England and Wales High Court (Administrative Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
DIVISIONAL COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		03/08/2018

LORD JUSTICE LEGGATT and MRS JUSTICE CARR DBE:

A. INTRODUCTION

1. This is another claim for judicial review of a decision by the Lord Chancellor to reduce the amount of money made available as legal aid for defending people accused of crimes. The decision challenged in these proceedings has reduced fees payable under a scheme called the Litigators' Graduated Fees Scheme under which most of the work done by "litigators" (typically solicitors) in preparing the defence of persons prosecuted in the Crown Court is paid. (There is a parallel scheme for advocates.) The principle of the Litigators' Graduated Fees Scheme – referred to for short in this judgment as "the Scheme" – is that a fixed "graduated" fee is paid for conducting a case irrespective of the number of hours spent working on the case. The fee is determined by a formula and depends on a number of factors, of which the most important are the nature of the offence charged, whether or not the defendant pleads guilty, the length of the trial (if the case goes to trial), the number of defendants in the proceedings and the number of pages of prosecution evidence ("PPE") served. These factors are used as rough measures, or proxies, of the complexity of the case and of the amount and difficulty of the work which it is likely to involve.
2. The decision under challenge was made on 24 October 2017 and implemented by the Criminal Legal Aid (Remuneration) (Amendment) Regulations 2017 ("the 2017 Regulations") which came into force on 1 December 2017. The effect of the decision was to reduce the maximum number of PPE served on the defence which can be counted in fixing graduated fees from 10,000 pages to 6,000 pages. The result is that, in any case where more than 6,000 PPE are served, the graduated fee is to be assessed in the same amount as if only 6,000 PPE had been served without taking account of any additional complexity that may be reflected by the fact that the volume of prosecution evidence is in fact greater. It is possible to claim "special preparation" fees for reading or viewing pages in excess of this limit but these do not cover any additional work, such as obtaining further evidence, arising out of the prosecution evidence.
3. Not all evidence actually served by the prosecution is included in the computation of PPE. In particular, evidence which has only ever existed in electronic form such as mobile phone records is not included unless the determining officer considers that in all the relevant circumstances (such as the importance of the evidence for the case) it ought to be included. The rationale put forward by the Lord Chancellor for the decision to reduce the maximum PPE allowed to 6,000 pages was, in summary, that there has in recent years been a steep increase in cases with high levels of PPE, which is said to have come about in part because a decision by a costs judge in a case called R v Napper [2014] 5 Costs LR 947 broadened the circumstances in which electronic evidence could be counted as PPE. According to the Lord Chancellor, the government had never intended to include such electronic evidence in the calculation. He considered that the result has been to increase the overall expenditure on the Scheme. The object of reducing the PPE limit was said to be to reverse this increase in cost and return the element of the Scheme driven by PPE to the level that it was before the Napper decision.
4. The Lord Chancellor's decision is challenged by the Law Society, the professional body which represents solicitors in England and Wales, on four grounds:

(1) It is argued that the consultation process undertaken before the decision was made was procedurally unfair because during the consultation the Ministry of Justice did not disclose the existence of the analysis on which it had relied to estimate the amount of the increase in expenditure said to have been caused by the Napper decision. The Law Society and other consultees therefore had no opportunity to comment on this analysis.

(2) It is said that the decision was irrational because it was allegedly based on some manifestly erroneous assumptions including a misunderstanding of the effect of the Napper case and the analysis already mentioned which is said by the Law Society to have used a flawed statistical method.

(3) It is said that the 2017 Regulations made in order to implement the decision constitute a disproportionate and unjustified interference with the right of access to justice protected by the common law.

(4) It is asserted – albeit without much enthusiasm – that the decision confounded a legitimate expectation that no further reduction in criminal legal aid fees would be made before the Lord Chancellor had undertaken a review which has yet to take place.

5. The Lord Chancellor's case, in summary, is that he reasonably concluded that the decision would restore compensation for PPE to about the levels previously seen before the definition of PPE was expanded by judicial decision; and also that this change was unlikely to affect the sustainability of the market for criminal defence work. He maintains that he took the decision on the basis of a careful quantitative analysis by officials and bearing in mind that (i) the effects of the decision would be monitored and (ii) criminal legal aid fees were to be more comprehensively reviewed shortly in any event. It is the Lord Chancellor's case that there was nothing procedurally unfair, irrational or otherwise unlawful about the decision.
6. The claim was commenced on 24 January 2018. On 9 April 2018 William Davis J ordered that the application for permission to proceed with the claim should be adjourned to a rolled up hearing, to be listed before the end of July. The judge also gave directions for further steps in the proceedings, including the service of evidence. These directions did not provide for any expert evidence which, pursuant to CPR 35.4(1) and 54.16(2), may not be adduced without the court's permission.
7. The Lord Chancellor's detailed grounds for resisting the claim and written evidence were served on 27 April 2018. When on 1 June 2018 the Law Society served evidence in reply, this included an expert's report from Professor Abigail Adams, an expert in econometrics. The Lord Chancellor responded by serving expert evidence on which he seeks to rely if the Law Society's expert report is admitted. The Law Society in turn served a supplemental report from Professor Adams commenting on the Lord Chancellor's expert evidence. As if that were not enough, the Lord Chancellor then served supplemental reports from his experts replying to the second report of Professor Adams. All this has been done without any permission from the court. We shall have to deal later in this judgment with the extent to which any of this evidence is admissible and should be admitted. We will also explain why the way in which the Law Society has gone about seeking to introduce expert evidence into the proceedings should not be followed in future.
8. But before addressing that question and the issues in dispute, we must explain some more of the factual and legal background.

B. FACTUAL AND LEGAL BACKGROUND

History of the Scheme

9. The Scheme replaced the standard fee and ex post facto payment scheme. The origins of the change can be found in the Carter Report on Legal Aid Procurement dated July 2006 ("the Carter Report"). This recommended the implementation of a new graduated fee scheme whereby case fees would be linked to a formula rather than the length of time spent preparing the case, offering an incentive for early resolution of cases where appropriate, and efficiency more generally. Additionally, given that additional payments in a graduated fee scheme for PPE and days at trial were a good proxy for complexity, those cases that were genuinely complex cases and required greater solicitor input would receive a proportionately higher total case fee.
10. The Carter Report led to the 2006 White Paper "Legal Aid Reform: The Way Ahead" (November 2006) (Cm 6993), announcing the introduction of the Scheme. The White Paper recognised the dynamic of "swings and roundabouts" in such a Scheme. A case that was more expensive to a firm than the standard fee would be balanced, "in the long run", by one that was cheaper.
11. In June 2007 the Legal Services Commission ("the LSC") published a consultation paper on fees payable under the Scheme. For this purpose an earlier extensive modelling exercise had been carried out, with the LSC sampling files as well as analysing data from some 273,000 claims. We have read various witness statements from solicitors describing their close involvement with the LSC in these early stages of devising the Scheme.
12. The LSC's consultation paper recognised that the graduated fee was not intended to be an accurate individual case payment system, but rather a means of providing a firm with "a reasonable overall payment for a large basket of cases", again a case of "swings and roundabouts". It confirmed that, on balance, case type, class of offence and PPE were effective proxies to predict the likely weight of a case in the Crown Court. Following consultation the LSC published its response, with revised fees. The definition of PPE was consulted on in November 2007.

The Scheme

13. The Scheme was then introduced on 18 December 2007 by the Criminal Defence Service (Funding) (Amendment) Order 2007 ("the 2007 Order") with effect from 14 January 2008. Guidance issued by the LSC and subsequently the Legal Aid Agency ("LAA") accompanied it. The Scheme is now contained in Schedule 2 to the Criminal Legal Aid (Remuneration) Regulations 2013 ("the 2013 Regulations") made pursuant to the powers in the Legal Aid, Sentencing and Punishment of Offenders Act 2012.
14. Where someone is eligible for criminal legal aid, the Scheme governs the remuneration of "litigators" for the provision of litigation services in respect of the majority of Crown Court proceedings. An "Advocates Graduated Fees Scheme" covers advocates' work in the Crown Court and there is a (pre-existing) separate scheme for "Very High Cost Cases" ("VHCC"). When the Scheme was first introduced, a significant number of heavy cases were dealt with under the VHCC scheme, for which the threshold was a trial length of 40 days. That threshold was subsequently increased to 60 days, leading to an increase of cases falling within the Scheme.
15. Litigators are paid a single graduated fee for dealing with each Crown Court case. The fee is set by reference to a mathematical formula with proxies for complexity, including the number of PPE served. The 2013 Regulations (at paragraph 20 of Schedule 2) also make provision for "special preparation fee" arrangements which allow for additional payments to be made for "reading" or "viewing" additional prosecution material outside qualifying PPE. The payments are discretionary, in each case depending on the assessment of the determining officer.
16. The PPE volume is an important proxy in the Scheme. The higher the PPE, the higher the graduated fee. Originally, the maximum number of PPE was 99,999. In 2009 it was reduced to 10,000. That was not a controversial move, since cases with more than 10,000 PPE were likely to fall within the VHCC scheme. It was made in full dialogue with the Law Society.
17. The number of pages is determined in accordance with paragraph 1(3) to (5) of Schedule 2 of the 2013 Regulations, which provide:

"(3) The number of prosecution evidence includes all –

(a) witness statements;

(b) documentary and pictorial exhibits

(c) records of interviews with the assisted person; and

(d) records of interviews with other defendants,

which form part of the served prosecution documents or which are included in any notice of additional evidence.

(4) Subject to sub-paragraph (5), a document served by the prosecution in electronic form is included in the number of pages of prosecution evidence.

(5) A documentary or pictorial exhibit which –

(a) has been served by the prosecution in electronic form; and

(b) has never existed in paper form,

is not included within the number of pages of prosecution evidence unless the appropriate officer decides that it would be appropriate to include it in the pages of prosecution evidence taking into account the nature of the document and any other relevant circumstances."

18. Schedule 2 paragraph 1(5) was originally introduced by the Criminal Defence Services (Funding) (Amendment) Order 2012 ("the 2012 Order"), amending the 2007 Order (which had defined PPE so as to exclude "documents provided on CDRom or other electronic means of communication"). LSC guidance at the time of the 2012 Order stated that if evidence was relied upon that would previously have been served in paper form, it should be included in the PPE count, including digital documents that formerly would have been printed off and included in a hard copy bundle.
19. There followed a number of costs cases debating whether a LAA officer should count particular electronic documents as part of the PPE: see R v Jalibaghodelehzi (SCCO 354/13); R v Fiaz (SCCO 57/14); R v Valton (SCCO 32/14), a topic also considered in R v Dodd [2014] 6 Costs LR 1131. Then came the case of Napper, already referred to above. There Costs Judge Simons allowed an appeal against the decision of a determining officer not to allow a document electronically served to count as PPE because it had never existed in paper form. At para 35 he stated:

"As Master Gordon-Saker stated in R v Jalibaghodelehzi, had it been intended to limit those circumstances only to the issue of whether the evidence would previously have been served in paper format, the Funding Order could easily so have provided. In my judgment, the regulation requires the Determining Officer to have a much wider role…He or she can take into account the nature of the document which clearly cannot be limited only to the physical nature of the document. Even if it did, the appropriate officer must also take into account all the relevant circumstances which clearly must go beyond whether or not the document previously existed only in paper form."

20. Napper was the first case in which oral and written submissions were made on behalf of the Lord Chancellor. Despite apparent expectations to the contrary, there was no appeal from the decision.
21. Revised guidance was issued by the LAA (in February 2015) to reflect the approach set out in Napper, including as follows:

"35. The [Napper] decision has interpreted the phrase "any other relevant circumstances" as including how important/integral the evidence was to the case and what work was required to consider this evidence. In other words, where there is sufficient evidence to establish that a page would previously have been served in paper form, in considering whether it would be appropriate to include it as a page of prosecution evidence regard should be had to how important/integral the evidence was to the case and what work was required to consider this evidence.

36. In the light of this, the LAA has revised the guidance both in order to comply with the findings of the Costs Judge in Napper and to preserve the policy intention reflected in the original 2012 guidance. That guidance is set out…below.

37. The LAA, in considering whether claims for electronic evidence should be paid as PPE, has confirmed that it considers that:

1. Whether the document would have been printed by the prosecution and served on paper form prior to 1 April 2012 is a relevant circumstance under paragraph 1(5) of Schedules 1 and 2 to the Regulations that the determining officer will take into account. If the determining officer is able to conclude that the material would have been printed prior to April 2012, it will be counted as PPE for both the litigator and advocate.

2. If the determining officer is unable to make that assessment, the determining officer will take into account "any other relevant circumstances" such as the importance of the evidence to the case, the amount and the nature of the work that was required to be done and by whom, and the extent to which the electronic evidence featured in the case against the defendant.

38. Some examples of documentary or pictorial exhibits that will ordinarily be counted as PPE are:

Scene of crime photographs

Prosecution analysis carried out on phone data

Bank statements

Raw phone data where a detailed schedule has been created by the prosecution which is served and relied on and is relevant to the defendant's case

Raw phone data if it is served without a schedule having been created by the prosecution, but the evidence nevertheless remains important to the prosecution case and is relevant to the defendant's case…

Raw phone data where the case is a conspiracy and the electronic evidence relates to the defendant and co-conspirators with whom the defendant had direct contact…"

The 2017 consultation

22. Various proposals and consultations with the objective of achieving savings in criminal legal aid expenditure emanated from the Lord Chancellor between 2013 and 2015, including the introduction of a 17.5% reduction to be introduced in two stages (of 8.75% reduction each) and a dual-contracting scheme which divided work between "own client work" and "duty provider work". On 28 January 2016 the then Lord Chancellor announced via a written Ministerial Statement that he would not be pressing ahead with a dual-contracting scheme and that he had therefore decided to suspend the second 8.75% fee reduction. He stated:

"I will review progress on joint work with the profession to improve efficiency and quality at the beginning of 2017, before returning to any decisions on the second fee reduction and market consolidation before April 2017."

23. Work between the Ministry of Justice and the Law Society on long term reforms to the Scheme terminated in late 2016.
24. However, on 10 February 2017 the Ministry of Justice published its consultation paper: "Litigators' Graduated Fees Scheme and Court Appointees", accompanied by a draft Impact Assessment and draft Equalities Statement. It is the decision taken following this consultation – which we will refer to as "the Decision" – that is the subject of the Law Society's challenge in these proceedings.
25. In the introduction to the consultation paper the Ministry of Justice stated under the heading "Longer Term Ambition":

"Over the next 12 months we want to work with the Law Society and other representative bodies to reform the scheme so that we measure the relative complexity of cases in a way that does not involve counting pages, but takes into account the totality of the evidence, whether paper or not. We would like to introduce a revised and future-proof scheme by early 2018."

The paper went on to refer to two specific immediate areas for reform under the heading "Short Term Pressures", one of which was the proposed amendment of the Scheme:

"…The Costs Judge decision in the case of Napper revised the interpretation of PPE, broadening the scope of what would be considered as PPE. This has contributed to LGFS costs rising substantially since 2013-14, while volumes of cases have fallen… In our view payments since Napper do not necessarily reflect the work actually done. Part one of this consultation paper sets out proposals for amending the LGFS to achieve a return to pre-Napper costs by lowering the point at which we stop counting PPE and start assessing work reasonably and actually done in relation to any additional pages under the "special preparation" provisions…"

26. The proposal put forward was to reduce the number of PPE used to calculate the graduated fee from 10,000 to 6,000. It also canvassed the possibility of restoring a second 8.75% reduction in all criminal legal aid fees.
27. The Ministry of Justice stated that the interpretation of the definition of PPE in Napper had contributed to the Scheme costs rising substantially since 2013-2014, whilst the volume of cases had fallen. Between 2013-14 and 2015-16 Scheme expenditure had increased from £292m to £341m (i.e. by £49m). The Ministry of Justice did not consider that the change in case mix accounted for the majority of the increase. There was no evidence that the rise in PPE reflected an equivalent increase in workload for providers. Reference was made to anecdotal evidence from caseworkers that large volumes of served material could be less relevant or capable of being searched electronically.
28. The draft Impact Assessment, amongst other things, also stated that the objective was to return Scheme expenditure to 2013-14 levels. Legal aid providers conducting cases with at least 6,000 PPE would receive around £26m to £36m less for Scheme payments.
29. The consultation period closed on 24 March 2017. A total of 1,005 responses were received, mostly from legal representatives or representative bodies. 97% of respondents opposed the proposals.

The Decision

30. On 24 October 2017 the Ministry of Justice published a Response to Consultation announcing the Decision, along with an Impact Assessment and Equalities Statement. Having summarised the responses, the Ministry of Justice stated that the government continued to believe that the proposals to limit PPE to 6,000 were sustainable. Around half the firms currently holding a contract would be unaffected. A return to pre-Napper level payments was sought. The Napper judgment did not reflect the original policy intention behind the Scheme. The second cut of 8.75% in fees would not be introduced.
31. The desire to introduce a revised and future-proof scheme in 2018 was repeated. (No progress has in fact been made in this regard to date, at least from the Law Society's point of view. Mr Chamberlain QC for the Lord Chancellor told us that the aim nevertheless remains for an overall revised scheme to be advanced this year. The review to that end will "begin shortly".)
32. The Impact Assessment included information as to (a) total volumes and expenditure of criminal legal aid fees broken down by reference to case category and (b) volume of cases and cost per case of criminal legal aid fees broken down by reference to case category, in each case over a three year period spanning the decision in Napper. It set out an analysis of cost changes by reference to different case sizes and the number and proportion of firms taking on work falling into those categories. It stated that expenditure under the Scheme had increased by around £44m from 2013-14 (pre-Napper) to 2015-16 (post-Napper).
33. The Decision was then implemented by the 2017 Regulations as set out above.
34. In the course of these proceedings various Ministerial submissions made prior to the Decision have been disclosed. On 15 September 2016 it was stated that the average costs had risen in part due to the decision in Napper but also in part due to a change in the mix of cases over time. It was stated that work was ongoing to "isolate the impact of expanded definitions of PPE on prices within this overall rise". Later Ministerial submissions of 11 November 2016, 14 December 2016 and 31 August 2017 cited a figure of £33m. This figure was derived from an analysis performed by the Legal Aid Agency ("the LAA analysis").

C. THE EXPERT EVIDENCE

35. We referred near the start of this judgment (at para 7) to the expert evidence on which each party has sought to rely in this case without permission from the court. As we will explain, as a result of the unsatisfactory way in which this expert evidence has been introduced, the extent to which it may be relied on is only now being decided by this judgment.

The applicable principles

36. The use of expert evidence in judicial review proceedings, as in all civil proceedings, in the High Court is governed by CPR Part 35. CPR 35.1 restricts expert evidence to "that which is reasonably required to resolve the proceedings." It follows from the very nature of a claim for judicial review that expert evidence is seldom reasonably required in order to resolve it. That is because it is not the function of the court in deciding the claim to assess the merits of the decision of which judicial review is sought. The basic constitutional theory on which the jurisdiction rests confines the court to determining whether the decision was a lawful exercise of the relevant public function. To answer that question, it is seldom necessary or appropriate to consider any evidence which goes beyond the material which was before the decision-maker and evidence of the process by which the decision was taken – let alone any expert evidence.
37. The classic statement of the extent to which evidence other than evidence of the decision under challenge is admissible in judicial review proceedings is that of Dunn LJ in R v Secretary of State for the Environment, ex parte Powis [1981] 1 WLR 584, 595. The categories identified in that case can be summarised as follows:

a) Evidence showing what material was before or available to the decision-maker;

b) Evidence relevant to the determination of a question of fact on which the jurisdiction of the decision-maker depended;

c) Evidence relevant in determining whether a proper procedure was followed; and

d) Evidence relied on to prove an allegation of bias or other misconduct on the part of the decision-maker.

38. Although these categories are a useful and well-established list, it would be wrong to treat them as if they were embodied in statute or as necessarily exhaustive. That is particularly so as public law has developed in ways which were not in contemplation when the Powis case was decided. In R (Lynch) v General Dental Council [2003] EWHC 2987 (Admin); [2004] 1 All ER 1159, Collins J was prepared to allow some extension of the possibility of admitting expert evidence beyond the Powis categories in a case where a decision is challenged on the ground of irrationality. The judge accepted that, where an understanding of technical matters is needed to enable the court to understand the reasons relied on in making the decision in the context of a challenge to its rationality, expert evidence may be required to explain such technical matters.
39. We would extend this principle to a situation where – as in the present case – it is alleged that the decision under challenge was reached by a process of reasoning which involved a serious technical error. It would be glib to suppose that, if an error in reasoning requires expert evidence to explain it, a challenge to the decision on the ground of irrationality cannot succeed. In Gibraltar Betting and Gaming Association Ltd v Secretary of State for Culture, Media and Sport [2014] EWHC 3236 (Admin); [2015] 1 CMLR 28, para 100, in the context of a challenge to a measure under EU law as "manifestly inappropriate", Green J said:

"An error which is far from being obvious or palpable may nonetheless prove to be fundamental. For instance, a decision or measure based upon a conclusion expressed mathematically might have been arrived at through a serious error of calculation. The fact that the calculation is complex and that only an accountant, econometrician or actuary might have exclaimed that it was an 'obvious' error or a 'howler', and even then only once they had performed complex calculations, does not mean that the error is not manifest… An error will be manifest when (assuming it is proven) it goes to the heart of the impugned measure and would make a real difference to the outcome."

40. The same point in principle applies, in our view, to a challenge based on irrationality. A decision may be irrational because the reasoning which led to it is vitiated by a technical error of a kind which is not obvious to an untutored lay person (in which description we include a judge) but can be demonstrated by a person with relevant technical expertise. What matters for this purpose is not whether the alleged error is readily apparent but whether, once explained, it is incontrovertible.
41. The corollary of this is that, as was recognised in the Lynch case, para 18, if the alleged technical error is not incontrovertible but is a matter on which there is room for reasonable differences of expert opinion, an irrationality argument will not succeed. This places a substantial limit on the scope for expert evidence. In practice it means that, if an expert report relied on by the claimant to support an irrationality challenge of this kind is contradicted by a rational opinion expressed by another qualified expert, the justification for admitting any expert evidence will fall away.
42. Two further issues are raised in these proceedings on which, in our view, expert evidence could in principle be admissible. The first is whether the consultation procedure was unfair, in particular because information of substantial importance to the decision was not disclosed. As discussed below, it is relevant in determining that issue to have regard to the potential impact of the proposal put out for consultation. Expert evidence may assist in showing this. We also think it relevant to consider whether consultees were prejudiced by the non-disclosure of significant information. Where, as in this case, the information in question is of a technical nature, and the claimant says that, had the information been disclosed, it would have wished to make technical submissions in response to the consultation, expert evidence may be admissible to illustrate the points which might have been made and to show their materiality.
43. The other issue raised in this case on which expert evidence could in principle, depending on its content, be admissible is the argument that the decision to reduce legal aid fees constituted an unlawful interference with the right of access to justice.

The procedure followed

44. If a party to judicial review proceedings wishes to rely on expert evidence in support or defence of the claim, it is essential that an application for permission to do so and for appropriate consequential directions should be made at the earliest possible opportunity. The points made in this regard in BAA Ltd v Competition Commission [2012] CAT 3, para 82, are equally applicable in judicial review proceedings generally. In the ordinary course, any such application by the claimant should be made when its grounds for bringing the claim are filed so that it can be considered by the judge who makes the decision whether to give permission to proceed with the claim. In the present case the Law Society points out – and we accept – that it was not until the Lord Chancellor filed his detailed grounds of resistance and written evidence on 27 April 2018 that his reliance on the LAA analysis and the significance of that analysis for the claim became clear. We accordingly accept that, in so far as the expert evidence of Professor Adams responds to that analysis, the need for that evidence could not reasonably have been identified sooner.
45. Once the Law Society had decided to instruct an expert in econometrics to prepare a report, however, notice of that decision should have been given to the Lord Chancellor's representatives. Simply to serve an expert's report with other evidence in reply, as the Law Society did on 1 June 2018, without having given any prior notice of the intention to do so was not consistent with the duty of cooperation which the parties owed to each other and to the court, particularly when the court had – at the Law Society's request – ordered an expedited hearing. Further, it was wrong in the application notice filed with the evidence to invite the court to deal with the application to admit expert evidence at the start of the substantive hearing and to give a time estimate for dealing with the application of five minutes. The Law Society should have requested that its application be placed before a judge (if possible, the Lord Justice who would be presiding at the hearing) as soon as possible. That judge might well have decided that the application, if opposed, should be dealt with at the start of the hearing. But it was not for the applicant to pre-judge that decision.
46. On 11 June 2018 the Government Legal Department on behalf of the Lord Chancellor wrote to the solicitors representing the Law Society setting out detailed reasons for opposing the Law Society's application to rely on Professor Adams' report. At that point if not before, it must have been obvious to the Law Society's representatives that the time estimate of five minutes for dealing with the application was completely unrealistic and a realistic time estimate should have been provided to the court. That would almost certainly have meant that the overall time estimate for the hearing needed to be increased beyond the two days that had been allowed, with additional time also made available for pre-reading.
47. The Law Society's skeleton argument for the hearing was filed on 20 June 2018. Reliance was placed on Professor Adams' report in the skeleton argument as if the report were already in evidence. That too was wrong. The facts that permission had not been obtained to rely on the report and that the Law Society's application to admit it was opposed should have been explained to the court. Indeed, it was reasonable to expect the skeleton argument to contain a short statement of the reasons relied on for asking the court to admit the report of Professor Adams. As it was, such reasons were given only in a supplementary skeleton argument, filed after the Lord Chancellor's skeleton argument had, in an appropriate fashion, set out in an annex his reasons for opposing the application.
48. Finally, it was probably in hindsight a mistake for the court not to insist that the Law Society's application to admit expert evidence and the Lord Chancellor's cross-application to rely on expert evidence if the Law Society's application was granted should be dealt with at the start of the hearing. However, as mentioned, no proper allowance for the time required to deal with the applications had been included in what was already a tight time estimate. The upshot was that arguments about whether or to what extent the expert evidence should be admitted were made by each side alongside arguments which relied on the expert evidence. That was an unsatisfactory way to proceed.
49. To disentangle those questions, we will first give our ruling on the application to admit expert evidence before addressing the grounds of the Law Society's claim.

Professor Adams' report

50. Professor Adams is an economist who specialises in econometrics. Part of her report (paras 15-54) points out what are said to be flaws in the methods used by the Lord Chancellor's officials to estimate the extent to which the Napper decision caused the costs of the Scheme to increase. The main focus of these criticisms is the LAA analysis. Save on one minor point (which relates to the choice of yearly rather than quarterly data to compare cost trends in the Scheme with cost trends in the Advocates' Graduated Fee Scheme) the Lord Chancellor's expert evidence does not respond to any of the criticisms made by Professor Adams. We infer that, save for the one point with which issue is taken, the validity of those criticisms is accepted by the Lord Chancellor's experts. In these circumstances we consider that this part of Professor Adams' report is admissible in connection with the Law Society's irrationality challenge as well as to support its case that the non-disclosure of the LAA analysis prejudiced consultees. Although some of Professor Adams' criticisms are directed at matters which were mentioned in the Impact Assessments published as part of the consultation process, we do not think it right to sever those criticisms from the rest of this part of Professor Adams' evidence as the significance of the LAA analysis depends in part on the nature and validity of other reasons relied on by the Ministry of Justice for attributing an increase in expenditure to the Napper decision.
51. We accept that there is good reason for the delay in obtaining this evidence. Although we have criticised the way in which the Law Society sought to introduce it, we think it right to admit this part of Professor Adams' report.
52. We are also prepared to admit paras 88-101 of the report, which contain some analysis of the effect which the decision to reduce the PPE cap can be expected to have on firm income. The figures have not been contested and form part of the relevant context in assessing the fairness of the consultation process.
53. The rest of Professor Adams' report presents independent analyses which she has carried out herself for the purpose of assessing the effect of the Napper decision. Her main analysis is said to show, in essence, that there was no statistically significant increase in costs following the Napper decision which is not explicable as part of a longer term trend.
54. We refuse permission to rely on this evidence for a number of reasons. First and most simply, the question of whether or not the Napper decision caused an increase in the cost of the Scheme is not a question for the court to decide. Nor does the fact that Professor Adams has reached different conclusions using different methods show that the methods used to perform the LAA analysis were improper. In any event, Professor Adams' methods and conclusions are heavily contested by the Lord Chancellor's experts and the court is in no position to resolve that dispute. Professor Adams' evidence also has no bearing on whether consultees were prejudiced as a result of the non-disclosure of the LAA analysis, as it is entirely independent of that analysis. In addition, we see no good reason why, if the Law Society thought it relevant to obtain an independent analysis of the effect of the Napper decision on the cost of the Scheme – an issue clearly flagged as critical to the proposal in the consultation documents – it could not have proposed the need for such an expert analysis at the start of these proceedings.
55. The other analyses carried out by Professor Adams are relied on to rebut a contention made in the Lord Chancellor's detailed grounds of resistance that his decision to reduce the PPE limit was not a "cut" in the conventional sense, as it was designed to return the element of expenditure driven by PPE to pre-Napper levels. We do not consider the debate about the validity or otherwise of that contention to be of any real relevance to the issues which the court has to decide. Counsel for the Law Society sought to argue that this part of Professor Adams' report is relevant to its case that the Decision violates the right of access to justice. But in our view this evidence is far too remote in its connection with that issue to be reasonably required to resolve it. Analysing the estimated effect of the Napper decision has no direct bearing on whether the reduction in the PPE cap will result in any category of defendants in criminal cases being denied adequate legal representation. In any case, we see no good reason for the delay in seeking to adduce expert evidence on this issue.
56. Accordingly, we admit in evidence paras 15-54 and 88-101 of Professor Adams' report (together with the corresponding parts of the executive summary). Permission to rely on the report is otherwise refused.

The Lord Chancellor's expert evidence

57. In response to Professor Adams' report the Lord Chancellor served a witness statement from Mr Darren McHale, a member of the Government Statistical Service who works for the Ministry of Justice, and an independent expert's report from Professor João Santos Silva, an economist who, like Professor Adams, specialises in econometrics. Although Mr McHale's evidence is contained in a witness statement – no doubt because he is an internal and not an independent expert, his evidence is opinion evidence which depends for its weight on his statistical expertise.
58. At the hearing Mr Chamberlain confirmed the Lord Chancellor's position as being that he was seeking permission to put in evidence the statement of Mr McHale and the report of Professor Santos Silva only if the Law Society were given permission to rely on Professor Adams' report, on the ground that in that event the Lord Chancellor should be permitted to respond to her evidence. If, however, the evidence of Mr McHale and Professor Santos Silva is admitted for this purpose, Mr Chamberlain indicated that he would also wish to rely on their evidence, if necessary, to support an argument based on section 31(2)(A) of the Senior Courts Act 1981 which we address later in this judgment.
59. With one exception, all of the written evidence of Mr McHale and Professor Santos Silva responds to parts of Professor Adams' evidence which we have declined to admit. In particular, they criticise aspects of methods used in her independent analysis of the effect of the Napper decision. Mr McHale also presents a further analysis which is said to show that, if carried out in a different and allegedly more appropriate manner, modelling of the kind undertaken by Professor Adams supports the original view taken by the Ministry of Justice that the Napper decision did have a significant effect on costs.
60. As we have refused permission to the Law Society to rely on the analyses and evidence to which the Lord Chancellor's expert evidence is responsive, it follows that permission to rely on that responsive evidence must also be refused. The evidence is in any case inadmissible, in our view, for similar reasons as the evidence to which it responds.
61. The one exception is a short passage in Mr McHale's witness statement (paras 45-49) which answers Professor Adams' criticism of the use by the Ministry of Justice of yearly rather than quarterly data. As this criticism is contained in the part of Professor Adams' report which we have admitted in evidence, we will also admit Mr McHale's response to it (and Professor Adams' reply to Mr McHale's response at paras 36-39 of her supplemental report). The remainder of the supplemental reports of Professor Adams, Mr McHale and Professor Santos Silva are responding to evidence which we have not admitted. Accordingly, permission to rely on those supplemental reports is also otherwise refused.
62. On this basis we turn to consider the four grounds on which the Law Society seeks to challenge the Lord Chancellor's Decision and the 2017 Regulations which implemented it.

D. LEGITIMATE EXPECTATION

63. It is convenient to dispose first of the Law Society's fourth ground, on which Ms Rose QC did not elaborate in her oral submissions. This is an argument that the Decision was unlawful because solicitors who undertake legally aided criminal defence work had a legitimate expectation that no reduction would be made in fees paid under the Scheme before the Lord Chancellor had carried out a review. This expectation is said to stem from the Ministerial Statement issued on 28 January 2016 (referred to at para 22 above), in which the then Lord Chancellor stated:

"I will review progress on joint work with the profession to improve efficiency and quality at the beginning of 2017, before returning to any decisions on the second fee reduction and market consolidation before April 2017."

The reference to the "second fee reduction" was to the second cut of 8.75% in all fees payable under the Scheme which had previously been proposed.

64. The Response to Consultation noted that:

"Some respondents said that at the time the second fee cut was suspended the then Lord Chancellor promised a 'review' before the final decision on the second fee cut was taken and that therefore a review should take place before pressing ahead with this proposal."

This was not, however, an argument which the Law Society itself made in its own response and in our view it is not a tenable argument. The Ministerial Statement could not reasonably have been understood as a commitment to carry out a formal review (of unspecified ambit) before any new proposal to reduce fees by altering any element of the Scheme was put forward. It was no more than a statement of intention to work with the profession to improve efficiency and quality and to consider the extent of progress on such work before reinstating the proposal which the Lord Chancellor had suspended in January 2016.

65. In our view, the only relevant legitimate expectation which law firms undertaking criminal defence work had was that proper consultation with the profession and with the public would take place before any significant change to the Scheme was made. That expectation was justified not only by the importance of legal aid to the livelihoods of those who serve the public interest by defending people accused of crimes who cannot otherwise afford legal assistance but by the necessity of ensuring that the public funds available to pay for such assistance are sufficient to guarantee that criminal proceedings are fair. The expectation also arose from the long-standing practice of consulting before significant changes in the provision of legal aid are made: see R (London Criminal Courts Solicitors Association) v Lord Chancellor [2014] EWHC 3020 (Admin); (2015) 1 Costs LR 7, para 34.

E. PROCEDURAL UNFAIRNESS

66. The ground which Ms Rose QC put at the forefront of the Law Society's case in oral argument is that the Decision was taken without proper consultation because neither the figure of £33m estimated by the Ministry of Justice to be the increase in annual expenditure attributable to the Napper decision, nor the LAA analysis on which the estimate was based, were disclosed during the consultation process (or when explaining the Decision).

What proper consultation requires

67. The essential requirements which any consultation process must satisfy were summarised by Lord Woolf MR, giving the judgment of the Court of Appeal in R v North and East Devon Health Authority ex parte Coughlan [2001] QB 213, para 108:

"To be proper, consultation must be undertaken at a time when proposals are still at a formative stage; it must include sufficient reasons for particular proposals to allow those consulted to give intelligent consideration and an intelligent response; adequate time must be given for this purpose; and the product of consultation must be conscientiously taken into account when the ultimate decision is taken."

These criteria were approved by the Supreme Court in R (Moseley) v Haringey London Borough Council [2014] UKSC 56; [2014] 1 WLR 3947, para 25, where Lord Wilson (who gave the lead judgment) endorsed them as "a prescription for fairness".

68. As that description indicates, although proper consultation also serves wider public interests such as ensuring public participation in decision-making and improving the quality of decisions, the duty to consult is generally conceived as an aspect of the common law duty of procedural fairness. In the Moseley case Lord Wilson characterised fairness as "a protean concept, not susceptible of much generalised enlargement" (para 24), and the point has often been made that the application of the duty is highly sensitive to the facts and context of the particular case: see e.g. R (EasyJet Airline Co Ltd) v Civil Aviation Authority [2009] EWCA Civ 1361, para 51; the London CCSA case, supra, para 34; R (United Co Rusal plc) v London Metal Exchange [2014] EWCA Civ 1271; [2015] 1 WLR 1375, para 28. As such, we do not think it useful to examine the facts of cases cited in argument on which each party sought to rely. The ultimate test has been expressed as whether the consultation process was so unfair as to be unlawful: see R (Baird) v Environment Agency and Arun District Council [2011] EWHC 939 (Admin), para 51; the London CCSA case, para 34; R (West Berkshire District Council) v Secretary of State for Communities and Local Government [2016] EWCA Civ 441; [2016] 1 WLR 3923, para 60.
69. The Law Society's complaint in the present case is focused on the requirement that consultation must include sufficient reasons for proposals to allow those consulted to give intelligent consideration and an intelligent response. Lord Woolf elaborated on this requirement in Coughlan, when he said (para 112):

"It has to be remembered that consultation is not litigation: the consulting authority is not required to publicise every submission it receives or (absent a statutory obligation) to disclose all its advice. Its obligation is to let those who have a potential interest in the subject matter know in clear terms what the proposal is and exactly why it is under positive consideration, telling them enough (which may be a good deal) to enable them to make an intelligent response. The obligation although it might by quite onerous, goes no further than this."

70. In R (Eisai) v National Institute for Health and Clinical Excellence [2008] EWCA Civ 438, para 26, Richards LJ, after citing this passage, observed:

"The mere fact that information is 'significant' does not mean that fairness necessarily requires its disclosure to consultees … nevertheless the degree of significance of the undisclosed material is obviously a highly material factor."

71. In Devon County Council v Secretary of State for Communities and Local Government [2010] EWHC 1456 (Admin), para 68, Ouseley J said:

"What needs to be published about the proposal is very much a matter for the judgment of the person carrying out the consultation to whose decision the courts will accord a very broad discretion … But, in my judgment, sufficient information to enable an intelligible [sic] response requires the consultee to know not just what the proposal is in whatever detail is necessary, but also the factors likely to be of substantial importance to the decision, or the basis on which the decision is likely to be taken."

72. All these observations were cited and applied by Burnett J in the London CCSA case in the context of a consultation by the Lord Chancellor on a previous proposal to reduce criminal legal aid fees.
73. In principle and consistently with these authorities, in judging whether non-disclosure of particular information made a consultation process so unfair as to be unlawful, relevant considerations in our view include:

(1) The nature and potential impact of the proposal put out for consultation;

(2) The importance of the information to the justification for the proposal and for the decision ultimately taken;

(3) Whether there was a good reason for not disclosing the information; and

(4) Whether consultees were prejudiced by the non-disclosure.

74. In relation to the last of these matters, whilst it is no part of the court's function to assess the merits of the arguments for or against the proposal, it is relevant in our view in judging whether the process was fair to consider whether the non-disclosure of information has prejudiced consultees by depriving them of the opportunity of making representations which it would have been material for the decision-maker to take into account.

The factual context

75. The point was made in the consultation paper, and emphasised by counsel for the Lord Chancellor in submissions, that the number of cases affected by the proposal was only around 2% of the total number of cases covered by the Scheme. In those cases which were affected, however, the effect was potentially severe. For a Class A offence Crown Court trial with at least 10,000 PPE, reducing the cap to 6,000 PPE would reduce the maximum 'final fee' by 37%. Furthermore, as mentioned earlier, the projected overall impact of the proposal on costs was a saving of between £26m and £36m a year. Those figures represent 7.6% and 10.6% respectively of the total expenditure of £341m under the Scheme in 2015-16. Thus, although affecting only a small number of cases, the overall size of the anticipated reduction in payments was not dissimilar to the fee cut of 8.75% which the Lord Chancellor had previously proposed.
76. In addition, Professor Adams has calculated that cases with over 6,000 PPE were the source of 24% of firm income on average in 2015-16 and that, for those firms which took on at least one such case, such cases were the source of 50% of firm income on average. Professor Adams also calculated that the firms which took on at least one such case between them handled more than 70% of all cases covered by the Scheme in 2015-16. It is thus apparent that the potential financial impact of the proposal on the funding and profitability of criminal defence work was substantial.
77. Another point made in the consultation paper and emphasised by counsel for the Lord Chancellor is that the reduction in the PPE cap was, and is, intended only as an "interim measure", since the Lord Chancellor's "longer term ambition" is to undertake a more comprehensive review and reform of the Scheme. It was said in the consultation paper that the Lord Chancellor wants to reform the Scheme so as to "measure the relative complexity of cases in a way that does not involve counting pages, but takes into account the totality of the evidence, whether paper or not." It was also said that "we would like to introduce a revised and future-proof scheme by early 2018." (In the event, as already noted, no proposal for such a revised scheme has yet been put forward.)
78. We consider that the expressed intention to introduce more comprehensive reforms in the longer term was a relevant factor for consultees to take into account in responding to the government's proposal. However, nothing said in the consultation documents was capable of creating a legitimate expectation that the proposed reduction in the PPE cap would be superseded by some other arrangement within any particular timescale or indeed at all. In these circumstances only limited weight could reasonably be given to the characterisation of the proposal as an interim measure.

The importance of the LAA analysis

79. The importance of the LAA analysis to the policy justification given for the Decision to reduce the PPE cap will already be apparent. In essence, the justification was that the decision in the Napper case in 2014 had allegedly broadened the definition of PPE and had resulted in electronic evidence being counted as PPE which the government had not intended to include within the definition; that this in turn had caused a significant increase in cost; and that intervention was necessary to reverse this increase in cost. As it was succinctly put in the Impact Assessment published with the Response to Consultation:

"The objective is to return the element of payment driven by PPE to the level it was at prior to the Napper case…"

80. In this regard it is important to note that the proposal was not to reverse the effect of the Napper decision directly by narrowing the definition of PPE to exclude from its scope electronic evidence of the kind which, according the Ministry of Justice, it had not been the government's intention to include. That option was rejected as impracticable. Instead, the proposal was to reduce the fees paid in cases with large volumes of PPE (by lowering the cap) by an amount which would reverse the increase in cost caused by the Napper decision. To achieve this policy objective it was therefore necessary to estimate what the increase in cost caused by the Napper decision was. That could not be done simply by comparing the costs of the Scheme before and after the case of Napper was decided since – as is obvious and as the Ministry of Justice recognised – other factors such as the number and type of cases prosecuted in the Crown Court, which naturally varies from year to year, would also have had an impact on the total expenditure. As was stated in the Response to Consultation:

"A reduction in spend because volumes have decreased or cases have become less serious overall is a natural consequence of the number or types of cases. That is not a 'saving' as we would expect to pay less for fewer or less serious cases. Equally, if volumes increase or case-mix becomes more serious we would expect to pay more. There is not a fixed budget that we would look to spend whether or not volumes or case-mix change.

Our PPE proposals are not designed to get us to a particular expenditure level (irrespective of volumes or case-mix). They are designed to return the element of payment driven by PPE to pre-Napper levels in line with our intended policy."

81. In order to design the proposals, it was thus essential to carry out an analysis to isolate the 'Napper effect' by estimating the amount of the increase (if any) in payments under the Scheme which was attributable to the change in the definition of PPE that followed the Napper decision, as opposed to other factors. This was the purpose of the LAA analysis. Mr MacMillan, the Head of the Criminal Legal Aid Policy Team in the Ministry of Justice who led the work to develop the policy which is the subject of this case, has explained in a witness statement how the LAA analysis was carried out and how it generated the figure of £33m which was set out in the Ministerial Submissions. Mr MacMillan also specifically confirmed that the figure of £33m (and hence the underlying analysis) was relied on by the Lord Chancellor in taking the Decision. Furthermore, although Mr MacMillan does not say so expressly, it is apparent from his explanation of how the policy was designed that the level of the revised PPE threshold was calibrated at 6,000 pages because it was estimated that this would achieve an annual saving approximately equal to the £33m figure.
82. For these reasons, the figure of £33m and the LAA analysis on which it was based were pivotal to the justification for the Decision.

The reason for non-disclosure

83. No explanation has been given on behalf of the Lord Chancellor for the absence of any reference to the figure of £33m and the analysis on which it was based in the consultation documents. It is therefore unclear whether this omission occurred through oversight or was the result of a deliberate decision not to disclose this information. All that Mr MacMillan has said on this question in his witness statement is that:

"… from our perspective, we considered it more important to set out: (i) the increase in costs between 2013-14 and 2015-16; (ii) our view that Napper had contributed to the overall change; and (iii) the projected impact of the proposed amendment."

84. Mr MacMillan did not explain why he and others responsible for the consultation documents thought it more important to set out their view that Napper had contributed to the overall change in expenditure than to set out the amount of the estimated contribution and how that amount had been calculated. He also did not explain why it was thought more important to set out the projected impact of the proposed amendment than to set out the estimated increase in costs which the amendment was designed to reverse, not least so that the two figures could be compared.
85. It is also instructive to note the two questions posed in the consultation paper to which consultees were invited to respond:

"Q1. Do you agree with the proposed reduction of the threshold of PPE to 6,000?

Please give reasons.

Q2. If not, do you propose a different threshold or other method of addressing the issue?

Please give reasons."

We find it impossible to see how the consultees could make an intelligent assessment of whether – judged on the government's own terms – the new proposed threshold of 6,000 PPE was an appropriate threshold without knowing the estimated amount of the cost increase which the proposal was seeking to reverse. Nor could consultees give intelligent reasons for disagreeing with the proposed threshold if they were not told how it had been arrived at.

86. In short, no reason – let alone a good reason – has been given for not disclosing during the consultation process the LAA analysis and its results.

Prejudice to consultees

87. The fact that the non-disclosure of the LAA analysis caused prejudice to consultees is demonstrated by Professor Adams' report. As discussed in more detail below, that report identifies two serious flaws in the LAA analysis. To put it no higher, therefore, the failure to disclose the analysis deprived consultees of the opportunity of making representations in response to the proposal which it would have been material for the Lord Chancellor to take into account in making his decision.

The Lord Chancellor's case

88. The factors considered above all point clearly to the conclusion that the failure even to refer in the consultation documents to the LAA analysis and the estimate produced by that analysis undermined the consultation process because this information was essential to allow those consulted to give intelligent consideration and an intelligent response to the proposal.
89. The main argument made on behalf of the Lord Chancellor to defend the fairness of the consultation process can, we hope fairly, be summarised as follows. The people most interested in the consultation and most likely to comment on the proposal were solicitors who undertake legally aided criminal defence work. They are a sophisticated audience, familiar with the Scheme. They would have understood from the draft Impact Assessment which accompanied the consultation paper that the objective of the proposal was to reduce legal aid expenditure by approximately the amount that such expenditure was believed by the Ministry of Justice to have increased as a result of the decision in Napper. The draft Impact Assessment indicated that the result of implementing the proposal would be to reduce payments under the Scheme by around £26m to £36m. From this information consultees could be expected to infer that the Ministry of Justice had estimated that the increase in expenditure resulting from the decision in Napper was somewhere in this range.
90. The fact that consultees would have understood this is confirmed, Mr Chamberlain submitted, by the fact that one of the consultees whose response is in evidence, the Criminal Law Solicitors' Association, described the reason put forward for the proposal (accurately) in its response as being that "the case of Napper has cost the Legal Aid Agency £30 million." As no such statement was made anywhere in the consultation documents, the authors of this response must have deduced that the Ministry of Justice was relying on such an estimate. If this was apparent to the Criminal Law Solicitors' Association, it would reasonably have been apparent to other consultees too. Mr Chamberlain submitted that consultees would further reasonably expect that the estimate of the cost increase caused by the case of Napper was based on some form of statistical analysis and that this expectation would have been reinforced by the fact that the Impact Assessment incorporated a certificate, signed by the responsible Minister, stating:

"I have read the Impact Assessment and I am satisfied that (a) it represents a fair and reasonable view of the expected costs, benefits and impact of the policy, and (b) the benefits justify the costs."

It was submitted that, in order to sign that certificate, the Minister would need to have been satisfied that there was a reasonable basis for the view that the Napper decision had caused an increase in costs of the order of £30 million. That meant, as consultees would reasonably have appreciated, that an analysis must have been undertaken which had generated such an estimated figure. If consultees wanted to see the analysis relied on by the Ministry, they could therefore have asked. This would have been simple, as the Impact Assessment gave the email address of Mr MacMillan, the official handling the consultation process, as a "contact for enquiries".

91. Counsel for the Lord Chancellor also made the point that the consultation documents squarely raised the issue of the extent to which, if at all, the Napper decision had caused or contributed to an increase in the element of legal aid expenditure driven by PPE. Accordingly, the Law Society and other consultees were on notice of the significance of that issue. If they had thought it important to analyse the effect of the Napper decision on costs for the purpose of responding to the consultation, they could have contacted the Ministry of Justice and requested the data needed for that purpose.
92. It is the Lord Chancellor's case that in these circumstances consultees were not prejudiced by the fact that the consultation documents did not refer to the LAA analysis or to the specific figure of £33 million generated by that analysis.

Conclusions

93. It is difficult to express in language of appropriate moderation why we consider these arguments without merit. The first point, which should not need to be made but evidently does, is that consultees are entitled to expect that a government ministry undertaking a consultation exercise will conduct it in a way which is open and transparent. In particular, they are entitled to expect that if, on the crucial question raised in the consultation paper, officials have carried out an analysis which forms the basis of the proposal, then that fact will be mentioned in the consultation documents and not left to be inferred.
94. It should also go without saying that consultees are entitled to expect that consultation documents will not be positively misleading. When a draft Impact Assessment is published which purports to set out the "evidence base" for the proposal, including an analysis of costs and benefits and a statement of key assumptions and risks, the reader would understand that any analysis relied on to estimate the increase in expenditure which it was the policy objective to reverse was described in the Impact Assessment. The fact that the responsible Minister has certified that the Impact Assessment "represents a fair and reasonable view of the expected costs, benefits and impact of the policy" would further reinforce that understanding. That certificate did not – as was suggested on the Lord Chancellor's behalf – justify an inference that a crucial part of the analysis of the costs, benefits and impact of the policy had not been disclosed in the Impact Assessment. Its significance is the exact opposite of what counsel for the Lord Chancellor submitted. The Minister's certificate encouraged and entitled consultees to believe that a full and fair statement had been made of the financial analysis underlying the proposal.
95. The fact that many consultees were likely to be knowledgeable and sophisticated is also not a reason for withholding important information from them. Again, if anything, the opposite is true. That fact gave all the more reason to disclose the analysis relied on to estimate the increase in expenditure which it was the aim of the proposal to reverse because it was a reason to expect that at least some consultees, such as the Law Society, would be able to provide an informed critique of that analysis – having commissioned expert assistance if necessary.
96. Perhaps the simplest and most telling answer to the suggestion that consultees could be expected to infer that a key piece of analysis relied on by the Ministry had not been mentioned in the consultation documents is that, so far as the evidence shows, none of them did. Instead, they evidently assumed – incorrectly as it now proves – that the description given in the Impact Assessment of the expected costs, benefits and impact of the proposal was fair and not misleading. Among the responses to the consultation which the Law Society has put in evidence is that of Mr Meyer, a solicitor who worked closely with the Ministry of Justice when the Scheme was designed. In giving his reasons for strongly disagreeing with the proposal to reduce the PPE limit, he wrote in his response:

"The proposal is reckless and is based on assumptions as to the reasons for an increase in the average cost of an LGFS claim which are not supported by any empirical evidence or data."

Thus, the conclusion that Mr Meyer drew – appropriately in our view – from the fact that the consultation documents did not state that a quantitative analysis had been undertaken to identify the cause of the increase in cost was that no such analysis had been undertaken. Nor did officials at the Ministry, when they read Mr Meyer's response, disabuse him of his belief and inform him that an analysis had in fact been carried out. The published Response to the Consultation was also silent on this point.

97. In short, the arguments advanced on behalf of the Lord Chancellor do not begin to justify the failure to disclose the key analysis relied on in making the Decision. Instead, those arguments only serve to underline why its non-disclosure was unfair. We conclude that the failure to disclose this information was a fundamental flaw in the consultation process which made it so unfair as to be unlawful.

F. IRRATIONALITY

98. The second ground on which the Lord Chancellor's Decision is challenged encompasses a number of arguments falling under the general head of "irrationality" or, as it is more accurately described, unreasonableness. This legal basis for judicial review has two aspects. The first is concerned with whether the decision under review is capable of being justified or whether in the classic Wednesbury formulation it is "so unreasonable that no reasonable authority could ever have come to it": see Associated Picture Houses Ltd v Wednesbury Corp [1948] 1 KB 223, 233-4. Another, simpler formulation of the test which avoids tautology is whether the decision is outside the range of reasonable decisions open to the decision-maker: see e.g. Boddington v British Transport Police [1998] UKHL 13; [1999] 2 AC 143, 175 (Lord Steyn). The second aspect of irrationality/unreasonableness is concerned with the process by which the decision was reached. A decision may be challenged on the basis that there is a demonstrable flaw in the reasoning which led to it – for example, that significant reliance was placed on an irrelevant consideration, or that there was no evidence to support an important step in the reasoning, or that the reasoning involved a serious logical or methodological error. Factual error, although it has been recognised as a separate principle, can also be regarded as an example of flawed reasoning – the test being whether a mistake as to a fact which was uncontentious and objectively verifiable played a material part in the decision-maker's reasoning: see E v Secretary of State for the Home Department [2004] EWCA Civ 49; [2004] QB 1044.
99. In this case the Law Society has directed a miscellany of arguments at the reasoning which underpinned the Lord Chancellor's Decision. Some of these arguments are more substantial than others. As the test of unreasonableness is a high one, we can confine our consideration to what seem to us to be the more substantial arguments made.
100. We have summarised earlier the justification given for the Decision in the response to Consultation and other relevant documents. To recap, the main steps in the reasoning are these:

(1) The decision in the case of Napper in 2014 broadened the definition of PPE and resulted in electronic evidence being counted as PPE which the government had never intended to pay for.

(2) The increase in the annual cost of the Scheme attributable to the change in definition of PPE that followed the decision in Napper was approximately £33m.

(3) Reducing the PPE limit to 6,000 pages would reverse this increase and return the amount of electronic evidence paid for to pre-Napper levels.

101. The arguments advanced by the Law Society attack each step in this reasoning.

The legal effect of Napper

102. Counsel for the Law Society submitted that the Lord Chancellor mischaracterised the legal significance of the decision in the Napper case. They submitted that the Napper case did not change the law. Rather, it followed earlier decisions and was consistent with the plain language of the 2012 Regulations. Furthermore, the Lord Chancellor was wrong to suggest that electronic evidence such as mobile phone records could not be counted as PPE before the Napper decision and to suggest that such material is necessarily counted as PPE now. Under the current criteria electronic evidence can only be counted as PPE if the LAA considers that it is appropriate to include it having regard to the relevant circumstances such as the importance of the evidence to the case. Accordingly, the consultation documents are wrong to suggest that material which is of little relevance to the case now falls within the definition of PPE.
103. The Lord Chancellor's response, shortly stated, is that Napper was the key decision which led to the publication of new guidance in February 2015. This departed from the previous guidance which had reflected the intention of the Ministry of Justice that whether electronic evidence should be included in the PPE count depended solely on whether the evidence would have been printed by the prosecution and served in paper form prior to 1 April 2012. Under the revised guidance electronic evidence is counted as PPE even if the LAA assessor is unable to conclude that the material would have been printed prior to 1 April 2012, provided that the assessor considers it appropriate to include the material in the PPE count taking into account any relevant circumstances. The Lord Chancellor maintains that there is evidence which suggests that in practice this has led to large volumes of electronic evidence being counted as PPE in some cases where it would not previously have been counted, with the result that the level of PPE is no longer an appropriate measure of complexity in such cases. Counsel for the Lord Chancellor further submitted that whether this result flowed from what was decided in the Napper case itself or from the cumulative effect of a line of decisions is of little significance: what matters from a policy-making point of view is how Napper was understood by the Legal Aid Agency and reflected in its practice.
104. The characterisation of the Napper case and its significance in the consultation documents seems to us on any reasonable view to have been simplistic. The consultation documents gave the impression that before the Napper case electronic material such as telephone records was not counted as PPE, and that the Napper decision changed the definition of PPE to include such material even when it is of little relevance and can be searched electronically. It is plain that the position is in fact more nuanced and more complex than that. The evidence served in these proceedings indicates that the critical change which has occurred since the Scheme was established is the change in practice on the part of the Crown Prosecution Service in 2012 from printing out evidence (other than certain raw data held on CD-ROMs) and serving it in paper form to serving all evidence electronically. The Law Society's evidence indicates that before this change in practice occurred evidence such as mobile telephone records would, often at least, be served in paper form. Although his evidence is not consistent on this point, this appears to be acknowledged by Mr MacMillan when he explains a number of decisions prior to Napper which held that telephone records should be counted as PPE as turning on findings of fact that the documents would previously have been served in paper format.
105. The difficulty highlighted in the Napper case was that it was not always possible to determine whether or not evidence served by the prosecution would or would not previously have been served in paper form, as in the case of some material such as telephone records there had not been a uniform or consistent practice in that regard. That meant that an approach which made the sole criterion whether a document would have been printed and served in paper form before 1 April 2012 was unworkable. An additional test was therefore needed. It is clear that this was accepted by the Ministry of Justice. Had it not been accepted or had the approach applied in Napper been considered inappropriate, the Ministry of Justice could readily have amended the criteria set out in the 2013 Regulations.
106. It is also clear that the test applied in the Napper case is not a licence to count electronic evidence as PPE even if it is of little relevance to the case and electronically searchable. If the test is properly applied by the Legal Aid Agency, it should not result in such material being counted as PPE in the way suggested, for example, in the Response to Consultation when it stated:

"It is understood that large volumes of served evidence can contain material of less relevance to the client's defence, and in many cases is capable of being searched electronically, e.g. downloads of the entire contents of a mobile phone including thousands of irrelevant SMS messages and pictures."

The implication was that such material now falls within the definition of PPE, when that is manifestly not the case.

107. Nevertheless, the Ministry of Justice had evidence to support its view that following the Napper decision there was in fact an increase in the number of cases in which large volumes of electronic evidence were counted as PPE. As set out in the consultation paper, comparison of the number and cost of cases with different levels of PPE in 2013-14 and 2015-16 showed that the number of cases with high levels of PPE increased very significantly during this period. Whereas the number of cases with less than 6,000 PPE increased by only 1%, the number of cases with between 6,000 and 9,999 PPE increased by 28% and the number with 10,000 PPE or more tripled over this period. Over the same period the proportion of cases with more than 10,000 PPE in which special preparation fees were awarded decreased from around 45% to around 20% of cases. These figures tended to suggest that there had been a significant increase in cases in which large volumes of material were being counted as PPE which did not involve a corresponding increase in workload.
108. The draft Impact Assessment published with the consultation paper also contained a breakdown of numbers of cases and expenditure by offence category. This showed large increases in expenditure and in the average cost per case in two categories in particular which were not matched by increases in other categories of case. These two categories were a category which included serious drug offences and the category which comprised offences of dishonesty of high value. These are the categories of case in which, of their nature, large volumes of electronic evidence are most likely to be served. A further analysis described in the Impact Assessment compared the average costs per bill for the Scheme over the period from 2012-13 to 2015-16 with the average costs per bill for the Advocates Graduated Fee Scheme over the same period. This comparison indicated that the average costs of the Scheme had increased since 2013-14, whereas the average costs of the Advocates' Scheme had remained broadly constant. The Ministry of Justice regarded this finding as significant, as PPE has less effect on cost in the Advocates' Scheme.
109. None of these analyses proved conclusively that there had been a significant increase in expenditure following the decision in Napper which was attributable to greater readiness to count electronic evidence as PPE. But they provided some support for that hypothesis. The fact that other interpretations of the relevant statistics are possible does not show that the view of the Ministry of Justice was unreasonable. In the circumstances we think it impossible to say that there was no or no proper evidence to support the belief that a change in the approach to electronic evidence following the decision in Napper was likely to have contributed to an increase in the cost of the Scheme.

Was the policy disproportionate?

110. It was argued on behalf of the Law Society that, even if there was reasonably perceived to be a problem as set out in the consultation paper that large volumes of electronic evidence were being inappropriately counted as PPE, the rational and proportionate solution to the problem was to target the perceived mischief by restricting the definition of PPE to exclude such evidence. It was submitted that the response which was instead adopted of reducing the PPE cap is inherently disproportionate because it applies to all evidence, including evidence which was not originally in electronic form, and therefore does not target the problem it is seeking to solve. It was further submitted that this approach was bound to unbalance the Scheme because in cases with more than 6,000 PPE the graduated fee would no longer reflect the complexity of the case. The fact that special preparation fees can in principle be claimed for time spent reading pages of evidence in excess of the 6,000 page limit cannot adequately mitigate the impact of reducing the limit – first, because such fees are often uneconomic to claim given the time and therefore cost involved in claiming them, the low rates paid and the risk that they will not be allowed and, second, because such payments do not compensate in any event for additional work generated by the evidence such as making enquiries and obtaining further evidence for the defence.
111. When the policy was being developed, officials at the Ministry of Justice considered the option of carving out certain types of evidence (e.g. telephone records and cell site analysis) from the definition of PPE but concluded in a Ministerial Submission dated 16 September 2016 that this was not a viable option "as there is less certainty about the extent to which this option would address the PPE pressure as we have no data on electronic evidence on which to cost it." The points were also made that this option would leave scope for argument in grey areas and would catch large volumes of lower value cases where there is a small volume of electronic material, which was not considered to be a problem. The view was taken that it would not be proportionate, either for litigators or for the Legal Aid Authority, to have to assess small value claims for reading electronic evidence under the special preparation provisions. The alternative option of reducing the PPE limit was preferred for the reasons that:

"It provides more certainty around returning average expenditure to pre-Napper levels, less room for argument about what is or what is not PPE and, in operational terms, it is much easier for case workers to assess."

112. We think it impossible to say that the reasons given for rejecting the option of restricting the definition of PPE to exclude certain types of electronic evidence were irrational and not reasons on which the Lord Chancellor could reasonably rely. The discussion in the Ministerial Submission, however, highlights the extent to which the approach adopted was based on inference rather than any direct evidence. In particular, the Ministerial Submission makes it clear that the Ministry of Justice had no data on the proportion of PPE that was accounted for by electronic evidence. It is therefore apparent that the hypothesis that in cases with high levels of PPE large volumes of electronic evidence of comparatively little relevance were being included was not supported by any empirical study.
113. We accept that in principle it was open to the Lord Chancellor to adopt a policy response which did not directly correspond to the problem which it was designed to meet. A policy-maker may reasonably decide that the disadvantages of a finely tuned solution to a problem outweigh its advantages and that a broader measure is preferable, even if the broader measure is both over- and under-inclusive in that it catches some cases in which there is no or no significant problem and fails to catch some cases in which the problem occurs. Such an approach is in any event consistent with the nature of the Scheme, which uses criteria such as PPE as proxies for the complexity of cases. It is inherent in the use of such proxies that they will result in under-compensation in some cases. But this does not cause unfairness if it is off-set by over-compensation in other cases. What matters is that overall a reasonable balance is struck.
114. Thus, in the present case there was nothing inherently unreasonable in responding to the perceived problem that some electronic material was being inappropriately counted as PPE by reducing the PPE threshold. The Law Society's submissions that this was inherently disproportionate and bound to unbalance the Scheme cannot be accepted. The Scheme would only become unbalanced if there was no reasonable relationship between the amount of the projected reduction in payments driven by PPE and the amount by which such payments had increased as a result of the Napper case.
115. The rationale and rationality of the policy response, however, depended critically on there being such a reasonable relationship. As already indicated, it was projected that reducing the PPE cap from 10,000 to 6,000 pages would reduce payments under the Scheme by between £26m and £36m a year (depending on the extent to which the reduction in the PPE limit resulted in an increase in special preparation payments). The justification for this policy thus crucially depended on the estimate made by the Legal Aid Agency, and relied on by officials and by the Lord Chancellor, that the amount of the increase in PPE payments attributable to a change in definition following Napper was approximately £33m and therefore broadly comparable to the saving that would be made as a result of the proposed measure.

The LAA analysis

116. Mr MacMillan has explained in his witness statement how the figure of £33m was calculated by the Legal Aid Agency. A comparison was made between the average costs of cases in two different periods. According to Mr MacMillan, the intention was to compare data for the financial year 2014-15 with data for the years 2015-16; but by mistake the comparison made was between data for the year 2013-14 and 2014-15. The data were split into different offence categories and, within each category of offence, between three types of case: jury trials, cracked trials and guilty pleas. The number of cases for each type in each offence category in the earlier year (2013-14) was compared with the number in the later year (2014-15) and the percentage increase or decrease was calculated. The expenditure for each type of case in each offence category was then adjusted by the percentage change in number of cases to estimate what the expenditure would have been in the later year if the number of cases had remained constant. This figure was then compared with the actual expenditure in the later year in order to work out what amount of money was attributable to "price changes" (that is, changes in average case cost). The assumptions were made, first, that any increases in price calculated in this way were a result of increases in PPE – since, according to Mr MacMillan, PPE is the "fundamental driver" of costs in the Scheme – and, second, that such increases in PPE were attributable to changes in the understanding of what should be counted as PPE following the Napper decision.
117. Mr MacMillan explained in his witness statement that in January 2018, in revisiting the LAA analysis in order to respond to points raised by the Law Society in a letter sent before commencing this claim, the "error in the input data" was identified and the calculation was then repeated comparing data in the periods July 2014 to June 2015 and July 2015 to June 2016. On this "slightly revised approach", it was estimated that the long-term increase in annual expenditure attributable to the widening of the scope of PPE following Napper was £31m a year rather than the figure of £33m which had been put to Ministers.

Flaws in the LAA analysis

118. The expert report of Professor Adams identifies what she describes as "two key flaws" in the methods used to calculate the figure of £33m (and the revised figure of £31m).
119. The first is that the LAA analysis took no account of the length of trials and the number of defendants, both of which are factors which have a significant effect on costs. Thus, in 2012-13 the mean cost of a case involving a trial of at least 40 days was £50,164 compared to £1,502 for cases lasting less than 40 days. Between 2012-13 and 2015-16 the number of serious drug and high value fraud cases with trials of 40 days or more almost doubled. Given that longer trials and multiple defendants are associated with higher average case costs, it was fallacious to assume that the price increases calculated by the LAA were the result of increases in PPE. That assumption could only properly have been made if the analysis had controlled for trial length and number of defendants, which the LAA analysis did not.
120. The second flaw identified by Professor Adams is that the analysis took no account of pre-existing cost trends. In her report, Professor Adams pointed out that in the period before the Napper case the average cost of certain types of case was already rising. For example, between 2012-13 and 2013-14 the average cost of all trials increased by 16% and the average cost of cases involving serious drug offences and high value fraud offences increased by 14%. In these circumstances it was fallacious to assume that price increases following the Napper decision were attributable to that decision (or to changes in the definition of PPE reflected in that decision) as opposed to representing the continuation of a pre-existing trend which, since it was already manifest before the Napper decision, cannot have been caused by that decision. It is apparent that this is an example of the logical fallacy known as the post hoc ergo propter hoc fallacy.
121. As Professor Adams pointed out, the very fact that the two calculations made by the LAA produced similar figures – £33m and £31m – although the periods compared were different, itself tends to suggest that there was a pre-existing cost trend. Since the Napper case was decided in November 2014 and the revised LAA guidance was published in February 2015, it might be expected that any effect of Napper in the 2014-15 financial year would be comparatively limited and the effect much greater thereafter. Yet the original LAA analysis relied on in making the Decision estimated a price increase of £33m between 2013-14 and 2014-15 which was assumed to be attributable to the change in definition of PPE which followed the decision in Napper.
122. As mentioned earlier, the expert evidence served on behalf of the Lord Chancellor in response to Professor Adams' report took issue with other aspects of her report but said nothing in answer to these fundamental criticisms of the LAA analysis. Nor was any other evidence served or argument made on behalf of the Lord Chancellor which sought to suggest that her criticisms of the LAA analysis were invalid or that the reliance placed on that analysis by the Ministry of Justice was reasonable or justifiable. In these circumstances, we see no escape from the conclusion that the LAA analysis was vitiated by methodological flaws and that no reasonable decision-maker could reasonably have treated the figure of £33m produced by that analysis as an estimate of increased expenditure attributable to the Napper decision on which reliance could reasonably be placed.
123. Instead of attempting to defend the LAA analysis, Mr McHale undertook a much more sophisticated multi-variate regression analysis. On two different versions of this analysis he estimated the "Napper effect" as £27.8m or £32.5m. There is a substantial dispute between Mr McHale (supported by Professor Santos Silva) and Professor Adams, who found no statistically significant effect, about the appropriate methodology for an analysis of this more sophisticated kind. That dispute could not be resolved in these proceedings and, for the reasons given earlier, we have refused each party permission to rely on the expert evidence in which the rival analyses and arguments addressing their respective merits are contained. The point that needs to be emphasised for present purposes is that none of this evidence is relevant to the issue that we have to decide or gainsays the conclusion that the Decision was unlawful because it was reached on the basis of a flawed analysis on which no reasonable authority would have relied. Indeed, the very nature of the modelling exercise undertaken by Mr McHale merely confirms that the exercise actually undertaken was defective. The fact – if it be the fact – that a similar conclusion could have been reached if a properly conducted analysis had been undertaken is nothing to the point.
124. We therefore conclude that the Decision to reduce the PPE cap to 6,000 was also unlawful for this further reason.

G. PREVENTING ACCESS TO JUSTICE

125. The Law Society's remaining ground of challenge is that the Decision violated the common law right of access to justice. The Law Society relies on what is now the leading case of R (UNISON) v Lord Chancellor [2017] UKSC 51; [2017] 3 WLR 409 in which seven Supreme Court justices unanimously held that a Fees Order made by the Lord Chancellor prescribing fees payable for bringing and proceeding with claims in employment tribunals and appeals to the employment appeal tribunal was unlawful. The essential reasoning of the Supreme Court was that the statutory power to prescribe tribunal fees had to be interpreted in the light of constitutional principles including the constitutional right of access to justice. So interpreted, the relevant statutory power did not authorise the making of an order if there was a real risk that its effect would be to prevent some people from having access to courts or tribunals or if it impeded such access to a greater degree than was justified by the legitimate objectives of the order. Lord Reed (who gave the judgment of the court) concluded that the Fees Order infringed these principles and was therefore ultra vires because it set fees at a level which some people could not reasonably afford and which also made it futile to bring many claims in view of the level of the fees and the prospects of recovering them.
126. Counsel for the Law Society argued that these principles are applicable in the present case. They submitted that, by analogy with article 6 of the European Convention on Human Rights, the common law right of access to justice includes the right of a person charged with a criminal offence who cannot afford to pay for legal assistance to receive legal assistance paid for by the state when the interests of justice so require. They submitted that section 1 of the Legal Aid, Sentencing and Punishment of Offenders Act 2012, which imposes an obligation on the Lord Chancellor to ensure that legal aid is made available in accordance with Part 1 and the powers to make regulations conferred by that Part of the Act must be interpreted in the light of the rights of criminal defendants to publicly funded legal assistance. On this basis they argued that the 2017 Regulations are unlawful because there is a real risk that they will effectively prevent some people charged with criminal offences from receiving effective legal representation or in any event will have a disproportionate effect on the right of access to justice of some criminal defendants.
127. In support of its case there is a such a risk, the Law Society relies in particular on studies undertaken in 2013-14 which concluded that the market for criminal defence work could not sustain a second fee cut of 8.75% and concerns raised by over a third of the respondents to the consultation, including the Law Society itself, that the proposed reduction of the fee cap to 6,000 PPE might make it no longer economic for firms to undertake large Crown Court cases or potentially to undertake any Crown Court cases since many firms rely on the profits from large cases to subsidise other work. Counsel for the Law Society also relied on what they contended was the complete failure of the Lord Chancellor to carry out any proper assessment of the likely impact of the proposed measure on the sustainability of firms engaged in criminal law practice.
128. On behalf of the Lord Chancellor, Mr Chamberlain submitted that the Law Society's argument under this head amounts to a challenge to a predictive judgment made by the Lord Chancellor that reducing the PPE cap will not make the market for legally aided criminal defence work unsustainable and will not materially harm access to justice by preventing defendants from obtaining proper advice and representation. He argued that such a challenge can only succeed if the Law Society can satisfy the court that this judgment was not one which any reasonable Lord Chancellor could have made.
129. We accept – and did not understand the Lord Chancellor to dispute – the Law Society's submission that the constitutional right of access to justice includes the right of those accused of criminal offences to be given publicly funded legal advice, assistance and representation when they cannot afford to pay for such services, if the interests of justice require it. Such a right is embodied in article 6(3)(c) of the European Convention on Human Rights and we see no reason to think that anything less is required by common law principles of procedural fairness. That such principles may require the provision of legal aid in particular categories of case is confirmed by the decision of the Court of Appeal in R (Howard League for Penal Reform) v Lord Chancellor [2017] EWCA Civ 244; [2017] 4 WLR 92. In that case the Court of Appeal held that the removal of legal aid from certain categories of decision-making involving prisoners was unlawful because it gave rise to an unacceptable risk of unfairness. The test applied by the Court of Appeal, described as a "high test", was whether the risk of unfairness was inherent in the policy or system itself, as opposed to there merely being a possibility of individual instances of unfairness: see paras 48-50. That assessment was to be made taking account, in particular, of the importance of the issues at stake, the complexity of the procedural, legal and evidential issues and the ability of individuals to represent themselves without legal assistance, having regard to their age and mental capacity: see paras 47 and 51.
130. Although the Howard League case was decided shortly before judgment was given by the Supreme Court in the UNISON case and was not cited in the UNISON case, the approach of the Court of Appeal seems to us to be consistent with that subsequently taken by the Supreme Court.
131. In the present case, the 2017 Regulations challenged by the Law Society do not provide for the removal of legal aid from any category of defendant or proceeding (as in the Howard League case), nor do they impose a cost on obtaining access to the courts (as in the UNISON case). Rather, the effect of the Regulations is to reduce certain fees paid to lawyers out of public funds for providing litigation services to defendants who qualify for legal aid in Crown Court proceedings. Applying the principles we have discussed, to establish that the 2017 Regulations are unlawful, it would need to be shown that there is a real risk that the fee reductions imposed by the Regulations will result in some criminal defendants being denied adequate legal assistance. It further seems to us that what would need to be demonstrated is not just a possibility of individual instances of unfairness which would not undermine the integrity of the system but a risk of systemic unfairness of the kind discussed in the Howard League case.
132. We do not agree with Mr Chamberlain's submission that the assessment of whether such a risk exists is solely a matter of predictive judgment for the Lord Chancellor to make and which is only susceptible to challenge if the court is satisfied that the Lord Chancellor's assessment was irrational. That, in our view, misconceives the nature of this ground of challenge. This ground does not involve an argument that the Lord Chancellor has failed to exercise his powers in accordance with common law duties which govern the exercise of public functions, including the duty to act reasonably. It is an argument that the measure adopted interferes with individual rights. As in any case where a rights violation is alleged, the question is not whether the decision-maker properly considered whether there would be an unlawful interference with individual rights but whether there has in fact been such an interference (or, as is alleged in this case, a real risk of systemic interference). That is a question for the court to decide for itself. In doing so, the court may give weight to opinions of the decision-maker which reflect its institutional expertise. But it is the court's judgment that is determinative. This is the well-established approach in relation to Convention rights: see e.g Belfast City Council v Miss Behavin' Ltd [2007] UKHL 19; [2009] 1 WLR 1420, paras 15, 31, 44. It is equally applicable to common law rights, as it follows from the constitutional role of the judiciary whereby questions of legal right are the province of the courts.
133. This was how the Court of Appeal approached the matter in the Howard League case (see para 38). Likewise, in the UNISON case the Supreme Court did not treat the impact of fees on court users as a matter for the judgment of the Lord Chancellor, reviewable only in accordance with a test of reasonableness. The court considered for itself the evidence relating to the impact of the Fees Order and concluded that the order effectively prevented access to justice.
134. In the UNISON case there was substantial evidence which showed that the fees imposed by the Lord Chancellor were likely in practice to be unaffordable by many prospective claimants on low to middle incomes, as well as evidence that the Fees Order had indeed deterred many people from exercising their rights to bring claims. By contrast, there is no evidence before the court in this case capable of substantiating the risk asserted by the Law Society.
135. In particular, the studies carried out in 2013-14 on which the Law Society has sought to rely were concerned with a different question: the capacity of the market to sustain a second fee reduction of 8.75% across the board. Those studies did not consider the impact of reducing the PPE cap, which was not then in contemplation. Nor does the fact that respondents to the consultation raised concerns that reducing the cap to 6,000 PPE would make criminal defence work or some categories of such work uneconomic prove the validity of those concerns. We also note that the 2017 Regulations have now been in force for several months and it was not suggested at the hearing that they have as yet had any material adverse effect on the availability of legally aided advice and representation.
136. In short, the evidence relied on by the Law Society does not show, or provide any reasonable basis for inferring, that there is a real risk that the 2017 Regulations will have the systemic effect of preventing some criminal defendants from obtaining adequate legal assistance. We accordingly reject this ground of challenge.

H. REMEDY

137. Although the grant of relief when a claim for judicial review succeeds is in the discretion of the court, it would normally follow from the conclusion that the challenged decision was unlawful that the decision should be quashed. However, this is now subject to section 31(2A) of the Senior Courts Act 1981, introduced in 2015, which provides that the High Court must refuse to grant relief on an application for judicial review:

"if it appears to the court to be highly likely that the outcome for the applicant would not have been substantially different if the conduct complained of had not occurred."

138. Although no reliance was placed on section 31(2A) in the Lord Chancellor's detailed grounds of resistance to the claim or main skeleton argument for the hearing, in addressing the question of what, if any, expert evidence should be admitted, the Lord Chancellor advanced an argument (mentioned earlier) that, if he was given permission to rely on the expert evidence served in response to Professor Adams' report, the court should infer from that evidence that it is highly likely that the outcome for the Law Society would not have been substantially different if the LAA analysis had been disclosed during the consultation process. We understood the argument to be that, if the LAA analysis had been disclosed and the Law Society had then made the criticisms of the methods used in that analysis which Professor Adams has made, it is highly likely that Mr McHale would have been consulted and would have carried out a modelling exercise of the kind which he carried out in responding to the evidence of Professor Adams in these proceedings. We have mentioned that the analysis undertaken by Mr McHale produced an estimate of the increase in cost attributable to the Napper decision of £27.8m or £32.5m (depending on which of two different modelling techniques was used). It is suggested that on this hypothesis it is highly likely that the Lord Chancellor would have made the same decision as he in fact made to reduce the PPE cap to 6,000 pages.
139. As we have not admitted in evidence the relevant part of Professor Adams' report and have refused the Lord Chancellor permission to rely on the evidence of Mr McHale which contains his own modified version of Professor Adams' independent analysis, the basis on which Mr Chamberlain indicated that the Lord Chancellor would seek to rely on section 31(2A) has not arisen. But we think it right to indicate why, even if Mr McHale's evidence (and the evidence of Professor Santos Silva supporting the validity of Mr McHale's methods) had been admitted, we would have rejected the argument based on section 31(2A).
140. First, we see no reason to assume that, if the LAA analysis had been disclosed in the consultation process, as we have held that it should have been in order to comply with the Lord Chancellor's duty of procedural fairness, this would have resulted in the Ministry of Justice carrying out an analysis of the kind that Mr McHale has carried out for the purpose of this litigation in order to respond to Professor Adams. Even if it is assumed that the Law Society would have included in its response to the consultation an expert report similar to the report prepared by Professor Adams for the purpose of these proceedings, it is impossible to know what representations other consultees would have made in response to the LAA analysis. Nor can it be inferred from the fact that Mr McHale has produced the analysis that he has in the context of adversarial litigation that he would have been asked to undertake an analysis or, if asked, would have produced a similar analysis to that which he has produced in response to Professor Adams' report in the hypothetical scenario under consideration. Moreover, if that would have happened and if it was proposed to rely on Mr McHale's analysis instead of the LAA analysis, further consultation might have been necessary and it is impossible to know what further representations consultees would then have made.
141. More fundamentally, the final decision was that of the Lord Chancellor, not his officials, and the Lord Chancellor would have had to consider the responses to the consultation with an open mind. It is one of the requirements of proper consultation set out in the Coughlan case at para 108 (quoted at para 67 above) that the product of consultation must be conscientiously taken into account when the ultimate decision is taken. It would be wrong in principle for the court in a case where the hypothetical decision would have been made on the basis of materially different information and advice from the actual decision to make a judgment expressed as a high likelihood about what the Lord Chancellor would have decided. To do so would involve trespassing into the domain of the decision-maker, as it was well put by Lindblom LJ in R (Williams) v Powys City Council [2017] EWCA Civ 427; [2018] 1 WLR 439, para 72.

Result

142. We have found that the Decision implemented by the 2017 Regulations was unlawful because the key analysis relied on in making the Decision (1) was not disclosed to consultees, rendering the consultation process unfair, and (2) used methods that were statistically flawed, making it irrational to rely on the analysis.
143. In the circumstances, the orders that we think it right to make are to give permission to proceed with the claim, declare the Decision to be unlawful and quash the 2017 Regulations. We will deal with any consequential orders, if not agreed, after considering written submissions.