Entrust Pension Ltd v Prospect Hospice Ltd & Anor

This case raises questions about the true meaning and effect of various provisions relating to benefits in the documentation governing an industry-wide occupational pension scheme in the healthcare and educational sector which was established in 1966 and is now called the Federated Flexiplan No. 1 (“the Scheme”). The claimant, Entrust Pension Limited (“Entrust”), is the present trustee of the Scheme (“the Trustee”), having taken over from the previous trustee, Capita Pension Trustees Limited, in 2007.

The Scheme was closed to the future accrual of benefits with effect from 1 February 2010, so it no longer has any active members in current pensionable service. As at 31 March 2011, there were exactly 3,000 current members of the Scheme, comprising 1,370 pensioners already in receipt of pension benefits and 1,630 “deferred” members who left pensionable service in the past with an entitlement to a future pension, typically due to become payable when they reach their normal retirement age. At the same date, there were 263 employers participating in the Scheme, and the Scheme’s assets were valued at £69,255,782.

The Scheme has an unusual benefits structure. It is neither a defined benefits scheme, nor a money purchase scheme, in the usual sense of either of those generic descriptions. Instead, and in broad outline, the Scheme provides a member with an entitlement, on retirement, to a cash sum which comprises the contributions paid by the member, and by the relevant employer on the member’s behalf, plus a guaranteed rate of interest on those contributions. It is convenient to refer to this sum as the member’s “Accrued Amount”. In addition, the Accrued Amount may be augmented by the Trustee adding to it an amount representing a share of any “surplus” in the Scheme. The extent to which there may be an entitlement to a share of surplus, and how and when it should be calculated, are questions which lie at the heart of the construction issues on which the court is asked to rule.

Whatever the answer to those questions may be, however, it is common ground that the Scheme was in fact administered from its inception until July 2006 in a way that replicated a conventional final salary scheme, that is to say a scheme under which benefits are computed by reference to a member’s salary when leaving the Scheme. Such benefits were described in the booklets and other explanatory literature circulated to members, and in the periodic actuarial reports commissioned by the Trustee, as “target benefits”, although that term appears nowhere in the Scheme rules themselves.

In the early days of the Scheme, members were informed that the expectation was that their target benefits would equate to one sixtieth of final remuneration for each year of pensionable service. Following the introduction of the state earnings-related pension (“SERPS”) under the Social Security Pensions Act 1975, a variation was introduced which gave employers the option of paying contributions into the Scheme at a lower rate (8%, instead of 10%, of the member’s salary), in order to reflect the fact that the Scheme could not be contracted out of SERPS, in which case target benefits would be reduced to one seventy-fifth of final remuneration for each year of pensionable service. Over the years, some further complications were introduced, which led to a number of different target benefit structures, but the basic pattern remained the same. In practice, whether the provision of a share of surplus by the Trustee was a matter of legal entitlement or whether it was purely discretionary, and whatever the ways in which it could in theory have been calculated, the Scheme operated in essentially the same way as a final salary scheme, and for as long as the employers were able and willing to continue to fund it on that basis, no problems arose. When a member retired, he would be granted an annuity equal to his target pension. This had the result that the retiring member’s Accrued Amount was topped up by the Trustee out of surplus by whatever amount was needed to provide that target pension.

One striking feature of the Scheme, which it is convenient to note at this point, is that even though members’ benefits were on any view largely fixed in amount (comprising not less than the Accrued Amount before retirement, and a fixed rate of annuity after retirement), there has never been any general covenant by the employers to pay the balance of the cost of the liabilities to the extent they exceeded the value of the Scheme assets. However, until the actuarial valuation as at 31 March 2002 this did not matter, because the Scheme was always estimated to have a surplus in excess of the amounts required to fund pensions in payment and Accrued Amounts.

The 2002 actuarial valuation disclosed for the first time a comparatively modest funding deficit of £1.323 million, and various steps were taken to address this. By 2005, however, the funding deficit in relation to target benefit pensions for active and deferred members had grown to £19.894 million. This unwelcome news led the Trustee to conclude that the provision of target benefits for active members on retirement was no longer sustainable, and the decision was taken to cease to provide target benefits for those who retired after 10 July 2006. The decision was communicated to members in written announcements, which informed them that, in the light of actuarial and legal advice received, the pension payable for members retiring after that date would be based on their guaranteed Accrued Amounts, not target pension.

On 15 May 2007 Entrust was appointed as the sole corporate trustee of the Scheme. Following consultations with employers and members, a recovery plan was agreed whereby the bulk of the contributions due from the employers would be collected from them over a period of up to ten years, and good communications between members and employers were restored. As I have already noted, the Scheme has now been closed to future accruals of benefit since 1 February 2010. The amount of the funding shortfall, and thus the length of time for which the employers will be required to make payments to eliminate it, depends on the outcome of the present proceedings, and in particular on the answer to the questions whether, and if so to what extent, members are entitled to benefits in excess of those secured by their Accrued Amounts. The question arises in a particularly acute form for deferred members who left the Scheme before the 2005 Rules came into force, and whose benefits become payable on retirement after 10 July 2006. In stark, if over-simplified, terms, are they entitled to a pension which reflects a share of the surplus which was thought to exist when they left service, and in which they would have shared on actual retirement at that date, or are they confined to a pension based on their Accrued Amounts, in common with active members who retired after 10 July 2006?

One such deferred member is the second defendant, Yvonne Hunter, who joined the Scheme on 27 August 1986 and left on 5 September 1990. She has agreed to become a representative defendant. The first defendant, Prospect Hospice Limited, is a registered charity which provides hospice services in North Wiltshire and South Gloucestershire. It has been a participating employer since at least 1988, and it too has agreed to be appointed as a representative defendant. Because of the complexity and interaction of the issues, it is agreed that the necessary representation orders will be issue rather than class-based, with the first defendant arguing for the outcomes which can in general be expected to favour the employers, and the second defendant arguing for the outcomes likely to favour the members.

I have had the benefit of very full and careful written and oral arguments from counsel for both defendants, to whom I will refer for convenience as counsel for the employers and counsel for the members respectively, as well as from Mr Michael Furness QC on behalf of Entrust. I express my gratitude to them all.

The issues which I am now asked to determine are, with one possible exception, issues of pure construction of the rules of the Scheme. The extrinsic evidence which is relevant and admissible for that purpose is limited, and I have not been asked to resolve any disputed questions of fact. Depending on the answers to those issues, it may subsequently be necessary for the court to decide various questions of fact, or to give directions to Entrust about the factual assumptions it should make when administering and winding up the Scheme. The parties therefore sensibly agreed, and the court has ordered, that there should be a split trial, with the first hearing dealing only with questions of construction. A second hearing has been fixed for October 2012 to determine such consequential questions of fact as it may prove necessary to answer. Everybody agrees that the latter exercise, should it arise, may be a very difficult one, because of the incomplete and inadequate nature of the Scheme records over a period of many years, and possible differences in the way in which individual members were treated. It is also rightly agreed that the potential difficulty of the factual enquiries which may arise is not, in itself, a relevant factor for the court to take into account on the questions of construction. Understandably, however, Entrust hopes that the court will find itself able to answer the questions of construction in the way which would cause least disturbance to the current benefit structure of the Scheme and to how it has in practice been administered in the past.

The claim form was issued on 18 June 2010 and has been twice amended. On the basis of the questions raised in paragraph 2 of the claim form, the parties have helpfully agreed a list of issues for the court to determine on this hearing (“the List of Issues”). The case has been argued by reference to the issues so identified, and references in this judgment to numbered Issues refer accordingly.

In broad terms, the Issues fall into the following categories. Issues 1 to 5 relate to the 1976 Rules of the Scheme, and raise questions about entitlement to, and the calculation of, shares of surplus. Issues 6 to 10 ask the same questions about similarly, but not identically, worded provisions in the 2005 Rules. Issues 12, 13 and 14 raise questions about the entitlement to surplus of deferred members who left service while the 1976 Rules were in force, and about the present duties of the Trustee if its predecessor acted on the wrong basis in the past. These last three Issues do not arise under the 2005 Rules. Finally, it is agreed that Issues 11 and 15 have no practical relevance and therefore need not be argued.

With this introduction, I will now turn in more detail to the history of the Scheme and the relevant provisions of the rules. I will also make reference, where appropriate, to the explanatory material provided to members, and to the actuarial reports. I will then deal briefly with the general approach to the construction of pension scheme documents, after which I will deal with the Issues in turn.

The Scheme was established on 12 August 1966 by an interim trust deed, and came into operation on 1 October 1966. No copy of the interim deed can now be found, but it appears from a recital to the first definitive trust deed executed on 12 November 1968 (“the 1968 Deed”) that the interim deed was in the minimalist form then customary and provided for a definitive deed to be executed within two years (a deadline which, again not untypically, was allowed to slip by a few months).

The 1968 Deed recited that the Trustee, which was then a company called Federated Pension Schemes, had for many years run a superannuation scheme for nurses and hospital officers, and that it now wished to establish a new scheme for the benefit of persons employed in those or similar capacities, comprising two funds (to be known as the Pension Fund and the Lump Sum Fund) each of which should be capable of approval under the relevant income tax legislation then in force. To this end, the 1968 Deed and the rules scheduled to it (“the 1968 Rules”) duly established the Scheme with separate rules for the two funds providing pension and lump sum benefits respectively.

For present purposes, it is enough to note the following key features of the rules relating to the Pension Fund. Members were to contribute five per cent of their remuneration every year, while principal employers were to contribute somewhat larger specified percentages of the aggregate remuneration of their male and female employees. Normal pension age was 60 for both men and women. Upon retirement at that age, a member was entitled to an annual pension for life, the amount of which was defined in rule 14 as follows:

“14.

(1) The amount of the Pension payable to a Member on retirement will depend on the amount of contributions and interest available.

(2)

In the case of every Member who becomes entitled to retire on Pension the Trustee shall ascertain the aggregate amount of all sums credited to or in respect of such Member in the accounts of the Pension Fund (being the contributions of the Member concerned … and of Participating Employers in respect of him [subject to certain deductions]) and there shall be added thereto (a) compound interest on each such respective sum calculated at [a specified rate] up to the time of retirement and (b) such an amount (if any) as in the opinion of the Trustee may properly be added thereto as representing the retiring Member’s share of any actuarial surplus arising in the Pension Fund as to which the decision of the Trustee shall be final and shall not be questioned. The total amount ascertained in accordance with the foregoing provisions is hereinafter referred to as the Member’s “Pension Capital”.

(3)

No part of a Member’s Pension Capital can be withdrawn by him but the whole amount thereof shall be used by the Trustee to provide a pension for the retired Member which may be provided by purchasing a Pension in the form of an annuity from an Insurance Company to which the Insurance Companies Act 1958 applies.

(4)

No Pension payable out of this Fund shall exceed the lesser of: -

(A)

£3,000 per annum …

(B)

Two-thirds (or such lesser fraction not being less than 1/60^th for each year of Service as the Commissioners of Inland Revenue may require) of the retiring Member’s Remuneration as at Normal Pension Age or earlier date of retirement …”

It can be seen from these provisions that the unusual benefit structure of the Scheme was present from the beginning, with entitlement to a basic amount of pension comprising the contributions made by or on behalf of a member plus compound interest (i.e. the member’s Accrued Amount), together with such amount (if any) as the Trustee might decide could properly be added thereto as a share of any actuarial surplus in the Pension Fund. The total amount thus ascertained was described as the member’s “Pension Capital”. The maximum amount of annual pension was, in broad terms, not to exceed the lesser of £3,000 and n/60ths of final remuneration for each year of pensionable service.

The rules applicable to the Lump Sum Fund provided for participating employers to make contributions to it which, when added to their contributions to the Pension Fund, came to 10 per cent of the relevant members’ aggregate remuneration. Rule 28 provided that every member who retired on pension would also be entitled to a lump sum payment equal to the contributions made to the Lump Sum Fund in respect of the member together with compound interest

“… and together also with such an amount (if any) as in the opinion of the Trustee may be properly added thereto as representing the retiring Member’s share of any actuarial surplus arising in the Lump Sum Fund as to which the decision of the Trustee shall be final and shall not be questioned.”

Part III of the 1968 Rules contained rules of general application to the Scheme. These included a power in rule 41 for the Trustee “in its absolute and uncontrolled discretion” to award supplementary annuities to compensate for rises in the general cost of living.

The first booklet for members which survives is dated June 1969. It described its purpose as being to give “a simple outline” of the Scheme, and emphasised at the outset that the Scheme was regulated by a trust deed and rules

“… and all matters relating to it will be determined by these rather than by this outline which has been prepared for the convenience of employees.”

Paragraph 7 of the booklet contained a brief description of the pension and lump sum benefits payable at normal pension age, and said:

“It is hoped, but not guaranteed, that (i) and (ii) together [i.e. pension plus lump sum] will approximate to a pension of 1/60^th of average final salary for each year of service under the Group Scheme.”

The next major development came in 1976, when by a second definitive trust deed and rules dated 9 September of that year (“the 1976 Deed” and “the 1976 Rules”) the provisions of the 1968 Deed and Rules were replaced with effect from 1 April 1976. The principal reasons for the adoption of the 1976 Deed and Rules appear to have been (a) changes in the legislation governing Inland Revenue approval of pension schemes contained in the Finance Acts of 1970 and 1971, which now allowed a single pension scheme to provide all the relevant benefits for members, and (b) the introduction from 1975 (by the Social Security Act 1973) of requirements for the preservation of benefits of deferred members who left service before their normal retirement date. The underlying nature of the Scheme, however, and its benefit structure, remained essentially unchanged. The previously separate Pension and Lump Sum Funds were merged into a single Fund, and the rules on contributions were simplified so that members continued to contribute 5% of their salary, while the employers made a unified contribution of 10% of the aggregate salary of all contributing members employed by them.

The benefits payable to a member on retirement were set out in rules 11 to 13, as follows:

“11.

On retirement at or after Normal Pension Age subject to any provision for reduction under Rule 14 [which provided for the partial surrender of pension to nominated dependants]

(1)

The amount of the pension payable to a Member on retirement will depend on the amount of contributions and interest available.

(2)

In the case of every Member who becomes entitled to retire on pension the Trustee shall ascertain the aggregate amount of all sums credited to or in respect of such Member in the accounts of the Fund … and there shall be added thereto (a) compound interest calculated at the rate of 4% per annum with annual rests in accordance with Rule 16(2) up to the time of retirement and (b) such an amount (if any) as in the opinion of the Trustee may properly be added thereto as representing the retiring Member’s share of any actuarial surplus arising in the Fund as to which the decision of the Trustee shall be final and shall not be questioned. The total amount ascertained in accordance with the foregoing provisions is hereinafter referred to as the Member’s “Pension Capital”.

(3)

[This enabled a member to elect to receive part of his Pension Capital in the form of a lump sum].

12.

Any member of the Group Scheme who has become in the opinion of the Trustee and the Participating Employer by whom he is employed incapable of discharging his duties by reason of incapacity and who has retired from the Service in consequence thereof shall (notwithstanding that he may not otherwise be eligible for a pension under the Rules) be entitled on retirement during life to an annual pension calculated as at the time of retirement by the Trustee in accordance with the advice of the Actuary PROVIDED ALWAYS that such pension shall not be less than that calculated in accordance with Rule 11.

13.

(1) Any Member of the Group Scheme who has attained age 50 may subject to the prior approval of the Participating Employer by whom he is employed be permitted by the Trustee to retire at any time and receive an annual pension calculated as at the time of retirement in accordance with Rule 11 …”

The provisions of the 1976 Rules relating to deferred pensions were contained in rules 18 and 19, and since these rules are of central importance to Issues 12 to 14, I will set out the relevant parts of them:

“18.

(1) Subject to Rule 23(2) if a Member (whether of Class A or of Class B) before becoming entitled to an immediate pension out of the Fund leaves the Service whether voluntarily or involuntarily and within 12 months thereafter (or such longer period as the Trustee may approve in any exceptional case) takes up employment with another employer who maintains a pension scheme (which qualifies as a Transfer Scheme) he shall have the option unless his contributions have been returned to him pursuant to sub-Rule (3) hereof by notice in writing to the Trustee not later than 12 months (or such longer period as the Trustee may approve in any exceptional case) after leaving the Service to elect that the Trustee shall make a transfer payment in accordance with Rule 25 hereof .

(2)

Subject to Rule 23(2) a Member who before becoming entitled to an immediate pension out of the Fund leaves the Service whether voluntarily or involuntarily but in respect of whom a Transfer Payment is not made pursuant to sub-Rule (1) hereof and

(a)

who (whether of Class A or Class B) has five or more Years of Qualifying Service and has reached the age of 26 or whose remuneration has at any time exceeded £5,000 per annum or

(b)

who is a Class A Member and within 12 months (or such longer period as the Trustee may approve in any exceptional case) takes up employment or becomes self-employed in his profession as the Trustee shall decide in each case

shall (unless his contributions have been returned to him pursuant to sub-Rule (3) hereof) be granted a deferred pension out of the Fund calculated as provided by Rule 19(1) and (2) payable at such time and on such terms as are specified in the Rules …

(3)

[This provided for the return of contributions in limited cases]

19.

(1) A deferred pension granted out of the Fund pursuant to Rule 18 shall be calculated in the manner set out in Rule 11 based on the Member’s Pension Capital at the date of leaving the Service.

(2)

A deferred pension shall include all additional benefits to which a Member would be entitled on attaining Normal Pension Age including …

(3)

Where a Member has become entitled to a deferred pension pursuant to Rule 18 then … such pension shall commence either

(a)

when the Member shall reach Normal Pension Age under this Scheme

(b)

at the Member’s request if and when the Trustee shall be satisfied that such Member is permanently incapacitated

(c)

after attaining the age of 50 at the Member’s request

whichever first occurs.

(4)

On the death of any Member who has become entitled to a deferred pension as aforesaid then if he shall have died before becoming entitled to receive his deferred pension the Trustee shall raise out of the Fund and hold upon Discretionary Trusts a sum equal to the Member’s Pension Capital calculated in the manner set out in Rule 11 as at the date of his death.”

The changes made by the 1976 Deed and Rules were explained by the Trustee to participating employers in a memorandum dated 16 March 1977. It stated that there was “very little difference” between the old and the new rules of the Scheme, but drew attention in particular to rule 18 “which now contains the provisions of the 1973 Social Security Act regarding preservation of benefits for those Members who leave your employment and for whom a transfer value cannot be paid”. The memorandum also explained that the Scheme could not be contracted out of SERPS, which had been introduced by the Social Security Pensions Act 1975 and would apply from April 1978 onwards, and indicated the Trustee’s intention of introducing an option for contributions to be paid at reduced levels of 8% from the employer and 4% from the member with corresponding reductions in benefits:

“If we are able to continue to pay benefits at 1/60^th of final salary for those for whom contributions total 15% we would similarly be able to pay benefits at a level of 1/75th of final salary for service during which contributions total 12%.”

The first surviving actuarial report and valuation of the Scheme is a report on the position as at 31 March 1978, prepared in August 1979 by the Scheme actuaries, Duncan C Fraser & Co. This report reviewed developments over the period of three years from 31 March 1975, when the previous actuarial valuation (which no longer survives) had been carried out.

By way of introduction, the actuaries provided the following description of the operation of the Scheme:

“The Scheme is based on a fixed level of annual contributions, payable one-third by the Member and two-thirds by the Employer. The fixed level is normally 15 per cent of Members’ salaries, but with effect from 1 April, 1978 a Participating Employer, with the agreement of the Trustee, may elect for a lower level of 12 per cent to apply.

The Trust Deed and Rules provide that the benefits available to a Member will be determined by accumulating the contributions paid by and in respect of him at specified rates of interest. The Trust Deed further provides that such benefits can be augmented by a share to be determined by the Trustee of any actuarial surplus. We are informed that it is the intention of the Trustee to apportion any such actuarial surplus so that the benefits will approximate, so far as the surplus permits, to those which would emerge from a final salary pension arrangement. In particular, the aim is to provide, for the period during which contributions are paid at the level of 15 per cent of salary, a pension on retirement at normal pension age of one-sixtieth of the highest average salary over any period of 12 calendar months within the last 3 years of service for each year of pensionable service. For periods during which the lower level of contribution has been paid, the benefits would be proportionately reduced. The Scheme does not, however, guarantee to provide benefits according to these final salary formulae.”

The actuaries then made their valuation on the assumption that the Trustee would continue to provide benefits on the same basis as before, and concluded, on that footing, that there was an excess of assets over liabilities of £586,000. In the final section of the report, headed “Comments and Recommendations”, they pointed out that average salaries had increased by 54% over the last three years, and continued:

“The Scheme is financed by means of a fixed rate of annual contribution and in consequence the benefits which can be provided must be those which the contributions can ultimately support. The ability of the Scheme to continue in the longer term to provide benefits calculated according to the final salary formulae will be influenced by the course of inflation and whether the number of active members grows or contracts. The Scheme is vulnerable to both inflation and a reduction in the number of active members.”

Despite this caveat, however, it was recommended that the Trustee should continue to augment the Accrued Amounts of members so as to provide a pension related to final salary until the next actuarial valuation, which was due to take place not later than 31 March 1980.

It is not clear exactly when or how the Trustee implemented the intended alteration of contribution levels from 1 April 1978, but the position was in any event regularised for the future by a further deed of amendment dated 26 January 1983, which altered the rules on contributions by members and employers by adding in each case the words “or such other percentage of Salary as the Trustee shall from time to time authorise”. Further changes to contribution rates were then made in 1988 and 1997, which had the effect of complicating the benefit structure, but not in any way which is material to the questions I now have to decide.

Meanwhile, the explanatory guides produced for members, and the surviving actuarial valuation reports, continued to state that it was the Trustee’s policy to provide target benefits. Thus, for example, the employees’ guide issued in January 1989 explained that the Scheme was a “targeted final salary” pension scheme, the aim of which was to provide members with an income in retirement which was related to their salary close to retirement and the number of years for which they had been members of the Scheme. The guide also included the following note:

“Unlike true final salary pension schemes the contributions to [the Scheme] are fixed. It is therefore possible that, at some time in the future, because of inflation, investment or other factors, the Plan may be unable to provide the target pension unless higher contributions can be agreed. In these circumstances there is a guarantee that the pension will not be less than that which can be purchased by the contributions paid plus 4% p.a. compound interest. However, it should be noted that the Plan has been going since 1966 and has never failed to pay the target pension.”

Statements to similar effect may be found, again by way of example, in the actuarial valuation of the Scheme as at 31 March 1989 by William Mercer Fraser, as the Scheme actuaries had by then become. The financial position of the Scheme remained healthy, with an excess of assets over target liabilities of £4.136 million. This position then remained essentially unchanged in the 1994, 1997 and 2000 actuarial reports, each of which included a statement that:

“The main purpose of the valuation is to assess whether the current contributions payable to the Scheme are sufficient to pay the target benefits in the long term including discretionary pension increases.”

The first cloud on the horizon appeared in the shape of the 2002 actuarial report by F P S Actuarial Services Limited. This revealed a shortfall of assets to past service target benefits of £1.323 million on an ongoing basis, in comparison with an excess of £18.2 million at the previous valuation in 2000. The reasons given for this decline in the funding position were poor investment performance and changed actuarial assumptions to reflect longer life expectancies. The immediate action taken by the Trustee to manage the funding problem was to change the calculation of target benefits from a final salary basis to (in effect) an average salary basis for the future, combined with an increase in contribution rates of one per cent from 1 April 2003. The decision was also taken to reduce the guaranteed rate of compound interest on contributions from 4% to 2%, although it is common ground that this decision was not effectively implemented until September 2005.

Between 2002 and 2005 the financial position of the Scheme continued to deteriorate, and the next actuarial report as at 31 March 2005 (now prepared by Capita Hartshead) revealed that the Scheme’s assets were “significantly below” those required to provide target benefits for active and deferred members, as well as for existing pensioners. The amount of the deficit in relation to past service target benefits on an ongoing basis had in fact increased to £19.894 million. The assets were, however, “broadly in line” with those required to continue to provide target benefits for existing pensioners, and to pay benefits for active and deferred members based on their Accrued Amounts. In the light of this report, and having taken legal advice, the Trustee decided that from 10 July 2006 target benefits would no longer be provided for future retirements.

Meanwhile, new definitive documentation for the Scheme was introduced in the form of a trust deed and attached rules dated 29 September 2005 (“the 2005 Deed” and “the 2005 Rules”). These replaced the 1976 Deed and Rules with effect from 1 October 2005.

The principal relevant provisions of the 2005 Rules are as follows.

Rule 8 deals with “Accounts”, and provides that:

“(1)

The Trustee maintains one or more Accounts in respect of each Member and Early Leaver. Any payment by or in respect of a Member … is credited to an Account …

(2)

Accounts are maintained solely for the purpose of calculating benefits and do not confer on any person any interest in the Plan or its assets which he would not otherwise have had …

…

(4)

Interest will be credited to a Member’s Account up to the date of retirement.”

Rule 1 defines “Interest” as meaning:

“(a)

in respect of contributions paid before 1^st April 2003, compound interest at the rate of 4% per annum with annual rests on 31^st March each year, and

(b)

in respect of contributions paid after 31^st March 2003, compound interest at the rate of 2% per annum with annual rests on 31^st March each year.”

In fact, however, no amendment to the 1976 Rules effecting this change in interest rates had been made before the 2005 Deed itself. It is agreed that the 2005 Deed could not have had retrospective effect, because of the terms of rule 41 of 1976 Rules and section 67 of the Pensions Act 1995, so (as I have already mentioned) the interest rate change was effective only from the date of the 2005 Deed itself.

Rule 11(3) operates in a similar way to rule 11(2) in the 1976 Rules, although it is differently worded:

“(3)

In the case of a Member who becomes entitled to retire on pension the Trustee shall determine his “Pension Capital” being the amount standing to the credit of his Account and such amount, if any, as in the opinion of the Trustee may properly be added thereto as representing the retiring Member’s share of any actuarial surplus arising in the Plan as to which the decision of the Trustee shall be final and shall not be questioned.”

Rule 11(4) then makes provision for the conversion of the Member’s cash entitlement into a pension:

“(4)

The amount of a pension which can be provided from that part of a Member’s Pension Capital which remains after the payment of any cash sum is decided by the Trustee after taking actuarial advice or in the form of an annuity with an insurer. The Trustee will use such annuity conversion factors (after taking actuarial advice) as the Trustee determines.”

This provision had no express precursor in the 1968 or 1976 Rules, but I accept the submission of counsel for Entrust that it does no more than make explicit what was already implicit in the earlier rules.

There was no significant disagreement between the parties about the general approach which the court should adopt to the interpretation of pension scheme documentation. I would accept and adopt the concise summary contained in paragraph 12 of the skeleton argument of counsel for the employers:

“The principles governing the construction of pension scheme rules are well-established and not controversial. The Court should avoid a narrow, literal, or overly technical approach and should instead adopt an approach which produces a reasonable and practical result. Although the Court should not be predisposed to arriving at any particular result, its approach should be influenced by the practical consequences of the possible rival interpretations: see National Grid [2001] 1 WLR 864 at 877-878; In Re Courage [1987] 1 WLR 495 at 505; Stevens v Bell (British Airways) [2002] EWCA Civ 672 at [26-32]; Armitage v Staveley Industries [2006] PLR 191 at 196-7; H R Trustees v Wembley PLC [2011] EWHC 2974 (Ch) at [23-25]. This reflects the principles applicable to construction more generally, as established by ICS v West Bromwich [1998] 1 WLR 896 at 912-914; Chartbrook Limited v PersimmonHomesLimited [2009] 1 AC 1101, particularly at [14-26]; Rainy Sky SA v Kookmin Bank [2011] UKSC 50 at [14-30].”

In a little more detail, I would refer to the helpful summary of the relevant principles given by Arden LJ (with whom Waller and Auld LJJ agreed) in the British Airways case at [26] to [32], from which I cite the following extracts:

“26.

There have been several reported cases about the interpretation of provisions of pension schemes in recent years. There are no special rules of construction but pension schemes have certain characteristics which tend to differentiate them from other analogous instruments. I mention some of those characteristics in the following paragraphs.

27.

First, members of a scheme are not volunteers: the benefits which they receive under the scheme are part of the remuneration for their services and this is so whether the scheme is contributory or non-contributory. This means that they are in a different position in some respects from beneficiaries of a private trust. Moreover, the relationship of members to the employer must be seen as running in parallel with their employment relationship. This factor too, can in appropriate circumstances have an effect on the interpretation of the scheme.

28.

Second, a pension scheme should be construed so [as] to give a reasonable and practical effect to the scheme. The administration of a pension fund is a complex matter and it seems to me that it would be crying for the moon to expect the draftsman to have legislated exhaustively for every eventuality … it is necessary to test competing permissible constructions of a pension scheme against the consequences they produce in practice. Technicality is to be avoided …

29.

Third, in pension schemes, difficulties can arise where different provisions have been amended at different points in time. The effect is that the version of the scheme in issue may represent a “patchwork” of provisions: see per Robert Walker J in the National Grid. Pension schemes are often subject to considerable amendment over time. The general principle is that each new provision should be considered against the circumstances prevailing at the date when it was adopted rather than as at the date of the original trust deed … Likewise, the meaning of a clause in the scheme must be ascertained by examining the deed as it stood at the time the clause was first introduced …

30.

Fourth, and as with any other instrument, a provision of a trust deed must be interpreted in the light of the factual situation at the time it was created. This includes the practice and requirements of the Inland Revenue at that time, and may include common practice among practitioners in the field as evidenced by the works of practitioners at that time. It has been submitted to us that the factual background is only relevant if the document is ambiguous. I do not accept this submission, which is inconsistent with the approach laid down by Lord Hoffmann in [the West Bromwich case] …

31.

Fifth, at the end of the day, however, the function of the court is to construe the document without any predisposition as to the correct philosophical approach …

32.

Sixth, a pension scheme should be interpreted as a whole. The meaning of a particular clause should be considered in conjunction with other relevant clauses. To borrow John Donne’s famous phrase, no clause “is an island entire of itself”.”

It is also worth quoting one key paragraph from the judgment of the Supreme Court, delivered by Lord Clarke of Stone-cum-Ebony JSC, in Rainy Sky [2012] Bus LR 313 at [21]:

“The language used by the parties will often have more than one potential meaning. I would accept the submission made on behalf of the appellants that the exercise of construction is essentially one unitary exercise in which the court must consider the language used and ascertain what a reasonable person, that is a person who has all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract, would have understood the parties to have meant. In doing so, the court must have regard to all the relevant surrounding circumstances. If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other.”

In so holding, the Supreme Court rejected a narrower principle, which had found favour with the majority in the Court of Appeal, that the court is obliged to give effect to “the most natural meaning of the words” unless it produces a result so extreme as to suggest that it was unintended. At [30], Lord Clarke said the essence of the correct approach was that:

“where a term of a contract is open to more than one interpretation, it is generally appropriate to adopt the interpretation which is most consistent with business common sense.”

Despite the general agreement between counsel about the right approach to construction of the Scheme, there were some relatively minor points of disagreement between them which I should briefly mention.

First, counsel for the members submitted that the starting point and “golden rule” in construing documents is that stated in Lewison, The Interpretation of Contracts, 5^th edition, para 5.01, namely:

“The words of a contract should be interpreted in their grammatical and ordinary sense in context, except to the extent that some modification is necessary in order to avoid absurdity inconsistency or repugnancy.”

The wording of this rule, however, predates the decision of the Supreme Court in Rainy Sky, and must in my judgment be read subject to it. With the greatest respect, I would question whether it is now helpful to begin with any kind of “golden rule” when what has to be performed is the single unitary exercise identified by the Supreme Court.

Secondly, there was some debate about the extent, if any, to which post-contractual conduct may be admissible as an aid to the construction of documents. As to this, I do not find it helpful to consider the problem in the abstract, but in general I am satisfied that the position is correctly stated, so far as English law is concerned, in Lewison at para 3.19:

“The court may not generally look at the subsequent conduct of the parties to interpret a written agreement. However, where the agreement is partly written and partly oral, subsequent conduct may be examined for the purpose of determining what were the full terms of the contract. In addition the subsequent conduct of the parties may be examined where an estoppel by convention is alleged; where it is alleged that the agreement was a sham; and probably for the purposes of determining the boundaries of an ambiguous grant of land.”

The basic principle was clearly stated by Lord Reid in Whitworth Street Estates Limited v Miller [1970] AC 583 at 603D:

“I must say that I had thought that it is now well settled that it is not legitimate to use as an aid in the construction of the contract anything which the parties said or did after it was made. Otherwise one might have the result that a contract meant one thing the day it was signed, but by reason of subsequent events meant something different a month or a year later.”

Issues 1 to 5 stem from the wording of the second component of the pension to which a member became entitled on retirement at normal pension age under rule 11(2) of the 1976 Rules (referred to in the List of Issues as “the Old Rules”, a definition which I have avoided using in this judgment because the oldest rules of the Scheme were of course the 1968 Rules). For convenience, I repeat the critical words:

“… and (b) such an amount (if any) as in the opinion of the Trustee may properly be added thereto as representing the retiring Member’s share of any actuarial surplus arising in the Fund as to which the decision of the Trustee shall be final and shall not be questioned.”

The Issues are as follows:

“1.

Under Old Rule 11(2), and when the Scheme is in surplus does the Trustee have a discretion as to whether or not to credit a retiring member with a share of surplus, or is it obliged to do so?”

2.

If the Trustee is obliged to, or decides to, credit a member with such a share of surplus, does the Trustee have a discretion over the quantum of the share of surplus, or is it obliged to adopt a particular method of computation (and if so what method)?

3.

Once the Trustee has determined that a member should be credited with a given share of surplus, does the Trustee have a discretion over the amount to be credited to or in respect of the member, or is it obliged to adopt a particular method of computation (and if so what method)?

4.

If the Trustee has discretion, what principles govern the Trustee’s exercise of discretion?

5.

(i) How is surplus to be determined by the Trustee for the purpose of Old Rule 11(2) – does the Trustee or the Actuary have a discretion over the choice of method, or do they have to adopt a particular method?

(ii)

Can, or must, the Trustee use the surplus disclosed in the most recent actuarial valuation of the Scheme?

(iii)

If that is permissible, does that excuse the Trustee or the Actuary from considering any alternative methods?

(iv)

If the most recent valuation includes more than one method or basis of valuation, how is the Trustee to select the method to apply?

(v)

Having regard to the methods actually adopted in past actuarial valuations for the Scheme is one method to be preferred to other methods?”

There is clearly a considerable degree of overlap between some of these questions, but in the broadest terms the first underlying question to which they give rise is whether each member is a priori entitled to a defined share of surplus, so that the Trustee’s role is essentially to exercise its judgment as to the amount which properly represents that share, or whether the Trustee has a wide discretion whether to award a share of surplus at all. On either basis, the Trustee then needs to know what is meant by “actuarial surplus” and the concept of a “share of actuarial surplus”, and how such surplus is to be ascertained. Further, once it has been ascertained, the Trustee needs to know whether it has any discretion about the amount of surplus to be allocated to each retiring member, and what principles should guide the Trustee in the exercise of any discretion it may possess. In general, it is likely to be in the financial interests of the employers to argue for the maximum amount of discretion for the Trustee in relation to these Issues, and in the interest of the members to argue for a fixed entitlement and fixed principles of computation.

I will begin with the first underlying question which I have identified, because the answer to it is of crucial importance for all that follows. Indeed, I would describe it as the single most important issue in the case. In the simplest terms, do members have an entitlement to a definable share of surplus in addition to their Accrued Amounts, or is the award of a share of surplus purely discretionary?

Counsel for the employers argued that the Trustee should be afforded a wide discretion in relation to the application of surplus, and that Issues 1 to 3 should be answered accordingly. They began their analysis with the wording of the rule, and then moved on to wider contextual and purposive considerations, while recognising that the submissions could not be rigidly compartmentalised, and that the exercise of construction is ultimately the single unitary one described in Rainy Sky.

On the wording of the rule, counsel for the employers rely on the words “such an amount (if any) …” as indicating that a member has no right to be awarded a share of surplus at all, even if a surplus exists. They point out that the words “if any” qualify “amount”, and that their limiting function is different from that performed by the word “any” in the phrase “any actuarial surplus arising”. The latter phrase reflects the obvious point that there may at the material time be no surplus at all, on whatever basis it is to be calculated. The former phrase presupposes that surplus exists, and makes it clear that the member’s right to any part of it depends on the prior exercise by the Trustee of a discretion in the member’s favour. This indication is then reinforced by the words “in the opinion of the Trustee” and the permissive “may”, which are classic indicators that the exercise to be performed is a discretionary one. The word “properly”, it is submitted, does not detract from this, because it merely makes express what is implicit anyway, namely that the Trustee is obliged to exercise its power to add a share of surplus in accordance with its fiduciary obligations (on which see Issue 4). Similarly, the words “as to which the decision of the Trustee shall be final and shall not be questioned” point clearly in the same direction, and are not consistent with the proposition that a member has an objectively ascertainable entitlement to a share of surplus, because if such an entitlement did exist, it would necessarily render the Trustee’s decision open to question. Furthermore, this discretion about the amount (if any) of surplus to be credited to the member contrasts with the lack of discretion regarding the other element of Pension Capital, namely the member’s Accrued Amount, in respect of which the language is mandatory (“the Trustee shall ascertain the aggregate amount of all sums credited to or in respect of such Member … and there shall be added thereto (a) compound interest …”), and no “decision” or “opinion” of the Trustee is required.

Counsel for the employers went on to argue that, if the rule had been intended to confer a fixed entitlement to a share of surplus, it is extraordinary that no methodology was prescribed by the draftsman for its ascertainment. Not only would one expect this as a matter of simple common sense, and by parity with the clear directions for calculation of the Accrued Amount, but it is also made necessary by the inherently notional nature of actuarial surplus, depending as it does on the application of a number of actuarial assumptions which are matters for expert judgment and which may change over time, as well as on the impact (predictable or otherwise) of external events. In the absence of a clear prescribed methodology, submit counsel, there can be no single, objectively verifiable basis for the ascertainment of surplus, and thus no pre-existing entitlement to a share of it.

Counsel sought to derive further support for this submission from the fact that any calculation of surplus is inevitably no more than an actuarial estimate, at a particular date, of the funding position of the Scheme. There is always a risk, as experience has indeed shown, that a healthy looking surplus may later prove to have been illusory. As a matter of common prudence, this uncertainty is a major factor that the Trustee must always take into account before deciding whether to take the irrevocable step of augmenting a member’s benefits (and therefore increasing the Scheme’s long term liabilities) by crediting the member with a share of that notional surplus. Considerations of this nature are easily accommodated within a discretionary framework, but they are much harder, if not impossible, to reconcile with the argument that a retiring member has a fixed entitlement to a share of surplus. If that argument is right, say counsel for the employers, the effect is to require the Trustee continually to ratchet up benefits out of actuarial surplus, thereby increasing the risk of future deficits. They might have added, in my view, that such an intention is a particularly implausible one to attribute to the architects of a scheme which imposed no balance of cost funding obligation on the employers.

I now turn to the submissions of counsel for the members. Mr Moeran began by emphasising the word “shall” in the phrase “shall be added thereto”, and submitted that its mandatory force must carry through into the second element of a member’s pension under the rule. He submitted that, if there was no obligation to credit retiring members with a share of surplus, then there must also be no obligation to credit them with compound interest. That is a submission which I would immediately reject. It seems to me that there can be no doubt about the mandatory force of “shall” in relation to the first component of pension, i.e. the Accrued Amount, and that the adoption of a discretionary interpretation of the second element would have no impact on this. Furthermore, I consider that the mandatory “shall” is easily reconcilable with a discretionary interpretation of the second limb, because it expresses an obligation to add to the member’s Accrued Amount such amount, if any, as the Trustee may in the exercise of its discretion decide should be added thereto under the second limb. In other words, there is no conflict, but merely an obligation which is itself dependant on the prior exercise of a discretion.

A more powerful point, to my mind, is derived from the words “as in the opinion of the Trustee may properly be added thereto as representing the retiring Member’s share”. If something represents a retiring member’s share, it must conceptually have a prior existence, and the function of the Trustee is confined to ascertaining it. For that purpose, the Trustee has to form an opinion, not exercise a discretion, and in substance the role of the Trustee is to act as an expert. The concluding words to the effect that their decision shall be final and shall not be questioned serve the purpose, on this interpretation, of making it clear that the Trustee’s expert determination is to be final and binding. On this analysis, the correct classification of the power conferred on the Trustee is that it is a “category 3 power” in the well-known classification of powers set out by Warner J in Mettoy Pension Trustees Ltd v Evans [1990] 1 WLR 1587, 1613-1614, at 1614C:

“Category 3 comprises any discretion which is really a duty to form a judgment as to the existence or otherwise of particular circumstances giving rise to particular consequences.”

Reliance was also placed on the use of the adverb “properly”, as being directed towards the process of correctly ascertaining a particular sum rather than exercising a discretion in an unobjectionable manner, and on the last sentence of rule 11(2), which refers to “[t]he total amount ascertained in accordance with the foregoing provisions”: to ascertain something, counsel submitted, you discover or determine what it is, you do not create or alter it. Mr Moeran also pointed to the contrast between the wording of the second limb of rule 11(2) and the language of unfettered discretion which may be found, for example, in rule 21 of the 1976 Rules:

“Any benefits … may from time to time be increased by the Trustee in its absolute discretion acting upon Actuarial Advice PROVIDED that any such increase shall be limited so that the new benefits payable shall not exceed the appropriate amounts specified in Rule 20.”

This suggests, said counsel, that if the draftsman of the 1976 Rules had intended to provide the Trustee with a dispositive discretion under rule 11(2), he would have done so expressly. Another example, to which the same comment applies, may be found in clause 29(1)(c) of the 1976 Deed, dealing with the application of the Fund in a winding up, where the Trustee was given an express discretion to augment the benefits referred to in paragraphs (a) and (b) of that sub-clause.

Counsel also referred in this context to rule 25 of the 1976 Rules, which provided for the making of a transfer payment to another pension scheme when a member left service. By virtue of rule 25(1)(e):

“The amount of the transfer payment shall be equal to the Member’s actuarial interest in the Fund or such other amount as the Trustee may approve on Actuarial Advice having regard to any special transfer arrangements applicable to the Transfer Scheme involved.”

The reference to a member’s “actuarial interest”, it was said, must have been intended to refer to more than the member’s Accrued Amount, and must therefore have envisaged some allocation of surplus, either on the same basis as under rule 11(2) or in some comparable manner. How, asked counsel rhetorically, could it be fair to give less to a member on his retirement than he would have been entitled to by way of a transfer payment upon leaving service at an earlier date?

Mr Moeran naturally emphasised that the members were not volunteers, and that the Scheme had in practice been operated from its inception until 2006 so as to provide retiring members with target benefits. He submitted that, viewed objectively, this had always been the aim and genesis of the Scheme, not in the sense that target benefits were guaranteed – that would be impossible with fixed-rate contributions – but in the sense that, if there were enough assets in the Scheme to provide them, it was mandatory to do so. Given that the benefits received by members under the Scheme were deferred remuneration for their services (see British Airways at paragraph [27]), the court should be very slow to conclude that their interest in surplus under rule 11(2) was purely discretionary. On the contrary, it was a fundamental benefit, and the fact that there might be difficulties in ascertaining a member’s share of actuarial surplus should not cause the court to lose sight of its essential nature as a core benefit. As to the alleged practical difficulties of ascertaining it, these were matters for the Scheme actuary to consider and advise on, and the 1976 Rules (like the 1968 and 2005 Rules) clearly proceeded on the basis that there would at any given time be a relevant and ascertainable actuarial surplus, unless of course the Scheme was in deficit. If this were not the case, the Trustee could not have been given power over it, even on the employers’ construction of rule 11(2).

In summary, therefore, it is the duty of the Trustee to identify the actuarial surplus in the Scheme when a member retires, and then to ascertain the member’s share of that surplus. In performing this duty, the Trustee is forming a judgment about the existence of a state of affairs, and is exercising a category 3 power in the Mettoy classification. The Trustee’s role is akin to that of an expert, and the language of the rule makes it clear that the Trustee’s determination is not to be open to challenge. Once the Trustee has gone through these steps, and the member’s share has been ascertained, the Trustee is under an obligation to add the share to the member’s Accrued Amount. Properly understood, there is no exercise of a discretion at any stage of this process, and this becomes even clearer if one looks at other contexts in the 1976 Deed and Rules where an unfettered discretion is conferred on the Trustee.

Pausing at this point, I will now express my preliminary conclusions on the arguments which I have so far reviewed. I do not pretend to find the question an easy one, and I agree with counsel for Entrust that either interpretation is arguably correct. On balance, however, I find the arguments in favour of the discretionary approach to be distinctly more persuasive. My reasons are briefly as follows.

First, I see considerable force in the linguistic points made by counsel for the employers, and the contrast with the simple mandatory language of the provisions for calculation of a member’s Accrued Amount. Secondly, I see even greater force in the absence of any prescribed criteria for ascertaining a member’s share of actuarial surplus, bearing in mind the inherent imprecision of the concept of surplus in a pension scheme which is still a going concern, and the fact that (as the present case illustrates) actuaries may well use more than one basis of valuation when preparing a report on the funding position of a scheme. So, for example, the 1977 valuation of the Scheme provided a valuation of surplus on a full service basis by reference to target benefits, and a valuation on a discontinuance basis, again by reference to target benefits. The valuations for 1989, 1991, 1994 and 1997 gave an ongoing valuation by reference to past service only, and a discontinuance valuation, again both by reference to target benefits. The 2000 valuation then did the same, but added a minimum funding requirement (“MFR”) valuation based on Accrued Amount benefits.

The great difficulty for the members, it seems to me, is that if a member is to be regarded as having an existing right to an ascertainable share of surplus when he retires, it must be possible to explain what is the correct basis of calculating that share in every case. In other words, if the role of the Trustee is merely to act as a quasi-expert valuer, it must be conceptually clear what it is that the Trustee has to value. This leads on to a related point: if the draftsman of the 1976 Rules had intended the Trustee to use and apply a particular basis of valuation of surplus, why on earth did he not make this explicit? The absence of prescription is in my view much more easily reconciled with the discretionary approach, whereby the choice of appropriate valuation techniques is left to the discretion of the Trustee (no doubt acting on actuarial advice), as is the choice whether or not to award an appropriate share of any surplus so ascertained to retiring members. Furthermore, this absence of prescription is not a peculiarity of the 1976 Rules. It goes back to the very beginning of the Scheme in 1968.

Nor is it an objection to the discretionary approach, in my judgment, that the Trustee sought from the beginning to provide target benefits, and in effect to mimic a final salary scheme. I feel little doubt that, in general terms, this was always the intention of the employers and the Trustee alike, but, as the funding of the Scheme was based on fixed contributions, the provision of target benefits could never be guaranteed. On the available evidence, this was clearly explained to members in the booklets and other literature provided to them, and was also well understood by the Scheme actuaries. If that is right, however, the mere fact that the financial position of the Scheme was healthy enough for this general intention to be fulfilled for nearly forty years cannot make any difference to the basic nature of the entitlement: it remained purely discretionary, and no amount of past practice can turn it into a fixed entitlement.

In the light of these considerations, although I acknowledge that there is considerable force in some of the linguistic and contextual points made on behalf of the members, I cannot regard them as strong enough to prevail. Equally, the fact that the members are not volunteers, and the fact that pension is a form of deferred remuneration, cannot be conclusive. A considerable degree of certainty is anyway provided by the fixed entitlement to Accrued Amounts, while the exercise by the Trustee of its discretionary powers over surplus is regulated by the fiduciary duties which they owe to the members. These duties are reinforced by the word “properly”. In practice, the Trustee could no doubt be relied upon to continue to provide target benefits for as long as the financial position of the Scheme permitted and no preferable alternative presented itself, and this is indeed what happened. It is worth observing, in this connection, that no complaint is made on behalf of the members about the way in which the Trustee has exercised its discretion under rule 11(2) by awarding target benefits, if the correct view is that the Trustee had a discretion to exercise.

Similarly, it cannot be concluded from the mere fact that clearly discretionary language is used elsewhere in the 1976 Deed and Rules that it was not intended to confer a discretion on the Trustee in rule 11(2). It has often been said that one should not hope to find perfect consistency of drafting in a pension scheme, and the present case is certainly no exception. In the course of the hearing I was referred to a number of obvious infelicities of drafting in the various iterations of the Scheme rules. This is not to say that there is no force in the contrast drawn by counsel for the members with other, more obviously discretionary, wording in, for example, rule 21, but merely that the assistance to be gained from such a comparison is likely to be limited, and the critical question must always be how to construe the wording of rule 11(2)(b) itself.

Having stated my preliminary conclusion in favour of the discretionary approach, I will now test it by reference to each of Issues 1 to 5.

Issue 1 presents no problem. If my preliminary conclusion is correct, the answer to Issue 1 is obviously Yes. The Trustee has a discretion whether or not to credit a retiring member with a share of surplus, but is not obliged to do so. I also agree with counsel for the employers that, if the intention had been to require the Trustee to add a share of surplus, this could have been much more simply expressed, and one would not expect to find the words “such an amount (if any) as in the opinion of the Trustee may properly be added thereto as representing”, which to my mind have a strong overall discretionary flavour. I accept that the use of such language could be a pointer, not to the existence of a fiduciary discretionary power (category 2 in Mettoy), but rather to a category 3 power; and this was indeed the analysis which Mr Moeran placed at the forefront of his oral submissions. I would agree with him that this is the most attractive way of putting the argument for the members on this issue, but the fatal objection to it, in my view, is the absence of any prescribed methodology for ascertainment of the retiring member’s share of surplus.

On Issue 2, the position of the employers is again straightforward. If the Trustee decides to credit a member with a share of surplus, the Trustee has a discretion over the quantum of the share, and is not obliged to adopt any particular method of computation. The critical point is that the share of surplus to be awarded to a member is itself the product of a broad discretion vested in the Trustee, and conceptually it had no prior existence. The difficulty for the members, in the absence of any prescribed methodology, is again how to construe rule 11(2) in a way which makes the quantum of the share objectively ascertainable. It is symptomatic of this difficulty that counsel for the members, in the proper discharge of their duty with an issue-based representation order, found it necessary to propound four mutually exclusive methodologies. In descending order of preference, they are briefly as follows. The primary position is that, on the true construction of the rule, there is either no discretion, or any discretion must in practice be exercised, so that the share of surplus is that which would give the member target benefits together with a share of the surplus disclosed by the actuarial reports over target benefits. The secondary position is as above, but with no further share of surplus in addition to target benefits. If that is wrong, the next fall back position is that the share of surplus is one which derives from or represents the member’s Accrued Amount; and if that too is wrong, the final fall back position is that “any discretion in assessing the member’s share of surplus must be exercised on a basis which is rational, consistent between members, and based on and takes into account contributions to the Fund”.

Counsel for the employers submit, and I agree, that the need to identify so many rival candidates for the correct approach to ascertainment of a member’s share of surplus is itself a strong indicator that the approach is misconceived. If it is impossible to say with any certainty what the share of surplus has to be, then it must be discretionary. The fundamental problem with the first two candidates is that they are based on target benefits, a concept which has no place in the 1976 Rules (or, before their adoption, in the 1968 Rules), despite its importance in the practical administration of the Scheme. No reference to “target benefits”, or anything like them, can be found anywhere in the 1976 Deed or Rules; and I am unable to see any legitimate way in which the consistent practice of the Trustee over the years could be of any probative value in construing language which, in its essentials, went back to the establishment of the Scheme in 1968. The fundamental problem with the third and fourth positions, quite apart from their total inconsistency with the first two positions, is that they do not begin to answer the question of how to calculate the share of surplus to be awarded to the member. Simply stating that the share must be one which is derived from, or represents, the contributions paid by or in respect of the member does not tell the Trustee how to calculate the share. In my view it is unnecessary to pursue these points any further. The conceptual difficulties faced by the members on this Issue are, to my mind, a strong indicator in favour of the discretionary approach, and although Mr Moeran deployed his arguments (here as elsewhere) with skill and tenacity, it seemed to me that he had no real answer to the central thrust of the employers’ case.

Before moving on, I should also deal with one discrete aspect of the argument for the members on Issue 2, to the effect that, even if the Trustee did have a discretion, it could as a matter of law be exercised in only one way, because to do anything else would be so unreasonable or capricious that no reasonable trustee could properly so act. An explanatory note at the end of the List of Issues identifies the facts and circumstances relied upon in support of this argument as being (a) that the Scheme was set up with, and was understood to have, the aim and purpose (albeit not a guarantee) of providing target benefits; (b) that the actuarial valuations of the Scheme until 2002 disclosed sufficient available assets to meet target benefits, and until 2005 were carried out with a view to identifying and calculating what target benefits would cost; and (c) that the contributions to the Scheme, from at least the late 1980s, were determined and provided on three different levels which were matched to three different target benefit levels.

In order for this argument to succeed, it would be necessary for the members to demonstrate that the de facto practical administration of the Scheme so as to provide target benefits had over time become so entrenched that the Trustee ceased to have any real choice in the matter and became obliged for the future to exercise its nominal discretion so as to achieve the same result. In my judgment this is an ambitious contention, which the evidence before me does not begin to bear out. The argument seeks to elevate the consistent past practice of the Trustee in the exercise of its discretion (by granting target benefits out of surplus, when this was considered to be affordable) into a fetter on the future exercise of the discretion, even in radically changed circumstances. Such a result would be inconsistent with any conventional analysis of the Trustee’s fiduciary obligations, and would also be inconsistent with the clear understanding conveyed to members that they had no entitlement to target benefits. I am therefore satisfied that this alternative way of putting the members’ case is untenable. For similar reasons, I am also unable to accept a variant of the same argument, based on the maxim that equity treats as done that which ought to be done, which was raised for consideration by counsel for Entrust in the course of the hearing, and was adopted by Mr Moeran in his closing submissions.

I can deal briefly with Issue 3, because it raises essentially the same questions as Issue 2. As before, the answer on the discretionary approach is straightforward. Once the Trustee has determined that a member should be credited with a given share of surplus, the Trustee still retains a discretion over the amount to be credited to or in respect of the member, and is not obliged to adopt any particular method of computation. The necessary protection for the member lies, not in the existence of any fixed entitlement or any mandatory approach to questions of computation, but rather in the Trustee’s obligation to act fairly and in accordance with its fiduciary duties, having regard to all the relevant circumstances. All attempts on behalf of the members to erect a higher hurdle than this, or to define some particular method of computation as a matter of construction, seem to me to face the same obstacles as those which I have already discussed under Issue 2.

I can also deal briefly with Issue 4, because the parties are agreed on the correct answer to it. The principles which should govern the Trustee’s exercise of discretion, if the Trustee has a discretion, are that the Trustee’s power should be exercised (a) in good faith, with the Trustee giving genuine and responsible consideration to the exercise of its power; (b) for the purpose for which it is given; and (c) giving proper consideration to the matters which are relevant and excluding from consideration those which are irrelevant. In this respect, the Trustee should make proper enquiries to inform itself about the matters which are relevant to the exercise of the discretion. See generally Edge v The Pensions Ombudsman [2000] Ch 602 at 607 C-F per Chadwick LJ delivering the judgment of the Court of Appeal (the other members of which were Peter Gibson and Ward LJJ).

Finally, I come to Issue 5, which raises the question of how surplus is to be determined by the Trustee. Even on the discretionary approach, it is necessary to have an answer to this question, because in order to exercise its discretion the Trustee must first ascertain whether there is any actuarial surplus in the Fund. Once again, however, this question is much easier to answer on the discretionary approach. According to the employers, it is for the Trustee, rather than anyone else, to determine what surplus, if any, exists within the Fund; and since rule 11(2) does not specify any method by which surplus is to be calculated, the Trustee may do so on any reasonable basis, which will in practice almost certainly, if not invariably, involve consultation with an actuary (probably, but not necessarily, the Scheme actuary appointed pursuant to clause 21 of the 1976 Deed). It is not surprising, say counsel for the employers, that there is no express provision in rule 11(2) prescribing the basis on which surplus is to be calculated, precisely because actuarial methods and assumptions are likely to change over time in the light of changing financial and demographic conditions, so it is only to be expected that the choice of valuation method, and the underlying assumptions, would be left open, affording the necessary flexibility to the Trustee and the actuary it consults to use the method and assumptions most appropriate to the prevailing circumstances. In practice, this is likely to mean that the Trustee would use the most recent triennial valuation, obtained pursuant to clause 25 of the 1976 Deed, in order to ascertain the surplus, but there is nothing in the wording of the 1976 Rules which compels the Trustee to do so. Similarly, if the Trustee does decide to use the most recent valuation report, it may, but is not obliged to, consider alternatives.

The members’ primary case on Issue 5 is that determination of the surplus under rule 11(2) is in principle a matter for the Trustee, but in exercising its power it must take actuarial advice (save, possibly, in extreme circumstances where it is plain on any basis that there is no surplus in the Fund). Otherwise, it would be impossible to say that the “actuarial” surplus referred to in the rule had been identified. In practice, this is said to mean that the Trustee is obliged to use the most recent actuarial valuation of the Scheme prior to the date of exercise of the power, updated where appropriate to the date of exercise. This is required either as a matter of construction, or because in all the circumstances the Trustee could not properly proceed in any other way.

I find this argument unconvincing, whichever way it is put. As a matter of construction, I am unable to spell out of the 1976 Deed or Rules an obligation on the Trustee to use the most recent actuarial valuation of the Scheme, although I readily accept that in practice this was likely to be what would happen. It is not as though the regular actuarial valuations could serve no other purpose, because (for example) they would help the Trustee to monitor the general solvency of the Scheme and the appropriateness of the current investment strategy. Further, it is not self-evident that the assumptions used in a triennial valuation would necessarily be appropriate for the purpose of deciding what, if anything, should be treated as a member’s share of surplus under rule 11(2). For example, at a time of financial uncertainty the Trustee may reasonably wish to adopt an extremely cautious approach, on a more conservative basis than is reflected in the most recent valuation, when deciding what share (if any) of surplus to allocate to a retiring member, with the effect of locking in the entitlement thus conferred on the member for the rest of the member’s life. A yet further problem is that the most recent actuarial valuation is likely to be made on more than one basis, but neither expressly nor by necessary implication does rule 11(2) tell the Trustee how to choose between them. Counsel for the members argued that in such circumstances the Trustee must, as a matter of fact, use “the target benefits surplus calculation”, but I agree with counsel for the employers that all this does, at best, is to identify the liabilities to be valued. It does not answer the critical question of how the valuation is to be performed.

As to the alternative way of putting the members’ primary case under Issue 5, it seems to me to face the same insuperable objections which I have already considered in relation to Issue 2. In my view the evidence falls well short of establishing that the Trustee had a discretion which in law it could exercise in only one way.

The members’ fall back position on Issue 5 is put as follows in the List of Issues:

“Given the use of the word “surplus” there must be a calculation by reference to excess of assets over liabilities. The only liabilities it is reasonable to look at (other than target benefits) is Accrued Amount … and the “surplus” must therefore be the excess of assets over this sum. This is not a matter for discretion, but a mathematical exercise.”

The difficulty with this approach, however, is that it is too simplistic and does not go far enough. Ascertainment of a member’s Accrued Amount down to the date of retirement is indeed a relatively simple mathematical exercise, which would not require the assistance of an actuary. But in order to place a value, for Scheme funding purposes, on a member’s Accrued Amount (for the future as well as the past), it is necessary to make various actuarial assumptions, for example about the rate of future investment return. It is only when this has been done that the extent of any actuarial surplus can be ascertained. Once again, however, rule 11(2) is silent about the assumptions which are to be adopted, and I am unable to see how they can be spelt out of past practice, whether as a matter of construction, or as a guide to the exercise of discretion which the Trustee is in practice obliged to adopt.

I agree with the submission of counsel for the employers that it is important not to lose sight of the essential nature of the task confronting the Trustee under rule 11(2)(b). In that context, “surplus” must mean an excess of assets over the amount needed to meet existing liabilities. Existing liabilities are pensions currently in payment, which of course include all pensions validly augmented to target benefit levels in the past, and, for members who have not yet retired, pensions based on each member’s Accrued Amount. The Trustee’s task is to consider whether to grant additional benefits in excess of the Accrued Amount. In order to know whether it would be appropriate to do this, the Trustee will need to know whether the assets of the Scheme are, and are predicted to remain, sufficient to meet the existing liabilities. For members who have not yet retired, that means benefits based on each member’s Accrued Amount. Accordingly, the members’ fall back case is correct to the extent that, for members who have not yet retired, the relevant liabilities to take into account in deciding whether there is any actuarial surplus out of which more generous benefits might be granted are the Accrued Amounts for each member. But that is not the end of the exercise, because the Trustee will then have to decide, taking such actuarial advice as it considers appropriate, how to value the assets and liabilities in question, which is in turn likely to involve making certain assumptions. The basic question raised under Issue 5 is whether the selection and making of those assumptions are left to the discretion of the Trustee, or whether they are in some way prescribed by the 1976 Rules. I accept the employers’ submission that, as a matter of construction, the choice of assumptions is left to the discretion of the Trustee, acting on appropriate actuarial advice.

Having now tested and considered my preliminary conclusion in the light of the detailed questions raised by Issues 1 to 5, I find that it provides a reasonably workable and coherent framework within which the Trustee could discharge its functions under rule 11(2), whereas the arguments for the members give rise to a range of conceptual and practical difficulties to which I can see no satisfactory solution. My final conclusion, therefore, is that Issues 1, 2, 3 and 5 should be answered in the sense contended for by the employers.

Issues 6 to 10 ask the same questions about rule 11(3) of the 2005 Rules as Issues 1 to 5 ask about rule 11(2) of the 1976 Rules. I have set out the main relevant provisions of the 2005 Rules in paragraphs 36 to 39 above. Despite some minor differences in wording, the general effect of the relevant provisions is the same as that of their counterparts in the 1976 Rules, and the critical wording in rule 11(3) (“and such amount, if any, as in the opinion of the Trustee may properly be added thereto as representing the retiring Member’s share of any actuarial surplus arising in the Plan as to which the decision of the Trustee shall be final and shall not be questioned”) is in all material respects identical to rule 11(2)(b) of the 1976 Rules. It is therefore common ground that Issues 6 to 10 should be answered in the same way as Issues 1 to 5, and neither side directed any separate argument to them. I agree, and my answers to these Issues will therefore be the same as my answers to Issues 1 to 5.

This Issue arises in respect of members who left service in the Scheme with a deferred pension before 1 October 2005, when the 2005 Rules came into force, but who have retired after 10 July 2006, or will do so in the future. The basic question is whether the Trustee was under a duty to consider allocating a share of surplus to these members when they left service. If the Trustee was under such a duty, it is clear that the relevant members should have been considered for allocation of a share of surplus, at least if the Scheme was in surplus at the time. If, however, the duty to consider allocation of a share of surplus to deferred members does not arise until they retire, it is equally clear that since July 2006 there has been no surplus in existence out of which an allocation could be made.

This basic question is the subject of Issue 12(1):

“12(1): Did the Trustee’s duty to add an amount in respect of a Member’s deferred pension pursuant to Old Rule 19(1) fall to be performed (a) at the date the Member left service or (b) at the date when he started to draw his pension?”

If the answer to the above question is that the duty fell to be performed at the date of leaving service, three further questions arise, as follows:

“12(2): If the answer to (1) is (a), did the Trustee’s duty to ascertain the level of pension payable from the Fund in respect of the Member’s pension capital fall to be performed (i) on the date the Member left service or (ii) on the date the Member retired?

12(3): If the answer to (1) is (a) how, if at all, is interest to be added to the Member’s pension capital in the period between leaving service and retirement?

12(4): If the answer to (1) is (a) and (2) is (i), at the date the Member leaves service is the Trustee obliged to re-compute pension in the event of the Member taking it before normal retirement date under Old Rule 19(3)?”

In the interests of clarity, I should explain that the reference in Issue 12(1) to “the Trustee’s duty to add an amount in respect of a Member’s deferred pension” is meant to refer to addition of a share of surplus pursuant to rule 11(2) of the 1976 Rules, as applied by rule 19(1). Further, in the light of my answer to Issue 1, the duty would be more accurately described as a duty to consider adding a share of surplus.

I have set out the relevant parts of rules 18 and 19 of the 1976 Rules in paragraph 24 above. For present purposes, the most important provisions are those contained in rules 18(2) and 19(1):

“18(2) … a member who before becoming entitled to an immediate pension out of the Fund leaves the Service whether voluntarily or involuntarily but in respect of whom a transfer payment is not made pursuant to sub-Rule (1) hereof … shall (unless his contributions have been returned to him pursuant to sub-Rule (3) hereof) be granted a deferred pension out of the Fund calculated as provided by Rule 19(1) and (2) payable at such time and on such terms as are specified in the Rules …

…

19(1) A deferred pension granted out of the Fund pursuant to Rule 18 shall be calculated in the manner set out in Rule 11 based on the Member’s Pension Capital at the date of leaving the Service.”

“Pension Capital” is defined in rule 1(c) as having “the meaning ascribed to it in Rule 11 calculated to the relevant date as specified in any Rule”, unless that meaning is “inconsistent with the subject or context” (see the opening words of rule 1(c)). In rule 11(2), it will be remembered, the term “Pension Capital” is used to describe the total amount ascertained in accordance with the provisions of that rule, i.e. a retiring member’s Accrued Amount plus any amount added thereto in respect of actuarial surplus.

In the light of these provisions, the argument for the members on Issue 12(1) is simplicity itself. The duty to consider adding a share of surplus must be performed when the member leaves service, because that is precisely what rule 19(1) requires. The words “at the date of leaving the Service” unambiguously designate that date as the one at which the calculation of the member’s Pension Capital has to be performed. Furthermore, it is impossible for a member to have a Pension Capital at the date of leaving service unless it is ascertained pursuant to rule 11(2) at that date, because the concept of Pension Capital is by definition the product of the process of ascertainment carried out by the Trustee under rule 11(2). If (as I have held) the addition of a share of surplus depends upon an exercise of discretion by the Trustee, it is obvious that no member can have a Pension Capital, in the defined sense, unless and until the Trustee has exercised its discretion over surplus, whether by deciding to make an addition to the member’s Accrued Amount, or by deciding not to do so.

Counsel for the employers, however, argue that the position is not as straightforward as it might at first blush appear, and for a number of reasons the construction of rule 19(1) which best accords with business sense is that it is primarily concerned with methodology, and the question of adding a share of surplus can only sensibly be addressed when the member retires. They do not shrink from the submission that there is a clear mistake in the drafting of rule 19(1), and that the reference to Pension Capital should be interpreted in this particular context as a reference to only the contributions element of Pension Capital, i.e. before the addition of interest and any share of surplus.

In support of this argument, reliance is placed on the following main points.

First, it is argued that, when a member leaves service, he becomes entitled, subject to meeting certain criteria, to a pension payable from a future date: rule 18(2). When the member leaves service, he has no right to present payment of the pension. He has only a right to future payment of a pension in an as yet unascertained amount. His right to a pension vests in interest, but not in possession. The date upon which the pension becomes payable (i.e. vests in possession), and the method for calculating the amount of the pension, are provided for by other rules: see rules 19(3), and 19(1) to (2), respectively.

Secondly, the words “in the manner” in rule 19(1) show that it is essentially concerned with the method of calculating a deferred member’s pension. The methodology to be adopted is that set out in rule 11. But rule 11 applies only when a member retires, and the calculation which it requires is dependant upon the member having retired.

Thirdly, the amount of a member’s pension will depend upon a number of factors which will not be known with certainty until the member actually retires. The obvious example is the date of retirement itself. Under rule 19(3), there are three possible dates upon which a member’s pension might become payable: (a) normal pension age; (b) a date requested by the member, if the Trustee is satisfied that the member is permanently incapacitated; and (c) after attaining the age of 50, at the member’s request. There is no way of knowing in advance which of these dates will apply, so for this reason alone it is impossible for the amount of a deferred member’s pension to be calculated at the date of leaving service.

Fourthly, rule 19(4) provides that, on the death of a deferred member before the commencement of his pension, the Trustee is to raise out of the Fund and hold on discretionary trusts “a sum equal to the Member’s Pension Capital calculated in the manner set out in Rule 11 as at the date of his death”. This provision clearly requires a calculation of Pension Capital as at the date of the member’s death. However, such a requirement would be strange and wasteful if a calculation of Pension Capital had already been made when the member left service; nor is there any indication of how the calculation on the member’s death should relate to the earlier one performed when he left service.

Fifthly, it is anyway not the case that a member becomes entitled to a deferred pension immediately upon leaving service. Rule 18(1) gives a member leaving service the right to require a transfer payment out of the Scheme if, within 12 months after leaving service (or a longer period approved by the Trustee), he takes up employment with another employer. The right is exercisable by notice in writing to the Trustee. There will therefore be a period of at least 12 months after the member leaves service when his entitlement to a deferred pension within the Scheme will remain in doubt. Similarly, rule 18(3) confers upon deferred members who have remained unemployed for three months or more after leaving service a right to a return of contributions plus interest, provided that within the first three months the member may elect to take a deferred pension in lieu of a return of contributions. Again, therefore, the position may remain unclear for three months after the date of leaving service.

Sixthly, if a deferred member’s Pension Capital has to be calculated at the date of leaving service, it follows that his Accrued Amount also has to be calculated at that date, and compound interest on his contributions under rule 11(2)(a) would cease running for the future. There is no provision in rules 11 or 19, or anywhere else in the 1976 Deed or Rules, for revaluation of deferred pensions during the period of deferment. As a result, a member who left service would apparently enjoy no protection against the effects of inflation during the period between his departure and his eventual retirement, which could be many years later. Further, it does not detract from the force of this point that statutory provisions for the revaluation of deferred benefits were introduced with effect from 1 January 1986 by the Social Security Act 1985. No such statutory provisions were in force when the 1976 Rules came into operation, and one would therefore expect the 1976 Rules to have made express provision for the continuing accrual of compound interest on the member’s contributions, or for the revaluation of his pension capital, if it was really intended that his pension capital should be ascertained when he left service.

Seventhly, the effect of requiring the Trustee to ascertain a deferred member’s share of surplus many years before his eventual retirement would be to make it impossible to take account of changes in financial circumstances during that period. It is improbable that this was intended, and greater flexibility would be preserved by postponing the calculation of Pension Capital until the member’s retirement. Furthermore, ascertainment at the date of leaving service could operate unfairly in relation to the deferred member if there was a deficit at that date, but a surplus at the date of retirement (generated in part by the investment return on his contributions). If the construction now contended for by the members is correct, a member in that position would not be entitled to share in the surplus when he retired. Given that both surplus and deficit are notional concepts based upon assumptions about future events, a more accurate financial picture is likely to be obtained by leaving the ascertainment of a deferred member’s Pension Capital until the latest possible date, namely the date of his retirement. It is true that there are some circumstances under the 1976 Rules where a deferred member’s pension does have to be calculated at a date earlier than retirement, namely where the member requests a transfer payment (rules 18(1) and 25), and where the member dies during the period of deferment (rule 19(4)). However, in those situations there is no choice about the date, and there is no practical alternative to ascertaining the member’s entitlement at the date of transfer or the date of his death respectively. By contrast, there is no need to ascertain the pension benefits to be paid to a living deferred member who remains in the Scheme before the first date when they become payable, namely upon retirement.

In their written submissions, counsel for the employers also sought to place some reliance on the leaving service provisions in the 1968 Rules, contending that they were admissible as aids to construction of the 1976 Rules, and that the 1976 Rules cannot have been intended to introduce any material changes to the relevant provisions given the limited reasons for the introduction of the 1976 Deed and Rules specified in recital (D) to the 1976 Deed. In his oral submissions, however, Mr Evans accepted that the relevant provisions of the 1968 Rules were not wholly clear, and that little help could be gained from a comparison with them. I agree, and I am also not satisfied that the changes introduced in 1976 were intended to be confined to those signalled in recital (D) to the 1976 Deed, not least because recital (C) says that “The Trustee desires that … certain amendments shall be made to the Group Scheme for inter alia the purposes set out in Recital (D) below” (my emphasis). It is clear, therefore, that the purposes set out in Recital (D) were not intended to be exhaustive. It cannot be inferred that no change of substance was intended to the leaving service provisions, especially since they were substantially recast.

The most that can be said, in my view, is that the 1968 Rules did not explicitly require a deferred member’s pension benefits to be ascertained at the date of leaving service, while rule 22 said that a member’s deferred pension should be based on the full amount of his actuarial interest in the Fund and should be of :

“such an amount as shall be certified by the Actuary to be appropriate having regard to such Member’s age and the amount of his contributions with interest and any other relevant facts and so that the Actuary’s decision shall not be questioned by any person in any circumstances.”

Counsel for the employers submitted that the passage which I have quoted applied only to members who left service on becoming permanently incapacitated, but it seems to me that the passage was intended to be of general application to all deferred pensions to which a member had become entitled pursuant to rule 21. Clearly, that machinery was very different from rule 19 of the 1976 Rules, and I am therefore doubtful whether any assistance can be gained from it beyond the negative point that there was no mandatory requirement for deferred pension to be ascertained at the time of leaving service.

As to the wording of rule 19(1) itself, Mr Evans emphasised two key points in his oral submissions. First, the rule does not state explicitly when the calculation has to be carried out, and its focus is instead on how the calculation should be performed. Secondly, the words “based on the Member’s Pension Capital at the date of leaving the Service” cannot be read literally, because Pension Capital is by definition the end result of the process of computation under rule 11, and it therefore makes no sense to say that the computation must be based upon it. He went on to submit that something must have gone wrong with the drafting of rule 19(1), and that the appropriate remedy is to read “Pension Capital” as referring to the component of Pension Capital which necessarily came to an end when the member left service, namely the contributions made by him or on his behalf to the Fund. On this interpretation, the calculation of deferred pension should not take place until the member retires, and it will then be “based on” his contributions up to the date of leaving service, plus interest under rule 11(2)(a) down to the date of retirement, and (if appropriate) a share of surplus as at the date of retirement. Mr Evans submits that the error in the drafting of the rule is capable of correction as part of the process of construction, because it is clear that something has gone wrong with the language, and it is also clear what a reasonable person would have understood the parties to the 1976 Rules to have meant: see Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] 1AC 1101, at [14] to [26] per Lord Hoffmann.

In his oral submissions on behalf of the members, Mr Moeran drew particular attention to the wording of rule 18(2) and submitted that it clearly contemplates the immediate grant of a deferred pension when a member leaves service and satisfies the relevant conditions. He emphasised the desirability, from the departing member’s point of view, of knowing where he stood, and the amount of his deferred pension, on leaving service. Mr Moeran also drew attention to rule 25(1)(e), which prescribes the transfer value payable when a member leaves service and requests a transfer payment to another scheme:

“The amount of the transfer payment shall be equal to the Member’s actuarial interest in the Fund or such other amount as the Trustee may approve on Actuarial advice having regard to any special transfer arrangements applicable to the Transfer Scheme involved.”

Mr Moeran submitted that the calculation of a member’s “actuarial interest in the Fund” under rule 25(1)(e) was intended to be the same as the calculation required under rule 19(1), and that the calculation would in effect serve a dual purpose: if the member chose to request a transfer, it would quantify the amount of the transfer payment, while if he chose to remain within the Scheme, it would be used to quantify his deferred pension.

As with Issues 1 to 5, I propose to state my provisional conclusion on Issue 12(1) and then, if that conclusion is in favour of the members, to test it by reference to the specific questions raised in Issues 12(2) to (4).

Despite the considerable force, both separately and cumulatively, of many of the arguments advanced by counsel for the employers, and despite my feeling that it would in many ways have been more sensible if consideration whether to award a share of surplus had been left until the date of a deferred member’s retirement, my provisional view is that the submissions for the members on this issue are clearly to be preferred. The natural meaning of rules 18(2) and 19(1) is that, subject to certain exceptions, leaving service triggers an immediate entitlement to a deferred pension, the amount of which is to be calculated by reference to the member’s Pension Capital at that date. This is the natural meaning of the language used, and it is reinforced by the definition of Pension Capital in rule 1(c). The relevant date for the calculation of Pension Capital specified in rule 19(1) is clearly the date of leaving service: that is, in terms, what rule 19(1) says. I do not accept the submission that rule 19(1) is concerned only with the method of calculation. It seems to me to be concerned with both the method and the date of calculation.

In the light of this clear and (to my mind) unambiguous language, it would require powerful considerations to persuade me that a mistake was made in the drafting of rule 19(1) and that Pension Capital should not be given its normal meaning. The points urged on me by counsel for the employers seem to me to fall well short of what would be needed to justify such radical surgery. The preferable approach, in my view, is to take the starting point which is apparently so clearly mandated by rule 19(1), and then see if, with common sense and appropriate adjustments, it can be made to work in practice. It is only if the machinery is clearly unworkable, or if it would give rise to consequences that could never sensibly have been contemplated, that a departure from the apparently clear language of the rule could be justified.

Adopting this approach, I am not satisfied that any of the objections raised by the employers would make the rule 19(1) machinery unworkable, or that their consequences are so drastic as to show that something different must have been intended. So, for example, the point that it cannot be known with certainty for 12 months whether a departing member will require a transfer payment could be accommodated in most cases by simply asking the member what his plans are, and if necessary deferring the calculation until the position has been clarified. Again, the provisions of rule 19(4) clearly require a calculation of Pension Capital to be made when a deferred member dies before receiving a pension, but there is no insuperable difficulty in the view that such a calculation must supersede the calculation of Pension Capital made when the member left service. The earlier work will in a sense be wasted, but the situation is one which will not often arise, and although the member will never receive his deferred pension, he will have had the assurance of knowing what it would be had he lived, and he may well have planned for his future on that basis. Similarly, I am not persuaded that it is impossible to calculate a member’s deferred pension merely because the date of his future retirement is uncertain. The calculation can be made on the assumption that the member will retire when he reaches normal pension age (which is not capable of extension under the Scheme). If he then retires early, or becomes incapacitated, suitable adjustments can be made to cater for the accelerated receipt of the pension. Nor am I persuaded that no answer can be found to the problem of revaluation, or interest on contributions, between leaving service and retirement. I will return to this point under Issues 12(2) and (3) below, but, to anticipate, the simplest solution is that the member’s Pension Capital as at the date of leaving service should be immediately applied in granting the member a deferred pension from normal pension age.

Having stated my preliminary conclusion that Issue 12(1) should be resolved in favour of the members, I now turn to Issues 12(2) to (4). The question raised by Issue 12(2) is whether the Trustee’s duty to translate a deferred member’s Pension Capital at the date of leaving service into a pension fell to be performed on the date the member left service, or on the date of the member’s retirement. In the light of the construction which I have placed on rules 18(2) and 19(1), the answer to this question appears to me to be clear. Upon leaving service a member is entitled to be granted a deferred pension based on his Pension Capital at that date. Although the ascertainment of the member’s Pension Capital, and its translation into the grant of a pension, could in theory take place at different dates, such a procedure would cause great administrative difficulties, and none of the parties displayed any enthusiasm for it. By contrast, I see no insuperable practical difficulties in granting a deferred pension when the member leaves service, based on the assumption that he will retire at normal pension age. I did not understand counsel for the employers to argue that this would be impossible, either conceptually or in practice. Their point was, rather, that a number of other assumptions would have to be made, and kept under review, to such an extent that the practical outcome would be little different from postponing the grant of the pension until the member’s retirement. That may well be so, but it does not make the exercise impossible, and on any view this is a pension scheme where much of the administrative detail was left to the Trustee to work out.

Issue 12(3) asks how, if at all, interest is to be added to the member’s pension capital in the period between leaving service and retirement. The answer to this question is in practice bound up with the answer to the question which I have just discussed. Counsel for the members argue, and I agree, that when a deferred pension is granted as at the date of leaving service, appropriate annual increases must in principle be built in to compensate the member for the deferred receipt of the pension. They explain this point with a simple example. If £10 purchases a £1 per annum annuity for a 60 year old, then it will purchase a larger annuity payable at 60 for a 40 year old. There are two reasons for this: first, the £1 will earn interest in the intervening 20 years; and, secondly, there is a chance that the 40 year old will die before reaching 60. The appropriate amount of the increases would be a matter for the Trustee to determine, with such actuarial advice as it thinks fit. In practice, this question does not arise for members whose pensions were in fact fixed when they left service, because statutory revaluation provisions apply: but appropriate increases would in my judgment be necessary for members whose pensions were not fixed when they left service. This solution is in my judgment preferable to saying that the member’s contributions down to the date of leaving service will continue to earn compound interest at the rule 11(2) rate. There are at least two difficulties with that approach. First, the calculation of a member’s Accrued Amount cannot be postponed beyond the date of leaving service, and there is no provision in the 1976 Rules for continuing to add interest on contributions after that date. Secondly, and in any event, this solution could not apply to any share of surplus included in the member’s Pension Capital.

Issue 12(4) asks whether the Trustee is obliged to recompute the deferred member’s pension if he decides to take it before normal retirement date pursuant to rule 19(3). This Issue is no longer in dispute. Although the 1976 Rules make no express provision for the scaling down of a deferred pension in these circumstances, to allow for its accelerated receipt, it is common ground (and I agree) that it can never have been contemplated that a member could obtain a windfall benefit of this kind by opting for early retirement. The appropriate rates of reduction for early payment must again be a matter for the Trustee to determine, with actuarial advice.

Having now reviewed Issues 12(2) to (4), I see no reason to depart from my provisional conclusion on Issue 12(1). I therefore answer Issue 12(1) in sense (a), and Issue 12(2) in sense (i). My answer to Issue 12(3) is that interest, as such, is not to be added to the member’s Pension Capital, but appropriate annual increases should be built into the deferred pension awarded to a member when he leaves service. My answer to Issue 12(4) is Yes.

Issue 13 is framed as follows:

“13.

In the event that the Trustee has failed to perform its duty under Old Rule 11(2) in respect of one or more Members who left service before 1 October 2005, does the Trustee still remain under a duty to comply with Old Rule 11(2)?”

The question has to be answered in the light of my decisions on the earlier Issues, including in particular my decision on Issue 1 that the power to make an award of surplus under rule 11(2) of the 1976 Rules was discretionary in nature, and my decision on Issue 12(1) that the Trustee’s duty to consider making such an award in favour of a deferred member arose when the member left service, not on the member’s subsequent retirement.

There can be no doubt that the rule 11(2) power was a fiduciary one, which was vested in the Trustee for the benefit of the members and which the Trustee could not have released. On any view, it formed a central part of the benefit structure under the Scheme, and it was the means whereby the Scheme was in fact able to provide target benefits for retiring members from 1968 until 2006.

In these circumstances, a convenient starting point for consideration of this question is provided by the observations of Park J in Breadner v Granville-Grossman [2001] Ch.523 at 540:

“50.

It is trite law that there is a distinction between two kinds of dispositive discretion which may be vested in trustees. There are discretions which the trustees have a duty to exercise (sometimes called “trust powers”) and discretions which the trustees may exercise but have no duty to exercise (sometimes called “mere powers”). The distinction is most familiar in the context of discretions to distribute income. In cases of trust powers the trustees are bound to distribute the income, but have a discretion as to how it should be divided between the beneficiaries. In cases of mere powers the trustees have two discretions: first, a discretion whether to distribute the income or not; and second, if they decide that they will exercise the first discretion, a further discretion as to how to divide the income between the beneficiaries. In the latter kind of case there will usually be a default trust which deals with the income if the trustees do not exercise their discretion to distribute it. Typically the default trust will provide for the undistributed income to be accumulated or to be paid as of right to a beneficiary whose interest in it is vested but defeasible by the trustees exercising their discretion to distribute.

51.

The distinction is explained by Lord Upjohn in In re Gulbenkian’s Settlements [1970] AC 508, 525 and illustrated by In re Locker’s Settlement [1977] 1 WLR 1323 (a trust power case), and In re Allen-Meyrick’s Will Trusts [1966] 1 WLR 499 (a mere power case).

52.

Sometimes the distinction does not matter, but there is an important difference between the two kinds of case if the trustees do not exercise the discretion to distribute income within the normal time for exercising it. That time is usually “a reasonable time”. If there is a trust power and, although the trustees are required to exercise it within a reasonable time, they do not do so, the discretion still exists. If the trustees are willing to exercise it, albeit later than they should have done, the court will permit them to do so. That is what happened in In re Locker’s Settlement. Alternatively the court will exercise the discretion itself. But if the discretion to distribute is a mere power, and the trustees do not exercise it within a reasonable time of the receipt of an item of income, the discretion no longer exists as respects that income. The default trusts take effect indefeasibly. That is what happened in In re Allen-Meyrick’s Will Trusts.”

The passage from the speech of Lord Upjohn in Gulbenkian, to which Park J referred in paragraph [51], reads as follows (at 525B):

“Again the basic difference between a mere power and a trust power is that in the first case trustees owe no duty to exercise it and the relevant fund or income falls to be dealt with in accordance with the trusts in default of its exercise, whereas in the second case the trustees must exercise the power and in default the court will. It is briefly summarised in Halsbury’s Laws of England, 3^rd ed., vol. 30 (1959) para. 445: “The court will not exercise or compel trustees to exercise a purely discretionary power given to them; but the court will restrain the trustees from exercising the power improperly, and, if it is coupled with a duty, the court can compel the trustees to perform their duty.” It is a matter of construction whether the power is a mere power or a trust power and the use of inappropriate language is not decisive: Wilson v Turner (1883) 22 Ch. D. 521, 525.”

In In re Allen-Meyrick Will Trusts a testatrix had directed her residuary estate to be held by her trustees “upon trust that they may apply the income thereof in their absolute discretion for the maintenance of [her husband] and subject to the exercise of their discretion upon trust for [her two godchildren] in equal shares absolutely”. The husband was an undischarged bankrupt, and although the trustees had made certain payments to him they were unable to agree how far they should use the trust income to support him. They asked the court to rule on a number of questions, including whether the trust declared by the will was a discretionary trust in favour of the husband and the godchildren, or whether it conferred a mere power on the trustees to pay income for the husband’s maintenance subject to which the income ought to be held for the godchildren; and, in the latter case, for how long the power remained exercisable. Buckley J held that the latter construction of the will was correct, and that after a reasonable period had elapsed from receipt of the trust income the trustees’ power ceased to be exercisable. The judge’s reasoning on the second point appears from this passage at 505A:

“So also I think that in the present case it is incumbent upon the trustees to make up their minds as income becomes distributable from time to time to what extent they will apply it for the maintenance of [the husband], and to the extent that they do not decide so to apply it the trust for the benefit of [the godchildren] attaches to the fund and it becomes theirs.

As I have said, the trustees have been making certain payments for the benefit of [the husband] and have accumulated in their hands a certain fund of undistributed income. So far as a reasonable period after receipt of any part of that accumulated fund has elapsed, I think that the trustees’ discretion in respect thereof must be treated as being at an end.”

In In re Locker’s Settlement, on the other hand, the income of the trust fund was distributable at the absolute discretion of the trustees among a wide class of individual, charitable and other institutional beneficiaries, but the trustees, paying, as Goulding J said, “too deferential a respect to the settlor’s subsequent wishes”, had omitted to distribute any income for several years, and in 1977 they sought the directions of the court as to whether they could still exercise their own discretion over a fund representing income from December 1965 to April 1968. After recording that it was common ground that it was the duty of the trustees to distribute the trust income within a reasonable time after it came into their hands, and that the court could, if necessary, execute a discretionary trust of that nature if the trustees failed to do so, the judge continued at 1325F:

“From that basis let me first express my own views as a matter of elementary principle and then see how far they can stand in the face of reported authorities. A court of equity, where trustees have failed to discharge their duty of prompt discretionary distribution of income, is concerned to make them as owners of the trust assets at law dispose of them in accordance with the requirements of conscience; that is, to give benefits to the cestuis que trust in accordance with the confidence that the settlor reposed in them, the trustees. In a case such as the present, where the trustees desire to repair their breach of duty, and to make restitution by doing late what they ought to have done early, and where they are in no way disabled from doing so, the court should, in my judgment, permit and encourage them to take that course. A tardy distribution at the discretion of the trustees is, after all, nearer to prompt distribution at the discretion of the trustees, which is what the settlor intended, than tardy distribution by the trustees at the discretion of someone else … That being my view of the matter on first impression and first principle, I turn to the arguments against it.

It has been argued by Mr Blackburne, on behalf of some of the objects of the trust, that once a reasonable time for the distribution of a particular item of income has elapsed, the trustees’ discretion over that income is extinguished and either cannot be revived, or ought not to be revived, by the court. That submission is founded on In re Allen-Meyrick’s Will Trusts [1966] 1 WLR 499 and In re Gulbenkian’s Settlements (No.2) [1970] Ch. 408. They, however, concerned permissive, as distinct from obligatory, discretionary powers, and in each case the trust instrument contained a subsisting trust to take effect in default of exercise of the power. The discretion of the trustees ought to be exercised promptly, if at all, where its exercise is optional, just as it ought to be exercised promptly in every case where its exercise is obligatory. But the consequences of non-exercise are to my mind quite different in the two situations. In the cases cited, failure to exercise the permissive power within the proper limits of time left the default trust standing. In the case of an obligatory power (in other words, a compelling trust to distribute), the failure to execute the trust promptly is an unfulfilled duty still in existence. Therefore, as it seems to me, the Allen-Meyrick and Gulbenkian cases do not carry me any further. It follows from Lord Wilberforce’s observations in the Baden case [1971] AC 424, that if the court appoints new trustees to remedy their predecessors’ default, they, the new trustees, can execute the neglected discretionary trust (being one of a mandatory character). A fortiori, it seems to me, the court can permit the existing trustees, if willing and competent to do so, to repair their own inaction.”

In Breadner v Granville-Grossman, the trustees of a discretionary family settlement made in 1973 made an appointment of the fund in 1976 on accumulation and maintenance trusts in favour of the settlor’s son, Jonathan, and Jonathan’s three cousins. Under the 1976 appointment each of the four principal beneficiaries had a life interest in a quarter share of the fund, subject to a power conferred on the trustees to override those trusts by a further deed of appointment executed before a “closing date” defined as one day before the day on which the first of the four children should attain 25. That date was 2 August 1989, the day before the 25^th birthday of the oldest of the cousins. In default of exercise of the overriding power, the fund was to be held in trust for Jonathan and his cousins. The last date for exercise of the overriding power was accordingly 1 August 1989, but it was not until the closing date itself that the trustees met, discussed the matter, and by a deed made in purported exercise of the overriding power appointed Jonathan and his children as the sole beneficiaries of the fund. In those circumstances, the trustees asked the court to determine whether the appointment of 2 August 1989 was valid. A number of arguments in favour of the appointment’s validity were advanced, but they were all rejected by Park J. For present purposes, the argument which matters was to the effect that the trustees had a fiduciary duty to consider exercise of the power before its expiry, and if they had considered the question in time they would certainly have decided to make an appointment in favour of Jonathan and his children. Thus, it was said, the court should permit the trustees to do late that which they should have done in good time. After making the general observations which I have quoted in paragraph 111 above, Park J rejected this argument for the reasons which he gave at paragraphs [53] and following:

“53.

The distinction between trust powers and mere powers is, as I have said, most commonly encountered in connection with powers to distribute income. But the distinction also exists in connection with other kinds of dispositive powers, including powers of appointment. Thus it existed in the case of the 1976 power contained in the 1976 appointment.

54.

The 1976 power was a mere power, not a trust power which the trustees had a positive duty to exercise. That is so notwithstanding the word “shall” … The clause 5 default trusts are capable of continuing until the perpetuity date in 2053, so the draftsman of the 1976 appointment clearly contemplated the possibility that the trustees would not appoint any trusts under clause 3 before their power to do so expired, leaving the 1976 appointment to carry on into the future regulated solely by the default trusts of clause 5. That, indeed, is what I think is happening now.

55.

Given that the 1976 power was a mere power which the trustees did not have to exercise, it ceased to be exercisable on 1 August 1989, and the fact that the trustees had failed to perform their duty to consider whether or not to exercise it cannot mean that it continued to be exercisable after all. The power had still expired, and it did not exist on 2 August 1989, when the trustees purported to exercise it.”

I have reviewed these cases at some length because, although they all concerned private family trusts, the combined experience of counsel was unable to refer me to any case where a similar question had arisen in the context of a pension scheme. Counsel for the employers submitted that, since the rule 11(2) power was a permissive and discretionary one, it lapsed after the expiry of a reasonable time for its exercise. They accepted that the power was fiduciary and could not be released, but submitted that this was not conclusive, any more than the argument that the members were not volunteers. The members had the right to have the exercise of the Trustee’s power over surplus considered, but it did not follow that the Trustee was under an obligation to distribute surplus, and although there were no default trusts of the type found in a normal family settlement, there were nevertheless many things that the Trustee could do with surplus if it was not allocated to members under rule 11(2). Once a reasonable period had elapsed without the power being exercised, the interests of certainty of administration and overall fairness to the beneficiaries required that the power should cease to be exercisable. One particular reason for this is that, unlike the income of a fund, the concept of actuarial surplus is a fluid and flexible one, which is likely to change over time, and if the power to allocate a share of it is not exercised within a reasonable time, the subject matter of the power will cease to exist in any identifiable form.

In urging the opposite view, counsel for the members placed emphasis on the general reasoning of Goulding J in Locker, and on the absence of clearly defined default trusts which played such an important role in the analysis in Allen-Meyrick and Breadner. Mr Moeran pointed out that rule 11(2) itself provides no express time limit for exercise of the power, and he naturally stressed both the central role which the power had played in the provision of target benefits and the value which members had provided, in the form of their services, in return for the benefits under the Scheme. Although actuarial surplus might in some respects be a nebulous concept, there was nothing unreal about the assets from time to time comprised in the Fund, and the deferred members who had left service at a time when target benefits were still being provided to retiring members would be unfairly treated in comparison with pensioners if it were no longer possible to award them a share of the surplus which existed at the time of their leaving service.

Although there is no authority directly in point, I do not in the end feel much doubt about the answer to this question. In the context of a pension scheme, and given the central role of rule 11(2) in the benefit structure, I think it would be wrong to equate the Trustee’s discretionary and fiduciary power to award a share of surplus with a mere power in a family trust, where on a conventional analysis the relevant trust property is regarded as vested in the beneficiaries entitled in default of exercise of the power, subject only to any valid exercise of the power. My conclusion is reinforced by, but not dependent upon, the consistent practice of the Trustee over nearly four decades in making use of the power to award target benefits. The proposition that deferred members would automatically lose any right to be considered for the award of a share of surplus, merely because the Trustee failed to address the question within a reasonable period after their leaving service, is to my mind a deeply unattractive one, and its adoption is not required by any existing authority. The circumstances seem to me to fit far more comfortably into the category of cases where a trust power will not be allowed to lapse for non-exercise, and the court will (as appropriate) either allow the power to be exercised late, or exercise the power itself in the manner best calculated to fulfil the purposes of the trust: compare the well-known observations of Lord Wilberforce in In re Baden’s Deed Trusts [1971] AC 424 at 456H-457B.

For these reasons, I answer Issue 13 in the affirmative.

I come finally to Issue 14, which reads as follows:

“14.

In the event that the answer to the preceding issue is “yes” is the Trustee obliged to perform that duty by reference to the circumstances prevailing and the actuarial surplus (if any) (a) at the date the Member in question left service or (b) at the date the duty is actually performed?”

Wrapped up in this question, in my view, is a more fundamental one, which was the subject of some discussion during the argument on Issues 13 and 14, and which Mr Evans for the employers clearly articulated as a separate submission in the course of his reply. The further question is whether the rule 11(2) discretion ceased to be exercisable in any event when there was no longer any actuarial surplus left in the fund once pensions in payment and the Accrued Amounts of all members had been provided for. That date had occurred at the latest by 10 July 2006. If the answer to this further question is that the discretion did then cease to be exercisable, the reason is not (or not only) that the Trustee must perform its duty under rule 11(2) by reference to the circumstances prevailing at the date when the duty is performed, but simply that, as a matter of construction, the duty necessarily ceases to be capable of performance when there is no longer “any actuarial surplus arising in the Fund” out of which, however belatedly, an award can be made.

As a matter of construction, there is in my judgment much to be said for this simple way of looking at the question. If there is no longer any actuarial surplus in the Fund, there is nothing upon which the rule 11(2) power can bite, and that is so even if the discretion should in principle be exercised today by reference to the circumstances which prevailed when the deferred members left service. There is no way, in either law or logic, by which the actual exercise of the Trustee’s discretion can be backdated. The exercise which has to be performed is not one of historical reconstruction, seeking to establish what would have happened if the Trustee had exercised its discretion when it should have done, but is rather a question of how the Trustee should exercise its discretion today in relation to the Fund as it is now constituted. Without additional funding from the employers, which the Scheme does not contractually oblige them to provide, the hard truth is that there is no longer any actuarial surplus available for distribution.

I find it helpful at this point to draw some further distinctions. In the first place, the present case is quite unlike those where trust income has in breach of trust not been distributed, and where a fund representing the undistributed income remains in existence. In cases of that type, exemplified by Locker, there is no problem in identifying the fund to which the discretion relates, and if the discretion remains exercisable the only question is how far account should be taken of intervening events. In Locker, Goulding J held at 1327 E-H that the power could not be used to benefit certain individuals and institutions which had been added to the defined class of beneficiaries in 1972, after the income in question should have been distributed. He regarded it as “plainly unacceptable in equity” that the default of a trustee should be able to benefit a beneficiary not otherwise interested in the relevant trust property. He added, however, that the trustees should not necessarily ignore intermediate events, and that it might be “too narrow a rule to say of such a trust as that under scrutiny that the trustees’ discretion can only be exercised in favour of a party existing when the particular income arose”. Goulding J did not give an example, but Mr Evans instanced the possibility of a previously overlooked beneficiary coming to light. By contrast, in cases of the present type the problem is simply that there is no longer any actuarial surplus over which the power can be exercised.

Secondly, this is not the kind of case where there is scope for the redistribution or scaling down of other benefits already in payment, or of members’ Accrued Amounts, as a way of providing funding for at least part of the shares of surplus which would no doubt have been awarded to deferred members had the Trustee exercised its rule 11(2) discretion when it should have done. The members’ Accrued Amounts are guaranteed, and as I have already explained once the decision was taken to award a retiring member target benefits, those benefits were locked in and could not subsequently be varied.

Thirdly, a historical reconstruction of the type which I have mentioned would be relevant if a deferred member were seeking to establish loss for the purposes of a breach of trust claim against the Trustee. But that is conceptually quite different from a belated exercise by the Trustee of a discretion which it should have exercised in the past. Further, it is one of the objections to answering Issue 14 in sense (a) that, if the correct approach is for the Trustee now to perform its duty by reference to the circumstances then prevailing, the exercise becomes a very unreal and artificial one, which is in practice likely to be indistinguishable from an exercise in historical reconstruction. To the extent that there may be a difference between the two exercises, the difficult question would arise of how far the circumstances then prevailing should be taken to be the circumstances as they then appeared to be, and how far hindsight may be taken into account. As counsel for Entrust observed, with the benefit of perfect hindsight it can probably be said that there has never been any actuarial surplus in the Fund. Thus, in order to succeed on this part of the case, the deferred members would have to establish not only that Issue 14 should be answered in sense (a), but also, in effect, that the Trustee should now exercise its discretion in the same way as its predecessor would have done then.

These difficulties, and others which were helpfully explored by counsel for the employers and for Entrust in their submissions, reinforce me in the view that the rule 11(2) discretion ceased to be exercisable in favour of deferred members when there was no longer any actuarial surplus in the Fund. If that is wrong, I would in any event hold that the discretion must be exercised by reference to circumstances as they are today in respect of the availability or otherwise of actuarial surplus. In practice, this would of course produce the same result.

There is an obvious, if superficial, attraction in the argument for the members that, as far as possible, the present Trustee should now be directed to do what its predecessor ought to have done when the deferred members left service. If the Scheme were still in surplus, I see no reason to doubt that this desirable objective could in principle be achieved, subject to argument on points of detail about the assumptions that the Trustee should now make. Unfortunately, however, the fact that the Scheme is no longer in surplus seems to me to make all the difference, and to rule out any possibility of remedial action by Entrust.

To conclude, therefore, my answer to Issue 14 is that the Trustee’s duty under rule 11(2) in respect of deferred members who left service before 1 October 2005 ceased to be capable of performance when there was no longer any actuarial surplus in the Fund, and (if that is wrong) the duty must in any event be performed by reference to the actuarial surplus (if any) at the date the duty is actually performed.