England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE ADMINISTRATIVE COURT
Mr Justice Davis
[2004] EWHC 3011 (Admin)

		Royal Courts of Justice Strand, London, WC2A 2LL
		1 March 2005

Lord Phillips of Worth Matravers MR : This is the judgment of the court, prepared by Brooke LJ.

1. Introduction

1. On 22^nd December 2004 we heard an application by the claimants for permission to appeal against an order made by Davis J two days earlier whereby he refused to grant them a protective costs order ("PCO") for the substantive two-day hearing of this judicial review application which was fixed to take place on 13^th-14^th January 2005. These proceedings were commenced on 29^th November 2004, and on 2^nd and 3^rd December Beatson J granted the claimants an initial PCO and directed that their application for such an order to cover the main hearing should be listed as soon as possible. He also directed a "rolled-up hearing" of the substantive application so that if the judge at that hearing granted the claimants permission to apply for judicial review he would immediately proceed to hear that application on its merits.
2. Although Davis J refused the claimants permission to appeal to this court, he extended their interim PCO to cover their equivalent application in this court. The matter was listed before us as an application for permission to appeal with the appeal to follow if permission was granted. At the end of the hearing, which lasted a full day, we said that we would grant permission to appeal and that we would allow the appeal. We directed that a PCO should be made which must include a cost-capping element, along the lines of that directed by this court in King v Telegraph Group Ltd [2004] EWCA Civ 613 (see in particular paras 101-2). The senior costs judge then arranged a hearing during the vacation at which he could fix the amount of the cap, but in the event the defendant's solicitors elected to consent to an order made in the maximum amount claimed by the claimants on the basis that they would be at liberty to challenge the reasonableness of the amount claimed in due course, if the need arose. On 13^th January 2005 a consent order was made disposing of the claimants' application for judicial review, so that the merits of their case never received a judicial determination..
3. In this judgment we give the reasons why we decided to allow the appeal against Davis J's order.

2. The nature of the hearing

4. We will start by explaining the status of the Public Law Project in this litigation. Because this was the first occasion on which issues relating to PCOs had been considered in depth by this court (since an appeal to this court was likely whatever decision the judge had made), the judge permitted the Public Law Project to intervene by placing before him a substantive generic submission settled by junior counsel (Mr Michael Fordham) which set out reasons, supported by authority, why the courts should now be willing to adopt a more relaxed approach than hitherto when invited to make PCOs in public law cases which raise issues of general public importance. On the hearing of the appeal we permitted Mr Drabble QC to make brief submissions to us by way of oral exposition of these contentions.
5. Both the hearing before the judge and the hearing in this court had to be arranged at short notice and under great pressures of time. For this reason, although over 40 authorities were placed before us, we had no opportunity for pre-reading, and it was only after we announced our decision that we had a proper opportunity to study the case-law in depth, and to follow up some of the leads to other cases that are suggested in the case-law.

3. The traditional approach to costs in private law litigation

6. It will be convenient to structure this judgment by considering the relevant law first, and then to explain why we considered that it was appropriate to grant a PCO on the facts of this particular case. The general purpose of a PCO is to allow a claimant of limited means access to the court in order to advance his case without the fear of an order for substantial costs being made against him, a fear which would disinhibit him from continuing with the case at all. In this jurisdiction the leading authority on this topic is currently the judgment of Dyson J in R v Lord Chancellor ex p CPAG [1999] 1 WLR 347 (for which see para 44 below).
7. As a general rule it has been traditionally accepted in the courts of England and Wales that costs follow the event. In British Columbia (Minister of Forests) v Okanagan Indian Band (2003) 114 CCR 2d 108 LeBel J said at para 19:

"The jurisdiction to order costs of a proceeding is a venerable one. The English common law courts did not have inherent jurisdiction over costs, but beginning in the late 13^th century they were given the power by statute to order costs in favour of a successful party. Courts of equity had an entirely discretionary jurisdiction to order costs according to the dictates of conscience."

8. In McDonald v Horn [1995] ICR 685 Hoffmann LJ took up the story at p 693.

"The court's jurisdiction to deal with litigation costs is based upon section 51 of the Supreme Court Act 1981, which, with some rearrangement of the words, is derived from section 5 of the Supreme Court of Judicature Act 1890:

'(1) Subject to the provisions of this or any other enactment and to rules of court, the costs of and incidental to all proceedings in . . . the High Court . . . shall be in the discretion of the court. . . . (3) The court shall have full power to determine by whom and to what extent the costs are to be paid. (4) In subsections (1) and (2) 'proceedings' includes the administration of estates and trusts.'

The background to the Act of 1890 is briefly as follows. In the old courts of common law, costs followed the event. The judge had no discretion. In the Court of Chancery, costs were in the discretion of the court but that discretion was exercised according to certain principles which I shall discuss later. The first Rules of the new Supreme Court of Judicature (enacted in 1875) adopted the Chancery practice. But in In re Mills' Estate (1886) 34 Ch D 24 the Court of Appeal decided that the Rules conferred a discretion only in cases in which before the Judicature Acts the courts would have had jurisdiction to make awards of costs. The Act of 1890 was intended to confer such jurisdiction in any case whatever.

In Aiden Shipping Co. Ltd. v. Interbulk Ltd [1986] AC 965 the House of Lords drew attention to the broad language of section 51 of the Supreme Court Act 1981. The policy, said Lord Goff of Chieveley, at p. 975, was to confer jurisdiction in wide terms:

'thus ensuring that the court has, so far as possible, freedom of action, leaving it to the rule-making authority to control the exercise of discretion (if it thinks it right to do so) by the making of rules of court, and to the appellate courts to establish principles upon which the discretionary power may, within the framework of the statute and the applicable rules of court, be exercised.'

The discretion conferred by section 51 is thus by no means untrammelled. It must be exercised in accordance with the rules of court and established principles."

9. In Aiden the House of Lords found it possible to interpret these wide statutory powers so as to confer a jurisdiction on a court, if it thought it just, to order the payment of costs by someone who was not even a party to the litigation. It considered that there was no justification for implying a limitation on the court's powers to the effect that costs could only be ordered to be paid by the parties themselves (see Lord Goff at pp 979-980).
10. In Davies v Eli Lilley & Co [1987] 1 WLR 1136 this court upheld an order made by Hirst J in the Opren litigation to the effect that the 1500 plaintiffs should contribute rateably to the costs incurred by the legally aided lead plaintiff in a test action. Lloyd LJ said at p 1144 that Order 62 Rule 3(3) was concerned with the manner in which, and not the time at which, the court's discretion as to costs should be exercised, and that there was nothing in the language of the rule to prohibit the exercise of the discretion at an earlier stage than the conclusion of the proceedings where the interests of justice so required. Sir John Donaldson MR congratulated the judge (at p 1143) on providing a very fair and workable order in a novel and highly complex situation.
11. In Steele Ford & Newton v CPS [1994] 1 AC 22 the House of Lords held that section 51 did not confer a power on a court to order the payment of a successful party's costs out of central funds in the absence of any express statutory power enabling such an order to be made. After referring to a number of situations in which a court was unable to achieve justice for a successful litigant, Lord Bridge said (at pp 40-41)

"I will not multiply examples, but I hope I have said enough to explain why I cannot attribute to the legislature any general willingness to provide the kind of publicly funded safety net which the judiciary would like to see in respect of costs necessarily and properly incurred by a litigant and not otherwise recoverable…

Some general legislative provision authorising public funding of otherwise irrecoverable costs, either in all proceedings or in all appellate proceedings, would no doubt be an admirable step in the right direction which the judiciary would heartily applaud. But this does not, in my opinion, justify the courts in attempting to achieve some similar result by the piecemeal implication of terms giving a power to order payment of costs out of central funds in particular statutes, which can only lead to anomalies.

The courts must always resist the temptation to engage, under the guise of statutory interpretation, in what is really judicial legislation, but this is particularly important in a sensitive constitutional area, such as that with which we are here concerned, where we should be scrupulous to avoid trespassing on parliamentary ground. I would hold that jurisdiction to order payment of costs out of central funds cannot be held to have been conferred by implication on the courts by any of the statutory provisions which I have examined. Indeed, I find it difficult to visualise any statutory context in which such a jurisdiction could be conferred by anything less than clear express terms."

12. In Aiden Lord Goff referred to the work undertaken by the appellate courts in establishing principles upon which the discretionary power to order costs might be exercised. So far as conventional private law litigation is concerned, a good example of this process at work can be seen in the judgment of Atkin LJ in Ritter v Godfrey [1920] 2 KB 47. In that case the trial judge had refused to award costs to a successful defendant in a clinical negligence action. He was mainly influenced in this regard by the attitude the defendant had adopted in response to a letter before action, which, in the words of the headnote to the report, he had written in a tone of levity and in somewhat insulting terms. In agreeing that his costs order should be overruled, Atkin LJ reviewed the relevant case law and then said:

"It is not easy to deduce from these authorities what the precise principles are that are to guide a judge in exercising his discretion over costs. And yet as the discretion is only to be exercised where there are materials upon which to exercise it, it seems important to ascertain the principles upon which a judge is to discern whether the necessary materials exist. In the case of a wholly successful defendant, in my opinion the judge must give the defendant his costs unless there is evidence that the defendant (1.) brought about the litigation, or (2.) has done something connected with the institution or the conduct of the suit calculated to occasion unnecessary litigation and expense, or (3.) has done some wrongful act in the course of the transaction of which the plaintiff complains."

13. There are echoes of these principles in the language of CPR 44.4 (a) and 44.5, whose meaning was recently explored by Brooke LJ in Groupama Insurance Co Ltd v Overseas Partners Re Ltd [2003] EWCA Civ 1846. In that case this court overruled the refusal of a trial judge to award any costs at all to a successful defendant in the Commercial Court and substituted an order that it be allowed 90% of its costs of the action (the discount of 10% being attributable to the conduct to which the judge had taken exception).
14. From time to time the judges in the Chancery Division tempered the effect of the "costs follow the event" principle in cases where there was what we will describe as a "private fund" available. This private fund might be the assets of a trust (In re Beddoe: Downes v Cotton [1893] 1 Ch 547); the assets of a company in a minority shareholders' action (Wallersteiner v Moir (No 2) [1975] QB 373); the assets of a pension scheme (McDonald v Horn); or the assets involved in the reorganisation of a life insurance business (In re Axa Equity & Law Life Assurance plc (No 1) [2001] 2 BCLC 447).
15. The judges started from the proposition that they must do nothing to inhibit the exercise of discretion as to costs which would be vested in the judge conducting the substantive hearing. At the time In re Beddoe was decided, Order LXV Rule 1 provided that:

"Subject to the provisions of the Acts and these rules, the costs of and incident to all proceedings in the Supreme Court, including the administration of estimates and trusts, shall be in the discretion of the Court or judge…"

16. The rule also provided for the right of an executor, administrator, trustee or mortgagee "who has not unreasonably instituted or carried on or resisted any proceedings" to costs out of the particular estate or fund to which he would be entitled according to the rules hitherto acted upon in the Chancery Division. It also made special provision for jury actions.
17. More recently RSC Order 62 rule 3 (3) provided that:

"(3) If the Court in the exercise of its discretion sees fit to make any order as to the costs of any proceedings, the Court shall order the costs to follow the event, except when it appears to the Court that in the circumstances of the case some other order should be made as to the whole or any part of the costs."

18. RSC Order 62 rule 6 was entitled "Cases where costs do not follow the event." It included in sub-rule (2) a revised version of the part of Order LXV Rule 1 which was concerned with Chancery practice. It also identified details of other occasions, not material for the purposes of the present judgment, in which the ordinary rule was not to be followed.
19. In McDonald v Horn [1995] ICR 685 Hoffmann LJ set out the general practice at p 694C-F:

"There are two relevant rules of court, both of which reflect well-established principles. The first is RSC Ord 62, r 3(3):

'If the court in the exercise of its discretion sees fit to make any order as to the costs of any proceedings, the court shall order the costs to follow the event, except when it appears to the court that in the circumstances of the case some other order should be made as to the whole or any part of the costs.'

This rule reflects a basic rule of English civil procedure, namely that, as Lord Halsbury LC said in Civil Service Co-operative Society v. General Steam Navigation Co [1903] 2 KB 756, a successful litigant has a prima facie right to his costs. In cases like Ritter v Godfrey [1920] 2 KB 47 the Court of Appeal has laid down more detailed principles limiting the circumstances in which a successful party can be deprived of his costs or ordered to pay the costs of the other party. Ord 62, r 3(3) is a formidable obstacle to any pre-emptive costs order as between adverse parties in ordinary litigation. It is difficult to imagine a case falling within the general principle in which it would be possible for a court properly to exercise its discretion in advance of the substantive decision. So in Wallersteiner v Moir (No 2) [1975] QB 373, 403 Buckley LJ rejected an application for an order protecting the plaintiff, Mr Moir, from being ordered to pay the costs of the defendant, Dr Wallersteiner, irrespective of the outcome of the case:

'I have never known a court to make any order as to costs fettering a later exercise of the court's discretion in respect of costs to be incurred after the date of the order. I cannot think of any circumstances in which such an order would be justified.'"

The other rule to which Hoffmann LJ made reference was Order 62 Rule 6(2) (see para 18 above).

20. In Wallersteiner v Moir (No 2) Buckley LJ went on to set out (at p 404) his solution (with which Scarman LJ agreed) to the dilemma that arose in that case:

"… [T]here are circumstances in which a party can embark on litigation with a confident expectation that he will be indemnified in some measure against costs. A trustee who properly and reasonably prosecutes or defends an action relating to his trust property or the execution of the trusts is entitled to be indemnified out of the trust property. An agent is entitled to be indemnified by his principal against costs incurred in consequence of carrying out the principal's instructions… The next friend of an infant plaintiff is prima facie entitled to be indemnified against costs out of the infant's estate … It seems to me that in a minority shareholder's action, properly and reasonably brought and prosecuted, it would normally be right that the company should be ordered to pay the plaintiff's costs so far as he does not recover them from any other party. In all the instances mentioned the right of the party seeking indemnity to be indemnified must depend on whether he has acted reasonably in bringing or defending the action, as the case may be: see, for example, as regards a trustee, In re Beddoe, Downes v Cottam [1893] 1 Ch 557. It is true that this right of a trustee, as well as that of an agent, has been treated as founded in contract. It would, I think, be difficult to imply a contract of indemnity between a company and one of its members. Nevertheless, where a shareholder has in good faith and on reasonable grounds sued as plaintiff in a minority shareholder's action, the benefit of which, if successful, will accrue to the company and only indirectly to the plaintiff as a member of the company, and which it would have been reasonable for an independent board of directors to bring in the company's name, it would, I think, clearly be a proper exercise of judicial discretion to order the company to pay the plaintiff's costs. This would extend to the plaintiff's costs down to judgment, if it would have been reasonable for an independent board exercising the standard of care which a prudent business man would exercise in his own affairs to continue the action to judgment. If, however, an independent board exercising that standard of care would have discontinued the action at an earlier stage, it is probable that the plaintiff should only be awarded his costs against the company down to that stage.

There is a well established practice in Chancery for a trustee who has it in mind to bring or defend an action in respect of his trust estate to apply to the court for directions: see In re Beddoe, Downes v Cottam [1893] 1 Ch. 557. If and so far as he is authorised to proceed in the action, the trustee's right to be indemnified in respect of his costs out of the trust property is secure. If he proceeds without the authority of an order of the court, he does so at his own risk as to costs. It seems to me that a similar practice could well be adopted in a minority shareholder's action."

21. See also Scarman LJ at p 407A-D, and in particular this passage (at A-B):

"The indemnity is a right distinct from the right of a successful litigant to his costs at the discretion of the trial judge; it is a right which springs from a combination of factors - the interest of the company and its shareholders, the relationship between the shareholder and the company, and the court's sanction (a better word would be 'permission') for the action to be brought at the company's expense. It is a full indemnity such as an agent has who incurs expense in the authorised business of his principal."

22. In McDonald v Horn, after explaining why it was appropriate to apply the same principles in favour of beneficiaries of a pension scheme who were concerned with alleged improprieties and breaches of trust by the pension fund trustees, Hoffmann LJ said at p 500 that there was a need for caution in making such orders. He went on to say:

"The court should not authorise any legal process until it has explored, as Vinelott J did in this case, the possibility of independent investigation by a person or persons acceptable to both parties. In the normal case I would not expect any proceedings to be authorised until such an independent investigation had been completed. It is unfortunate that in this case the proposal was unsuccessful. I have the impression from what I have seen of the evidence and the way the case was put by [counsel for the defendants] that the defendants have taken the view that, provided the fund was in surplus, the way in which it was invested and administered was none of the plaintiffs' business. This attitude is unacceptable: the whole fund is a trust fund, whatever may be the beneficial interests on a winding up, and the members are entitled to openness in the way it is run."

23. Practice in the Family Division has also departed from the "costs follow the event" principle in significant respects: and see CPR 44.3(2). The recent judgment of Rex Tedd QC, sitting as a deputy High Court judge, in C v FC (Children Proceedings: Costs) [2004] 1 FLR 362 brings together helpfully in one place a number of recent cases on this topic. These include Gojkovic v Gojkovic [1992] Fam 40; Sutton LBC v Davis (No 2) [1994] 1 WLR 1317; Keller v Keller and Legal Aid Board [1995] 1 FLR 259; and R v R (Costs: Child Case) [1997] 2 FLR 95. In the last of these cases Hale J was concerned to analyse the reasons why costs orders were generally not made in cases involving children.

4. Some distinctive features relating to costs in public law litigation

24. These then, were the prevailing trends over the years in private law litigation in the civil and family courts. There have been some distinctive features in the past, so far as the Crown and other public bodies were concerned. Official bodies, for instance, would often appear or intervene in public law proceedings on the basis that they were present to assist the court in an amicus curiae role, even if they were respondents in the proceedings, and in that capacity, in a court which traditionally ordered only one set of costs, it would neither apply for costs nor expect an order for costs to be made against it, even if its submissions favoured one side more than the other. Examples of this practice were recently given by Brooke LJ in R (Davies v Birmingham Deputy Coroner [2004] EWCA Civ 207, [2004] 3 All ER 543: justices, tribunals, coroners and the Central Arbitration Committee were cited as examples.
25. From time to time leave to appeal to the House of Lords was given to a public body like the Inland Revenue on terms that they would pay both sides' costs in the House of Lords and not seek to disturb the orders for costs made in the court below. Chief Constable of North Wales Police v Evans [1982] 1 WLR 1164 provides just one example of this practice. Similar orders have been made in this court recently in cases where the appellants wished to have a point of law authoritatively determined and might not have been granted permission to appeal in the ordinary course of things.
26. Sometimes, the Crown, when successful, does not apply for an order for costs in its favour: Liversidge v Anderson [1942] AC 206, 283 is a good example of this practice. In New Zealand Maori Council v Attorney-General of New Zealand [1994] 1 AC 466 the Privy Council went one step farther, and declined to make an order for costs against the unsuccessful appellants where they were not pursuing the proceedings out of any motive of private gain, but "in the interests of taonga which is an important part of the heritage of New Zealand", and the judgments in the Court of Appeal had left an undesirable lack of clarity in an important area of the law which it was important for the Privy Council to examine.
27. Less than two years later Lord Lloyd of Berwick, who as a member of the Privy Council in the New Zealand Maori case, went so far as to say, in Bolton MDC v Secretary of State for the Environment [1995] 1 WLR 1176, 1178:

"As in all questions to do with costs, the fundamental rule is that there are no rules. Costs are always in the discretion of the court, and a practice, however widespread and longstanding, must never be allowed to harden into a rule."

Against that background the House enunciated new guidance about the incidence of costs in multi-party planning appeals.

5. Protective costs orders: the historical setting

28. The present appeal is concerned not with the incidence of costs in private law civil or family litigation or with statutory (or other) appeals, but with the incidence of costs in a judicial review application at first instance. Over the last 20 years there has been a growing feeling in some quarters, both in this country and in common law countries abroad which have adopted the "costs follow the event" regime, that access to justice is sometimes unjustly impeded if there is slavish adherence to the normal private law costs regime described by Buckley LJ in Wallersteiner v Moir (No 2) and by Hoffmann LJ in McDonald v Horn.
29. The radical overhaul of RSC Order 53 in 1977, followed by the enactment of section 51 of the Supreme Court Act 1981 ("the 1981 Act"), were accompanied by the liberalisation of the rules as to standing in judicial review cases by the House of Lords in IRC v National Federation of Self-Employed and Small Businesses Ltd [1982] AC 617. Lord Diplock justified the modern approach to standing and identified the purpose of judicial review (to vindicate the rule of law and to get unlawful conduct stopped) in these words (at p 644E-G):

"It would, in my view, be a grave lacuna in our system of public law if a pressure group, like the federation, or even a single public-spirited taxpayer, were prevented by outdated technical rules of locus standi from bringing the matter to the attention of the court to vindicate the rule of law and get the unlawful conduct stopped. The Attorney-General, although he occasionally applies for prerogative orders against public authorities that do not form part of central government, in practice never does so against government departments. It is not, in my view, a sufficient answer to say that judicial review of the actions of officers or departments of central government is unnecessary because they are accountable to Parliament for the way in which they carry out their functions. They are accountable to Parliament for what they do so far as regards efficiency and policy, and of that Parliament is the only judge; they are responsible to a court of justice for the lawfulness of what they do, and of that the court is the only judge."

30. The members of the House, however, considered that questions of standing should not be treated as a preliminary issue but should be decided at the substantive hearing in the legal and factual context of the whole case (see 630 D-E, 645D and 654A-B). At that time there was no provision by which a respondent was afforded the opportunity now given by CPR 54.8 of making representations to the court in every case before leave was granted.
31. In 1989 Toohey J, a member of the High Court of Australia, raised a quite new question in his address to a conference of the Australian National Environmental Law Association. He observed that the awarding of costs was a factor that loomed large in any consideration to institute litigation, and that addressing the issue of standing on its own in what he called "public interest" cases was grossly insufficient without considering issues of costs:

"Relaxing the traditional requirements of standing may be of little significance unless other procedural reforms are made. There is little point in opening doors to the courts if litigants cannot afford to come in… The fear, if unsuccessful, of having to pay the costs of the other side - with devastating consequences to the individual or environmental group bringing the action - must inhibit the taking of the case to court."

32. This early stirring of the germ which was to become known as a protective costs order did not feature in the suggestions about costs which were canvassed by the Law Commission in its Consultation Paper No 126, Administrative Law: Judicial Review and Statutory Appeals (1993) at paras 11.1 – 11.14. The consultation process threw up suggestions that judges should have power to award costs out of central funds in civil cases, particularly where there was no other source from which they could be paid and the interests of justice so required, and that the court should be empowered to grant legal aid for the application for leave or for the substantive hearing. The Commission contented itself by recommending that costs should be available from central funds (a) in favour of a successful party, at the judge's discretion or (b) in favour of an unsuccessful applicant where a case had been allowed to proceed to a substantive hearing on the basis of either a public interest challenge or for the purpose of seeking an advisory declaration.
33. The Government did not accept either of these recommendations. This lacuna in the court's ability to do justice led this court to order a coroner to pay the costs of a successful claimant when directing a new inquest into the death of his wife when there was no other means of indemnifying him for the expense to which he had been put, even though the coroner was a judicial officer who had conducted himself impeccably (see R v Inner London Coroner ex p Touche [2001] EWCA Civ 383, [2001] QB 1206 at paras 54-49 and the discussion in R (Davies) v Birmingham Deputy Coroner (see para 24 above) at paras 38-48.
34. In 1990 a judgment of Schiemann J in R v Secretary of State for the Environment ex p Rose Theatre Trust Co [1990] 1 QB 504 raised questions as to the extent to which the rules on standing had or should be liberalised. In order to put the matter beyond doubt the Law Commission recommended that an application should not be allowed to proceed to a substantive hearing unless the court was satisfied that the applicant had been or would be adversely affected, or the High Court considered that it was in the public interest for the applicant to make the application (see Law Com No 226 (1994) at paras 5.16 – 5.22). In the event section 31 of the 1981 Act has not been amended in the way the Commission suggested, but the Administrative Court is now very willing to permit "public interest challenges" in appropriate cases.
35. In R v Foreign Secretary ex p World Development Movement Ltd [1995] 1 WLR 386 the Divisional Court permitted the applicants to challenge the legality of the expenditure of aid and trade provision on the Pergau Dam project in Malaysia pursuant to the Overseas Development and Co-operation Act 1980. After referring to the increasingly liberal approach to standing that the courts had adopted in the previous 12 years, Rose LJ said, at p 395F, that the merits of the challenge were "an important, if not dominant, factor when considering standing." He then cited Professor Wade in Administrative Law (7^th edition) (1994) at p 712:

"{t}he real question is whether the applicant can show some substantial default or abuse, and not whether his personal rights or interests are affected."

36. Apart from the merits of the case, Rose LJ identified five other considerations which militated towards the court's decision that the applicants had a sufficient interest to challenge the lawfulness of this expenditure:

i) The importance of vindicating the rule of law;

ii) The importance of the issue raised;

iii) The likely absence of any other responsible challenger;

iv) The nature of the breach of duty against which relief was sought;

v) The prominent role of the applicants in giving advice, guidance and assistance with regard to aid.

37. We were shown three other decisions between 1995 and 1998 by judges in what is now called the Administrative Court which demonstrate how the role of a public authority in public law proceedings and the way in which the court exercises its discretion as to costs in cases containing a genuine public interest element present significant differences from the usual practice in private law litigation.
38. Thus in Coventry City Council v Finnie (1997) 29 HLR 658 Scott Baker J held that the grant of an injunction in favour of a local authority performing law enforcement duties did not necessarily carry with it a cross-undertaking on damages of a type that is familiar in private litigation.
39. In R v Secretary of State for the Environment ex p Shelter [1997] COD 49 Carnwath J refused to make a costs order against Shelter on the grounds that:

(i) there were already pending before the court a sequence of individual cases raising precisely the same issue;

(ii) the legal question raised was of genuine public interest;

(iii) the applicant's involvement had assisted the court in determining the issue speedily; and

(iv) had the matter been determined in separate proceedings, it was likely that any applicant would have been legally aided, and thus the burden of his/her costs would have fallen upon the tax payer and the respondent would not have obtained an order for his costs.

40. In R v Merthyr Tydfil Crown Court ex p the Chief Constable of Dyfed Powys Police (COT 9^th November 1998) Lightman J quashed an order for costs that had been made against the chief constable in the Crown Court in favour of a successful appellant at a licensing appeal. He said that the "costs follow the event" principle did not apply in a case where the police were merely placing before the court matters which it was material for the court to know. Such an order could only be made if it could be shown that the police's position had been totally unreasonable or prompted by some improper motive.
41. Some of the authorities that we have considered thus far demonstrate a trend towards protecting litigants, who reasonably bring public law proceedings in the public interest, from the liability to costs that falls, as a general rule, on an unsuccessful party. The making of a PCO was a substantial further step in the same direction.
42. As early as 1989 the Ontario Law Reform Commission suggested that the following criteria might be adopted by a court considering whether to make a PCO:

(i) The litigation must raise issues of importance beyond the immediate interests of the parties;

(ii) The plaintiff must have no personal, proprietary or pecuniary interest in the outcome, or if such an interest does exist, it clearly does not justify the litigation economically;

(iii) The litigation does not present issues which have previously been judicially determined against the same defendant;

(iv) The defendant must have a clearly superior capacity to bear the costs of the proceedings.

43. In 1995 the Australian Law Reform Commission addressed the same issues in its report entitled "Costs shifting - who pays for litigation" (ALRC 75 (1995)). In Chapter 13 the Commission discussed what it called public interest costs orders. It observed that public interest litigation was an important mechanism for clarifying legal issues to the benefit of the general community (para 13.1), and commented that what it described as the costs indemnity rule generally had a deterrent effect on this type of litigation (para 13.8; for the benefits the Commission ascribed to litigation of this type see para 13.6). It recommended that courts or tribunals should have power to make a public interest costs order at any stage of the proceedings, and suggested criteria which should be taken into account when determining what type of order to make. It might, for example, direct that each party should bear his or her own costs, or that

"the party applying for the public interest costs order, regardless of the outcome of the proceedings, shall

not be liable for the other party's costs

only be liable to pay a specified proportion of the other party's costs

be able to recover all or part of his or her costs from the other party."

6. Protective Costs Orders in the High Court and the Court of Appeal

44. In R v Lord Chancellor ex p CPAG [1999] 1 WLR 347 Dyson J heard two applications for PCOs at the same time. The Child Poverty Action Group sought a PCO to enable it to continue judicial review proceedings for the purpose of requiring the Lord Chancellor to reconsider the way he exercised his power under section 14(2) of the Legal Aid Act 1985 in relation to the extension of legal aid to cover at least some cases before social security tribunals and commissioners. At the same time Amnesty International UK sought a similar order in relation to its legal challenge to a decision made by the Director of Public Prosecutions not to prosecute two individuals for possession of an electro-shock baton without the requisite licence.
45. It was conceded by both respondents that the court possessed jurisdiction to make a PCO, but there was no agreement as to the principles on which the jurisdiction should be exercised. It was common ground, following McDonald v Horn, that a PCO would not be available in a private law action. Dyson J said (at p 349F) that the main question of principle he had to determine was whether different considerations of public policy applied in cases which could aptly be characterised as "public interest" challenges.
46. After analysing the arguments, Dyson J said that it was only in the most exceptional circumstances that the discretion to make a PCO should be exercised in a case involving a public interest challenge. He went on to say (at p 358 C-E) that:

(i) the court must be satisfied that the issues raised are truly ones of general public importance;

(ii) the court must be satisfied, following short argument, that it has a sufficient appreciation of the merits of the claim that it can be concluded that it is in the public interest to make the order;

(iii) the court must have regard to the financial resources of the applicant and respondent, and the amount of costs likely to be in issue;

(iv) the court will be more likely to make an order where the respondent clearly has a superior capacity to bear the costs of the proceedings than the applicant, and where it is satisfied that, unless the order is made, the applicant will probably discontinue the proceedings, and will be acting reasonably in so doing.

47. On the facts of the two cases before him, in the CPAG case he had his doubts about (i) above, and was unable to assess the merits sufficiently to be able to arrive at the conclusion required by (ii) above. (iii) and (iv) appeared to be satisfied. The Amnesty case failed to satisfy any of the tests he had laid down.
48. In Hodgson v Imperial Tobacco Ltd [1998] 1 WLR 1056 Lord Woolf MR noted at p 1068, without comment, the order made by Dyson J in ex p CPAG when he said that the court was not suggesting that the court had no power to make a debarring order of the type sought by the plaintiffs in that case. In R v Hammersmith and Fulham LBC ex p CPRE (CAT 26^th October 1999) Richards J applied Dyson J's tests when declining to make a PCO. He said, in passing, that it did not seem to him that the overriding objective laid down in the Civil Procedure Rules (which were now in force) affected or undermined those principles:

"…I accept that in exercising discretion with regard to costs … I should seek to give effect to the overriding objective and should have particular regard to the need, so far as practicable, to ensure that the parties are on an equal footing and that the case is dealt with in a way which is proportionate to the financial position of each party. Those aspects of the overriding objective seem to me to be embedded in any event in the principles laid down in ex p CPAG."

49. In R v The Prime Minister ex p CND [2002] EWHC 2712 (Admin) the Divisional Court (Simon Brown LJ and Maurice Kay J) made a PCO in favour of the claimants to the extent that any award of costs against them should be capped in the sum of £25,000. They were seeking an advisory declaration to the effect that UN Security Council Resolution 1441 did not authorise the use of force against Iraq in the event of a breach of that resolution. Although the order was being sought before permission to apply for judicial review had been granted, Simon Brown LJ found that all the CPAG tests were satisfied, and that it was right to afford the claimants the relatively limited security that the order would afford them. Maurice Kay J, agreeing, suggested a procedure by which applications of this kind should be made in future.
50. In R (Refugee Legal Centre) v Home Secretary [2004] EWCA Civ 1296 this court set a day aside to consider whether a PCO should be granted in favour of the claimants in relation to a substantive appeal in a matter in which they had been protected by an undertaking by the Home Office not to seek an order for costs against them at first instance. In the event the court made a PCO by consent. The previous week Brooke LJ had made a PCO in their favour to cover the PCO hearing before the full court, on the clear understanding that they would not be looking for their costs against the Secretary of State if they were to win. The claimant's lawyers had been acting pro bono throughout, and their clients were an independent not-for-profit charity which had overall responsibility for ensuring the delivery of quality legal services to those seeking human rights protection. What was under challenge was the fairness of the very streamlined new arrangements for processing asylum-seekers' claims at Harmondsworth (see the judgment at [2004] EWCA Civ 1239 at [3]-[9] for a description of the scheme) which clearly warranted the scrutiny of this court.
51. Although the Centre's appeal ultimately failed, it served an important public purpose because it enabled this court to make it clear that the scheme required to be backed by a clear stated policy which recognised that it would be unfair not to enlarge the standard timetable in a variety of instances (see the judgment of Sedley LJ at [2004] EWCA Civ 1481 at [18]). Although the court held that the arrangements were inherently fair, it was critical of the fact that the Home Office had formulated no test or standard for the adaptation of the abbreviated timetable to individual needs (see paras 16-19 and 24-25).
52. This is a good example of the way in which PCOs can be harnessed in cases of general public importance where it is in the public interest for the courts to review the legality of novel acts by the executive in a context where it is unreasonable to expect that anyone would be willing to bear the financial risks inherent in a challenge. In his earlier judgment Brooke LJ said (at paras 18-20):

"18. … [Counsel for the |Home Office] submitted that there was no reason why the ordinary judicial review scenario should not apply whereby individual asylum seekers who feel that they are being treated unfairly should obtain a legal aid certificate and bring proceedings in which there will be no prospect of the Home Secretary recovering his costs even if the matter has not been settled by compromise before it came to a hearing.

19. If the challenge is to systemic unfairness, as this challenge is, it would be necessary to have 10 or 12 individual cases for the court to look at in order to test whether the system is generally unfair. That would mean 10 different legal aid certificates, perhaps 10 different solicitors' firms, and counsel and solicitors acting on terms on which they now act in legal aid cases, so that if they win the Home Office would be liable to pay the costs of lawyers acting in a legal aid case who are successful at reasonable and not restricted rates of pay. But [counsel] submits that this is the appropriate way to go forward and not by the route that the Refugee Legal Centre has selected.

20. I told [counsel] during the course of argument that I was less than convinced that this would be a satisfactory way of resolving issues relating to systemic unfairness if they can be substantiated by the evidence."

7. Recent developments in Ireland, Canada and Australia

53. There are some developments overseas of which we should take note. In Village Residents' Association Ltd v An Bord Pleanala (No 2) [2000] 4 IR 321 Laffoy J was concerned with (and dismissed) the first application for a PCO in the High Court of Ireland. She said that she was satisfied that there was jurisdiction to make such an order, but that it was difficult in the abstract to identify the type or types of cases in which the interests of justice would require the court to deal with costs in the manner indicated by a PCO and it would be unwise to attempt to do so.
54. She said that the principles set out in Dyson J's judgment in ex p CPAG seemed to meet the fundamental rubric that the interests of justice should require a PCO to be made. When the Irish Law Reform Commission visited the question of PCOs in its 2004 report on Judicial Review Procedure (LRC 71-2004), it recommended that this jurisdiction should be exercised only in exceptional circumstances (which it did not attempt to define) "and that where any doubt exists the court should instead simply indicate the approach to be taken in relation to costs at the conclusion of the judicial review proceedings, while not committing itself absolutely on the issues." Because the making of a PCO pre-dated the determination of fact at the trial, they carried an inherent risk that an inappropriate order might be made.
55. In British Columbia (Minister of Forests) v Okanagan Indian Band (2003) 114 CCR 2d 108 the Supreme Court of Canada, by a 6-3 majority, went one step further and made an order directing the respondents to pay the costs of the appellants as the proceedings went on, on a strictly controlled basis. What was in issue was a challenge by Indian Bands to a prohibition on logging on their lands without prior authorisation. They asserted aboriginal title to the land in question and complained of a breach of their constitutionally protected aboriginal rights.
56. LeBel J, giving the judgment of the majority of the Supreme Court, said that concerns about access to justice and the desirability of mitigating severe inequality between litigants featured prominently in the rare cases in which interim costs orders were made. The power to order interim costs was inherent in the nature of the equitable jurisdiction as to costs, in the exercise of which the court might determine at its discretion when and by whom costs were to be paid. The following principles were identified in the judgment:

(i) The party seeking the order must be impecunious to the extent that without such an order that party would have been deprived of the opportunity to proceed with the case (para 36);

(ii) The claimant must establish a prima facie case of sufficient merit to warrant its pursuit (para 36);

(iii) Public law cases, as a class, were different from ordinary civil disputes, and the case must fall into a sub-category where the special circumstances that justified an award of interim costs were related to the public importance of the questions at issue in the case (para 38);

(iv) It was for the judge at first instance to determine whether a particular case, which might be classified as special by its very nature as a public interest case, was special enough to rise to the level where the unusual measure of ordering costs would be appropriate (para 38).

57. In the Canadian context LeBel J said (at para 39) that it was possible (although still unusual) for costs to be awarded in favour of the unsuccessful party in a dispute between the government and an individual Charter claimant of limited means if the court considered that this was necessary to ensure that ordinary citizens would not be deterred from bringing important constitutional arguments before the courts. This practice attenuated the concerns that might otherwise arise about prejudging the issues at an interim stage. He concluded this part of his judgment by saying (at para 40):

"With these considerations in mind, I would identify the criteria that must be present to justify an award of interim costs in this kind of case as follows:

1. The party seeking interim costs genuinely cannot afford to pay for the litigation, and no other realistic option exists for bringing the issues to trial – in short, the litigation would be unable to proceed if the order were not made.

2. The claim to be adjudicated is prima facie meritorious, that is, the claim is at least of sufficient merit that it is contrary to the interests of justice for the opportunity to pursue the case to be forfeited just because the litigant lacks financial means.

3. The issues raised transcend the individual interests of the particular litigant, are of public importance, and have not been resolved in previous cases."

58. Although these were necessary conditions that had to be met, it was for the court to determine in the exercise of its discretion whether the particular case was such that the interest of justice would be best served by making the order in a particular case. LeBel J went on (at para 41) to explain how the implementation of such an order, if made, should be carefully reviewed over the course of the proceedings to ensure that concerns about access to justice were balanced against the need to encourage the reasonable and efficient conduct of litigation, which was also one of the purposes of costs awards.
59. Five years earlier, in Oshlack v Richmond River Council [1998] HCA 11, a majority of 3-2 in the High Court of Australia restored the refusal of a judge at first instance to order costs in favour of a council who were the successful respondents to a challenge to a planning consent. The appellant had been concerned about the habitat of the endangered Koala, and complained about the absence of any fauna impact statement before the consent was granted. The judge considered that there were "sufficient special circumstances to justify a departure from the ordinary rule as to costs". These were to be found in the following considerations:

(i) The appellant had nothing to gain from the litigation "other than the worthy motive of seeking to uphold environmental law and the preservation of endangered fauna";

(ii) A significant number of members of the public shared the appellant's stance, so that in that sense there was a public interest in the outcome of the litigation;

(iii) The challenge had raised and resolved significant issues as to the interpretation and future administration of statutory provisions relating to the protection of endangered fauna and the present and future administration of the development consent in question, which had implications for the council, the developer and the public.

60. In that case the minority (Brennan CJ and McHugh J) were influenced by the difficulty in identifying criteria for "public interest litigation" such as would justify the courts, in the absence of legislation, in identifying particular cases in which a successful litigant was nevertheless deprived of its costs (see paras 1-3 and 90-97).
61. In a joint judgment Gaudron and Gummow JJ challenged the proposition that rules of practice applicable in other species of litigation had hardened so much that they looked like rules of law and thus rendered the matters which the judge had taken into account irrelevant to the exercise of the discretion as to costs conferred on him by section 69 of the Environmental Planning and Assessment Act 1979 (NSW) (see paras 36-47).
62. In the course of their judgment they showed (at para 33) how in England the First Report of the Commissioners and the Judicature (1869-70), vol 25 at p 15, had compared the "full power over the costs" in the Court of Chancery, the Court of Admiralty, and the Courts of Probate and Divorce with the "absence of this power" in the Courts of Common Law, which "often occasioned injustice". For Chancery practice they cited the judgment of Fry LJ, giving the judgment of this court in Andrews v Barnes (1888) 39 Ch D 133, 138:

"The Jurisdiction of the Lord Chancellor in costs was essentially different from that at common law. 'The giving of costs in equity', said Lord Hardwicke in Jones v Coxeter (1742) Atk 400' is entirely discretionary and is not at all conformable to the rule at law.' 'Courts of Equity', said the same great Judge in another case 'have in all cases done it' (ie dealt with costs) 'not from any authority (ie as we understand, from any statutory or delegated authority) – but from conscience and arbitrio boni viri, as to the satisfaction on one side or other on account of vexation' (Corporation of Burford v Lenthall (1743) 2 Atk 551, 552)."

63. In his supporting judgment Kirby J referred to the broad language of the discretion conferred on the court by section 69 of the 1979 Act and pointed out that guidance afforded by appellate courts in general terms as to the considerations which the decision-maker can take into account should not confine him to "a rigidly mechanical approach" (para 134 (3)). Although there are "rules" or ordinary principles which will guide the donee of the power in the exercise of discretion, "they cannot extinguish the element of discretion. They must not be allowed to harden into rigid or inflexible rules" (para 134 (4)). On the proper interpretation of the powers conferred by the statute, the judge was entitled to take into account the considerations which influenced him when he made no order as to costs.

8. Protective Costs Orders: the governing principles and some practical guidance

64. Since the CPR came into force in this country in 1999 the court's jurisdiction as to costs has been governed by section 51 of the 1981 Act, which provides that:

"(1)   Subject to the provisions of this or any other enactment and to rules of court, the costs of and incidental to all proceedings in—

a) the civil division of the Court of Appeal;

b) the High Court; and

c) the county court

shall be in the discretion of the court.

(2)   Without prejudice to any general power to make rules of court, such rules may make provision for regulating matters relating to the costs of those proceedings …

(3)   The court shall have full power to determine by whom and to what extent the costs are to be paid."

and by the rules contained in CPR Parts 43-48 and the Practice Direction about Costs. CPR 44.3 is of particular importance:

"44.3(1) The court has discretion as to –

(a) whether costs are payable by one party to another;

(b) the amount of those costs; and

(c) when they are to be paid.

(2) If the court decides to make an order about costs –

(a) the general rule is that the unsuccessful party will be ordered to pay the costs of the successful party; but

(b) the court may make a different order.

…..

(4) In deciding what order (if any) to make about costs, the court must have regard to all the circumstances…"

65. CPR 3.2(m) gives the court an unqualified power to "take any other step or make any other order for the purpose of managing the case and furthering the overriding objective."
66. CPR 1.1 and 1.2 provide, so far as material, that:

"1.1(1) These Rules are a new procedural code with the overriding objective of enabling the court to deal with cases justly.

(2) dealing with a case justly includes, so far as is practicable –

(a) ensuring that the parties are on an equal footing;

(b) ensuring that [the case] is dealt with … fairly…

1.2 The court must seek to give effect to the overriding objective when it

(a) exercises any power given to it by the Rules; or

(b) interprets any rule."

67. In King v Telegraph Group Ltd [2004] EWCA Civ 613 this court called in aid (at paras 82-83) a combination of CPR 3.2(m) and CPR 1.1 when it identified the source of its power to make cost-capping orders at an early stage of civil proceedings. A relatively impecunious claimant in a libel action had sued the defendant newspaper group with the benefit of a conditional fee agreement ("CFA") (which might have borne a success fee as high as 100%) without the benefit of "after the event" insurance cover. Brooke LJ identified (in para 99) "the obvious unfairness" of these arrangements from the defendant's perspective and in paras 101-2 he set out the court's solution (with which Jonathan Parker and Maurice Kay LJJ agreed):

"101. In my judgment the only way to square the circle is to say that when making any costs capping order the court should prescribe a total amount of recoverable costs which will be inclusive, so far as a CFA-funded party is concerned, of any additional liability. It cannot be just to submit defendants in these cases, where their right to freedom of expression is at stake, to a costs regime where the costs they will have to pay if they lose are neither reasonable nor proportionate and they have no reasonable prospect of recovering their reasonable and proportionate costs if they win.

102. If this means, now that the amount at stake in defamation cases has been so greatly reduced, that it will not be open to a CFA-assisted claimant to receive the benefit of an advocate instructed at anything more than a modest fee or to receive the help of a litigation partner in a very expensive firm who is not willing to curtail his fees, then his/her fate will be no different from that of a conventional legally aided litigant in modern times. It is rare these days for such a litigant to be able to secure the services of leading counsel unless the size of the likely award of compensation justifies such an outlay, and defamation litigation does not open the door to awards on that scale today. Similarly, if the introduction of this novel cost-capping regime means that a claimant's lawyers may be reluctant to accept instructions on a CFA basis unless they assess the chances of success as significantly greater than evens (so that the size of the success fee will be to that extent reduced), this in my judgment will be a small price to pay in contrast to the price that is potentially to be paid if the present state of affairs is allowed to continue."

68. Miss Carss-Frisk QC, who appeared for the Secretary of State, did not contend that the court possessed no jurisdiction to make a PCO. This was an important concession, although in our judgment it was correctly made, because there is nothing in the House of Lords' interpretations of the wide words contained in section 51 of the 1981 Act or in CPR 44.3 to preclude the court from making such an order as to costs as affects only the parties to the case (as opposed to central funds) as it considers necessary in the interests of justice. The requirements of the interests of justice was a concept invoked twice by Laffoy J in the Village Residents Association case (see para 52 above); and the interests of justice were also invoked by LeBel J in the Okanagan Band case (see para 54 above). Lloyd LJ used the expression "where the interests of justice so require" in the Davies v Eli Lilley case (see para 10 above).
69. We are satisfied that there are features of public law litigation which distinguish it from private law civil and family litigation. The House of Lords identified one important difference in R v Home Secretary ex p Salem [1999] 1 AC 450 when Lord Slynn acknowledged (at pp 456-7) that the House possessed discretion to hear an appeal concerned with an issue involving a public authority as to a question of public law even when the parties to the appeal had ended the "lis" between them. He said that there must be a good reason in the public interest for doing so, and cited, as an example, a case:

"where a discrete point of statutory construction arises which does not involve detailed consideration of facts and where a large number of similar cases exists or are anticipated so that the issue will most likely need to be resolved in the near future."

70. The important difference here is that there is a public interest in the elucidation of public law by the higher courts in addition to the interests of the individual parties. One should not therefore necessarily expect identical principles to govern the incidence of costs in public law cases, much less the "arterial hardening" of guidance into rule which the majority of the High Court of Australia eschewed in the Oshlack case.
71. Miss Carss-Frisk was content with the CPAG guidelines, and encouraged us not to depart from them in any way. While we are in broad agreement with all but one of the guidelines, we consider that it is possible to reformulate them with greater precision and that we should suggest changes in the procedure whereby they are sought. Experience has shown that they are cumbersome to operate and that the achievement of justice is thwarted because they are so cumbersome. Thus in the Refugee Legal Centre case Brooke LJ had to hold a two-hour hearing to determine whether a PCO should be available to the claimants to protect them at a one-day hearing at which the full court was to consider the making of a PCO for the substantive appeal. And Dyson J's requirement that the court should have a sufficient appreciation of the merits of the claim after hearing short argument tends to preclude the making of a PCO in a case of any complexity.
72. Dyson J emphasised that the guidelines related to public interest challenges, which he defined at p 353. We believe that this definition can usefully be incorporated into the guidelines themselves. Dyson J said that the jurisdiction to make a PCO should be exercised only in the most exceptional circumstances. We agree with this statement, but of itself it does not assist in identifying those circumstances.
73. We endorse the first, third and fourth of the CPAG guidelines. We consider, however, that the second guideline needs to be recast. It commonly happens when a court has to take an important decision at an early stage of proceedings that it must do no more than conclude that the applicant's case has a real (as opposed to a fanciful) prospect of success, or that its case is "properly arguable". To place the threshold any higher is to invite heavy and time-consuming ancillary litigation of the type that disfigured the conduct of civil litigation 25 years ago. We realise that in CPR Part 54 the rule-maker prescribed no explicit criterion for the grant of permission to apply for judicial review, but we consider that no PCO should be granted unless the judge considers that the application for judicial review has a real prospect of success and that it is in the public interest to make the order.
74. We would therefore restate the governing principles in these terms:

1. A protective costs order may be made at any stage of the proceedings, on such conditions as the court thinks fit, provided that the court is satisfied that:

i) The issues raised are of general public importance;

ii) The public interest requires that those issues should be resolved;

iii) The applicant has no private interest in the outcome of the case;

iv) Having regard to the financial resources of the applicant and the respondent(s) and to the amount of costs that are likely to be involved it is fair and just to make the order;

v) If the order is not made the applicant will probably discontinue the proceedings and will be acting reasonably in so doing.

2. If those acting for the applicant are doing so pro bono this will be likely to enhance the merits of the application for a PCO.

3. It is for the court, in its discretion, to decide whether it is fair and just to make the order in the light of the considerations set out above.

75. A PCO can take a number of different forms and the choice of the form of the order is an important aspect of the discretion exercised by the judge. In the present judgment we have noted:

i) A case where the claimant's lawyers were acting pro bono, and the effect of the PCO was to prescribe in advance that there would be no order as to costs in the substantive proceedings whatever the outcome (Refugee Legal Centre);

ii) A case where the claimants were expecting to have their reasonable costs reimbursed in full if they won, but sought an order capping (at £25,000) their maximum liability for costs if they lost (CND);

iii) A case similar to (ii) except that the claimants sought an order to the effect that there would be no order as to costs if they lost (CPAG);

iv) The present case where the claimants are bringing the proceedings with the benefit of a CFA, which is otherwise identical to (iii).

76. There is of course room for considerable variation, depending on what is appropriate and fair in each of the rare cases in which the question may arise. It is likely that a cost capping order for the claimants' costs will be required in all cases other than (i) above, and the principles underlying the court's judgment in King at paras 101-2 will always be applicable. We would rephrase that guidance in these terms in the present context:

i) When making any PCO where the applicant is seeking an order for costs in its favour if it wins, the court should prescribe by way of a capping order a total amount of the recoverable costs which will be inclusive, so far as a CFA-funded party is concerned, of any additional liability;

ii) The purpose of the PCO will be to limit or extinguish the liability of the applicant if it loses, and as a balancing factor the liability of the defendant for the applicant's costs if the defendant loses will thus be restricted to a reasonably modest amount. The applicant should expect the capping order to restrict it to solicitors' fees and a fee for a single advocate of junior counsel status that are no more than modest.

iii) The overriding purpose of exercising this jurisdiction is to enable the applicant to present its case to the court with a reasonably competent advocate without being exposed to such serious financial risks that would deter it from advancing a case of general public importance at all, where the court considers that it is in the public interest that an order should be made. The beneficiary of a PCO must not expect the capping order that will accompany the PCO to permit anything other than modest representation, and must arrange its legal representation (when its lawyers are not willing to act pro bono) accordingly.

77. In this jurisdiction we do not consider that a court would have any power to make the type of order which was made in the Okanagan Band case, whereby the defendants were obliged to finance the claimant's costs at first instance as the litigation proceeded. This would be to trespass into judicial legislation in a way which was proscribed by the House of Lords in the Steele Ford & Newton case (see para 11 above).
78. We consider that a PCO should in normal circumstances be sought on the face of the initiating claim form, with the application supported by the requisite evidence, which should include a schedule of the claimant's future costs of and incidental to the full judicial review application. If the defendant wishes to resist the making of the PCO, or any of the sums set out in the claimant's schedule, it should set out its reasons in the acknowledgment of service filed pursuant to CPR 54.8. The claimant will of course be liable for the court fee(s) for pursuing the claim, and it will also be liable for the defendant's costs incurred in a successful resistance to an application for a PCO (compare Mount Cook Land Ltd v Westminster City Council [2003] EWCA Civ 1346 at para 76(1)). The costs incurred in resisting a PCO should have regard to the overriding objective in the peculiar circumstances of such an application, and recoverability will depend on the normal tests of proportionality and, where appropriate, necessity. We would not normally expect a defendant to be able to demonstrate that proportionate costs exceeded £1,000. These liabilities should provide an appropriate financial disincentive for those who believe that they can apply for a PCO as a matter of course or that contesting a PCO may be a profitable exercise. So long as the initial liability is reasonably foreseeable, we see no reason why the court should handle an application for a PCO at no financial risk to the claimant at all.
79. The judge will then consider whether to make the PCO on the papers and if so, in what terms, and the size of the cap he should place on the claimant's recoverable costs, when he considers whether to grant permission to proceed. If he refuses to grant the PCO and the claimant requests that his decision is reconsidered at a hearing, the hearing should be limited to an hour and the claimant will face a liability for costs if the PCO is again refused. The considerations as to costs we have set out in paragraph 78 above will also apply at this stage: we would not expect a respondent to be able to demonstrate that proportionate costs exceeded £2,500. Although CPR 54.13 does not in terms apply to the making of a PCO, the defendant will have had the opportunity of providing reasoned written argument before the order is made, and by analogy with CPR 52.9(2) the court should not set a PCO aside unless there is a compelling reason for doing so. The PCO made by the judge on paper will provide its beneficiary with costs protection if any such application is made. An unmeritorious application to set aside a PCO should be met with an order for indemnity costs, to which any cap imposed by the PCO should not apply. Once the judge has made an order which includes the caps on costs to which we have referred, this will be an order to which anyone subsequently concerned with the assessment of costs will be bound to give effect (see CPR 44.5(2)).
80. We have not yet referred to the position of any party other than the defendant. If an interested party such as a developer in a planning dispute wishes to resist the making of a PCO, it is likely to be entitled to its costs attributable to that part of its acknowledgment of service that relates to this issue, subject to the same considerations as to the proportionate costs of any resistance we have set out in paragraphs 78 and 79 above. The judge should not normally allow more than one set of additional costs because he will expect different interested parties to make common cause on this issue. Similar considerations will apply to any application to set the PCO aside, although this should be a very rare event.
81. It follows that a party which contemplates making a request for a PCO will face a liability for the court fees, a liability (which should not generally exceed a proportionate total of £2,000 in a multi-party case) for the costs of those who successfully resist the making of a PCO on the papers, and a further liability (which should not generally exceed a proportionate total of £5,000 in a multi-party case) if it requests the court to reconsider an initial refusal on the papers at an oral hearing. We hope that the Civil Procedure Rules Committee and the senior costs judge may formalise these principles in an appropriate codified form, with allowance where necessary for cost inflation in due course.

9. Corner House and the ECGD

82. We now turn to consider the facts.
83. The Export Credit Guarantee Department ("ECGD") of the Department of Trade and Industry ("DTI") helps to provide finance or security to UK exporters in the field of international trade. Its operation and powers are governed by the Export and Investment Guarantees Act 1991 ("the 1991 Act"). It currently supports trade worth around £3 billion per annum. Exporters favoured by ECGD backing include Rolls Royce, Airbus and British Aerospace. One of ECGD's published aims is to help to eradicate or minimise incidents of corruption and bribery in overseas trade.
84. In this context it conducted a 12-month review of its "mission and status" in the late 1990s, and in 2000 it issued a new procedural code which laid more stringent requirements on applicants for ECGD support. Indeed, this code was tougher than that recommended by the Organisation for Economic Co-operation and Development ("OECD") which was also interesting itself in anti-corruption processes.
85. The development of this new code had not been preceded by any formal consultation process. In November 2000, however, the Cabinet Office published a new Code of Practice on Written Consultation. It said that the Committee on Standards in Public Life had drawn attention in its Sixth Report to the importance of consultation with a wide section of the public, without which the openness and accountability of government could be impaired and the dangers of privileged access magnified. The purposes of consultation had to be borne in mind throughout the development of a policy or a service. The main objective of the new code was to improve decision-making. Effective consultation ought to ensure that so far as possible everyone concerned felt that they had had their say. In his Introduction the Prime Minister said that real changes in behaviour were needed, and that written consultation documents were simply one tool in the consultation process.
86. ECGD published its own consultation policy in September 2001. It set out its commitment to consult in three parts:

i) Formal consultation in relation to issues and policies likely to lead to a fundamental change in the way ECGD operates;

ii) Less formal consultation conducted in accordance with the principles set out in the paper in cases involving a major change to ECGD's existing policies and practices;

iii) The use and maintenance of the existing dialogue it had established with its stakeholders to obtain advice on minor issues.

87. ECGD said that it would consult prior to major decisions being taken, and give sufficient time for comments to be submitted and considered, together with an account of the view expressed and the main reasons for decisions finally taken (including an explanation of the reason why any significant alternative options had not been adopted).
88. All these processes were to involve ECGD's stakeholders, an expression defined to mean parties with a legitimate interest in its operations.
89. ECGD's policies for openness include the publication of the minutes of the Export Guarantee Advisory Council ("EGAC"). These minutes are generally published at the time of EGAC's next meeting, when the minutes of the previous meeting are approved. EGAC is a statutory body charged with the responsibility of advising the Secretary of State in the exercise of her functions relating to ECGD, and some of its minutes featured in the evidence in this case.
90. There were two non-governmental organisations ("NGOs") who took a particular interest in issues concerned with bribery and corruption in connection with the award of major contracts in the international market. One of these was called Transparency International (UK) Ltd ("TI"). The other was Corner House Research ("Corner House").
91. Corner House was incorporated a few years ago as a non-profit making company limited by guarantee,. It has a particular interest and expertise in examining the incidence of bribery and corruption in international trade. In this context it has had a long-standing interest in the role of export credit agencies.
92. In June 2003 it published a 79-page booklet by Dr Susan Hawley entitled "Turning a Blind Eye: Corruption and the UK Export Credit Guarantee Department". This was an expanded version of a report that was published two years earlier. Corner House described itself in the 2003 booklet as

"a UK-based advocacy, research and solidarity group that aims to support the sustainable use of resources and the growth of a vibrant, democratic, equitable and non-discriminatory civil society in which communities have control over the resources and decisions that affect their lives and means of livelihood, as well as the power to define themselves rather than be defined solely by others."

93. Section Two of the report contained case studies of nine ECGD-backed contracts in the 1990s. These studies led the author to conclude that throughout that decade the ECGD had displayed the following weaknesses:

i) A persistent failure to take notice of corruption allegations and a deep reluctance to investigate them;

ii) Inadequate investigatory procedures;

iii) An unwillingness to pass on corruption allegations to the appropriate external investigatory authorities;

iv) Inadequate due diligence regarding the potential for corruption in the projects it backed, coupled with complete disregard for international concerns about corruption in countries in which it supported projects;

v) Inadequate vetting of UK companies and inadequate due diligence regarding consortia, partners and agents used by UK companies;

vi) Lack of openness and accountability regarding whether it had backed certain products.

94. In Section Three Dr Hawley described the changes introduced by ECGD in 2000. While welcoming these changes, she said it was perhaps too early to tell how much impact they were having. Experience showed that it could take several years for evidence of corruption to emerge, and several more for it to be investigated, let alone brought to court. She adduced evidence to suggest that since its new procedures had been introduced, ECGD had backed at least one project that had become shrouded in significant corruption allegations. She added that Corner House had learned of another case in which a UK company was alleged to have paid bribes on a project backed by ECGD since its new warranty procedures had come into force.
95. While accepting that ECGD's new anti-corruption measures were a vast improvement on its previous laissez-faire approach to corruption, she said that there was still significant room for improvement. She suggested four steps in particular which ECGD ought to take. She also made a series of detailed and closely reasoned recommendations for the future. The main body of the report ended like this:

"Since [the new measures in late 2000] were introduced, the ECGD has backed at least one project with a corruption problem, and there are some serious weaknesses in the new measures. The ECGD must urgently re-address the issue of corruption and make some qualitative leaps forward. If not, more developing and transition countries are likely to suffer from over-priced products and greater debt burdens, and the UK's reputation in tackling corruption will be tarnished."

96. The report contained a two-page response from ECGD. It said it would prepare a more detailed response in due course to the criticisms contained in the report. In these initial comments it remarked that it had always considered bribery and corruption to be unacceptable in the conduct of international business. Its rigorous scrutiny procedures had been developed over many years, and in September 2000 major improvements had been made to strengthen their checks and balances. Since then it had made further amendments in order to maintain or enhance their rigour.
97. It said that it had been aware of the allegations contained in Dr Hawley's historical case-studies before it had overhauled its procedures. The recommendations made in this report provided some further suggestions which it would consider in due course, along with its own continuing review of this area. After setting out details of the measures it had already introduced and asserting its full commitment to greater transparency and accountability, it ended its response by saying that while it was satisfied with its current stance in respect of bribery and corruption issues, it was by no means complacent. It was committed to maintain and enhance its already rigorous standards. It welcomed the publication of the report as a stimulus to further discussion. A little earlier, after describing how it now published details of cases with potentially high impacts that it was actively examining, together with a list of the guarantees that it issued in support of UK exporters, it said:

"These actions have been taken in responses to consultations with stakeholders, including NGOs and customers, and such consultations now play a major role in our external relations. We are regularly in contact with Corner House, and co-operated extensively in providing information and clarification during the drafting of the report."

98. In her witness statement Dr Hawley described how Corner House asked ECGD on several occasions between June and November 2003 when its detailed response would be forthcoming. It appears that what was described as a "comprehensive reply" was discussed at EGAC's September 2003 meeting, when ECGD justified the fullness of its response by saying that the Corner House report was in the public domain and that this was being treated as a public affairs issue. It described Corner House as an influential group with links to much larger organisations and the media, and said that it had now begun to engage more closely with them. It added that the ECGD response, with which EGAC was generally satisfied, focused on the recommendations in the Corner House report, as opposed to the allegations it contained.
99. On 13^th November 2003 ECGD's External Affairs Manager told Dr Hawley that ECGD was holding a meeting the following week to discuss the changes it was proposing to make to its application form and procedures. He said that he would reflect the outcome of this meeting in the draft response, which he would put up for approval shortly after that meeting. EGAC's January 2004 minutes then show that the Minister of Trade asked ECGD to strengthen this response.

10. ECGD's new procedures in March 2004 and the changes then made to them

100. No such response ever saw the light of day. The next thing that happened was that, without any prior consultation with anyone, on 4^th March 2004 ECGD sent a letter to its customers announcing new and improved anti-bribery and corruption procedures. These were to come into effect on 1^st May. It said that these new measures were introduced to reflect lessons it had learned, and to ensure that it continued to play its part in the Government's drive to root out wrongdoing in international business transactions.
101. The nature of this initiative was described in a Press Release issued on 1^st April 2004. This contained the following statement by the Minister of Trade:

"These new measures ensure that the Government continues to play its part in rooting out wrongdoing in international business transactions. ECGD is at the forefront of Export Credit Agencies which are keen to eliminate unethical and illegal practices. And this further package of measures means the Department is being even more effective.

Bribery is not only wrong, it is bad for business. A culture of corruption is a disincentive to trade and investment and payment of bribes just makes the corrupt officials worse.

This is a balanced package. It will be to the ultimate benefit of all UK companies."

102. The Press Release identified five distinct features of the new measures:

"ECGD will:

i) Obtain additional information from applicants to ensure that no improper payments involving agents have been made to win contracts;

ii) Have greater rights to inspect exporters' documents relating to winning contracts and any payments made to agents;

iii) Require applicants to provide ECGD with a copy of their Codes of Conduct if they have one, and to sign a declaration that they will not engage in corrupt activity and will take action against anyone found guilty of such. Applicants must also show that they have precautions in place to prevent corrupt activity and monitor compliance with their Codes of Conduct (or similar procedures);

iv) Extend the range of various declarations regarding corruption to include affiliates - ie any company which is a member of the same group of companies, or that is a party to any joint venture or consortium - as well as directors and employees in those companies;

v) Require applicants to warrant that neither they nor to the best of their knowledge, their affiliates, have been convicted of, or admitted to, an offence of money laundering."

103. ECGD would also remind applicants of their legal obligations – all new ECGD application forms would contain a statement of the UK laws on bribery, corruption and money-laundering – and tell them that allegations of bribery and corruption and money-laundering would be referred to the appropriate authorities, including the National Criminal Intelligence Service.
104. In a letter dated 20^th March 2004 Dr Hawley asked the Minister of Trade a number of questions about French investigations into bribery allegations in Nigeria. A consortium which included a French partner had secured a valuable contract in that country, and in the autumn of 2003 a Paris-based investigation into bribery had attracted the attention of the British Press. A UK subcontractor, backed by ECGD, was involved in some parts of this contract. In his response dated 5^th May the minister said that Dr Hawley's letter raised a number of issues on a topic which he was keen to continue tackling. He said that someone carrying out corrupt practices would undoubtedly attempt to conceal them, but that ECGD and other Government Departments made strenuous efforts to deter and detect this sort of behaviour. After answering her questions and making it clear that ECGD did not know any detail of the French investigation due to the French rule of secrecy in these matters, he ended his letter by saying that he would be happy to meet her to discuss more broadly the subject of combating bribery and corruption, and to explain the steps ECGD was already taking in this area. Corner House then told ECGD that they would be interested in such a meeting.
105. In the event no such meeting ever took place. One opportunity arose in July, but as it happened the projected event – a two-hour afternoon seminar at ECGD involving the minister – was cancelled due to a clash of dates. ECGD's letter of invitation, addressed to Dr Hawley and a representative of TI contained this passage:

"You two represent our primary NGO partners on this topic and we are very keen that you form the core audience… The focus here is bribery and corruption… and I am keen to make this as constructive a meeting as possible…"

106. Dr Hawley believed that the purpose of this event was to go over what ECGD was doing and to provide an opportunity for the NGOs to raise further concerns about areas where they felt that ECGD had not gone far enough. In July, however, Corner House first learned that ECGD might be holding private discussions with its customers about the new application forms. On 21^st July, for instance, the minutes of EGAC's May meeting were published, and these showed that some major customers had expressed "negative views" about the changes. EGAC had resolved to revisit this subject at its September meeting.
107. On 29^th September Dr Hawley asked the head of ECGD's Business Principles Unit whether ECGD was likely to be changing its application form in the near future. She also asked about the date for the postponed meeting with the minister. On 11^th October ECGD's new external affairs manager replied. He said, rather vaguely, that a meeting on bribery and corruption was definitely still an option, and that they hoped to be able to sort something out soon. He then told her that ECGD had been having discussions with its customers and trade associations regarding its revised procedures on bribery and corruption. Nobody, he said, was questioning the need for effective measures. ECGD's aim was to ensure that its robust measures deterred bribery but did not place burdens on industry which might damage its competitiveness. He said that ECGD would be in a position to provide further information once these discussions were complete.
108. Matters then came to a head, so far as Corner House were concerned, in the third week of October. On 17th and 18^th October the Financial Times published articles by its political correspondent headed "UK ministers back down on bribery controls" and "Ministers 'massage' anti-bribery export credit guidelines after objections." The first of these articles revealed that the new rules had run into sustained opposition from British exporters, and that an action group formed by several business bodies had been locked in talks with ECGD since May. The Government was understood to have agreed to virtually all their demands. It had also secretly allowed some of the new anti-bribery rules to be flouted while it negotiated with industry. The article contained a brief statement by Dr Hawley urging ECGD to stick to its guns. The second article contained some of the justifications being put forward for the changes and a suggestion that ministers and business would both play down the effect of the changed wording when the revised rules were announced.
109. Corner House was now extremely concerned, because it appeared that ECGD was conducting a detailed and extensive consultation process with corporate customers, but was not prepared to hear the view of NGOs. The minutes of EGAC's July meeting, published on 21^st October, revealed that the CBI had been acting in a co-ordinating role for the complaints that had been received from exporters and banks. They described how a "major problem" had arisen in relation to ECGD's new requirements for the disclosure of agents' identities, payments and activities.
110. In the fourth week of October the new Minister for Trade gave written answers to a number of Parliamentary questions about these matters. He revealed that ECGD officials had held eight meetings since January 2004 with customers and representative trade associations to discuss its anti-corruption procedures. He also revealed that the Secretary of State and other ministers had been involved in the meetings with UK trade exporters and their trade bodies, and that these matters had been discussed on a number of occasions at such meetings. Since May 2004 ECGD had approved interim arrangements in respect of three transactions for the supply of Airbus aircraft, and one other transaction which could not yet be identified. The minister averred that these arrangements were consistent with international best practice, and that they incorporated elements of ECGD's previous anti-bribery and corruption procedures which were themselves robust. He added that informal representations had also been made by a number of NGOs. In a later Parliamentary answer on 15^th November it was said that representatives of British Aerospace and Rolls Royce had been present at six of these meetings, and representatives of Airbus at five.
111. Dr Hawley said in her witness statement that she was unclear what was meant by the reference to representations by NGOs. Corner House had certainly been given no opportunity to make any representations. This, too, was the burden of her complaint when she wrote to the minister on 27^th October praising the new procedures introduced in May and saying that Corner House was very concerned to learn that various changes were likely to be made to them. She added:

"We would like to register our disappointment that despite the fact that both ourselves and other anti-corruption groups such as Transparency International (UK) have been in discussion with ECGD about corruption and bribery issues for some years, there has been no effort to include our organisations in the discussions about the anti-corruption procedures. We believe that it would be in the interests of ECGD to ensure that any changes introduced to their procedures now are the result of broad consultation and not of untransparent meetings between the ECGD and a select group of customers and industry groups."

112. She said that the DTI and ECGD should be encouraging British exporters to be competitive through excellence, and not through weakening anti-corruption procedures. And she hoped that the promised meeting would be possible in the near future.
113. This letter, which was copied to ECGD and a number of other influential bodies within Government, had received no reply by the time that ECGD published details of its revised documentation a week later, which it said would be available from 8^th November. It explained that the changes were being introduced in response to "feedback" received from its customers. There is no Press Release equivalent to the April Press Release (see para 100 above) with the court's papers.
114. On 18^th November representatives of Corner House attended a meeting at ECGD's offices at which ECGD officials told them that the new procedures would not be suspended pending detailed consultation with NGOs. On the same day the minister wrote a brief letter to Dr Hawley expressing the hope that the meeting had helped to reassure her that ECGD's procedures, compared with those of its leading counterpart export credit agencies, remained among the most effective in the world. On 19^th November Corner House' solicitors wrote a letter before action to ECGD.

11. The issues between the parties

115. The main thrust of its complaint, which it carried forward into the judicial review proceedings themselves, was that the consultation between March and November 2004 had been one-sided, and that the new procedures and forms had effected fundamental changes to ECGD's anti-corruption requirements. All the changes had been made in one direction, and they weakened anti-bribery and anti-corruption protection. Two particular examples were given:

i) A definition of "applicant" had been introduced. Where the state of mind of a company was in issue, any knowledge of bribery or corruption held by senior executives or managers was now to be ignored. Only the state of mind of the directors and the signatory of the form was to be taken into account. The board was not now required to make any reasonable inquiry to ensure that its belief had any factual basis. Provided that the main board directors were not informed of improper conduct, the company was safe.

ii) Under the March 2004 procedures there was a condition of the guarantee agreement that the exporter would comply with what was required of it. If there was improper conduct, ECGD would be entitled to terminate the agreement without payment. The word "condition" had now been altered to "term". As a result ECGD had no automatic right to terminate and would ordinarily have to claim damages. Since it would normally suffer no loss or damage as a result of corruption, there was now no incentive for an exporter to comply with the anti-corruption procedures, which were thereby rendered nugatory.

116. During the hearing of the appeal Lord Lester of Herne Hill QC, who appeared for Corner House, drew our attention to another feature of the changes which had come to light in the documents disclosed by ECGD in these proceedings.
117. In 2003 TI had been arguing strongly for greater transparency in relation to the size of agents' commissions. In a paper submitted to OECD's Working Party on Export Credits and Credit Guarantees in April 2003 a member of TI's board had said that the employment of a middleman or agent was traditionally one of the most common vehicles for passing on not only the bribe but also the "sordid bribery action" itself to a third party for whose action the exporter often believed that he could not be held responsible. The commission very often contained not only a legitimate agency fee but also the amounts to be used for "unspecified contract acquisition purposes" (a euphemism for bribery).
118. At that time TI was suggesting that the names and addresses of the agent(s) and the level of agents' commission should be identified, and that any percentage higher than 5% should "raise the red flag" and require increased due diligence procedures to take place. Later in the year TI welcomed the opportunity ECGD had afforded it of commenting on a Best Practices paper, commending it for "its diligence in consulting upon these proposals." It said that ECGD's then current practice of requiring details on agents' commissions was a considerable step forward in seeking to stem this very obvious way in which foreign bribery might take place. It continued, however:

"A percentage is a poor indicator because even a very small percentage of a very large contract sum amounts to a lot of money that should not receive support from public funds if there is any suspicion of illicit purpose. What matters is the actual amount payable relative to the actual services to be provided. TI would like to see best practice address a range of practical issues that may be covered by enhanced due diligence in appropriate cases. An applicant's non-bribery declaration may be false. A customer who is willing to break the law to pay a bribe will probably also be willing to provide a false declaration. Due diligence on the agency agreement therefore needs to be undertaken to minimise the risk of false declarations."

119. The schedule attached to the revised application forms issued by ECGD in March 2004 addressed this issue directly. In a passage headed "Agents' Commission" ECGD observed that because agents could act as a conduit for improper payments, it was important that it received detailed information about any agent involved with a contract for which cover was being sought. After requiring details of any agent or other intermediary who would be involved in the process leading to the award of a contract, and of the services that they were performing, the form contained the following questions:

"Will you or any Affiliate or anyone acting on your or any of your Affiliates' behalfs make any payments to the agent or intermediary in respect of any contract or any related agreement, undertaking, consent, authorisation or arrangement of any kind? Yes/No.

If Yes, are all such payments included in the contract price shown at section … above? Yes/No.

If No, please give details below of the value of any payments that are not included in the contract price together with an explanation."

120. After he had studied the initial letters of complaint from leading exporters and the CBI, Mr Weiss, who was then ECGD's acting chief executive, told the CBI on 19^th May 2004 that ECGD had no valid reasons for changing the new forms it had so recently introduced. These complaints were then escalated to ministerial level. At a meeting attended by the Minister for Trade on 5^th July concessions were made on some issues, but no agreement was reached about the requirements in relation to greater transparency on the subject of agents. At that meeting ECGD officials expressed surprise that companies were now refusing to provide the additional information on agents' commission that it required, since most of these details had been specified in its application forms since April 2003. It regarded the provision of such information as important in ensuring that the business it supported was not tainted in any way. (It will be recalled that EGAC was told of a "major problem" at its July meeting: see para 109 above).
121. During the negotiations about the wording of the forms that then ensued ECGD initially maintained a very firm stance on its requirements for details of agency arrangements, although it expressed a willingness to be flexible in discussions about the steps it should take to prevent any leakage of this information. In a letter to the CBI dated 12^th August Mr Weiss said that there would be difficulty for ministers in changing this part of the system, given that the requirement for providing this information to ECGD had now been in place for over a year. He recognised that the safeguard arrangements he now proposed did not address the rather different objections raised by Airbus, a company in multi-national ownership.
122. Following further exchanges of views, on 13^th September Mr Crawford, ECGD's new chief executive, told the CBI that there had now been extensive consultation with ministers. The Secretary of State now wished certain changes to the arrangements to be adopted. An applicant was now to be at liberty to explain why it was unable to provide the name and address of any agent who was involved. If a commission was not included in the contract price, or covered by ECGD support, the applicant should declare whether or not it exceeded £2 million or 5% of the contract price, if less, and if it did, it should give details of the amount and the services in respect of which it was paid.
123. On 24^th September the CBI stated that even if the provision relating to a percentage were to be acceptable, an absolute amount would not be, given the very large contract values for which major UK exporters were responsible. A further meeting then took place on 7^th October, at which the following discussion was minuted:

"Companies will accept the 5% threshold but there will be major difficulties if ECGD also insists on the £2m figure as well. Commission above this amount is not unreasonable particularly for larger deals and those spread over a longer horizon. The £2m is an arbitrary figure that ECGD is trying to impose on industry without any prior consultation. It has no legal validity and unlike the 5% cannot be justified by any reference to the thresholds mentioned by NGOs. An actual figure at this level will catch too many deals, increasing the amount of commercially sensitive detail that companies would be required to provide, and adding to the risk that agents would be identified by competitors. Industry also had concerns that the £2m would be in the public domain and therefore be a focus for the attention of NGOs. This would also be a risk for ECGD as well, as it could encourage NGOs that amounts above this level were somehow suspect."

124. In response, ECGD officials, led by Mr Weiss, reiterated the case for including a base figure for the threshold as well as a percentage. They suggested that a possible way forward would be to increase the threshold to £5m or £10m. They said that ECGD would have to present the Secretary of State with valid reasons for dropping the base figure altogether.
125. In his letter dated 29^th October which concluded these exchanges Mr Weiss told the CBI that he could now "confirm" that in relation to applications where ECGD cover for the agent's commission was not being sought, the application form would now only request information about the existence of the agent, the amount of the commission, and the services rendered by the agent, where the agent's commission was more than 5% of the contract price. Any reference to a base figure (whether of £2m, £5m or £10m) had disappeared completely.
126. On 16^th November 2004 Mr Weiss appeared before the Trade and Industry Committee of the House of Commons. He did not mention this change at all when he gave examples of some of the legal issues that had led to changes in the forms. When he was asked why ECGD had not thought to consult the other interested parties who had welcomed the original changes, Mr Weiss replied:

"…There was an enormous amount of detailed negotiation around specific points in the documentation which I did not think would have been of any interest to the other parties."

127. In his witness statement Mr Weiss gave examples of the sort of things he had in mind when he made this comment. In paragraphs 48-52 he explained why ECGD had been willing to accept the 5% figure and referred to the "red flag" imagery in TI's April 2003 paper (see para 118 above). As to the absence of a base figure, he said that in the case of high value contracts, or contracts where there was a suspicion of bribery and corruption, ECGD had the ability to revert to the exporter and request details of any commissions paid to agents which would not be covered by ECGD's support and which were below the 5% threshold.
128. In its grounds for applying for judicial review Corner House contended that ECGD's failure to consult it (or other interested organisations), when it was carrying out extensive and detailed consultation with its corporate customers and their representatives, was a serious breach of basic public law standards of fairness and of ECGD's own published consultation policy. It asserted that the changes were far-reaching and fundamental, and were all in one direction, and that ECGD must have known that Corner House would wish to comment on the proposed revisions.
129. Section 4(1) of the 1991 Act provides that:

"Transactions entered into in pursuance of arrangements made under section 1 to 3 of the Act may be on such terms and conditions as the Secretary of State considers appropriate."

130. Corner House contended that where the Secretary of State was determining her policy as to standard terms and conditions, and where ECGD had published its policy for consultation on major issues (such as the weakening of the procedures it had recently introduced for combating bribery and corruption in connection with the contracts it supported), both public law fairness and ECGD's own policy required consultation with appropriate stakeholders. Given that ECGD's customers would have an interest in minimising their obligations and duties, ECGD would not receive a balanced or complete presentation of the arguments when determining how to perform its statutory duties unless it was willing to consult organisations like Corner House which were in practice the only bodies able to make representations to ECGD on the other side of the argument and to challenge the views of ECGD's customers.
131. Lord Lester also referred to arguments founded on Article 10 of the European Convention on Human Rights and on equality of treatment, but his most powerful contention, at any rate for the purpose of establishing a case for a PCO, was that the Secretary of State had acted ultra vires and in breach of her own consultation policy by undertaking a partial, one-sided consultation on a major issue, and ignoring the promise to consult all interested parties.
132. ECGD's response to this complaint was to the effect that public law imposes no general duty to consult – it cited Administrative Law, Craig (5^th Edition, 2003, pp 381-3) and English Public Law, ed Feldman (2004), para 15.28 for this proposition – and that it considered that the revisions it had made were neither fundamental nor major. It is easy to understand why they were not considered fundamental. On the other hand there was a seriously arguable case that some of them, at least, should be considered major. In the case of agent's commissions (for instance), which ECGD had described as a "major problem" in July, ECGD (and in due course its ministers) had resisted the change which was eventually conceded on four separate occasions (18^th May, 5^th July, 13^th September and 7^th October). Corner House argued with some force that in the knowledge of this history it was disingenuous of ECGD to suggest that this was not a major issue. In any event the history showed that even if it were shown that this was only a minor issue, ECGD had nevertheless acted in breach of its consultation policy by failing to use and maintain its existing dialogue with Corner House when it knew that Corner House would have had a lot to say if only it had been approached. In this context we would endorse the following comment of the European Court of Human Rights in Steel & Morris v UK (Judgment 15 February 2005):

"The Court considers, however, that in a democratic society even small and informal campaign groups, such as London Greenpeace, must be able to carry on their activities effectively."

12. The judgment of Davis J

133. In an admirable judgment – especially so in view of the pressures of time – Davis J said (at para 41) that on the relatively limited material actually drawn to his attention he would have formed the view that an arguable case for permission to proceed had been shown. He would not have formed the view that something likely to lead to a fundamental change in the way in which ECGD operated was involved here (para 44). Nor would he have taken the view that there was here a major change to ECGD's existing policies and practices (para 44). On the other hand the issues were of some importance at least, and arguably gave rise to an obligation, pursuant to ECGD's published policy, to take advice (if that was the right word) from Corner House given that advice was being taken from the CBI and its customers.
134. He then turned to consider Dyson J's four criteria in CPAG. His conclusion on the third, on which there was no issue on the appeal (so that we do not have to set out the evidence), was that Corner House's financial resources were very limited and the Department had massive resources available to it (para 53). On the other three issues his conclusions were that:

i) The case was not of sufficient general public importance to satisfy the first test;

ii) He did not feel himself to be in a position to say that Lord Lester's arguments were very strongly arguable, or ones that were very likely to succeed; such consideration could not therefore make up for the deficiency in the general public importance criterion so as to enable him to say that it was in the public interest that he should make a PCO;

iii) Although Corner House asserted that the proceedings would be discontinued if a PCO was not made, its lawyers might well continue to act if the case was indeed as important and as strong as it contended, and it was also appropriate to consider whether or not there might realistically be a pro bono alternative.

135. On the final issue he also suggested that it somewhat told against the "equality of arms" approach if, unlike the Refugee Legal Centre, Corner House would be asking for costs if it won while seeking a protection against an adverse costs order if it lost.

12. The reasons for our decision on the appeal

136. We had the opportunity of hearing rather fuller argument than the judge, and this enabled us to form a view of the case for making a PCO based on a rather fuller study of the available documents.
137. We had no hesitation in concluding for two quite different reasons that the case raised issues of general public importance. The first reason was that it relates to the way in which major British companies, supported by credit guarantees backed by the taxpayer in accordance with a statutory scheme, do business abroad. Obtaining contracts by bribery is an evil which offends against the public policy of this country. When the interests of the taxpayer are involved, the question whether or not companies are obliged to provide details of money paid to middlemen, such as were required by ECGD with the strong endorsement of the relevant minister before the changes were made, is a matter of general public importance.
138. The second reason is that the case raised important issues arising out of the implementation or non-implementation of ECGD's published consultation policy. As we have observed (see para 85 above), one of the drivers for its open consultation policy was the Sixth Report of the Committee on Standards in Public Life, which stated (at para 7.56):

" …[T]he Government needs to maintain a high quality of consultation with all groups and individuals. The openness and effectiveness achieved by some departments in consulting the public should become the norm for consultation right across Government. There is a need for reassurance that privileged access is not being granted during the development of policy. One solution could be to require greater consistency and transparency in the recording of all contributions from the outside world. The promulgation of stronger rules on consultation, requiring compliance with the principles of the document on written consultation to central government… would be more effective than maintaining the current position where they are simply recommended as best practice."

139. The papers before the court appear to evidence the mischief which the Committee on Standards in Public Life was concerned to address. The judge was influenced by the consideration that Corner House's challenge related to procedural unfairness and not to any alleged irrationality in the eventual outcome, and he took note of the fact that the issue in the centre of the case was whether or not ECGD should have consulted Corner House in the circumstances of this particular case.
140. Procedural issues, however, are often of greater importance than issues of substantial law. It is in our judgment a matter of general public importance if a division of a department of state publishes and adopts an open consultation policy of general application and then reverts to a timeworn practice of privileged access, particularly on an issue as obviously sensitive as measures to combat bribery and corruption in connection with the attainment of major contracts abroad.
141. On the third of these issues the judge plainly misdirected himself. What Corner House was mainly worried about was whether it would have to pay the Secretary of State's costs (and possibly the costs of an interested party, too) if it lost. Whether its lawyers might be willing to act pro bono or whether they had great confidence in the strength of their clients' case were matters which did not eliminate that risk. Mr Nicholas Hildyard, the director of Corner House responsible for its financial management, told the court that the company possessed slightly more than £8,000 in unrestricted funds, which was already fully committed, and that all the alternative funding sources which it had approached were unable to help. He gave unequivocal evidence to the effect that without the benefit of a PCO the company would have no option but to withdraw the claim, and the judge should have made findings to that effect on the evidence before him.
142. We turn back to the second issue. On the CPAG guidelines it would be impossible to fault the judge's approach. He said (at para 52):

"… I do not find myself in any position to say that [Lord Lester's arguments] are very strongly arguable or ones that are very likely to succeed, such that such consideration can make up for the deficiency in the public importance criterion. Certainly, on the arguments advanced before me, I would take the view that his arguments might succeed but certainly also the arguments of Miss Carss-Frisk advanced before me led me to the view that they might also fail."

143. We have explained (at para 73 above) why we consider that Dyson J set the threshold test too high in this regard. In a case as complex as this – there were nearly 1,000 pages of documents before the court – it would be very difficult for a judge to reach the kind of view on the merits that Dyson J required in what must necessarily be a short period. Using our substituted test we considered that Corner House had a real prospect of success in the sense that that phrase is used in CPR Parts 24 and 52. Since the judge understandably did not use this test, we were entitled to substitute our own view in this respect.
144. Finally, we considered that the public interest required that these issues should be litigated, and since Corner House had no private interest in the outcome of the case, and since our fourth and fifth principles (see para 74 above) were both satisfied, we considered in the exercise of our discretion that it was appropriate to permit Corner House to proceed with the benefit of a PCO, and that this was one of those exceptional cases in which such an order should be made. Corner House had a real prospect of showing that they had been wronged. Whether ECGD's procedural principles promised them consultation or dialogue, they had received neither. In 2003 they had been promised a substantive response to their report, and they never received it. In 2004 they were offered a meeting with the minister, and the offer ran into the sand. ECGD told them (and TI) that it regarded them as their primary NGO partners on the topic of bribery and corruption, yet what occurred in the spring, summer and early autumn of 2004 was the antithesis of partnership. And all through 2004 ECGD was affording privileged access to the representatives of commerce and banking which it wholly denied to Corner House, despite its acknowledged expertise in the topic and in the face of ECGD's own consultation policy.
145. In R v Somerset County Council ex p Dixon [1998] Env LR 111 Sedley J said that "public law is not about rights, even though abuses of power may and often do invade private rights; it is about wrongs – that is to say misuses of public power." In the present case Corner House asserted that it had been wronged, and if all the criteria for the grant of a PCO were otherwise met, we were satisfied that it was necessary in the interests of justice that it should be permitted to continue with the proceedings with the protection of a PCO. If we had not taken that course, the issues of public importance that arose in the case would have been stifled at the outset, and the courts would have been powerless to grant this small company the relief that it sought.
146. If we had not been under such time pressures we would no doubt have explored with the parties the possibility of making a PCO which had the effect, say, of requiring Corner House to meet the first £10,000 of the defendants' costs if its substantive application had been dismissed in due course. In general a PCO in that form, or in the form in which one was made in the Refugee Legal Centre case (in which the claimants undertook to seek no order for costs from the defendants if they won) are preferable to a PCO in the form in which we made it on the evening of 22^nd December.
147. Our order as drawn provided that:

"…

4, The court directs that the Defendant is not permitted to recover its costs of the judicial review proceedings from the Claimant;

5. The Claimant's costs are to be capped applying the decision of the Court of Appeal in the case of King v Telegraph Group Limited [2004] EWCA Civ 613 at paras 101-2;

6. The Claimant to apply to the senior costs judge to set the level of the court's cap."

148. These are the reasons why we made this order.

ORDER: Appeal allowed. Respondent to pay appellant's costs of appeal and application for a protective costs order before Mr Justice Davis on a standard basis subject to detailed assessment in not agreed.

(Order does not form part of approved Judgment)