pakistan) and by Hesham Ali, we are obliged to have regard. In Boultif the ECtHR held in terms that it was necessary in the assessment of the proportionality of deportation to take into account “the nature and seriousness of the offence committed by the applicant” (para. 48); and it is clear from para. 51 of its judgment that in coming to the conclusion that the applicant’s deportation was disproportionate it took into account the fact that his sentence was comparatively low. Since it was established in KO that the relative seriousness of the offence cannot be taken into account in considering Exception 2 (see para. 43 above), it must be capable of being deployed by the potential deportee at the second stage.

In making this point I do not wish to be misunderstood. It cannot be the case that an appellant can rely on the fact that his offence attracted a sentence of, say, “only” twelve months as sufficient by itself to constitute very compelling circumstances for the purpose of section 117C (6): that would wholly subvert the statutory scheme. But if there were other compelling circumstances in his case the fact that his offence was comparatively less serious could form an element in his overall case that the strong public interest in deportation was outweighed.

The UT did of course purport to take the seriousness of HA’s offending into account:

It remains to be seen whether this point has any significance on remittal, assuming that section 117C (6) is in play at all. But it is not inconceivable that it might. If the UT were to regard HA as only having failed by a small margin to bring himself within Exception 2 and/or if there were other circumstances weighing against deportation, the fact that his offence, as measured by his sentence, had been near the bottom of the scale of the seriousness might make a material difference to how the balance was struck.

I would therefore allow HA’s appeal and remit the issues under sub-sections (5) and

RA is an ethnic Kurd and Iraqi by nationality. We were told, though there is no express finding to this effect, that he is originally from Kirkuk. He came to this country with his brother clandestinely in 2007, when he was aged 14. His claim for asylum was refused in October 2009, but he was given discretionary leave until 1 September 2010. An application to extend his leave was refused in July 2011 but he remained in the UK without leave.

In 2012 RA married a British citizen, likewise of Kurdish Iraqi descent, to whom I will refer as “KI”. They have a daughter, to whom I will refer as “Y”, born in 2013, who is a British citizen. KI’s parents live nearby. In June 2016 RA was given limited leave to remain on the basis of his family life.

RA’s mother and his sister live in Erbil, in autonomous Iraqi Kurdistan, sometimes known as the Iraqi Kurdish Region (“the IKR”). We were told, though again there is no finding to this effect and we were given no details, that they had had to leave their own home in Kirkuk, which is not part of the IKR, because it had become unsafe for them. RA’s mother is in poor health and lives not in her own home but in accommodation provided to her by a charity as a displaced person. His sister is married and has her own home: we were told that beyond that there was no evidence about her circumstances.

KI has maintained links with Iraq. She speaks Kurdish Sorani. On three occasions in recent years she has travelled to the IKR – in 2010, 2014 and 2017. On the latter two occasions she took Y with her and visited her mother-in-law, RA’s mother, in Erbil.

On 10 August 2016 RA was convicted, on his plea of guilty, of an offence of under section 4 of the Identity Documents Act 2010 and sentenced to twelve months’ imprisonment. Only brief details of the circumstances of the offence were before us, but the essence is that he was sent a forged Iraqi passport by his mother so that he could come and visit her in Iraq. When he presented the passport to the authorities in order to enable him to travel the forgery was detected. The Judge’s very short sentencing remarks acknowledge that RA was of good character but say that an immediate custodial sentence was necessary because of the nature of the offence. He gave him maximum credit for his guilty plea.

In September 2016 the Secretary of State decided to deport RA in accordance with the foreign criminal provisions.

RA appealed to the First-tier Tribunal. His appeal was heard by FTTJ Mensah on 18 May 2018. She reserved her decision. By a decision and Reasons promulgated on 12 June she allowed his appeal. Again, because we are concerned only with the decision as re-made by the UT, I need not set out her reasoning.

The Secretary of State appealed. At a hearing in January 2019 RA accepted that the decision of the FTT was vitiated by a misdirection, of which I need not give details; and it was accordingly set aside. The UT decided to re-make the decision itself, and at the hearing on 13 February, at which he was represented by Mr Bazini, it heard evidence from RA and his wife.

The first part of the UT’s Reasons is concerned with setting out the relevant law. I have dealt with this above. At paras. 34-49 it summarises the background facts and

The Tribunal starts by recording, at para. 51, that it was conceded that RA had a genuine and subsisting relationship with KI and a subsisting parental relationship with Y, and that both were British citizens. Accordingly section 117C (5) was engaged, and the question was whether the effect of his deportation would be unduly harsh on his wife and/or his daughter. In principle, as the drafting of paragraph 399 (a) of the Rules recognises, the answer to that question might depend on whether they would go with him to Iraq or stay in this country. The Tribunal considers both alternatives in turn.

Taking first the alternative that both would accompany RA to Iraq, the Tribunal decides at para. 53 that that would not be unduly harsh for his wife. It says:

“So far as the wife alone is concerned, it would not be unduly harsh to expect her to live in northern Iraq with the appellant. The appellant's wife speaks the local language and, although her situation there with the appellant would be much less pleasant than it is in the United Kingdom, it would not be unduly harsh, applying the test approved in KO (Nigeria). The appellant's wife is in good health, as is the appellant.”

It deals with the effect on Y at paras. 54-55, as follows:

“54.

It would plainly not be in the best interests of the appellant's British daughter for her to be expected to live in northern Iraq. She would not only lose the opportunity of being educated in the United Kingdom but would also face a challenging physical environment. She would, in addition, have quickly to master Kurdish Sorani, although the evidence indicates that she has exposure to that language as a result of the presence of her parents, grandparents and other relatives in the United Kingdom.

55.

Looking at matters in the round, we conclude, albeit with some degree of hesitation, that it would not be unduly harsh for the daughter to live with both parents in northern Iraq. The child is still relatively young. The security position is considerably improved, compared with the position when her mother decided to take her there on a visit. She would be with both parents, in a loving relationship. There would be other family support to call on in the country, in the form of her aunt, even if the grandmother may not be able to offer much practical assistance. There is, in any event, no reason why the appellant cannot secure employment in Erbil. Overall, expecting the daughter to live in Iraq would not be unduly harsh, applying the test approved in KO (Nigeria).”

At paras. 56-60 the Tribunal considered the alternative that KI and Y would stay behind. At para. 57 it rejects the evidence of KI that she would in that case commit suicide. Paras. 58-60 read:

“58.

If the appellant were deported, life for the appellant's wife and the daughter would, we find, be hard. It would, however, be far from being unduly harsh. The appellant's wife and daughter live in very close proximity to family members, who already provide assistance and who can be expected to help the appellant's wife with the consequences of the appellant's removal.

59.

The appellant's wife has, until recently, worked part-time. She told us that she stopped because of the forthcoming tribunal hearing. She did not explain, however, why she was expected to do so much in connection with that hearing as to be unable to continue such work, particularly given the involvement of the appellant's solicitors. In any event, following the appellant's deportation, it can reasonably be expected that the appellant's wife can work part-time, as do very many mothers with children of her daughter's age. If, as has already occurred, the appellant's wife has to have recourse to benefits, that would not be a matter that would cause or contribute to undue harshness.

60.

We agree with Mr Bazini that reliance upon modern means of communication, such as Skype, is no substitute for physical presence and face-to-face contact. We do not, however, believe that, in the event of deportation, such face-to-face contact would not be possible. The appellant's wife has made several visits to northern Iraq in the past, including two with her (then very small) daughter. There is no suggestion that, at that time, the family's financial circumstances were markedly better than they are at present or would likely be in the future. Accordingly, it would be entirely possible for the appellant to see both his wife and daughter on a face-to-face basis in Iraq.”

On that basis, the requirements of Exception 2 were not satisfied, whether RA’s family returned with him to Iraq or not. The Tribunal proceeded to address the question under section 117C (6) (as applied to medium offenders in accordance with NA (Pakistan)). Paras. 62-66 read:

“62.

We have regard to the fact that the appellant's sentence of imprisonment is at the bottom of the range covered by section 117C (3). We give that due weight. We do, however, take account of the fact that credit was given for the appellant's guilty plea. We also take account of the fact that, as the Sentencing Judge pointed out, the offence was a serious one. Given that the appellant has never been found to have had any legitimate reason to come to the United Kingdom, the fact that he should decide to engage in criminal behaviour, having only just regularised his former unlawful presence, counts against him. The weight of the public interest, bearing in favour of deportation, therefore remains high.

63.

So far as concern factors bearing on the appellant's side of the proportionality balance, we have regard to the fact that, as mentioned in section 117B (4) (b), the appellant's relationship with his wife was established in 2012, at a time when the appellant was in the United Kingdom unlawfully.

64.

At all material times, the appellant has not had indefinite leave to remain and, accordingly, section 117B (5) indicates that little weight should be given to the appellant's private life in the United Kingdom. In this regard, we observe that the appellant's history of employment in the United Kingdom is, in any event, exiguous.

65.

We accord, however, significant weight to the appellant's relationship with his daughter and to her own best interests, as a child. We accept, as we have already stated, that the appellant's deportation would have serious adverse effects upon his daughter and that, despite the opportunities to meet outside the United Kingdom, the appellant's daughter will clearly miss the appellant's daily presence in her life.

66.

Notwithstanding those factors in favour of the appellant, we conclude that the weight of the public interest is such that it cannot be said that there are very compelling circumstances, as required by section 117C (6), which would make deportation a disproportionate interference with the Article 8 rights of the appellant, his wife, or daughter. That is so, looking at each of their positions both individually and together.”

I should note that although the Tribunal says at para. 65 that it has “already stated … that the appellant's deportation would have serious adverse effects upon his daughter” there is in fact no earlier statement to that effect beyond the opening sentence of para. 58. The decision contains no particularised discussion of the effect on Y of being separated from her father.

Again, the grounds of appeal were reformulated following the grant of leave under three heads. The three heads do not so much represent distinct grounds as focus on the UT’s conclusions on the two scenarios identified in paragraph 399 (a) and on the article 8 proportionality assessment, with various discrete points being made under each head. I take them in turn.

The scenario where KI and Y relocate to Iraq

This “ground” reads as follows:

“The UT erred in failing to take account, alternatively adequate account, of material factors in concluding that it would not be unduly harsh for the Appellant’s child and partner [sic] to relocate to Iraq with the Appellant, including but not limited to [a] the importance of the child's British citizenship, [b] the FCO advice against travel to Iraq and [c] the situation facing the Appellant and his family on return.”

I have included the letters [a]-[c] to identify the specific complaints made about the UT’s reasoning.

As to [a] – the importance of Y’s British citizenship – Mr Bazini referred us to wellknown passages in the judgments of Lady Hale and Lord Hope in ZH (Tanzania). That case concerned the proposed removal to Tanzania of a mother who was not herself British but who had three children who were British citizens and who would in practice have had to accompany her if she were removed: it was not a deportation case, but Lady Hale noted at para. 1 of her judgment that the issue of principle arose in deportation decisions as well. The Supreme Court held that the removal of the mother would constitute a disproportionate breach of the article 8 rights of the children, having regard also to the requirements of article 3.1 of the UNCRC. An important element in its reasoning was the fact that the children were British citizens. The importance of that factor is explained at paras. 29-32 of Lady Hale’s judgment. I need not set those paragraphs out in full. I should note, however, that at para. 30 Lady Hale says:

“Although nationality is not a ‘trump card’ it is of particular importance in assessing the best interests of any child. The UNCRC recognises the right of every child to be registered and acquire a nationality (Article 7) and to preserve her identity, including her nationality (Article 8). In Wan, the Federal Court of Australia, pointed out at para 30 that, when considering the possibility of the children accompanying their father to China, the tribunal had not considered any of the following matters, which the Court clearly regarded as important:

‘(a) the fact that the children, as citizens of Australia, would be deprived of the country of their own and their mother's citizenship, “and of its protection and support, socially, culturally and medically, and in many other ways evoked by, but not confined to, the broad concept of lifestyle” (Vaitaiki v Minister for Immigration and Ethnic Affairs [1998] FCA 5, (1998) 150 ALR 608, 614);

(b)

the resultant social and linguistic disruption of their childhood as well as the loss of their homeland;

(c)

the loss of educational opportunities available to the children in Australia; and

(d)

their resultant isolation from the normal contacts of children with their mother and their mother's family.’”

At para. 32, after referring to various particular consequences that the children in the case before the Court would suffer from being removed, she says:

“Nor should the intrinsic importance of citizenship be played down. As citizens these children have rights which they will not be able to exercise if they move to another country. They will lose the advantages of growing up and being educated in their own country, their own culture and their own language. They will have lost all this when they come back as adults. …”

Likewise Lord Hope said, at paras. 40-41:

“40.

It seems to me that the Court of Appeal fell into error in two respects. First, having concluded that the children's British citizenship did not dispose of the issues arising under article 8 … they did not appreciate the importance that was nevertheless to be attached to the factor of citizenship in the overall assessment of what was in the children's best interests. …

41.

The first error may well have been due to the way the mother's case was presented to the Court of Appeal. It was submitted that the fact that the children were British citizens who had never been to Tanzania trumped all other considerations …. That was, as the court recognised, to press the point too far. But there is much more to British citizenship than the status it gives to the children in immigration law. It carries with it a host of other benefits and advantages, all of which Lady Hale has drawn attention to and carefully analysed. They ought never to be left out of account, but they were nowhere considered in the Court of Appeal's judgment. The fact of British citizenship does not trump everything else. But it will hardly ever be less than a very significant and weighty factor against moving children who have that status to another country with a parent who has no right to remain here, especially if the effect of doing this is that they will inevitably lose those benefits and advantages for the rest of their childhood.”

Mr Bazini submitted that the Tribunal had failed to take any account of that guidance. The glancing mention of Y’s British nationality in the first line of para. 54 of the Tribunal’s Reasons was inadequate because it was not in the context of the undue harshness assessment. He acknowledged that the Tribunal does refer to Y losing the opportunity of being educated in England, which is mentioned in Lady Hale’s judgment as one of the advantages of British nationality. But that is only one particular advantage (and not one which is in fact dependent on British citizenship), and Mr Bazini’s essential complaint is that the Tribunal made no reference to what Lady Hale calls “the intrinsic importance of citizenship”.

I do see some force in this submission. I fully accept that this Court should not overturn the decision of a tribunal, let alone as experienced a tribunal as this, merely because it has not expressly mentioned a factor which it can fairly be assumed that it has taken into account. The Tribunal mentions Y’s British citizenship not only in para. 54 but also at para. 51, as part of its finding that she was a qualifying child, and I see the force of the argument that, having taken account of her nationality at that stage of its decision-making, it is unlikely to have left it out of account when making the assessment of undue hardship. However, the degree to which a tribunal needs to spell out its thinking must be sensitive to the circumstances of the particular case. The Tribunal’s reference to having reached its decision “with some degree of hesitation” shows that it regarded its decision on this issue as near the borderline. That must,

Mr Pilgerstorfer submitted that the fact that the Tribunal had taken Y’s British citizenship into account in finding that Exception 2 was engaged was enough: that first stage in the structured decision-making imposed by the statute was part of the overall process of assessment under article 8. I do not accept that. Under the statutory scheme British citizenship does indeed operate to open the Exception 2 gateway. But in my view it does not follow that it can thereafter be ignored when making the assessment of whether the effect of relocation would be unduly harsh: such a mechanistic approach risks artificially restricting the holistic exercise required by article 8.

Mr Pilgerstorfer also submitted that it was unnecessary for the Tribunal to spell out in full the factors to which it had had regard in its assessment of undue harshness because its conclusion was in RA’s favour – that is, that it was “plainly not in [Y’s] best interests” to live in Iraq. I do not accept that either. It is in truth pretty obvious

I turn to elements [b] and [c]. The Tribunal proceeded on the basis that if they returned to Iraq the family would live in Erbil, where RA’s mother and sister live and where his wife had family and had gone on her three previous visits. That was entirely reasonable: although, so we were told, RA himself is from Kirkuk, it was found in the country guidance case of AA (Article 15 (c)) Iraq [2015] UKUT 00544 (IAC) (“AA”) – which is broadly endorsed, though updated, in AAH (Iraqi Kurds – internal relocation) Iraq [2015] UKUT 00544 (IAC) (“AAH”) – that Kurds from elsewhere in Iraq can readily relocate to the IKR. It is clear from the country guidance that conditions in the IKR are safer and more settled than in the rest of Iraq, and although we were shown, in support of element [c], warnings in the FCO guidance about some continuing risk to visitors from remnants of Daesh, I can understand why the Tribunal did not find that residual risk a significant factor in the assessment of whether relocation would be unduly harsh for Y.

However, the country guidance does raise concerns about access to accommodation and employment for relocating Kurds who do not have family support: in particular, there is a finding that 70% of Kurds who are originally from outside the IKR are unemployed. Mr Bazini submitted that if RA was unable to secure proper accommodation or work, so that the family became effectively destitute, that would have a very serious impact on Y’s welfare. He pointed out that those issues are only referred to in the most general terms in para. 55, and the essence of element [c] under this ground is that the situation facing RA and his family on return is simply not adequately dealt with.

Taking elements [a] and [c] together, I have come to the conclusion that the UT’s conclusion in this part of its decision is indeed not sufficiently reasoned. Economy in giving reasons is generally a virtue, but, as I have said, what is required depends on the particular case. The Tribunal’s conclusion was that it would not be unduly harsh for a child of five to be removed to the IKR in circumstances where she would lose for the rest of her childhood at least the benefits of being a British citizen and where there were, on the evidence, real questions about RA’s ability to find decent accommodation and a job. Such a conclusion required, in my view, a full explanation which demonstrated that all the material considerations had indeed been fully taken into account.

I mention for completeness one other factor that featured in the argument before us. Peter Jackson LJ drew attention to the absence of any reference in the undue harshness assessment of the impact on Y of losing contact with her maternal grandparents, with whom on the evidence she had a close relationship. I must say that I too would have expected that to be a significant factor, particularly as the assistance of the grandparents was treated at para. 58 of the decision as a factor mitigating the impact on KI and Y in the stay scenario. But Mr Pilgerstorfer pointed out that that was not a matter on which any reliance had been placed in the grounds of appeal or the joint skeleton argument; and he also told us, without demur from Mr Bazini, that it had not been relied on in the Tribunal.

The scenario where KI and Y stay in the UK

I can take this aspect more shortly. I have set out the entirety of the UT’s reasoning on the issue of whether, if KI and Y stayed in the UK, the effect on them, and more particularly on Y as a five-year old child, of RA’s deportation would be unduly harsh. Mr Bazini submitted that it was unclear from that reasoning what factors had been taken into account in considering the issue of undue harshness and that that made it impossible to see whether the best interests of Y had indeed been treated as a primary consideration.

I have to say that I believe that this is a fair criticism. Paras. 58-60 of the decision do not in my view amount to the kind of particularised consideration that it is clear from Zoumbas is necessary in a case of this kind. In contrast to what we saw in HA’s case, there is simply no indication of the kind of role that RA played in the life of his daughter, from which it would be possible to make a considered assessment of the degree of harshness that separation from him would entail.

I am aware that the degree of detail in a tribunal’s reasoning may reflect the way the case was put before it. However, we were shown Mr Bazini’s skeleton argument before the UT. This relied on, and quoted from, the decision of the UT in MK (Sierra Leone), to which I have already had occasion to refer (see in particular paras. 45-46

I do not say that Mr Bazini’s reliance on MK was well-founded (as to this, see para. 129 below), but I quote this passage to show that the substantive points based on the impact on Y of separation from RA were squarely made in the UT; and I believe that they needed to be directly addressed.

We were also shown the witness statement from RA that was before the UT, and it is fair to say that it goes no further than claiming that he “has always been a very involved father”. There was apparently a witness statement from his wife but we were not shown this. However, even if it contained nothing further of any substance as regards Y, both RA and his wife gave oral evidence, and this aspect could and should have been explored with them further to the extent that it was not adequately covered in cross-examination.

In short, the Tribunal’s conclusion on the stay scenario is in my judgement insufficiently reasoned. Since I have reached the same conclusion in relation to the go scenario the result is that HA’s appeal will have to be allowed and the case remitted to the UT for reconsideration.

That is all that it is strictly necessary, but I think it would be useful to say something more about Mr Bazini’s reliance on MK (Sierra Leone) because it raises a point about the use of “factual precedents” which is of some general significance and which the

I start with the last point. What Lord Carnwath actually says at para. 27 of his judgment is that he does not “understand the conclusion on the facts of that case to be controversial”. Even if that is regarded as an implicit endorsement, the conclusion which he had just quoted related to only one element in the UT’s decision, being its conclusion that it would be unduly harsh for the children to have to relocate to Sierra Leone: see para. 46 above. But the relevant question for our purposes is about the stay scenario. Lord Carnwath does not allude to the UT’s finding on that issue, and I do not think it can be assumed that he had it in mind when he said what he did at para. 27 of his judgment.

I turn to the question whether, even if it was not approved in KO, the UT’s conclusion on the stay scenario in MK (Sierra Leone) should nevertheless have been treated by the UT in this case as having some kind of authoritative status. I agree with the Tribunal that it had no such status. I am not so austere as to say that a tribunal may not sometimes find it useful to consider the outcomes in other apparently similar cases as a cross-check on a conclusion which it is minded to reach. But the exercise can only ever be valuable up to a point. Ultimately the tribunal has to make its own evaluation of the particular facts before it. As the UT put it at para. 14 of its decision, in response to the same submission from Mr Bazini:

“Although the application of a legal test to a particular set of facts can sometimes shed light on the way in which the test falls to be applied, it is the test that matters. If this were not so, everything from the law of negligence to human rights would become irretrievably mired in a search for factual precedents.”

I would add that it is often difficult to be sure that the facts of two cases are in truth substantially similar. And, even where they are, the assessment of “undue harshness” is an evaluative exercise on which tribunals may reasonably differ. If this kind of factual comparison were legitimate it might indeed be deployed against RA, since in KO Exception 2 was held not to apply on facts that were at least as close to those of his case as those in MK: see para. 83 above.

The proportionality assessment

As in HA’s case, if RA eventually succeeds in bringing himself within the terms of section 117C (5) it will be unnecessary for him to rely on section 117C (6). But in his case also his challenge to the UT’s conclusion raises issues which it would be useful for me to address. The pleaded ground reads:

“The UT erred in concluding that there were no very compelling circumstances within the meaning of section 117C (6) of the 2002 Act, failing to take into account all material factors including [a] the Appellant’s immigration history, [b] his rehabilitation, and [c] the circumstances of his offence.”

Again, I have inserted [a]-[c] to denote the particular complaints made about the UT’s reasoning.

As to [a], RA’s point as developed in the skeleton argument is that it was an important factor weighing in his favour that he had been in this country since the age of 14 and that he had been given leave to remain as a minor and again, later, following his marriage. The UT did not, however, regard those as the most significant points about his immigration history and in fact regarded it as weighing against him. It pointed out at para. 62 that he had come here without any lawful basis (he is not a refugee) and that he remained in this country unlawfully between the expiry of his leave to remain as a minor until he was granted discretionary leave in 2016, which was only shortly before his conviction. It also pointed out at para. 63 that he had no leave to remain at the time that he married his wife. I can see nothing wrong in that reasoning.

As to [b], the UT dealt with the issue of rehabilitation in Section F of its decision, as part of its discussion of the general law. It said:

“32.

As the Court of Appeal pointed out in Danso v Secretary of State for the Home Department [2015] EWCA Civ 596, courses aimed at rehabilitation, undertaken whilst in prison, are often unlikely to bear material weight, for the simple reason that they are a commonplace; particularly in the case of sexual offenders.

33.

As a more general point, the fact that an individual has not committed further offences, since release from prison, is highly unlikely to have a material bearing, given that everyone is expected not to commit crime. Rehabilitation will therefore normally do no more than show that the individual has returned to the place where society expects him (and everyone else) to be. There is, in other words, no material weight which ordinarily falls to be given to rehabilitation in the proportionality balance (see SE (Zimbabwe) v Secretary of State for the Home Department [2014] EWCA Civ 256, paragraphs 48 to 56). Nevertheless, as so often in the field of human rights, one cannot categorically say that rehabilitation will never be capable of playing a significant role (see LG (Colombia) v Secretary of State for the Home Department [2018] EWCA Civ 1225 (Footnote: ¹¹)). Any judicial departure from the norm would, however, need to be fully reasoned.”

RA’s case is squarely that that self-direction was wrong. The joint skeleton argument contends that it

“… undermines a long line of authority including the recent case of LG (Colombia) …, in which the courts have held that matters such as the applicant’s age on arrival, length of time in the UK, private and family ties formed and evidence of rehabilitation were sufficient to establish very compelling circumstances”.

That submission faces the difficulty that paras. 32-33 of the UT’s summary were quoted in full by Hamblen LJ at para. 84 of his judgment in Binbuga v Secretary of State for the Home Department [2019] EWCA Civ 551 (with which Floyd LJ agreed) in support of the proposition that “rehabilitation involves no more than returning an individual to the place society expects him to be … [and] will generally be of little or no material weight in the proportionality balance”. However it is fair to say that Hamblen LJ does not discuss the issue any further, and I do have some difficulty with that way of putting it. I think that a rather fuller consideration of the authorities would be useful.

I should say by way of preliminary that the core idea behind the concept of

“rehabilitation” in this context is the elimination, or at least the substantial reduction, of the risk of future offending. That can of course never be definitively assessed, but various forms of evidence of it, of varying cogency, may be adduced. One is simply that the criminal has committed no offences since his release: how cogent that is will depend on the circumstances. Others may include formal assessments of the risk of future offending and/or the taking of courses or other measures designed to address the causes of the offending behaviour. Occasionally, foreign criminals may be able to show evidence of exceptional positive contributions to society since release: that too can be described as “rehabilitation” but it may involve different considerations.

Although the skeleton argument refers to “a long line of authority” in relation to the effect of the very wide combination of factors there listed, we are concerned only with authorities specifically relating to an appellant’s rehabilitation. As to that, there are several authorities that refer to rehabilitation in passing, or in general terms, as relevant considerations in the overall proportionality assessment – see, for example, NA (Pakistan) (at para. 112 – “absence of any real risk of re-offending”) and Hesham Ali (per Lord Reed at para. 38 – “conduct since the offence was committed”). The applicant’s rehabilitation also played an important part in the seminal Strasbourg case of Boultif: see para. 51 in the judgment of the majority and para. O-15 in the concurring judgment of the minority. But we were only referred to three judgments in which the question has been given fuller consideration, being those to which the UT itself refers. I take them in date order.

The first is SE (Zimbabwe) v Secretary of State for the Home Department [2014] EWCA Civ 256. I have to say that I find this of limited general application. What it decides is only that a potential deportee cannot rely for article 8 purposes on an argument that his deportation would make it more difficult to complete a programme of rehabilitation on which he is engaged.

More useful is Danso v Secretary of State for the Home Department [2015] EWCA Civ 596. The appellant was a sexual offender who had been sentenced to 4½ years’ imprisonment. His appeal against a deportation order was dismissed by both the FTT and the UT, and their decisions were upheld in this Court. One of the factors on which the appellant relied in his appeal was that he had undergone courses in prison designed to address his offending behaviour and reports on the likelihood of reoffending, including one from a consultant psychologist which said that the formal assessment of his risk of further sexual offending as “medium” was “probably overstated” and that the risk of violent offending on his part was low. At para. 20 of his judgment (with which the other members of the Court agreed) Moore-Bick LJ said:

“[Counsel for the appellant] submitted that the tribunal should have placed much greater weight on the appellant's rehabilitation and the fact that he did not pose a significant risk of re-offending. He suggested that far too little importance is attached to factors of that kind, with the result that those who commit offences have little incentive to co-operate with the authorities and make a positive effort to change their ways. I have some sympathy with that argument and I should not wish to diminish the importance of rehabilitation. It may be that in a few cases it will amount to an important factor, but the fact is that there is nothing unusual about the appellant's case. Most sex offenders who are sentenced to substantial terms of imprisonment are offered courses designed to help them avoid re-offending in future and in many cases the risk of doing so is reduced. It must be borne in mind, however, that the protection of the public from harm by way of future offending is only one of the factors that makes it conducive to the public good to deport criminals. Other factors include the need to mark the public's revulsion at the offender's conduct and the need to deter others from acting in a similar way. Fortunately, rehabilitation of the kind exhibited by the appellant in this case is not uncommon and cannot in my view contribute greatly to the existence of the very compelling circumstances required to outweigh the public interest in deportation.”

That passage clearly accepts that rehabilitation is in principle a relevant consideration. However it makes it equally clear it will not generally be a factor carrying great weight – “it may be that in a few cases it will amount to an important factor”. One reason given is that rehabilitation of the kind exhibited by the appellant – i.e. the successful completion of a sex offender’s course – was “not uncommon”. MooreBick LJ’s other reason is that the prevention of the risk of offending is only one element in the public interest in deportation. His reference to “the need to mark the public's revulsion at the offender's conduct and the need to deter others from acting in a similar way” is plainly derived from such well-known cases as N (Kenya) v Secretary of State for the Home Department [2004] EWCA Civ 1094, OH (Serbia) v Secretary of State for the Home Department [2008] EWCA Civ 694, and DS (India) v

The third case is LG (Colombia) [2018] EWCA Civ 1225. In that case the FTT had allowed an appeal against deportation in the case of a medium offender who had lived in the UK for thirty years since coming here as a child. The offence had been committed five years earlier and he had in that time wholly abandoned his previous chaotic lifestyle. At para. 29 of his judgment McFarlane LJ summarised the FTT’s conclusion as being that

“… [his] age at arrival, the length of time that he had lived in this country, the family and private life ties he had with his family, his new partner and more generally and, finally, the evidence of reform and rehabilitation were sufficient to establish very compelling circumstances”

and added that in relation to rehabilitation the FTT had held in terms that that factor “did not carry much weight, but was still of some significance”. That is less full than the discussion in Danso but entirely consistent with it: rehabilitation is an admissible consideration but not one that will normally carry great weight on its own.

What those authorities seem to me to establish is that the fact that a potential deportee has shown positive evidence of rehabilitation, and thus of a reduced risk of reoffending, cannot be excluded from the overall proportionality exercise. The authorities say so, and it must be right in principle in view of the holistic nature of that exercise. Where a tribunal is able to make an assessment that the foreign criminal is unlikely to re-offend, that is a factor which can carry some weight in the balance when considering very compelling circumstances. The weight which it will bear will vary from case to case, but it will rarely be of great weight bearing in mind that, as Moore-Bick LJ says in Danso, the public interest in the deportation of criminals is not based only on the need to protect the public from further offending by the foreign criminal in question but also on wider policy considerations of deterrence and public concern. I would add that tribunals will properly be cautious about their ability to make findings on the risk of re-offending, and will usually be unable to do so with any confidence based on no more than the undertaking of prison courses or mere assertions of reform by the offender or the absence of subsequent offending for what will typically be a relatively short period.

That summary may come to much the same thing in practice as the UT’s proposition that “no material weight … ordinarily falls to be given to rehabilitation in the proportionality balance”; but I think, with respect, that it is more accurately expressed, and I cannot in any event adopt its reasoning that “rehabilitation will … normally do no more than show that the individual has returned to the place where society expects him … to be”, notwithstanding its endorsement (not, I think, as a matter of ratio) in Binbuga. I do not think that it properly reflects the reason why rehabilitation is in principle relevant in this context, which is that it goes to reduce (one element in) the weight of the public interest in deportation which forms one side of the proportionality balance. It is not generally to do with being given credit for being a law-abiding citizen: as the UT says, that is expected of everybody, but the fact that that is so is not a good reason for denying to an appellant such weight as his rehabilitation would otherwise carry.

RA’s case on rehabilitation amounts simply to the fact that he has not committed any further offence and there is no reason to believe that he is likely to. The UT did not expressly put that factor into the proportionality balance. I think it should have done, but it follows from what I have said above that it is unlikely that it would carry great weight, and I am far from saying that it would necessarily have made a decisive difference to the outcome.

I turn finally to element [c], “the circumstances of [RA’s] offence”. Mr Bazini’s submission before us was that the offence which led to RA’s imprisonment was unusual and committed in extenuating circumstances. He did in fact have the right both to enter Iraq and, following the recent grant of leave to remain, to re-enter the UK, so he was not seeking to use the forged passport to obtain a benefit to which he was not entitled. He only sought to use it because he wanted to visit his mother and was unable to do so in any other way. Mr Bazini pointed out that a tribunal was entitled to take into account mitigating circumstances of that kind even though they will have been taken into account in arriving at the sentence: see para. 94 above.

I am far from sure that this argument was advanced before the UT. It does not appear in Mr Bazini’s skeleton argument below, and in his witness statement before the UT RA made the different point (which on the face of it was inconsistent with his plea of guilty) that he was unaware that the passport that his mother sent him was not a genuine document. In any event the UT does not address the point. In those circumstances I prefer to say nothing more about it, save to observe that I doubt if it can add much to the fact that is in any event available to RA, namely that his offence is at the very bottom of the scale of seriousness.

There is, however, another criticism of the UT’s reasoning as regards the seriousness of the offence which emerged during the oral argument. It says at para. 62 that it gives due weight to the fact that RA’s sentence “is at the bottom of the range covered by section 117C (3)”. It was right to do so: see para. 92 above. But it then notes as countervailing factors (a) that the Judge gave credit for a guilty plea and (b) that “as the sentencing judge pointed out, the offence was a serious one”. It was submitted that it was wrong to treat those points as diminishing such weight as RA could otherwise put on the shortness of the sentence.

As to (a), I appreciate the logic of the UT’s point. If the importance of the sentence is as an indicator of the seriousness of the offence, then that is more accurately reflected in the level of sentence pre-discount. On the other hand, the statutory provisions themselves make no distinction between discounted and undiscounted sentences, which suggests that this degree of refinement is rather out of place. It might also be thought wrong that the fact that RA had acted responsibly and acknowledged his guilt was not allowed to be put into the proportionality balance. I think the UT should have proceeded without qualification on the basis that his sentence was at the very bottom of the relevant range.

As to (b), I think that the observation that the offence was “serious” was inappropriate for the reasons given at para. 94 above: of course offences of this kind are serious, but the authoritative measure of the degree of seriousness is the sentence imposed. I would add that the Tribunal was also wrong to say that the sentencing judge had himself described the offence as serious. We have seen the sentencing remarks, in which he says simply that an immediate custodial sentence is appropriate because travelling on false documents undermines “the immigration and travel pillars upon

Again, I do not wish to be understood as saying that the fact that RA’s sentence was at the very bottom of the relevant range is capable by itself of outweighing the strong public interest in the deportation of foreign criminals. I say only that it is, as indeed the UT recognised, a material consideration in striking the relevant proportionality balance.

For the reasons given I would allow both appeals and remit the underlying appeals to the UT for re-determination.

I agree with the clear and far-reaching analysis that Underhill LJ has carried out in relation to these important matters. I also agree that these appeals should be allowed and, with full respect to this specialist tribunal, that the underlying appeals must be reheard. The existing decisions do not in my view adequately address the circumstances of the children of these foreign criminals in the way that is required under Exception 2 or under the proportionality assessment.

Parliament has enacted two important public interests in cases involving children. Section 117C of the Immigration Act 2014 enshrines the public interest in the deportation of foreign criminals. Section 55 of the Borders, Citizenship and Immigration Act 2009 requires the Secretary of State to make arrangements to ensure that in her immigration functions she has regard to the need to safeguard and promote the welfare of children, which translates into a duty to make the best interests of the child a primary consideration. The decision-maker must bring both of these elements into play in accordance with the complete statutory code, applying (as may be appropriate in the individual case) the fulcrum of undue harshness, or of very compelling circumstances or of proportionality. A resulting decision to deport a parent may produce hugely detrimental consequences for a child but, provided his or her best interests have been adequately identified and weighed in the balance as a primary consideration, the decision will be lawful. But a decision that does not give primary consideration to the children’s best interests will be liable to be set aside.

The practical effect of Section 55 has been summarised in Zoumbas. I draw particular attention to the final parts of Lord Hodge’s summary, reproduced for convenience:

“(5)

It is important to have a clear idea of a child's circumstances and of what is in a child's best interests before one asks oneself whether those interests are outweighed by the force of other considerations;

(6)

To that end there is no substitute for a careful examination of all relevant factors when the interests of a child are involved in an article 8 assessment; and

(7)

A child must not be blamed for matters for which he or she is not responsible, such as the conduct of a parent.”

To these I would respectfully add that the Section 55 duty falls on the decision-maker. A child will not usually be in a position to urge his or her point of view and the decision-maker cannot treat the child as if he or she had some burden of proof.

The assessment that has to be carried out is therefore one that is adequately informed and specific to the individual child as a person distinct from the offending parent. It requires the decision-maker, as part of the overall assessment, to look at matters from the child’s point of view – in the case of Exception 2, the question explicitly concerns undue harshness to the child.

There are two broad ways in which it seems to me that a decision-maker may inadvertently be deflected from giving primary consideration to the best interests of the child of a foreign criminal. One is by focusing on the position of children generally rather than on the best interests of the individual child. The other is by treating physical harm as intrinsically more significant that emotional harm. I will take these in turn.

In order to maintain focus on the individual child, it will be helpful for the decisionmaker to apply the words of the statutory tests themselves. By their nature, commentaries on the tests may be illuminating, but they are not, as Underhill LJ has shown at [56], a substitute for the statutory wording. For example, Lord Carnwath’s reference in paragraph 23 of KO (Nigeria) to undue harshness to “any child” cannot have been intended to set up a notional comparator, if only because it is not possible to know what the circumstances of such a child might be. For some children the deportation of a largely absent parent may be a matter of little or no real significance. For others, the deportation of a close caregiver parent where face to face contact cannot continue may be akin to a bereavement. A decision that gives primary consideration to the best interests of the child will instead focus on the reality of that child’s actual situation and the decision-maker will be more assisted by addressing relevant factors of the kind identified by Underhill LJ at the end of [56] than by making generalised comparisons. Likewise, as explained in the footnote to [48], the aphorism “That is what deportation does” is an important truth, but it is not a substitute for a proper consideration of the individual case. The full citation from Sedley LJ in Lee makes this clear:

“The tragic consequence is that this family, short-lived as it has been, will be broken up for ever because of the appellant's bad behaviour. That is what deportation does. Sometimes the balance between its justification and its consequences falls the other way, but whether it does so is a question for an immigration judge.”

It can be seen that the aphorism frames the question; it does not provide the answer. In that case, the parent was a supplier of Class A drugs who had received a seven year sentence and was facing deportation to Jamaica, and the balance not surprisingly fell in favour of deportation despite the impact on the two young children. In the same way, I agree with Underhill LJ’s observations at [34] and [35] that decision-makers

My other general observation concerns the treatment of emotional harm. Section 31(9) of the Children Act 1989 defines harm as ill-treatment or the impairment of health or physical, intellectual, emotional, social or behavioural development. Reflecting our contemporary understanding of the importance of emotional development and mental health, there is no hierarchy as between physical and nonphysical harm. It must therefore always be recognised that for the child the consequences of going with both parents may be experienced as far less harsh than staying with one parent. Despite this, it may be easier for decision-makers to envisage the harm that may be done by expecting a family to experience precarious or even dangerous physical conditions than to factor in at full worth the lifelong emotional harm of terminating the relationship between a child and a close parent during the child’s minority and possibly forever. Both situations are grim but for the child neither is intrinsically grimmer than the other. Provided the decision-maker faces up to the reality of the child’s situation and gives it primary consideration, the public interest in deportation may prevail, but it will not do to minimise the emotional impact on the child of the severing of ties by reference to the doubtful prospect of maintaining relationships over many years by indirect means only, or by reciting the fact that this is what deportation does.

Turning to the present cases, at the time of the Upper Tribunal decision the children of

The Tribunal was less sanguine about the maintenance of the parental relationship from abroad. It found that if their father was deported it would have “a significant impact” on the children, they would be separated from him “for at least 10 years, if not permanently” and that maintaining contact by Skype or telephone was “far from ideal”. I agree with Underhill LJ when he says at [82] that the “any child” approach taken by the Tribunal at paragraph 71 was not correct. I also agree, for the reasons given above, that the decision in KO itself is not to be treated as factual precedent.

In the case of RA, the child was aged 5. I do not wish to add anything to what

The decision therefore turned on matters of physical care, financial support, community ties and parental deserts (“You have, through your offending behaviour, effectively exempted yourself from forming part of your child's day-to-day life.”) There is no reference to the real losses to the child of her father’s deportation, whether or not they turned out to be outweighed by the public interest. The Upper Tribunal’s decision itself has little to say about Y’s situation, beyond the fact that it would be hard but far from unduly harsh. It is said that there would be serious adverse effects on her, but these are not explored. It is said that opportunities to meet may be possible, but that she would clearly miss her father’s daily presence in her life. I agree that this reasoning is insufficient to underpin the conclusion. It is also interesting to note that whilst the Tribunal felt some hesitation about the “go scenario”, it did not feel the same about the “stay scenario”. Returning to my discussion about emotional harm, it is not possible to understand why it held these differing views about Y’s best interests. When taken alongside the emphasis given to the seriousness of the offence (see [145] to [147] above), I am clear that RA’s underlying appeal must also be reheard.

I agree with both judgments.