Royal Courts of Justice
Strand, London, WC2A 2LL
Before :
MR JUSTICE JULIAN KNOWLES
Between :
THE QUEEN ON THE APPLICATION OF CATHERINE HARVEY | Claimant
|
- and - |
|
(1) LONDON BOROUGH OF HARINGEY (2) SECRETARY OF STATE FOR HOUSING, COMMUNITIES AND LOCAL GOVERNMENT | Defendants |
Nigel Giffin QC and Patrick Halliday (instructed by Bindmans LLP ) for the Claimant
James Cornwell (instructed by Borough Solicitor ) for the First Defendant
Julian Milford and Christopher Knight (instructed by GLD ) for the Second Defendant
Hearing dates: 26 and 27 June 2018
Judgment Approved
The Honourable Mr Justice Julian Knowles:
Introduction
This is an application for judicial review with the permission of Her Honour Judge Cooke sitting as a Deputy High Court Judge in respect of the refusal of the First Defendant (‘the Council’) of 11 July 2017 to pay Ms Catherine Harvey, the Claimant, a survivor’s pension following the death of her co-habiting partner, Mr Stephen Roe. Mr Roe was an employee of the Council until October 2003, when he was made redundant. At that point he became a pensioner member of the Local Government Pension Scheme (‘LGPS’/‘the Scheme’), into which he had paid during his career in local government. He continued to receive his pension until his untimely death in 2016.
The Claimant maintains that the Council’s refusal to pay her a pension is discriminatory and in violation of Article 14 of the European Convention on Human Rights (‘ECHR’/‘the Convention’), read with Article 1 of Protocol 1 to the Convention (‘A1P1’), and thus that the Council’s refusal is unlawful as a matter of domestic law pursuant to s 6 of the Human Rights Act 1998 (‘HRA 1998’).
At the hearing the parties were agreed that the remedial issues which would arise if discrimination is established might be relatively complicated if they had to be considered on the basis of different potential hypotheses. They were agreedthat it would be easier and more efficient if any argument about them is addressed, if I rule in the Claimant’s favour, once I have determined the basis upon which any such discrimination exists. Accordingly, the parties agreed that the hearing be limited to the question of whether there is unlawful discrimination in this case. I agree that is a sensible way to proceed.
The facts in outline
The Claimant is 66 and retired. Her evidence is that she lived with Mr Roe as his unmarried cohabiting partner from 1987, when they bought a house together, until his death in June 2016, although towards the end of his life he lived in a nursing home. She says that they lived together as husband and wife and were financially interdependent throughout their relationship. She explains that there was no legal impediment to them marrying, and that their decision not to do so was a matter of personal choice, in particular because of her feminist views. I accept her evidence in full.
Mr Roe worked for the Council from about 1979 until 2003, when he was made redundant. Thereafter he received his pension under the LGPS, effectively being treated as if he had retired. He received his full pension, even though he was being treated as retiring early. He also received a lump sum entitlement. Following his death, the Council asked the Claimant to complete a form confirming she was his next of kin. She returned the form in April 2017, having agreed with Mr Roe’s family that she was his next of kin for these purposes, he having died intestate.
In May 2017 the Council wrote to the Claimant to inform her that there had been an overpayment of pension, and that no further pension payments were due to Mr Roe’s estate. On 30 June 2017 she wrote to the Council asking it to confirm whether she was eligible for any survivors’ benefits. The Council responded as follows on 11 July 2017:
“Unfortunately for a co-habiting partner to be entitled to receive a survivor’s pension, the member must have paid into the LGPS on or after 1 April 2008. The late Mr Roe left the LGPS on 27 October 2003; therefore there is no spouse pension entitlement ( sic ).”
As I shall explain later in this judgment, upon the ordinary meaning of the applicable rules of the LGPS, this statement was correct, and is conceded by the Claimant to have been so: under the rules, she was not entitled to a survivor’s pension following Mr Roe’s death. The rules provide the survivor of a person in Mr Roe’s position with a pension if they are a married spouse (whether same-sex or different sex) or a civil partner, but not if they are an unmarried cohabiting partner.
The Claimant’s ultimate complaint is against these rules, which are contained in the various statutory instruments which govern the LGPS. The Secretary of State, the Second Defendant, is responsible in law for this legislation, made under the Superannuation Act 1972 (‘SA 1972’). From time to time, the Secretary of State has introduced a new set of rules for the LGPS, each set of rules constituting a new pension scheme with differing benefits. Such new schemes were introduced in 1997 (‘the 1997 Scheme’), in 2008 (‘the 2008 Scheme’), and in 2014 (‘the 2014 Scheme’). In this judgment when I refer to ‘the Scheme’ I am referring to these three schemes collectively unless the context otherwise indicates. Under the 1997 Scheme, a survivor’s pension was not payable to an unmarried cohabiting partner; under the 2008 and 2014 Schemes, it was and is payable. Mr Roe’s position was governed by the 1997 Scheme, because that was the Scheme in place when he was made redundant in 2003.
As I have said, the Claimant alleges that the refusal to pay her a survivor’s pension constitutes unjustifiable discrimination contrary to Article 14 read with A1P1 of the Convention, as given effect in domestic law by the HRA 1998. She relies upon her status as an unmarried person cohabiting with a partner whose active membership of the LGPS ceased before 1 April 2008 (a ‘Pre-2008 Cohabitee’). On that basis, she asserts that she has suffered unlawful discrimination in comparison with:
Spouses of persons whose active membership of the LGPS ceased before 1 April 2008 (‘Pre-2008 Spouses’);
Unmarried persons who cohabit with partners who remained active members of the LGPS on or after 1 April 2008 (‘Post-2008 Cohabitees’).
It is common ground that both of these suggested comparators are entitled to receive a survivor’s pension.
The Claimant also alleges indirect age discrimination against what she terms ‘older surviving cohabitees’.
The Council and the Secretary of State resist the claim. Their principal arguments are that:
The Claimant is not in a relevantly similar position to either of the classes of person to whom she compares herself, and because that is a pre-requisite of an Article 14 claim, the Claimant’s case fails at the outset.
The Claimant’s age based complaint is without substance, in that the LGPS does not define benefits by reference to age. It is not indirect age discrimination for a pension scheme to be revised prospectively with effect from a certain date.
In any event, any difference in treatment has an objective and reasonable justification.
The LGPS and the statutory framework
The LGPS
The LGPS is one of the largest defined benefit schemes in the world, with over five million members. These may be ‘active members’, ie, members who are in pensionable service under the Scheme; ‘pensioner members’, ie, persons who are entitled to the present payment of pension or other benefits; or ‘deferred members’, ie, persons other than an active or pensioner members who have accrued rights under the Scheme, ie, persons who have left LGPS employment after a period of pensionable service but before their pension benefits are payable (see s 124(1), Pensions Act 1995).
The Scheme comprises around 90 separate funds, each of which is run by an administering authority. These are mainly, although not exclusively, local authorities. They are responsible for investing assets and applying the Scheme’s rules so that individuals receive their correct pension entitlement under the Scheme. The Council was the administering authority for the fund from which Mr Roe’s pension was paid.
The LGPS’s funds are maintained by the administering authorities separately from the funds which they hold in other capacities. These funds are comprised of contributions from scheme members, scheme employers, and returns from the assets in which those contributions are invested. Together, as at March 2017, the administering authorities held assets of around £259 billion invested in a range of UK and global equities and bonds. The Council’s pension fund comprised about £1.3 billion, or 0.5% of the LGPS’s funds at that date.
As I have explained, strictly speaking, the LGPS consists of a series of different pension schemes in particular, the 1997 Scheme, the 2008 Scheme, and the 2014 Scheme. The way in which they operate can be summarised as follows. When a new scheme comes into force active members in the old scheme are transferred into the new scheme, subject to transitional protections for their accrued rights. Thus, it is likely that long-term local government employees will have been members of more than one scheme during their employment. Non-active (ie, deferred or pensioner) members at the time of the introduction of a new scheme remain in the previous scheme (although transitional provisions generally allow deferred members to become members of the new scheme if they rejoin LGPS employment after it has come into force). The old scheme remains in existence to allow payments to be made to pensioner members, but is closed to new members. Thus, when Mr Roe was made redundant in 2003 and became a pensioner member of the 1997 Scheme he remained a member of that scheme and was entitled to the benefits payable under it. He did not become an active member of the 2008 Scheme when that came into force on 1 April 2008.
Statutory provisions
It is now time to set out the principal statutory provisions. All of the relevant regulations concerning the 1997, 2008 and 2014 Schemes were made by the Secretary of State under ss 7 and 12 of the SA 1972.
(i) The SA 1972
Section 7(1) provides:
“
Superannuation of persons employed in local government service, etc.
(1) The Secretary of State may by regulations make provision with respect to the pensions, allowances or gratuities which, subject to the fulfilment of such requirements and conditions as may be prescribed by the regulations, are to be, or may be, paid to or in respect of such persons, or classes of persons, as may be so prescribed, being—
(a) persons, or classes of persons, employed in local government service; and
(b) other persons, or classes of persons, for whom it is appropriate, in the opinion of the Secretary of State, to provide pensions, allowances or gratuities under the regulations.”
Mr Giffin QC for the Claimant pointed to s 12, which allows the regulations to make retrospective provision for the payment of pension benefits:
“ 12 Further provisions as to regulations.
(1) Any regulations made under section 7, 8(2), 9 or 10 of this Act may be framed so as to have effect as from a date earlier than the making of the regulations.
(2) Subject to subsection (4) below, any regulations made under section 7, 9 or 10 of this Act may be framed—
(a) so as to apply in relation to the pensions which are being paid or may become payable under the regulations to or in respect of persons who, having served in an employment or office service in which qualifies persons to participate in the benefits for which the regulations provide, have ceased to serve therein (whether or not they have subsequently recommenced any such service) or died before the regulations come into operation; or
(b) so as to require or authorise the payment of pensions to or in respect of such persons.”
(ii) The 1997 Scheme
The 1997 Scheme is contained principally in the Local Government Pension Scheme Regulations 1997 (SI 1997/1612) (‘the 1997 Regulations’).
Regulations 40 and 41 provide respectively for short-term and long-term pensions to be paid to a ‘surviving spouse’. They state (as far as relevant):
“ Surviving spouse’s short-term pension
40(1) If an active or pensioner member dies leaving a surviving spouse, the spouse is entitled to a short-term pension.
(2) It is payable for three months after the member’s death…
…
(5) Where the deceased was a pensioner member, the short-term pension is equal to his retirement pension immediately before death…
Surviving spouse’s long-term pension
41(1) If a member dies leaving a surviving spouse, the spouse is entitled to a spouse’s long-term pension.
(2) If the deceased was an active or pensioner member, the long-term pension is payable from the end of the period for which the short-term pension is payable.
(3) The long-term pension payable on a deferred member’s death is payable from his death…
…
(6) If the deceased was a pensioner member, the long-term pension is equal to half of his retirement pension immediately before the date of death…”
The pension payable to the widow of a male member is calculated by reference to the member’s full length of service: see reg 20. In relation to the widower of a female member, only service on or after 5 April 1988 is taken into account: reg 42(2). As I will discuss later in this judgment, in R (Cockburn) v Secretary of State for Health [2011] EWHC 2095 (Admin), similar provisions in the NHS Pension Scheme were held not to violate Article 14 read with A1P1.
On 5 December 2005, by virtue of the Local Government Pension Scheme (Civil Partnership) (Amendment) (England and Wales) Regulations 2005 (SI 2005/3069), regs 40 and 41 were amended to include reference to civil partners as well as spouses. The amendment applied retrospectively: it took effect as from 1 April 1998 (the date on which the 1997 Scheme came into force). However, for the purposes of calculating the benefit to which a surviving civil partner was entitled, account is only to be taken of service after 5 April 1988: see reg 4A of the Local Government Pension Scheme (Transitional Provisions) Regulations 1997 (SI 1997/1613).
The 1997 Regulations did not provide for a survivor’s pension to be paid to an unmarried co-habiting partner.
By reg 26 of the 1997 Regulations, a member aged 50 or more who retired from local government employment, whose employing authority certified that the reason for his dismissal was redundancy, was entitled to an immediate pension and to a retirement grant. Mr Roe was 52 when he was made redundant, and he received these payments.
For completeness, it is to be noted that from 13 March 2014, when the Marriage (Same Sex Couples) Act 2013 came into force, the term ‘spouse’ in the 1997 Regulations includes same-sex spouses.
(iii) The 2008 Scheme
The 2008 Scheme is governed by the Local Government Pension Scheme (Benefits, Membership and Contributions) Regulations 2007 (SI 2007/1166) (‘the 2007 Regulations’) and the Local Government Pension Scheme (Administration) Regulations 2008 (SI 2008/239) (‘the 2008 Administration Regulations’). The 2007 Regulations came into force on 1 April 2008. Transitional provisions were made by the Local Government Pension Scheme (Transitional Provisions) Regulations 2008 (SI 2008/238) (‘the 2008 Transitional Regulations’).
The 2007 Regulations provide as follows.
Regulation 2 provides that employees of relevant bodies listed in the 1997 Regulations (ie, active members of the 1997 Scheme) became active members of the 2008 Scheme. This regulation, read with regs 4 and 13 of the 2008 Administration Regulations, produced the result that in order to become a member of the 2008 Scheme, an individual had to be in LGPS employment (ie, employed by a body listed in Part 1 of Sch 2 to the 2008 Administration Regulations) on or after 1 April 2008. The effect of this was that any member of the 1997 Scheme whose LGPS employment had already ceased by 1 April 2008 (and who did not rejoin LGPS employment after that date) did not become a member of the 2008 Scheme but remained entitled to the pension benefits under the 1997 Scheme. That was Mr Roe’s position.
The 2007 Regulations introduced a survivor’s pension for a cohabiting partner in certain circumstances. Regulation 36 contains the right to a cohabitee’s pension for a cohabiting survivor of a pensioner member, and provides as follows (as amended):
“ Survivor benefits: pensioners
36(1) If a pensioner member dies leaving a surviving spouse, nominated cohabiting partner or civil partner, that person is entitled to a pension, which shall come into payment on the day following death.
(2) The pension is calculated by multiplying his total membership by his final salary and divided by 160.
(3) If there is more than one surviving spouse, they become jointly entitled in equal shares under paragraph (1).”
Regulation 24 contains a right equivalent to that in reg 36, applicable to cohabiting survivors of active members of the 2008 Scheme, and reg 33 contains a right equivalent to that in reg 36, applicable to cohabiting survivors of deferred members of the 2008 Scheme.
Regulation 25 defines the phrase ‘nominated cohabiting partner’ as follows, so far as material:
“ Meaning of ‘nominated cohabiting partner’
25(1) “Nominated cohabiting partner” means a person nominated by a member in accordance with the terms of this regulation.
(2) A member (A) may nominate another person (B) to receive benefits under the Scheme by giving to his administering authority a declaration signed by both A and B that the condition in paragraph (3) is satisfied for a continuous period of at least 2 years which includes the day on which the declaration is signed.
(3) The condition is that-
(a) A is able to marry, or form a civil partnership, with B,
(b) A and B are living together as if they were husband and wife or as if they were civil partners,
(c) neither A nor B is living with a third person as if they were husband and wife or as if they were civil partners, and
(d) either B is financially dependent on A or A and B are financially interdependent …
…
(8) In this regulation, “member” means an active member or a former active member who has become a deferred or pensioner member in accordance with these Regulations.”
In 2014 the requirement for a cohabiting partner to have been nominated by the member prior to death was removed by the Local Government Pension Scheme (Transitional Provisions, Savings and Amendment) Regulations 2014 (SI 2014/525) for those who were still active LGPS members. Subsequently, in In Re Brewster [2017] 1 WLR 519, the Supreme Court held the requirement for nomination in the equivalent Northern Irish pensions legislation infringed Article 14 read with A1P1.
It is to be noted that in some ways the 2008 Scheme was less generous than the 1997 Scheme. By regs 24, 33 and 36, the provision for short-term survivor pensions in the 1997 Scheme was removed. Also, by reg 19, a member dismissed by reason of redundancy/business efficiency, who had attained the age of 55, was entitled to immediate payment of their retirement pension (previously it had been 50, as I have explained). I will have more to say about the differences between the two Schemes later in this judgment when I consider the evidence.
The 2008 Transitional Regulations provide for transition between the 1997 and 2008 Schemes. Regulations 3, 4 and 6 provide:
“Membership accrued before 1st April 2008: active members
3 . —(1) This regulation applies to a person who was an active member of the 1997 Scheme and becomes a member of the Scheme by virtue of regulation 2 of the Benefits Regulations [ie, by remaining in LGPS employment on 1 April 2008].
(2) Notwithstanding the revocations effected by regulation 2, the regulations listed in the Schedule continue to have effect, subject to regulation 4, so far as is necessary so that—
(a) The person’s total membership accrued in the 1997 Scheme in respect of, or calculated by reference to, his service before 1st April 2008, and the pension rights accrued at that date, are preserved; and
(b) his benefits under the 1997 Scheme are payable immediately where benefits become payable without reduction under regulations 16, 17, 19 and 20 of the Benefits Regulations, or with the appropriate actuarial reduction in line with guidance produced by the Government Actuary where benefits become payable under regulations 18 or 30 of the Benefits Regulations.
(3) But his pay, for the purposes of any calculation of benefits under paragraph (2), is calculated in accordance with regulations 8 to 11 of the Benefits Regulations.
(4) And, for the purposes of regulations 24, 33 or 36 of the Benefits Regulations as they apply to nominated cohabiting partners or civil partners, only periods of membership after 5th April 1988 are to be taken into account.
Membership accrued before 1st April 2008: deferred members
4 . (1) This regulation applies to a person—
who was a deferred member of the 1997 Scheme;
(b) who becomes a member of the Scheme by virtue of regulation 13 of the Administration Regulations [ie, by becoming an LGPS employee again after 1 April 2008]; and
(c) whose preserved benefits under the 1997 Scheme have not yet come into payment.
(2) He may choose, by notice in writing to the appropriate administering authority within 12 months of becoming a member of the Scheme under paragraph (1)(b), to be treated as if regulation 3 applied.”
…
Survivor benefits
6(1) Subject to paragraphs 2 and 3, any survivor benefits payable in respect of a person to whom regulation 3 or 4 applies are calculated as if the Benefits Regulations applied to all of his membership, whether accrued before or after 1 st April 2008.
(2) The survivor benefits payable in respect of a deferred member to whom regulation 4 does not apply are calculated as if his total membership before 1 st April 2008 was a separate period of membership to which the conditions in force at the time he became a deferred member apply.
…”
In summary, therefore, the effect of the 2007 Regulations and the 2008 Transitional Regulations as far as material to this claim is as follows:
The 1997 Regulations were revoked, save for specified purposes, so that no person could join the 1997 Scheme after 31 March 2008 (reg 2 of the 2008 Transitional Regulations).
Active members of the 2008 Scheme who were previously active members of the 1997 Scheme had their total membership and pension rights accrued in the 1997 Scheme preserved, save that their pay for the purposes of such benefits was to be calculated in accordance with the 2007 Regulations (see reg 3 of the 2008 Transitional Regulations).
Survivor benefits payable in respect of such persons were calculated as if the 2007 Regulations applied to their whole membership of the LGPS, whether accrued before or after 1 April 2008. However, by reg 3(4) of the 2008 Transitional Regulations, only periods of membership after 5 April 1988 were to be taken into account for the purposes of survivors’ pensions for nominated cohabiting partners or civil partners.
Those who were deferred members of the 1997 Scheme (eg, those who left LGPS employment but were not then eligible for a pension) who rejoined LGPS employment after 1 April 2008 could within 12 months of rejoining choose to become a member of the 2008 Scheme as if reg 3 applied to them.
The survivor benefits payable to active members of the 2008 Scheme or deferred members of the 1997 Scheme who rejoined LGPS employment are determined by reg 6 of the 2008 Transitional Regulations.
It is common ground that, human rights considerations aside, these statutory provisions did not entitle the Claimant to a survivor’s pension. The reasons are as follows. Such a pension is available to members of the 2008 Scheme. But Mr Roe did not (and could not) become a member of the 2008 Scheme because of reg 2 of the 2007 Regulations, he having left his LGPS employment prior to 1 April 2008 and become a pensioner member of the 1997 Scheme. Therefore, the survivors’ benefits which applied to him were those contained in the 1997 Regulations, and these did not provide for a survivor’s pension for a cohabiting partner. The 2008 Transitional Provisions did not operate to afford the Claimant a pension because, as I have explained, in relation to service before 1 April 2008 they only provided survivors’ benefits to active members of the 2008 Scheme (active either by virtue of being in LGPS employment on that date, or by being a deferred member of the 1997 Scheme who rejoined LGPS employment after that date). Neither scenario applied to Mr Roe. It was for these reasons that the Council informed the Claimant in July 2017, following her enquiry, that although she was Mr Roe’s unmarried cohabiting partner, she was not entitled to a survivor’s pension.
(iii) The 2014 Scheme
I will deal with the 2014 Scheme for completeness.
The 2014 Scheme was introduced with effect from 1 April 2014, and is governed by the Local Government Pension Scheme Regulations 2013 (SI 2013/2356) (‘the 2013 Regulations’). Like the 2007 Regulations, the 2013 Regulations provide for survivors’ benefits for active, deferred and pensioner members (see regs 41, 44 and 47 of the 2013 Regulations), payable at a rate of 1/160th of pensionable pay.
The Local Government Pension Scheme (Transitional Provisions, Savings and Amendment) Regulations 2014 (SI 2014/525) (‘the 2014 Transitional Regulations’) provided for transition between the 2008 and 2014 Schemes. The 2008 Regulations were revoked, save for specified purposes, so that no person could join the 2008 Scheme after 31 March 2014 (reg 2). For the purposes of calculating eligibility of a person to survivor benefits under the 2008 Scheme and any earlier Schemes, the definition of ‘cohabiting partner’ in Sch 1 to the 2013 Regulations was substituted for the definition of ‘nominated cohabiting partner’ in reg 25 of the 2007 Regulations. In other words, the cohabiting partner of a person who died after 1 April 2014 would not need to have been nominated, in order to receive a survivor’s pension under the 2008 Scheme: see reg 17(9) of the 2014 Transitional Regulations.
(iv) Summary
The position so far as survivors’ benefits in the LGPS is concerned can therefore be summarised as follows:
1997 Scheme: initially survivors’ pension payable to different-sex spouses, with pension based on full length of service (for female spouses of male members; the pensions of male spouses of female members based on service back to 5 April 1988). Later extended to civil partners (2005) and to same sex spouses (2014), but only in respect of service on or after 5 April 1988. These latter changes had retrospective effect, in the sense that they were applied to pensioner and deferred members of the 1997 Scheme. (I will say more about retrospectivity later in this judgment when I discuss the evidence).
2008 Scheme: survivor’s pension introduced for nominated co-habiting partners of active members and deferred members of 1997 Scheme who became LGPS employees again after 1 April 2008 (and elected to become members of the 2008 Scheme). Requirement for nomination removed in 2014. Length of qualifying service for these purposes is back dated to service after 5 April 1988 (at the earliest). Having become a pensioner member of the LGPS in 2003, Mr Roe did not qualify for a survivor’s pension for the Claimant, his co-habiting partner.
2014 Scheme: provides for survivor benefits including for co-habiting partners for active, deferred and pensioner members.
The evidence
The nature of the LGPS as it applies in England and Wales (there are different schemes in Scotland and Northern Ireland) is addressed in detail in the first witness statement of Sophie Broadfield on behalf of the Secretary of State and in the first witness statement of Thomas Skeen on behalf of the Council. Ms Broadfield is the Deputy Director of the Local Government Finance Reform and Pensions Division in the Ministry of Housing, Communities and Local Government and is responsible, among other things, for policy and advice in relation to the LGPS. Mr Skeen is the Council’s Head of Pensions. There is also a witness statement from John Bayliss, an actuary with the Government Actuary’s Department, who has particular expertise in relation to the LGPS and advises the Government on it.
Ms Broadfield’s evidence
Ms Broadfield explains that the LGPS is a funded, rather than an unfunded, pension scheme. An unfunded scheme is one where there is no ‘pot’ of assets; pensions are met by the Exchequer out of tax revenue as they fall due. The LGPS is not funded by the Exchequer, and the benefits for members are the responsibility of individual administering authorities to be met out of the Scheme’s assets. If an individual authority is facing a deficit in its particular fund, the burden of meeting that deficit falls on the participating employers. As local authorities account for 80% of the value of the fund, these costs would predominantly fall on local taxpayers through rises in Council Tax. As at 31 March 2016, the LGPS’s liabilities over the Scheme as a whole were estimated at £254 billion, indicating an overall funding level of 85%. There is an estimated deficit of £37 billion. Employee contributions are on average 6.5% of pensionable pay and the standard employer contribution is 16.9%, which comes from public funds (principally, Council Tax receipts).
Pension benefits are a form of deferred pay, in which employees forego current income in the expectation of benefits in the future. Pension terms and conditions are an important part of recruitment for some roles in the public sector, and can offset the lower salaries which are generally on offer for those roles.
In a defined benefit scheme, such as the LGPS, the member is entitled to a defined level of pension benefit on retirement, which is calculated according to a formula set out in the rules of the scheme. Defined benefit schemes may also provide a range of other retirement benefits, such as lump sum payments and survivor benefits.
Inevitably, that means some pensioners may benefit more than others, notwithstanding they have contributed the same amount. For example, some members may die before or shortly after qualifying for their pension, whereas others may live for a long time and receive their pension throughout their retirement. Mr Roe was able to take an early retirement pension when he was made redundant because of his age.
Thus, Ms Broadfield says that it is unhelpful to look at the terms of a defined benefit scheme like the LGPS from the point of view of one member. The Scheme has to be designed and costed by reference to its coherence as a whole. The funding of the Scheme has to be designed to fund those benefits accrued at the point at which the Scheme is to commence, and those estimated will accrue over the lifetime of the Scheme, on the basis of the best available assumptions at the time it is being designed. The funding responsibility for the LGPS is a matter for the administering authorities, such as the Council, which are obliged to scrutinise carefully the Scheme’s deficits to ensure its long-term sustainability.
The content and effect of the legislation governing the LGPS has been changed at various points in time, relevantly, in 1997, 2008 and 2014. This has been done in order to maintain the affordability and effectiveness of the LGPS, as well as to reflect other changes considered appropriate in the Scheme’s terms and conditions and matters of governance. Different schemes have thus been created at different times, and collectively they form the LGPS. Sometimes the new scheme has very different rules, benefits and contribution levels to the previous scheme; sometimes the changes have been less significant. Each administering authority will, however, have a single fund to provide benefits under all the schemes taken as a whole.
Employees who have joined the LGPS and who work in local government for a lengthy period of time will therefore be likely to have been members of a number of different Schemes if they remain in post for long enough. This is because when a new scheme is implemented, individuals continuing to make employee contributions (ie, active members) will transfer into the new Scheme, subject to transitional protections for their accrued rights. They continue to contribute to the new Scheme and their contributions form part of the overall calculations as to its affordability.
However, those who have left their employment and have started to receive their pension payments (ie, pensioner members) are not transferred into the new scheme: their rights have already accrued and started to be paid. The older schemes are left in existence to allow these payments to be made to pensioner members, but are closed to new members. A pensioner member is no longer contributing to the Scheme; rather, he is drawing benefits from it. As a matter of fairness to the active members still making contributions, the pensioner member is not transferred into a new Scheme, with its different balance of benefits and liabilities. The pensioner member cannot be expected to have planned to receive a package of benefits only applicable after his retirement. Similarly, active members cannot be expected to pay contributions to fund benefits for someone who only ever paid contributions in exchange for a different package of benefits and liabilities.
A central aspect of the LGPS, therefore, is that administering authorities should be able to assess the amount of potential payments to members on the basis of fixed and predictable rules, which do not create unexpected liabilities that put the Scheme’s funding at risk. Those potential pension benefits (representing deferred pay), along with the current pay on which they are based, form the majority of the remuneration package. That package is designed to reflect the particular needs of the time, having regard to the need to recruit and retain employees, to be affordable for employers and the tax payers who ultimately fund the Scheme, and to operate fairly between members. That, says Ms Broadfield, is the most important reason why changes to benefits are only very exceptionally made retrospectively, and such changes are generally applied only to benefits to be earned in the future.
At [21] et seq of her witness statement Ms Broadfield explains the Government’s policy on survivors’ benefits and retrospectivity (a term which she defines and discusses later). She says that there are two relevant elements of Government policy in relation to survivors’ benefits in public sector pension schemes which have been consistently maintained since (at least) the late 1990s. The first is that widening the scope of survivors’ pension benefits beyond spouses was acceptable, providing that members made clear that they wanted such changes, and that the cost of such a benefit was borne by the membership through changes to the terms of the scheme, and not by the public purse. The second is that new benefits would not be applied retrospectively.
She illustrates the first point by reference to the Government’s 1998 Green Paper, ‘A New Contract for Welfare: Partnership in Pensions’ which said that if the general membership of a public service pension scheme wanted in future to extend eligibility for survivors’ pensions to unmarried partners and were prepared to meet the additional costs, the Government would be prepared to consider how to achieve it. Accordingly, Government policy was not that survivor benefits to cohabiting partners were to be opposed in principle, but was that the provision of such benefits had to be the desire of the membership, and the membership had to be prepared to pay for them, so that the costs of this benefit did not fall on the taxpayer. She also refers to a note produced by the Treasury in January 2001 on the provision of unmarried survivor benefits by public service schemes, which pointed out that the requirement set by the Green Paper that the costs of extra benefits should be met by scheme members through higher employee contributions or other scaling back of member benefits, ‘effectively rules out retrospection to cover non-active members’.
Ms Broadfield refers to another Treasury document, namely a briefing note produced in December 1998 in preparation for the Green Paper. This said at [12] – [14]:
“12. The principle of no retroactivity is fundamental to the development of pensions policy. Otherwise each evolution in the detail of pension benefits would carry a potentially huge cost in terms of accrued liabilities at the point of change, as well as higher costs accruing in the future. For instance, GAD [the Government Actuary’s Department] estimate that if all public service pension schemes extended survivor pensions to unmarried partners and backdated the change to cover all past service credits, the immediate impact on accrued liabilities would be of the order of £10 billion. Not only is it out of the question for this cost to be laid on public funds, but to make it a requirement of financing by scheme members would effectively block the option of changing scheme rules in this way, where it might for other reasons be desirable.
13. ‘No retroaction’ means that most serving employees would be able to build up only a fractional entitlement to a pension for an unmarried survivor by the time of their retirement. Retired and deferred pensioners would build up no entitlement at all. Unless this were generally understood by scheme members there would be a risk of any review or decision on survivor benefits raising false expectations.
14. No retroaction also means that where a new scheme were ( sic ) introduced incorporating more expensive survivor benefits financed by higher employee contributions, it would only be possible for members to transfer accrued service credits into the new scheme on less favourable terms than a year for year basis (if the rest of a scheme’s benefit provisions remained unchanged, providing one-for-one past service credits will be tantamount to a retroactive application of the new benefits).”
In her witness statement Ms Broadfield says that this passage explains the different senses in which the terms retrospectivity/retroaction can be used in the pensions context. When she refers to ‘retrospectivity’, she says that she means it in the sense of changes conferring new benefits upon persons who have left active membership of the LGPS. She says that is the sense which applies to the benefit which the Claimant is seeking, because unmarried survivor pensions were not a benefit under the 1997 Scheme of which Mr Roe was a pensioner member. The alternative meaning refers to changes in the benefits applicable which can be applied to active members for both current and past service, ie, including for a period of time for which they were not paying contributions at a level assessed to cover that benefit. She says the two senses of the term ‘retrospectivity’ are importantly different. Application to past service of active members is a step which can be taken because in the actuarial calculation of the cost envelope of the scheme as a whole, costs can be calculated so that active members will make payment contributions covering the extension to their own past service. Clearly, not every active member will make contributions which in their own case will pay for past service, and Ms Broadfield provides an example of an active member who retires a short time after the new scheme takes effect; however, she says that such individual variations are inevitable in any defined benefit pension scheme. The scheme and its contribution levels will have been set on the basis of assumptions and actuarial calculations which mean that the active membership as a whole will have both taken the benefit of the change, and ensured that it is paid for.
Ms Broadfield goes on to say that, in contrast, that balance cannot be achieved if a new benefit is conferred upon a pensioner member who has ceased to make any contributions at all. In principle, that would be asking active members to subsidise the benefit for a generation of pensioner members who have made no contribution towards the cost of that benefit, in effect reducing the active members’ pay (through setting their employee contributions at a higher level than would otherwise be the case) in order to increase the deferred pay of those who are no longer employees. Ms Broadfield says it is highly unlikely to be fair to impose that cost on active members, who would already be paying for their own entitlement to that new benefit. She says that a lack of fairness in how benefits and obligations are allocated and paid for would disincentivise membership of the LGPS and recruitment into local government employment.
Ms Broadfield then explains that the extension of provision to unmarried partners of members who had already left the 1997 Scheme would not have assisted recruitment, retention or motivation. It would have provided a further benefit to members in circumstances where the 1997 Scheme was already more generous than the 2008 Scheme, because of something called ‘the Rule of 85’; this was a rule which allowed someone whose age plus years of pensionable service exceeded 85 to take early retirement without their pension being actuarially reduced. The 1997 Scheme also provided other additional benefits not in the 2008 Scheme, including a lump sum payment in certain circumstances.
In support of her evidence that the ‘no-retroactivity’ approach to pension benefits has remained Government policy, Ms Broadfield quotes from a Government Review, the ‘Review of Survivor Benefits in Occupational Pension Schemes’ (June 2014). Paragraphs 2.24-2.26 reiterates the policy and explains the rationale for it:
“2.24 In conducting a valuation of a funded or unfunded pension scheme, the scheme actuary must make many assumptions in order to put a value on the benefits that will be earned in the future - for example the longevity of members, and the rate of return that will be earned on the scheme’s assets. However, the actuaries must also consider the scheme’s rules and any relevant legislation which is in place at the time of the valuation. Actuaries would not make any allowance for any future change in the scheme rules, or in the relevant legislation, unless they were certain that this change would occur.
2.25 It follows, therefore, that any improvements to benefit entitlements which take place after the period in which they have been earned and paid for will increase the size of the scheme’s liabilities in a way which was not expected when contributions to the scheme were paid. Without any corresponding increase in the scheme’s assets, retrospective improvements such as these would, all else being equal, create a deficit in the scheme’s funding. In a funded trust based scheme, the additional cost of funding this kind of shortfall would initially fall on employers (and so in the case of public service schemes, on the Exchequer), and in some cases on scheme members. Given the potential for retrospective changes to create scheme deficits, successive Governments have maintained a policy presumption against making or mandating retrospective changes to pension schemes. This particularly applies when a pension has already been awarded, as there is then no scope for offsetting reductions in the benefits that have been awarded or for the member to make additional contributions.
2.26 When making changes to scheme rules, or to legislation which will affect the value of the members’ pension, the general position has therefore been to make these changes prospectively. This means that these changes will apply to benefits which were earned after the changes are made - but not necessarily those which are only paid in the period following any change. This avoids the risks to scheme funding that would be created if schemes are required to meet legal obligations to pay benefits which did not exist at the time those benefits were accrued.”
In further support, Ms Broadfield quotes from a letter dated 12 June 2002 from the Head of the Local Government Pensions Division on behalf of the Government responding to a proposal from the Local Government Pensions Committee (representing employers and trade unions) dated 24 July 2001 proposing the extension of survivors’ benefits to cohabiting partners. The letter reiterated the policy position set out in the Green Paper and addressed the element of the proposal that would have extended the benefit retrospectively in the sense that past membership would be taken account of, but not applied to pensioner or deferred members:
“We agree that any extension of survivor benefits should apply only in respect of active members. It has been the consistent policy of consecutive governments that scheme benefit improvements should as a general rule not be made retrospective but should apply from a given date in respect of future service. However, for partners’ benefits it is agreed that past membership can be taken into account, provided that the member meets the cost. You would need to formulate practical proposals to achieve this.”
In summary, therefore, Ms Broadfield’s evidence is that it is not objectionable in principle to change a defined benefit pension scheme so that it confers a benefit on active members in respect of a period of past service even though at the time of that service they were not paying for the benefit. That is because the actuarial calculations for the new scheme can be performed so as to ensure that active member contributions taken as a whole going forward cover the cost of the benefit for the period of past service (and which those paying members will receive), as well as for the benefit in relation to the period of service going forward. What is objectionable, and what the Government through its policy approach has set its face against, is to change a pension scheme so that it confers a benefit on pensioner members who are no longer paying contributions but are drawing benefits. That is because, if such a change were made, it would result in those pensioner members receiving a benefit that they have not and will not be required to pay for, but which will have to be paid for by other employees through their pension contributions, in effect, reducing their pay.
Later in her statement at [53]-[54] Ms Broadfield discusses what she accepts was an exception to this general policy in relation to civil partnerships. She says the decision was taken that, with the coming into force of the Civil Partnerships Act 2004, all public sector pension schemes would provide survivors’ benefits for civil partners to be calculated on the basis of any service after 5 April 1988. She says the Government was always clear that this was an exception to its ordinary policy but that Ministers considered that an exception was appropriate because of the radical change in the law which the Act represented, rectifying the historic wrong which had prevented same-sex couples from entering a legally recognised partnership. The pension change was a reflection of that commitment (unlike, comments Ms Broadfield, heterosexual couples who choose not to marry).
Although Ms Broadfield accepts (as the Claimant accepts) that the 1997 Scheme did not in terms provide a survivor’s pension for an unmarried partner, whereas it did so for a surviving spouse, she goes on to explain that it did contain a mechanism whereby a pensioner member could, within a month of retiring, nominate a dependent to receive part of their pension, and for them to surrender part of their pension entitlement so to provide. Ms Broadfield’s evidence is that administering authorities would regard a long-term cohabiting unmarried partner as a dependent for these purposes. She says at [37]:
“In relation to the 1997 Scheme, the Government’s position at the time this Scheme was created was that it was sufficient that Scheme members wanting to provide for unmarried partners were allowed to surrender part of their own pension and for this to be put towards a pension for their dependent after death. Further, they were entitled to leave their lump sum death grant to a person of their choice under regulation 38. I am not aware of whether or not Mr Roe took advantage of these rights.”
In the next section of her statement Ms Broadfield describes the policy work which took place in relation to public sector pensions, and survivor benefits in particular, in the late 1990s, leading up to the December 1998 Green Paper. In November and December 1998 the GAD estimated that the costs across all public sector pension schemes for permitting all active members to nominate an unmarried survivor would be around £600 million annually. If past service were included there would be an additional capital cost of £2 billion, with the potential to rise to £10 billion. GAD did not additionally calculate the cost of extending the provision to pensioner members.
The Green Paper of December 1998 set out the Government’s view that the significant potential costs of extending survivor benefits would have to be borne by the membership and be limited by a rule providing for no retrospection. There would also have to be clear, workable and economical administrative rules for determining who was eligible. At [47] Ms Broadfield comments that although different views appear to have been expressed in relation to the taking into account of past service of active members, so far as she has been able to discover, there was no suggestion during this time that it would be fair, appropriate or financially viable to extend survivor benefits to unmarried partners of pensioner members.
Ms Broadfield then describes wider policy work which went on within the Government in the early 2000s. At that time the Government wished to introduce a new civil service pension scheme including survivor benefits for unmarried partners, which the membership was strongly in favour of and was prepared to pay higher contributions for. In late 2001 the Treasury prepared a note on pension developments in various areas, and reiterated that Government policy remained that there should be no retrospective improvements to benefits for non-active members. Other work done around this time showed that to extend benefits would result in an added cost to public sector payrolls of between £200 million and £1 billion, and an addition of £3 billion or more if past service were included. That figure, says Ms Broadfield, did not appear to include pensioner members.
At [75] Ms Broadfield discusses representations that were received from trade unions during an LGPS ‘stocktake’ in 2003. These supported the extension of unmarried partners’ survivor benefits, but none of them called for them to be extended in relation to past service, or extended to non-active members.
In 2004 Ministers decided there should be a consultation in readiness for an anticipated replacement for the 1997 Scheme. Ms Broadfield addresses this at [76] et seq of her statement. The consultation was a long one and ran until March 2005, which she says reflected the range and complexity of the matters being consulted upon. The extension of survivor benefits to unmarried partners was one of the matters consulted on. Pensioner and deferred members would retain their benefits under the 1997 Scheme. The position taken in the consultation was that cohabiting partners’ pensions would be based upon membership accrued after the date the provision was introduced. It stated that it was a question for consultees whether the past service of existing members should be taken into account, but if it was, then it could only be at member cost. As I have explained by reference to the statutory provisions, the 2008 Scheme did in the event include past service. Ms Broadfield says at [82] – [83]:
“82. There was no suggestion that survivors’ benefits should be retrospectively applied to pensioner or deferred members.
83. A variety of views were expressed by consultees on the introduction of cohabiting partners’ pensions, but I have reviewed the detailed spreadsheet of consultation responses running to over 200 pages and none proposed that the Government should introduce cohabiting partners’ pensions on a retrospective basis for non-active scheme members. This is also reflected in the consultation response summary document produced at the time. Unions representing scheme members, who had been particularly vocal in supporting the addition of survivors’ benefits for cohabiting partners during the stocktake, responded that they were in favour of the proposal. None proposed extending it to deferred or pensioner members. In terms of recognising past service of active members, only the GMB response argued for active members’ past service to be counted towards benefits.”
In 2005 – 2006, following the consultation, work continued on a new Scheme. It was agreed that the ‘Rule of 85’ would be abolished, and that the cost savings would be used to provide improved benefits for Scheme members. At that stage the LGPS had developed a substantial deficit because of underlying investment issues (in particular the ‘dot.com’ crash of 2000), and this led to significant increases in employer contributions (funded predominantly through Council Tax).
On 30 June 2006 the Government published a consultation paper, ‘Where next ? Options for a new-look Local Government Pension Scheme in England and Wales’. The paper noted in Chapter 12 that as at 2004, the average funding level of the individual local funds was running at 74%, meaning that there were significant LGPS deficits. It was intended that the new Scheme would be introduced on 1 April 2008, and the consultation ran until the end of September 2006.
Chapter 13 of the consultation explained that the ordinary LGPS approach would be adopted of transferring existing members into the new Scheme. It was not proposed that pensioner or deferred members would be transferred into the new Scheme.
At [108] of her statement Ms Broadfield states that following receipt of the extensive responses to the 2006 consultation, officials prepared a lengthy internal document summarising them. She says that no consultee argued that cohabiting partner survivor benefits should be retrospectively extended to pensioner members.
The Government announced its intentions to the House of Commons in November 2006, and consultation on the draft Regulations began in December 2006. These provided for survivor benefits for unmarried partners of Scheme members, with service after 5 April 1988 being taken account of for the calculation of those benefits.
On 4 February 2007 officials wrote to stakeholders and indicated that transitional provisions would confirm that membership between 1988 and 2008 would count for the purposes of calculating unmarried partner pension benefits and that ‘In the case of a pensioner at 1 April 2008 continue to pay survivor benefits as under the 1997 Regulations’ (sic).
At [121] of her statement Ms Broadfield describes the 2008 Scheme, which came into force on 1 April 2008. At [127] she explains that in many respects the package provided in the 2008 Scheme was of greater attraction and benefit to members than the 1997 Scheme and comments that had that not been the case then the Government would have failed to achieve the aim of maintaining the LGPS as a valuable recruitment and retention tool. However, she goes on to say that not every alteration benefitted members. In particular, she emphasises that under the 1997 Scheme Mr Roe was entitled to draw down his LGPS pension at the age of 52. Under the 2008 Scheme the minimum age was raised to 55. She concludes:
“Mr Roe’s contributions to the 1997 Scheme were set at a level which sought to fund a package entitling him to draw down his pension from the age of 50, but not for an unmarried partner to acquire survivors’ benefits. Had Mr Roe been a member of the 2008 Scheme, his contributions would have been set at a different level, which sought to fund a package which did provide for unmarried partner survivors’ benefits, but which would not have permitted him to receive his pension until he was 55.”
Mr Skeen’s evidence
Mr Skeen explains that Mr Roe was employed by the Council from 1979 until October 2003, when he was made redundant aged 52. He was a member of the 1997 Scheme. Under the LGPS rules in force at the time, upon his redundancy Mr Roe was eligible to retire and to access his pension immediately, with no actuarial reduction for drawing his pension early before his normal retirement age. In addition to the pension membership he had built up, the Council awarded him four years of additional pension (‘added years’), which was permitted under the rules at the time.
Mr Roe’s annual pension was based upon 1/80 of his final pay for each year of service with the Council; he was also entitled to a lump sum of three times his annual pension. He also exercised his right to convert some of his pension into a lump sum.
Mr Roe received his monthly pension payments until his death in June 2016, equating to about 12 ½ years of payments.
In his statement at [15] et seq Mr Skeen provides information about the LGPS, some of which I set out earlier. He explains that as a funded scheme, the LGPS is potentially affected by fluctuations in the value of its investments. However, while the fund’s assets fluctuate in value in line with market movements, the benefits payable to pensioners are fixed. In essence, this is the distinction between a defined benefit scheme and a defined contribution scheme. The overriding aim is to invest the fund’s assets successfully in order to reach a 100% or more funding level (ie, so that assets equal or exceed liabilities), however Mr Skeen says that at the present time most funds are in deficit. Ultimately, it would fall to the employing authorities for a particular pension fund to make good any shortfall through increased contributions.
The LGPS provides a comprehensive suite of benefit entitlements, which may change from time to time due to emerging societal or demographic changes. This is the norm for pension schemes generally: pension entitlements must evolve and keep pace with such changes to ensure that they provide appropriate and adequate pension benefit packages for employees and pensioners, and impose sustainable burdens on contributing employees and employers.
Mr Skeen explains that survivors’ benefit entitlements under the LGPS have evolved significantly over the decades. Originally no survivors’ benefit entitlement existed. In 1954 widows’ pensions were introduced in a fairly limited set of circumstances. In 1972 widows’ pensions for married female spouses were introduced and made widely available to the male membership of the LGPS, however no corresponding widowers benefit for married male spouses existed for the majority of female scheme members. Married male members of the scheme had to contribute extra towards widows’ pension entitlements on pre-1972 service and had the option to do this via an increased contribution rate, or a reduction in their lump sum on retirement.
In 1988 survivors’ benefits for married male spouses of female LGPS members were introduced (ie, widowers’ pensions) although only payable based on service after 5 April 1988, unless the member elected to buy extra years.
Mr Skeen then deals with the changes that were made for same-sex couples and the introduction in 2008 of survivors’ benefits for cohabiting partners, which I have already described.
The LGPS benefit entitlements are designed to form a framework of individual benefits and protections, which active members of the scheme may weigh up in order to choose whether to join or stay in the Scheme and contribute towards those benefits, or choose not to, and opt out of the Scheme. Entitlements include (or have in the past included) regular pension payments; a lump sum payment on retirement; survivor benefits to eligible children, spouses and partners; and ill-health pension entitlements.
At [28] – [29] Mr Skeen discusses the benefits available under the 2008 Scheme as compared with the 1997 Scheme. The 2008 Scheme was more advantageous in one way, in that the accrual rate was increased from 1/80 to 1/60 of final salary per year of service. On the other hand, it was less advantageous in that the automatic lump sum entitlement on retirement was removed, and the earliest age for retirement on the grounds of redundancy was increased from 50 to 55. The Rule of 85 was also removed (with transitional provisions applying).
At [30]-[31] Mr Skeen says:
“30. As the LGPS schemes have evolved over time, they have been deliberately designed to provide for clear and predictable rules, and to avoid retrospectivity. Benefit entitlements are only awarded under the schemes to which a member of the schemes contributed: ie a member of the 2008 Scheme who worked until 2013 and then became deferred or retired, would not benefit from an entitlement introduced in 2014 under the 2014 Scheme (but equally that member’s entitlements would not be affected by any reductions in entitlements under the new Scheme) …
31. Members of a scheme will understand the suite of benefits they have bought into when they are active members of the scheme, as they are contributing towards the cost of these benefit entitlements through regular deductions from their salary. It would therefore be inappropriate for a member of the Scheme to be awarded a benefit introduced under a new scheme they did not contribute towards, or alternatively, to lose a pension benefit that was taken away under a subsequent scheme of which they had not become a member. The suite of pension benefits that a member of the scheme has understood and accepted, when they have agreed to regular deductions to be taken from their salary cannot be retrospectively diminished or enhanced.”
At [38] et seq Mr Skeen compares the differences in benefit packages pre-2008 and post-2008 and attempts to compare what Mr Roe would have received on the assumption that he was born 10 years later, had the same number of years of LGPS service, and retired aged 52 as a member of the 2008 Scheme. It is unnecessary to set out the detail, however Mr Skeen points out that although the 2008 Scheme would have provided the Claimant with a survivors’ pension, Mr Roe himself would have received less annually by way of pension payments. The general point which Mr Skeen makes is that the two schemes had different benefit packages, both in substance and in value. He goes on to say that the benefit which the Claimant seeks to obtain in this litigation was not costed into the 1997 Scheme, and therefore was not contributed to by Mr Roe during his working life or through his employer’s contributions.
In relation to the Claimant’s assertion of discriminatory treatment as compared with the spouses of 2008 Scheme members, Mr Skeen says at [60] that the past pension scheme entitlements which he described earlier in his statement show a number of differences in treatment between, in particular, males and females. He says that it can be assumed that at the time the relevant legislation was written it was accepted that the difference in treatment was due to a genuine and reasonable demographic trend.
Mr Bayliss’ evidence
Mr Bayliss’ witness statement contains some fairly detailed and technical actuarial evidence about the valuation of the LGPS and about how proposed changes in the Scheme over time have been costed. It is not necessary for the purposes of this claim to set out this evidence in detail. It is sufficient to note that Mr Bayliss has attempted to estimate the cost of providing survivors’ benefits to unmarried co-habiting partners in the Claimant’s position. He says that part of the difficulty in doing so lies in identifying accurately how many such persons there might be. To make the estimate, societal trends in living arrangements as identified by the Office for National Statistics have to be used. Mr Bayliss estimates future costs at £600 million. The total costs of providing backdated payments he estimates at £300 million (see at [100]-[101]). These figures are based on an assumption that members’ full service is taken into account. If service is limited to post-1988 then the figures are £350 million and £150 million respectively ([103]-[104]). However, these are only estimates, and the confidence intervals from the Office of National Statistics show that the upper cost limits in each scenario could be as much as £1.1 billion and £600 million respectively.
Article 14 and A1P1: the approach
The relevant provisions of the Convention
A1P1 provides:
“Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law.
The preceding provisions shall not, however, in any way impair the right of a state to enforce such laws as it deems necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties.”
Article 14 of the Convention states:
“Prohibition of discrimination
The enjoyment of the rights and freedoms set forth in this Convention shall be secured without discrimination on any ground such as sex, race, colour, language, religion, political or other opinion, national or social origin, association with a national minority, property, birth or other status.”
Preamble
Article 14 and A1P1 were recently considered in the context of the compensation scheme for the victims of crime by the Court of Appeal presided over by the Master of the Rolls in JT v First Tier Tribunal [2018] EWCA Civ 1735. Judgment in that case was given after the hearing in the present case, and the parties have addressed it in written submissions. Parts of my discussion below are gratefully adapted from the judgment of Leggatt LJ (with whom the Master of the Rolls and Sharpe LJ agreed).
Shortly after I had prepared this judgment in draft and circulated it to the parties in the usual way, the Supreme Court gave judgment in In Re McLaughlin [2018] 1 WLR 4250. In that case the Court, overturning the decision of the Court of Appeal of Northern Ireland, held that the restriction of widowed parent’s allowance to married parents or civil partners whilst denying it to unmarried parents such as Ms McLaughlin infringed Article 14 read with Article 8 of the Convention. Although at the hearing neither side had invited me to postpone the consideration of the Claimant’s case until In Re McLaughlin had been decided by the Supreme Court, I thought it right to invite the parties to provide further written representations on the decision if they wished to do so. The Claimant, the Council and the Secretary of State provided brief written submissions. The Claimant submitted that while there were aspects of the decision which supported her case, it could not be said to be directly on point, and she did not submit that there was anything in it which by itself demanded either the reversal of the conclusions I had reached in the draft judgment, or the fundamental rewriting of that judgment. The Secretary of State’s and the Council’s observations were to the effect that the decision supported my reasoning and conclusions.
I made some amendments to my draft judgment to take account of what is said in the majority’s judgments but I am satisfied that, subject to those relatively minor changes, the Supreme Court’s decision does not require any reconsideration of the views which I expressed in my draft judgment and the overall conclusion that I reached.
Discussion
The cases show that there are different ways of analysing what must be demonstrated in order to establish a violation of Article 14, although each analytical route crosses the same bridges and ends up in the same place. In Ghaidan v Godin-Mendoza [2004] 2 AC 557, [133] – [134], Lady Hale analysed the requirements of Article 14 by reference to five questions:
“133. It is common ground that five questions arise in an article 14 inquiry, based on the approach of Brooke LJ in Wandsworth London Borough Council v Michalak [2003] 1 WLR 617, 625, para 20, as amplified in R (Carson) v Secretary of State for Work and Pensions [2002] 3 All ER 994, 1010, para 52; [2003] 3 All ER 577. The original four questions were: (i) Do the facts fall within the ambit of one or more of the Convention rights? (ii) Was there a difference in treatment in respect of that right between the complainant and others put forward for comparison? (iii) Were those others in an analogous situation? (iv) Was the difference in treatment objectively justifiable? Ie, did it have a legitimate aim and bear a reasonable relationship of proportionality to that aim?
The additional question is whether the difference in treatment is based on one or more of the grounds proscribed—whether expressly or by inference—in article 14. The appellant argued that that question should be asked after question (iv), the respondent that it should be asked after question (ii). In my view, the Michalak questions are a useful tool of analysis but there is a considerable overlap between them: in particular between whether the situations to be compared were truly analogous, whether the difference in treatment was based on a proscribed ground and whether it had an objective justification. If the situations were not truly analogous it may be easier to conclude that the difference was based on something other than a proscribed ground. The reasons why their situations are analogous but their treatment different will be relevant to whether the treatment is objectively justified. A rigidly formulaic approach is to be avoided.”
In R (Carson) v. Secretary of State for Work and Pensions [2006] AC 173, [64], Lord Walker said that this five-step approach might not always be the best one, and counselled caution about an over-rigid stepwise process, especially in the search for a comparator; see also Lord Hoffmann at [28] – [34].
In R (JS) v Secretary of State for Work and Pensions (Child Poverty Action Group intervening) [2015] 1 WLR 1449, [8], Lord Reed took a four-stage approach, and said that the following must be shown to establish a breach of Article 14. There must be (a) a difference in treatment; (b) of persons in a relevantly similar position; (c) which does not pursue a legitimate aim, or (d) there is not a reasonable relationship of proportionality between the means employed and the aim sought to be realised. He derived this approach from what the Grand Chamber of the European Court of Human Rights said on the application of Article 14 in Carson v United Kingdom (2010) 51 EHRR 369, [61]:
“… in order for an issue to arise under article 14 there must be a difference in the treatment of persons in analogous, or relevantly similar, situations. Such a difference of treatment is discriminatory if it has no objective and reasonable justification; in other words, if it does not pursue a legitimate aim or if there is not a reasonable relationship of proportionality between the means employed and the aim sought to be realised.”
In Mathieson v Secretary of State for Work and Pensions [2015] 1 WLR 3250, [16] – [47], Lord Wilson analysed the Article 14 issue arising in that case by way of three questions: (a) Did the difference in treatment fall within the ambit or scope of a Convention right ? (b) Did the difference in treatment arise on one of the grounds in Article 14 ? (c) Was the discrimination justified ? Lord Wilson considered the question of persons in a relevantly similar position at the second stage.
Most recently, in In Re McLaughlin , supra, Lady Hale said at [15] that Article 14 raised four questions: (a) Do the circumstances ‘fall within the ambit’ of one or more of the Convention rights ? (b) Has there been a difference of treatment between two persons who are in an analogous situation ? (3) Is that difference of treatment on the ground of one of the characteristics listed or ‘other status’ ? (4) Is there an objective justification for that difference in treatment?
These different modes of analysis simply demonstrate, as Lady Hale said in Ghaidan and again in In Re McLaughlin , [15], that there is room for overlap in the various questions that have to be considered in relation to Article 14 and that a rigidly compartmentalised approach is to be avoided.
Dealing with the first question concerning ambit, as its opening words make clear, Article 14 is not a freestanding prohibition of discriminatory treatment. It applies only in the context of securing the rights and freedoms set forth in the Convention. But this does not mean that the scope of Article 14 is limited to cases where there has been a breach of another Convention right. The European Court of Human Rights has held that, where a contracting state goes further than the Convention requires in protecting any of the rights set forth in the Convention, it must do so in a manner compatible with Article 14. In the phrase favoured by the Court, Article 14 applies to those additional rights falling ‘within the ambit’ of any Convention article for which the state has voluntarily decided to provide. It is not in issue that pension rights fall within the ambit of A1P1: In Re Brewster , supra, [45]; Cockburn , supra, [2]; Runkee v United Kingdom , Application 42949/98, [34]-[36]. Thus, although the United Kingdom is not obliged by A1P1 to provide pension benefits to local government employees, because it has done so, it must do so in a manner that is compatible with Article 14.
The next question is whether there is a difference in treatment. Again, that is not in issue in this case, although I will say more about it later.
I deal next with ‘status’. Article 14 requires that the difference in treatment complained of be on the grounds of the applicant’s ‘status’, as that term is to be understood in relation to Article 14. The Article contains a list of grounds on which discrimination is prohibited. But the wording of Article 14 also makes it plain that the list is illustrative and not exhaustive. Thus, the list is preceded by the words ‘on any ground such as’ and ends with the words ‘or other status’. The approach of the European Court of Human Rights has been to interpret that phrase (‘ toute autre situation ’ in the French text) broadly. As interpreted, Article 14 is not restricted to grounds such as sex or race which are particularly ‘suspect’ because they are commonly or historically associated with prejudice and discriminatory treatment. In Mathieson, supra, Lord Wilson called these the ‘core’ grounds, and went on to say, by reference to what Lord Walker said in R (RJM) v Secretary of State for Work and Pensions (Equality and Human Rights Commission intervening) [2009] AC 311, [5]:
“Lord Walker offered the simile of a series of concentric circles and suggested that these core grounds fell within the circle of the narrowest diameter. But then there was a wider circle which included acquired characteristics, such as nationality, language, religion and politics. Indeed, so Lord Walker suggested, there was an even wider circle which included, for example, the homeless appellant then before the House; which also included the complainant in the Carson case 51 EHRR 369, who had chosen a particular country of residence; and which even included the complainant in Sidabras v Lithuania (2004) 42 EHRR 104, who had previously been employed by the KGB. The value of Lord Walker’s simile lies in what he then added [2009] AC 311, [5]:
‘The more peripheral or debateable any suggested personal characteristic is, the less likely it is to come within the most sensitive area where discrimination is particularly difficult to justify.’”
In Carson , supra, Lord Walker said at [55]-[58]:
“55. The proposition that not all possible grounds of discrimination are equally potent is not very clearly spelled out in the jurisprudence of the Strasbourg Court. It appears much more clearly in the jurisprudence of the United States Supreme Court, which in applying the equal protection clause of the 14th Amendment has developed a doctrine of "suspect" grounds of discrimination which the court will subject to particularly severe scrutiny. They are personal characteristics (including sex, race and sexual orientation) which an individual cannot change (apart from the wholly exceptional case of transsexual gender reassignment) and which, if used as a ground for discrimination, are recognised as particularly demeaning for the victim.
…
58. In its judgments the European Court of Human Rights often refers to "very weighty reasons" being required to justify discrimination on these particularly sensitive grounds. This appears, for instance (in relation to cases of discrimination on the ground of sex) in Abdulaziz, Cabales and Balkandali v United Kingdom (1985) 7 EHRR 471, 501, para 78; Schmidt v Germany (1994) 18 EHRR 513, 527, para 24; Van Raalte v Netherlands (1997) 24 EHRR 503, 518-519, para 39. When Harris, O'Boyle and Warbrick's valuable work, Law of the European Convention on Human Rights , was published in 1995, the authors recognised that the Strasbourg Court had its own suspect categories, identifying them as discrimination on the grounds of race, gender or illegitimacy. Since then religion, nationality and sexual orientation have, it seems, been added: see Jacobs and White, European Law of Human Rights , 3rd ed (2002), pp 355-6, citing Hoffmann v Austria (1994) 17 EHRR 293, 316, para 36; Gaygusuz v Austria (1997) 23 EHRR 364, 381, para 42 and Salgueiro da Silva Mouta v Portugal (2001) 31 EHRR 1055, 1071, para 36 …”
Lord Wilson observed in Mathieson , supra, [22], that if the alleged discrimination falls within the scope of a Convention right, the European Court is generally reluctant to conclude that nevertheless the applicant has no relevant status, with the result that the inquiry into discrimination cannot proceed. Its preferred approach is to take the nature of the ground into account at the subsequent stage of deciding whether the difference in treatment complained of amounts to discrimination. In other words, steps (c) and (d) in Lord Reed’s four stage test are, in practice, combined. The point was put by Lord Reed in JS , supra, [9], as follows:
“In practice, the analysis carried out by the European Court of Human Rights usually elides the second element—the comparability of the situations—and focuses on the question whether differential treatment is justified. This reflects the fact that an assessment of whether situations are “relevantly” similar is generally linked to the aims of the measure in question: see, for example, Rasmussen v Denmark (1984) 7 EHRR 371, para 37.”
As I shall explain, the Claimant maintains that the status from which her difference in treatment flows has three different aspects. First, she says it arises because she is unmarried. Second, she says it is because she is the cohabitee of a member of the 1997 Scheme, as opposed to the 2008 Scheme. Third, she says it has occurred on account of her age. The Defendants accept that the first and third of these fall within Article 14, but not the second.
I consider next the question of analogous position. For an issue to arise under Article 14, there must be a difference in the treatment of persons in analogous, or relevantly similar, situations, although the comparator does not need to be exact: In Re McLaughlin , supra, [24]; British Gurkha Welfare Society v UK , Application 44818/11, [62]. As Lord Hoffmann said in Carson, supra, [14], discrimination means a failure to treat like cases alike, and there is no discrimination when the cases are relevantly different. However, the Strasbourg court does not lay great emphasis on there being an exact comparator, and only in the most obvious cases does an Article 14 challenge fail on the absence of an analogous comparator: see eg AL (Serbia) v Secretary of State for the Home Department [2008] 1 WLR 1434, [24]. As the Court of Appeal recently pointed out in Smith v (1) Lancashire Teaching Hospitals NHS Foundation Trust (2) Lancashire Care NHS Foundation Trust (3) Secretary of State for Justice [2018] 2 WLR 1063, [83], in many cases this issue is swept up within the issue of justification. There is a considerable overlap between the questions whether the situations to be compared are truly analogous, whether the difference in treatment was based on a proscribed ground, and whether it had an objective justification. Unless there are very obvious relevant differences between the two situations, it is generally better to concentrate on the reasons for the difference in treatment in relation to the question of whether that differential is objectively justified. The Court said ([88]) that whether or not a cohabiting couple who are neither married nor civil partnered or the survivor of such a couple are in an analogous position with spouses for the purposes of Article 14 depends on the precise context in which the Article 14 issue arises. To the same effect is what Lady Hale said in In Re McLaughlin, supra, [26]:
“It is always necessary to look at the question of comparability in the context of the measure in question and its purpose, in order to ask whether there is such an obvious difference between the two persons that they are not in an analogous situation.”
Finally, I turn to the question of objective justification. Where the claimant shows she has been treated differently by reason of her status from a class of persons whose situation is relevantly similar, the question is whether there is an objective and reasonable justification for the difference in treatment (often called ‘the justification test’).
Because of the different ways this test had been expressed in the case law (to my mind, at least), I invited the parties to prepare written submissions on this specific topic, however those can now largely be taken to have been superseded by Leggatt LJ’s analysis in JT, supra, [81]-[86].
The way in which the justification test has been expressed in the Strasbourg jurisprudence is that, in order for a difference in treatment to be justified, the state must provide an ‘objective and reasonable justification’ for the difference in treatment, judged by whether it has a legitimate aim and there is a ‘reasonable relationship of proportionality’ between the aim and the means employed to realise it: see eg Rasmussen v Denmark (1985) 7 EHRR 371, [38]; Petrovic v Austria (2001) 33 EHRR 14, [30].
The test for justification in relation to Article 14 is therefore one of proportionality. The most often cited formulation of the proportionality test is now that of Lord Reed in Bank Mellat v HM Treasury (No 2) [2014] AC 700, [74], where he identified the assessment of proportionality as involving four questions:
“(1) whether the objective of the measure is sufficiently important to justify the limitation of a protected right, (2) whether the measure is rationally connected to the objective, (3) whether a less intrusive measure could have been used without unacceptably compromising the achievement of the objective, and (4) whether, balancing the severity of the measure's effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, the former outweighs the latter.”
Lord Reed dissented in the result of the appeal, but his analysis of the requirements of proportionality was common ground among the members of the Supreme Court: see [20] per Lord Sumption.
In JT, supra, [83], Leggatt LJ said that put more shortly, Lord Reed’s fourth question is whether the impact of the right’s infringement is disproportionate to the likely benefit of the impugned measure. He said that another way of framing the same question is to ask whether a fair balance has been struck between the rights of the individual and the interests of the community.
In his post-hearing written submissions Mr Cornwell for the Council tentatively queried whether proportionality, at least as explained in Bank Mellat , was indeed the test for justification in relation to Article 14. However, Lord Reed’s four-stage test was applied in relation to Article 14 in the pensions context by the Supreme Court in In Re Brewster ¸ supra, [66]. Also, in R (SC) v Secretary of State for Work and Pensions [2018] EWHC 864 (Admin), [133], Ouseley J said that the approach to the assessment of justification for a difference in treatment under Article 14 is the same as for an assessment of the proportionality of an interference with a Convention right. It is therefore the test which I must apply.
It is well settled in the case law of the European Court of Human Rights that states have a certain margin of appreciation in applying the test of proportionality, the breadth of which will vary according to ‘the circumstances, the subject matter and the background’: see eg Rasmussen v Denmark , supra, [40]; Petrovic v Austria (2001) 33 EHRR 14, [38].
The concept of a margin of appreciation is integral to the jurisprudence of the Strasbourg court. It arises because of the need for an international court, charged with the task of determining whether a domestic measure has infringed an individual’s Convention rights in a particular case, to have appropriate respect for the choice which a democratic state has made with respect to the measure in question, having regard to its own national conditions, and which it, rather than the Court, is best placed to assess. Where the context is social or economic policy then states have a wide margin of appreciation. In its judgment on the merits in Stec v United Kingdom (2006) 43 EHRR 74, [52], the Grand Chamber said:
“A wide margin is usually allowed to the state under the Convention when it comes to general measures of economic or social strategy. Because of their direct knowledge of their society and its needs, the national authorities are in principle better placed than the international judge to appreciate what is in the public interest on social or economic grounds, and the Court will generally respect the legislature’s policy choice unless it is ‘manifestly without reasonable foundation’.” [citations omitted]
The rationale for the margin of appreciation does not directly apply where a national court is considering whether a domestic measure has led to a violation of the Convention: R (Steinfeld and Keidan) v Secretary of State for International Development [2018] 3 WLR 415, [28]. Nonetheless, there is a broadly analogous principle that in some circumstances it is appropriate for the courts to recognise that there is an area of judgment within which the judiciary will defer, on democratic grounds, to the considered opinion of the elected body or person whose act or decision is said to be incompatible with the Convention: R v Director of Public Prosecutions ex parte Kebilene [2000] 2 AC 326, 381; In Re McLaughlin , supra, [34]. Accordingly, the Supreme Court held in Humphreys v Revenue and Customs Commissioners [2012] 1 WLR 1545, [15]-[20], that the ‘manifestly without reasonable foundation’ test is also the test to be applied by a UK domestic court when examining a justification advanced for a difference in treatment in a matter of economic or social policy. This has been confirmed by the Supreme Court in a number of subsequent cases: see JS, supra, [11], [93]; Mathieson , supra, [26-27]; R (Tigere) v Secretary of State for Business, Innovation and Skills [2015] 1 WLR 3820, [27], [75-77]; R (MA and Carmichael) v Secretary of State for Work and Pensions [2016] 1 WLR 4550, [36]-[38]. In relation to this, the term ‘margin of discretion’ is more apt than ‘margin of appreciation’, a term which Lord Mance said in In Re Medical Costs for Asbestos Diseases (Wales) Bill [2015] AC 1016, [44], does not apply at the national level.
How does the ‘manifestly without reasonable foundation test’ link with the four-stage Bank Mellat proportionality test ? The cases disclose two different approaches. In the Medical Costs case [46]-[52], Lord Mance (giving the lead judgment in the Supreme Court) discussed this question at some length. He concluded that the test is applicable at the first stage, when asking whether the measure has a legitimate aim, and possibly at the second and third stages. However, at the fourth stage where the court is required to weigh the benefits of the measure against its impact on individual rights, it may be appropriate to give significant weight to the choice made by the legislature but ‘the hurdle to intervention will not be expressed at the high level of ‘manifest unreasonableness’ ([46] and [52]). Lord Thomas CJ, who gave the other judgment, agreed with Lord Mance on this issue ([114]). Subsequently, the point was clearly expressed by Lord Wilson in giving the majority judgment in R (A) v Secretary of State for Health (Alliance for Choice and others intervening) [2017] 1 WLR 2492, [33]:
“… it is now clear that, while this criterion may sometimes be apt to the process of answering the first question, and perhaps also the second and third questions, it is irrelevant to the question of fair balance, which, while free to attach weight to the fact that the measure is the product of legislative choice, the court must answer for itself: see In Re Recovery of Medical Costs for Asbestos Diseases (Wales) Bill [2015] UKSC 3, [2015] AC 1016, para 46, Lord Mance.”
The second approach taken in some of the cases has been to apply the ‘manifestly without reasonable foundation’ test to the fourth Bank Mellat question. This was the approach adopted by the unanimous seven-judge Supreme Court in MA and Carmichael, supra , which applied ‘manifestly without reasonable foundation’ to the whole of the proportionality analysis (see [32]-[38] and [41] per Lord Toulson and [66] per Lady Hale). That judgment was approved by the Supreme Court in R (HC) v Secretary of State for Work and Pensions [2017] 3 WLR 1486 at [32] per Lord Carnwath. In SC, supra , at [140] Ouseley J applied the test to the fourth stage in reliance on MA and Carmichael , supra, and HC , supra.
The Secretary of State points out that in JT , supra, Leggatt LJ did not consider it necessary to decide whether the Medical Costs or the MA and Carmichael approach was to be preferred, pointing out that the Supreme Court might in due course revisit and clarify the correct analysis (see at [86]). It does not appear that either the approval of MA and Carmichael , supra, in HC , supra, or the judgment in SC, supra, or the approval of Medical Costs, supra , by Lord Wilson in A , supra, was cited to or considered by Leggatt LJ. The Secretary of State submits that in SC , supra , Ouseley J reached the correct conclusion on the basis of the authorities, and points out that in any event Lord Mance’s judgment in the Medical Costs case, supra, holds that at the fourth stage the courts ‘must attach appropriate weight to informed legislative choices at each stage in the Convention analysis’ (see [54]). I consider that the Secretary of State is correct in these circumstances – and in particular, in a context where on any view the legislative choice of the national authorities is to be afforded weight – to suggest that the distinction between the two approaches to the application of the ‘manifestly without reasonable foundation’ test at the fourth stage of the Bank Mellat analysis may be immaterial. However, like Leggatt LJ in JT , I propose to apply both versions of the test.
Although the ‘manifestly without reasonable foundation’ standard of intensity review has been most commonly applied to challenges in state welfare and benefits contexts, the same test has been applied or otherwise approved in the context of challenges to public sector pension schemes: see In Re Brewster¸ supra, at [55] per Lord Kerr; R (Turley) v London Borough of Wandsworth [2017]1 HLR 337, [39] per Underhill LJ (agreeing with [55] of In Re Brewster); Cockburn, supra, [38] per Supperstone J; British Gurkha Welfare Society v UK, supra, [62].
Finally, it should be pointed out that different factors may enhance the intensity of the court’s scrutiny even within a context such as social or welfare policy in which the ‘manifestly without reasonable foundation’ test applies. The principles were set out by Leggatt LJ in JT, supra, [87] – [90]. He said that although broad, the margin of discretion which the court should afford to a policy choice on a matter of economic or social strategy is nevertheless not without limit. As Lord Neuberger stated in RJM, supra, [57]:
"Of course, there will come a point where the justification for a policy is so weak, or the line has been drawn in such an arbitrary position that, even with the broad margin of appreciation afforded to the state, the court will conclude that the policy is unjustifiable."
In determining where this point comes, Leggatt LJ said that three important considerations emerge from the case law.
First, the fact that the hurdle for intervention is a high one does not mean that justifications put forward for the measure in question should escape careful scrutiny. On analysis, such justifications may lack a reasonable basis: see Humphreys¸ supra , [22]. Nor can this displace the fundamental principle that ultimately it is for the court to decide whether or not there has been a breach of a Convention right.
Second, the strength of the reasons required to justify a difference in treatment will vary according to the nature of the ground on which the difference in treatment is based. In MA and Carmichael , [37], Lord Toulson accepted that there are cases involving state benefits in which the European Court of Human Rights has spoken of a need for weighty reasons to justify a difference in treatment. He gave the example of Andrejeva v Latvia (2009) 51 EHRR 28, where state pension rules discriminated against the applicant on grounds of her nationality. Lord Toulson said that in that case ‘there was, on the face of it, no reasonable foundation for such discrimination, and in those circumstances, it was for the state to produce a good reason to justify it. In other words, the nature of the ground on which the difference in treatment is based will affect the readiness of the court to find that there was manifestly no reasonable foundation for it.
Third, a further important factor is whether or to what extent the values or interests relevant to the assessment of proportionality were actually considered when the policy choice was made. Thus, it is clear that where the public authority has addressed the particular issue before the court and has taken account of the relevant human rights considerations in making its decision, a court will be slower to upset the balance which the public authority has struck. But where there is no indication that this has been done, ‘[t]he court's scrutiny is bound to be closer and the court may have no alternative but to strike the balance for itself, giving due weight to such judgments as were made by the primary decision-maker on matters he or it did consider’: Belfast City Council v Miss Behavin' Ltd [2007] 1 WLR 1420, [26], [37], [47], [91]; Tigere , supra, [32]; and In Re Brewster , supra, [50]-[52].
Mr Giffin submits that these principles are in play in the present case, and that the approach I should adopt ought not to be as wide as the ‘manifestly without reasonable foundation’ test. I will return to this later.
The parties’ submissions
The Claimant’s case in summary
The Claimant has been denied a survivor’s pension by the Council in its capacity as the administering authority of the relevant LGPS fund. The Claimant’s case is that this denial amounts to discrimination against her in a manner contrary to the HRA 1998, in that the Council’s refusal breaches her rights under Article 14 read with A1P1 of the Convention.
The Claimant submits that the issue in question, namely pension rights, falls within the ambit of A1P1 for the purposes of Article 14. As I have said, this is common ground and is demonstrated by In Re Brewster , supra, among other cases.
Next, the Claimant submits that the difference in treatment of which she complains – her lack of entitlement to a survivor’s pension – is by reason of a qualifying status under Article 14, and that there is a difference in treatment between her and persons in analogous, or relevantly similar, situations. She puts the matter in two principal ways:
First, she says that she loses out on any survivor’s benefits because she and Mr Roe were not married. If she had been married to him, and he had served for exactly the same period of time at exactly the same dates in the same job, she would have been entitled to a survivor’s pension calculated by reference to his entire length of LGPS membership. Hence, she says her status for these purposes is her unmarried state, and she says that the pool of relevant comparators are ‘Pre-2008 spouses’, ie, spouses of pensioner members of the 1997 Scheme, who are eligible for a survivor’s pension.
Second, she says she loses out on any survivors’ benefits because of the requirement in the 2008 Scheme for the member of the 1997 Scheme to have been active as at 1 April 2008 (or to have been a deferred member of the 1997 Scheme as at that date and rejoined LGPS employment after that date) so as to become eligible for a survivor’s pension for an unmarried partner under the 2008 Scheme. She says that her status for these purposes is as the partner of a non-active 1997 Scheme member as at 1 April 2008, and that the pool of relevant comparators are ‘Post-2008 cohabitees’, ie, cohabitees of active members as at 1 April 2008.
Additionally, she says that the way in which the Scheme operates discriminates indirectly on the grounds of age, in that she says that a person in her position is likely to be older than the partner of a person who was still a member of the 1997 Scheme as at 31 March 2008 and who transitioned into the 2008 Scheme on 1 April 2008 (and thus whose co-habiting partner became eligible for a survivor’s pension).
In response to the Defendants’ contention that the Claimant is not in a relevant comparable position to either Pre-2008 spouses or Post-2008 cohabitees, the Claimant points to dicta (some of which I have already referred to) which she says show that the Strasbourg court does not lay great emphasis on there being an exact comparator and that only in the most obvious cases (of which this is not one) does an Article 14 challenge fail because of the absence of an analogous comparator: see eg AL (Serbia) , supra, [24].
In this case, the Claimant says that the notional comparator spouse and cohabitee are both individuals who have been in a significant relationship with a former LGPS member. Certainly, in the case of a cohabitee of the kind who has rights under the post-2008 Scheme, the relationship will be an enduring and financially interdependent one, and that will generally be true of the spouse as well. In each case, the individual's partner has paid for pension rights under the LGPS, both with his or her labour as an employee, and with his or her contributions. The level of those contributions is not affected by their marital status at the time or in the future. All that distinguishes them is the status accorded by society to their relationship. Whether such status makes the differential treatment lawful is a matter that goes to justification, and not to comparability. She relies in this regard on Smith , supra, and the approach taken there by the Court of Appeal to the question of analogous position which I have already set out. In short, the Claimant submits that the real issue is justification.
In relation to justification, and the suggested difference in treatment between the Claimant and Pre-2008 spouses, the Claimant accepts that the test is that of proportionality as explained in Bank Mellat , supra. She contends, however, that in the circumstances of the present case the ‘manifestly without reasonable foundation’ test is not the appropriate test. She says that there is no evidence of the Secretary of State ever having given serious contemporary consideration to whether the discrimination between spouses and stable cohabitees was justified on grounds of social policy, or otherwise. Indeed, she says the evidence shows that the Secretary of State, certainly by 1998 (a full 10 years before LGPS survivor benefit rights were actually extended to cohabiting partners), did not believe that this discrimination on the ground of marital status was right or justifiable. She also submits that discrimination on grounds of marital status should be treated as a ‘suspect’ ground calling for close scrutiny from the court. She maintains that neither the Council nor the Secretary of State has identified the ‘legitimate aim’ pursued in discriminating between cohabitees and spouses, either before or since 2008. Further, the Claimant submits that cost is no answer if discrimination is shown.
In relation to justification and the difference in treatment between the Claimant and Post-2008 Cohabitees, the Claimant maintains that her position as the unmarried partner of member of the 1997 Scheme is a relevant status for the purposes of Article 14, and says the words ‘other status’ in Article 14 are to be interpreted broadly: RJM , supra, [36] - [47].
In response to the Defendants’ contention that a Pre-2008 Cohabitee like the Claimant is not in a properly comparable position to a Post-2008 Cohabitee, the Claimant submits, again, that only in the most obvious cases should an Article 14 claim fail on the basis that the claimant’s situation is not analogous with that of her comparator. In any event, she says that the reality is that this case involves a direct disparity of treatment between a Pre-2008 Cohabitee and a Post-2008 Cohabitee in terms of the benefits payable in respect of exactly the same period of service (ie, service pre-2008), occurring at a time when the same Scheme was in force, and in relation to which exactly the same contributions were made by their respective partners. She says it is wrong to suggest that two such individuals are not in an analogous position as regards that period of service. Thus, as before, she says the true issue is justification.
As to that, the Claimant rejects the Defendants’ arguments that extending survivors’ pensions to Pre-2008 Cohabitees would be a windfall for persons who made no contribution to the costs of such pensions; that the cost of extending survivors’ pensions to Pre-2008 Cohabitees would be prohibitively expensive; that LGPS administering authorities must be able to assess their liabilities on the basis of fixed and predictable rules, and that granting survivors’ pensions to Pre-2008 Cohabitees would create unexpected liabilities; and that granting survivors’ pensions to Pre-2008 Cohabitees would pose practical difficulties for administering authorities in applying any cohabitation test. She also makes the general point, again, that there is no evidence of the Secretary of State ever giving any contemporaneous consideration to the discrimination between Pre- and Post-2008 Cohabitees in respect of service between 1988 and 2008, or to whether such discrimination was justified.
I turn to the Claimant’s case on age discrimination. She says that under the LGPS Regulations, cohabiting partners are entitled to a survivor’s pension for service from 1988 to 2008 only if their partner remained in LGPS employment, or rejoined LGPS employment, on or after 1 April 2008. She submits that it is plain that older cohabiting partners of LGPS members are less likely to fulfil this criterion than younger cohabiting partners of LGPS members. She says that is because, on average, most cohabiting couples are likely to be a similar age to one another; and the older a member of the LGPS is, the more likely it is that he or she will have permanently ceased LGPS employment before 1 April 2008, since it is more likely that he or she will have retired from work by that date. Hence, she submits that the rule that a cohabiting partner will not qualify for a survivor’s pension for service from 1988 to 2008 if his or her partner ceased LGPS employment before 1 April 2008 has a disproportionate adverse impact on older cohabitees and/or on cohabitees who were aged over 50 on 1 April 2008.
The Defendants’ case in summary
On behalf of the Secretary of State, Mr Milford did not dispute that pension entitlements fall within the ambit of A1P1 for the purposes of an Article 14 challenge, and he accepted that there has been a difference in treatment, but submitted that the claim should be dismissed for the following reasons.
First, he submits that the Claimant’s claim based on her status as a Pre-2008 Cohabitee fail at the outset because she does not compare herself to persons in relevantly similar positions for the purposes of Article 14 ECHR. Specifically:
As concerns the comparison between ‘Pre-2008 Cohabitees’ and ‘Pre-2008 Spouses’, it is well-established in the case law of the European Court of Human Rights that marriage confers a special status upon those who enter into it: hence, Mr Milford says, the Claimant (who chose not to marry) is not in a relevantly similar position to a Pre-2008 Spouse.
As concerns the comparison between ‘Pre-2008 Cohabitees’ and ‘Post-2008 Cohabitees’, the date upon which an individual’s partner retired, Mr Milford says this does not give rise to a relevant ‘status’ for the purposes of Article 14. In any case, Pre-2008 and Post-2008 Cohabitees are not in a materially similar position: their partners belonged to entirely different Schemes with different suites of benefits (the 2008 Scheme being more favourable than the 1997 Scheme in some respects, and less favourable in others).
In relation to the Claimant’s age claim, Mr Milford submits that the Claimant’s case is not in reality related to her status as an ‘older person’. By fixing a date for the implementation of the 2008 Scheme the Government was not discriminating against her on grounds of age.
Moreover, if any of the above comparisons were to disclose differential treatment between the Claimant and persons in a relevantly similar position, Mr Milford says that the hurdle for showing justification is lowered in circumstances where (a) the Claimant’s status and/or ability to claim under Article 14 is unclear in the first place; and (b) none of the forms of discrimination alleged amounts to discrimination on a ‘suspect’ ground for the purposes of Article 14.
If, contrary to the above, the Secretary of State is required to show justification, Mr Milford says that the approach taken in the legislation to survivor’s pensions for cohabiting partners is and has always been justified. The relevant test here is whether the legislative choice was ‘manifestly without reasonable foundation’. It is a test which affords the Secretary of State a wide margin of discretionary judgment because the context is socio-economic policy and the allocation of scarce public resources. Mr Milford says that there is no basis upon which the Secretary of State’s legislative choice can properly be impugned:
The approach taken in 2007-2008 to creating a new LGPS scheme which did not retrospectively extend the right to cohabitee pensions to pensioner members of previous schemes was justified by reference to a number of important legitimate aims. Those included (i) the establishment of a new Scheme which implemented desired benefit structures at a stable and affordable cost; (ii) managing and reflecting intergenerational fairness through the provision of benefits only to those who would pay for them through contributions; and (iii) adopting a Scheme which could readily be administered and which protected existing active members. It applied a clear, but necessarily bright line, rule in order to create a suitably funded overall package of benefits and costs.
The Claimant now wishes to take one of the benefits of the 2008 and 2014 Schemes without any of the costs, leaving those costs to be borne by members (who did not obtain the benefits of the earlier scheme) and by the taxpayer. The Secretary of State was plainly entitled to take the view when introducing the cohabitee benefit in the 2008 Scheme that such a retrospective windfall was not appropriate. Still less is he now required to redesign the 2014 Scheme to include a retrospective windfall of this type, at a cost of up to £1.1 billion for the LGPS alone (and far greater costs, were the same principles to be applied to all other public sector pension schemes), in a context where LGPS funds are heavily in deficit; where the costs of pension provision for an aging population are ever-increasing; and where local authorities are under huge financial pressure.
On behalf of the Council, Mr Cornwell supports Mr Milford’s arguments on behalf of the Secretary of State. He submits that:
The Claimant is not entitled to a pension on the ordinary meaning of the rules pertaining to the 1997 Scheme.
Article 14 does not produce a different outcome. The Claimant is not in analogous position to her suggested comparators. Alternatively, if she is, then any difference in treatment is justified. Also, says Mr Cornwell, the Claimant’s suggested status as a Pre-2008 Cohabitee is not a status for the purposes of Article 14.
There is a well developed body of Strasbourg case law supporting the proposition that married spouses and unmarried cohabitees are not in analogous positions in respect of pensions or other fiscal measures, and in any event, any difference in treatment is easily justified. Also, the Claimant is essentially seeking to avail herself of the benefits of the 2008 Scheme to which Mr Roe never belonged, without accepting the disadvantages of the later Scheme as compared with the advantages of the 1997 Scheme which he benefitted from.
The Claimant’s age claim is unsupported by any evidence, and in any event is parasitic upon her claim as a Pre-2008 Cohabitee versus a Post-2008 Cohabitee.
Discussion
The question with which I am concerned is whether the Claimant has been the victim of a violation of Article 14 read with A1P1 because she has been denied a survivor’s pension. It is well-established that the test for determining whether there has been discrimination under the Convention must be applied at the time the claim falls for determination, not as at the date when legislation was enacted or came into force: see eg Ghaidan , supra, [23] per Lord Nicholls. Hence, I am not concerned with the compatibility or otherwise of the 1997 Scheme with the Convention (which may have raised difficult questions in light of the non-retrospective nature of the Human Rights Act 1998), but with the question of whether the LGPS taken as a whole has resulted in a violation of the Claimant’s Convention rights.
What is not in dispute
(i) Ambit
As I have already said, it is not disputed that pension rights fall within the ambit of A1P1 for the purposes of Article 14: In Re Brewster , supra, [44] – [45]. In that case, the Supreme Court unanimously held that the requirement in the Northern Irish LGPS that a co-habiting partner be nominated in advance before she could receive a survivor’s pension violated Article 14 read with A1P1. In Cockburn, supra, Supperstone J rejected a challenge under Article 14 read with A1P1 in relation to a version of the NHS Pension Scheme, but it was common ground that the Scheme fell within the ambit of A1P1 for the purposes of Article 14 ([2]). In Runkee v United Kingdom, supra , the European Court of Human Rights said in relation to widows’ pensions (‘WP’):
“33. The Court considers that the applicants' complaints about the non-payment to them of WP fall within the scope of Article 1 of Protocol No. 1 (see Stec and Others v. the United Kingdom (dec.) [GC], nos.65731/01 and 65900/01, § 54, ECHR 2006- ...) …
34. The Court recalls that Article 1 of Protocol No. 1 does not include a right to acquire property. It places no restriction on the Contracting State's freedom to decide whether or not to have in place any form of social security scheme, or to choose the type or amount of benefits to provide under any such scheme. If, however, a State does decide to create a benefits or pension scheme, it must do so in a manner which is compatible with Article 14 of the Convention (see the Stec and Others decision, cited above, §§ 54-55).”
(ii) Difference in treatment
The facts of the present case show a difference in treatment. The Claimant has been denied a survivor’s pension whereas she would have received one had, for example, she been married to Mr Roe. That is a difference in treatment; whether it violates Article 14 is the question I have to decide. I have set out Mr Skeen’s evidence, which is to the effect that the Claimant is not necessarily in a worse off position than she would have been had Mr Roe become a pensioner member of the 2008 Scheme. However, I am prepared to assume for the purposes of this claim that she is indeed adversely affected by not being entitled to a survivor’s pension, and no-one suggested to the contrary.
The issues to be determined
The principal issues which fall for determination on this application are as follows:
Does the difference in treatment of which the Claimant complains come about because of one of the grounds proscribed, expressly or by inference, in Article 14; in other words, does it occur because of the Claimant’s ‘status’ as that term is interpreted in relation to Article 14 (‘status’) ?
If so, are the others put forward by the Claimant for comparison in an analogous situation (‘comparability’) ?
If so, is the difference in treatment objectively justifiable, ie, does it have a legitimate aim and bear a reasonable relationship of proportionality to that aim (‘justification’) ?
(i) Status and comparability
These two issues can be analysed together. That was Lord Wilson’s approach in Mathieson, supra: see at [19] – [23].
I take first the Claimant’s complaint that she is being treated differently on account of her unmarried status as compared with the married spouse of a member of the 1997 Scheme.
I accept that being unmarried qualifies as a status for the purposes of Article 14, and indeed the Defendants do not argue to the contrary. They were right not to do so. In Re G (Adoption: Unmarried Couple) [2009] 1 AC 173, Lord Hoffmann said at [8] that he had ‘…no difficulty with the concept of being unmarried as a status within the meaning of Article 14’. And in In Re McLaughlin , supra, [31], Lady Hale said that it is well-established in the case-law that not being married can be a status just as being married can be.
However, where I consider that the Claimant runs into difficulty is in her suggestion that she is, for these purposes, in a relevantly similar position to the married spouse of a member of the 1997 Scheme, in other words, that such a comparator is in an analogous position to herself.
The first question is whether being married, by itself, means that the suggested comparator cannot be properly compared to the Claimant in the way required by Article 14. As to this question, context is all important. Married and unmarried persons may, or may not, be comparable for the purposes of Article 14 depending on the context: In Re McLaughlin , supra, [26]; Smith, supra, [88].
Cases where married and unmarried persons have been held not to be in an analogous position for the purposes of Convention rights, at least so far as tax relief and state benefits are concerned, include Lindsay v United Kingdom, Applcn 11089/84, 1 January 1987, Comm Dec 1.11.86, DR 49, p181; Shackell v United Kingdom , Applcn 45851/99, 27 April 2000, p5; Burden v United Kingdom (2008) 47 EHRR 38, [63] – [65]; see also Van der Heijden v Netherlands (2013) 57 EHRR 13, [69].
In Lindsay’s case the married applicant sought to compare herself to an unmarried person in relation to a complaint about the UK’s income tax regime. The European Commissions rejected the comparison, holding that (p181):
“… these are not analogous situations. Though in some fields, the de facto relationship of cohabitees is now recognised, there still exist differences between married and unmarried couples, in particular, differences in legal status and legal effects. Marriage continues to be characterised by a corpus of rights and obligations which differentiate it markedly from the situation of a man and woman who cohabit.”
In Shackell’s case the applicant was the unmarried partner of a man who died in 1995. There were children of the partnership. She submitted a claim for widow’s benefits to the Benefits Agency. The claim was refused because she was not a widow. In her application to the European Court of Human Rights she claimed that the lack of provision for unmarried survivors was discriminatory. The application was rejected as inadmissible. The court accepted, at p5, that ‘there may well now be an increased social acceptance of stable personal relationships outside the traditional notion of marriage’ but continued: ‘However, marriage remains an institution which is widely accepted as conferring a particular status on those who enter it.’ Later the court said:
“The court considers that the promotion of marriage, by way of limited benefits for surviving spouses, cannot be said to exceed the margin of appreciation afforded to the respondent Government.”
I agree with the Claimant’s post-hearing submissions that, following the decision in In Re McLaughlin , supra, Shackell cannot now assist the Defendants, who placed some reliance upon it. For the reasons given by Lord Mance in his concurring judgment at [48] – [49] (with which Lady Hale, Lord Kerr and Lady Black agreed) it should be regarded as wrong or should not be followed, at least in cases involving children, in so far as it holds that married and unmarried persons are not in comparable positions in relation to a measure which adversely affects the children of an unmarried couple (eg, by denying a benefit intended for the support or maintenance of children).
In Burden, the applicants were unmarried sisters. They had lived together, in a stable, committed and mutually supportive relationship, all their lives; and for thirty-one years in a house built on land inherited from their parents. Under domestic law, when one of the sisters died, the survivor would be liable to pay inheritance tax on any assets received under the will. The tax rate was nil for the first £300 000 and 40% thereafter. However, property which passed from one spouse to the other in the case of same sex couples was exempt from tax. The applicants, obviously, were not able to marry or form a civil partnership, and they complained of a breach of Article 14 read with A1P1. The Grand Chamber of the European Court of Human Rights said at [63]-[65]:
“63 Moreover, the Grand Chamber notes that it has already held that marriage confers a special status on those who enter into it. The exercise of the right to marry is protected by Art.12 of the Convention and gives rise to social, personal and legal consequences. In Shackell , the Court found that the situations of married and unmarried heterosexual cohabiting couples were not analogous for the purposes of survivors' benefits, since “marriage remains an institution which is widely accepted as conferring a particular status on those who enter it”. The Grand Chamber considers that this view still holds true.
…
65 As with marriage, the Grand Chamber considers that the legal consequences of civil partnership under the 2004 Act, which couples expressly and deliberately decide to incur, set these types of relationship apart from other forms of cohabitation. Rather than the length or the supportive nature of the relationship, what is determinative is the existence of a public undertaking, carrying with it a body of rights and obligations of a contractual nature. Just as there can be no analogy between married and Civil Partnership Act couples, on one hand, and heterosexual or homosexual couples who choose to live together but not to become husband and wife or civil partners, on the other hand, the absence of such a legally binding agreement between the applicants renders their relationship of cohabitation, despite its long duration, fundamentally different to that of a married or civil partnership couple. This view is unaffected by the fact that, as noted at [26] above, Member States have adopted a variety of different rules of succession as between survivors of a marriage, civil partnership and those in a close family relationship and have similarly adopted different policies as regards the grant of inheritance tax exemptions to the various categories of survivor; states, in principle, remaining free to devise different rules in the field of taxation policy.”
Van der Heijden , supra, was not a property or benefits case but nonetheless is instructive in this context. T he applicant alleged a violation of Article 8 of the Convention, taken both alone and together with Article 14 of the Convention, in that an attempt had been made to compel her to give evidence in criminal proceedings against her long-standing companion, Mr A, with whom she was in a stable family relationship. The applicant’s case was that although she and her partner were not married and had not entered into a registered partnership ( geregistreerd partnerschap ), she and he had been cohabiting for eighteen years in a relationship out of which two children had been born, both of whom had been recognised by him. The applicant argued that on the basis of this relationship she should be regarded as entitled to the testimonial privilege ( verschoningsrecht ) afforded to suspects’ spouses and registered partners under the relevant provisions of Dutch law. At [69] the Grand Chamber rejected the suggested basis of comparison, saying:
“69. The Court does not accept the applicant’s suggestion that her relationship with Mr A., being in societal terms equal to a marriage or a registered partnership, should attract the same legal consequences as such formalised unions. States are entitled to set boundaries to the scope of testimonial privilege and to draw the line at marriage or registered partnerships. The legislature is entitled to confer a special status on marriage or registration and not to confer it on other de facto types of cohabitation. Marriage confers a special status on those who enter into it; the right to marry is protected by Article 12 of the Convention and gives rise to social, personal and legal consequences (see, mutatis mutandis , Burden v. the United Kingdom [GC], no. 13378/05 , § 63, ECHR 2008-...; and Şerife Yiğit v. Turkey [GC], no. 3976/05, § 72, ECHR 2010-...). Likewise, the legal consequences of a registered partnership set it apart from other forms of cohabitation. Rather than the length or the supportive nature of the relationship, what is determinative is the existence of a public undertaking, carrying with it a body of rights and obligations of a contractual nature. The absence of such a legally binding agreement between the applicant and Mr A. renders their relationship, however defined, fundamentally different from that of a married couple or a couple in a registered partnership (see Burden , cited above, § 65). The Court would add that, were it to hold otherwise, it would create a need either to assess the nature of unregistered non-marital relationships in a multitude of individual cases or to define the conditions for assimilating to a formalised union a relationship characterised precisely by the absence of formality.”
Accordingly, the Court rejected her complaint under Article 8 and Article 14 ([84]).
On the other hand, there are situations where married and unmarried persons have been regarded as being in comparable situations for the purposes of Article 14. There are a number of cases, both at the Strasbourg level and domestically, in which it has been held or agreed by the parties that unmarried couples are, in the particular context, in an analogous position to married couples: see eg In Re G (Adoption: Unmarried Couple) , supra, [132] – [133]; Sahin v Germany (2001) 36 EHRR 43, PM v United Kingdom [2005] STC 1566; In Re Brewster , supra; In Re McLaughlin, supra.
In the pensions context, unmarried and married people have in some circumstances been regarded as being in comparable positions. In Ratcliffe v Secretary of State for Defence [2009] ICR 762, the Court of Appeal decided that the claimant unmarried partner of a deceased Royal Navy officer was in an analogous position to a spouse for the purposes of Article 14 in conjunction with A1P1 in the context of a war pension. Hooper LJ said at [72] of his judgment, with which the other members of the Court agreed, that:
“… the decision whether a married and unmarried couple are in an analogous situation must be made in the light of the scheme under examination. By the end of 2003 unmarried couples were being treated substantially the same as married couples for the purposes of the occupational pension scheme and the government had announced that it would by 2005 be treating them the same for the purposes of the 2005 Order. This distinguishes the present case from the situation in Burden’s case 47 EHRR 38. Thus in 2004 it would, in my view, be wrong to say that they were not, in the context of armed forces benefits, in an analogous position for the purposes of article 14.”
It is clear that this aspect of the decision in Ratcliffe turned in part on the special facts in relation to armed forces’ pensions. As Hooper LJ explained at [30], the evidence showed that on 20 March 2003, at the time of the start of the invasion of Iraq and following political pressure, the Government announced in the House of Lords that where a member of the armed forces after that date died in service or by reason of service related to conflict, ex gratia payments equivalent to the benefits payable to spouses under the occupational pension scheme would be made to unmarried partners in a substantial relationship. On 15 September 2003 the Government announced that the requirement that the service was related to conflict no longer applied from that date. The necessary detailed changes were made in a Defence Council instruction issued in January 2004. It was for that reason that the Court was able to conclude that by the end of 2003 (prior to the time when Ms Ratcliffe’s claim arose) unmarried couples were being treated substantially the same as married couples for the purposes of the occupational pension scheme. The court went on, however, to hold that the difference in treatment was justified and that there was no violation of Article 14 read with A1P1 ([89]).
To the same effect is In Re Morrison’s Application [2010] NIQB 51, which also concerned death benefits under an occupational scheme. There, the High Court of Northern Ireland relied upon Ratcliffe and held (at [26] – [36]), in the context of the particular regulations in that case which, like those in Ratcliffe , sought to equate certain cohabitees with spouses/civil partners, that it would be wrong to say that the applicant was not, at the relevant time, in an analogous position to a spouse or civil partner.
In Smith, supra, a man lived with the claimant in the same household as husband and wife for 11 years until his death as a result of the admitted negligence of the defendant NHS trusts. The claimant brought proceedings against the NHS trusts for dependency damages under section 1 of the Fatal Accidents Act 1976, as substituted and amended, which made provision for claims by cohabitees of more than two years. After the claim was compromised the NHS Trusts played no further part in the proceedings. The Secretary of State for Justice was joined as the third defendant to enable a claim for bereavement damages to be pursued, since the express terms of section 1A(2)(a) of the 1976 Act, as inserted and amended, did not extend claims for bereavement damages to cohabitees. The claimant argued this breached Article 14 read with A1P1 of the Convention. At [90]-[91] the court held that unmarried and married people were in analogous positions so far as bereavement damages were concerned, because the grief and sense of loss which the damages were intended to compensate was the same in both situations:
“ 90 I agree with the judge that, in the context of bereavement damages under section 1A of the FAA, the situation of someone like Ms Smith, who was in a stable and long-term relationship in every respect equal to a marriage in terms of love, loyalty and commitment, is sufficiently analogous to that of a surviving spouse or civil partner to require discrimination to be justified in order to avoid infringement of article 14 in conjunction with article 8. In the context of this particular scheme, it is not the special legal status and legal consequences of marriage and civil partnership that are material, in the sense of providing a rational distinction with other people and relationships: cf, for example, Burden’s case, in which the ECtHR rejected the complaint of two unmarried sisters, who had lived together all their lives, that the liability to inheritance tax payable on the death of one of them, which would not be faced by the survivor of a marriage or civil partnership, would violate their rights under article 14 read with A1P1. Rather, it is the intimacy of a stable and long-term personal relationship, whose fracture due to death caused by another’s tortious conduct will give rise to grief which ought to be recognised by an award of bereavement damages, and which is equally and analogously present in relationships involving married couples and civil partners and unmarried and unpartnered cohabitees.
91 In making that analogy, it is plainly material that Parliament has treated 2 years + cohabitees as being in a stable and long-term relationship comparable to that of spouses and civil partners for the purposes of dependency damages, and that neither in Parliament nor in any evidence before the court has any member of the Government provided any justification for the different treatment of 2 years + cohabitees under section 1A. As the judge said [2017] PIQR P4, paras 34 and 109 the Secretary of State’s position has a degree of incoherence.”
I do not consider that the Claimant can derive significant support from this decision. That is for three main reasons. First, the right in issue was Article 8, as in Re G (Adoption: Unmarried Couple) , supra, which concerned a statutory provision that an adoption order could only be made to a couple if that couple were married. It was held that there was a breach of Article 8 taken with Article 14. The House of Lords held that being unmarried was a status for the purposes of Article 14 and that restrictions on adoption engaged Article 8. It said that the exclusion of all unmarried couples was irrational and contradicted the fundamental statutory principle of adoption law that the most important consideration was the best interests of the child. Lord Mance noted the decisions in Burden and Shackell and stated at [133]:
“The Shackell and Burden cases were decisions in the context of taxation and social benefits, where the right to which the alleged discrimination related was the right to protection of property under article 1 of the First Protocol to the Convention. The present appeal arises in the different context of the right to respect for private and family life, in relation to which it is clear that distinctions between married and unmarried persons may be unjustifiably discriminatory.”
Second, the court plainly regarded it as significant that Parliament had legislated to provide for one type of damages for unmarried partners, but not for bereavement damages, without there being any rational justification for the distinction. I accept the Secretary of State’s submission that such reasoning is not applicable to the position of Pre-2008 Cohabitees and Pre-2008 Spouses. The evidence that I have set out shows that their positions were kept consistently different, in accordance with Government policy that public sector pension schemes could extend the provision of survivor benefits providing they were wanted by the membership and the costs of those benefits were met by the membership, and that those who have left the LGPS should not be eligible for new benefits because they have not paid for them. I do not accept Mr Giffin’s submission that there is any incoherence in the Government extending provision to unmarried partners in the 2008 Scheme. On the contrary, the evidence is quite clear that the Government introduced this because members of the LGPS indicated that they wanted it during detailed consultation processes and were prepared to pay for it, and that no-one suggested it should be extended to pensioner members’ partners.
Third, as Lady Hale explained in In Re McLaughlin, supra , [3], [26], the bereaved unmarried Ms McLaughlin argued at first instance that the denial to her of a bereavement payment and a widowed parents’ allowance violated Article 14 read with Article 8 and/or A1P1. The judge dismissed her claim in relation to the bereavement payment on the grounds that she was not in a relevantly similar position as a married person so far as that benefit was concerned because of the absence of a ‘public contract’ in the form of marriage. However, he allowed her claim in respect of the widowed parents’ allowance (see [2016] NIQB 11). The Court of Appeal allowed an appeal by the respondent Department, holding that the denial of the latter benefit did not violate the Convention. There was no appeal by Ms McLaughlin in relation to the judge’s decision on the bereavement payment. Ms McLaughlin’s appeal to the Supreme Court was therefore solely in relation to the widowed parents’ allowance, and her appeal was allowed. However, at [27] of her judgment Lady Hale approved the judge’s analysis of the question whether married and unmarried people are in an analogous position for the purposes of the two benefits. To that extent, I do not consider too much weight can be placed upon Smith , supra.
Nor does PM v United Kingdom (2006) 42 EHRR 45 assist the Claimant, notwithstanding it was an Article 14/A1P1 case. There, the applicant complained that he was not allowed to deduct from his taxable income the payments he made for the maintenance of his daughter solely because he had not been married to the mother. The court said, at [27]–[29]:
“27. … This applicant differs from a married father only as regards the issue of marital status and may, for the purposes of this application, claim to be in an relevantly similar position.
28. The justification for the difference in treatment relied on by the Government is the special regime of marriage which confers specific rights and obligations on those who choose to join it. The court recalls that it has in some cases found that differences in treatment on the basis of marital status has had objective and reasonable justification … It may be noted however that as a general rule unmarried fathers, who have established family life with their children, can claim equal rights of contact and custody with married fathers … In the present case, the applicant has been acknowledged as the father and has acted in that role. Given that he has financial obligations towards his daughter, which he has duly fulfilled, the court perceives no reason for treating him differently from a married father, now divorced and separated from the mother, as regards the tax deductibility of those payments. The purpose of the tax deductions was purportedly to render it easier for married fathers to support a new family; it is not readily apparent why unmarried fathers, who undertook similar new relationships, would not have similar financial commitments equally requiring relief.
29. The court concludes therefore that there has been a violation of article 14 of the Convention in conjunction with article 1 of the First Protocol in this case.”
Hence, as I read the court’s decision, it was indicating that in the particular context of relationships with their children, unmarried and married fathers could be regarded as being in the relevantly similar position because as a general rule unmarried fathers, who have established family life with their children, can claim equal rights of contact and custody with married fathers. That reasoning is not inconsistent with the reasoning in Lindsay’s case, for example, that marriage continues to be characterised by a corpus of rights and obligations which differentiate it markedly from the situation of a man and woman who cohabit.
Thus, there are cases on both sides of the line. Where does the present case fall ? In my judgment, the unmarried Claimant is not to be regarded as being in an analogous situation to a spouse in the context with which I am concerned. As Lady Hale said in In Re McLaughlin , supra, [26], it is important precisely to delineate that context. The comparison here is not simply between spouses and unmarried persons. It is between a spouse whose deceased husband was a member of the 1997 Scheme and an unmarried person whose unmarried partner was a member of the 1997 Scheme . The italicised words are crucially important in identifying the right context. The Claimant’s complaint is that a wife whose husband had precisely the same length of service from 1979 – 2003 as Mr Roe, and who died on the same day in 2016 as he did, would get a pension, whereas she does not. But the answer is that is because the deceased husband paid for that benefit for his wife, whereas Mr Roe did not pay for that benefit for the Claimant. I have explained by reference to the evidence that when a pension scheme is being designed, the package of benefits to be provided are costed and are reflected (in part) in the level of contributions which members make to the Scheme. Hence, the comparative spouse is merely receiving what her husband paid for through his contributions to the Scheme. The Claimant is not in a relevantly comparative position as the comparator wife for the simple reason that Mr Roe did not pay for such a benefit to be afforded to her, because that was not part of the 1997 Scheme and so was not reflected in the contributions which Mr Roe paid. Someone who does not receive a benefit which has not been paid for in their case is not in a comparable position to someone who does receive the benefit which has been paid for in their case.
The issue of comparability in this case is fundamentally different to the situation in In Re Brewster, supra, where comparability of married and unmarried persons was conceded ([47]). There, Ms Brewster was denied a benefit under a Scheme which had been costed to provide survivors’ benefits for nominated unmarried partners. Her partner Mr McMullan had paid for her to have that benefit via his contributions. But she was denied it simply because she had not been nominated whereas she would have received it had she been married to Mr McMullan. There was hence true comparability.
I turn to the alternative comparator put forward by the Claimant, namely post-2008 cohabitees.
Again, the first question is whether for these purposes the Claimant has a status within Article 14. The suggested status in her Skeleton Argument ([20(i)]) is ‘the cohabiting partner of someone who was an active LGPS member at a particular date, ie, 1 April 2008’. I think it is better expressed as the cohabiting partner of a pensioner member of the 1997 Scheme, the suggested comparator being the cohabiting partner of an active member of the 2008 Scheme.
The Secretary of State argues that the comparison that the Claimant draws between a Pre-2008 Cohabitee and a Post-2008 Cohabitee is not one based upon her status as a cohabitee at all (because her comparator has the same status) and says the point of distinction is simply one of time. The Council makes the same submission.
I do not accept this submission. I accept for the purposes of this case that being the cohabitee of a pensioner member of the 1997 Scheme qualifies as a status for the purposes of Article 14. As I have set out, that term has been interpreted broadly to include many categories which might not readily be thought of as a status: RJM, supra , [5].
But, again, I consider that the Claimant’s argument breaks down because a Post-2008 Cohabitee is not in a relevantly similar situation to the Claimant, essentially for same reasons that Pre-2008 Spouses are not comparable to her position. Post-2008 Cohabitees are entitled to a survivor’s pension because it was a costed benefit in the 2008 Scheme which their partners paid for via their contributions. In her Skeleton Argument at [79] the Claimant disputes that this benefit was costed into the 2008 Scheme, but the evidence shows that it was. Mr Bayliss explains that although the target pension cost of 19.4% of payroll did not include the costs of this benefit, it was allowed for via the costs savings made from the abolition of the Rule of 85 (see his witness statement at [80] – [81] and see also Ms Broadfield’s statement at [119]).
Mr Giffin submitted that it was unfair and anomalous that the unmarried partner of a person who was a member of the 2008 Scheme for only a few days after 1 April 2008 (for example) is eligible for pension based on service going back to April 1988, whereas the Claimant is eligible for nothing. The Claimant’s Skeleton Argument makes clear that she does not complain about such a person getting additional benefits in respect of a period when he was an employee and Mr Roe was not, but says that the problem lies in respect of pre-April 2008 service, when such a person would get full entitlement to survivors’ benefits, but the Claimant would not.
In my judgment the answer to this point is three-fold. First, in relation to pensions, it is wrong to focus on the position of just one person, here, the comparator partner whose employment lasts only days after 1 April 2008. In her evidence Ms Broadfield accepts that not every active member will make contributions which in their own case pay for past service, and she also provides the example of an active member who retired a short time after the 2008 Scheme took effect. However, she says that such individual variations are inevitable in any defined benefit pension scheme. Her evidence is that the scheme and its contribution levels will have been set on the basis of assumptions and actuarial calculations which mean that the active membership as a whole will have both taken the benefit of the change (or at least be eligible for it) and ensured that it is paid for. As she says in her witness statement at [15]:
“… it is unhelpful to look at the terms of a defined benefit pension scheme from the point of view of one particular member. The scheme has to be designed and costed by reference to its coherence as a whole. The funding of the scheme has been designed to fund those benefits accrued at the point at which the scheme is to commence, and estimated to accrue over the lifetime of the scheme on the basis of the best available assumptions at the time.”
If I may respectfully say so, this is the point which Lord Hoffmann made in Carson , supra, [21] when he said that the proper approach under Article 14 to a scheme (there, the UK’s social security scheme) was to regard it as a corpus of rights and obligations of which it would be artificial to isolate one specific aspect (quoting Van der Mussele v Belgium (1984) 6 EHRR 163, [46]).
The second answer is that, as I have already explained, the 2008 Scheme was costed to include benefits related to service going back as far as April 1988. Hence, active members of the Scheme as at 1 April 2008 who have gone on to take a pension based on past service (perhaps as far back as 1988) have not thereby received a gratuitous benefit but have simply received what had been allowed for in the costs envelope of the 2008 Scheme (including their own contributions, those of their employers, and the savings made from the abolition of the Rule of 85, the benefits of the last of which they otherwise would have received.)
The third answer to Mr Giffin’s point is that, to the extent that there is any unfairness in the situation he cites, which I do not accept, then such is an inevitable product of having bright line rules in the application of the LGPS. A hard case which falls on the wrong side of a bright line rule does not invalidate that rule if judged in the round it is beneficial: R (Animal Defenders International) v Secretary of State for Culture, Media and Sport [2008] AC 1312, [33], per Lord Bingham.
Overall, the reason why a Post-2008 Cohabitee is not in a relevantly similar position to a Pre-2008 Cohabitee is because they are subject to different legal regimes. The evidence (in particular from Mr Skeen) shows that although in general terms the 2008 Scheme was more generous than the 1997 Scheme, direct comparison between the two is not straightforward and in some respects Mr Roe received benefits from the 1997 Scheme that he would not have received under the 2008 Scheme. But the more general reason why the Claimant’s case fails on this aspect is because a person who is subject to one legal regime at a particular point in time is not in an analogous situation with a person subject to a different legal regime at a later point in time. In Minter v United Kingdom , App 62964/14, the applicant had been sentenced to an extended sentence of imprisonment. He complained that there had been a breach of Article 14 read with Article 8 because after he had been sentenced, the sentencing regime had changed so that for persons sentenced after 14 July 2008 an extended sentence was not available for the offences for which he had been given an extended sentence. At [67] the European Court of Human Rights rejected this argument, saying that in Massey v United Kingdom (dec) No 14399/02:
“… the Court considered that no discrimination was disclosed by legislative measures being prospective only or by a particular date being chosen for the commencement of a new legislative regime … In this regard, it has noted that the use of a cut-off date creating a difference in treatment is an inevitable consequence of introducing new systems which replace previous and outdated schemes.”
I therefore reject the suggested comparison between the Claimant and a Post-2008 Cohabitee. They are not in relevantly similar situations for the reasons I have given.
I turn to the Claimant’s third suggested basis for comparison, namely her age. This is accepted by the Defendants to be a status for the purposes of Article 14. She submits that she is the victim of indirect discrimination on the basis of age. She argues that the partner of a member of the Scheme who remains in service after 1 April 2008, and who therefore qualifies for a survivor’s pension for his unmarried partner, is likely to be younger than she is. She argues that on average most co-habiting couples are likely to be a similar age to one another and the older a member of the LGPS a person is, the more likely it is that s/he will have permanently ceased to belong to LGPS before 1 April 2008 and not qualify for a survivor’s pension for their partner.
It seems to me that this argument is simply another way of arguing that the Claimant is similarly situated to a Post-2008 Cohabitee and is entitled to be treated in the same way, and I conclude the Council is essentially correct to say that the age claim is parasitic on that aspect of the Claimant’s case. Even assuming in the Claimant’s favour the factual premise of the argument (which the Defendants do not), I do not accept that there is here indirect discrimination on the basis of age. The 1997 Scheme does not define pension entitlements by reference to the age of a member’s spouse or civil partner. The Claimant has not been refused a pension on the basis of her age, but is not entitled to it because that was not a benefit which was costed or paid for under the 1997 Scheme. In contrast, it is payable in relation to members of the 2008 Scheme as it has been costed into the benefits which are payable under it.
In my judgment the answer to the Claimant’s age claim is that, in reality, what she is complaining about is a difference in treatment arising from different rules applying to different schemes at different times; however, that is not an age-based reason for the different treatment complained of. In Ackermann and Fuhrmann v Germany (App No 71477/01, 8 September 2005) the European Court said:
“Insofar as the applicants further complained about discrimination on ground of age, alleging that earlier generations of pensioners received considerably higher pensions than they themselves would on reaching pension age, the Court notes that the applicants have not established that their own situation is comparable to that of earlier pensioners. In this respect, it has to be taken into account that the State must be in a position to adapt the pension system to the change of socio-economic circumstances. Accordingly, the applicant cannot claim equal treatment ‘in time’”.
In R (Gurung) v Secretary of State for Defence [2008] EWHC 1496 (Admin) the claimants were Gurkhas challenging army pension arrangements. Gurkha pensions from 1 July 1997 were paid at a rate equivalent to that of other army soldiers; Gurkha pensions under the Gurkha Pension Scheme established in 1949 were paid at a much less favourable rate. The claimants asserted that this amounted to indirect age discrimination, on the basis that (a) Gurkhas with more years of service were disadvantaged; and (b) those Gurkhas would on the whole be older. Ouseley J was not convinced that such an argument gave the claimants any ‘other status’ for the purposes of Article 14 at all; but if it did, he considered that the matter was correctly approached on the basis that the Secretary of State had a wide margin of discretion; that the selected date of 1 July 1997 was not irrational; and that the challenge failed accordingly.
He also observed that this type of differentiation between one group and another, based upon a particular cut-off date, was inevitable whenever there was a transition from one welfare scheme to another; and this could not form a strong basis for challenging a decision on social and economic policy:
“74. There may be differences of view about whether “age” is a suspect ground for discrimination, requiring a more intense scrutiny, or whether “old age”, which is not quite the same, can be. But the grounds of differentiation here, not wholly characterised as those of age, are not suspect grounds. The grounds of difference do not arise because someone is above or below a particular age, but because the introduction of changes which are not directly age related are defined by dates, and years of service. The drawing of lines, by reference to dates, around schemes which help some and not others is an inevitable part of many legislative or policy changes; this is the more so where a past disadvantage or even wrong is being remedied retrospectively. Of course, this means that either the older or the younger will be affected; the date itself will import an indirect differentiation on age grounds. But that is a weak starting point for an assertion of indirect discrimination on age grounds. In any event, if there is a rational basis for the selection of the date as at which the changes are made, that disposes of the Article 14 challenge.”
This, it seems to me, is largely dispositive of the Claimant’s case on age. As the Defendants rightly point out, by their nature pensions are linked to age. The application of an indirect age discrimination claim to a pension scheme context can be problematic because the membership and benefits of all pension schemes are determined by rules based directly on, or otherwise related to, age. They are right to submit that pension schemes are necessarily linked to age and age-related concepts, such as length of service. If the Claimant were correct then it seems to me that to introduce a new pension scheme with improved benefits would nearly always be liable to give rise to age-based claims, because it will generally be the case that older people are more likely to be members of the older and less advantageous pension scheme. But it is clear that pension schemes cannot stay static and have to be updated from time to time to reflect macro-economic and societal and demographic changes, among other reasons. That suggests that she cannot be correct. But that is not to say that pensions can never give rise to age-based claims: see eg McCloud v Lord Chancellor and Secretary of State for Justice, Appeal [2018] ICR 1039, where age discrimination was conceded in relation to the New Judicial Pension Scheme, but the Lord Chancellor argued it was objectively justifiable. There, however, the rules expressly defined benefits by reference to age, with older judges being eligible for more generous benefits than younger judges. It was not a case where a new pension scheme had been introduced with effect from a certain date.
In JT , supra, the claimant complained that the ‘same-roof’ rule which excluded her from receiving compensation for sexual offences committed before a certain date was age-discriminatory because victims of such offences were likely to be older ([76]). The Court of Appeal rejected that argument and, in my judgment, similar reasoning applies in the present case.
For all of these reasons, I reject the suggested comparators put forward by the Claimant as being in a relevantly similar position. They are not, for the reasons that I have given. And I also reject the Claimant’s case on age.
However, I recognise that the analysis carried out by the European Court of Human Rights usually elides the comparability of the situations and focuses on the question whether differential treatment is justified. This reflects the fact that an assessment of whether situations are ‘relevantly’ similar is generally linked to the aims of the measure in question. Thus, in accordance with this approach, I go on next to consider the question of justification.
(ii) Justification
The question is whether the decision to introduce a survivor’s pension for unmarried co-habiting partners of active members of the LGPS as at 1 April 2008, based upon future service and past service back to 5 April 1988, but not to extend such a benefit to the unmarried cohabitee pensioner or deferred members of the LGPS, was a proportionate one, assessed according to the four stage Bank Mellat test (see above). The Claimant does not assert that the 1997 Scheme should have provided for equality for spouses and cohabitees from its inception (a claim which would be prohibited by the lack of retrospective application of the Human Rights Act 1998). Rather, she claims that the differential treatment in the 1997 Scheme could not properly continue once the 2008 Scheme had been introduced.
It is right to point out that the burden of showing justification falls upon the Defendants, and in particular the Secretary of State. But, equally, it needs to be emphasised that I am here concerned with pensions and with suggested grounds of discrimination which are not ‘suspect’, in the sense of requiring especially strong justification. Marital status and age (as it arises in the context of pension schemes at least) are not suspect grounds for the purposes of Article 14, and I reject the Claimant’s submissions to the contrary. Marital status is a matter of choice rather than a personal characteristic of a type which is one of the core grounds in Article 14. Although age is a personal characteristic, it is a characteristic that is common to us all, and so, at least in this context, it is simply an ‘aspect of ordinariness’: Carson, [60]. I agree with what Ouseley J said about age and pensions in Guring , supra, [74].
The significance of this, and the relevance of the pensions context, to the question of justification is to be found in Ratcliffe , supra, [89], which as I have already set out was a case involving an unmarried claimant:
“89 At the end of the day this case, in my view, falls squarely within the now well-established principle that where alleged discrimination in the field of pensions is based on non-suspect grounds, courts will be very reluctant to find that the discrimination is not justified. Whatever the position today, historically the distinction in the war pension scheme between married and unmarried partners and between unmarried partners who fell within the very narrow criteria for a pension and other unmarried partners was justified. In 2003 the government recognised that the distinction was no longer justified, altered the occupational pension scheme prospectively and announced its intention to make changes to the war pension scheme from some time in the future but also prospectively. The decision, from what point in time unmarried partners are put in an analogous position to spouses in the field of pensions, is a decision for the government and is a decision with which the courts will not normally interfere. In the words of Laws LJ in Carson’s case , para 73 (referred to at para 51 above):
‘In the field of what may be called macro-economic policy, certainly including the distribution of public funds upon retirement pensions, the decision-making power of the elected arms of government is all but at its greatest, and the constraining role of the courts, absent a florid violation by government of established legal principles, is correspondingly modest.’”
Hence, whilst I must carry out the proportionality analysis in the way that I have outlined, I must do so whilst affording the decision maker a wide margin of deference, in other words, being faithful to the test of whether what the decision maker has done is manifestly without reasonable foundation.
As I have already noted, whilst accepting that generally this was the right approach, Mr Giffin suggested that this was not the correct test on the facts of the present case. He suggested a more intensive standard of review. He submitted, firstly, that the justifications put forward in the evidence are ex post facto justifications and therefore should be given less weight. He says that there is no evidence that the Secretary of State ever gave serious contemporaneous consideration to whether the discrimination between spouses and stable cohabitees was justified, on grounds of social policy or otherwise. Second, he submitted that discrimination on grounds of marital status should be treated as a suspect ground calling for close scrutiny from the court rather than the opposite.
I do not accept these submissions. As to the first point, the evidence shows that the Secretary of State did give detailed consideration at the time to whether spouses and cohabitees should be treated in the same way, and the circumstances in which this should happen. That is the plain thrust of Ms Broadfield’s evidence. The Government was fully aware of the difference in treatment in the 1997 Scheme, and from at least 1998 onwards (when the Green Paper was published) indicated its willingness to bring forward proposals to extend coverage to cohabiting partners if that was what the membership of the LGPS wanted and were willing to pay for. As to the second point, I consider it to be contrary to authority and in particular to Turley , supra, [22], per Underhill LJ, and Ratcliffe at [89] per Hooper LJ, which I have already quoted. In the former case, Underhill LJ said:
“22. Mr Lask also pointed out that the status which was the basis of the difference in treatment – being a “common law” as opposed to a legal spouse – was not a “suspect ground” in the sense explained by Lord Hoffmann in the well-known passage of his speech in R (Carson) v Secretary of State for Work and Pensions [2006] 1 AC 173 (see paras. 15-16 (pp. 182-3)), and that the court could and should apply a less rigorous standard of review. He referred us to a similar passage in the judgment of the ECtHR in Bah [ v UK (2012) 54 EHRR 21], where it said (again, at para. 47 (p. 790)):
“The Court recalls that the nature of the status upon which differential treatment is based weighs heavily in determining the scope of the margin of appreciation to be accorded to Contracting States. As observed above … immigration status is not an inherent or immutable personal characteristic such as sex or race, but is subject to an element of choice. In the applicant's case, while she entered the United Kingdom as an asylum seeker, she was not granted refugee status. She cannot therefore be described as a person who was present in a Contracting State because, as a refugee, she could not return to her country of origin. Furthermore, she subsequently chose to have her son join her in the United Kingdom. Given the element of choice involved in immigration status, therefore, while differential treatment based on this ground must still be objectively and reasonably justifiable, the justification required will not be as weighty as in the case of a distinction based, for example, on nationality.”
I turn, then, to the four-stage Bank Mellat test for proportionality in order to determine whether the decision not to introduce survivors’ pensions to unmarried co-habiting partners such as the Claimant was objectively justified.
(1) Is the objective(s) of the measure sufficiently important to justify the limitation of the protected right ?
The objective of the measure in question, namely the changes made by the 2008 Scheme in relation to cohabitees, was to introduce survivors’ pensions for cohabitees of active scheme members so as to place them on the same footing as married couples. This was identified in In Re Brewster, supra, [34], as the purpose behind the changes in Northern Ireland in 2009 which introduced survivors’ cohabitees pensions there, and the same must be true for England and Wales. A further objective, as Ms Broadfield explains, was to give effect to members’ wishes as revealed in the various consultations which took place before 2008. Allied to these objectives was the overarching desire of the Government that this change should take place in the context of an LGPS which implemented desired benefit structures at a stable and affordable cost; which did not impose the costs of the LGPS upon the national or local taxpayer, by remaining within a fixed costs envelope; but which provided a package of benefits that was attractive, and so aid recruitment and retention of high quality staff.
The specific aspect of the measure which the Claimant objects to, namely the lack of retrospective application of this benefit to pensioner members of the 1997 Scheme by comparison to spouses and cohabitees of active members of the 2008 Scheme, seems to me on the evidence to have had a number of objectives. Firstly, there was the need to avoid imposing unexpected pension liabilities upon administering authorities by introducing benefits for pensioner members which had not been paid for via their membership of their Scheme, and which might jeopardise the Scheme’s viability. As the Government put it in a 2014 document on survivor benefits in occupational pension schemes, reiterating existing policy (quoted in Ms Broadfield’s statement at [30]):
“2.26 When making changes to scheme rules, or to legislation which will affect the value of the members’ pension, the general position has therefore been to make these changes prospectively … This avoids the risk to scheme funding that would be created if schemes were required to meet legal obligations to pay benefits which did not exist at the time that those benefits were accrued.”
Second, there was the need to avoid unfairness to existing Scheme members, whose contributions going forward would have to pay (at least in part) for the benefit to pensioner members. As Ms Broadfield explains at [28] – [29] of her witness statement, to extend the benefit retrospectively would in effect have required active members to subsidise pensioner members who would have made no contribution to that benefit, in effect reducing the deferred pay of active members through higher contributions whilst at the same time increasing the deferred pay of pensioner members. This, says Ms Broadfield, would not have assisted recruitment, retention or motivation. The way it is put by the Secretary of State, which I accept, is that one objective was managing and reflecting inter-generational fairness through the provisions of benefits to those who, taking the Scheme as a whole, would pay for them through their contributions going forward.
This, it seems to me, is the central and core justification running throughout this case. The short point is this: spouses in the 1997 Scheme and cohabitees of 2008 Scheme members are entitled to a pension because it was costed into the relevant Schemes and paid for. Mr Roe did not pay for the Claimant to receive that benefit, nor was it otherwise costed into the 1997 Scheme. That is the reason she in the position that she is in. Mr Roe got what he was entitled to under the 1997 Scheme, and the Claimant, by virtue of her financial interdependence with Mr Roe, benefitted accordingly.
Third, and linked to the second objective, was the need to avoid a ‘windfall’ for pensioner members. As it was put in 2003 in a memorandum by the Treasury to the Chief Secretary ([12]):
“The guiding principle is that members of public service pension schemes should receive the benefits which they are promised during their service in return for that service and their contributions to the scheme. Explicitly, the scheme does not promise that years after members have moved to another employer or retired, their benefits will be enhanced to reflect what is judged reasonable to recruit and retain future generations of public servants in different social climates.”
To this might be added especially not where, as in Mr Roe’s specific case, he had received advantages and benefits under the 1997 Scheme which were not available under the 2008 Scheme.
In my judgment, taken together, these objectives were sufficiently important to justify the limitation on the Claimant’s right to receive a survivor’s pension to the extent that that right is engaged. In making the change in the way that it did the Government ensured the stability of the LGPS and ensured that active members (both current and prospective) did not have imposed upon them a levy to fund a benefit for pensioner members. In seeking to satisfy these objectives it cannot be said that the Government’s choice was manifestly without reasonable foundation.
Mr Giffin submitted that there was no objection of principle to retrospection, and argued that that is demonstrated by the fact that pre-2008 service of active members was taken into account in the calculation of benefits. But in my judgment that argument confuses the two senses in which the term ‘retrospection’ can be used in the pensions context. Ms Broadfield explains the difference in her witness statement, as I have set out.
Next, Mr Giffin submits that there cannot have been any proper objection in principle to retrospection (in the true sense) because pension benefits were retrospectively applied to civil partners and same-sex spouses. But the evidence shows that these were exceptional steps which were taken to correct the historic injustice that people of the same sex were simply not able to make a solemn and public declaration of their love and fidelity, and so take upon themselves the contractual responsibilities inherent in marriage, however much they wanted to, because the law then simply did not permit it. The Government applied retrospection in these cases in the knowledge that it was acting contrary to long-standing policy because of the very special circumstances. That was, it seems to me, pre-eminently a matter for Parliament to determine bringing together, as it did, a complex range of socio-economic and other factors. The fact the step was taken did not mark an abandonment of the Government’s underlying long-standing policy, nor was it a recognition that somehow the policy could no longer be justified.
(2) Is the measure rationally connected to the objective(s) ?
The measure in question was rationally connected to the achievement of these objectives. The non-retrospective extension of cohabitees’ pensions avoided the need for the LGPS to meet many millions of pounds of liabilities which would otherwise have been inccurred; it avoided existing members being asked to pay for a benefit for pensioner members at direct expense to themselves; and it enabled the Government in 2008 to introduce a revised LGPS which it believed achieved the various policy goals it wanted to achieve. Again, the question for me is whether it was manifestly without reasonable foundation to conclude that the measure is rationally connected to the objectives that I have identified, and I do not consider that that is the case.
It is of note that it has consistently been Government policy that new benefits in public sector pension schemes should only be applied prospectively, save in wholly exceptional circumstances such as the introduction of same-sex marriages and civil partnerships, when the change was made retrospective to remedy perceived historic injustice. There is nothing in the evidence to show that this policy goal has ever been questioned. Also, and furthermore, the evidence shows, that when the 2008 Scheme was being consulted upon, no consultee suggested that new benefits should be extended retrospectively to pensioner members.
In his submissions the Secretary of State relied heavily on the decision in Cockburn, supra, where (as I have already noted) Supperstone J considered an Article 14/A1P1 challenge to provisions in the NHS Pension Scheme which limited a widower’s survivors pension to service after 5 April 1988, whereas a widow in the same position is entitled to a pension based on her husband’s full length of service. The difference in treatment between men and women was held to be objectively justified. The case involved issues, such as gender equality and the historically disadvantaged position of widows as compared with widowers, which do not directly arise in the case before me. However, the judgment is nonetheless of relevance because at several points the judge noted the Government’s policy as expressed in the evidence before him that improvements to public service pension schemes are generally not given retrospective effect: see eg [71]. The judge also said that it was relevant to the question of justification for the Secretary of State to have taken into account the lack of support among NHS Pension Scheme members for proposed equalisation of benefits if they had to bear the costs. The same consideration applies here.
(3) Could a less intrusive measure have been used without unacceptably compromising the achievement of the objective ?
The Claimant has not suggested that a less intrusive measure short of granting her a survivor’s pension based on Mr Roe’s service after 5 April 1988, which she would have received had he been in post on 1 April 2008, could have been applied without compromising the achievement of the objectives that I have set out. That is plainly right. There is in this context no room for a half-way house measure. The Claimant contends for nothing less than that which the co-habiting spouse of a member of the 2008 Scheme receives.
I conclude, again, that it was not manifestly without reasonable foundation for the Government to make the changes that it did in the way that it did.
(4) Balancing the severity of the measure's effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, does the former outweigh the latter ?
Another way of posing this question (suggested by Leggatt LJ in JT , supra, at [83]) is whether the impact of the right’s infringement is disproportionate to the likely benefit of the impugned measure.
In my judgment, whether I decide this matter for myself, or do so by reference to the ‘manifestly without reasonable foundation’ test (see above for discussion of the two contrasting approaches in this context), the answer is the same, and that is that the impact on the Claimant is not disproportionate to the benefits of the absence of any survivor’s pension in her case and that of others similarly situated.
True it is that the Claimant is not entitled to a pension under the 2008 Scheme. But the 2008 Scheme must be considered as a whole. As I have already mentioned, and as Mr Skeen describes in detail, Mr Roe and the Claimant, through their financial interdependence, received advantages and benefits under the 1997 Scheme which they would not have received had Mr Roe been made redundant as a member of the 2008 Scheme. This is not therefore a case where matters are all one way. Looking at the 1997 Scheme as a whole, in comparison with the 1997 Scheme taken as a whole, there is good and bad from the Claimant’s perspective.
On the other hand, the broader impact on the public purse, and on other LGPS members, if the Claimant were to receive the benefit she contends for, would be very significant. I have set out the figures earlier in the judgment and on any view they are substantial, as I think Mr Giffin in the end accepted. However, Mr Giffin said that cost was not relevant. As he puts it in his Skeleton Argument ([92]), if a benefit is to be limited to save costs then it must be limited in a non-discriminatory way, and he relies on what was said in R (Coll) v Secretary of State for the Justice [2017] 1 WLR 2093, [40]. I reject this submission. I am concerned here with the choices that have been made in a complex policy area where many different factors are in play and specifically with the 2008 Scheme which, as I have discussed, followed a widespread consultation process during which members of the Scheme did not express a wish to pay for the benefit which the Claimant now claims. Against that background, the choice faced by the Executive was whether to fund that benefit by bringing the costs of doing so to bear on the broader public purse and/or to load the cost on to existing and future members of the Scheme through their contributions but against their expressed wishes. By enacting the 2008 Scheme, the Executive chose not to pursue either of these options. I have set out Ms Broadfield’s evidence that to have required existing and future Scheme members to have funded the benefit would have had the practical effect of reducing their pay in order to benefit pensioners. There are strong reasons for concluding that that would not have been fair; be that as it may, in my judgment the choice made by the Secretary of State and enacted in the legislation cannot be characterised as a policy choice that was manifestly without foundation. So far as requiring the public purse (in reality, Council Tax payers) to fund the benefit, the cost of doing so is relevant to justification, and there is ample authority to support that proposition. I have already quoted Laws LJ in Carson’s case, [73], which was cited in Ratcliffe, supra, [91], and which Supperstone J relied on in Cockburn at [68]. It was not manifestly without reasonable foundation for the Executive and Parliament not to have imposed an upfront capital cost of many millions of pounds on the LGPS which the evidence shows would have been the outcome if survivors’ benefits had been extended to the Claimant and others in her position. Further or alternatively, asking myself the question I have set out without reference to the ‘manifestly without reasonably foundation’ test, I reach the same conclusion.
Taking a step back and assessing matters in the round, I am here concerned with matters of economic and social strategy, in relation to which the national authorities are entitled to a broad area of discretionary judgment: see Cockburn, supra, at [77]. In my judgment the Secretary of State was fully entitled not to provide retrospective survivors’ pensions for members of the 1997 Scheme in circumstances where members had not indicated a wish to pay for this benefit, which meant that if the Secretary of State had taken that course, the result would have been the imposition of the cost on members against their wishes and/or the imposition of significant extra liabilities on the public purse.
There is nothing in In Re McLaughlin, supra, which undermines this analysis. It appears from the majority’s judgments that the legitimate aim for the measure in question put forward by the respondent Department was ‘to promote the institutions of marriage and civil partnership by conferring eligibility to claim only on the spouse or civil partner of the person who made the contributions’: see [36]. The majority’s view was that the denial of widows’ pension allowance to Ms McLaughlin was not a proportionate means of achieving this legitimate aim because, irrespective of whether the deceased and the claimant were married to each other:
“…the allowance exists because of the responsibilities of the deceased and the survivor towards their children. Those responsibilities are the same whether or not they are married to or in a civil partnership with one another.”
See [39] per Lady Hale and also [52] per Lord Mance (accepting, however, that a policy in favour of marriage could constitute justification for differential treatment, when children were not involved).
In my judgment the Secretary of State is correct to submit that this reasoning has no ‘read-across’ to the justifications advanced by him and by the Council in the present case.
Conclusion
For all of these reasons, this application for judicial review is dismissed. It is therefore unnecessary for me to consider the question of remedy.