IN THE HIGH COURT OF JUSTICE
ADMINISTRATIVE COURT
Sitting at:
Leeds Combined Court
1 Oxford Row
Leeds
West Yorkshire
LS1 3BG
Before:
HIS HONOUR JUDGE KAYE QC
Between:
ABEL OGE-DENGBE | Claimant |
- and - | |
SECRETARY OF STATE FOR JUSTICE | Defendant |
(DAR Transcript of
WordWave International Limited
A Merrill Communications Company
101 Finsbury Pavement London EC2A 1ER
Tel No: 020 7422 6131 Fax No: 020 7422 6134
Official Shorthand Writers to the Court)
Mr Stanbury (instructed by Grayson Willis Bennett Solicitors) appeared on behalf of the Claimant
Mr Murray (instructed by the Treasury Solicitor) appeared on behalf of the Defendant
Judgment
HHJ KAYE QC:
The claimant is a 52-year-old national of Sierra Leone and was at the material time a prisoner at HMP Wellingborough. He had the advantage of refugee status. He challenges the decision of the Governor of the prison dated 29 April 2010 refusing to reclassify him as a category D prisoner. Although proceedings were issued also against the Governor of HMP Wellingborough, the effective defendant is the Secretary of State for Justice.
Originally this challenge was mounted on two grounds:
First, that the provisions of the relevant Prison Service Order (“PSO”), PSO 4630, were ultra vires rule 7 of the Prison Rules 1999 in that it promotes consideration of the deportation process as paramount rather than the requirements of maintaining good order and facilitating training.
Second, the decision was based solely on the claimant’s deportation status and failed to take account of other relevant policies and is accordingly irrational and unreasonable.
Permission to make the application was granted by Judge Langan QC (sitting as a Judge of the High Court) on 7 September 2010 on paper. He considered that it was arguable based on the papers before him that the Governor’s decision not to reclassify the claimant to category D status was based solely and not just mainly on his immigration status and was therefore unlawful. He directed the claimant’s skeleton argument to be lodged not less than 21 days before the hearing and the defendants’ skeleton arguments not less than 14 days before the hearing. This, then, is the substantive hearing.
On 2 November 2010 an application to add a third ground to include a challenge that PSO 4630 was incompatible with the Race Relations Act 1976 was refused by Langstaff J but with a caveat that the application might be renewed at the hearing. At the same time, however, it was conceded (in my judgment rightly, having regard to R (Omoregbee) v SSJ [2010] EWHC 2658 (Admin)) that the first ground, the ultra vires ground, was no longer arguable, leaving the second.
The claimant’s counsel, in his skeleton argument lodged on 16 January 2011, albeit not in conformity with the directions of Judge Langan, also refers to a proposed application to add a fourth ground, namely that the claimant ought to have had a six-monthly review. This was on the basis that, as a person within the last 30 months of his sentence, the claimant was entitled to a categorisation review every 6 months in accordance with Prison Service Instruction (“PSI”) 16/2008. On this basis the claimant was entitled to a further categorisation review on or about 20 October 2010 but it has not taken place.
Mr Stanbury, who has appeared for the claimant, however, has quite properly made it clear in his skeleton argument he was not pursuing the amendment refused by Langstaff J. As to his other proposal, this was more in the nature of drawing a complaint to the attention of the defendant, who was in turn, no doubt, awaiting the outcome of this present claim.
The background is as follows.
The claimant is, as stated, a citizen of Sierra Leone. In 2002 he was granted asylum and refugee status. He appears to have a wife in the United Kingdom. This may be in some doubt, having regard to matters I refer to later.
The claimant was sentenced on 27 July 2006 to 11 years imprisonment at Manchester Crown Court for conspiracy to import class A drugs. His conditional release date is 22 March 2011 and his sentence expiry date is 22 September 2016. On 17 June 2009 he was notified of his liability to deportation under section 32(7) of the UK Borders Act 2007 (“the 2007 Act”).
For some time the claimant has been applying for re-categorisation to category D status. This was prolonged owing to the prison authorities seeking clarification of his refugee status from the UK Border Agency since at least prior to 21 September 2009, when his reclassification review was deferred pending receipt of the clarification. The prison authorities were aware of his concerns because his then solicitors wrote complaining of the deferment. Following the receipt on 21 April 2010 of notification by the UK Border Agency of its intentions in this respect, on 29 April 2010 the decision was made to keep him at HMP Wellingborough with his prison categorisation at its present level, category C, on the basis that deportation would “manifestly increase the risk that you would comply with open conditions”. This was notwithstanding, as the defendant concedes, that the previous view of the review board convened on 20 April 2010, the day before the receipt of the notification on 21 April 2010, was to approve his application noting his “very positive aspects” and (in substance) that his exemplary custodial behaviour, course work undertaken to reduce risk and the fact that he had refugee status therefore presented as a low risk of absconding. Indeed, the defendant confirmed in his Grounds of Defence that it was the receipt on 21 April 2010, the following day, of the UK Border Agency’s indication that the claimant’s refugee status was under consideration that led to the review board being quickly reconvened, resulting in the change of mind that he now, in light of that, gave rise to a “significantly increased risk of absconding”. According to the re-categorisation review form completed by the prison authorities, this seems to have been the only reason given for revision of the previous view. Indeed, the reason stated was as follows:
“Recent communication from the UK Border Agency indicates that they are actively considering deportation in your case. This would significantly increase the risk that you would not comply with open conditions.”
On 28 July 2010 the claimant was notified by the UK Border Agency of its intention to take away or terminate his refugee status on the basis of a change of circumstances in his country of origin, Sierra Leone.
It appears from the letter notifying him that the claimant was granted asylum on the basis that he was claiming that his father was the former leader of the SLPP in Sierra Leone, which is now the party of opposition (though his status as the son of an alleged leader of the SLPP I am told is now a matter of dispute). The UN High Commissioner for Refugees’ report of June 2008 recommended that those who fled Sierra Leone should cease to enjoy refugee status and it appears to be this which may have prompted the UK Border Agency’s decision.
On 17 August 2010 the United Kingdom formally informed the UN High Commissioner for Refugees of intention to withdraw the claimant’s refugee status. As Mr Stanbury correctly points out, the indication that the UK Border Agency is considering terminating his status is just that. Until the UK Border Agency formally revoked his status he continued to have that status. In his response, the claimant indicated that he fears for his life if returned to Sierra Leone and will fight deportation.
This has important consequences. Under section 32(4) of the 2007 Act, the claimant is liable to automatic deportation on completion of his sentence. By section 32(2)(a), however, an exception exists where removal would breach the individual’s Convention rights. Under section 32(2) (b) a further exception exists where removal would breach the UK’s obligations under the Refugee Convention. In the first case (“exception (a)”) the burden is on the individual to show he qualifies for the exception. To do that he must apply to the First Tier Tribunal. This is the reason why a notification of deportation carries with it also a notification of a right of appeal. It appears no such application has yet been made, presumably on the ground that until very recently, as will be seen, no formal decision had as yet been made.
In the second case (“exception (b)”) the burden rests on the state to demonstrate removal would not breach its obligations. Mr Stanbury submits that the removal of a person with refugee status would fall within exception (b). Hence, unless and until the United Kingdom formally terminates the claimant’s refugee status, he falls, it is argued, potentially within exception (b). This is an interesting point but not the one that falls before me, even though the background and the statutory context is illuminating.
Mr Stanbury points out, further, that the most recent Home Office report records that SLPP supporters have since been subjected to inhumane conduct constituting a breach of their human rights. The whole point of this line of submission, as I understand it, shows, he says, there is at least an arguable case that the claimant would be at risk if he was removed and deported to Sierra Leone and by implication may well support a right of appeal or a case for appealing any deportation notice. On the other hand, if the risk of removal is removed, there is a powerful incentive for the claimant not to abscond in the now remaining two months of his sentence. If he did abscond, the likely effect on his refugee status (i.e. termination and deportation) would be disproportionate to any perceived benefit to be derived from absconding. He has, in short, every reason, it is argued, not to abscond on the basis, as I understand it, that he would wish to appeal and to preserve his refugee status in this country.
The factors or reasons which go into weighing the balance as to whether to re-categorise the claimant from category C to category D are quite complex. It is these very facts, Mr Stanbury submits, which the defendant appears to have overlooked or failed to consider in his refusal to re-categorise the claimant as category D on 29 April 2010. Instead, the defendant, he argues, (whatever might now be said) appears to have been overly influenced by the stated intention of the UK Border Agency to remove the claimant’s refugee status.
The legislative background is as follows. The Prison Act 1952 (as amended) empowers and enables the Secretary of State, amongst other things, to commit prisoners to prisons as he may from time to time direct (section 12) and to make rules for the regulation and management of prisons and for the classification and control of prisoners (see section 47) with the purpose of encouraging and assisting them to lead a good and useful life (see section 3). The Prison Rules 1999 (as amended) require prisoners to be classified in accordance with the directions of the Secretary of State (Rule 7).
From time to time, this has led to a number of Prison Service Orders (PSOs) and Instructions (PSIs) made pursuant to Rule 7 in which the Secretary of State has set out mandatory guidelines and instructions for management of prisoners, staff and prisons, including their categorisation and allocation to prison.
The current system is that PSOs were issued until 31 July 2009 and those then in force remain in force until cancelled and replaced by a new system of PSIs gradually being brought into operation from 1 August 2009. PSIs (and indeed PSOs) appear to have an expiry date but many remain in force and are applied even though on the face of them they appear to have expired.
Categorisation is concerned with the individual security risk assessment of each prisoner. Allocation is the system by which a prisoner is allocated to a prison and a prison of itself the appropriate category to that prisoner. Those relevant to the issue of categorisation and allocation of prisoners in this case are PSO 0900 and PSI 03/2009. Both are mandatory instructions which the prison authorities are required to observe.
PSO 0900 (which came into force on 1 September 2000) sets out the principles of allocation and categorisation and outlines in an introduction the key role of the instructions:
“Categorisation and allocation of prisoners is a critical task. Effectively assigning prisoners to the correct security category and allocating them to an appropriate prison helps to ensure that they do not escape or abscond or threaten the control of establishments. It also means that prisoners are not held in conditions of security higher than are necessary.”
For present purposes the main relevant provisions are as follows. Paragraph 1.1.1, dealing with categorisation, lists category C and D prisoners as follows:
“Category C
Prisoners who cannot be trusted in open conditions, but who do not have the resources and will to make a determined escape attempt.
Category D
Prisoners who can be reasonably trusted in open conditions.”
Paragraph 1.2.1 deals further with categorisation: it provides that:
“prisoners must be categorised objectively according to the likelihood that they will seek to escape and the risk that they would pose should they do so.”
Paragraph 1.2.3 provides:
“Every prisoner must be placed in the lowest security category consistent with the needs of security and control. A prisoner must be assigned to the correct security category even if it is clear that it will not be possible to allocate him to a particular establishment for prisoners in that category.”
This is supplemented by PSI 03/2009, which deals with re-categorisation of prisoners from category C to D. It came into force on 24 May 2009 and on the face of it expired on 25 May 2010, though it still seems to be applied. It re-stated guidance formerly contained in PSI 45/2004 (itself the subject of proceedings, see R (Bryant) v Secretary of State for the Home Department [2005] EWHC 1663 (Admin)). It appears to be intended to incorporate it into PSO 1000, the National Security Framework, so is likely to have some relevance in fact for some time. Again, it is impossible to set out the whole document but the relevant parts are as follows:
“Paragraph 3: “The specific instruction on recategorisation to category D highlights the importance of weighing time left to serve in the assessment for category D and in particular the extent of any impact on public confidence should a long sentence prisoner abscond. Cases must be decided on their individual merits but to help those making the decisions the guidance is that prisoners should not normally spend any longer than 2 years in open prison before their expected release date.”
…
8.1 “The purpose of the recategorisation process is to determine whether, and to what extent, the risks a prisoner presented at his or her last review have changed and to ensure that the prisoner continues to be held in the most appropriate conditions of security.” To this end new or additional information impacting on the original categorisation is obviously of relevance and potential importance.
…
8.3 “It must be the aim that prisoners are held in the lowest possible security consistent with preventing escape or risk of harm to the public or to the security of the prison. However, for operational reasons, prisoners may be held in a prison of a higher security category, although the numbers of such prisoners must be limited by agreement between Regional Managers Custodial Services and PMU. On no account must a prisoner be allocated to a prison of a lower security category than the category assigned to the prisoner. “
Paragraph 14 of PSI 03/2009 deals with re-categorisation, as I say, from C to D and is worth setting out in full:
“14. Re-categorisation to category D
14.1 It is essential that prisoners must be assessed as trustworthy and sufficiently low risk before being allocated to open conditions. In making the decision, governors must keep in mind the particularly challenging management issues associated with the low physical security and supervision levels of the open estate and that the environment and regime opportunities available in open prison may not be suitable for a prisoner who is still many years away from possible release.
14.2 In addition to the risk assessment issues listed above (under Process) it is important to bear in mind the damage to public confidence in the Criminal Justice System if a prisoner serving a lengthy sentence were to abscond, particularly if the prisoner had spent a very short period of time in closed conditions and/or still has many years left to serve.
14.3 The risks to be assessed may conflict. Likelihood of abscond and risk of harm to the public and damage to public confidence if an abscond occurs will not necessarily be the same, and long sentence prisoners who statistically present an average or lower likelihood of abscond may represent a disproportionately high risk of harm to the public should they abscond and/or a high risk of damage to public confidence in the Prison Service’s ability to safeguard the public by keeping prisoners in safe custody.
14.4 When assessing long sentence prisoners for open conditions it is vital to balance the risks involved if the prisoner were to abscond against the likely benefits to the prisoner of going to open conditions at this stage. Governors will need to consider whether the prisoner has made sufficient positive and successful efforts to reduce risk levels and that the benefits he or she would gain from allocation to open prison are worthwhile at this particular stage in sentence. Consultation with the Police Intelligence Officer should be an integral part of the assessment of any long-term prisoner.
14.5 Every case must be considered on individual merit but, in general, long sentence prisoners should not be recategorised and allocated to open prison until they have served a sufficient proportion of their sentence in a closed prison to enable them to settle into their sentence and to access any offending behaviour programmes identified as essential to the risk reduction process.
14.6 In addition, prisoners should generally not be allocated to open prison :-
with more than 2 years to serve before their earliest release date ; and
in the case of prisoners subject to the release provisions of the Criminal Justice Act 1991 to whom the new duty to release at the half-way point in section 33(1A) does not apply (i.e. ‘unconverted’ 1991 Act prisoners with a PED), they must also be within 5 years of non-parole date (NPD).
Where prisoners are more than 2 years away from earliest release date they must still have their categorisation reviewed in line with the normal process and consideration given to whether there are exceptional circumstances to justify allocation to open prison at this stage. There is no right to have 2 years in open conditions before possible release.”
PSI 35/2002 removed a blanket ban on foreign national prisoners being allocated to category D. Instead, whilst deportation status was to remain a major factor in the risk assessment process, that process was to be applied in an individual basis in the same way as other prisoners.
The general guidance is now supplemented in the case of foreign national prisoners by PSO 4630. This explains in paragraph 14 that categorisation is applied in accordance with the principles I have previously referred to, but continues at paragraph 14.3 as follows:
“14.3. Before a foreign national prisoner who meets the deport criteria ... is classified, the individual risk must be assessed on the assumption that the deportation will take place unless a decision not to deport has already been taken by [what is now the UK Border Agency]…
14.4. Each case must be individually considered on its merits but the need to protect the public and ensure the intention to deport is not frustrated is paramount. Category D will only be appropriate where it is clear that the risk is very low.”
In R (Manhire) v Secretary of State for Justice [2009] EWHC 1788 (Admin), HHJ Langan QC had quashed a decision by the prison authorities declining to downgrade the claimant foreign national prisoner to category D given his exemplary prison record and settled family. Further, there were special circumstances in that case: there was no prospect of the claimant being removed to Zimbabwe since the UK authorities had decided not to deport foreign national prisoners to that country due to the local conditions there then existing.
Thus Mr Stanbury argues that the apparent reliance solely upon the claimant’s deportation status is irrational and unreasonable. In substance, his submissions amount to this: the prison authorities have simply reacted to the notification of the intended change of refugee status and applied the “paramount” consideration without any real anxious or proper consideration of any countervailing or balancing factors such as his exemplary record, the presence of his wife (he claims to be married but has no children), the likelihood of his appeal against a deportation order for the reasons previously given and matters of that kind. Some doubt as to the status of his wife has been cast in evidence put forward on behalf of the defendant and by a recent notification from the UK Border Agency of its decision to deport the claimant, to which I refer at the end of this judgment, all of which suggests that the claimant’s wife may be living in the UK illegally. This doubt seems to have arisen from information received after the decision of the review board on 29 April 2010.
Mr Murray on behalf of the defendant argues, equally succinctly, essentially as follows:
The decision should not be quashed as the claimant requests, but is entirely rational, reasonable and based on a careful assessment of risk in accordance with relevant policy, particularly paragraphs 14.3 to 14.4 of PSO 4360. His potential deportation impacted on and raised his risk of absconding to the extent the claimant could no longer be considered for open conditions. This was not a case of potential deportation to Zimbabwe as in Manhire but to Sierra Leone.
The decision may have been expressed succinctly in the form, but nevertheless, as the evidence of Mr Steele, the Prison Governor, filed in these proceedings makes clear, a balanced decision was reached after careful consideration of all the relevant factors including the positive aspects of the claimant’s case.
In any event, had the claimant appealed or requested he would have been informed more fully of the reasons for refusal to re-categorise him.
The decision is likely to be academic since a new review board now might reach the same conclusion, especially in light of the recent decision of the UK Border Agency to deport the claimant.
Whilst I accept that the powers and discretions conferred on the prison authorities under the framework I have outlined are wide, I have to say I am not persuaded by Mr Murray’s arguments.
First, I well appreciate that risk management is a matter for the prison authorities but, despite the further evidence from Mr Steele (previously mentioned) that the matter was given careful consideration and included such matters as his previous positive record, his wife being in the UK and a reference to variations of his name (which as it turns out had never been put to him and could be attributable to misspellings), the whole tenor of the facts as put forward on behalf of the Secretary of State (including in the Grounds of Defence as referred to previously) gives at least the appearance (even if it is not the fact) that a hurried review decision was made following the UK Border Agency’s indication of its intention to remove the claimant’s refugee status and it was this fact and this fact alone which caused the change of mind from that previously, a matter which is entirely underlined by the completion of the re-categorisation review setting out the reasons for the refusal to re-categorise him to category D. In fact, more accurately, it might be viewed as a re-categorisation from D to C, in view of the decisions that had been made shortly before. Certainly, as I have said, the form setting out the board’s conclusions appears only to reflect the single factor, the potential change of refugee status, and appears to reflect that solely as the determining factor. It does not, on the material presented, appear that this decision was put in any context at the time, despite the careful witness statement of Mr Steele, particularly balanced against the other factors previously noted: e.g. that the claimant had an “exemplary” record. Such context as appears suggests the prison authorities were throughout waiting a determination from the UK Border Agency of the claimant’s refugee status. I cannot and do not say Mr Steele is incorrect or misleading the court when he intimates a balanced decision was taken but in short, the expressed reasons for the decision had all the appearance that the UK Border Agency’s statement of intention led to the application of a blanket policy not to re-categorise foreign national prisoners at risk of deportation. If it was done, it does not, as I say, at least appear that any kind of proper balancing exercise was done. Justice is not only required to be done but seen to be done.
Second, I entirely accept that reasons need only be stated succinctly. But the essential elements need to be stated. Unlike R (Omoregbee) v Secretary of State for Justice, to which I have previously referred and on which Mr Murray relies as showing a succinct statement was sufficient, particularly where later evidence shows a balanced decision was in fact taken, it does not appear the point now taken was argued in that case. In any event, the decision form did note and did indeed refer on the face of it to the prisoner’s excellent reports (see paragraph 14 of the judgment).
Third, nor do I, regrettably, find it impressive that the prisoner is expected to appeal or to ask for further information to be given detailed reasons. Let me make it clear. I find it perfectly reasonable and practical that the detailed reasons should expand on the reasons stated, but the reasons stated still ought to be clear, even if they are succinct. I am not criticising succinctness. I am criticising absence. There is no reference in the form stating the decisions in the present case that the positive aspects were even considered or reconsidered. As I say, it has all the hallmarks of a change of decision prompted solely by notification of the UK Border Agency’s intent.
Fourthly, even if the decision is or now may be academic, that is not a reason for denying the claimant the order he seeks, as was the case in R (Smith and Mullally) v Governor HMP Lindholme [2010] EWHC 1356 (Admin).
In that case I observed at paragraph 52 in response to submissions from counsel for the defendant:
"The Secretary of State's policies as set out in the [prison] instructions are intended to provide guidance in the exercise of administrative discretion and how the discretionary powers are to be exercised.”
At paragraph 56 I added:
“The guidance [given by the PSOs and PSIs] however, is just that. I have already stated that there is flexibility built in to the policy. This means each case must be carefully and appropriately considered. The guidance in the PSOs and PSIs is not a "rule book". It is not to be followed blindly, slavishly, or even parrot fashion.”
As happened in that case, this case, too, has all the appearance of the UK Border Agency’s decision to consider withdrawal of refugee status being the overriding determinative matter to the exclusion of all else. The fact that something is “paramount’ does not mean all other factors are to be ignored or not given proper weight. Accordingly, in my judgment, the decision of 29 April 2010 does have the appearance of unreasonableness. This is not to say that the prison authorities, after a careful review of all the facts, might not still come to the same conclusion, despite the fact that the claimant has only two months left to serve of his sentence before conditional release, and particularly if the residence position of his wife has been further clarified in the meantime. But it is enough, in my judgment, to persuade me for present purposes that Mr Stanbury is right that the decision of 29 April 2010 ought, subject to one further matter, to be quashed.
The further matter is this. I was informed late yesterday and referred to previously of a notification from the UK Border Agency. This is to the effect that the UK Border Agency has now decided to deport the claimant and that he has been served with the appropriate notification, a copy of which I have seen. He was served with it, I was told, on 4 January last. He has, as the accompanying notice makes plain, a right of appeal against this decision. The notification will obviously have an impact on any future consideration of the claimant’s categorisation status, but whether he appeals may also be a matter which may or may not have an impact on any future consideration of his categorisation status. But these are not matters that are sufficient to alter my decision as regards the decision that was taken on 29 April 2010 which is the subject of the present application. These further factors are, in my judgment, matters for future consideration, even though I recognise (as, to be fair, did Mr Stanbury) that the result may be the claimant’s successful arguments in the present case may (but not necessarily must) be rendered academic and his victory somewhat pyrrhic.
Nevertheless, for the reasons given, although I shall quash the decision complained of, namely that of 29 April 2010 it is right to point out that technically Mr Stanbury was seeking declaratory relief only. The substance of his complaint on behalf of the claimant was that the decision of 29 April 2010 appeared irrational and unreasonable. I repeat I am not challenging nor disregarding what Mr Steele said, but what he said about the way the actual decision was reached is not consistent with the appearance of the decision as represented both to the claimant and (at least until Mr Steele’s witness statement) to the court. Thus presented the decision appeared unreasonable in that, as it appeared, the prison authorities did not weigh the new information in the balance. If they did, they failed to express it. That seems to me to be both unreasonable and unfair to the claimant.