Bahta & Ors, R (on the application of) v Secretary of State for the Home Department & Ors

These are appeals against decisions as to costs made by High Court Judges in immigration cases following consent orders. AK appeals against an order of Mitting J dated 15 September 2010, TZ (Eritrea) against an order of Lloyd Jones J dated 23 September 2010, RO against a decision of Mitting J dated 20 December 2010, Filmon Bahta against an order of King J dated 21 December 2010 and KD against an order of Edwards-Stuart J dated 17 February 2011. Following orders whereby disputes between the appellants and the Secretary of State for the Home Department (“the Secretary of State”) were resolved without a contested court hearing, each judge declined to make an order for costs in the appellants’ favour.

The appeal raises a question of general application, given the bulk of cases, in the immigration jurisdiction but applies across the judicial review public law jurisdiction. Public authorities do on occasions grant the relief sought by the claimant, or other substantial relief, at some stage in the proceedings and substantial costs may have been incurred. For the appellants, Mr Wilson QC submitted that, in refusing costs to the appellants, the judges have erred in law. It was submitted that they have not correctly applied the principles stated by Scott Baker J in R (Boxall) v Waltham Forest LBC 21 December 2000 (2001) 4 CCL Rep 258 and, further or in the alternative, the test stated in that case should be modified in the light of current circumstances and the recommendation of Jackson LJ, following his Review of Civil Litigation Costs: Final Report, dated December 2009 (“the Jackson Report”). There is also an issue as to whether the court has jurisdiction to hear these appeals.

The appellants were seeking, by judicial review, either permission to work in the United Kingdom, or indefinite leave to remain, the grant of which includes permission to work. The applications were based on rights granted by Immigration Rule 360 and article 11 of the Reception Directive. For present purposes there is no need to consider the instruments in detail. After 12 months an applicant for asylum should be allowed to work, the right in the United Kingdom not including self-employment or involvement in business.

In the course of proceedings, the appellants were granted what they sought and their applications were withdrawn by consent. Little turns on the differences of fact between the cases and I will keep references to a minimum. A cause of the delay in granting the relief sought was that the Secretary of State was awaiting the outcome of the decision of the Supreme Court in ZO and Others v Secretary of State for the Home Department [2010] UKSC36; [2010] 1 W.L.R. 1948. The issue in that case was whether the right to work was available to repeat or subsequent asylum seekers, as distinct from primary asylum seekers. The Secretary of State contended that it was not. Four of the appellants were repeat asylum seekers; KD was a primary asylum seeker and not subject to the ZO issue. Reversing High Court decisions, this court found against the Secretary of State in ZO and Others v Secretary of State for the Home Department [2009] EWCA Civ 442; [2009] 1 W.L.R. 2477 on 20 May 2009. On 28 July 2010, the decision of the Court of Appeal was upheld in the Supreme Court.

Indefinite leave to remain was granted to AK by the Secretary of State on 15 March 2010, to TZ on 10 March 2010, to RO on 16 August 2010 and to Bahta on 23 December 2009, following permission to work granted shortly before. All four fresh applications for asylum were made in 2007 so that the relief was granted in each case substantially later than the period of 12 months contemplated in the Directive and well after the Court of Appeal decision on 20 May 2009. KD was granted permission to work on 6 September 2010.

The judges differed in their views as to the date at which the claim for costs should be assessed, between the date of commencement of proceedings (RO and TZ), the date the claim was withdrawn (Bahta), any date before the Supreme Court judgment in ZO (AK), and the date on which it was decided to adjourn the judicial review claim for an oral hearing (KD). The appellants submitted that the correct date was the date on which the decision as to costs was made. In seeking permission to work, the appellants relied on paragraph 360 of the Immigration Rules and article 11 of the Reception Directive.

In refusing the appellants’ applications for costs, the judges either referred to, or plainly had in mind, the decision of Scott Baker J in Boxall. Scott Baker J stated, at paragraph 22:

“Having considered the authorities, the principles I deduced to be applicable are as follows:

(i)

the court has power to make a costs order when the substantive proceedings have been resolved without a trial but the parties have not agreed about costs.

(ii)

it will ordinarily be irrelevant that the Claimant is legally aided;

(iii)

the overriding objective is to do justice between the parties without incurring unnecessary court time and consequently additional cost;

(iv)

at each end of the spectrum there will be cases where it is obvious which side would have won had the substantive issues been fought to a conclusion. In between, the position will, in differing degrees, be less clear. How far the court will be prepared to look into the previously unresolved substantive issues will depend on the circumstances of the particular case, not least the amount of costs at stake and the conduct of the parties.

(v)

in the absence of a good reason to make any other order the fall back is to make no order as to costs.

(vi)

the court should take care to ensure that it does not discourage parties from settling judicial review proceedings for example by a local authority making a concession at an early stage.”

Those principles have been approved in subsequent cases, including in this court. The expression “plain and obvious” does not appear in the Boxall guidelines but has since been used when applying principles (iv) and (v).

In AK, Mitting J stated:

“Until the Supreme Court handed down its judgment in ZO on 28 July 2010, this was not a ‘plain and obvious case’. The grant of indefinite leave to remain on 15/03/10 made it academic.”

In TZ, Lloyd Jones J stated:

“I do not consider that the outcome of the proceedings when commenced was plain and obvious. The proceedings became academic because of the grant of ILR. That decision was taken independently of the grounds of challenge.”

In RO, Mitting J stated:

“The Claimant’s grounds of claim were vindicated by the unanimous judgment of the Supreme Court in ZO Somalia v SSHD (2010) UKSC 36; but at the stage at which the claim was brought, it was not ‘plain and obvious’ that it would succeed. It was issued at a time when the Defendant’s appeal to the Supreme Court was pending. Different Judges sitting in the Administrative Court made different decisions upon applications for permission and interim relief in similar cases. My own practice was to stay permission applications and refuse interim relief, pending the decision of the Supreme Court. The Defendant did not defy the order of Ian Dove QC, on the papers, to grant the Claimant permission to work. She applied, in accordance with paragraph 2 of the order, to discharge it on 48 hours written notice. Her application was not determined before judgment was given by the Supreme Court in ZO. The reality is that the Claimant gained nothing by these proceedings which he would not have gained by awaiting upon that decision.”

In Bahta, King J stated:

“I start from the premise that where as here, the substantive proceedings have been resolved without a trial, the overriding objective in any order as to costs is to do justice between the parties, with the court always remembering to take care to ensure that it does not discourage parties from settling judicial review proceedings.

Notwithstanding the authority of the existing Court of Appeal decision in ZO at the material time, I am not persuaded that this is a plain and obvious case in which the claim to Judicial Review would have bound to have succeeded at the time the claim was withdrawn by consent on the 14^th of July 2010 or at the earlier time when the principal claim for relief became academic by virtue of the grant of ILR on the 17^th and 23^rd of December 2009.

There is nothing to suggest that the decision to grant ILR was influenced by the judicial review claim or the grant of interim relief by Ouseley J on the 9^th of December 2009. I accept that that was a decision taken outside the grounds of challenge on the application for Judicial Review and was one which was an awaited one to settle the claimant’s immigration status and which in the event was determined exceptionally outside the Immigration Rules.

This was a case in which at the time of settlement, the decision of the Supreme Court in ZO had not yet been delivered. Had it not become academic, its outcome would have been determined by that decision. The defendant should not be penalised for making a sensible decision on settlement, by virtue of the fact that in the event the Supreme Court upheld the Court of Appeal, any more than the claimant should have been penalised in costs, in the event the Supreme Court had ruled to the contrary. I also accept that in any event the claimant would not necessarily have been entitled to an automatic grant of PTW.

The fall back position on costs is the appropriate one here.”

In KD, Edwards-Stuart J stated:

“1.

In the light of the guidance in Boxall and subsequent cases, I have to consider whether this is a case in which it is obvious which side would have won had the substantive issues being fought to conclusion.

2.

Whilst I have considerable sympathy with the Claimant’s position, given the position under the EC reception directive, I note that Mr Rabinder Singh QC was not prepared to give permission when the matter was before him on 12 August 2010. If he thought that this was a clear and obvious case, he would have given permission. Accordingly, I consider that it would not be appropriate to make any order in respect of the costs of this claim before that date.

3.

Any costs incurred thereafter will have been very modest. Accordingly, I consider that the only course open is to make no order for costs.”

We were referred to similar cases, in which there were decisions in writing the other way. I refer to three of them. In SS (CO Ref: 53/2010) Blake J, on 25 August 2010, stated:

“Once the CA had decided that the EU Reception Directive applied to second asylum claims, and there was no application to suspend the effect of that decision it was incumbent on the defendant to apply the law as it was declared to be even if only on a contingent basis pending any further appeal.

. . .

This was therefore a plain case when the law should have been applied and not deferred and the reasonable costs of challenging a plainly unlawful decision should be allowed.

The subsequent grant of ILR and the dismissal of the appeal to the Supreme Court are irrelevant, but costs are limited to 25 January in the light of the change of approach presaged on that date.”

On 12 October 2010, in RS (CO/5300/2010), Sales J stated:

“The claimant properly advanced his claim before commencing proceedings. The defendant refused to grant him permission to work and so it was necessary to commence proceedings . . . It is plain and obvious that on the authority of the Court of Appeal the claimant would have won the judicial review. It is fair that the defendant should pay the claimant’s reasonable costs.”

On 10 May 2010, His Honour Judge Anthony Thornton QC, sitting as a High Court Judge in YS (CO/14385/2009), stated in the course of his reasons for awarding costs:

“The claimant was justified in bringing judicial review proceedings since he has a long outstanding claim which had not been determined and in consequence he wanted to work, he needed permission to work, the decision of the Court of Appeal had held that someone in his position was entitled to receive permission to work pending the determination of his substantive asylum claim but the defendant had not granted him permission despite it being requested and had not provided a decision but only a statement that the decision would not be considered at the present time.”

Having set the scene, I consider the jurisdictional issue. There were slight variations between the paragraphs in the consent orders specifying how the decision on costs was to be made but they were in similar form. In Bahta, it was stated:

“Liability as to costs shall be determined by a judge on the basis of the filed written submissions only.”

On behalf of the respondent, Mr Swift QC submitted that the effect of the order is to make the decision of the judge on the papers final without the possibility of further recourse to the court, or to this court. He relied on the decision of this court in R (RS (Sri Lanka)) v Secretary of State [2011] EWCA Civ 114. In construing a similar order, Maurice Kay LJ, with whom Thomas LJ agreed, stated; at paragraph 8:

“the parties are putting their trust in the judge [in that case a single Lord Justice] to produce a binding decision as to where the costs should fall.”

What was sought in that case (paragraph 9) was an oral hearing, with more than one judge, under CPR r.52.16(6). Mr Swift submitted that the present situation is indistinguishable.

Mr Wilson QC, for the appellants, noted the absence of the expression “and shall be final” from the order. He relied on section 16 of the Senior Courts Act 1981, which in terms confers jurisdiction on this court “to hear and determine appeals from any judgment or order of the High Court”. In Mendes v London Borough of Southwark [2009] EWCA Civ 594, the order was for “costs to be decided by a single [High Court] judge on the basis of written submissions”. Sedley LJ, with whom Moore-Bick LJ agreed, stated, at paragraph 14:

“There is of course no judgment of the High Court on the merits but there is a valid order for costs. . . . In that situation section 16 of the 1981 Act appears to me to give the clearest possible jurisdiction for this court to entertain an appeal against the order if such an appeal is arguable . . .”

In R (Jones) v Nottingham City Council [2009] EWHC 271 (Admin), where a further hearing in the High Court was sought, following a decision on the papers, Collins J stated, at paragraph 10:

“The appropriate route if there is dissatisfaction with such an order is an appeal to the Court of Appeal. It is ‘a final order’ and any appeal lies to the Court of Appeal.”

The present point was not argued in that case.

In my judgment, High Court orders in the present form do not exclude the right of appeal to this court. Plain words would be required in a consent order if the parties intended to exclude the right of appeal to this court granted in section 16 of the 1981 Act. The words used in these orders do not have that effect. The court in RS was considering the different question of the right to an oral hearing before the same court and did not have the present situation in mind. This court has jurisdiction.

The general rule, stated in CPR r.44.3(2) is:

“The unsuccessful party will be ordered to pay the costs of the successful party.”

The court may make a different order (CPR r.43.3(2)(b)) and must “have regard to all the circumstances” (CPR r.44.3(4)). By virtue of CPR r.44.3(5) the conduct of the parties includes:

“(a)

Conduct before, as well as during, the proceedings, and in particular the extent to which the parties followed the Practice Direction (Pre-Action Conduct) or any relevant Pre-Action Protocol.” [“PAP”]

In the reasons given for the decisions appealed, no reference is made to compliance with the Protocol by the appellants and non-compliance by the respondent.

The application of the Boxall principle was considered in this court in R (On the application of Scott) v London Borough of Hackney [2009] EWCA Civ 217 where the claimant challenged by judicial review the adequacy of a care plan adopted by the Borough. The care plan was reviewed and the application withdrawn. Mr Kenneth Parker QC, sitting as a High Court Judge, made no order as to costs.

In a judgment dismissing the appeal, with which the Chancellor and Richards LJ agreed, Hallett LJ stated, at paragraph 44:

“It should not be forgotten that applications of this kind are usually dealt with on paper and in relatively short form. It will rarely be proportionate to enter into an overly detailed forensic analysis of the issues and the evidence and assessment of which party would have won or lost on which issue. The discretion is a broad one, and it is exercised in broad fashion.”

Hallett LJ analysed the conduct of the parties in that case. She acknowledged, at paragraph 41:

“Plainly, the compliance with or breach of the pre-action protocol must be a relevant factor to be taken into account.”

Reasons for dismissing the appeal were stated at paragraph 51:

“For my part, the furthest I would be prepared to go along the path urged upon us . . . would be to urge all judges to bear in mind that, when an application for costs is made, a reasonable and proportionate attempt must be made to analyse the situation and determine whether an order for costs is appropriate. I emphasise a reasonable and proportionate attempt, bearing in mind the pressures on the Administrative Court, yet another hard pressed institution. A judge must not be tempted too readily to adopt the fall back position of no order for costs. Having said that, in my view there is no reason whatsoever to suppose that that is what is happening in the Administrative Court generally or that Mr Kenneth Parker fell into that temptation on this occasion. On the contrary, as I see it, he has considered the issues carefully and come to a reasoned decision. He exercised his discretion in proper fashion, acknowledging his task was to produce, if he could, a fair (or just) outcome. Thus, when he resorted to the fall back position, he only did so having conducted the appropriate exercise which was well within the permitted generous ambit of his discretion.”

Concern had been expressed, reflecting paragraph 12 of Boxall, about the levels of pay in publicly funded work. I will return to this subject. Hallett LJ stated, at paragraph 50:

“It is not for this court to interfere and set aside a perfectly proper order because the rates of pay of publicly funded work are said to be too low. I understand the expressed concerns. It would be a sad day if society lost the services of lawyers prepared to act in publicly funded cases for the most vulnerable in society. It would also, I note, be a sad day if hard pressed local authorities found themselves unable to care for the vulnerable and needy in their areas, in the way they would wish, because they have wasted too many precious resources on unmeritorious claims.”

In Mendes Sedley LJ, with whom Moore-Bick LJ agreed, reversed the decision of the judge and found that the claimants should recover their costs of the judicial review proceedings. The court took as the relevant point of time the moment the application for costs came before the judge for decision. When the case came before the judge, it was plain that the claimant would have won and “there was no sufficient reason in the documents before [the judge] to resort to the default position of no order as to costs”. That conclusion was reached notwithstanding that it did not involve “any criticism of the local authority’s lawyers” (paragraphs 24 and 25).

A further factor is the problem identified by Lord Hope in Re appeals by Governing Body of JFS [2009] 1 WLR 2353. Lord Hope stated:

“24.

As has already been noted, Ms Rose declined to seek an order that each side should be liable for its own costs in any event on the ground that to do so would be wrong in principle. As Scott Baker J observed in [Boxall] para 12, the failure of a legally aided litigant to obtain a costs order against another party may have serious consequences. This is because, among other things, the level of remuneration for the lawyers is different between a legal aid and an inter partes determination of costs. This disadvantage is all the greater in a case such as this. It is a high costs case, for which lawyers representing publicly funded parties are required to enter a high costs case plan with the Legal Services Commission. It is a common feature of these plans that they limit the number of hours to an artificially low level and the rates at which solicitors and counsel are paid to rates that are markedly lower than those that are usual in the public sector. Mr Reddin has indicated that, as they are defending a win, E’s solicitors would not be expected to be paid at risk rates. Nevertheless the rate of remuneration that is likely to be agreed for this appeal will be considerably lower than that which would be reasonable if costs were to be determined inter partes.

25.

It is one thing for solicitors who do a substantial amount of publicly funded work, and who have to fund the substantial overheads that sustaining a legal practice involves, to take the risk of being paid at lower rates if a publicly funded case turns out to be unsuccessful. It is quite another for them to be unable to recover remuneration at inter partes rates in the event that their case is successful. If that were to become the practice, their businesses would very soon become financially unsustainable. The system of public funding would be gravely disadvantaged in its turn, as it depends upon there being a pool of reputable solicitors who are willing to under take this work. In [Boxall] Scott Baker J said that the fact that the claimants were legally aided was immaterial when deciding what, if any, costs order to make between the parties in a case where they were successful and he declined to order that each side should bear its own costs. It is, of course, true that legally aided litigants should not be treated differently from those who are not. But the consequences for solicitors who do publicly funded work are a factor which must be taken into account. A court should be very slow to impose an order that each side must be liable for its own costs in a high costs case where either or both sides are publicly funded. Had such an order been asked for in this case we would have refused to make it.”

Mr Wilson submitted that following the Jackson Report there needs to be a change of landscape and of culture. Boxall was decided before the implications of the CPR and of PAPs had emerged. The judicial review protocol took effect in March 2002, that is after Boxall. Those submissions were developed by Miss Lieven QC, for the Public Law Project (“PLP”), the intervener.

Miss Lieven supported the appellants’ approach to the Jackson Report. To avoid further litigation about costs, this court should give clear guidance as to the proper approach in circumstances such as the present, she submitted.

Miss Lieven submitted that, in cases where relief has been granted, it was for the defendant to rebut the inference of defeat. Whether there had been compliance with the PAP was the appropriate starting point for a consideration of costs in circumstances such as the present. Boxall principle (v) should be amended accordingly. A defendant, in present circumstances, needs to show a good reason for failing to comply, if he is not to pay costs. A party who has complied with the PAP and obtained the result sought in the judicial review should not be seriously out of pocket by reason of the absence of an order for costs in his favour.

The problem is exacerbated, Miss Lieven submitted, by the very considerable difference between the level of publicly funded costs, still paid at 1994 rates, and costs inter partes. The order that there shall be no order as to costs has a chilling effect on the ability of lawyers to act, both for publicly funded and privately paying litigants. This, it was submitted, has significant constitutional implications.

Miss Lieven referred to paragraph 3.1 of the PAP for judicial review, which is dated October 2006. Having urged parties to consider some form of alternative dispute resolution procedure, it is stated:

“The Courts take the view that litigation should be a last resort, and that claims should not be issued prematurely when a settlement is still being actively explored. Parties are warned that if the protocol is not followed (including this paragraph) then the Court must have regard to such conduct when determining costs.”

It is also correctly noted that a claim for judicial review “must be filed promptly and in any event not later than three months after the grounds to make the claim first arose”. A potential claimant cannot dally.

Paragraph 13 of the PAP provides:

“Defendants should normally respond within 14 days using the standard format at Annex B. Failure to do so will be taken into account by the court and sanctions may be imposed unless there are good reasons.”

At paragraph 14, the PAP makes provision for a request for a reasonable extension of the 14 day period. Reasons for such a request are required. The Secretary of State has failed completely to comply with the Protocol in these cases. No requests for extension were made. It was suggested that the policy was to sit back and wait upon a decision of the court granting permission to the applicant. Defendants must not be left with the impression that, provided they concede before the final hearing, they will not be ordered to pay costs.

Mr Douglas, in his statement on behalf of the respondent, is guilty of special pleading, it was submitted, in referring to problems arising from the volume of work and funding. There can be no special rule for a government department. The Protocol imposes on both parties an obligation to be reasonable. The claims were meritorious and reliance is not placed by the appellants on special consideration for publicly funded parties.

By way of analogy, Miss Lieven referred to CPR r.38.6(1) which deals with discontinuance and provides that, unless the court orders otherwise, a claimant who discontinues is liable for the defendant’s costs to the date of the discontinuance. The claimant bears the burden of justifying a departure from the presumption (in re Walker Wingsail Systems Plc [2005] EWCA Civ 247, [2006] 1 WLR 294 and R v Liverpool City Council ex parte Newman (1992) 5 Admin LR 669). This situation, submitted Miss Lieven, is the mirror image of the approach in the Jackson Report.

For the respondent, Mr Swift submitted that there is no compelling reason to replace the Boxall principles. They fit with the CPR, which are cited in the judgment. The PAP concept is clearly recognised in the judgment (paragraph 14). Scott Baker J adopted the statement of Simon Brown J in Ex parte Newman that “it would seldom be the case that on discontinuance this court would think it necessary or appropriate to investigate in depth the substantive merits of what had by then become an academic challenge. That ordinarily would be a gross misuse of the court’s time and further burden its already over-full list”. Simon Brown J also stated, however, that, as a general rule, discontinuance can often be equated with defeat or an acknowledgement of defeat.

Mr Swift does not suggest that the issue of compliance with the PAP in public law cases falls outside the Boxall principles, a possible, but in my judgment unjustified, reading of the last two sentences of paragraph 41 of Hallett LJ’s judgment. The court need not go outside the principles stated in Scott, it was submitted by Mr Swift.

Reference was made to the breadth of the discretion available to judges in circumstances such as the present, as expressed by Rimer LJ in Coyne v DRC Distribution Ltd [2008] EWCA Civ 488, [2008] BCC 612, at paragraph 69. A wide range of options is available. In exceptional cases, where costs are very considerable, Rimer LJ stated that it may be “proportionate to direct oral evidence on such factual issues as would enable [the court] to resolve the costs issues.”

Mr Swift referred to the judicial review seminar held by Jackson LJ in July 2009 while preparing his Report. Boxall was discussed but Scott was apparently not mentioned. Jackson LJ did not of course receive at the seminar the structured submissions addressed to this court. Jackson LJ referred, at paragraph 3.21 of his Final Report, to the representations of PLP at the seminar:

“PLP points out that Boxall was decided before the protocol came into effect. PLP states that research shows that approximately 60% of judicial review cases are now settled following the letter of claim. Nevertheless some authorities wait to see whether proceedings will in fact be issued and whether permission will be granted before settling. Furthermore, many judicial review claims settle following the grant of interim relief, such as interim accommodation or an order for community care assessment. Yet the effect of Boxall is that claimants seldom recover costs in these cases. PLP propose that, if C has followed the protocol but D has not, there should be a presumption that D should pay C’s costs. This would encourage reasonable litigation behaviour on the part of defendants. Also it would transfer the costs burden in many cases from the legal aid fund to the defendant authorities. Similar arguments are advanced by the firms of claimant solicitors mentioned above.”

In his Final Report Jackson LJ recommended, at paragraphs 4.12 and 4.13:

“The Boxall approach made eminently good sense at the time that case was decided. However, now that there is an extremely sensible protocol in place for judicial review claims, I consider the Boxall approach needs modification, essentially for the reasons which have been urged upon me during Phase 2.

. . . in any judicial review case where the claimant has complied with the protocol, if the defendant settles the claim after (rather than before) issue by conceding any material part of the relief sought, then the normal order should be that the defendant pays the claimant’s costs. A rule along these lines would not prevent the court from making a different order in those cases where particular circumstances warranted a different costs order.”

In its response to the Jackson Report in March 2011, the Ministry of Justice did not refer to the recommendation at 4.13.

Mr Swift submitted that a judge’s general discretion should be maintained. Compliance with the PAP should not be the decisive, or even the pre-eminent consideration, in deciding on a costs order. That would be to take the wrong starting point. CPR r.44.3 requires all considerations to be taken into account. For example, non-compliance with the PAP would not be relevant if it was obvious that the defendant would have succeeded in the action. To start with a presumption based on compliance or non-compliance with a PAP would be to move away from principle.

Mr Philip Douglas, Director in the Appeals and Removals Directorate of United Kingdom Border Agency (“UKBA”) referred in his statement to the financial resources available to UKBA, and to the number of claims to be considered, most of which were found to be without merit. In the absence of a response on the merits at the PAP stage, the Secretary of State will, of course, escape without having incurred any significant costs in cases where permission is subsequently refused.

Mr Swift submitted that the court should take into account the reason for and the consequences of the failure to comply with PAPs. To move the decision as to reasonableness back to the PAP stage would be to relocate the problem and not confront it. Mr Douglas invited the court to maintain the Boxall guidance. The court should have regard to the particular concerns of UKBA. UKBA would, for example, wish to be able to argue that no order for costs is appropriate where it settles a claim for “purely pragmatic reasons”, despite having an arguable or even strong defence.

In written submissions on behalf of the Bar Council, Mr Clayton QC and Ms Jackson submitted that the court should adopt Jackson LJ’s recommendation at paragraph 4.13. Compliance with the PAP is a “highly pertinent factor” in favour of the presumption recommended by Jackson LJ. If there is focus on the extent to which the parties have complied with their protocol obligations, claimants will need to demonstrate that they have identified the issues in dispute with sufficient clarity and defendants will need to demonstrate that they have addressed the issues in dispute and, when requested, provided a fuller explanation for their decision and disclose any relevant documentation. This will have the effect of weaker cases not proceeding and stronger cases being settled.

Mr Wilson accepted that the clarity of a letter of claim requires scrutiny. In his written submissions, Mr Swift made suggestions as to how letters of claim should be framed in order to facilitate the Secretary of State’s response, with emphasis on compliance with Annex A to the PAP. No criticism is made of the adequacy of the letters of claim in these cases. As to defendants, the PAP provides, at paragraph 13:

“Defendants should normally respond within 14 days using the standard format at Annex B. Failure to do so will be taken into account by the court and sanctions may be imposed unless there are good reasons.”

In relation to Lord Hope’s judgment in JFS, Mr Swift submitted that it was given in a different context. Moreover, the principle stated at the end of paragraph 25 is of restricted application. None of the individual cases considered in this appeal are high costs cases.

In JFS, the challenge was to a decision of the Legal Services Commission not to continue to provide funding for a litigant who had been successful in the Court of Appeal and the unsuccessful parties had been granted permission to appeal. It was held that in all the circumstances a decision to refuse public funding was so unreasonable as to be unlawful. Lord Hope went on to consider, in paragraphs 24 and 25, the suggestion that a solution would be to order that each side be liable for its own costs in any event. I comment now that, while the context is different, I regard Lord Hope’s statement that “the consequences for solicitors who do publicly funded work are a factor which must be taken into account” is intended to be of general application.

It was submitted for the appellants and interested parties that the long term impact of the ability of the public to obtain appropriate advice and representation in judicial review proceedings will be imperilled unless the Boxall principles are amended. The long term impact of those principles is to undermine the financial viability of lawyers who now undertake public law work. Reference was made to research which shows that a significant number of cases settle after permission is granted.

That may restrict access to legal advice to such a degree that it breaches the right of access to the courts. It was submitted that where a claimant has complied with the PAP and the defendant settles the claim after, rather than before, issue of proceedings by conceding a material part of the relief sought, the normal order should be that the defendant pays the claimant’s costs.

It was submitted, for the Secretary of State, that in each of the decisions challenged, the judge’s decision was made within the bounds of the broad discretion in costs available to the judge and should not be disturbed. The submissions on behalf of the Secretary of State on individual cases were short, the main concern being as to the principles involved. It was submitted:

(i)

The grant of ILR was wholly unrelated to the course of the judicial review proceedings.

(ii)

In support of the submission that the result was not “plain and obvious”, Mr Swift relied on the High Court decision in the Secretary of State’s favour in ZO and on the grant of leave to appeal by the Supreme Court. It was not “plain and obvious” that the Supreme Court would dismiss the appeal.

(iii)

In some of the cases, compromises were reached prior to the decision of the Supreme Court.

(iv)

It was also submitted that because, under article 11(2) of the Directive, the Secretary of State retained a discretion “to decide conditions for granting access to the labour market”, it was entirely possible that the appellants would be refused permission to work in the United Kingdom.

I comment on each of those submissions. As to (i), it was unsubstantiated. In any event, the appellants were entitled to the relief claimed and had to commence proceedings to obtain it. As to (ii), I do not accept that the respondent was entitled to deny relief until the Supreme Court had ruled. The Court of Appeal had ruled that applicants in the position of the appellants (apart from KD who as a primary asylum seeker was entitled to relief either way), were entitled to relief on the basis of the decision of the Court of Appeal. That was the law unless and until a higher court, or Parliament, ruled otherwise. In the absence of a stay on the Court of Appeal decision, the respondent could not require the appellants to wait for relief until the Supreme Court had ruled. It was suggested in one case that the appellant “gained nothing by these proceedings which he would not have gained by awaiting upon” the decision of the Supreme Court. What he stood to gain was the right to work at an earlier date.

As to (iii), there was in some cases a change of front by the Secretary of State before the Supreme Court had given judgment. That was admirable but in the absence of submissions that the appellants proceeded rather than settled when they had obtained relief, it does not affect the outcome in costs.

As to (iv), no attempt has been made to substantiate this submission.

I do not accept the suggestion made in some of the decisions appealed that ILR might have been given for reasons unconnected with ZO. There is no evidence to support that submission. I accept there may be cases where relief is granted for reasons unconnected with the grounds of relief relied on but I am not prepared to treat the present cases as such. In the case of Bahta, for example, a stay in the judicial review proceedings was sought “in relation to the outstanding applications for permission to work until the issue was decided by the Supreme Court.”

Mr Swift submitted that the decision should be based on the situation at the date of commencement of proceedings. The question was whether, at the time of issue, proceedings should have been commenced.

In my judgment, it is the date at which the application for costs is determined that is the relevant date for assessment. However, a consideration of what order should be made requires consideration of the whole sequence of events and the conduct of the parties throughout. That includes the conduct of the parties after the defendant has told the claimant that relief is being offered and what the relief is.

What is not acceptable is a state of mind in which the issues are not addressed by a defendant once an adequately formulated letter of claim is received by the defendant. In the absence of an adequate response, a claimant is entitled to proceed to institute proceedings. If the claimant then obtains the relief sought, or substantially similar relief, the claimant can expect to be awarded costs against the defendant. Inherent in that approach, is the need for a defendant to follow the Practice Direction (Pre-Action Conduct) or any relevant Pre-Action Protocol, an aspect of the conduct of the parties specifically identified in CPR r.44.3(5). The procedure is not inflexible; an extension of time may be sought, if supported by reasons.

Notwithstanding the heavy workload of UKBA, and the constraints upon its resources, there can be no special rule for government departments in this respect. Orders for costs, legitimately made, will of course add to the financial burden on the Agency. That cannot be a reason for depriving other parties, including publicly funded parties, of costs to which they are entitled. It may be, and it is not of course for the court to direct departmental procedures, that resources applied at an earlier stage will conserve resources overall and in the long term.

In the case of publicly funded parties, it is not a good reason to decline to make an order for costs against a defendant that those acting for the publicly funded claimant will obtain some remuneration even if no order for costs is made against the defendant. Moreover, a culture in which an order that there be no order as to costs in a case involving a public body as defendant, because a costs order would only transfer funds from one public body to another is in my judgment no longer acceptable.

Equally, it is not an acceptable reason to make an order for costs in favour of a claimant, and neither the appellants nor the interested parties have suggested it is, that publicly funded lawyers are, or are claimed to be, inadequately remunerated. Whether to make an order for costs depends on the merits of the particular application. However, both the warning in Scott against too ready resort to making no order as to costs, and the indication by Lord Hope in JFS, cited at paragraph 28, in relation to publicly funded parties, demonstrate the need for analysis of the particular circumstances.

I have serious misgivings about UKBA’s claim to avoid costs when a claim is settled for “purely pragmatic reasons”. My reservations are increased by the claim, on the facts of the present cases, that the right to work was granted for pragmatic reasons. I am unimpressed by suggestions made in the present cases that permission to work was granted for reasons other than that the law required permission to work to be granted. There may be cases in which relief may be granted for reasons entirely unconnected with the claim made. Given the Secretary of State’s duty to act fairly as between applicants, and the duty to apply rules and discretions fairly, a clearly expressed reason would be required in such cases. The expression “purely pragmatic” covers a multitude of possibilities. A clear explanation is required, and can expect to be analysed, so that the expression is not used as a device for avoiding an order for costs that ought to be made.

In addition to those general statements, what needs to be underlined is the starting point in the CPR that a successful claimant is entitled to his costs and the now recognised importance of complying with Pre-Action Protocols. These are intended to prevent litigation and facilitate and encourage parties to settle proceedings, including judicial review proceedings, if at all possible. That should be the stage at which the concessions contemplated in Boxall principle (vi) are normally made. It would be a distortion of the procedure for awarding costs if a defendant who has not complied with a Pre-Action Protocol can invoke Boxall principle (vi) in his favour when making a concession which should have been made at an earlier stage. If concessions are due, public authorities should not require the incentive contemplated by principle (vi) to make them.

When relief is granted, the defendant bears the burden of justifying a departure from the general rule that the unsuccessful party will be ordered to pay the costs of the successful party and that the burden is likely to be a heavy one if the claimant has, and the defendant has not, complied with the Pre-Action Protocol. I regard that approach as consistent with the recommendation in paragraph 4.13 of the Jackson Report.

I do not accede to the request to tack on words to the Boxall guidelines to meet the appellants’ submissions. Such a formula would carry the danger of being used mechanistically when what is required is an analysis of the circumstances of the particular case, applying the principles now stated. These include the warning in Scott that a judge should not be tempted too readily to adopt a fall back position.

The circumstances of each case do require analysis if injustice is to be avoided. Such analysis will not normally be difficult if the parties have stated their cases competently and clearly and if the statement of reasons required when a consent order granting relief is submitted to the court genuinely and accurately reflects the reason for the termination of proceedings.

I accept that the principle of proportionality, and the workload of the courts, require that limits are placed on the degree of analysis which is appropriate but judges should not too readily be deterred. If they find obscurity, or obfuscatory conduct by the parties, that can be reflected in the order made. A willingness to investigate is likely to promote clarity in future cases.

Where relief is granted by consent, CPR r.54.18 provides a procedure whereby the court may decide the claim for judicial review without a hearing. That procedure should be followed wherever possible. It requires the filing of a document signed by all the parties “setting out the terms of the proposed agreed order together with a short statement of the matters relied on as justifying the proposed agreed order and copies of any authorities or statutory provisions relied on” (CPR PD 54A, para.17.1).

In the present cases, the licence granted by the judges to the Secretary of State to await a judgment of the Supreme Court and still avoid costs was unjustified. The decision paid no regard to the rule that what the Court of Appeal says is the law, is the law, unless and until overruled by a superior court or by Parliament. The Court of Appeal found that appellants such as these were entitled to work and there was no justification for withholding that right from them. Proceedings were instituted and pursued because the right was withheld. Once the right was granted by the Secretary of State, as was required, the claims were withdrawn within a reasonable time and it has not been argued otherwise in the cases before this court. I see no justification for resorting to the fall back position stated in Boxall in these cases. As I have said, I am unimpressed by the suggestion that the right to work was granted for some reason other than that the law required it.

I would allow these appeals and order costs against the Secretary of State in each case.

I agree.

I agree that these appeals should be allowed for the reasons given by Pill LJ in his judgment with which I am in full agreement. I presume to make three observations of my own from the perspective of one with wide experience of litigation generally but as a stranger in the world of judicial review.

The first concerns our jurisdiction to hear this appeal. It is clear to me that the provision in the consent order for the determination of costs by a judge on the basis of written submissions was a sensible and convenient procedural tool but that is just what it was. It did not have the effect of depriving any party of a substantive right (in this case to seek permission to appeal); to do that would require clear, express words to that effect of which they were none here. In my view this Court clearly had jurisdiction to entertain these appeals.

Secondly I am satisfied that the general approach, that the merits of a case insofar as they are relevant to costs fall to be assessed when the issue of costs is determined, should apply in judicial review as elsewhere. Of course the history of the case and the litigation conduct of the parties throughout may be highly relevant or even determinative to the exercise of the judicial discretion as to costs but merits should be assessed as at the date of determination of costs.

Thirdly it is clear to me that Boxall is a well-established guide in the area of judicial review but like all guides it must be applied both to the particular facts of the instant case and it must take account of procedural developments like PAPs. Compliance with PAP, whilst not determinative in itself, must now be a highly relevant factor in the exercise of the judicial discretion as to costs.