THE POLICE SUPERINTENDENTS’ ASSOCIATION (R on the application of) v HER MAJESTY’S TREASURY

In these proceedings the Police Superintendents’ Association (“PSA”) challenges the legality of the consultation on “Public service pension schemes: changes to the transitional arrangements to the 2015 schemes” (“the consultation”) and the decision announced on 4 February 2021 to close legacy public service pension schemes, including the police schemes, and move all active members to reformed pension schemes from 1 April 2022 (“the closure decision”). As I will come on to address in more detail, the relevant decision-maker was the Chief Secretary to the Treasury (“CST”). The Interested Party supports the Defendant’s responses in these proceedings and has not advanced separate submissions.

The PSA is concerned about the impact of the closure decision on officers in the Police Pension Scheme 1987 (“the 1987 Scheme”) and the New Police Pension Scheme 2006 (“the 2006 Scheme”), both of which would be closed from 1 April 2022, with active members transferred to the reformed police scheme established by the Police Pension Regulations 2015 SI 2015/445 (“the 2015 Regulations” and “the 2015 Scheme”). Under transitional arrangements contained in Schedule 4 of the 2015 Regulations, certain members of the 1987 Scheme and the 2006 Scheme (collectively, “the police legacy schemes”) were able to remain in those schemes for what was then an unlimited period. The PSA rely on representations made that members with this transitional protection (“TP”) could remain in their police legacy schemes until they retired, even if this was after 1 April 2022 (“the representations to police”). In the Detailed Grounds of Resistance (“DGR”), the Defendant accepted that these representations were made.

Permission to apply for judicial review was granted on the papers by Cheema-Grubb J. on 9 August 2021. At that stage five grounds of challenge were relied upon. In summary, the four grounds now pursued are that: the consultation undertaken was unlawful (“Ground 1”); there was a failure to comply with the public sector equality duty contained in s.149, Equality Act 2010 (“PSED”) (“Ground 2”); the representations to police gave rise to a substantive legitimate expectation that has been unlawfully breached by the closure decision (“Ground 3”); and the closure decision was flawed by material error of fact, in that it was premised on the basis that all members of the public service pension schemes who had received TP would have reached their Normal Pension Age in their respective legacy schemes by 1 April 2022 (“Ground 4”). The Claimant no longer relies upon the fifth ground, namely that there was a failure to take into account the substantive legitimate expectation. By way of relief, the Claimant seeks an order quashing the consultation and the consultation response and/or declarations that the consultation was unlawful and that the decision to close the police legacy schemes from 1 April 2022 was unlawful.

The Defendant takes issue with each of the grounds and asserts that in any event the court should refuse to grant relief, by reason of s.31(2A), Senior Courts Act 1981 (“SCA 1981”) and/or infringement of Parliamentary privilege. The latter contention arises from the fact that closure of the public service pension legacy schemes is being implemented by Clause 76 of the Public Service Pensions and Judicial Offices Bill (“the Bill”) that is currently proceeding through Parliament (Footnote: ¹). It is common ground that these proceedings will be academic if they are not determined before the enactment of this provision. For the avoidance of doubt, the Defendant accepts that following the grant of permission, the claim is justiciable and that the court should determine the Claimant’s grounds of challenge.

The bundles of documents before the court comprised nearly 1,600 pages. The Claimant relies upon three witness statements from Chief Superintendent Daniel Murphy, the National Secretary of the PSA, made on 11 May 2021 (“Murphy 1”), 13 October 2021 (“Murphy 2”) and 15 November 2021 (“Murphy 3”); and a statement from Superintendent Emma Richards made on 11 May 2021 (“Richards 1”). The Defendant relies upon three witness statements from Eleanor Tack, the Deputy Director, Workforce Pay and Pensions in HM Treasury (“HMT”), made on 1 October 2021 (“Tack 1”), 1 November 2021 (“Tack 2”) and 5 November 2021 (“Tack 3”).

It is apparent from the parties’ respective submissions that the substantive issues for the court’s determination are as follows:

was the consultation undertaken at a time when the relevant proposal was still at a formative stage;

was adequate information provided to consultees to enable them to properly respond to the consultation exercise;

did the decision-maker give conscientious consideration to the consultees’ responses;

was there a breach of the PSED in relation to the closure decision;

did the representations to the police give rise to an enforceable legitimate expectation;

if so, was the Defendant entitled to depart from it in making the closure decision;

in making the closure decision did the Defendant wrongly believe that all members of the affected public service pension schemes who had received TP would have reached their Normal Pension Age in their respective legacy schemes by 1 April 2022;

if so, was this a material error (Footnote: ²);

should the court refuse relief pursuant to s.31(2A) SCA 1981 on the basis that it is highly likely that the outcome for the Claimant would not have been substantially different if the conduct complained of had not occurred;

would granting the relief sought impermissibly interfere with Parliamentary privilege; and

if the grant of relief is not otherwise precluded, what relief, if any, should the court grant in its discretion?

The proceedings were filed on 4 May 2021. When permission was granted the Claimant’s request for expedition was refused and directions were made for the Defendant to file its DGR and any written evidence relied on within 35 days of service of the Order. By Order of 13 September 2021, the timetable was extended by Swift J, permitting the Defendant to file its DGR and evidence by 1 October 2021, with any application by the Claimant to rely on evidence in reply to be filed by 13 October 2021. It was accepted that the substantive hearing should take place this term (given the proceedings would become academic after the Bill was enacted) and a hearing in the week commencing 15 November 2021 was directed. Directions were also given for the Claimant’s Skeleton to be filed and served by 1 November 2021 and the Defendant’s by 8 November 2021. The DGR and Tack 1 plus exhibits were filed in accordance with this timetable.

After considering the Defendant’s disclosure, the Claimant’s solicitors wrote to the Government Legal Department (“GLD”) saying: “We had previously understood that the individual who took the decision subject to challenge was the Chancellor of the Exchequer however, in light of the recent disclosure, it appears that the decision maker was actually the First Secretary to the Treasury (see, inter alia the Submissions to Ministers of 28 January 2020 and 17 April 2020). Please confirm that is the case.” The GLD replied on 13 October 2021: “The Chief Secretary to the Treasury (CST) is the Minister leading on public service pensions. The consultation, including the response, was by collective agreement and write-around (Footnote: ³) across government, as set out in” Tack 1.

The Claimant filed Murphy 2 in accordance with Swift J’s directions and filed its Skeleton on 1 November 2021. Reference was made therein to the recent correspondence concerning the identity of the decision-maker. The Claimant also emphasised that the Defendant had not provided any Ministerial Submissions that post-dated the consultation document, nor any other record of the decision-maker taking the closure decision after the consultation.

On 1 November 2021 the Defendant applied to rely on Tack 2 and its supporting exhibits. Amongst other matters, Tack 2 responded to material in Murphy 2 relating to the impact of the closure decision. However, it also referred to and exhibited documentation relating to the decision-making under challenge, including a Ministerial Submission dated 10 December 2020, an email chain between HMT officials in December 2020, the CST’s letter to the Prime Minister of 16 December 2020 and a letter dated 13 January 2021 from the Cabinet Secretariat to the CST.

On 5 November 2021 the Defendant applied to rely on Tack 3. This statement addressed the question of who was the decision-maker, saying (para 7): “…the CST is the key decision-maker in relation to the public service pensions and it is the CST who decides on policy proposals. This includes the decision under challenge.” Reference was also made to the collective agreement process by “write-round”. Para 14 onwards of the statement dealt with the CST’s consideration of the consultees’ responses and in particular made detailed reference to the 10 December 2020 Ministerial Submission and related documents and email chains. Further material on the extent of the adverse impact of the closure decision was also provided.

The Claimant objected to the applications relating to Tack 2 and Tack 3, pointing out that they had been served significantly outside of Swift J’s timetable. It was submitted that the exhibited documentation should have been disclosed earlier; that the failure to do was a breach of the duty of candour; and that the Claimant was prejudiced given the limited time before the hearing and the fact it had already served its Skeleton.

I considered the applications relating to Tack 2 and Tack 3 on the papers. By this time, the Defendant’s Skeleton, dated 8 November 2021 had been filed. As I observed in my Order made on 10 November 2021, the late filing of the Defendant’s additional evidence was “highly unsatisfactory”. I rejected the assertion that it had been necessitated by the Claimant raising new grounds, since the legitimacy of the closure decision and its relationship to the consultation process had been in issue from the outset. It was clear to me that the documentation relating to the decision-making process should have been disclosed earlier in accordance with Swift J’s directions. Nonetheless, I gave the Defendant permission to rely on Tack 2 and Tack 3, observing in para 3 of my Reasons that given the nature of the Claimant’s grounds of challenge, “it is in the interests of justice for the Court to have as comprehensive a picture as is reasonably possible of the decision-making process that was undertaken. The Defendant’s additional evidence bears directly on these matters.” However, in granting permission I was concerned to ensure that the Claimant was not unfairly prejudiced. I therefore gave permission for the Claimant to rely on evidence in response to Tack 2 and Tack 3 and/or a supplementary Skeleton, if filed by noon on 15 November 2021 (the day before the hearing). As I observed at para 4 of my Reasons, in the circumstances the Defendant could have no grounds for complaint about the relatively short time it would have to consider these materials. Nonetheless, I did stress that any supplementary material served by the Claimant “must be confined to responding to” Tack 2 and Tack 3.

Murphy 3 was filed and served in accordance with my Order. The Claimant also filed and served a Supplementary Skeleton within the stipulated timescale. Much of it responded to the Defendant’s additional evidence, as my Order had permitted. However, the Claimant also contended (at paras 39 and 43) that the decision-maker was actually the Domestic and Economic Implementation Committee (“the DEI Committee”), rather than the CST, who, it was said, had only made a provisional decision. In turn, this was said to support the Grounds, for example, as the Defendant could not show that conscientious consideration of the consultation responses had been undertaken by the DEI Committee, as there was no documentary evidence indicating it had been supplied with the relevant material (para 46). Footnote 14 accepted: “this argument was not advanced in either the statement of facts and grounds or skeleton argument”.

At the outset of the hearing before me on the morning of 16 November 2021, the Defendant objected to the Claimant raising the argument that the decision-maker was the DEI Committee at this stage, contending that permission would be required to advance this, which I should refuse. The Defendant also applied to rely on some additional documentation, which the Claimant opposed. I then heard submissions on these matters. The parties agreed with my proposal that I would give oral rulings with summary reasons at that stage, with a fuller account to be given in this judgment (including the procedural history I have just recited), in order to avoid further delay or disruption to the progress of the hearing. The hearing had been listed for two days. In the event, addressing these matters took up the first two hours of the hearing. I therefore extended the hearing for a further half-day to enable sufficient time for oral submissions on the substantive matters. I will now address the rulings that I gave. For the avoidance of doubt, the time for any application for permission to appeal these rulings will run from the date when this judgment is handed-down.

I refused to permit the Claimant to advance a case based on the proposition that the relevant decision-maker was the DEI Committee, ruling that the hearing would proceed on the basis that the CST was the relevant decision-maker. My reasons are set out in the paragraphs that follow.

I bore in mind the “highly unsatisfactory” position I had identified in my earlier Order in relation to Tack 2 and Tack 3, in particular the Defendant’s delay in disclosing documentation relating to what would be expected to be a key period in the decision-making process and, in turn, the difficulties this had caused for the Claimant.

However, the permission I had granted in the 10 November Order to file a Supplementary Skeleton so close to the hearing date was explicitly confined to responding to Tack 2 and Tack 3 and was aimed at rectifying the consequential prejudice to the Claimant occasioned by the late service of that material. Importantly, the new argument which the Claimant now sought to advance did not arise from Tack 2 and/or Tack 3.

The proposition that decisions were made by the CST and were then subject to collective agreement was contained within Tack 1, see in particular paras 109 and 114 and references to the Ministerial Submissions exhibited to it. Mr Sharland QC pointed out that Tack 1 did not deal in detail with the post-consultation decision-making, nor exhibit documents from that period (as opposed to CST’s decisions regarding the launch of the consultation). However, in so far as that gave rise to any uncertainty over whether the same decision-making structure applied, this was put beyond doubt by the GLD’s letter of 13 October 2021 that I have already described. Although Mr Sharland suggested that there was still some ambiguity, the Claimant’s solicitors did not write again seeking any further clarification, despite corresponding on various other issues. The 13 October letter from GLD was over a fortnight before the Claimant’s Skeleton was due.

Nonetheless, the proposition that the decision was made by collective agreement and was unlawful because the Defendant had provided insufficient evidence to show that the consultation responses and/or the Equality Impact Assessments were considered by the collective decision-maker, was not put forward in the Claimant’s Skeleton of 1 November 2021. For the reasons I have just identified, if the Claimant wished to rely on this contention, it could have been advanced at that juncture. This position was not negated by the fact that additional material that potentially bore on this point was served subsequently (in particular the 13 January 2021 letter from the Economic and Domestic Affairs Cabinet Secretariat exhibited to Tack 2). Further, in so far as significance was placed upon them for these purposes, the Cabinet Manual and Ministerial Code exhibited to Tack 3 were publicly available documents.

Additionally, it was very late in the day to raise a new argument that had not been pleaded. The Supplementary Skeleton was filed and served less than 24 hours before the start of the hearing. If permission had been granted to advance this contention, the hearing timetable would have been significantly disrupted, in circumstances where it was not possible to simply adjourn the hearing, given the considerations that had prompted this early listing. The new argument had not been formulated in proposed amended Grounds. The scope of the parties’ submissions and the material relied on meant that, even with the existing arguments and the Court’s pre-reading, it would be tight to complete the hearing within the available time. The Defendant had indicated it would want to respond to the new argument in writing if the Claimant was permitted to advance it; and in these circumstances it would also want to apply to rely on further documentation.

The Defendant would be prejudiced if permission was given, in light of the very short notice. Although Ms Callaghan QC indicated she was in a position to “make a stab” at dealing with it, if the Court did grant permission, she indicated she had not had sufficient time to prepare a detailed response, in circumstances where the contention raised complex constitutional issues as to the respective responsibilities of ministers and Cabinet in terms of collective decision-making and who was to be regarded in law as the decision-maker and for what purposes. I accepted this. The prejudice was self-evident.

Although it was desirable for the court to have the full legal and evidential picture and the Defendant had been remiss in respect of the Tack 2 and Tack 3 material, when I made the 10 November 2021 Order, I was able to make directions mitigating the consequential prejudice to the Claimant. At this very late stage it was not possible to mitigate the prejudice to the Defendant if I permitted the argument to be advanced.

At the start of the afternoon, Mr Sharland asked me to clarify whether my ruling meant that he was precluded from relying on certain material in the document bundles. I confirmed that he was not restricted in the material he referred to, provided it was not used to advance a submission that the decision-maker was other than CST.

The additional material which the Defendant sought to rely upon was as follows: (i) a Ministerial Submission to the CST dated 10 November 2020 and a related email dated 3 December 2020 referring to the Minister’s decision; and (ii) documents that were attached to an email dated 21 January 2021 seeking the CST’s final clearance to publish the consultation response and related documentation. The email itself had been exhibited to Tack 3. I refused the Defendant’s application. My reasons are set out below. I firstly identify the reasons that applied to both sets of documents.

Disclosure of these documents was made at an extremely late stage. They were provided to the Claimant’s legal team just before the start of the hearing that morning. Mr Sharland was due to make his substantive submissions that day and would inevitably be prejudiced by this late documentation, that he would not have the chance to consider properly. Whilst it was desirable for the Court to have as full an evidential picture as possible, the prejudice that would arise could not be effectively addressed in the circumstances.

I accepted that the November 2020 Ministerial Submission did not necessarily fall within the scope of material that should have been disclosed in accordance with the Defendant’s duty of candour, as it did not concern future pension provision (the subject of the closure decision) but related to a recommendation that the CST accept the deferred choice underpin (Footnote: ⁴) (“DCU”), as the best method of addressing the retrospective period from April 2015 to 31 March 2021. The Defendant wanted to rely on this document and the related email as part of its response to the Claimant’s contention (advanced in the Supplementary Skeleton) that the December 2020 Ministerial Submission showed that the closure decision had already been made before the CST had been furnished with the consultation responses. Ms Callaghan accepted that the new documents did not address this point directly, but she suggested that they may explain what officials had in mind when referring to policy decisions that had already been taken in that Ministerial Submission.

However, even if the lack of earlier disclosure involved no breach of the duty of candour, the perceived need to produce these documents at this very late stage arose because of the Supplementary Skeleton served the previous day, which, in turn, arose solely because of the Defendant’s late provision of Tack 2 and Tack 3. If the Defendant had undertaken timely compliance with Swift J.’s directions, this position would not have arisen. The difficulty the Defendant was now in (in so far as it was a difficulty) was entirely of its own making.

Furthermore, the new documents were, at best, of marginal relevance given that there was no explicit indication in the December 2020 Ministerial Submission that adoption of the DCU was the earlier decision referred to.

As regards the attachments to the 21 January 2021 email, Ms Callaghan accepted that it “would have been preferable” for them to have been exhibited to Tack 3, along with the email itself. This was something of an understatement. On the face of it, the documents bore directly on the decision-making process concerning the consultation and the closure decision. In my view, the Defendant’s duty of candour required that they should have been disclosed in accordance with Swift J.’s timetable. Ms Callaghan explained that she wished to rely on the documents as supporting the proposition that a final decision had yet to be made in light of the reference to “final clearance” being sought from the CST. Given the Defendant’s history of non-compliance and of late disclosure (which I have already summarised) and the prejudice to the Claimant if the documents were admitted, the interests of justice indicated that the application should be refused.

In June 2010 the government established the Independent Public Service Pensions Commission, chaired by Lord Hutton, to undertake a fundamental structural review of public service pension provision. The Commission published its final report (“the Hutton report”) in March 2011. It recommended replacing existing final salary pension schemes with career average revalued earnings schemes (“CARE schemes”) and moving existing members of the current public service pension schemes into the new schemes in respect of future accruals, whilst protecting accrued pension rights up to the date of reform. The Hutton report did not recommend TP. The Government accepted the recommendations of the Hutton Report, but decided to implement TP.

The main public service pension schemes that the UK Government is responsible for in addition to those relating to police officers, include the Civil Service Pension Scheme for England, Wales, Scotland, and home civil servants in Northern Ireland; the Teachers’ Pension Schemes in England and Wales; the National Health Service Pension Schemes in England and Wales; the UK Armed Forces Pension Schemes; and the Firefighters Pension Schemes in England. The specifics of implementation was addressed by the various Government Departments responsible for the workforce in question, including the Home Office in relation to the police and firefighters.

The public service pension schemes other than the 1987 Scheme had a Normal Pension Age (“NPA”), that is the minimum age at which the scheme member can retire voluntarily and take their pension without reduction for early payment. In relation to these schemes, it was agreed that members who were within 10 years of their scheme’s NPA on 1 April 2012, would be exempt from the reforms and allowed to stay in their existing schemes. This meant that by 1 April 2022 all of these members would have been able to retire and take an unreduced, immediate pension (albeit some of them may have chosen to remain in employment to maximise the full pension benefits that they could realise). There was also a form of tapered protection for members between 10 and 13.5 / 14 years of NPA as of 31 March 2012. Those who were not eligible for TP or tapered protection were to be transferred to the new schemes.

The 1987 Scheme contained in the Police Pension Regulations 1987 (SI 1987/257) (“the 1987 Regulations”) provides benefits to members who joined the police before 1 April 2006. There is no NPA. Unreduced, immediate benefits are payable on members reaching their voluntary retirement age (“VRA”) where one exists (Footnote: ⁵); or on reaching both 25 years’ service and an age in excess of 50 years; or on reaching 30 years of service at any age. The 1987 Scheme is a final salary scheme. Pensions are calculated as 1/60^th of the final salary for every year of pensionable service up to 20 years, plus double accrual at 1/30^th of the final salary for every year of pensionable service thereafter, up to a maximum of 30 years’ service. The entitlement to retire on immediate, unreduced pension after 30 calendar years’ service applies to both part-time and full-time officers.

The 2006 Scheme contained in the Police Pension Regulations 2006 (SI 2006/3415) (“the 2006 Regulations”) provides benefits to members joining the service between 1 April 2006 and 1 April 2015. This scheme has an NPA of 55. It is also a final salary scheme. Pensions are calculated as 1/70^th of the final salary for every year of pensionable service up to a limit of 35/70ths. The accrual rate is therefore less generous than the 1987 Scheme; the maximum pension under the 1987 Scheme is 2/3rds of final average pensionable pay (typically attained after 30 years’ service), whereas the maximum pension under the 2006 Scheme is half of final average pensionable pay, typically attained after 35 years’ service.

It is worth emphasising at this stage that the point at which a member can retire with unreduced immediate pension is not necessarily the same as the date at which they have earned maximum pension entitlement, as explained in paras 12-14, Tack 1. I will return to this topic. Amongst other instances the latter situation could arise where the scheme member has worked part-time.

As the 1987 Scheme did not have an NPA, it was decided that TP would be designed to provide roughly equivalent protection by ensuring that a large cohort of officers who were within 10 years of retirement would be protected. In September 2012 the Home Office published the “Police Pension Scheme: Reform Design Framework” (“the RDF”) setting out the government’s final position on the main elements of the new police pension scheme to be introduced from April 2015. In relation to TP, it stated:

Para 7.4 of the Home Office Circular 014/2013, published on 19 November 2013 said of the police legacy schemes:

The Defendant accepts that the statements I have underlined amounted to clear representations that members who had the benefit of TP would be able to remain in their legacy schemes until they retired. I will address the statements made in respect of the other public service pension schemes when I consider Ground 3.

The TP afforded to members of the 1987 Scheme was not exactly equivalent to the TP afforded to those in the other public sector schemes. As the RDF indicated, those qualifying for TP included those who had 10 years or less to age 55. However, not all officers in this category would reach a point that enabled them to access unreduced, immediate pension within 10 years, specifically if the individual had joined the police at a relatively late age and so would not yet have built up the requisite years in service (and did not have the benefit of a VRA of 55).

The pension reforms were implemented by the Public Service Pensions Act 2013 (“PSPA 2013”), which came into force in April 2015. It made provision for the establishment of the reformed schemes and the closure of the legacy schemes. Section 18(1) stated that: “No benefits are to be provided under an existing scheme to or in respect of a person in relation to the person’s service after the closing date”. The closing date for all the main legacy schemes (other than the local government scheme for England and Wales) was 31 March 2015. Section 18(5)–(7) set out the power to enact TP as an exception to the closure of the legacy schemes for persons who were members of an existing scheme, providing that it could “in particular, be framed by reference to the satisfaction of a specified condition (for example, the attainment of normal pension age under the existing scheme or another specified age) before a specified date”.

The 2015 Regulations made pursuant to PSPA 2013 established the reformed police pension scheme, which came into force on 1 April 2015. The 2015 Scheme provides for payments of an unreduced, immediate pension from age 60. Benefits under this scheme are calculated by reference to the average salary that the member earned throughout their time in the scheme, increased by reference to a prescribed formula. Pensions are calculated as 1/55.3 of average salary for each year of pensionable service within the scheme. The majority of active members of the legacy schemes did not qualify for TP and were transferred to the 2015 Scheme. I will refer to members who did not qualify for TP as “unprotected members” and those who qualified for full TP (as opposed to the tapered form) as “protected members”.

The TP for police officers, contained in Schedule 4 to the 2015 Regulations, reflected the arrangements set out in the RDF, which I have described. The transitional provisions were not time limited; there was no date by which those who qualified for TP would be required to move to the reformed schemes.

The trigger for the consultation was the Court of Appeal’s judgment in Lord Chancellor v McCloud [2018] EWCA Civ 2844, [2019] ICR 1489 (“McCloud”), handed down in December 2018. A large number of public service employees (including firefighters, police officers and judges) who did not benefit from the TP arrangements brought Employment Tribunal proceedings alleging that those provisions amounted to unlawful direct age discrimination in treating older scheme members more favourably than younger members. Claims for equal pay and sex and race discrimination were also made. The government accepted that transitional arrangements were prima facie discriminatory on grounds of age but contended that they were a proportionate means of achieving the legitimate aim of protecting those closest to retirement from the financial effects of the pension reforms. The Court of Appeal rejected that contention. In short, this was because the perceived need to do so was unsupported by any evidence, in circumstances where younger members of the schemes stood to lose the most financially as a result of the 2015 reforms: see in particular, paras 89, 92 and 157-164 in the court’s judgment (Longmore LJ, Sir Colin Rimer, Sir Patrick Elias). It followed from this conclusion that there was no objective justification defence to any of the claims. The court’s judgment related only to the TP arrangements and did not affect the validity of the reformed schemes themselves introduced by the PSPA 2013 and the related regulations.

On 27 June 2019 the Supreme Court refused permission to appeal. The government accepted that the McCloud judgment had implications for all the schemes established under the PSPA 2013. It confirmed that it would address the difference in treatment across all of the schemes, for all members with relevant service and not just for those who had lodged discrimination claims.

In January 2020 the CST was the Rt Hon. Rishi Sunak MP. On 28 January he was provided with a submission prepared by officials in HMT’s Workforce Pay and Pensions team entitled “Public Service Pension Reform”. It is agreed that knowledge acquired by Mr Sunak cannot be attributed to his successor. However, the document provides some context in terms of subsequent events.

The front page summary indicated: “As a result of the McCloud judgment and to avoid future discrimination challenges we need to equalise future treatment in public service pension schemes. This provides an opportunity for further reform.” The submission recommended that the CST “equalise future treatment by moving all members into 2015 schemes from a future date”. The document said that officials intended to announce policy in the “McCloud consultation” in Spring 2020. It explained there was a need to address discrimination occurring from 2015 (when TP had come into effect), as well as equalising treatment from a future date. I will refer to the plan to address past and ongoing discrimination as “the retrospective remedy” and the period under consideration as “the remedy period”. I will refer to the plan for the future as “the prospective policy”.

This submission was focused on the prospective policy. Five options were identified for the Minister (para 4 and following). In brief summary, they were: to allow the retrospective remedy to continue; to place all members into the old legacy schemes; to place all members into the reformed schemes introduced in 2015; to place all members into the 2015 schemes and reduce its generosity; or to move all members into brand new schemes. The document recommended the third option, namely placing all members into the 2015 schemes. Broadly, two supporting reasons were identified, fiscal considerations and fairness.

The submission noted that the trade unions would be opposed to ending TP early for those working beyond their NPA, but that these members would generally have final salary rights in respect of their accrued legacy scheme pensions (para 14). It also observed that using primary rather than secondary legislation was highly desirable as members with TP “could argue they have a legitimate expectation that they would remain protected (i.e. in their final salary scheme) until retirement” (para 15). Officials indicated that if the Minister agreed with the recommendation, it was proposed to announce plans relating to both the retrospective remedy and the prospective policy in parallel “as part of a wider consultation on the McCloud remedy” (para 18).

The same paragraph observed that the new equalised treatment could be introduced from 2022 at the earliest. “At this point, the cohort of members still under the original transitional protection would be limited to those working beyond their Normal Pension Age”. Annex A included a worked example of a member working beyond their NPA. However, the statement I have just quoted was not accurate in terms of the remaining members of the 1987 Scheme; as I have explained, there was no NPA, and some officers would not have reached the point at which they could obtain an unreduced, immediate pension by 2022 (paras 37 & 40, above). In oral submissions Ms Callaghan accepted that the statement was incorrect in this respect. However, she submitted that this was not significant for reasons I will address when I consider the Grounds.

A short email sent between HMT officials on 10 February 2020 said that CST “agreed with your recommendation to move all members into the 2015 schemes”. Mr Sharland submitted this indicated that the closure decision had to all intents and purposes already been made at this stage before the consultation was even formulated or begun, given the absence of any explicit qualification. I reject that proposition. I accept that this text was simply a shorthand reference to the Minister’s agreement to the course proposed in the January Ministerial Submission, which entailed a preferred policy (of ending TP and moving all to the 2015 schemes) but a consultation on the plans. The context was a brief email between officials familiar with the proposals.

On 2 April 2020 there was a meeting of the Public Pensions Steering Group (“PPSG”) within HMT. By this time the CST was the Rt. Hon Stephen Barclay MP. The PowerPoint slides for the meeting included that: “Advice was put to the new CST asking whether he agrees with the decision taken by the former CST (ie to end the remedy period as soon as possible by placing all members in 2015 schemes)”. It was noted that CST had “come back with some questions which were due to be addressed in the teach-in”. This underscores that matters remained under consideration at this juncture.

A further meeting of the PPSG took place on 29 April 2020. The PowerPoint slides said that at a meeting with the CST on 23 April “he accepted our recommendations to: end the remedy period and equalise future treatment by placing all members in 2015 schemes…” (bolded emphasis in the original). I accept that this was also a shorthand reference between officials to the Minister’s agreement to the proposed course; the very next slide referred to the public consultation that would be “outlining two of our main policies for dealing with the McCloud judgment” namely the retrospective remedy and the prospective policy. A subsequent slide headed “Approach to Consultation – Presentational Risks” noted that “Members across all schemes could complain about the small section of originally TP’d membership who have their protection removed before they retire”. Accordingly, this was an issue that officials at least were aware of, and it was not confined to the police legacy schemes.

The CST was provided with a Ministerial Submission dated 17 April 2020 headed “Public Service Pension Reform”. It appears from the dates that he would have had an opportunity to consider this document before the 23 April 2020 meeting I referred to. The front page summary said: “On 10 February 2020, the previous CST agreed to equalise future public service pension treatment by moving all members into 2015 Hutton schemes from a future date (option C below)”. The stated recommendation was for the Minister to “confirm this position”. The section on “timing” said: “We intend to set out this position in the ‘McCloud consultation’ in late spring / summer 2020” (page 1). The five options identified in the January 2020 document were set out again, along with some information about costings. The previous CST’s decision was referred to (para 7), with the indication that officials still believed the option of placing all members in 2015 schemes was the correct way to address the removal of the discrimination identified in McCloud, noting that it would end the £4 billion per annum remedy costs, and provide members with clarity. Para 14 stated in bold: “We therefore recommend that you confirm the decision to move all members to the 2015 schemes and confirm this position in the consultation”.

These materials indicate that in April 2020 the CST agreed with his officials’ recommendation, and the previous CST’s view, that moving all members to the 2015 schemes was the preferable option and the one that would be put forward in the forthcoming consultation. It is clear that it was the strong preference in terms of policy options, but I do not consider that the documents indicate that the CST had made a final decision that it would be adopted at this stage.

The April 2020 Ministerial Submission did not refer to the position of members with TP working beyond their NPAs. Nor did it include any specific reference to the position of those in the 1987 Scheme.

On 16 June 2020 the CST wrote to the Prime Minister and the Leader of the House of Commons indicating that he was seeking clearance from the Domestic Affairs and the Union Committee (“DAU”) “to commence a public consultation on proposals to end the discrimination in public service pension schemes identified by the Court of Appeal in 2018”. He referred to the need for primary legislation to end the discrimination and he sought clearance for a commitment to implement legislation by April 2022. The response has not been made available, but the parties agree it should be inferred that clearance was forthcoming given the launch of the consultation shortly afterwards.

Under the sub-heading “Public consultation for discrimination remedy and future pension provision”, the CST referred to the ongoing costs of remedying the discrimination identified in McCloud, saying: “Resolving this is a priority for me; the costs increase in billions each year. I therefore intend to close this remedy period in April 2022…Given the costs involved it is important to target the earliest possible date” (para 3). The Minister then explained that in terms of the remedy period, the consultation would identify two options for removing the discrimination. Both options entailed giving qualifying members a choice over whether they wished to have accrued benefits from their legacy scheme or from their reformed scheme for these years. The first option involved an immediate choice being made and the second option was the DCU, meaning that the choice would be deferred until the benefits became payable (para 4). The Minister said he did not intend to express a preference between these options (para 5). He went on to say that he intended to launch the consultation in July 2020 (para 8).

The Claimant places particular emphasis upon text that appeared under a new sub-heading “Pensions policy post 2022” in para 10. It said: “The proposals above will ensure that discrimination is removed and will address the court’s judgment but keeping the legacy schemes is not a sustainable solution for the longer-term. I propose to confirm that all active pension scheme members who are not in reformed schemes will move or return to 2015 reformed schemes from April 2022” (emphasis added). The Minister went on to say that these schemes represented generous pension provision and were based on the Hutton report’s recommendations.

The sentence beginning “I propose to confirm…” is capable of being read either as the Minister indicating that he proposes to confirm that this will be the policy adopted, come what may with the consultation; or that he proposes to confirm that closure of the legacy schemes will be the preferred policy identified in the consultation. The former is a more natural reading that would explain the reference to his future actions, whereas the Minister had already confirmed to his officials the policy that should be set out in the consultation. However, this text must be viewed in the context of my assessment of what the Minister had decided (and had not decided) in April 2020 and in the context of the contents of the consultation document that followed. I return to this topic when I set out my conclusions on Ground 1.

On 23 June 2020 Home Office officials prepared a Ministerial Submission for the Home Secretary, consequent upon HMT’s write round on the proposed changes to public service pension schemes and the intention to launch the consultation. Officials recommended that she wrote to the CST “outlining reservations regarding both legitimate expectations” and the projected timeline. As regards the former, the document said that officials had identified a risk of challenge in relation to legitimate expectation created by the terms of the ‘Heads of Agreement’ letter (“HOA”) for the firefighters and also for the police pension schemes (para 5). HOA was then used as a shorthand in the remainder of the paragraph, but the authors were clearly referring to the position of police officers as well. The text continued: “Officials have flagged to HMT that under the terms of this agreement, members who were originally afforded full transitional protection would be allowed to remain in their legacy pension scheme until their retirement, even if that was beyond April 2022. The current proposals do not reflect the HOA”.

The Home Secretary clearly agreed with the recommendation as she wrote to the CST on 30 June 2020 (again, referring to the HOA as a shorthand for representations made to both firefighters and police). The letter said:

Accordingly, the representations to police that the Claimant relies upon as founding the legitimate expectation were raised in terms with the CST, in relation to firefighters as well as police.

On 7 July 2020 officials emailed the following for the CST’s attention: the intended final version of the consultation document (“this includes minor changes since the previous version we shared”); a proposed Written Ministerial Statement (“WMS”); and the Equality Impact Assessment due to be published with the consultation “(subject to minor drafting changes & proofreading)”. A reply sent on the Minister’s behalf on 9 July 2020 indicated: “CST’s happy with everything set out and agrees to all recs”; that he had “Noted all impacts and content to proceed” in relation to equalities; and that he was happy to sign the foreword to the consultation and to put his name to the WMS. Tack 3, para 12 and the 7 July 2020 email indicate that a Ministerial Submission was also provided. The Claimant submits that inferences should be drawn from its non-disclosure. I return to this when I consider Ground 1.

The consultation was presented to Parliament by the CST. The Minister signed the foreword, explaining that the document set out two proposed options for the retrospective remedy and that it “also sets out proposals for moving all active members into the reformed schemes after this period”. The CST said that the “final decisions will need to take full and careful account of the views of all stakeholders”. The Executive Summary noted that whilst those who saw their earnings increase considerably during their employment would likely benefit from a final salary scheme, others including lower paid members were likely to be better off in CARE schemes. Reference was made to the annual cost of £41.8 billion in paying out public service pension benefits. As to future pension provision this section said: “The government therefore believes that the reformed schemes initially introduced in 2015 provide an appropriate level of public service pension provision. All public servants in scope of this consultation will be placed in these pension schemes in respect of employment from 1 April 2022 onwards. This consultation seeks views on that proposal”.

Under a section headed “Purpose of this consultation” (after addressing the retrospective remedy options), the document explained: “This consultation also sets out proposals for future service beyond the remedy period. The government remains of the view that the schemes established in 2015 currently represent an appropriate level of pension provision. To ensure the schemes remain appropriate and affordable while treating members equally for future service, all active members will be placed in the reformed schemes in respect of employment from 1 April 2022”.

The consultation document set out the government’s clear preference as to the prospective policy but did so in the context of seeking views on this proposal. This is also the impression given by Chapter 3, which I will come on to.

The document noted that approximately 2 million individuals would be in scope of the pension provision changes from 1 April 2022 (para 1.22). The government’s intention to bring forward primary legislation as soon as practicable was explained (paras 1.29 – 1.30). Consultees were given 12 weeks to respond, with the consultation closing at midnight on 11 October (para 1.36).

Question 1 asked: “Do you have any views about the implications of the proposals set out in this consultation for people with protected characteristics as defined in section 149 of the Equality Act 2021? What evidence do you have on these matters? Is there anything that could be done to mitigate any impacts identified?” The preceding passages noted that some of the proposals may have differential impacts “but the government’s current view is that these will not have a disproportionate or otherwise unjustified impact on individuals” (para 1.33). Reference was then made to the Equality Impact Assessment published alongside the consultation (“EIA1”).

Chapter 3 addressed future pension provision. Under this section, Question 9 asked: “Does the proposal to close legacy schemes and move all active members who are not already in the reformed schemes into their respective reformed schemes from 1 April 2022 ensure equal treatment from that date onwards?” The government’s plans for closing the legacy schemes from 1 April 2022 were explained from para 3.7 onwards. Paras 3.9-3.10 said that 1 April 2022 was the earliest point at which the necessary primary legislation and administrative arrangements could be in place, that 1 April was the normal date on which pension changes were implemented and this date allowed sufficient time for the government to consult on the proposals and “subject to decisions taken following the consultation, introduce the necessary legislation”. It was also noted that: “Members of the legacy schemes will have more than 20 months’ notice of the government’s plans”. Paras 3.11 and 3.12 said as follows:

The passages I have underlined are relied upon by the Claimant as constituting a material error of fact. The statement was incorrect for some police officers in the 1987 Scheme because of the absence of an NPA and the particular TP criteria that were used, as I have explained earlier (paras 37 & 40 above). In oral submissions Ms Callaghan accepted this. I will address the legal significance when I consider Ground 4. I will return to some of the reasoning set out in this document when I consider the Defendant’s case on Ground 3. The contents of para 3.11 indicates a general awareness that some members who had chosen to remain in legacy schemes would want to continue in employment beyond 1 April 2022. The document did not contain any specific reference to the police legacy schemes.

The contents of Chapter 3 and Question 9 indicate a concern not to adopt a prospective policy that will fail to ensure equal treatment and/or give rise to further discrimination issues. I return to these matters when I address Ground 1.

EIA1 said it had been prepared to be read alongside the consultation (para 1.1). The document indicated that the government did not consider that the proposals set out in the consultation document would result in unjustifiable differential impacts on individuals with the protected characteristics in the Equality Act 2010 (para 2.1). It said the government “welcomes input from stakeholders – particularly whether there are any further potential impacts that have not been considered in this document” (para 1.2). Details of the data used were set out in Annex A. Para 1.11 explained that public service pension schemes held data on sex and age, but not on the other protected characteristics. Annex A explained that the information available from the schemes was based on their 2016 actuarial valuation data.

Future pension provision in relation to the protected characteristic of sex was addressed at paras 2.60-2.62. It was emphasised that all members were to be moved to the reformed schemes and that the proposals applied regardless of sex. Para 2.61 noted that data was not available on the “remedy cohort”, that is to say the cohort of members who would be moved from the legacy schemes to the reformed schemes on 1 April 2022. It continued:

I will consider the adequacy or otherwise of that approach when I address Ground 2. However, I observe that the number of scheme members who would elect to stay in the legacy schemes during the remedy period was inevitably unknown at this juncture. The final point made in this section of EIA1 was that a CARE scheme may offer relatively fairer outcomes to women who have tended to experience lower salary progression (para 2.62). There was no separate consideration of the position of police officers in the legacy schemes in terms of sex discrimination. In the next section of the document (which addressed the implications for the protected characteristic of age), para 2.65 said: “by 1 April 2022, all members who were offered transitional protection in 2012 will have reached their Normal Pension Age”. The Claimant submits that this was a further incorrect statement so far as the position of members in the 1987 Scheme was concerned.

HMT received 3,144 responses to the consultation, including 3,016 from individuals and 128 from organisations. Those who provided responses included the PSA and the Police Pensions Scheme Advisory Board (“PPSAB”). The PPSAB’s constituent bodies included the National Police Chiefs Council (“NPCC”), the Chief Police Officers Staff Association, the National Association of Retired Police Officers and the Police Federation of England and Wales. The constituent bodies, including the NPCC, also provided their own responses.

In her oral submissions Ms Callaghan accepted that the responses from police bodies raised the following points (amongst others): (i) their members had a legitimate expectation that they would remain in their legacy schemes until retirement (“the legitimate expectation issue”); (ii) those in the 1987 Scheme were in a different position to others, because of the service-based requirements for triggering the entitlement to a pension, rather than reaching an NPA (“the NPA issue”); and (iii) the prospective policy was likely to impact adversely on those who worked part-time or had taken career breaks as they were mostly likely to be women (“the indirect sex discrimination issue”). Ms Callaghan said she also acknowledged that these were important submissions that needed to be carefully considered and grappled with.

In light of the Defendant’s position in relation to those matters I can refer to the consultation responses from the PSA, PPSAB and NPCC quite briefly. However, I will give a flavour of the first two.

The PSA’s response proposed that, given the complexities of double accrual and the absence of an NPA in the 1987 Scheme, a specific Equality Impact Assessment should be conducted for the affected police officers as part of the consultation process (para 3). It said that no adequate explanation had been given as to why the remedy period cut-off date was 31 March 2022 (para 13). Para 14 observed that: “some police officers with full protection, mainly those aged 55 who have less than 30 years service with the police, including those who have taken career breaks, or worked part-time, or left the service and re-joined within five years, will remain in service at this date and be transferred to the reformed scheme. This is because the 1987 legacy police scheme requires an individual to work for 30 years in order to obtain full benefits”. The response continued that these members had believed until now that they would retain the full benefits of their legacy scheme until retirement and had made life choices and financial decisions on that basis (para 15). It was said that ending the remedy period on 31 March 2022 may indirectly discriminate against women and in any event “is contrary to the affected individuals’ legitimate expectation and is clearly unfair” (paras 15 and 16). Para 18 referred to the position of officers who had taken career breaks in the belief that they were fully protected and would remain in their legacy scheme. The essence of the response to Question 9 of the consultation was that “the proposed end of the remedy period should be adjusted so that the potentially discriminatory impacts are removed and the legitimate expectations of those identified are satisfied”.

The PPSAB response listed 12 equality issues that were said to arise in relation to the police schemes. The sixth point was: “For those transferred to legacy schemes in 2022 (Footnote: ⁹) without having achieved 30 years’ service. These will be mainly female workers whose part-time service has impacted on their ‘reckonable/pensionable’ service. The gender/age profile of this cohort may give rise to claims for discrimination and therefore consideration should be given to identifying an alternative outcome for these individuals”. The response to Question 9 amplified this concern. It will suffice to quote from para 9.2:

who are fully protected members aged 45 or over on 31 March 2012 who will not have reached 30 years of full pensionable service under the 1987 scheme by the end of the Remedy period on 1 April 2022. (There is no normal pension age in the 1987 scheme).

who were part-time members aged 45 or over on 31 March 2012 and part-time members aged 38 or over with at least 20 years of service who will not have reached full pensionable service by 1 April 2022.

who took career breaks while covered by the transitional protections who will not have reached full pensionable service in their legacy scheme by 1 April 2022.”

HMT officials analysed the responses received over August to November 2020. They also held meetings with various stakeholders, including a meeting with representatives of the PPSAB on 18 September 2020. Mr Murphy of the PSA attended. According to the minutes he said that they “would look to challenge, especially as they believe officers who are part time or have taken career breaks would be affected worse”. Mr Peri’s response was that “the government needs to ensure there are no special arrangements for any members to ensure no further discrimination” (paras 47 and 48). Reference was also made to the belief that “the government should honour the commitment they previously made” (para 49). Accordingly, the legitimate expectation issue and the indirect sex discrimination issue were raised at this meeting. However, as I address when I set out the relevant legal principles, it is common ground that the knowledge of officials cannot be attributed to the Ministerial decision-maker. It is also agreed that the CST was not required to read the individual consultation responses, as opposed to a summary or distillation of them.

On 10 December 2020 officials emailed a Ministerial Submission and “a number of documents for the CST to review and approve in advance of the publication of the consultation response”. The email said that the submission “summarises the key policy announcements and seeks approval for updated implementation timings for the retrospective elements of remedy”. It also attached a consultation response document, outlining the analysis of the responses received “as well as the government’s policy positions on prospective and retrospective remedy”; and a write round letter seeking clearance from the DAU for “our prospective and retrospective remedy policies”. The Defendant does not suggest that the CST received information summarising the consultation responses before this communication. This draft version of the consultation response has not been disclosed. Ms Callaghan invites the inference that it was in materially similar terms to the subsequent finalised version given the email indicated that the finalised version would have added sections on tax and an Annex A covering technical issues. The email also asked that CST: “Notes the updated summary of the equalities impacts, considering views raised in response to the consultation”. This was at Annex B to the Ministerial Submission.

An email in response sent on 16 December 2020 indicated that “CST agreed with all the recommendations and is content to issue the WR. He also noted the updated equalities impacts”.

The Ministerial Submission itself was headed “Final clearance to write round on McCloud policies. For decision”. The “Issue” for decision was described as “Consequential decisions on technical matters and clearance of WR letter”. The front page summary said that the CST was asked to write to DAU seeking agreement to implement a DCU, to move all affected public service employees into the reformed schemes from 1 April 2022 and to waive cost control mechanism (“CCM”) ceiling breaches for completion of the 2016 valuation process. The Minister’s agreement was also sought on updated implementation timings. The recommendations set out on page 2 were for the CST to: (1) “write round on the consultation responses, noting the foreword in your name and noting that drafting changes are likely during write-round as we iterate with departments”; (2) note the updated summary of the equalities impacts; and (3) agree that timescales for implementing the retrospective remedy regulations should be extended to 2023, but the remedy window should close in 2022.

Under the heading “Overview & key announcements” the document said: “The attached consultation response includes the analysis of responses we received and sets out the final policy decisions that you have taken” (para 1). I have underlined the words that the Claimant places particular emphasis on. The same paragraph went on to note that the consultation response “confirms that the legacy schemes will be closed on 31 March 2022”; and announces the implementation of a DCU for the remedy period. It was said that officials were continuing to work through some technical aspects, as detailed in Annex A. Para 2 then addressed the CCM ceiling breaches point. Paras 3 – 4 and Annex A covered technical issues. Paras 5 – 6 and Annex B addressed equalities impacts. Annex B referred to the key objective of removing the discrimination identified in McCloud and ensuring equal treatment. Various points raised in consultation responses concerning potential age discrimination were summarised. No reference was made to the indirect sex discrimination issue. The remaining parts of the submission dealt with timelines.

The Claimant submits that this Ministerial Submission shows that CST had already made the closure decision by this stage. I return to this contention when I address Ground 1. For now, I note the following from the documentation I have just summarised:

Neither the covering email, nor the description in the Ministerial Submission of the “Issue” for CST’s decision, nor the officials’ recommendations, indicate that an outstanding decision is required from the Minister as to the prospective policy to adopt in light of the consultation responses;

The content of the Submission focuses on the areas that are identified as outstanding, including the CCM ceiling breaches point, the technical issues and the timeline;

The reference in para 1 refers to “final policy decisions” that the CST “has taken” in the past tense.

By letter dated 16 December 2020 the CST wrote to the Chancellor of the Exchequer seeking agreement from the DEI Committee to publish the government’s response to the consultation. The CST said that the “case for closing the legacy pension arrangements was clear” (para 9). He referred to the need to ensure the schemes were affordable; that pension benefits were provided on a fairer basis, including removing subsidies to higher earners; and to recognise significant improvements in life expectancy. He continued that “some responses to the consultation have argued for continued membership of legacy schemes”, but that the rationale for their closure stood (para 10). Further reasons he then identified were that allowing the legacy schemes to continue would place an unfair burden on taxpayers, many of whom did not have access to generous pension arrangements themselves; and that the majority of public service employees were already members of the reformed schemes. The CST said he proposed “to confirm our position that the remedy window will close on 31 March 2022” (para 11).

A letter dated 13 January 2021 from the Economic and Domestic Affairs Cabinet Secretariat indicated that the DEI Committee had given clearance to proceed.

On 21 January 2021, HMT officials emailed a number of documents for the CST’s clearance. The attachments listed in the email included a covering submission containing recommendations and a second version of the consultation response which was said to account for additional comments from Whitehall and “minor changes to the ministerial forward which the CST has previously approved”. Ms Callaghan invites the court to infer that this version of the consultation response would not have been significantly different to the published version. Additionally, the email attached an Equality Impact Assessment. It is unclear whether or not the text of this was the same as the version published in February 2021, but the email suggested it was close to finalisation (it “will be proof-read over the next few days, so there may still contain minor spelling or formatting errors”). A WMS was attached which CST was asked to lay in Parliament on the same day as the publication of the consultation response. The email asked that the Minister noted the analysis of the equalities impacts and recommended that he: “Agrees to the publication of the final consultation response, noting that there have been minor changes to the document during the write-round process”.

An emailed response sent on 25 January 2021 indicated that CST agreed to the publication of the final consultation response, approved the WMS, and noted the equality impact assessment. Ms Callaghan submits that this was the date of the closure decision, which was subsequently published on 4 February 2021.

The consultation response was presented to Parliament by the CST. The WMS was made to Parliament by the CST on 4 February 2021 summarising the legislative plans set out in the response document. A second Equality Impact Assessment (“EIA2”) was published with the response.

The Foreword to the consultation response was signed by the CST. This section indicated that the DCU had been chosen as the better option in terms of the retrospective remedy and confirmed that the legacy schemes would close on 31 March 2022. The Minister said that responses had been listened to and “have been indispensable in refining our proposals”. The Executive Summary said that responses had been considered when making final policy decisions on future pension provision.

Chapter 3 dealt with future pension provision in more detail. Under the section summarising responses from individuals, the document said that 34% broadly agreed with the government’s proposals and 66% broadly disagreed (para 3.5). It said that a number of individuals had argued that equality would only be achieved by keeping the legacy schemes beyond 2022 for some members (para 3.6); and also that insufficient notice had been given (para 3.9). There were explicit references to the inaccurate statement in para 3.12 of the consultation (albeit not as specifically affecting police officers). The text said: “Several respondents argued that paragraph 3.12 within the consultation document is incorrect…The respondents state that they were protected, but remain weeks, months or years away from their NPA. Most of these respondents have requested to remain within their legacy schemes until their NPA.”

Under the section summarising responses from organisations, reference was made to an alleged legitimate expectation. It was said that a number of respondents had “claimed that members who were given full protection in 2015 have a legitimate expectation to stay in the legacy schemes indefinitely if they wish to work beyond their NPA”. In this regard, reference was made to the responses from the NHS Pension Scheme Advisory Board and from the PCS union (paras 3.14 – 3.15). There was no specific mention of the police legacy schemes or representations from the police bodies in this context. However, this passage shows that widespread concern was raised about the government resiling from earlier promises made to those in legacy schemes.

The government’s response was set out at paras 3.19-3.56. Para 3.29 referred to the importance of arrangements for future provision ensuring equal treatment and that if some members remained in different schemes, this would not be achieved. Issues raised in relation to age discrimination were addressed. At para 3.33 the document said: “By 1 April 2022 those who were offered full transitional protection by virtue of being within 10 years of their NPA in the legacy schemes will have reached that NPA. Where those members choose to remain in employment from 1 April 2022 they will do so with an entitlement to be members of reformed schemes, like all other members” (emphasis added). The Claimant relies on the passage I have underlined as a further inaccurate statement in respect of some members of the 1987 Scheme. The Defendant does not accept this, saying the wording was literally correct, as it only referred to those who did have NPAs. The Claimant counters that the phrasing was misleading, implying that this was the position for all. I return to this when I consider Ground 4.

The earlier representations that legacy members could remain in their schemes until their retirement were addressed at para 3.36:

I will return to these issues when I consider Ground 3. The next paragraph said that when TP was introduced in 2015 it was intended to be limited because the arrangements “were applied only to members who were within 10 years of their NPA under the legacy schemes, and the majority of those members are expected to have retired already or to do so in the coming years”. Both parties rely on this passage. The Claimant does so on the basis that it contains a further erroneous statement that TP only applied to members who were within 10 years of their NPA; and the Defendant does so because it is said to show an awareness that not all legacy scheme members would have retired by 1 April 2022.

It is common ground that there was no specific relevant reference to the position under the police legacy schemes save in para 3.46, which said:

The document addressed an age discrimination point and then re-emphasised that everyone would be treated equally under the reformed schemes (paras 3.47–3.49). Para 3.50 acknowledged that many respondents wanted to maintain their current arrangements up to when they retired: “but the government does not believe it would be fair to allow some members, and not others, to continue under different arrangements and as members of different schemes, after the discrimination has been addressed and the remedy period ends”.

The introduction to EIA2 said that the document updated EIA1 and that further analysis had been undertaken since the closure of the consultation and thought had been given to whether policies should be changed or altered in light of the identified equality impacts (paras 1.9-1.10). The explanation of the data relied upon at paras 1.11-1.13 was broadly equivalent to the comparable section in EIA1. The document indicated that individual pension schemes would consult on specific implementation details when they published their draft regulations and that the government would further consider the impacts and policy for each scheme at that stage (para 1.36). Although there was some emphasis placed upon this point in the Defendant’s written materials, Ms Callaghan accepted that as the closure decision had now been made, this future stage would not involve re-considering whether police legacy schemes would close from 31 March 2022.

Chapter 2 contained the overview. The Claimant relies upon the text of para 2.38 as repeating the erroneous indication that I have earlier highlighted. The Defendant says that this statement, like the passage in the consultation response, is literally true as it only referred to those who did have an NPA. The text said: “By 1 April 2022, all members who were eligible to be offered full transitional protection in 2012 will have reached their legacy NPA”.

Paras 2.57-2.67 addressed future pension provision in relation to the protected characteristic of age. Whilst no age discrimination issue is raised in these proceedings, some parts of this section are relied upon by the parties. Para 2.61 contained a statement analogous to that in para 2.38 (“by 1 April 2022, all members who were offered transitional protection in 2012 will have reached their Normal Pension Age”), as did para 2.62. In para 2.65 it was said that the government was aware “of a scheme specific issue raised in some responses, that occurs due to the service length-based specifics of some schemes (namely older police and firefighter schemes)”. The age discrimination implications of this were then discussed.

The next section concerned the protected characteristic of sex. It did not make specific reference to the representations regarding indirect discrimination made by the police bodies. It was noted in para 2.76 that: “A small number of individuals also felt that the proposals to move members in scope of remedy into the reformed schemes after the remedy period would negatively impact those who work part time, who they felt would be more likely to be female”. Para 2.78 contained a quote from one of the individual responses that said the changes would be discriminatory for those who had taken a career break or worked part-time, who would be predominantly women. By way of response, the document referred to CARE schemes offering fairer outcomes to women with lower salaries (para 2.78). The document accepted that women were more likely to take a career break and to work part-time than men (paras 2.84-2.85). Ms Callaghan accepts both those propositions in these proceedings. Future pension provision was addressed specifically at paras 2.86-2.90. It was acknowledged that: “Many responses also suggested that moving members in scope of remedy to the reformed schemes after the remedy period ends would create sex discrimination.” As in EIA1, the lack of data concerning the composition of those who may want to remain in legacy schemes after 1 April 2022 was then highlighted (paras 2.88-2.89). Para 2.90 reiterated that CARE schemes were likely to benefit lower earning women.

Mr Sharland submits that EIA2 failed to address the indirect discrimination issue raised by the police bodies with any or any sufficient specificity. He also contends that there was meaningful data that should have been obtained, as described in Murphy 2 and Murphy 3. Ms Callaghan says that the contents of EIA2 shows that there was awareness of the indirect discrimination issue; that it was unnecessary to deal with it specifically in terms of the police, as it had been raised by consultees in other public pension schemes too; that the response adequately addressed it; and that there was no additional meaningful data that could reasonably be obtained in relation to this. I return to these matters when I address Ground 2. I summarise the evidence relating to additional data at the end of this section of my judgment.

The government announced in the Queen’s Speech on 11 May 2021 that it would legislate to implement changes across all the main public service pension schemes. The Bill was introduced into the House of Lords on 19 July 2021. The second reading of the Bill took place on 7 September 2021. At the hearing I was told that the report stage would be on 29 November 2021. The government’s intention is that the Bill will receive the Royal Assent in time for implementation by 1 April 2022.

Clause 76 of the Bill has the effect of confirming that everyone who qualifies for the retrospective remedy can have accrued service in their respective legacy schemes up to 31 March 2022. It then removes the powers in s.18(5)-(8) PPSA 2013 to make exceptions to the closing date of the legacy schemes, with effect from 1 April 2022. Accordingly, TP arrangements previously made under those powers will no longer have effect from 1 April 2022.

Clause 77 creates two exceptions for relevant purposes relating to transfers of service from certain public and private sector schemes which are not within PSPA 2013 and weighted accrual. The Claimant submits this illustrates that exceptions can be made for those who remain in police legacy schemes on 31 March 2022. However, neither of these exceptions allow for continued accrual in the main legacy public service pension schemes beyond that date.

A third Equality Impact Assessment (“EIA3”) was published in July 2021 when the Bill was introduced in Parliament. On any view, this assessment post-dated the closure decision and the issue of these proceedings. However, the Defendant relies upon para 4.16, where the indirect discrimination issue identified by the police bodies in their consultation responses is referred to directly:

I will summarise the material relied upon by the parties. It is not suggested that it was obtained before the closure decision was made. I address the Claimant’s contention that further inquiries should have been made in relation to the indirect discrimination issue when I consider Ground 2 below.

As set out in Richards 1, Superintendent Richards joined the MPS in 1988, aged 19 and is a member of the 1987 Scheme. She attained 30 years of calendar service on 15 August 2018. However, because she had worked part-time she had earned less and made reduced pension contributions, when compared to an officer working full-time. Consequently, she is unable to retire on maximum pension benefits under this scheme (if it remains open) until February 2026. She said that in reliance on the representations to police, she had planned to delay her retirement until 2026. She said that if she is able to stay in the 1987 Scheme until that time she will receive an annual pension income of £42,724.50 and a commutation lump sum of £305,195.30; whereas if she is transferred to the 2015 Scheme from 1 April 2022 she will receive an annual pension of £40,839.15 and a total commutation lump sum of £280,569.40.

In Tack 1, para 94 Ms Tack said that the Government Actuary’s Department (“GAD”) had estimated that 63 members of the 1987 Scheme were affected by the closure decision in the sense that they would not have reached the point at which they could retire with an unreduced, immediate pension by 31 March 2022. I will refer to this as “the narrower cohort”. Ms Tack said this assessment was based on 2016 data, but that in any event the figure was likely to be between 60 – 70 members.

Mr Murphy took issue with that assessment in Murphy 2. He said that responses to freedom of information requests (“FOIA”) by 24 police forces (Footnote: ¹⁰) indicated that 3,141 officers in the legacy schemes would be adversely affected by the abandonment of the representations to police. He said that extrapolating this figure to include the membership of the remaining forces, indicated that between 8,000 – 10,000 officers were adversely affected (paras 13-14). Further, that HMT could have ascertained this information had reasonable inquiries been made. I will refer to the group of officers that Mr Murphy relies upon as “the wider cohort”.

Murphy 2 also included a worked example concerning a Superintendent with 26 years and 67 days of service under the 1987 Scheme on 31 March 2022. Mr Murphy said this showed that if moved to the 2015 Scheme, the officer would receive £3,046.60 less annual pension than under the 1987 Scheme; or £12,515.72 less cash and £1,181.82 less annual pension. Ms Tack disputed some of these specific figures in Tack 2, paras 27 – 29.

In Tack 2, Ms Tack explained that the GAD’s figure of 63 did not include the wider cohort of officers who can retire by 31 March 2022 but do not wish to do so because they have yet to achieve maximum pension benefits (para 15). She said that there was insufficient data to provide a reliable estimate of this wider cohort, but that in any event Mr Murphy had wrongly conflated the two groups (para 16). She exhibited an email relating to the GAD estimate, explaining that the figure of 63 was derived from those of the 1,667 protected members in the 1987 Scheme who will not have more than 25 years’ service by the closure date (and thus will not be eligible for an unreduced, immediate pension under the criteria I referred to in para 34 above). She referred to a report prepared by GAD addressing this in more detail, provided to the Claimant on 27 October 2021. Ms Tack also disputed Mr Murphy’s figures of 8,000 – 10,000 as the size of the wider cohort. In particular she said that figures from one police force could not be extrapolated to another; that the FOIA figures only dealt with a snapshot in time; and that the MPS (which accounts for 25% of all police in England and Wales) should be excluded, as the vast majority of MPS protected members would have retired by age 55 and thus before 1 April 2022 (given the MPS’ VRA: see para 34, above).

In Murphy 3, Mr Murphy continued to dispute: the Defendant’s reliance on the narrower cohort; HMT’s failure to obtain more extensive data from the police forces; and the basis upon which the figure of 63 was arrived at. He said that police constables and sergeants and MPS officers should not have been excluded from that figure. He also gave a more precise figure of 7,750 officers for the wider cohort, which he arrived at by adding the 3,141 officers referred to in Murphy 2 to a similar percentage of officers in the other forces.

I will address the parties’ respective positions as to the appropriate cohort of police legacy scheme officers at this stage, because, in turn, it may impact upon submissions made under several of the Claimant’s Grounds. The Claimant not only submits that the wider cohort of officers is the relevant group for the court to consider in terms of impact, but also relies on the proposition that these officers were in a unique and qualitatively different position, as compared to individuals in the other public service pension schemes, because of the absence of an NPA in the 1987 Scheme. In turn, the latter proposition informs the Claimant’s submissions that aspects of the decision-making process were flawed because there was insufficient recognition of this unique position of police officers, which, it is said, warranted separate and specific consideration and treatment. It is also relevant to the legitimate expectation issues.

However, I accept Ms Callaghan’s submissions (in turn based on Ms Tack’s witness evidence) on these matters. The unique feature of the 1987 Scheme for present purposes was the absence of an NPA. I have already explained why that meant that eligibility for TP for those in the 1987 Scheme was determined by its own specific criteria and why, in turn, this had the effect that unlike the schemes with an NPA, not all protected members will have reached the point at which they could retire with unreduced, immediate pension before 1 April 2022 (paras 37 and 40 above). Accordingly, the number of officers who were in this position, that is to say, the narrower cohort, does provide a point of distinction from the other schemes.

By contrast, the fact that some protected individuals would likely choose to remain in their legacy scheme if it remained open beyond 31 March 2022 in order to accrue maximum pension benefits is not restricted to or special to the 1987 Scheme. Although officers have reached the point at which the scheme permits them to retire with unreduced immediate pension, maximum pension benefits may not have been acquired for a number of reasons, including part-time working, or taking a career break. The Claimant is correct in saying that such persons in the wider cohort will also be adversely affected by the closure decision. Superintendent Richards is an example of an officer who comes within this wider cohort (but is not within the narrower cohort because she has attained 30 years’ service). However, this position is not unique to the 1987 Scheme. As the material summarised in the consultation response shows, some members of the other legacy pension schemes will be in a similar position of wanting to remain in their scheme post 31 March 2022 in order to obtain maximum pension benefits. (See also Tack 1, paras 110 and 133.) In some instances, these situations will also have arisen because those individuals worked part time and/or took career breaks. In other words, the indirect sex discrimination issue is not limited to or specific to the 1987 Scheme or to police officers. The related respect in which the Claimant suggests that these officers are in a unique position relates to the representations made to police. However, as I conclude when I come to address Ground 3, equivalent representations were made to members of the other public service pension schemes.

In his oral reply, Mr Sharland also contended that officers in the 1987 Scheme were more adversely affected by the closure decision than those in other public service legacy pension schemes, because of the particularly generous provisions of that Scheme, including the double accrual rate for service between 20 and 30 years. However, I do not consider this to be a point of uniqueness or qualitative distinction, rather than one of degree; protected members under the various public service pension schemes who have yet to attain maximum pension entitlements will be affected to varying extents by their particular scheme rules in terms of the amount by which the closure decision will reduce their previously expected pension. Furthermore, it is the absence of an NPA, rather than any other distinction, that was placed at the heart of the Claimant’s case, as set out in the Claimant’s pleadings and Skeletons.

In terms of the size of the narrower cohort, in so far as it is necessary to form a view on this dispute, the points made in Tack 2, in turn relying on the GAD’s report, appear to me to be well-founded. Officers with a VRA of 55 who were protected members of the 1987 Scheme will have reached the point at which they can retire with an unreduced, immediate pension before 1 April 2022 (paras 34 & 37 above) and so are correctly excluded from that cohort.

R (National Association of Health Stores) v Department of Health [2005] EWCA Civ 154 (“National Association of Health Stores”) concerned a challenge to measures adopted under statutory powers by the Secretary of State for Health restricting the use and sale of an herbal anxiolytic, on the basis that the Minister’s decision was made in ignorance of a relevant fact. Sedley LJ (Keene LJ and Bennett J agreeing) rejected the proposition, accepted by Crane J. below, that knowledge of departmental officials could be attributed to the Minister. Ms Callaghan does not take issue with this or with its applicability to the issues before me. Sedley LJ said a para 26:

A duty of consultation may arise under express statutory provision or in certain circumstances as part of the common law duty of procedural fairness. It is not suggested that consultation was a matter of statutory requirement in this instance. In any event, it is uncontroversial that once it is embarked upon, consultation must be undertaken in accordance with the recognised requirements: for example, see R v North and East Devon Health Authority ex p Coughlan [2001] QB 213 (“Coughlan”), para 108.

As is well-known, these requirements comprise four elements, originally identified in R v Brent London Borough Council ex p Gunning (1985) 84 LGR 168 (at p.189) and approved by the Supreme Court in R (Moseley) v Haringey London Borough Council [2014] UKSC 56, [2014] 1 WLR 3947, para 25 (“Moseley”), namely:

consultation must be undertaken at a time when proposals are still at a formative stage (“Gunning (1)”) (Footnote: ¹¹);

it must include sufficient reasons for particular proposals to allow those consulted to give intelligent consideration and an intelligent response (“Gunning (2)”);

adequate time must be given for these purposes (“Gunning (3)”); and

the product of the consultation must be conscientiously taken into account when the ultimate decision is taken (“Gunning (4)”).

As I indicated when listing the issues, the Claimant submits that Gunning (1), Gunning (2) and/or Gunning (4) were breached in this case.

In R (Electronic Collars Manufacturers Association) v The Secretary of State for Environment, Food and Rural Affairs [2019] EWHC 2813 (Admin) (“Electronic Collars”) Morris J reviewed the authorities and identified a series of general propositions at para 27. They included the following:

The principles relating to Gunning (1) were summarised by Morris J at para 139 as follows:

Gunning (2) requires that: “those with 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”: Electronic Collars, para 141. It also requires that the presentation of the information must be fair, but “the decision-maker may present for consultation his or her preferred option” (para 142).

As regards Gunning (4), Morris J observed at para 151 that this did not amount to an obligation to adopt the submission made by any particular respondent, nor to adopt the majority view: “The decision-maker is entitled to consider the whole range of responses and then to form his own view, independently of the views of any particular consultees. Further there is no obligation to consider each and every specific item of detail: West Berkshire CC §§62-63”. Nonetheless, “there should be evidence of consideration of important points made by consultees” (para 153).

All of these principles identified by Morris J are common ground between the parties. I record for completeness that there is one point of controversy (albeit, in light of my conclusions on Gunning (4) in this case, it is unnecessary for me to resolve it). It concerns how the adequacy of the information placed before the decision-maker is to be assessed. Morris J held that a Wednesbury approach is to be applied. He explained this in para 152 as follows:

The passage in Sir Ross Cranston’s judgment in Langton that Morris J referred to also cited the judgment of Elias J (as he then was) in R (Khatib) v Secretary of State for Justice [2015] EWHC 606 (Admin), paras 49 – 53. Mr Sharland submits that the passages in National Association of Health Stores and in Khatib were concerned with the decision-maker’s alleged failures to take relevant considerations into account and not with an alleged failure to comply with Gunning (4). He says that as the requirements of the consultation duty stem from the need for procedural fairness, the court must make its own evaluation of each aspect, rather than applying a Wednesbury test to the degree of information furnished to the decision-maker following the consultation. Ms Callaghan, on the other hand, submits that the approach identified in Langton and approved in Electronic Collars is correct; that it matters not that the ground of challenge was different in the earlier cases and was not related to procedural fairness, as the common issue which the courts were addressing in each instance was the sufficiency of a summary provided to the decision-maker (in circumstances where they could not be expected to consider all of the underlying material).

The relevant protected characteristics for the purposes of the PSED include sex: see s.149(7). Section 149, Equality Act 2010 provides (as relevant):

eliminate discrimination, harassment, victimisation and any other conduct that is prohibited by or under this Act

advance equality of opportunity between persons who share a relevant protected characteristic and persons who do not share it;

foster good relations between persons who share a relevant protected characteristic and persons who do not share it;

Having due regard to the need to advance equality of opportunity between persons who share a relevant protected characteristic and persons who do not share it involves having due regard, in particular, to the need to –

The importance of compliance with the PSED as an essential preliminary to decision-making by public bodies has been emphasised many times: see R (Adiatu) v HM Treasury [2020] EWHC 1554 (Admin), [2020] PTSR 2198 (“Adiatu”), para 203 and the cases cited therein.

In a very well-known passage, McCombe LJ summarised the principles that he drew from the authorities to date in R (Bracking) v Secretary of State for Work and Pensions [2013] EWCA Civ 1345 (“Bracking”), para 26. I will simply refer to the aspects that are directly material to the parties’ submissions in this case. In terms of the responsibility on the decision-maker, McCombe LJ said (para 26(3)):

In terms of timing, McCombe LJ said (para 26(4)):

In this regard, McCombe LJ referred to the points identified by Aikens LJ in R (Brown) v Secretary of State for Work and Pensions [2008] EWHC 3158 (Admin), including that “the duty must be fulfilled before and at a time when a particular policy is being considered”; that the duty must be “exercised in substance, with rigour, and with an open mind”; and that it was non-delegable and continuing.

McCombe LJ also cited from paras 77, 78 and 89-90 in the judgment of Elias LJ in R (Hurley & Moore) v Secretary of State for Business, Innovation and Skills [2012] EWHC 201 (Admin) (Divisional Court). For present purposes, the key points he made (at para 26(8)) were twofold. Firstly, that provided there had been a rigorous consideration of the duty “it is for the decision maker to decide how much weight should be given to the various factors informing the decision”; and “the court cannot interfere with the decision simply because it would have been given greater weight to the equality implications of the decision than did the decision maker”. Secondly, as to the extent to which the PSED entailed a duty of inquiry. Elias LJ said: “If the relevant material is not available, there will be a duty to acquire it and this will frequently mean than [sic] some further consultation with the appropriate groups is required”.

The obligation to investigate was summarised by Lewison LJ in R (Ward) v Hillingdon London Borough Council [2019] EWCA Civ 692, [2019] PTSR 1738 (“Ward”) at para 71 as: “Compliance with the PSED requires the decision-maker to be informed about what protected groups should be considered. That will involve a duty of inquiry, so that the decision-maker is properly informed before making a decision”. The extent of this obligation was described in the judgment of the court (Sir Terence Etherton MR, Dame Victoria Sharp P and Singh LJ) in R (Bridges) v Chief Constable of South Wales Police [2020] EWCA Civ 1058, [2020] 1 WLR 5037 as follows:

The principles emerging from the authorities were also summarised in the court’s judgment (Bean LJ and Cavanagh J) in Adiatu. They included that: the PSED is concerned with procedure, not with outcome (para 204); the statutory phrase ‘due regard’ means that which is “appropriate in all the circumstances” (para 205); and the PSED “does not require a detailed analysis of the sort that might be undertaken by leading counsel in the course of submissions in legal proceedings” (para 207).

As explained by Laws LJ in R (Bhatt Murphy) v The Independent Assessor [2008] EWCA Civ 755 (“Bhatt Murphy”), para 32: “a substantive legitimate expectation arises where the court allows a claim to enforce the continued enjoyment of the content – the substance – of an existing practice or policy, in the face of the decision-maker’s ambition to change or abolish it”.

The expectation must be based on a representation that is clear, unambiguous, and devoid of any relevant qualification: R v Inland Revenue Commissioners ex p MFK Underwriting Agents Ltd [1990] 1 WLR 1545, p.1570B. In this case, as I have already indicated, the Defendant accepts that representations of this nature were made. However, there is a dispute as to whether the representations are enforceable, in light of the size of the class to which they were made; and, in terms of whether the expectation is legitimate in light of the McCloud decision. I will therefore summarise the caselaw that bears on these issues.

R v Secretary of State for Education and Employment ex p Begbie [2000] 1 WLR 1115 (“Begbie”) concerned a statement the government had made whilst in opposition that it would abolish the state-funded assisted places scheme, but children who already had places under the scheme would continue to receive funding. When it came to power, the government enacted legislation providing that those holding places would only be funded until they had completed their primary education. The main basis on which the substantive legitimate expectation challenge failed was that it would require the Secretary of State to act inconsistently with the legislative intention; but Laws LJ (with whom Sedley LJ agreed) made some more general observations, including the following at 1130F-1131D:

In his judgment in Bhatt Murphy Laws LJ observed that the representation “must constitute a specific undertaking, directed at a particular individual or group, by which the relevant policy’s continuation is assured” (para 43). After referring to Ex p Khan [1985] 1 AER 40 and to Coughlan, he said at para 46:

In Rainbow Insurance Company Limited v The Financial Services Commission [2015] UKPC 15 (“Rainbow Insurance”) Lord Hodge observed that the courts would enforce an expectation only if it is legitimate, as “what is at stake here is the principle of legality” (para 52). He referred to a line of authorities concerned with tax legislation that established that nobody could have a legitimate expectation that they will be entitled to an ultra vires relaxation of a statutory requirement. He also referred to the basis on which the challenge in Begbie, had failed namely that “there could be no legitimate expectation that the Secretary of State would act contrary to statute” per Peter Gibson LJ at p.1125D-G, Laws LJ at p.1129E and Sedley LJ at p.1132B.

I can deal quite briefly with the circumstances in which the public authority can resile from a substantive legitimate expectation that would otherwise be enforceable, as the parties are agreed that the test is whether the authority has shown that it has good reasons, judged by the court to be proportionate, to resile from it: for example, United Policy Holders Group v AG of Trinidad [2016] 1 WLR 3383 at paras 120-121. In the latter paragraph, Lord Carnwarth JSC observed that in judging proportionality “the court will take into account any conflict with wider policy issues, particular those of a ‘macro-economic’ or ‘macro-political’ kind”.

The parties are agreed that the applicable criteria were identified in E v Secretary of State for the Home Department [2004] EWCA Civ 49, [2004] QB 1044, para 66 as follows (“the E criteria”):

there must have been a mistake as to existing fact (including a mistake as to the availability of evidence on a particular matter);

the fact or evidence must have been ‘established’ in the sense that it was uncontentious and objectively verifiable;

the appellant (or his advisers) must not have been responsible for the mistake;

the mistake must have played a material (not necessarily decisive) part in the decision-maker’s reasoning.

As relevant, s.31(2A) SCA 1981 provides:

must refuse to grant relief on an application for judicial review,

…

Application of the statutory test involves an evaluation of the counterfactual world in which the identified unlawful conduct by the public authority is assumed not to have occurred: for example, R (PCSU) v Minister for the Cabinet Office [2017] EWHC 1787 (Admin), [2018] ICR 269, para 89. It is well established that the onus is on the defendant and that the threshold is a high one: for example, R (Plan B Earth) v Secretary of State for Transport [2020] EWCA Civ 214, [2020] PTSR 1446, para 273. The same paragraph of the court’s judgment (Lindblom, Singh and Haddon-Cave LJJ) sounded the following note of caution:

As regards consultation, Mr Sharland relied on the judgment of the court, Leggatt LJ (as he then was) and Carr J (as she then was), in the Law Society case at para 141 where it was said that:

In relation to Ground 2, the Defendant highlighted cases where the court has refused relief despite finding a breach of the PSED, on the basis that it is highly likely that the same conclusion would have been reached if there had been compliance with the duty, as evidenced by an appropriate equalities assessments undertaken after the material decision was made: for example, R (Utilita Energy Limited) v Secretary of State for Business, Energy and Industrial Strategy [2019] EWHC 2612 (Admin), para 77; and R (Durand Education Trust) v Secretary of State for Education [2020] EWCA Civ 1651, para 72-76. However, Mr Sharland submitted that the latter case, where there was specific evidence from the decision-maker on this point, underscores the absence of comparable evidence in this instance.

The classic description of Parliamentary privilege was given by Stanley Burnton J (as he then was) in Office of Government Commerce v Information Commissioner [2008] EWHC 774 (Admin), para 46. Following a review of the caselaw, he said:

It is the second of these principles that is relevant to the present case. R (Wheeler) v Office of the Prime Minister [2008] EWHC 1409 (Admin) (“Wheeler”) concerned an alleged legitimate expectation arising from the Prime Minister’s promise to hold a referendum on whether the United Kingdom should ratify the European Union’s treaty establishing a Constitution for Europe. In the event, the European Council signed the Treaty of Lisbon instead. The British government indicated it would not hold a referendum and introduced the European Union (Amendment) Bill into Parliament to bring this treaty into effect in domestic law. The relief claimed was a declaration that the refusal to hold a referendum was unlawful. The court noted counsel’s submission that he sought “no more than to require the executive to introduce into Parliament a Bill…providing for a referendum…” (para 23) (emphasis added). By the time of the hearing, Parliament had passed the European Union (Amendment) Act 2008 and in so doing had rejected amendments that would have provided for a referendum. The court: rejected the proposition that an enforceable promise had been made (paras 40-41); concluded that the relief sought was futile, given the passing of the Act (para 52); and held that interference by the court with the proceedings of Parliament was a further “decisive reason” why the claim must fail (para 51).

Undoubtedly the relief sought in Wheeler went further than that claimed in the present proceedings. Ms Callaghan accepted as much. Nonetheless, the statements of principle are instructive. Richards LJ cited Sir John Donaldson MR’s judgment in R v Her Majesty’s Treasury ex p Smedley [1985] QB 657, 666C-E that it “behoves the courts to be ever sensitive to the paramount need to refrain from trespassing upon the province of Parliament or, so far as this can be avoided, even appearing to do so”; and “it would clearly be a breach of the constitutional conventions for this court, or any court, to express a view, let alone take any action, concerning the decision to lay this draft Order in Council before Parliament or concerning the wisdom or otherwise of Parliament approving the draft”.

In para 49, Richards LJ observed that the introduction of a Bill into Parliament forms part of the proceedings within Parliament. Later in the same paragraph he considered what would be the “practical effect” of the declaration sought:

In R (Unison) v Secretary of State for Health [2010] EWHC 2655 (Admin) (“Unison”) the claimant claimed that a legitimate expectation of consultation had been frustrated by the presentation of a White Paper to Parliament on the restructuring of the National Health Service (“NHS”). The court observed that the relief sought would require that consultation be undertaken and this, in turn, would impact upon the Parliamentary timetable (paras 8 and 11). After referring to para 49 of Richards LJ’s judgment in Wheeler and noting that the courts could not require a Bill to be laid before Parliament (para 10), Mitting J concluded: “The converse must also be true. The courts cannot forbid a Member of Parliament from introducing a Bill. To do so would be just as much an interference with Parliamentary proceedings…”

The claimant in R (Gill) v Cabinet Office [2019] EWHC 3407 (Admin) (“Gill”) was concerned about the absence of a Sikh ethnic tick box on a forthcoming census. A declaration was sought that “it would be unlawful for Her Majesty to make an Order in Council which follows the reasoning of the White Paper” in not including this option. The application was refused on the basis that it was premature (para 88); and because the declaration sought would not respect the separation of powers between legislature and judiciary (para 89). After reviewing the authorities, Lang J summarised the position at para 95: “It is well-established that a declaration which has the effect of requiring a minister to introduce, or prohibiting a minister from introducing draft legislation to Parliament, other than on the terms laid down by the court, is an impermissible interference with the proceedings of Parliament”. She concluded that granting the declaration would have the effect of preventing the Minister for the Cabinet Office from laying any draft Order and regulations before Parliament which did not include the option sought in the census questionnaire (para 94). She also referred to the statement of principle of Lloyd-Jones LJ (as he then was) in R (Yalland) v Secretary of State for Exiting the European Union [2017] EWHC 630 (Admin), para 19 that “[w]hether any legislation is to be introduced and the form it is to take is entirely a matter for Parliament itself, and not a matter for the courts” (para 101). I note that in Lang J’s reasoning the focus is again upon what would be the effect of the relief sought, albeit, in a context where the declaration sought would have pre-empted the introduction of particular legislation.

Adiatu concerned a challenge to the scope of the Coronavirus Job Retention Scheme. Under the scheme, assistance was provided to furloughed employees who were within the PAYE scheme and eligible for statutory sick pay. The latter was not payable unless an employee earned above a lower earnings limit; a position set out in primary legislation that could only be altered by amendment. The court held that the PSED did not apply to decisions that are given effect by primary legislation (paras 229 – 238). The judgment of the court included the following at para 230:

I accept that when the consultation was initiated, proposals as to the prospective policy were still at the formative stage in the sense used in the caselaw I have summarised. My earlier description of the material sequence of events shows that the CST and his officials had a strongly held preference for all the relevant public service pension legacy schemes to close in April 2022. Undoubtedly it would have taken a powerful argument to alter or modify this view, in circumstances where, on the face of it, there were strong reasons identified for the preferred option. But it has not been shown that the consultation on the prospective policy was initiated in bad faith or with a fixed and closed mind. As I explained earlier, the authorities indicate that this stage of affairs is sufficient to meet the requirements of Gunning (1).

In arriving at this conclusion, I have looked carefully at each stage of the chronology. When setting out the material circumstances I have explained why I conclude that no fixed decision had been made by the CST in either February 2020 or in April 2020. I noted the terms of the 16 June 2020 letter from the CST to the Prime Minister in paras 59-60 above. However, on balance, I am not persuaded that the CST had pre-determined the issue at this stage. Firstly, this letter was written when the consultation was imminent and in that context. The consultation is referenced in the text and the policy plans set out should be seen in that light. Secondly, the CST had not made a fixed determination in April 2020 (para 55 above), and I have not been made aware of anything occurring in the interim that had altered his state of mind. Thirdly, whilst I take Mr Sharland’s point that generalised reassurance in a consultation document along the lines that careful consideration will be given to responses, does not in itself carry significant weight, the contents of this consultation in respect of the prospective policy, including question 9, indicated a concern to understand the equalities implications of the proposal. Does the “I propose to confirm…” sentence in the 16 June 2020 letter indicate I should arrive at a different interpretation? In my judgment it does not outweigh the features I have just identified; as Ms Callaghan submitted, it is capable of being read as no more than an indication of the preferred policy that will be set out in the consultation (even though that is not the most natural reading, as I have observed earlier).

I do not consider it appropriate to draw an inference of pre-determination from the non-disclosure of the July 2020 Ministerial Submission. The terms of the 7 July 2020 email and the available attachments, indicate that the CST was being asked to approve the contents of the consultation, EIA1, and the WMS at this stage. None of this is suggestive in itself of pre-determination and to infer that the Submission gave a contrary indication would be no more than speculation.

Mr Sharland submitted I should follow the approach of Beatson LJ in Regina (VC) v Secretary of State for the Home Department [2018] EWCA Civ 57, [2018] 1 WLR 4781 (“VC”) (para 68), agreeing with the statement of Sales J (as he then was) in R (Das) v Secretary of State for the Home Department [2013] EWHC 682 (Admin) (“Das”) that: “Where a Secretary of State fails to put before the court witness statements to explain the decision-making process and the reasoning underlying a decision they take a substantial risk. In general litigation, where a party elects not to call available witnesses to give evidence on a relevant matter, the court may draw inferences of fact against that party. The basis for drawing adverse inferences of fact in a judicial review proceedings will be particularly strong…”. Reference was then made to the duty of candour. In my judgment, the position here is distinct. In both VC and Das the Home Secretary had chosen not to put in any evidence to explain the decision to detain. In this instance the Defendant has provided witness evidence and documentation, albeit in a “highly unsatisfactory” way, as I indicated in my earlier Order. Where gaps remain, the Defendant does indeed take the risk that inferences will be drawn. This is underscored by my reasoning when I address Gunning (4). However, it does not mean that I should mechanistically draw adverse inferences from the absence of a particular document when, as here, the available material points to a different conclusion.

Mr Sharland’s submission in respect of Gunning (2) was closely allied to his contention that the consultation outcome was pre-determined. He submitted that this undisclosed pre-determination impermissibly confined the actual consultation to narrow issues of implementation, with the result that consultees were unable to meaningfully respond. Accordingly, my conclusion that the requirements of Gunning (1) were met is dispositive of the Claimant’s case on Gunning (2) in these circumstances.

As I have indicated, Ms Callaghan acknowledged that the responses received from the Claimant and from the other policing bodies required careful consideration. The Defendant’s case is that this occurred between the closure of the consultation and 25 January 2021, when CST’s decision was made. She submitted that the documentary material showed that HMT officials gave detailed consideration to the responses, amongst other things, reading and summarising their contents and holding meetings with stakeholders. However, she accepted that the crucial matters for present purposes were what the CST was provided with, whether it accurately reflected the consultation responses and when he was provided with it relative to the decision-making timeline. She took issue with the significance Mr Sharland attached to the December 2020 Ministerial Submission.

If the CST made the decision to close the public service pension legacy schemes from 1 April 2022 before he received a summary of the consultation responses, it is evident that the Gunning (4) requirement of affording them conscientious consideration would not be met. The Defendant does not suggest that the CST received a summary of the responses any earlier than the 10 December 2020 email attaching (amongst other documents) the draft consultation response. The content of the accompanying Ministerial Submission is therefore particularly significant. The content is such that Ms Callaghan acknowledged: “I appreciate this does not read as if a decision is yet to be taken on the prospective policy”. She was right to do so. I have already highlighted the particular features of the Submission and the email to which it was attached that, taken together, strongly give the impression that the CST had already made his decision (para 86 above). Furthermore, I am unable to identify anything in this material that credibly points to a different interpretation. If, as Ms Callaghan suggested, the CST was being asked at this stage to agree the proposed prospective policy, I would expect this to be reflected in the “Issue” he was asked to determine and in the contents that followed in this briefing document. Her suggestion is also inconsistent with the reference in the past tense to “final policy decisions” (in the plural) “that you have taken”. Given the Defendant’s fragmented and incomplete disclosure, I cannot infer from the absence of decision-making documentation, that no relevant decisions on the prospective policy were taken by CST in the period before 10 December 2020.

Ultimately I am driven to conclude that the December 2020 Ministerial Submission provides a clear indication that the CST had made his final decision on the prospective policy before he was shown the summary of the consultation responses. There is nothing in the subsequent documentation or known events that negates this. I accept that little weight can be attached to generalised statements in the consultation response that the consultee’s views were taken into account. It seems likely that they were taken into account by the officials who drafted the document. However, for reasons I have already explained, it is the Minister’s state of knowledge that is crucial. Accordingly, there was a clear breach of the requirement to give conscientious consideration to the consultation responses. I accept that the ultimate test is one of clear unfairness (para 125 above). This amounted to clear unfairness.

Ms Callaghan also submitted that a final decision could not have been made prior to 10 December 2020 as there remained the need for collective Cabinet approval. Whilst this is correct as a matter of factual chronology, I agree with Mr Sharland’s objection that having rested the Defendant’s case on the proposition that the CST was the relevant decision-maker for the purposes of this litigation and having objected (successfully) to the Claimant’s late attempt to contend that the decision-maker was the DEI Committee, rather than the CST, the Defendant was not in a position to rely on outstanding collective approval as a basis for disputing that a decision had been made on the prospective policy before 10 December 2020. As I suggested to Ms Callaghan, there was an element of the Defendant “trying to have its cake and eat it” in this submission. In any event, as the material decision-maker is the CST, the question for me is when he made his decision on the prospective policy. For the reasons I have identified, this was before he received the summary of the consultation responses.

As I have found that consultees’ responses were not considered by the CST when the key decision was taken, I do not need to address the more nuanced submissions made as to whether the consultation response document fairly summarised the points made by the police consultees (save to the extent that I address this in respect of Grounds 2 and 3).

For the avoidance of doubt, the Defendant did not dispute the applicability of the PSED to the functions under consideration in this litigation. Mr Sharland submitted that the PSED was breached in three respects:

whilst the Defendant asserts that the CST was provided with drafts of both EIA1 and EIA2, the draft versions of these documents had not been disclosed. Accordingly, the Defendant had not shown what was before the Minister prior to the consultation being launched or prior to the closure decision being taken and in these circumstances the court should draw adverse inferences as to what the CST actually saw;

in any event, as the material decision to close the legacy schemes was made prior to the CST’s receipt of the draft of EIA2, the duty was not complied with at a time when the policy was still under consideration;

alternatively, even if the CST did consider EIA1 and EIA2 at an appropriate stage, their contents were inadequate in that the specific issues raised by the Claimant and other policing bodies were not referred to or addressed. Additionally, there was a failure to carry out reasonable enquiries into the numbers of officers who would be affected by closure of the police legacy schemes and thus the equalities implications were not properly examined.

As set out in my summary of the material sequence of events, an email was sent on 7 July 2020 attaching a version of EIA1 that was “subject to minor drafting changes and proofreading” (para 64 above). The response sent on 9 July 2020 (before publication of the consultation) indicated that the CST had considered this. None of the available documentation suggests that in the event more extensive changes were made before the EIA1 was finalised. This is sufficient for me to conclude that the CST saw and approved the relevant contents of EIA1 prior to its publication with the consultation.

As regards EIA2, I have already noted that the email sent on 21 January 2021 indicated that the draft sent to the CST at this stage was close to finalisation, save for proof-reading corrections (para 89, above). In these circumstances I consider it is sufficiently clear that the CST was provided with a version of EIA2 that reflected the published version in the relevant respects.

Accordingly, I do not consider it appropriate to draw the inferences that Mr Sharland invited me to from the absence of the drafts of EIA1 and EIA2.

However, Mr Sharland’s second submission is well founded. I have already concluded that there was a breach of Gunning (4) in that the CST made his decision on the prospective policy before he was provided with a summary of the consultation responses on 10 December 2020. It follows that he also made the decision before he saw a draft of EIA2 or a summary of the indirect sex discrimination issue or of any other equalities concerns raised by consultees. Ms Callaghan does not suggest that the CST was provided with any of that material prior to 10 December 2020. As my earlier summary of the relevant principles indicates, the PSED is placed on the Minister personally and what matters is what he or she took into account and knew, not what his officials read or summarised or discussed. As I have indicated when setting out the legal principles, the duty must be fulfilled at a time when a particular policy is under consideration, rather than after it has been adopted. Accordingly, there was a breach of the PSED in this regard.

I will address Mr Sharland’s third submission more briefly than I otherwise would have done, as I have already found that there was a breach of the PSED. I will focus on the post-consultation EIA2. I have summarised the parties’ competing contentions at para 104 above. I address this on the basis of the counterfactual situation that the CST did consider the draft of EIA2 before making his decision to close the legacy schemes. In this alternative scenario, I do not consider that there was a breach of the PSED in respect of the implications for those in the police legacy schemes.

I have already concluded that the indirect sex discrimination issue raised in response to the consultation was a cross-scheme issue, rather than one confined to the police legacy schemes (para 118, above). In these circumstances, considering the points on a global basis, rather than with a specific focus on the police legacy schemes, did not, in my judgment, indicate a failure to have “due regard” to the matters identified in s149. Accordingly, absence of detailed references to the position of officers in the police legacy schemes in EIA2 or in other material placed before the Minister does not bear the significance that Mr Sharland suggested.

EIA2 did indicate that consultees had raised indirect sex discrimination (and age discrimination) in relation to the prospective policy, including the impact on those who had worked part-time or taken career breaks. As my earlier summary of this documents indicates, this was referred to in paras 2.65, 2.76, 2.78 and 2.86-2.90. Given the number of public service pension schemes under consideration and the number of responses received from consultees it is unsurprising that examples were selected, rather than references made to each of the responses that raised these points.

Furthermore, EIA2 recognised that women were more likely than men to have taken a career break or worked part-time (paras 2.84-2.85), but identified, in summary form, the reasons why, notwithstanding this, it was considered appropriate to close the legacy schemes. The document was to be read with the consultation response. Taken together, the rationale provided included that: future arrangements had to ensure equal treatment circumstances where the continuation of legacy schemes for protected members had been found to constitute age discrimination; fairness required that all remaining members in the legacy public service pension schemes were moved to the reformed schemes at the same time (and the majority of members were already in the reformed schemes); maintaining the legacy schemes was very costly; and CARE schemes would offer a better outcome for women with lower salaries. The PSED is concerned with procedure, not with outcome, and I do not detect an absence of “due regard” in this analysis. As I have noted earlier, the kind of detailed analysis that expert counsel would undertake in legal proceedings is not required (para 138 above).

I note Mr Sharland’s point that EIA2 underplayed the strength and extent of the responses on this topic in the “small number of individuals” reference in para 2.76. However, this passage should not be viewed in isolation; para 2.86 referred to “Many responses” having raised sex discrimination in relation to the prospective policy. Moreover, this reference does not assist the Claimant given, as I have described, the indirect sex discrimination issue was analysed and responded to, rather than treated dismissively.

The contention that there was a failure to make reasonable inquiries is based on the proposition that the number of police officers adversely affected by closure of the legacy schemes could and should have been ascertained (paras 112 and 115 above), so that the data used was inadequate. I do not accept this submission. Firstly, because the Defendant was entitled to approach the indirect sex discrimination issue on a cross-scheme basis, for the reasons I have explained. Secondly, as Ms Tack says at Tack 3, paras 28-29, at this stage of cross-scheme consultation and assessment, it would not have been practical or proportionate to have commissioned data with the equivalent level of granularity in relation to each of the affected pension schemes and each protected characteristic, yet it would not have been fair to do so in relation to only the police legacy schemes. Thirdly, there would have been inevitable uncertainties around the numbers of members in the legacy schemes at the relevant time, as noted in EIA1 and EIA2 (paras 74-75 and 103 above). Fourthly, because there is force in the points made by Ms Tack as to the difficulties with Mr Murphy’s extrapolated figures regarding the size of the wider cohort for police officers (para 114 above).

Ms Callaghan submitted that the acknowledged representations to police did not give rise to an enforceable legitimate expectation for two reasons. Firstly, similar representations were made to members of the other public service pension schemes to the effect that protected members could remain in their legacy schemes until they retired, irrespective of the date. In terms of those who could benefit from the TP arrangements, this amounted to over 500,000 people across the public service pension schemes (as explained at Tack 1, paras 58-59) and accordingly the class was too broad for the statements to have that effect. Secondly, McCloud meant that the representations made were incapable of forming a legitimate expectation thereafter. I will deal with these points in turn and then with the alternative submission that, in any event, it was proportionate in all the circumstances for the Defendant to resile from the representations.

The promises that were made in respect of the other public service pension schemes are detailed in Tack 1, paras 51-56. On 20 December 2011, the then Secretary of State for Health made a statement in Parliament on the NHS pension scheme. It included the following:

On the same date, the Minister for the Cabinet Office and Paymaster General made a statement to Parliament on civil service pensions in which he said:

In May 2012, the Department for Communities and Local Government published the “Firefighters’ Pension Scheme: Proposed Final Agreement” which said of the TP arrangements it set out: “This protection will be achieved by the members remaining in their current scheme until they retire, which could be beyond 31 March 2022”. Similar statements were also made in March 2012 in respect of teachers’ pensions and those for the armed forces.

In each of the schemes, as Ms Tack indicates, TP was made available to members who were 10 years or less from being able to retire with an unreduced, immediate pension, not to those who were within 10 years of their maximum pension (which may be a later date for the reasons I have explained). Accordingly, each of the legacy schemes had members who might well want to remain in employment and in the legacy scheme, accruing additional pension rights after that 10 year period. And they were all told that they would be able to remain in these schemes until their actual retirement. I therefore conclude that the representations made were materially the same in relation to each scheme, meaning, on Ms Tack’s figures, that the representations were made to at least half a million people.

Mr Sharland advanced three reasons for treating the representations to police as distinct and distinguishable. Firstly, he relies on the fact that only the 1987 Scheme lacked an NPA. However, in seeking to identify the size of the class of officers who are adversely affected by the Defendant resiling from the representation, he relies on the wider cohort of about 7,500 officers who will not have reached full pension entitlements before 1 April 2022, rather than the narrower cohort of about 63 officers who will not have reached an age where they can retire with unreduced, immediate benefits before 1 April 2022. It is easy to understand why he does so, but this underscores that the change affected legacy scheme members much more widely than those without an NPA. This is also reinforced by the consultation response, which indicates that widespread concern around this issue was raised (para 94 above). Secondly, he says that the representations to police were made at a later date. The RDF was published in September 2012 and so whilst this appears to be factually correct (albeit, in most cases only by a matter of months), I cannot see that it affords any material distinction. Thirdly, he points out that police officers are prevented from taking strike action. Here as well, I do not see how the nature and extent to which scheme members can express disagreement with the government’s pension proposals provides any material point of distinction when ascertaining the size of the class to whom the representations were made.

As Laws LJ recognised in both Begbie and in Bhatt Murphy in the passages I cited earlier (paras 141 - 142 above), whilst in theory there may be no limit to the number of beneficiaries of a promise, in reality the larger the class, the less likely it is that the statement/s made will generate a legally enforceable representation. Here the size of the class is very large, and the subject matter concerns the macro-economic and political field. Mr Sharland relied on the decision of Cox J in R (HSMP Forum (UK) Limited) v Secretary of State for the Home Department [2009] EWHC 711 (Admin), that it would be unlawful for the Home Secretary to resile from a substantive legitimate expectation that the terms on which people had joined the Highly Skilled Migrant Programme, would be the terms on which they qualified for settlement. The Judge observed that the issue affected a specific well-defined group of people and did not lie within the macro-political field. She noted that in Begbie (at 1131D) the group of between 1,200 – 1,500 affected children was said to constitute a “relatively small, certainly identifiable, number of persons”; and commented that the number of skilled migrants affected in the present case was “considerably smaller and is clearly identifiable” (para 71). As such, I do not consider this authority assists the Claimant. Cox J applied the approach identified in Begbie, as I have done. In addition to the Judge’s characterisation of the subject-matter, the size of the class was much smaller. As Ms Callaghan pointed out, even if the cohort is limited to officers in the legacy schemes who were potentially affected by the representations to police (rather than members of the other public sector legacy scheme as well), the correct number is around 30,000, rather than the Claimant’s 7,750 officers, because the position for these purposes, should be considered at the time when the representations were made.

For these reasons I conclude that the representations to police did not give rise to an enforceable legitimate expectation. However, if I am wrong about this, both the numbers involved, and the subject matter are highly relevant to the question of whether the Defendant can lawfully resile from the representations.

Mr Sharland pointed out that the representations to police (and those made to members of the other legacy schemes) were not found to be unlawful in McCloud. However, this does not assist his argument; the legality of the representations was not before the Employment Tribunals or the Court of Appeal, so this is both unsurprising and insignificant.

The court’s reasoning in McCloud indicates that the whole basis upon which TP was provided to those who were generally the older members of the public service pension schemes, was seriously flawed and based on a completely unevidenced premise. On the face of it, continuing to differentiate for any of the groups of workers involved, by allowing some to remain in the, usually, more generous legacy schemes where the majority had to be in the reformed schemes was also likely to be viewed as unjustifiable age discrimination. When I put this to Mr Sharland, rather than suggesting that non-discriminatory arrangements could be devised for those who wanted to remain in the police legacy schemes, he said that an Equality Act challenge could be avoided by enacting the measures he sought in primary legislation. As I pointed out, that would not necessarily avoid a challenge under the Human Rights Act 1998. More importantly, I consider that the court should be very slow to recognise as enforceable a representation that post McCloud appears to involve unjustifiable discrimination. I agree with Ms Callaghan’s submission that the government should not be held to acting in a way that, on the face of it, would be knowingly discriminatory. Thus, there can be no legitimate expectation to that effect. This approach is supported by Lord Hughes’ analysis in Rainbow Insurance (para 143, above).

Accordingly, I conclude that there is no legitimate expectation that can be relied upon. However, if I am wrong on this point, then the discriminatory implications are in any event highly relevant to the question of whether the Defendant can lawfully depart from the expectation.

This issue only arises if I am wrong in my conclusion that the representations to police have not given rise to an enforceable legitimate expectation. As I set out earlier, the agreed test is whether the Defendant has good reason to resile from the expectation, judged by the court to be proportionate. In my judgment, the reasons that have been identified in the Defendant’s evidence and documentation do establish that departing from the expectation is proportionate in all the circumstances. In particular:

The issue arises in the macro-economic and political field and affect large groups of people. The class of those to whom relevant representations were made was over 500,000 members of the legacy schemes. Indeed, the impact is wider still, given the very large costs involved and the consequential impact on taxpayers. The consultation referred to the extension of TP post McCloud for the remedy period costing the government £17 billion for all public service schemes; and each additional year of the remedy costing around £2.5 billion (para 2.58). Accordingly, there are a number of significant public interest considerations in play and matters of political policy and priorities, which the government was in the best position to evaluate;

The reformed schemes themselves are lawful and unaffected by McCloud (which concerned the TP arrangements). As I have summarised earlier, the reformed schemes stemmed from legislation consequent upon the acceptance of the recommendations of the Hutton Report, which, in turn, followed a very detailed examination of the way forward for public service pensions;

Preserving membership of the legacy schemes for some and not others is likely to entail unjustifiable discrimination, as I have already highlighted. Even if legal action could be avoided by use of primary legislation, as the consultation response emphasises, considerations of fairness and non-discrimination were key objectives. The consultation response and EIA2 stressed that all members of the schemes would be treated equally from 1 April 2022 (for example, see the consultation response at paras 3.19-3.29 and 3.47-3.50);

Accordingly, retaining future legacy scheme membership for some only was not a viable option. Avoiding discriminatory and/or unfair treatment if some form of legacy membership was to be retained, would likely lead to extending the remedy period for all or providing equivalent benefits. For reasons I have already identified, I do not consider that there is a material distinction between the police legacy schemes and the other legacy schemes for these purposes. On the face of it, any such changes would be expected to apply to those schemes too. In turn, the costs to the public purse of taking the course that maintained the objective of equal treatment would be very substantial;

As was said, for example, in the CST’s letter to the Prime Minister of 16 December 2020, many taxpayers have access to less generous pension arrangements themselves; and the majority of public service employees are already in the reformed schemes;

In so far as members of the 1987 Scheme are in a distinct position through having service-based criteria, rather than an NPA, this only applies to the narrower cohort, comprising approximately 63 officers (paras 117-118). Taking steps to address consequential detriment to this small group of officers would not be proportionate given the wider countervailing considerations and implications that I have identified. Furthermore, as I have indicated, the Defendant considers that any remedial action would likely need to be on a broader basis given the fairness and non-discrimination considerations I have already highlighted. Whilst I do not in any way wish to minimise the difficulties caused to those affected, the scale of the financial loss identified in Mr Murphy’s worked example and in Richards 1 is limited in comparison with the wider considerations identified by HMT.

As for the notice that members of the legacy schemes have been given, the period is 13 months, rather than 20 months, if it is treated as running from the time when the closure decision was actually announced, as opposed to when there was the possibility of closure. Accordingly, whilst members of the police legacy schemes have been given some notice and there is some force in this point, I do not consider that it is as strong as the Defendant suggests. Nonetheless, taken with the cumulative weight of the points I have already identified, the Defendant has clearly established that resiling from the representations is a proportionate course.

Mr Sharland submitted that failure to properly consider the expectation at the time of the decision-making means that the reasons identified for resiling from it should carry little weight. In addition to this submission coming very close to, if not trespassing upon, the ground of challenge that was abandoned, it again rests on the proposition that the wider cohort of affected police officers are in a special position that required specific attention in the consultation response and EIA2. I have addressed why that is not the case. In any event, the contemporaneous documentation indicates an awareness of the legitimate expectation issue and an identified response to it: see my summary of the Home Secretary’s letter to the CST dated 30 June 2020; and my summary of paras 3.14-3.15, 3.36 and 3.46 of the consultation response.

As I identified when setting out the material circumstances: the January 2020 Ministerial Submission; paras 3.11-3.12 of the consultation; para 2.65 of EIA1; para 3.37 of the consultation response; and paras 2.38, 2.61 and 2.62 of EIA2, erroneously stated or indicated that by 1 April 2022 all members who were offered TP would have reached their NPA (the point at which they could retire with unreduced, immediate pension benefits). This was not correct in respect of all members of the 1987 Scheme because of the absence of an NPA (para 40, above). I accept that the reference at para 3.33 of the consultation response was literally true, as it only referred to those who did have an NPA. In any event, the errors I have identified show that the first of the E criteria is satisfied.

There is no dispute that the second and third elements of the E criteria were met, the question is therefore whether the errors played a material part. Mr Sharland contended that the Defendant was not entitled to dispute this issue as the DGR confined submissions to whether there was an error of fact. However, this point is not well founded; DGR, paras 65 and 66 also disputed that any error was material in the sense that it did not influence the decision made.

As with the other issues before me, it was accepted that it was not in point to show that HMT officials understood the detailed position if the CST was misled by such errors. In addition to the errors themselves, Mr Sharland says that there was no explicit correction in the documentary materials I have summarised.

As I noted earlier, the error was pointed out in the submission from the PPSAB. Further, the consultation response itself said that respondents had identified the alleged error in para 3.12 of consultation (para 93 above). In itself, this does not assist the Defendant, given the absence of evidence that the CST saw these materials before making his decision on the prospective policy to be adopted.

More importantly, as the issues were being considered at a global cross-scheme level (which I have found to be a legitimate approach), I do not consider that this error was in any sense material. Had the CST been made aware that about 63 members of the 1987 Scheme would not have reached the point at which they could retire with unreduced immediate pension before 1 April 2022, it is highly improbable that this would have affected his decision to close all the legacy schemes from that date, given the powerful considerations in favour of the approach adopted that I have described (and which were, in summary form, reflected in CST’s letters of 16 June 2020 to the Prime Minister and 16 December 2020 to the Chancellor of the Exchequer.)

The Claimant also suggested that the error was material in that the 1 April 2022 was chosen as the closure date because it was wrongly believed that all those who received TP would have reached their NPA before then. However, the evidence indicates that this date was chosen as it was the earliest by which the necessary changes could be implemented, and the very expensive remedy period brought to an end. This is reflected in the contents of the CST’s letter to the Prime Minister dated 16 June 2020 and the passages in the consultation document I have cited earlier; and it is supported by Tack 1, paras 128-132 and Tack 2, para 41.

For the reasons I have identified, I accept aspects of Grounds 1 and 2 and reject Grounds 3 and 4. The requirements of Gunning (1) and Gunning (2) were satisfied, but the consultation was so unfair as to be unlawful because the material decision was made in advance of the CST considering the consultation responses. The PSED was breached for related reasons, namely that the decision was made before the implications detailed in EIA2 were considered. However, I do not accept the Claimant’s challenge to the contents of EIA2 or that there was a breach of the duty to make reasonable inquiries. The representations said to give rise to the substantive legitimate expectation were made to a large class of public service workers in legacy pension schemes and were not limited to the police legacy schemes. In light of this and/or because of the discrimination in the TP arrangements established in McCloud, I do not accept that there is an enforceable legitimate expectation. Alternatively, the Defendant has shown that it is proportionate to depart from the expectation that the Claimant relies upon. There were factual errors in the consultation and other documentation in terms of indicating that all protected members in the legacy schemes had an NPA which they would have reached before 1 April 2022, but this error was not material as it did not impact on the closure decision that was made.

The Claimant submitted that the Defendant has not met the high threshold required by the statutory test, or alternatively that this is an instance of “exceptional public interest” so that the court should disregard s.31(2A). I will consider the position in turn in relation to each of the grounds that I have upheld.

As I indicated when setting out the legal principles, application of the statutory test involves an evaluation of the counterfactual scenario in which the identified unlawful conduct is assumed not to have occurred. Accordingly, in relation to Ground 1 this involves assessing what the position would have been if the CST had conscientiously considered a summary of the responses to the consultation before making his decision. The onus is on the Defendant and the threshold is a high one (para 147 above).

Although Mr Sharland placed particular reliance on para 141 of the judgment of the court in the Law Society case (para 148, above), the circumstances are materially different. The Lord Chancellor’s decision to reduce the amount payable to criminal solicitors under the Litigators’ Graduated Fee Scheme was found to be unlawful because an analysis conducted by the Legal Aid Agency (“LAA”) was relied upon by the decision-maker, but not referred to in the consultation document, so that consultees did not have the opportunity to consider and respond to this material. The s.31(2A) submission rested on a scenario in which the consultation had included this analysis, the Law Society had responded with an expert’s critique of its methodology (as per a report prepared for the proceedings) and the Lord Chancellor, in turn, had obtained an expert report which addressed those points (as per a report from a Mr McHale obtained for the proceedings) (para 138). The court did not admit these parts of the experts’ reports, but considered what the position would have been under s.31(2A) had it done so. The court held there was “no reason to assume” that on this scenario the equivalent of the McHale analysis would have been undertaken; and, if it had been, that the content would have equated to that produced in the context of adversarial litigation. Further, it was impossible to know what responses other consultees would have made to the LAA material (para 140). Hence the reference in para 141 of the judgment to the hypothetical decision being made on “materially different information and advice”.

By contrast, in the present case, I have found no basis for impugning the consultation document or the opportunity that consultees had to respond to it. In turn, the contents of those responses are known, as is the way that they were then summarised in the final consultation response. Given the way the draft consultation reports provided to the CST were described in the accompanying emails, I accept it can be safely inferred that the relevant sections were not materially different in the versions that he was supplied with on 10 December 2020 and 21 January 2021 (paras 82 and 89 above). Accordingly, in this case, the court is in a position to know what was before the Minister at the time when conscientious consideration should have been given.

In all the circumstances I conclude that the Defendant has shown it is highly likely that the CST would have made the same decision in this counterfactual scenario. This is because:

The closure decision was the strongly preferred policy position throughout the period January 2020 – February 2021, as I have described when setting out the material circumstances. This was the consistent direction of travel at all relevant times;

The closure decision was the strongly preferred policy option detailed in the consultation;

As I have explained, the content of the responses from consultees is known;

The documentation indicates that HMT officials who reviewed the consultation responses, met with stakeholders, prepared the draft consultation response and advised the Minister, did not consider that any basis had been shown to depart from the preferred prospective policy. The draft consultation response is not an ‘after the event’ document prepared in the context of adversarial litigation; it provides a contemporaneous record of the assessment of consultees’ submissions that was placed before the Minister;

The CST was already strongly in favour of the preferred prospective policy, as shown, for example, by his 16 June 2020 letter to the Prime Minister (para 59, above). He remained so thereafter, as shown, for example, by his 16 December 2020 letter to the Chancellor of Exchequer (para 87, above);

The CST did approve the consultation response containing the closure decision and laid it before Parliament with the WMS in February 2021;

Without trespassing into the forbidden realms of the policy-maker, I have had to assess the proportionality of departing from the representations to police when considering Ground 3. In finding that it was proportionate to do so, I have identified the reasons that underpin that conclusion. They also go to show why the CST would have made the same decision if Gunning (4) had been complied with.

I turn then to consider the position in relation to Ground 2, again on the counterfactual scenario that the CST did give conscientious consideration to EIA2 or a materially similar draft before making his decision. My reasoning and my conclusions are analogous to those I have just set out in relation to the unlawful consultation. The contents of EIA2 and the responses from consultees that informed this are known and remain unchanged in the counterfactual scenario. When addressing Ground 2, I concluded that no breach of the PSED has been shown in terms of the contents of EIA2 and in particular in terms of how the indirect sex discrimination issue was addressed. I also concluded that there was no breach of the duty to make reasonable inquiries. Accordingly, the contents of EIA2 provide a clear indication of what would have been considered by the Minister if events had proceeded without a breach of duty. In terms of outcome, EIA2 supported the preferred prospective policy as the most proportionate means of addressing the equalities issues that arose. For the reasons I have already identified when addressing consultation, it is highly likely that the CST would have made the same decision had consideration been given to this material before it was made.

For the avoidance of doubt, my conclusion in this regard is not dependent upon EIA3, which was produced after this litigation was underway. It tends to reinforce the assessment I have already reached (para 108 above), but my conclusion is not based upon it.

The issues raised by this litigation are undoubtedly important ones for those who are affected, but in my judgment they do not come close to the relatively limited circumstances that would constitute matters of “exceptional public interest”. Accordingly, the application of s.31(2A) SCA 1981 is mandatory, and I must refuse to grant relief.

Strictly speaking it is unnecessary for me to consider the impact of Parliamentary privilege in light of the conclusion regarding relief I have just arrived at, however I will do so in summary form, given the importance of the issue and in case I am wrong in relation to s.31(2A) SCA 1981. This issue could only arise if the mandatory terms of that statutory provision are not met and the grant of relief remains a live issue. The relief sought in the Claim Form, is for: an order quashing the consultation and the consultation response; a declaration that the consultation was unlawful; and a declaration that the closure decision was unlawful.

Although this area was only addressed briefly in the DGR and in the Defendant’s Skeleton Argument, it was accorded considerable prominence by Ms Callaghan in her oral submissions. Counsel sensibly agreed that Mr Sharland would defer addressing this issue until his Reply, after he had heard how Ms Callaghan developed her submissions on this topic. As she addressed it at the outset of her submissions on Day 2, Mr Sharland was able to prepare a written note in response and make oral submissions on Day 3.

As Ms Callaghan submitted, the authorities I reviewed earlier concerning Parliamentary privilege show that the court will not interfere in Parliamentary proceedings by making orders which either directly or indirectly require a Member of Parliament to introduce a Bill or which would prevent a Member of Parliament from doing so. Further, that doing so ‘indirectly’ includes making a declaration that has this effect. I also agree with her submission that this line of cases shows that the courts must not do anything directly or indirectly that would delay the passage of a Bill through Parliament, or which would directly or indirectly tell Parliament what the form or content of a Bill should be.

The next step of Ms Callaghan’s submission is more contentious. She submitted that it followed from the propositions I have just summarised that it would be equally constitutionally impermissible for the court to do anything that would give rise an expectation that: (a) the government would withdraw or abandon a Bill; (b) progress to a Bill would be delayed, for example to allow a consultation process to be re-opened; or (c) Parliament would take certain matters into account when considering a Bill.

In response, Mr Sharland submitted that the whole objection was entirely formalistic, given that that the issues were justiciable, and the court was entitled, indeed obliged, to give a reasoned judgment in relation to the Claimant’s grounds. Further, that the relief sought in this case, in contrast to the authorities cited by Ms Callaghan, did not require Parliament to do or refrain from doing anything; the relief sought would simply mean that it was better informed. Parliament could then choose whether to proceed with the current timetable in respect of the Bill or, for example, to abandon the Bill to enable further consultation to be undertaken. Furthermore, an expectation that Parliament might act in a certain way did not interfere with its proceedings as the expectation could be dashed.

As I have indicated, this issue arises in a context where the Defendant accepts that in light of the grant of permission, the Claimant’s grounds are justiciable, that the court should address them and I have duly addressed them, upholding them in some instances. In these circumstances, there are some attractions to Mr Sharland’s “formalistic” objection. However, as Parliamentary privilege has been raised in relation to relief, I consider the court should address it on its merits, rather than simply dismissing the objection for this reason. In judicial review proceedings, there are numerous well-established bases upon which a reasoned judgment in a claimant’s favour does not preclude the refusal of relief.

If the court was to quash the consultation then this would inevitably result in disruption to the Parliamentary timetable given the current passage of the Bill. My earlier citations from Unison, from Gill and from Adiatu show that this would amount to an impermissible interference with proceedings in Parliament. Accordingly, granting the quashing order that is sought would infringe Parliamentary privilege.

The position is more borderline in respect of the declarations. I prefer to consider the matter as one of effect, rather than expectation. I bear in mind the focus in this line of authorities on the “practical effect” of granting a declaration and the concern that the court should not indirectly provide a form of relief that would plainly trespass upon the privilege if expressed in a peremptory form (para 153 above). A declaration that the closure decision was unlawful, would be tantamount to saying that the decision embodied in Clause 76 of the Bill is unlawful. Yet, the form of primary legislation is a matter for Parliament and not for the courts (para 155 above). Furthermore, Parliamentary privilege would be breached if a court were to declare that the procedure leading to legislation being enacted was unlawful (para 156 above). However, that would be the effect of granting a declaration that the consultation preceding the Bill was unlawful. It would also entail disruption to the Parliamentary timetable in the sense that the Defendant would be under a public law duty to then do something about the consultation. Declaratory relief is sought in these proceedings precisely because the Claimant wants to influence the course of the Bill.

For these reasons I conclude that Parliamentary privilege would be infringed if I were to grant the relief sought, in the counterfactual situation that s.31(2A) did not apply.

In the circumstances although I have upheld aspects of Ground 1 and Ground 2, I do not grant relief. It follows that the claim for judicial review is dismissed.

After considering the judgment in draft, Mr Sharland submitted that the court’s order should include recitals referring to my conclusions that the requirements of Gunning (4) had not been met, that there had been a breach of the PSED and that s.31(2A) SCA 1981 applied. I do not consider that this is appropriate. If the order were to summarise my findings, in the interests of accuracy and balance it should include my conclusions on each of the issues before the court. However, this would make the order unduly lengthy and unwieldly. In any event, this is unnecessary as the judgment speaks for itself.