Browne v The Parole Board of England & Wales

Introduction

In this Judgment, I shall refer to the appellant as “Mr Browne” and to the respondent as “the Parole Board”.

This appeal was presented as raising a potentially important issue as to the correct test to be applied on a judicial review of a decision of the Parole Board. On a proper analysis, however, that issue is, at best, subsidiary to the main points in this appeal, particularly given the extremely limited scope of the appeal itself. Furthermore, for the reasons set out below, I consider that – as is all too common in public law cases - the appellant’s case has not been assisted by the production of over-long and contradictory skeleton arguments at every stage of the process, and the desire to pursue every possible point (whether good, bad or indifferent), regardless of the issues raised in the Detailed Statement of Facts and Grounds of Claim (“DSFGC”). This appeal has again demonstrated the need for claimants in the Administrative Court to demonstrate the same procedural and analytical rigour that can be found in other areas of the Civil Justice system.

The Parole Board’s Decision

Mr Browne was born on 6 November 1982. He has a history of serious and violent offending dating back to 2004, when he was convicted of possessing an offensive weapon (a knife) and other offences, for which he was sentenced to imprisonment. The following year, 2005, he was sentenced to imprisonment for possessing ammunition. In 2006 he was convicted of threatening behaviour, and in 2007 he was convicted of attempted robbery and sent to prison for 30 months.

On 15 June 2011 he was sentenced to 6 years and 3 months imprisonment for burglary and assault causing actual bodily harm. On the face of it, this offence was under-charged, since it involved Mr Browne and his two associates, dressed in balaclavas and hoods, attacking the owner of the property, who was a single female. The judge’s sentencing remarks referred to “great planned violence towards the victim…inside her own garden”; the judge said that Mr Browne and his associates had subjected the victim to “intensive violence”, and called the attack “every householder’s worst nightmare”.

Mr Browne was released on licence on 25 March 2014. Condition 5(1) of the licence required him to:

“…be well-behaved, not commit any offence and not do anything which would undermine the purposes of your supervision, which are to protect the public, prevent you from re-offending and help you re-settle successfully into the community.”

On 11 November 2014 Mr Browne missed his appointment with his Probation Officer, saying that he feared that, had he attended, he would have been recalled to prison as a result of an allegation of assault by his ex-partner. Beyond the fact that Mr Browne’s ex-partner had a clump of hair missing, and one of his daughters was said to have been present during the assault, there was little information about this event. Mr Browne’s ex-partner did not wish to press charges. She did however obtain, without notice, a non-molestation order in the Family Court on 11 December 2014, which was served upon Mr Browne on 19 December 2014. The order prevented him from, amongst other things, “communicating with [the ex-partner] whether by letter, text or otherwise except through solicitors”.

Despite the terms of the order, and despite express advice from Mr Browne’s Offender Manager not to have any contact with his ex-partner, he called and texted his ex-partner on 24 December 2014 and telephoned her again on three occasions on 8 January 2015. It was common ground that these events constituted a clear breach of the non-molestation order and, on 12 January 2015, Mr Browne was convicted on two counts of breach. He was sentenced to 56 days imprisonment and subsequently recalled to prison for breach of condition 5(1) of his licence.

There was a hearing before the Parole Board on 27 July 2015 at which Mr Browne gave oral evidence. In a Decision Letter (“DL”) dated 11 August 2015, the Parole Board concluded that the risk he presented was not manageable in the community for the period remaining until his determinate sentence expired. It therefore did not direct his release. The DL is a detailed document, setting out the evidence that was considered; an analysis of the appellant’s past offending; the relevant risk factors; the circumstances leading to the appellant’s recall; and his progress in custody.

Under the heading ‘Panel’s assessment of current risk’, the DL referred to the results of the OASys Assessment commonly used by the Probation Service for the assessment of the risk posed by offenders in England and Wales and went on:

“You have been assessed as presenting a low static risk of re-offending (OGRS), a medium risk of general re-offending (OGP) and a low risk of violent re-offending (OVP). You have also been assessed as presenting a high risk of causing serious harm to a known adult, a medium risk of serious harm to children, members of the public and staff members (presumably because of your offence of assaulting the police officer by running over his foot).

In coming to its own assessment of risk the panel took account of the very serious and violent nature of the index offence of burglary that involved pre-planning and targeted offending. The panel was particularly concerned about the violence that was used against the victim in order to get her to comply. Little is known by your Offender Supervisor or Offender Manager about the extent of the violence and your part in it. In your evidence to the panel you minimised your responsibility for running over the police officer’s foot, describing it as an accident, and that you reacted out of shock, not knowing who the person was. You have previous convictions for offences of violence, albeit that they are not for such serious offences.

The panel had concerns about the allegation of assaulting your ex-partner. Although you have not been convicted of any violence against your ex-partner, nonetheless a non-molestation order was imposed as a result of an incident that occurred in November 2014. You continue to demonstrate significant hostility towards your ex-partner, accusing her of having drugs in the house, of fabricating the complaint and that she threatened you for money. You demonstrate limited insight into the effects of your own behaviour.

You have yet to undertake any accredited thinking skills work or interventions to address your involvement in the use of violence during the index offences of burglary and ABH.

The panel considered that your risk of violence would not have included the violence used during the burglary. As such the panel considered that your OVP underestimated your present risk of future violent offending. Taking all factors into account the panel considered that you continue to present at least a medium risk of violent re-offending with a high risk of causing serious harm to an intimate partner through domestic violence. The panel noted that your risk of serious harm to the public had been reduced to medium whilst you were on licence as a result of the period of 10 months that you spent in the community without further violent offending.”

Thereafter, the DL went on to set out its evaluation of the effectiveness of the plans to manage that risk. As to that, the Parole Board concluded:

“The panel considered that the proposed risk management plan was robust and likely to cover all the relevant areas of risk. However the panel is not satisfied that external risk management proposals would not, on their own, be sufficient to manage the risk. The panel also have significant concerns about your motivation and ability to comply with your licence conditions, particularly if they interfere with your ability to see your daughter. Your behaviour in committing further offences on licence suggest that you may comply superficially with supervision but continue to behave in maladaptive ways to achieve your own objectives.”

Accordingly, having balanced the current risk posed by Mr Browne and the plans to manage that risk, the Parole Board set out its detailed conclusions as to why the appellant’s risk was not manageable in the community for the remaining period of his sentence.

The Application for Judicial Review

The application for Judicial Review was supported by the DSFGC dated 25 October 2015 drafted by Mr Rule. Although it was long (16 pages and 42 paragraphs), it identified just three grounds of challenge. Ground 1 (paragraph 28) was the alleged failure on the part of the Parole Board to apply a presumption in favour of release. Ground 2 (paragraph 30) raised a suggestion of procedural unfairness because it was said that there had been no proper investigation of the allegation of assault on his ex-partner in November 2014. Ground 3 (paragraph 39 and following) was said to be linked in part to the second ground, and concerned “the assessment of risk of serious harm”. Paragraph 40 accepted that “the proper threshold” for the challenge to the assessment of risk was that it was “irrational” but suggested that, in this case, the risk assessment was irrational because it was based on the alleged assault of the ex-partner which had not been properly investigated. Ground 3 also involved an alternative argument that the assessment for suitability for release ought to have been based on the risk management that would immediately follow release, not on the anticipation there might be a later change in circumstances. Nowhere in the DSFGC was there even a suggestion that the DL was not proportionate or that the test to be applied by the reviewing court was proportionality: on the contrary, as noted above, the challenge accepted that the only basis of challenge was on rationality grounds.

In his skeleton argument dated 1 July 2016, produced before the hearing, which ran to 22 pages and 69 paragraphs, Mr Rule again said (paragraph 52) that Ground 3 concerned “the assessment of a high risk of serious harm”. At paragraph 60 of the skeleton, he raised for the first time what he said was “an important issue of law” as to the correct approach for the reviewing court when “determining whether the finding of a high risk of serious harm is or can be justified on the material available”. This new argument, which was not set out in the DSFGC, contradicted what was said there (that the correct test was rationality), and asserted that the court was obliged to test whether the conclusion as to risk was “wrong on the evidence” (paragraph 64). That suggested that the right approach was a full merits review, an argument which Mr Rule expressly reiterated at paragraph 67 of the skeleton argument. He did this by reference to Youssef, an authority discussed below. At paragraph 68, there was the one and only reference to “proportionality”, but that was in a different context. In his oral submissions in reply, Mr Rule accepted that, in relation to the proper and clear setting out of Mr Browne’s case, “things had not been properly done”.

The Judgment

HHJ McKenna (sitting as a judge of the High Court), heard the judicial review application on 20 July 2016. His judgment was dated 31 August 2016 ([2016] EWHC 2178 (Admin)). Having set out the facts and the legal framework, from [32] onwards, the judge went through each of the grounds. He rejected the claim on Ground 1 (the alleged presumption in favour of release) and concluded at [37]:

“To my mind when the DL is considered in the round, it is plain that the Defendant reached a conclusion having considered all the available material, both favourable and adverse to the Claimant, including the dossier, the Offender Supervisor’s report and the oral evidence of the Claimant’s Offender Supervisor at HMP Mount and the Claimant’s Offender Manager and the Claimant himself. Moreover the Claimant has not identified any words in the DL that suggest the Defendant was applying anything other than a presumption in favour of release. It follows in my judgment that ground 1 is not made out.”

As to Ground 2, the procedural fairness challenge, the judge again dismissed the criticisms advanced. His reasoning is summarised at [40]:

“For my part, I do not accept the submission that the Defendant inferred that there was a genuine assault. That is to mischaracterise the Defendant's reasoning. On a fair reading of the DL, it is clear that the Defendant was careful not to make any assumptions about the truthfulness or otherwise of the allegation of assault. Thus, the DL expressly notes that it was an allegation of assault, that the Claimant denied the allegation and that there were few facts available relating to the incident. Moreover, the DL fairly made it clear that the Claimant had no previous convictions for domestic violence and that there was no evidence to suggest that the police had ever attended in respect of any previous domestic incidents between the Claimant and his ex-partner. Although the Claimant suggests that the words "although" and "nonetheless" are indicative that the Defendant proceeded upon the assumption that the allegations were true, for my part I do not read the DL in that way. The words provide no basis for the suggestion that the Defendant assumed the allegation was true. Rather, they indicate a recognition on the Defendant's part that a non-molestation order did not amount to a confirmation of the accuracy of the allegation.”

The judge then widened that out into a consideration of the Parole Board’s overall conclusions as to risk. The relevant paragraph is [43]:

“It is plain from a fair reading of the DL as a whole that the basis of the Defendant's decision on risk, rather than being an inference of guilt in respect of the allegation of assault, was grounded in a large number of elements namely, the violent and premeditated nature of the index offences of burglary and ABH; an escalation in the seriousness of the Claimant's convictions over time; his failure to attend a supervision appointment; the fact of the granting of a non-molestation order and of his pleading guilty to two offences of breaching that non-molestation order; the Claimant's ignoring of the advice of his Offender Manager to have no contact with his ex-partner (notwithstanding the acknowledgement of the explanation given by the Claimant that the contact was for the purposes of arranging payment of money for their daughter); the Claimant's mixed behaviour since returning to custody, including refusing a transfer; the offender assessment system ("OASys") assessment of the Claimant as presenting a high risk of causing serious harm to a known adult and a medium risk of serious harm to children, members of the public and staff members; his presentation at the hearing including his continued demonstration of hostility to his ex-partner, accusing her of having drugs in the house, of fabricating the complaint and threatening him for money and his "limited insights into the effects of his own behaviour"; and a lack of any accredited thinking skills, work or intervention to address his involvement in the use of violence during the index offences of burglary and ABH and the fact that the low risk of violent reoffending assessed by the OASys violence predictor ("OVP") didn't take into account the violence used in the burglary and therefore underestimated that risk.”

The judge dealt with Ground 3 starting at [47]. As to the specific complaint made about the assessment that Mr Browne posed a high risk of serious harm to his ex-partner, he dealt with this shortly at [47] and [48]:

“47.

I can deal with this ground shortly since this ground is linked to the second ground in that it relates to the assessment of risk of serious harm and where it is said that it was irrational for the Defendant, on the basis of the evidence before it, to conclude that the Claimant posed a high risk of serious harm to his ex-partner since there was no evidence that enabled the Defendant properly so to conclude. It was also suggested that the assessment of suitability for release ought to have been based on risk management immediately following release and not on what might happen further down the line.

48.

The first point to make about this ground is that, being a rationality challenge, the Claimant faces a high hurdle given the expert nature of the Defendant tribunal and essentially for the reasons set out in respect of ground 2 the Claimant gets nowhere near reaching the requisite threshold. The decision on assessment of risk was based on a significant number of elements which taken together plainly constituted a rational decision.”

The judge went on to reject at [49] the alternative submission concerning the management of risk in the community. He said:

“49.

So far as the alternative submission is concerned, this ignores the Defendant’s significant concern about the Claimant’s motivation and ability to comply with his licence conditions. That concern is not limited to residence outside of Approved Premises and to my mind therefore there is nothing in that alternative submission either.”

As to the argument that the test to be adopted in considering the risk assessment was somehow lower than or different to rationality, the judge introduced the point at the start of [50] in the following terms:

“Again for the sake of completeness, I should deal with the Claimant's alternative submission put forward for the first time in his skeleton argument that the test to be met in respect of this rationality challenge is in fact lower than Wednesbury unreasonableness. Rather it is a question as to whether the conclusion is seen to be wrong on the evidence with the court conducting a sufficiently rigorous assessment of its own to ascertain the correct conclusion on the basis of that evidence. This submission is based on dicta in recent decisions of the Supreme Court and in particular R (Yousef) v Secretary of State for Foreign and Commonwealth Affairs [2016] 2 WLR 509 and in particular in extracts from the judgment of Lord Carnwath JSC at paragraphs 55 and following:…”

In Youssef, although it was urged upon them, the Supreme Court declined to authorise a general shift away from the rationality test in judicial review cases. In consequence, Judge McKenna’s conclusion on Ground 3 was straightforward. At [51] he said:

“The difficulty for the Claimant with this submission as it seems to me is that the Supreme Court itself has yet to undertake the necessary comprehensive review of the approach to challenging decisions engaging fundamental rights and to my mind, in advance of such a review, it would be premature for this court to proceed down that particular path.”

It should therefore be noted that Judge McKenna thought that Mr Rule’s point about the test to be adopted relied on the alleged requirement for a full merits review, not proportionality (which is not a concept that is mentioned in the judgment). In my view, that simply reflected Mr Rule’s DSFGC and his skeleton argument (see paragraphs 11 and 12 above).

Grounds of Appeal and the Limited Permission Granted

Mr Browne sought to appeal again. The Grounds of Appeal reflected the three grounds in the DSFGC (paragraph 11 above). Thus Ground 1 concerned the alleged presumption in favour of re-release, and Ground 2 concerned procedural fairness. Ground 3 was in these terms:

“(3)

The Judge was wrong to dismiss ground (3). There was an irrational or wrongful assessment by the single-member panel of a high risk of serious harm to the Appellant’s ex-partner rendering it necessary to detain and/or failure to have regard to material evidence and consideration. Further or alternatively, the Judge applied the wrong test to the review of a decision that concerns liberty. The nature of his review is inconsistent with the Supreme Court’s approach for example in R (Youssef) v Secretary of State for Foreign and Commonwealth Affairs [2016] 2 WLR 509. This raises an important issue of public law not yet considered at appellate level in this context of the safeguarding of liberty.” (Emphasis supplied).

It is to be noted that Ground 3 again makes no mention of proportionality. The lengthy skeleton in support of the application for permission to appeal (23 pages and 64 paragraphs) makes passing references to proportionality (see for example paragraphs 38 and 46) but Mr Rule was still determined to run as many different points as he could: thus, at paragraph 38, he asserted that “the judge ought to have gone on to determine the proportionality, or the merits, of the decision evaluated against the factors and evidence properly relevant to the determination” (emphasis added). The skeleton still sought to rely on Youssef, despite the fact that it was not authority for the proposition which Mr Rule advanced.

Floyd LJ refused permission to appeal on Ground 1, Ground 2 and the first part of Ground 3. His reasoning was as follows:

“Ground 1. The judge was right that the failure to mention the presumption in favour of release did not amount to an error of law. The presumption would have been material if the panel had approached the issue of the applicant’s release by asking whether he had discharged a burden of proof which rested on him. The panel did not do so. Instead they considered all the evidence, both for and against release, before concluding that his risk was not manageable. The panel’s decision is therefore not open to criticism on this ground.

Ground 2. As the judge had explained, the panel did not make any finding of assault, and was not obliged to do so. There was other material available on which the panel could properly conclude that there was a risk of re-offending, see judgment paragraph [44]. The panel’s decision does not therefore disclose any error of law on this ground, or involve procedural unfairness.

Ground 3. As the judge said at [48] the decision on assessment of risk was based on a significant number of elements which taken together plainly constituted a rational decision…”

In this way, permission to appeal in respect of each of the criticisms which had been made in the DSFGC had been rejected. However, on the point in bold at paragraph 19 above, which was not set out in the DSFGC, which had not been clearly advanced before HHJ McKenna, and which was very unclear even in the skeleton seeking permission to appeal, Floyd LJ said:

“However it is arguable that a proportionality as opposed to a rationality approach should be adopted, and it is not possible to say that the panel would necessarily have reached the same conclusion had they applied a proportionality test.”

The confusion created by Mr Rule's unfocussed approach is evident from Floyd LJ’s reference to the test applied by the Parole Board. No-one had ever suggested that the Parole Board had ever adopted anything other than a proportionate approach to their decision; the issue in the second part of Ground 3 (such as it was) purported to go to the test to be applied by the reviewing court.

In the result, therefore, the only point on which Mr Browne has been granted permission to appeal was in respect of an argument which was raised for the first time (and even then obliquely) in Mr Rule’s skeleton argument to this court, and in circumstances where that permission was itself based on an apparent misunderstanding engendered by the confused arguments to which I have already referred. On every other point, Mr Browne has been unsuccessful.

However, his procedural difficulties do not end there. During the hearing before this court, a major dispute arose between counsel as to the precise nature of the challenge being mounted under the second part of Ground 3. Mr Rule said that this was an attack on the decision of the Parole Board; Mr Cross maintained, by reference to the documents to which I have referred at Sections 3, 4 and 5 above, that it was directed solely at the assessment that Mr Browne posed a high risk of causing serious harm. In my view, Mr Cross was right about that: Ground 3 is and always was a specific attack on the assessment of Mr Browne as posing a high risk of serious harm. That is how it was presented originally (see in particular the italicised passages in paragraph 12 above); that is how it was dealt with by HHJ Mckenna (paragraphs 15-18 above); and that is how it is described in the grounds of appeal (paragraph 19 above). The odd reference to the “decision” in dense paragraphs of overlong skeleton arguments did not somehow turn the specific attack on the risk assessment into a wider criticism of the DL itself.

Where then does that leave this appeal? In my view, since I have reached the conclusion that each of the submissions put forward by Mr Rule is equally hopeless, it is appropriate to deal with all of them, notwithstanding my views as to what is properly open to him and what is not. Accordingly, I start at Section 6 with a note of all the various matters which cannot be challenged on this appeal. At Section 7, I address the criticisms of the risk assessment. Thereafter, I address the test to be applied by a court reviewing a decision of the Parole Board (Section 8) and then apply that test to the DL itself (Section 9).

The Confines of The Appeal

As a result of the findings in HHJ McKenna's judgment; the matters within the DL which have never been criticised; and the refusal of permission to appeal on Grounds 1, 2 and the first part of Ground 3; the following features of the DL must be regarded as unchallengeable in this court.

First, in the DL, the Parole Board applied the correct legal test and correctly asked the appropriate question, namely whether it was necessary for the protection of the public that the appellant should continue to be detained. The DL properly recognised a balancing exercise between the rights of the appellant to liberty and the protection of the public from the risks that he posed. To the extent that there is a “presumption” in favour of liberty, it was properly made: see HHJ McKenna at paragraph 13 above; Floyd LJ at paragraph 21 above.

I should say, by way of completeness, that the respondent does not necessarily accept that, as a matter of law, there is a presumption in favour of release in the terms outlined by the appellant in the grounds of appeal. Since the point does not arise on this appeal it is unnecessary to decide the point, but I understand Mr Cross’ submission to be that the alleged presumption is based on the decision in R (Sim) v Parole Board [2003] EWCA Civ 1845, which was concerned with the recall of a prisoner on an extended sentence. In a case like this, which is concerned with a determinate sentence only, the test is that in R (King) v Parole Board [2016] EWCA Civ 51, namely that a prisoner must not be released “unless the secretary of State is satisfied that it is not necessary for the protection of the public that P should remain in prison”.

Secondly, when assessing the risk posed by Mr Browne, the Parole Board was entitled to and did take into account his history of offending; the particular circumstances of the offence which led to the lengthy prison term (the violent premeditated burglary); his admitted breaches of the non-molestation order; and all the circumstances of his recall. In addition, the Parole Board properly took into account the OASys assessment of risk, which assessed Mr Browne as presenting a high or medium risk of violent re-offending: see paragraphs 9-10 above and HHJ McKenna at paragraph 14 above.

Thirdly, the Parole Board did not make any findings of fact, let alone any findings of guilt in respect of the assault allegation. There was no procedural unfairness of any kind: see HHJ McKenna at paragraph 14 above; Floyd LJ at paragraph 21 above.

Fourthly, the Parole Board properly considered whether the risk posed by Mr Browne was capable of being managed in the community and reasonably and rationally concluded that it could not be: see HHJ McKenna at paragraph 16 above; Floyd LJ at paragraph 21 above.

As a result of these matters, it cannot be said that the respondent reached an irrational decision. On the contrary, both HHJ McKenna at [48] (paragraph 15 above) and Floyd LJ (paragraph 21 above) concluded that the Parole Board’s decision not to release Mr Browne constituted a rational decision.

The Criticisms of the Risk Assessment

Was the Risk Assessment ‘wrong’?

Throughout his various skeleton arguments, Mr Rule repeatedly suggested that the risk assessment was “wrong” and was therefore susceptible to judicial review on this basis. As noted in paragraphs 12, 18 and 20 above, this appeared to be linked to the suggestion in those same skeleton arguments that the reviewing court should undertake a full merits review of the Parole Board’s decision.

In my view, this criticism is hopeless. In order for a reviewing court to decide that a multi-factorial risk assessment was “wrong”, it would need to break down that risk assessment into its component parts and redo the exercise from scratch. There is no authority that suggests that a reviewing court is required to carry out a “full merits review” of the decision of the Parole Board. That would mean that the reviewing court was exercising not just an appellate jurisdiction, but would be exercising precisely the same fact-finding powers as the original tribunal. Such a proposition is misconceived.

Mr Rule eventually accepted, in his oral submissions on the first day of the appeal, that he could not put his case that high (despite the fact that that is precisely how he had put the case previously). I do not accept his submission that this change of case was clear at any time prior to those oral submissions.

Was the Risk Assessment rational?

The risk assessment was plainly rational: both HHJ McKenna and Floyd LJ have said so. Moreover, that conclusion is plain from the face of the DL itself. I have set out the relevant sections at paragraphs 9-10 above. The assessment was based on a number of things including the serious nature of the index offence, and in particular the violence that was used against the victim; the appellant’s attempts to minimise his responsibility; his previous convictions for violence; the existence of the non-molestation order and the appellant’s significant hostility towards his ex-partner; his breaches of that order; his limited insight into the effects of his offending; and the fact that he had yet to undertake any accredited thinking skills work or other related rehabilitation.

Most important of all, there was the OASys assessment which identified the appellant as posing a medium risk of general re-offending; a high risk of causing serious harm to a known adult (his ex-partner); and a medium risk of causing serious harm to children and other members of the public. That is a significant and independent piece of evidence which makes plain the level of risk that the appellant posed, in particular to his ex-partner, but also to the wider public. On one view, it might be said that the result of that independent assessment alone justified the Parole Board’s conclusions as to risk. But certainly, when taken together with all of the other material, it is impossible to say that the assessment as to risk was anything other than rational.

Was the Risk Assessment disproportionate?

I do not accept that proportionality is the relevant test: see Section 8 of this Judgment below. However, even if it were the appropriate way to scrutinise the DL, it is a singularly inapt test when applied to the assessment that Mr Browne posed a high risk of serious harm to his ex-partner.

Mr Rule summarised the authorities on proportionality to identify the four factors which are generally considered when considering this question (Footnote: ¹). They can be summarised as:

In answer to questions from Lord Justice Singh during the course of the oral argument, Mr Rule accepted that (a) and (b) did not arise in respect of the risk assessment in this case. That leaves (c) and (d).

The principal problem with this part of the case is that these questions are simply inapplicable to a risk assessment. The questions are concerned with a decision: they are a way of testing whether the end justifies the means. They are not applicable to one particular part of the evidence relied on in reaching that decision. An assessment of high risk of causing serious harm cannot be ‘necessary to accomplish an aim’ or ‘strike a fair balance’. Moreover, such an assessment is in many ways a prediction of the future. In my view, it is impossible to direct these broad proportionality questions to a multi-factorial and predictive assessment of risk of the kind routinely undertaken by the Parole Board.

Some support for this conclusion can be found in the judgment of Lord Sumption in R (Lord Carlile of Berriew) v SSHD [2015] AC 945where, at [32] he said:

“Nonetheless, there are cases where the rationality of a decision is the only criterion which is capable of judicial assessment. This is particularly likely to be true of predictive and other judgmental assessments, especially those of political nature. Such cases often involve a judgment or prediction of a kind whose rationality can be assessed but whose correctness cannot in the nature of things be tested empirically…”

And went on to at [33] to cite Lord Bingham of Cornhill in an earlier case:

“any prediction about the future behaviour of human beings (as opposed to the phases of the moon or high water at London Bridge) is necessarily problematical. Reasonable and informed minds may differ, and a judgment is not shown to be wrong or unreasonable because that which is thought likely to happen does not happen. It would have been irresponsible not to err, if at all, on the side of safety.”

Summary

Accordingly, for the reasons that I have given, there can be no sensible or justified attack on the risk assessment undertaken by the respondent. It cannot be said to be wrong. It was plainly rational. The proportionality or otherwise of the assessment that Mr Browne posed a high risk of serious harm to his ex-partner simply cannot arise.

On my primary view of this case, that is the end of the appeal. But if I am wrong in my analysis of the second part of Ground 3, and this appeal instead raises the proportionality of the DL itself, I go on to address that criticism too. I do so by setting out the law, and then considering the proportionality of the DL.

The Common Law Test for Judicial Review

The Decisions of the Parole Board

The law relating to the approach of a reviewing court to the decisions of the Parole Board is well-settled. I set out below some of what seem to me to be the important authorities.

In R v Parole Board ex parte Watson [1996] 1 WLR 906, Sir Thomas Bingham said at page 916H – 917A:

“In exercising its practical judgment the board is bound to approach its task under the two sections in the same way, balancing the hardship and injustice of continuing to imprison a man who is unlikely to cause serious injury to the public against the need to protect the public against a man who is not unlikely to cause such injury. In other than a clear case this is bound to be a difficult and very anxious judgment. But in the final balance the board is bound to give preponderant weight to the need to protect innocent members of the public against any significant risk of serious injury. This is the test which section 34(4) (b) prescribes, and I think it is equally appropriate under section 39(4).”

In R (Alvey) v Parole Board [2008] EWHC 311 (Admin) at [26], Stanley Burnton J, as he then was, said:

“The law relating to judicial review of this kind may be shortly stated. It is not for this court to substitute its own decision, however strong its view, for that of the Parole Board. It is for the Parole Board, not for the court, to weigh the various considerations it must take into account in deciding whether or not early release is appropriate. The weight it gives to relevant considerations is a matter for the Board, as is, in particular, its assessment of risk, that is to say the risk of re-offending and the risk of harm to the public if an offender is released early, and the extent to which that risk outweighs benefits which otherwise may result from early release, such as a long period of support in the community, and in some cases damage and pressures caused by a custodial environment.”

In R (Brooke) v Parole Board [2008] 1 WLR 1950 at [53], Lord Phillips CJ said:

“53.

Judging whether it is necessary for the protection of the public that a prisoner be confined is often no easy matter. The test is not black and white. It does not require that a prisoner be detained until the Board is satisfied that there is no risk that he will re-offend. What is necessary for the protection of the public is that the risk of re-offending is at a level that does not outweigh the hardship of keeping a prisoner detained after he has served the term commensurate with his fault. Deciding whether this is the case is the Board's judicial function. We turn to consider the extent to which the independence of the Board in performing this function has been put in doubt.”

In R (James) v Secretary of State for Justice(Parole Board intervening) [2010] 1 AC 553, Lord Judge CJ stated at [134]:

“In expressing myself in this way, I am not to be taken to being encouraging applications by prisoners for judicial review on the basis that the prisoner may somehow direct the process by which the Parole Board should decide to approach its section 28(6) responsibilities either generally, or in any individual case. These are question pre-eminently for the Parole Board itself. Although possessed of an ultimate supervisory jurisdiction to ensure that the Parole Board complies with its duties, the Administrative Court cannot be invited to second-guess the decisions of the Parole Board, or the way it chooses to exercise its responsibilities. Your Lordships were told that the Board is frequently threatened with article 5(4) challenges unless it requires the Secretary of State to provide additional material. Yet it can only be in an extreme case that the Administrative Court would be justified in interfering with the decisions of what, for present purposes, is the "court" vested with the decision whether to direct release, and therefore exclusively responsible for the procedures by which it will arrive at its decision.”

Some of these passages were set out in the recent judgment of the Divisional Court in R (DSD & MVB) v Parole Board of England & Wales [2018] EWHC 694. In his judgment, the President of the Queen’s Bench Division, Sir Brian Leveson, stressed that “the evaluation of risk, central to the Parole Board’s judicial function, is in part inquisitorial” [117]. He said at [118] that the courts have emphasised on numerous occasions the importance and complexity of the role performed by the Parole Board and how slow the courts should be to interfere with the exercise of judgment “in this specialist domain”. He stressed that the reviewing court was not an appellate jurisdiction at [133]:

“A risk assessment in a complex case such as this is multi-factorial, multi-dimensional and at the end of the day quintessentially a matter of judgment for the panel itself. This panel's reasons were detailed and comprehensive. We are not operating in an appellate jurisdiction and the decision is not ours to make. We are compelled to conclude that the decision of the panel must be respected. It follows that the irrationality challenge, in the terms in which it was advanced by Ms Kaufmann (adopting also those made by Mr Squires), cannot be upheld.”

The test applied by the Divisional Court in DSD, and in all the other authorities noted above, is whether the decision of the Parole Board could be said to be irrational in accordance with the classic test set out in Associated Provincial Picture Houses v Wednesbury Cooperation [1948] 1 KB 223 at [229]. For the reasons set out by Sir Brian Leveson in DSD, despite the fact that the Divisional Court considered that the Parole Board’s decision about Worboys was “surprising and concerning”, they could not say that it was irrational.

At no time has it been suggested, in any of these authorities relating to the decisions of the Parole Board, that the test for the reviewing court was anything other than irrationality. The only minor modification to this approach identified in the authorities is to be found in certain first instance decisions such as R (Bayliss) v The Parole Board [2008] EWHC 3127 (Admin) and R (Salter) v Secretary of State for Justice [2009] EWHC 1497 (Admin), where the court applied what is sometimes called an enhanced rationality test – that is to say they considered the challenge with anxious scrutiny – because the liberty of the claimant was at stake. Speaking for myself, that seems to me to be a sensible and appropriate development, particularly given that, in judicial review cases, context is everything (Footnote: ²).

However, other than this minor modification, I can see no basis for this court to depart from the conventional approach to the review of Parole Board decisions. The relatively high threshold of irrationality is appropriate when the Administrative Court is reviewing the decisions of the Parole Board. It properly reflects the Parole Board’s judicial function, its inquisitorial role, its specialist expertise, and the important and complex role that it performs.

Mr Rule argued that the common law has now adopted a different test for judicial review that goes beyond rationality in cases concerning fundamental rights. I do not agree. On the contrary, there is the highest authority for the proposition that the basic test to be applied in domestic judicial review cases which involve neither EU law nor Convention rights remains that of rationality. Again the only modification is that, in cases involving fundamental rights, the reviewing court is required to consider the rationality of the original decision with “the most anxious scrutiny”. Authority for these propositions can be found in the following:

These authorities make plain that proportionality is not a part of English domestic law (that is to say, in cases which do not involve EU law or Convention rights). Whether such a change should be sanctioned was expressly considered by the Court of Appeal in R (Association of British Civilian Internees: Far East Region) v Secretary of State for Defence [2003] EWCA Civ 473. In his judgment in that case, Dyson LJ (as he then was) identified at paragraph 34 the movement towards the recognition of proportionality and the end of the rationality test. But he then went on at paragraph 35 to say:

“But we consider that it is not for this court to perform its burial rites. The continuing existence of the Wednesbury test has been acknowledged by the House of Lords on more than one occasion. The obvious starting point is R v Secretary of State for the Home Department ex p Brind [1991] 1 AC 696. The Home Secretary had issued directives to the BBC and IBA prohibiting the broadcasting of speech by representatives of proscribed terrorist organisations. The applicant journalists challenged the legality of the directives on the ground that they were incompatible with the ECHR, and also on the ground that they were disproportionate in a sense going beyond the established doctrine of reasonableness. Mr Pannick submits that Brind does not stand in the way of this court holding that proportionality has supplanted the Wednesbury test in English domestic law, even where no human right or Community Law issues are raised. We do not agree. It is true, as Mr Pannick points out, that Lord Bridge and Lord Roskill left the door open for the possible future introduction and development of the doctrine of proportionality into English domestic law. But all of their Lordships rejected the proportionality test in that case, and applied the traditional Wednesbury test. In other words, they closed the door to proportionality in domestic law for the time being.”

In two more recent cases, the Supreme Court have also considered this issue and expressly declined to decide it. Thus, in Pham v Home Secretary [2015] 1 WLR 1591 the question of proportionality was raised by issues (ii) and (iii), as identified by Lord Carnwath at paragraph 31 of his judgment. But he then expressly declined to deal with the issue, a reticence shared by the other judges of the Supreme Court. Lord Reed expressly said at paragraph 115 of his judgment that Pham was not the occasion to review Brind and Smith. More recently still, in R (Youssef) v Secretary of State for Foreign and Commonwealth Affairs [2016] UKSC 3, the court again declined to address the question, in part because, as Lord Carnwath said at paragraph 59, the appellant had failed to highlight any particular aspect of the original reasoning which was open to challenge “even applying a proportionality test”.

In my view, the current position is summarised by Lord Neuberger in R (Keyu & Others) v Secretary of State for Foreign and Commonwealth Affairs [2015] UKSC 69. There, Lord Neuberger identified the issue and again declined to address it:

“131.

The appellants raise the argument that the time has come to reconsider the basis on which the courts review decisions of the executive, and in particular that the traditional Wednesbury rationality basis for challenging executive decisions should be replaced by a more structured and principled challenge based on proportionality. The possibility of such a change was judicially canvassed for the first time in this jurisdiction by Lord Diplock in Council of Civil Service Unions v Minister for the Civil Service [1985] AC 374, 410E, and it has been mentioned by various judges in a number of subsequent cases – often with some enthusiasm, for instance by Lord Slynn in R (Alconbury Developments Ltd) v Secretary of State for the Environment, Transport and the Regions [2001] UKHL 23, [2003] 2 AC 295, para 51. In other words, the appellants contend that the four-stage test identified by Lord Sumption and Lord Reed in Bank Mellat v HM Treasury (No 2) [2013] UKSC 39, [2014] AC 700, paras 20 and 74 should now be applied in place of rationality in all domestic judicial review cases.

132.

It would not be appropriate for a five-Justice panel of this court to accept, or indeed to reject, this argument, which potentially has implications which are profound in constitutional terms and very wide in applicable scope. Accordingly, if a proportionality challenge to the refusal to hold an inquiry would succeed, then it would be necessary to have this appeal (or at any rate this aspect of this appeal) re-argued before a panel of nine Justices. However, in my opinion, such a course is unnecessary because I consider that the appellants' third line of appeal would fail even if it was and could be based on proportionality.

133.

The move from rationality to proportionality, as urged by the appellants, would appear to have potentially profound and far-reaching consequences, because it would involve the court considering the merits of the decision at issue: in particular, it would require the courts to consider the balance which the decision-maker has struck between competing interests (often a public interest against a private interest) and the weight to be accorded to each such interest – see R (Daly) v Secretary of State for the Home Department [2001] UKHL 26, [2001] 2 AC 532, para 27, per Lord Steyn. However, it is important to emphasise that it is no part of the appellants' case that the court would thereby displace the relevant member of the executive as the primary decision-maker – as to which see per Lord Sumption and Lord Reed in Bank Mellat (No 2) at paras 21 and 71 respectively. Furthermore, as the passages cited by Lord Kerr from Kennedy v Charity Commission (Secretary of State for Justice intervening) [2014] UKSC 20, [2015] AC 455, paras 51 and 54, and Pham v Secretary of State for the Home Department (Open Society Justice Initiative intervening) [2015] UKSC 19, [2015] 1 WLR 1591, paras 96, 113 and 115 show, the domestic law may already be moving away to some extent from the irrationality test in some cases.”

Accordingly, these authorities spell out the simple proposition that, for now at any rate, the common law test for judicial review is based on the underlying principle of rationality. Whilst there is some support for adopting a proportionality test in particular cases concerned with fundamental rights (see for example Kennedy), there is a recognition that a more widespread change would require a major review by the Supreme Court and the necessary overruling of Brind and Smith.

Accordingly, whether the review of the law is limited to the authorities concerned with challenges to the Parole Board, or is widened to a consideration of the domestic common law authorities, the result is the same: there is no basis in law for applying anything other than a rationality test when reviewing the decisions of the Parole Board. Since both HHJ Mckenna and Floyd LJ have ruled that the DL of 11 August 2015 was rational, that is again the end of the appeal.

The Criticisms of the Decision

Introduction

For the reasons set out at paragraph 24 above, I am firmly of the view that the limited ground for permission to appeal in respect of Ground 3 related solely to the risk assessment. For the reasons set out at paragraph 59 above, I am also firmly of the view that the test to be applied by any court reviewing a decision of the Parole Board is one of rationality. However, if I am wrong on both these issues, and the second part of Ground 3 is an appropriate attack on the proportionality of the DL itself, I am in no doubt that the DL was proportionate.

Striking a Fair Balance

The DL considers, on the one hand, the appellant’s right to liberty, and on the other, the necessary protection of the public. That was the balance that needed to be struck. Moreover, for the reasons summarised by HHJ McKenna at [43] (paragraph 14 above), and explained at paragraphs 29 and 36-37 above, I consider that this balance was properly and fairly struck, by an expert tribunal carefully considering all the relevant issues.

Because Mr Rule’s written and oral arguments were primarily concerned with the law, and very little concerned with the facts of this case, he was unable to articulate clearly how or why he said that the DL did not strike a fair balance. One suggestion he made was that the Parole Board had somehow misunderstood the background to the alleged assault on the ex-partner, a point on which he did not obtain permission to appeal, which was comprehensively rejected by HHJ McKenna, and which would go to the merits and not proportionality anyway. On being pressed during his oral submissions, Mr Rule’s only other suggestion was that the DL failed to have proper regard to the alternative of managing the risk in the community. That gives rise to the separate issue as to whether the decision “is no more than is necessary to accomplish the aim”.

No More than is Necessary

The first difficulty for Mr Browne on this issue, which in my view is insurmountable, is that it was separately argued before HHJ McKenna that the DL had failed to take into account the question of risk management in the community. The judge expressly rejected that submission at [49] (paragraph 16 above). Mr Browne did not seek and therefore does not have permission to challenge that conclusion. That is the end of the point.

Moreover, the complaint based on risk management immediately following release ignores the Parole Board’s significant concerns about Mr Browne’s motivation and ability to comply with his licence conditions. This was the very point made by HHJ McKenna at [49]: as he went on to note, that concern was not limited to residence outside of approved premises, but was much more widely based.

Thirdly, this attack ignores the fact that the Parole Board heard oral evidence from Mr Browne. Unlike HHJ McKenna, and unlike this court, the Board was therefore in the best possible position to assess the appellant and his motivation. At least part of that assessment would have gone to the question of whether, if the appellant was released, the management of his risk in the community was likely to be successful. The Parole Board reached a negative view on that issue. It is impossible on judicial review for a court, who did not have the benefit of that evidence, to come to a different conclusion.

Finally, Mr Rule’s difficulty in advancing any coherent criticism of the DL was highlighted by his admission, made during his submissions in reply, that in respect of the relevant passages in the DL dealing with the measures to govern risk, he could not say “that anything was right or wrong”. In my view, no sustainable criticism was or could be made.

Would the Result be Different?

Mr Rule accepted early on in his submissions that he needed to show that, if proportionality was the right test, it would lead to a different result. This is important. A number of the authorities (such as R (Alconbury Developments Ltd v SoS for the Environment [2003] 2 AC 295, and R (Daly) v SSHD [2001] 2 AC 532) stress that there is often no great difference in the outcome whether the test applied is one of rationality or one of proportionality. One of the reasons why the Supreme Court did not wish to make the shift from one to the other in Youssef was because, which ever test was applied, the result was the same.

In my view, the same is true here. The Parole Board’s careful balancing between the competing interests of Mr Browne on the one hand and the public (and his ex-partner in particular), on the other, demonstrates beyond peradventure that their decision was proportionate.

Summary

Accordingly, for the reasons set out above, I consider that if, contrary to my primary view, the limited permission to appeal allowed the appellant to attack the DL itself and if, also contrary to my view, such an attack is permissible in law, I conclude that the Parole Board’s decision of 11 August 2015 was proportionate.

For all these reasons, therefore, I would dismiss this appeal.

I agree.

Save that I would prefer to express no view as to whether proportionality is or is not the relevant test, I agree that, for the reasons given by Coulson LJ, this appeal should be dismissed.