The Secretary of State for the Home Department v AA (Poland)

This appeal is about a decision made by the Secretary of State (SSHD) to deport AA, an EU citizen who had committed serious sexual offences in this country.

The First-tier Tribunal (FtT) allowed AA’s appeal against the SSHD’s decision, holding that AA’s removal would infringe his rights under the EU Treaties as implemented by the Immigration (European Economic Area) Regulations 2016 (the 2016 Regulations) and his right to respect for private and family life under Article 8 of the European Convention on Human Rights (the Convention). The Upper Tribunal (UT) dismissed an appeal by the SSHD, holding that the FtT had made no error of law and that its conclusions were sufficiently reasoned. The SSHD brings this second appeal with the permission of Asplin LJ.

The grounds of appeal raise two main issues: did the FtT err in law by (1) misapplying the 2016 Regulations and (2) treating the application of the 2016 Regulations as effectively decisive of AA’s claim to remain in the UK on the basis of his human rights?

AA is anonymised as he has been throughout the proceedings. That is not for his own sake but only because it is a necessary measure for the protection of his daughter (V). V was, as an infant, a victim of the relevant offending and benefits from the right to lifetime anonymity provided for by the Sexual Offences (Amendment) Act 1992.

AA is a Polish national, born on 29 May 1981. In 2006 he moved to the UK. In April 2007 he met his wife K who had moved to the UK from Poland earlier that year. They were married in 2012. In January 2014, AA gained a Master’s degree in Aeronautical Engineering. In July of the same year V was born. At this point AA began openly questioning his gender identity. (I use male pronouns because AA has indicated he prefers this).

On 23 February 2016, police attended AA’s home with a search warrant, acting on intelligence that someone at the address had used the internet to access indecent images of children. On AA’s devices the police found some 1,450 such images, including some 300 in Category A, the most serious. Some of the Category B and C images were of V. AA was arrested and charged. He pleaded guilty to three counts of making indecent images but contested charges of sexually assaulting his daughter, taking two indecent Category B images of her, and taking 17 indecent Category C images of her including a video. On 4 May 2018, in the Crown Court at Isleworth, he was convicted of all those charges.

On 18 June 2018, AA was sentenced to a total of five years’ imprisonment comprising four years in respect of the sexual assault on V, concurrent terms for the other offending against V, and 12 months consecutive for the offences to which he had pleaded guilty. The sentencing judge made a wide-ranging Sexual Harm Prevention Order (SHPO) for 10 years. As a further consequence of his conviction AA was subject to the sex offender registration and notification requirements for 10 years and was liable to be placed on the barring list by the Disclosure and Barring Service.

On 2 July 2018, AA was warned by the SSHD that he could be liable to deportation pursuant to the 2016 Regulations and invited to submit representations. On 16 December 2020, the custodial portion of AA’s sentence came to an end and he was released on licence. On 20 December 2020, having considered AA’s representations that his deportation would be contrary to the 2016 Regulations and his Article 8 rights, the SSHD decided that he should be deported (the Decision). A supplementary decision letter of 6 April 2021 affirmed the Decision on slightly different grounds.

Section 32 of the UK Borders Act 2007 requires the SSHD to make a deportation order in respect of a person who is not a British Citizen and is convicted in the UK of an offence and sentenced to at least 12 months’ imprisonment (a “foreign criminal”) unless the case falls within an exception under s 33 of the 2007 Act. At the material times this case would have fallen within one of the exceptions if AA’s removal would breach his right to respect for private and family life under Article 8 (Exception 1, s 33(2)(a)) or his rights under the EU Treaties (Exception 3, s 33(4)).

The main focus of attention in this case has been on AA’s rights under the EU Treaties.

The expulsion of EU citizens by Member States is governed by Directive 2004/38/EC, known as the Citizens’ Rights Directive. Article 27 of the Directive authorises restrictions on freedom of movement on grounds of public policy, public security or public health, subject to certain preconditions. Among these are that the measure must comply with the principle of proportionality and be based exclusively on the personal conduct of the individual concerned which “must represent a genuine, present and sufficiently serious threat affecting one of the fundamental interests of society.”

Article 28 sets out three levels or tiers of protection against expulsion. In every case the Member State must take account of “considerations such as how long the individual concerned has resided on its territory, his/her age, state of health, family and economic situation, social and cultural integration into the host Member State and the extent of his/her links with the country of origin” (Article 28(1)). If the individual concerned has the right of permanent residence in the host State an expulsion decision must not be taken against them “except on serious grounds of public policy or public security” (Article 28(2)). If the individual has “resided in the host Member State for the previous ten years” an expulsion decision must not be taken against them “except if the decision is based on imperative grounds of public security, as defined by Member States” (Article 28(3)).

Decisions of the CJEU have made clear that the ten-year residence provision in Article 28(3) is not to be interpreted or applied literally or mechanically. It is not just a question of arithmetic. The factors specified in Article 28(1) are relevant. Where the individual in question has been subjected to a period of imprisonment that may break integrative links previously forged with the host Member State; to decide whether that is so it is necessary to conduct an overall assessment of the situation at the time of the relevant decision: see Land Baden-Wurtemburg v Tsakouridis (Case C-145/09) [2011] 2 CMLR 11 [33], B and Vomero [2019] QB 126 [72]-[75]. In Secretary of State for the Home Department v Viscu [2019] EWCA Civ 1052 [44] Flaux LJ identified the following points as established by the European jurisprudence:

“…(i) that the degree of protection against expulsion to which a Union national resident in another member state is entitled under the Directive is dependent upon the degree of integration of that individual in the member state; (ii) that, in general, a custodial sentence is indicative of a rejection of societal values and thus of a severing of integrative links with the member state but (iii) that the extent to which there is such a severing of integrative links will depend upon an overall assessment of the individual's situation at the time of the expulsion decision.”

The 2016 Regulations transposed the provisions of the Citizens’ Rights Directive into domestic law with added details. Although the 2016 Regulations have now been revoked, they continue to apply to this case because AA was protected at the time of the Decision: see regulation 2(3) and Schedule 1 of the Citizens’ Rights (Restrictions of Rights of Entry and Residence) (EU Exit) Regulations 2020.

Regulation 23(6) of the 2016 Regulations authorises the removal of an EEA national who has entered the UK if the SSHD “has decided that the person’s removal is justified on grounds of public policy, public security or public health in accordance with regulation 27”. Such a decision is termed a “relevant decision”.

Regulation 27 contains the following provisions about such decisions:

“(3)

A relevant decision may not be taken in respect of a person with a right of permanent residence under regulation 15 except on serious grounds of public policy and public security.

(4)

A relevant decision may not be taken except on imperative grounds of public security in respect of an EEA national who—

(a)

has a right of permanent residence under regulation 15 and who has resided in the United Kingdom for a continuous period of at least ten years prior to the relevant decision; …

…

(5)

The public policy and public security requirements of the United Kingdom include restricting rights otherwise conferred by these Regulations in order to protect the fundamental interests of society, and where a relevant decision is taken on grounds of public policy or public security it must also be taken in accordance with the following principles—

(a)

the decision must comply with the principle of proportionality;

(b)

the decision must be based exclusively on the personal conduct of the person concerned;

(c)

the personal conduct of the person must represent a genuine, present and sufficiently serious threat affecting one of the fundamental interests of society, taking into account past conduct of the person and that the threat does not need to be imminent;

(d)

matters isolated from the particulars of the case or which relate to considerations of general prevention do not justify the decision;

(e)

a person’s previous criminal convictions do not in themselves justify the decision;

(f)

the decision may be taken on preventative grounds, even in the absence of a previous criminal conviction, provided the grounds are specific to the person.

(6)

Before taking a relevant decision on the grounds of public policy and public security in relation to a person (“P”) who is resident in the United Kingdom, the decision maker must take account of considerations such as the age, state of health, family and economic situation of P, P’s length of residence in the United Kingdom, P’s social and cultural integration into the United Kingdom and the extent of P’s links with P’s country of origin.”

Regulation 27(5) reflects and expands on Articles 27(1) and (2) of the Citizens’ Rights Directive. Regulation 27(6) reflects Article 28(1); Regulation 27(3) corresponds to Article 28(2) (serious grounds protection); and Regulation 27(4) corresponds to Article 28(3) (imperative grounds protection).

In Restivo (EEA – prisoner transfer) [2016] UKUT 449 (IAC) the UT considered the right approach to the assessment of whether a person’s conduct represents a “genuine, present and sufficiently serious threat” for the purposes of the Regulations. (Footnote: ¹ ) The FtT had held that it was not and could not be established that such a threat was posed by a convicted murderer whom the trial judge had described as “a man capable of inhuman depravity” as he would remain in prison for at least a further 37 years. The UT held that this was an error of law which wrongly conflated the assessment of risk with the question of management of that risk. The UT summarised its conclusions at [34]:

“Where the personal conduct of a person represents a genuine, present and sufficiently serious threat affecting one of the fundamental interests of society, the fact that such threat is managed while that person serves his or her prison sentence is not itself material to the assessment of the threat he or she poses. The threat exists, whether or not it cannot generate further offending simply because the person concerned, being imprisoned, has significantly less opportunity to commit further criminal offences.”

Regulation 27(8) provides:

“A court or tribunal considering whether the requirements of this regulation are met must (in particular) have regard to the considerations contained in Schedule 1 (considerations of public policy, public security and the fundamental interests of society, etc)”

The considerations contained in Schedule 1 include the following:

“Considerations of public policy and public security

1.

The EU Treaties do not impose a uniform scale of public policy or public security values: member States enjoy considerable discretion, acting within the parameters set by the EU Treaties, applied where relevant by the EEA agreement, to define their own standards of public policy and public security, for purposes tailored to their individual contexts, from time to time.

Application of paragraph 1 to the United Kingdom

2.

An EEA national or the family member of an EEA national having extensive familial and societal links with persons of the same nationality or language does not amount to integration in the United Kingdom; a significant degree of wider cultural and societal integration must be present before a person may be regarded as integrated in the United Kingdom

3.

Where an EEA national or the family member of an EEA national has received a custodial sentence, or is a persistent offender, the longer the sentence, or the more numerous the convictions, the greater the likelihood that the individual’s continued presence in the United Kingdom represents a genuine, present and sufficiently serious threat affecting one of the fundamental interests of society.

4.

Little weight is to be attached to the integration of an EEA national or the family member of an EEA national within the United Kingdom if the alleged integrating links were formed at or around the same time as-

a)

the commission of a criminal offence;

b)

an act otherwise affecting the fundamental interests of society;

c)

the EEA national or family member of an EEA national was in custody.

5.

The removal from the United Kingdom of an EEA national or the family member of an EEA national who is able to provide substantive evidence of not demonstrating a threat (for example, through demonstrating that the EEA national or the family member of an EEA national has successfully reformed or rehabilitated) is less likely to be proportionate.

6.

…

7.

For the purposes of these Regulations, the fundamental interests of society in the United Kingdom include-

a)

preventing unlawful immigration and abuse of the immigration laws, and maintaining the integrity and effectiveness of the immigration control system (including under these Regulations) and of the Common Travel Area;

b)

maintaining public order;

c)

preventing social harm;

d)

preventing the evasion of taxes and duties;

e)

protecting public services;

f)

excluding or removing an EEA national or family member of an EEA national with a conviction (including where the conduct of that person is likely to cause, or has in fact caused, public offence) and maintaining public confidence in the ability of the relevant authorities to take such action;

g)

tackling offences likely to cause harm to society where an immediate or direct victim may be difficult to identify but where there is wider societal harm (such as offences related to the misuse of drugs or crime with a cross-border dimension as mentioned in Article 83(1) of the Treaty on the Functioning of the European Union);

h)

combating the effects of persistent offending (particularly in relation to offences, which if taken in isolation, may otherwise be unlikely to meet the requirements of regulation 27);

i)

protecting the rights and freedoms of others, particularly from exploitation and trafficking;

j)

protecting the public;

k)

acting in the best interests of a child (including where doing so entails refusing a child admission to the United Kingdom, or otherwise taking an EEA decision against a child);

l)

countering terrorism and extremism and protecting shared values.”

Turning to the application of Article 8 of the Convention, deportation is likely to represent an interference with a person’s right to respect for private or family life or their home. Where that is so, the question will arise of whether it would be justified under Article 8(2). If deportation would be lawful and pursue a legitimate aim the question will arise of whether it is proportionate to that aim. Part 5A of the Nationality Immigration and Asylum Act 2002 lays down rules for judges to apply when determining that question (referred to as “the public interest question”).

Section 117A(2) of the 2002 Act provides that:

“in considering 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”

Section 117B(1) provides that “the maintenance of effective immigration controls is in the public interest.” Section 117B(5) provides that “little weight should be given to a private life established by a person at a time when the person’s immigration status is precarious.” Although this provision does not apply on the facts of this case it is relevant for reasons that will appear.

Section 117C(6) provides that in the case of a foreign criminal (C) who has been sentenced to a period of imprisonment of at least four years (conventionally referred to as a “serious offender”) “the public interest requires deportation unless there are very compelling circumstances, over and above those described in exceptions 1 and 2”. Those exceptions are set out in s 117C(4) and (5):

“(4)

Exception 1 applies where—

(a)

C has been lawfully resident in the United Kingdom for most of C's life,

(b)

C is socially and culturally integrated in the United Kingdom, and

(c)

there would be very significant obstacles to C's integration into the country to which C is proposed to be deported.

(5)

Exception 2 applies where C has a genuine and subsisting relationship with a qualifying partner, or a genuine and subsisting parental relationship with a qualifying child, and the effect of C's deportation on the partner or child would be unduly harsh.”

The purpose of Part 5A of the 2002 Act is to reflect “the Government’s and Parliament’s view of how, as a matter of public policy, the balance should be struck” and “an assessment of all the factors relevant to the application of article 8 …”; the provisions are designed “to provide clear guidelines to limit the scope for judicial evaluation” and “to narrow rather than to widen the residual area of discretionary judgement for the court to take account of public interest or other factors not directly reflected in the wording of the statute”: KO (Nigeria) v Secretary of State for the Home Department [2018] UKSC 53, [2018] 1 WLR 5273 [12]-[15] (Lord Carnwath JSC, with whom the other Justices agreed).

The SSHD decided to deport AA pursuant to the 2016 Regulations. The decision letter of December 2020 concluded that he constituted a genuine, present, and sufficiently serious risk to the public to justify his deportation on grounds of public policy and public security, and that it was proportionate to order his removal to Poland on those grounds. The SSHD initially proceeded on the footing that AA had not shown a right of permanent residence but said that the decision would have been the same if he had. In the supplementary decision letter the SSHD accepted that AA did have a right of permanent residence and, further, that he had resided in the UK for a period of at least 10 years between 2006 and 2018. The SSHD did not accept, however, that AA was entitled to “imperative grounds” protection. This was because, applying the “integration” test identified in Tsakouridis, AA was insufficiently integrated into the UK community. The SSHD accepted that the case fell within the “serious grounds” category but concluded that this threshold was met because of the real risk of reoffending and the risk of harm if that occurred.

The SSHD rejected AA’s human rights claim on the basis that, although the 2002 Act did not apply directly to the case because AA was an EEA national, the provisions of Part 5A were relevant because Article 8 applies equally to everyone regardless of nationality and “it would not be fair to consider Article 8 claims from EEA nationals either more or less generously than claims from non-EEA nationals.” The SSHD concluded that AA had not shown that either of the exceptions in s 117C(4) and (5) applied to him or that there were “very compelling circumstances” over and above those exceptions within the meaning of s 117C(6).

AA’s appeal to the FtT raised four main issues. The first was the level of protection to which AA was entitled (the level of protection issue). The answer turned on whether AA was sufficiently integrated in the UK. The FtT Judge (the FtTJ or the Judge) considered evidence from AA, statements from some character witnesses, and a report on AA prepared by the probation service using the Offender Assessment System known as OASys. The Judge held that the SSHD had erred on this point. He held that AA was entitled to imperative grounds protection because his offending and imprisonment had led to his earlier integration being “weakened but not so broken as to mean that he ceased to be entitled to the highest level of protection”.

The Judge’s reasons for that conclusion included that AA’s offending was “a secret aspect of his life that was not apparent to those around him”, his wider life being “productive and faultless”: [30]. The Judge went on (at [33]) to say this:

“The offending that [was] going on in the background had little active effect on the remainder of his public social and cultural life, even though it is by its nature antisocial. While I do not diminish the harm this type of offending causes, in this appeal it has had a less destructive effect on integration than, for example, would be the case with physically violent conduct in public, gang membership, and so on. It is insidious and conducted behind closed doors, but this also means that normal life continued around it. While at first blush that analysis may seem unpalatable, in my view it is necessary to give effect to the purpose of the present exercise: assessing integration.”

The second issue was the level of threat which AA’s personal conduct represented (the level of threat issue). The FtTJ addressed this question by reference to the OASys report and the protective measures imposed as part of AA’s sentence. The Judge rejected a submission for the SSHD that Restivo showed that such measures should be disregarded for this purpose. He distinguished the case as concerned only with the narrow question of whether incarceration should be disregarded when assessing risk at the point of a deportation decision. The Judge said that what was required was “a ‘real world’ risk assessment” and that he was “entitled to assess the risk posed by the appellant on release in light of the measures to be put in place on release to both manage the risk he poses and facilitate his rehabilitation.” Those measures referred to included the SHPO, the notification requirements, and the licence conditions to which AA would be subject upon his release from custody.

The Judge identified the “high watermark” of the SSHD’s case as the OASys Assessment that he posed a “medium” risk of serious harm to children when in the community. But he held that this had to be considered “in light of the low risk of re-offending and the protective measures provided by the SHPO”, the likelihood that AA would comply with that order, and the existence of “multi-agency protection” against such offending in any event. Applying that approach the Judge held that while AA did pose a genuine and present risk to public security “it cannot be said that this is particularly serious such as to mean that his removal is justified on imperative grounds of public security.”

The third issue was whether AA’s removal would comply with the principle of proportionality. The FtTJ’s decisions on the first two issues made it unnecessary for him to decide that question. He said however that if it had been necessary he would have found that AA’s removal would be disproportionate. The Judge explained that AA’s gender identity issues had little weight in this calculation; he reached his decision mainly because of the effect that removal would have on AA’s “rehabilitation and risk of future offending”. He said that AA’s social rehabilitation in the state in which he had become generally integrated was in the public interest. A balance was to be struck “between AA’s personal conduct and the risk of compromising that rehabilitation”. There was no evidence of the support provisions available in Poland, and (the Judge observed) he should not simply assume that they must be worse. Yet it was “vanishingly unlikely” that AA would face “anything like the same scrutiny and access to assistance” in Poland as he would in the UK. Deportation would increase almost all the risk factors identified in the OASys report.

The fourth issue was whether removal would breach AA’s human rights (the human rights issue). The FtTJ allowed the appeal against the SSHD’s decision on this issue on the basis that the outcome was dictated by the answer to the other grounds of appeal, and no separate analysis was required. At [20] he said that:

“Given that the appellant is a ‘serious offender’ and not entitled to rely on either statutory exception at s 117C of the 2002 Act, he would be required to demonstrate very compelling circumstances such that his removal would be a disproportionate interference with the right to respect for his private and family life afforded by Article 8(1). The regulations already require a proportionality assessment which takes into account the factors specified at reg 27(5) and (6) and which gives effect to Article 7 of the Charter of Fundamental Rights. If the appeal under the regulations is dismissed then the appellant will already have had his Article 8 ECHR proportionality case considered at its highest within the regulations. If the appeal under the regulations is allowed then, applying s 117B(1) of the 2002 Act, the maintenance of effective immigration controls will require that he be permitted to remain in the UK. That will be positively determinative of his human rights appeal.”

The FtT granted the SSHD permission to appeal to the UT on three grounds: that the FtTJ materially misdirected himself in law on the question of AA’s integration; failed to give adequate reasons for his finding that the threat posed by AA was not sufficiently serious to provide imperative grounds of public security for deporting AA; and failed to give adequate reasons for his findings on proportionality, both in respect of the 2016 Regulations and Article 8.

The UT accepted that the SSHD’s concerns about the case were “unsurprising” given that AA had received a substantial prison sentence for “serious and abhorrent criminal behaviour” and yet the FtTJ had ruled that he would not be deported. As I have said, however, the UT concluded that no error of law had been demonstrated.

On the level of protection issue the UT saw no error in the FtTJ’s identification of the relevant legal principles. It said the question for the UT was whether the Judge had failed to apply those principles properly to the facts before him. The Judge had taken account of the relevant factors, conducted the necessary “overall assessment”, and reached an evaluative conclusion. His finding that AA was, “otherwise than with regard to his offending, strongly integrated” was “well within the range of conclusions open to the judge” (emphasis in original). The UT said the nub of this aspect of the SSHD’s appeal was a challenge to the Judge’s handling of the offending itself, and in particular his paragraph [33] (quoted at [29] above). The SSHD’s case was, said the UT, that the approach adopted in that paragraph “radically underplays the profoundly counter-social nature of this offending” and that the Judge’s “failure to give proper weight to this fundamentally important factor led him into public law error.”

The UT rejected these contentions for three main reasons. First, it said the passage in question had to be considered in the context of the judgment as a whole. Secondly, the UT concluded that in that passage “the Judge was making a more limited point about the nature of AA’s offending than the SSHD fears”. The Judge had accepted that the offending was “actively repudiatory of the social and cultural norms underpinning integration”. In the disputed passage he was making the limited, and fair, point

“that while some forms of offending are almost inevitably inconsistent or incompatible with the development of other strong integrative links, some are not, and AA’s fell into the second category.”

On the level of threat issue, the UT summarised the FtTJ’s approach, noting that he had regard to “the measurement of his inherent risk of offending, as set out in the OASys report, and the measures to be put in place on release to facilitate his rehabilitation and manage the continuing risk he did pose.” The UT noted that the SSHD’s appeal on this issue was based on “insufficiency of reasoning”. It rejected that challenge on the basis that the judgment made clear that the Judge had considered the OASys report carefully and in considerable detail and had conducted a legitimate “real world” factual evaluation. The SSHD disagreed with the Judge’s assessment of the threat but that was not the test. The UT was satisfied that the Judge’s decision was “properly approached, and one to which he was properly entitled, and which he sufficiently explained.”

In these circumstances, said the UT, the proportionality issue did not arise. The third ground of appeal became academic since the Decision could not stand in any event.

The four grounds of appeal for which permission was granted reflect the issues raised before the FtT.

These grounds raise some procedural questions. They are all criticisms of the FtTJ. They do not map neatly or precisely onto the issues addressed by the UT in the decision under appeal. AA’s Respondent’s Notice objected to Ground 4 being relied upon in this court, contending that none of the points now made were advanced before the FtT or on appeal to the UT. AA’s Skeleton Argument added that the SSHD’s case on Ground 2 was not put to the UT. In her oral submissions, Ms Hirst broadened the objection to encompass Ground 1, contending that this court should not allow the SSHD to rely on any point that was not raised below.

The scope of the appeal is affected in other ways by AA’s Respondent’s Notice. The main thrust of this is to contend that this is a second appeal from the decision of a specialist fact-finding tribunal which should be respected unless it is quite clear that the tribunal has misdirected itself, which is not the position here. But the Respondent’s Notice also seeks to rely on different or additional grounds for upholding the decisions below.

I propose to consider these aspects of the case as I address the four issues in turn.

Having scrutinised the written grounds of appeal from the FtT, which the FtT gave permission to argue in the UT, I conclude that both the points now taken by the SSHD were sufficiently covered by the grounds. Ground 1(b) on this appeal was squarely before the UT. That is plain from the reasoning of the UT which I have summarised above. Ground 1(a) was put more broadly in the grounds of appeal to the UT. These alleged that the FtTJ had “failed to have regard to” the relevant provisions of the 2016 Regulations, including Schedule 1. Specific attention was not drawn to paragraph 4 of Schedule 1. As Ms Hirst submitted, this point was “not advanced to the UT in the form it is now”. But it was addressed, albeit tangentially, by the UT when it held that the FtTJ had addressed himself correctly to the principles to be applied in conducting the “overall assessment”.

In my judgement, the FtTJ did fall into legal error in his approach to this issue in both the ways relied on by the SSHD and the UT erred in law in upholding his decision.

Paragraph 4 of Schedule 1 is in mandatory terms, requiring “little weight … to be attached” to any integrating links formed “at or around the same time as” any offending or imprisonment. Regulation 27(8) of the 2016 Regulations is also in mandatory terms. It is true, as Ms Hirst points out, that what Regulation 27(8) requires the court or tribunal to do is to “have regard to” the matters listed in Schedule 1. But I cannot accept Ms Hirst’s submission that these provisions are “not binding” on the FtT which is free to decide how to have regard to the specified matters in any given case. That is too broad a view.

The statutory language is similar to that of s 117B(5) of the 2002 Act, quoted above. In Rhuppiah v Home Secretary [2018] UKSC 58, [2018] 1 WLR 5536 [49] Lord Wilson JSC (with whom the other Justices agreed) characterised s 117B(5) as an “instruction” by Parliament to “have regard to the consideration that little weight should be given” to a private life formed by a person when their immigration status was precarious. He concluded that this affords the decision-maker only a small and “limited degree of flexibility”; the statutory provisions provide “generalised normative guidance that may be overridden in an exceptional case by particularly strong features” of the case. I would apply that reasoning to Regulation 27(8) and Schedule 1 paragraph 4 of the 2016 Regulations.

The FtTJ recited those provisions, and this court will always take a benevolent approach to judgments of specialist tribunals, but I think it is clear enough that the Judge did not give effect to these provisions in deciding this issue. That was a material error. In the circumstances of this case the provisions were capable of having real importance. The issue for consideration was the degree of integration achieved by AA in the ten years prior to the Decision in December 2020. For the last 30 months of that period (June 2018 to December 2020) AA was in prison serving a sentence for serious sexual offending. For a period before that he was committing the offences. Although the papers before us do not allow precise identification of that period we do know that it ended with his arrest in February 2016, we know the scale of the offending, we have the sentencing remarks, the OASys report, and the expert evidence provided to the sentencing judge. Having regard to this material I am satisfied that it would be legitimate to infer that AA had a long history of accessing child pornography and that the offending for which he was sentenced in 2018 spanned a period of at least a year before his arrest, that is to say from no later than February 2015. Certainly, it would be fair to conclude that any integrating links formed in the 12-month period from February 2015 onwards were formed “at or around the same time as the commission of a criminal offence”.

It has not been suggested that this was an exceptional case involving particularly strong features that would allow the tribunal to depart from the general normative guidance laid down by Regulation 27(8) and paragraph 4 of Schedule 1. On that footing, if AA formed any integrating links with UK society in the 68 months immediately preceding the Decision the FtTJ was duty bound to attribute “little weight” to any such links formed during 42 of those months. In fact, the FtTJ attributed significant weight to integration during the period when AA was offending, as is clear from his paragraph [33]. In that paragraph the Judge placed reliance on outward manifestations of integration at a time when the offending was “going on in the background”. In paragraph [34] the Judge also attached real and significant weight to AA’s conduct when in prison, saying that he had “engaged as much as he could with the activities on offer.” In my judgement this reasoning was inconsistent with Regulation 27(8) and paragraph 4 of Schedule 1.

The reasoning in paragraph [33] of the FTT judgment is also wrong in principle. I am not persuaded by the UT’s narrow interpretation of this passage. But I do not think the FtTJ’s assessment can withstand scrutiny even on the UT’s analysis. The distinction identified by the UT is not a legitimate one. The reason why offending counts against the proposition that a person is integrated into society is that it shows disregard for fellow citizens and the rejection of core values of UK culture. The reasoning of Hamblen LJ in Binbuga v Secretary of State for the Home Department [2019] EWCA Civ 551[56]-[57] is apposite:

“Social integration refers to the extent to which a foreign criminal has become incorporated within the lawful social structure of the UK … [and] to the acceptance and assumption by the foreign criminal of the culture of the UK, its core values, ideas, customs and social behaviour. That includes acceptance of the principle of the rule of law”.

I would add some words about the intermediate period between AA’s arrest and sentence (February 2016 to June 2018). In my view it is arguable that the Judge should have attached little weight to any integrative links formed during this 28-month period also. This was a time when AA was no longer offending and was at liberty; but throughout this period he was subject to criminal proceedings and stringent bail conditions. More significantly perhaps, it appears that the delay was largely due to the fact that AA was falsely denying his guilt of the offences involving V. His was not a technical, legal defence. AA’s case, rejected by the jury, was that the touching of V and the creation of images of her were all innocent and not sexually motivated. It could perhaps be said that this period, or parts of it, fell at “around the same time as” the offending or the imprisonment so that paragraph 4 of Schedule 1 applied directly. Whether or not that is so, it would seem incongruous to attach more than “little weight” to any integrative links formed under those circumstances. The Judge evidently did so, however, finding (at [34]) that during this period AA “threw himself into therapy and addressing his gender identity and the possibility of imprisonment”.

I can understand why AA objects to pursuit of this ground of appeal. It is certainly fair to say that neither of the points now advanced was directly addressed in the judgment of the UT. I infer that this is because (at the minimum) neither was pressed clearly in oral argument. The mere fact that permission to appeal has been granted does not preclude such an objection: Mullarkey v Broad [2009] EWCA Civ 2 [9]. I would however permit the SSHD to pursue this ground, for three reasons. The first is that again, on a careful reading of the grounds of appeal to the UT for which permission was granted by the FtT, Ground 3(a) falls within their scope. Paragraph 12 of those grounds, although somewhat diffuse, does assert that the FtTJ was wrong to have “regard to measures that will be put into place on the appellant’s release” as reducing “the seriousness of the threat”. Secondly, the UT held that the FtTJ’s decision that the threat was not sufficiently serious was “one which was properly approached, to which he was properly entitled”. If the SSHD’s Ground 3 is sound that was an error of law.

Thirdly, I would permit reliance on this ground of appeal even if, on a proper analysis, it involved new points not taken below. The applicable criteria are well-known and need no repetition. Thy are summarised in Singh v Dass [2019] EWCA Civ 360 and Notting Hill Finance Ltd v Sheikh [2019] EWCA Civ 1337, [2019] 4 WLR 146. The points now raised are pure points of law of general importance. They do not turn on any evidence that was not adduced before the FtT, and there is no reason to believe that AA’s case would have been conducted differently if the points had been more squarely raised below. It is not unfair to him to allow them to be raised now. He has a fair opportunity to answer them. Any financial impact can be adequately addressed by the court’s discretion as to costs.

The passages in Restivo which I have cited above are clearly in point. As the FtTJ noted, that case involved a challenge to a deportation order which the SSHD had made with a view to exploring the possibility of transferring the offender to Italy to serve his life sentence there. The offender challenged the order on the footing that there were legal impediments in Italy to effecting such a transfer. The UT addressed that question, holding that such impediments were irrelevant to the validity of the deportation decision. None of this is relevant here. But the passages I have cited from Restivo address a separate and distinct question, about the right approach to the assessment of threat. That aspect of the decision is directly material to the present case. In Restivo the issue was the relevance of the fact that the person concerned would remain imprisoned for many years after the date of the decision under scrutiny. But that is not a satisfactory ground of distinction. The underlying question of principle is the same: should a sentence or similar measure imposed to prevent or mitigate a threat which an individual’s personal conduct poses to the interests of society be taken into account when assessing the seriousness of that threat?

In my view, Restivo was rightly decided on this point and the reasoning applies to the present case. Measures such as imprisonment, licence conditions on release, SHPOs, and notification requirements are all put in place because a person poses a threat to one of the fundamental interests of society. The existence of such measures is relevant because they involve a recognition of that threat and the need to prevent, manage, or mitigate it. But the preventative or mitigating effects that such measures may have are not themselves material to the question of what level of threat exists. The FtTJ’s “real world” approach in this case therefore involved the same error as was perpetrated by the FtT in Restivo. The seriousness of the threat that AA’s personal conduct represents should have been assessed without regard to the mitigating measures on which the FtTJ placed weight.

The OASys report was clearly a significant item of evidence for this purpose. The FtTJ was justified in placing reliance upon it. He conducted a careful analysis of its content. But I think his reasoning suffered from an error of approach. This was to begin with the OASys assessment that AA posed a “medium” risk of serious harm to children when in the community and then rely on features of the sentencing scheme to which AA was subjected as matters tending to reduce that risk. Mr Pilgerstorfer KC has taken us carefully through the OASys report and persuaded me that this approach did involve the error of “double discounting” alleged by the SSHD and that it fed into the Judge’s overall conclusion on this issue, that the risk could not be said to be “particularly serious”.

The OASys risk assessment was arrived at “on the basis that [the offender] could be released imminently back into the community” ignoring the length of the sentence left to serve. That is an appropriate starting point for the assessment required for the purposes of the 2016 Regulations. It was also an appropriate starting point on the facts of the case. But the OASys Assessment was not carried out for the purposes of the Regulations nor was its methodology in harmony with the principle I have identified in my discussion of Restivo above.

The report makes clear what it means by a “medium risk of serious harm”. It means that there are “identifiable indicators of risk of serious harm”, and that the offender “has the potential to cause serious harm”, but that he is “unlikely to do so unless there is a change in circumstances.” The instructions contained within the report make clear that the likelihood of serious harm is assessed by considering a range of items of information logged earlier in the report. These items include factors likely to increase risk and those likely to reduce it. The former include such matters as “unsupervised and unchallenged access” to the internet and to children. The latter include “compliance with Sex Offender Register, [SHPO] and licence conditions”. The overall risk identified in the OASys assessment was therefore one that had been discounted to take account of the likelihood of compliance with protective or mitigating measures, and the reduction in risk that this would bring about.

It follows from what I have said about Ground 2(a) that if the OASys risk assessment was to be used as a benchmark for the purposes of assessing “threat” under the 2016 Regulations that discount had to be stripped out or ignored. In the passages cited at [29] above, the FtTJ did the opposite, by applying a discount to the OASys assessment. He did so on the basis of factors that had been taken into account in arriving at the OASys assessment. This was double-counting which must necessarily have led to an under-estimate of the seriousness of the threat that AA posed to the fundamental interests of society. This error further vitiates the Judge’s assessment.

For these reasons, I would allow the appeal on this ground.

I add two points, which will merit attention when this case is reconsidered. First, for the reasons given, it may be that the standard against which the seriousness of the threat posed by AA falls to be measured is not the exceptional “imperative grounds” standard applied by the FtTJ but the less demanding “serious grounds” standard. Secondly, it appears to me that the Judge may have misunderstood the OASys report in another respect. The copy which he subjected to analysis was dated 27 January 2020. He said this had been prepared “in anticipation of [AA] being eligible for release into the community” and that it updated an earlier report. I am not sure this is right. We have two copies of the document. They seem to be identical except for the date at the top and bottom of each page. Internal evidence suggests that in each case this is the date on which the document was printed. At all events, each copy records the date of completion and signing as 5 November 2018. The information, or most of it, appears to have been obtained at the time of sentence in June of that year, some 30 months before AA’s release and the Decision, both of which took place in December 2020.

Neither the FtTJ nor the UT made a dispositive decision on this issue, but the FtTJ expressed a view on the matter and it is (at least) very likely that the issue will need to be addressed when this case is reconsidered. It is therefore important that we decide whether the SSHD’s critique of the FtTJ’s reasoning on the point is justified. In my opinion it is.

The first main criticism is that the FtTJ asked and answered the question of whether AA’s removal “would be disproportionate” in the abstract, without addressing the question begged by this formulation, namely “disproportionate to what?” The SSHD contends that the Judge’s focus was on the risk that removal would pose to AA’s prospects of rehabilitation with no, or no proper, regard to the countervailing consideration namely the grounds for removing a person whose conduct posed a serious threat to the fundamental interests of the public here. Secondly, it is said that it is implicit in the Judge’s reasoning that he was apportioning weight to the effect on the public in Poland of receiving an offender whose rehabilitation had been compromised, which is clearly wrong in law. Further, it is said, the Judge’s assessment of the risk to AA’s rehabilitation was arrived at on the basis that his prospects would be worse in Poland than here, which he had rightly recognised as an impermissible line of reasoning.

To some extent the criticism is unfair. The Judge did identify that a balance had to be struck. However, he described it as a balance between “the appellant’s personal conduct” and “the risk of compromising [his] rehabilitation”. That, in my judgement, is an incomplete and materially inaccurate statement of the true position, and the SSHD’s criticisms are otherwise valid.

In EU law the principle of proportionality “involves a consideration of two questions: first, whether the measure in question is suitable or appropriate to achieve the objective pursued; and secondly whether the measure is necessary to achieve that objective, or whether it could be attained by a less onerous method”: R (Lumsdon) v Legal Services Board [2016] AC 697 [33] (Lords Reed and Toulson JSC).Here, there was no doubt that the measure of removing AA to Poland was apt to achieve at least some of the legitimate UK public interest objectives identified in Schedule 1 to the 2016 Regulations. The real issue was whether removal was necessary or excessive, having regard to the countervailing aim of protecting AA’s private life rights under Article 7 of the Charter. (His family life rights were unaffected).

The relevance of AA’s “personal conduct” was that it posed a threat to one of the fundamental interests of society in the UK which, ex hypothesi, was serious enough to justify his removal to Poland on (at least) “serious grounds”. AA’s personal interest in rehabilitation was relevant on the other side of the balance, in the context of his private life rights. But the key question in that context was whether and to what extent his rehabilitation would be compromised by deportation. In my opinion the Judge had no adequate basis for concluding that this would be so. There was, as he acknowledged, no evidence of the support provisions that would be available in Poland. The OASys report identified that as an issue but all that it said was that deportation “may impact on [AA’s] access to offending behaviour programmes”. The report contained no assessment of whether that would in fact be so, still less did it attempt any form of comparison. In my judgement the Judge’s conclusion that AA would not receive similar assistance in Poland was essentially speculative. It follows that he had no basis for finding (as implicitly he did) that there was some UK public interest that counted against removing AA and instead ensuring his rehabilitation within the context of UK society.

The SSHD accepts that the Judge was right to find that removal would represent an interference with AA’s Article 8 rights. The Judge considered it followed from his finding that removal was contrary to the 2016 Regulations that the interference was “incapable of justification”. In one sense he was right. An interference can only be justified under Article 8(2) if it is “in accordance with the law”. If deportation could not be justified under the 2016 Regulations it could not have been justified by reference to s 32 of the 2002 Act either, and it would have had no lawful basis. That would be the end of the human rights argument. But that is not how the Judge approached the matter. It is clear from paragraphs [20] and [55] that he went on to consider the public interest question and concluded that it was answered by the proportionality assessment he had already conducted for the purposes of the 2016 Regulations.

The SSHD’s case is that this was an error of law: the proportionality assessment required by the 2016 Regulations is a matter of EU law that is separate and distinct from the public interest question which arises when considering a human rights claim, the latter being governed by Part 5A of the 2002 Act. Mr Pilgerstorfer argues that s 117B(1) is a red herring; it does not detract from the need for an assessment under s 117C(6). He submits that the Judge failed properly to conduct such an assessment, and that a proper application of s 117C(6) would have led inevitably to the dismissal of the human rights appeal.

I can understand AA’s objection to this line of argument being raised. It did initially appear to be another new point. It is certainly not addressed in the UT’s judgment. The SSHD has however pointed to a contention in paragraph 15 of the Grounds of Appeal to the UT, that the FtTJ erred by failing to follow the approach identified in Badewa v Secretary of State for the Home Department [2015] UKUT 329 (IAC). In Badewa the UT (Thirlwall J and UTJ Storey) held at [24] that:

“… the correct approach to be applied by tribunal judges in relation to ss 117A-D [of the [2002 Act] in the context of EEA removal decisions … is: (i) first to decide if a person satisfies [the] requirements of the Immigration (European Community Area) Regulations 2006. In this context ss 117A-D has no application; (ii) second where a person has raised Article 8 as a ground of appeal ss 117A-D applies.”

As with Ground 1, it seems to me that Ground 4 is sufficiently covered by the written grounds of appeal to the UT, albeit the point was not presented to the UT in quite the way it has been put in this court. I would in any event allow reliance on Ground 4 because this is a pure point of law, the merits of the case will have to be reconsidered in any event, and it would be (to say the least) highly unsatisfactory for this aspect of the matter to be excluded from consideration.

In my judgement, the correct approach is as indicated in Badewa. The application of the 2016 Regulations is a legally distinct exercise from the assessment of a human rights claim. Where both arise, they should be addressed separately and in turn. The 2016 Regulations should be addressed first, including the assessment required by Regulation 27(5)(a) of whether deportation would comply with the EU principle of proportionality. The provisions of Part 5A of the 2002 Act have no part to play at that stage. But they must be addressed as part of the human rights assessment, if the public interest question arises.

The public interest question will not necessarily arise. Although deportation will commonly interfere with Article 8 rights that will not invariably be the case. If it is, the second question arises: whether deportation would be in accordance with the law. That will not be so if deportation would be contrary to the 2016 Regulations. In such a case the human rights analysis need go no further. But if deportation would be consistent with the 2016 Regulations and otherwise lawful the tribunal should address the public interest question in the way that Parliament has prescribed in Part 5A of the 2002 Act. Where, as here, the appellant is a “serious offender” the tribunal will have to apply s 117C(6).

It follows that, given his conclusions on the first two issues, it was unnecessary for the Judge to address the provisions of Part 5A. In doing so, however, he adopted a flawed approach in two respects. First, by placing weight on s 117B(1). On the facts of this case that provision was either neutral or, if relevant, it favoured removal rather than the opposite. Secondly, the Judge erred by treating his conclusion that deportation would breach the EU principle of proportionality as making out the “very compelling circumstances” required by s 117C(6). That was not a legitimate line of reasoning. He should have explicitly asked and answered the question of whether the case featured any and if so what “very compelling” circumstances “over and above” those specified in ss 117C(4) and (5). I would therefore allow the appeal on this ground.

AA is now 40, having come here at the age of 25. He has not spent most of his life here. His wife and child are now in Poland. Against that background, and given what I have said on the issue of rehabilitation, it is easy to see the force of the SSHD’s argument that neither of the exceptions referred to in s 117C(4) and (5) could be said to apply in this case and that the desirability of AA’s rehabilitation proceeding in the UK could not amount to a very compelling circumstance “beyond” those exceptions. But I do not think it would be appropriate for us to address those issues at this stage. The question of how those issues should now be resolved remains for consideration.

For the reasons I have given I would allow the appeal on all four grounds. If the other members of the court agree then I would invite submissions on the question of relief. My provisional view is however that we should set aside the order made by the UT and substitute an order that the appeal be allowed and the case be remitted to the FtT for a fresh decision in accordance with the law as set out by this court.

I agree.

I also agree.