Betafence Ltd v Veys & Ors

This action concerns the meaning and legal effect of two amendments to the rules (“the Rules”) of an occupational pension scheme, the TWIL Group Pension Fund, (“the Scheme”), one made in 1991 (“the 1991 Amendment”) and one in 1993 (“the 1993 Amendment”). The claimant Betafence Limited (“the Claimant”), which is the Principal Employer under the Scheme, is represented by Mr Robert Ham QC (I shall refer to the Principal Employer from time to time under the Scheme as “the Employer”). The first to sixth defendants, who are the trustees of the Scheme (“the Trustees”), are represented by Mr Brian Green QC. The seventh to ninth defendants, Mr Ferris, Mr Holmes and Mr Crookes who are representative beneficiaries under the Scheme (“the Beneficiaries”), are represented by Mr Michael Furness QC. All counsel have provided valuable assistance.

The 1991 Amendment was an amendment to the Rules lacking any formality made in response to the decision given on the 17^th May 1990 of the European Court of Justice (“the ECJ Judgment”) in the case of Barber v. Guardian Royal Exchange [1991] 1 QB 344 (“Barber”). The 1993 Amendment was an amendment made by deed intended to formalise the 1991 Amendment.

The primary issue between the parties is one of fact as to the terms of the 1991 Amendment and in particular whether (as contended for by the Beneficiaries) the 1991 Amendment enabled male and female members to retire between the ages of 60 and 65 without any deduction from their pension entitlement or whether (as contended for by the Claimant) the 1991 Amendment required the members, if there was to be no such deduction, to obtain the consent of the Employer to their retirement between those ages. The secondary issues are issues of law relating to the validity of the 1993 Amendment which only arise if the Beneficiaries succeed on the primary issue. I shall say a word on the most important secondary issues. The 1993 Amendment expressly required members first to obtain the consent of the Employer to their retirement between the ages of 60 and 65 if there was to be no deduction from their entitlement. When executed the 1993 Amendment was intended in this regard to do no more than formalise what the Trustees and the Employer at the time believed to have been the effect of the 1991 Amendment. There was no intention in this regard to change the 1991 Rules. If the 1991 Amendment entitled members to retire between the ages of 60 and 65 without any deduction without the consent of the Employer, there is raised in respect of the 1993 Amendment the issue of law whether the absence of any intention on the part of the Trustees or the Employer to change the 1991 Rules in this regard invalidates the 1993 Amendment. If it does so, it is common ground that for the same reason it also invalidates the provision to like effect in the 1997 and 2000 Rules which remained in effect until 2005 when replaced by an intentional and undoubtedly valid amendment to the same effect.

For the purposes of this judgment the following words have the following meaning:

“active member” means a member of the Scheme still in service employed by the Employer;

“retire” means start to draw pension. Retirement may be from service or deferment;

“pensioner” means a member of the Scheme who has retired i.e. started drawing his pension;

“early leaver” means a member of the Scheme who has left before normal pension retirement date;

“deferred member” means a member of the Scheme who is no longer employed by the Employer or who has left the Scheme but has yet to retire and draw his ‘deferred’ or ‘preserved’ pension.

The 1991 and 1993 Amendments were made pursuant to Rule 23 (“Rule 23”) of the 1988 Rules. Rule 23 of the 1988 Rules provided as follows:

There are three significant features about Rule 23. The first is that without the consent of members it is not possible to take away benefits which members have already secured or earned by their pensionable service up to the date of the amendment. The second is that no formal document is required to effect an amendment. There is not even a requirement that the decision to amend the Scheme should be in writing. The only reference to writing is the requirement to notify members of the amendment. This lack of a requirement of any formality is the occasion for the dispute about the 1991 Amendment. The third is that Rule 23 provides for a three stage process: (a) a decision of the Trustees to make the amendment; (b) a decision of the Employer to consent to the amendment; and (c) notification in writing of any amendment to a Member who is affected by it. In the ordinary case minutes of the decisions of the Trustees and Employer and the written notification will afford the primary (though not necessarily the sole) evidence of the amendment made.

In this case the primary evidence of the 1991 Amendment must be: (a) the minute of the resolution of the Trustees at their meeting of the 25^th March 1991 to adopt the 1991 Amendment; (b) the minute of the resolution of the board of the Employer of the 16^th May 1991 to give consent to the 1991 Amendment; and (c) the announcement made to members on the 14^th August 1991. Each of them must of course be read in their context and in particular in the context of events and the other documents. But the Claimant was not content for the courts to decide the issue by reference to these documents and the further documents to which they referred but relied on witness statements by two witnesses, Mr David Young (“Mr Young”) and Mr Arthur Frederick Dewhurst (“Mr Dewhurst”) to the effect that the Trustees and the Employer when making and agreeing to the 1991 Amendment had the intention to require the Employer’s consent for early retirement between 60 and 65 without any deduction. Their evidence was challenged in cross-examination. I shall consider later in this judgment what weight I should give to their evidence.

As at the 31^st December 2003 the Scheme had 602 active members, 1549 deferred members and 949 pensioners in receipt of pensions. As at that date the Scheme had assets with a market value of £80 million.

The Scheme was established by an interim deed dated the 1^st April 1974 for the benefit of hourly paid (works) employees. There was a separate deed establishing a smaller scheme for staff. With effect from the 1^st January 1988 the staff scheme was merged into the Scheme.

From 1988 until 1991 the Scheme was governed by the provisions of the 1988 Rules and was an exempt approved contracted out non-contributory balance of cost final salary retirement benefit scheme. Under the terms of the 1988 Rules (in common with most occupational final salary schemes at the time) provision was made for payment to employees of a pension on “Normal Retirement Date” (“NRD”) which was defined in clause 1 as meaning “the Member’s 65^th birthday if male and 60^th birthday if female”. Rule 5.2 however provided for an “Early Retirement Pension”. This was only payable to a member on retirement before the NRD (in case of retirement for any reason other than permanent incapacity) after his or her 50^th birthday with the Employer’s consent, and if such consent was obtained there was imposed (if the period was ten years or less before NRD) a reduction of the pension entitlement of one third of one percent for each complete month between the date of entitlement and NRD (the equivalent of 4% per annum) or (if such period was more than ten years) a reduction of an amount determined by the Administrator (a term defined as meaning the Trustees). A deferred pensioner had the right to early payment of his pension actuarially reduced to reflect early payment.

Article 119 of the Treaty of Rome (now Article 141 of the EU Treaty) provides that:

On the 17^th May 1990 in Barber the ECJ held, in answering a number of questions that had been referred to it, that:

The effect of Barber was that there could no longer be any discrimination between male and female pension entitlement under the Scheme and from the 17^th May 1990 male members of the Scheme became entitled to unreduced pensions from the age of 60 until benefits under the Scheme were equalised in some other way. Barber however left it uncertain how far (if at all) the ECJ Judgment had retrospective effect. This question was only resolved by the ECJ in December 1993 in Coloroll Pension Trustees v. Russell [1994] PLR 211 (“Coloroll”). Barber accordingly required the Trustees and the Employer as a matter of some urgency and in an uncertain state of the law to consider the impact of the ECJ Judgment and what changes ought to be made to the Scheme to secure the required equalisation. Equalisation could take the form of a single NRD for both sexes at 60 or 65 but subject to the proviso that it was not possible to adopt the single NRD of 65 for both sexes with retrospective effect so as to prejudice previously secured or accrued entitlement of female members.

On the 11^th July 1990 Mr Dewhurst circulated a discussion paper (“the 1990 Discussion Paper”) to present the position on the two major current pension issues. The second of these issues was the Social Security Act 1990. The first was Barber. After stating that a common retirement age had to be applied to male and female members, he went on:

In Appendix 1 to the 1990 Discussion Paper Mr Dewhurst (on expert advice) set out the four main possibilities where the limits might be placed on the retrospective effect of Barber as follows:

Possibility (a) (as Mr Dewhurst reported) was unlikely and (as Mr Dewhurst accepted in cross-examination) he was not referring to possibility (a) as being a reasonable interpretation of Barber. Possibilities (b), (c) and (d) (as Mr Dewhurst accepted) were all reasonably possible interpretations. Of these, possibility (c) is the narrowest: it gives males the benefit of Barber in respect of service after the date of the Barber Judgment. This interpretation involved the conclusion that males may earn benefits by reference to two different normal retirement ages. In respect of service before Barber they will earn benefits by reference to the age 65 and in respect of service after Barber they will earn benefits by reference to the age of 60. Possibilities (b) and (d) assume that, if a member is entitled to benefit from Barber, he will benefit in respect of all his service both before and after Barber. The difference between them is that possibility (b) applies to all pensions which start after the 17^th May 1990 and therefore embraces anyone whose pension has not started on that date i.e. both active and deferred members. Possibility (d) is limited to those who are active members on or after the 17^th May 1990 and leave the Scheme to become deferred members or pensioners thereafter.

Mr Dewhurst went on to consider the options for NRD as follows:

The term “flexible” in relation to retirement age is introduced but not defined in the 1990 Discussion Paper and an issue has arisen as to its meaning. Its natural meaning in this context is “flexible at the option of the member but not of the Employer”. That is confirmed by the costing exercise. The cost of equalising NRD at 60 is given at 3% and the cost implications of a flexible age between 60 and 65 are stated to be similar “with possible small savings if women work on to the age of 65”. The cost would inevitably have been lower if there was equalisation at 65 with a right to retire earlier but only with the consent of the Employer. The reservation and exercise of the right to refuse consent would have enabled the Employer to limit and save costs. No reference to any right to refuse consent was made in the Discussion Paper. I am satisfied that “flexible” meant and was intended to mean flexible at the option of the member and that this remained the position throughout the period when the 1991 Amendment was under consideration and when the 1991 Amendment was made and announced to members. The first suggestion of any requirement for the consent of the Employer was in a curious passage in a minute dated the 20^th August 1991 to which I must shortly refer.

In October 1990 a draft Board Paper was prepared for presentation to the Group Management Committee on the 18^th October 1990. It stated that increasing the NRD to 65 would cause contractual problems with female members who would be likely to be successful in claims for breach of contract or wrongful dismissal if the increase was made; and that an option open to the Scheme was to have a flexible retirement age from 60 to 65 so that men retiring from the age of 60 did not have the 4% early retirement deduction applied; and that if this option was taken, this would satisfy the “worst possible interpretation of retrospection” i.e. possibility (c). The cost would be 3%. No mention is made of the need for the Employer’s consent before the member selected his flexible retirement age.

Also in October 1990 a paper was prepared for the full board. The paper stated that the “current” interpretation of Barber was that it only affected service after the 17^th May 1990 (i.e. possibility (c)) which would cost 2%, but that the Pension Consultants Godwins recommended further improvements if the Scheme was to be more in line with best UK practice and to this end the paper proposed that a flexible retirement age from 60 to 65 could be introduced so that men retiring from the age of 60 did not have the 4% per annum “early retirement deduction” applied to any period of service, and not just the post Barber service. The additional cost of this would be 1% to the funding rate “and it would also satisfy the ‘Barber’ costs given the worst interpretation of how the judgment should be applied” (i.e. possibility (b)). The paper stated that in view of the level of improvements it was not unreasonable to expect that some of the increased costs should be borne by members.

By a letter dated the 8^th January 1991 the Employer wrote to Bekaert to inform Bekaert of the proposed improvement to pension benefits for members aimed to bring the Scheme more in line with other major UK schemes. The relevant improvement was described as follows:

On the 1^st February 1991 Guardian Royal Exchange (“GRE”), the Trustees’ as actuaries to the Scheme, sent to Mr Dewhurst a report on the cost of the proposed improvement of benefits and in particular of: “[h]aving a floating retirement age of 60 to 65 whereby any member who retires in this period will receive a pension without actuarial discount”.

In February 1991 Mr Dewhurst on behalf of the Trustees issued an announcement to employees that the Employer was looking at the overall package of benefits and further improvements to the Scheme which would require contributions from members to meet the substantial extra funding required.

A report to the Trustees for their meeting on the 25^th March 1991 was prepared and distributed (“the March 1991 Report”). The March 1991 Report stated that the options were either to do the minimum necessary and await clarification of Barber or to take the opportunity to introduce improvements necessary to bring the benefits up to today’s requirements and remain in line with comparable companies and at the same time to introduce a contributory element to help fund what may eventually become a compulsory requirement. The recommendation was that the second option be chosen and that improvements be introduced together with a contributory element. As a result the Scheme would be fully in line with better UK practice with benefits that will automatically cover the impact of any reasonable interpretation of Barber i.e. possibilities (b), (c) and (d). These included a pension increase of 5% (or RPI if less) on total pension and a flexible retirement age between 60 and 65. The Report goes on:

This last sentence in brackets is of critical importance. If there was a requirement for consent to retirements between the ages of 60 and 65, this must apply to male and female members alike and accordingly female members would be “worse off” than before.

Appendix II to the Report sets out the four possible limits on retrospectivity which appeared in Appendix I in the 1990 Discussion Paper. Appendix III gave some indication as to comparable companies’ positions at the time, referring to a NAPF survey which showed that 82% of “schemes with an equal age of 65 [for males and females] operated no actuarial reduction on early retirement which is in essence a flexible retirement age”. Indeed the scheme of one of the Claimant’s shareholders (British Steel Corporation’s) permitted male members to retire at 60 without any actuarial reduction or the need for employer’s consent.

The meeting of the Trustees took place on the 25^th March 1991. Mr Young (though a trustee) did not attend. The minutes recorded the Trustees’ decision as follows:

The proposed amendment then had to proceed for a decision by the board of the Employer whether to give its consent. The pension committee of the Employer’s Belgian parent company NV Bekaert (“Bekaert”) reviewed the proposal on the 8^th April 1991. A memorandum of that date records that: (1) management was convinced that Barber was applicable to all service years; (2) management found it impossible to increase the retirement age of women to 65; (3) within these assumptions the pension committee found the proposal in balance; and (4) if the Employer had to give the same retirement benefits to men (at 60 years) as were then available for women, then it was an ideal moment to review the system and make the pension plan contributory.

On the 19^th April 1991 Mr Rankin of GRE wrote to Mr Dewhurst regarding commutation factors. The last paragraph of the letter read as follows:

A board paper or agenda was prepared by Mr Dewhurst for the Employer’s board meeting of the 16^th May 1991 which was broadly similar to the paper provided to the Trustees save that it omitted option (a) namely to make only such changes as were required by law. The paper stated that after detailed consideration a recommendation was submitted to the board for approval. After setting out the four possible limitations on the retrospective operation of Barber (which were set out in Appendix II), the recommendation read:

Appendix II, after setting out the four possibilities, went on:

It is clear (notwithstanding the evidence of Mr Dewhurst) that the words (“d above”) are in error for (“b above”). The meaning of the sentence to which “(d above”) is appended corresponds to possibility (b) and not (d); the reading as (“b above”) alone accords with the expressed intention to satisfy any reasonable interpretation of Barber; and the reading as (“b above”) again alone accords with the recent decision of the Industrial Tribunal referred to, namely Roscoe v. Hick Hargreaves & Co Ltd [1991] PLR 51 given on the 13^th March 1991 in Manchester which (in paragraph 7) held that “equality of treatment will apply after 17.5.1990 to the full amount of all pensions for retirements after that date”.

Appendix V set out the recommended improvements: no mention was made of any requirement of consent for retirement from the age of 60.

The relevant minute of the board meeting reads as follows:

Careful handling was most particularly required because the change from a non-contributory to a contributory scheme was a sensitive issue. Mr Dewhurst thereupon carefully and after full consultation and circulation of drafts prepared an announcement to employees of the amendment (“the Announcement”) for circulation to male and female members in accordance with Rule 23. The Announcement was made on the 14^th August 1991. Headed “Introduction to the TWIL Group Pension Fund Improvements”, the Announcement read:

It is to be noted that the Announcement makes no reference to any requirement for consent: elementary concern for the interests of members and indeed common honesty would have required the disclosure of any such requirement if any such requirement was to be imposed. Further the Announcement does not limit flexible retirement to retirement from service.

On the 20^th August 1991, one week after the Announcement a meeting was held in Edinburgh attended by Mr Dewhurst, Mr White of Godwins and Mr Thornton (and another) of GRE. No trustee or board member was present. The minute of the meeting states that its purpose was to “clarify certain features of the TWIL benefit improvement package”. The minute records that all costings had been calculated on the assumption that members retired at 63; that this meant that any member retiring before this age would cause a strain on the fund; and that the current benefits on early retirement could be considered over generous. It is apparent that there were second thoughts about the improvement in benefits, for the minute records: “It was agreed [at the meeting] that the current benefits received from early retirement could be over generous.” After considering the position of existing members the minute goes on to consider two other classes of members, namely members retiring from service prior to attaining age 60 and early retirement at any age for members with deferred pensions. As regards the latter it was minuted:

At the end of the paragraph dealing with the last of these classes of members there is a curious paragraph which reads as follows:

In October 1991, an “information guide” (“the Guide”) was circulated to members of the Scheme by Mr Dewhurst. Mr Dewhurst confirmed that it “should be read in conjunction with the Announcement issued in August 1991” (which did not contain any requirement for consent). He also stated that “this information guide is only a brief summary of the benefits provided and cannot override the Trust Deed and Rules”. In the section of the Guide headed “Cost”, Mr Dewhurst stated that “you will contribute 6% of your Pensionable Earnings”. Given that both the Trustees and the Employer had decided (on the 25^th March 1991 and the 16^th May 1991 respectively) that existing members would contribute 3% and only future members would contribute 6%, it is clear that the Guide was directed at new members and I am not satisfied that the Guide was in fact circulated to members other than new members. An existing member who saw it would assume that it was not directed at him. In summary, the Guide stated that normal retirement age was 65 and that a member’s pension entitlement would be 1/60^th of final pensionable earnings for each year of pensionable service. The Guide also stated that a member could retire from age 60 without loss of pension with the Company’s consent. This is the first time (in the context of the 1991 Amendment) that this requirement for consent had been included in a communication to members. As I have said existing members were not sent the Guide and, if they saw it, would not have read it as having any application to them.

More than a year after the Announcement, a Trustee meeting took place on the 26^th August 1992. At that meeting, concerns were raised about the “current and potential costs of the improvements to the Scheme in 1991”. There was “some concern that early retirement without penalty would tend to occur more at age 60 rather than the assumed age 63”, Notwithstanding the terms of the Guide, the minutes do not record any mention being made of a requirement for Company consent in the context of these concerns. Clearly the concerns being expressed could have been addressed by the Company in the way it exercised any discretion which it provided whether to grant consent. In the section of the minutes headed “Any other business”, it is recorded that Mr Dewhurst “expressed concern about current early retirement policy with particular reference to individuals funding for early retirement with AVCs on the basis that company consent to early retirement would be given. It was generally felt by the trustees that these were exceptional situations and there was a desire to maintain the 4% per annum early retirement reduction factor because of its inherent flexibility”.

An actuarial valuation of the Scheme as at the 6^th April 1992 was prepared by Godwins dated February 1993. At page 15 of that valuation, Godwins discussed the background to and effect of the Barber Judgment. At page 22, when setting out the “Benefits and Contributions” of the Scheme, it stated that normal retirement age was 65 for both males and females and noted that a member could retire from service after reaching age 50, with the consent of the Company. It also noted the removal of an actuarial reduction in the event of retirement between ages 60 and 65.

Ahead of adoption of the 1993 Rules on the 1^st July 1993 Mr Dewhurst prepared a note of issues for consideration regarding the purpose of the 1993 Rules: “The existing rules state that with the company’s consent a member may retire from age 50 but the benefit will reduce by ⅓% for each complete month for the period between date of retirement and the Normal Pension Date, or where such period is more than ten years by an amount determined by the Administrator.” Mr Dewhurst goes on to note that “rule amendments cannot make benefits less attractive” and that “the 4% PA reduction [this is the ⅓% per month reduction referred to in his opening paragraph] is well known by the members and the changes made in October were “sold” on the basis of the improved Flexible Retirement. Mr Dewhurst does not explicitly say that the “Flexible Retirement” changes made in 1991 removed the 4% per annum reduction in benefits in the event of retirement between ages 60 and 65. Mr Dewhurst then goes on to outline the “issues for consideration”. At heading 1, he addresses the “New Rules”. He makes two points in relation to these:

“These reflect previous practice and ‘spirit’ of October 1991 improvements.”

“The company’s consent is still required where early retirements are calculated using the 4% pa reduction with the discretion to use “actuarial factors” between age 50 to 55.”

A Trustees meeting took place on the 2^nd July 1993. Two of the items on the agenda for consideration were the “Scheme Rules” (item vi) and the “Early Retirement Policy” (item vii). In the minutes of the meeting Mr Dewhurst noted that the 1993 Rules were ready for execution, that he has discussed the Company’s policy towards early retirement with Godwins, and that “this should be kept as broad as possible and should only be provided with the Company’s consent as at present. This policy should be reviewed if circumstances change.” These minutes were approved at a subsequent board meeting on the 29^th April 1994 at which it was also noted that the 1993 Rules had been executed.

The 1993 Rules were adopted on the 17^th November 1993 by a Deed of the same date. The 1993 Amendment formed part of the 1993 Rules. (I have not been taken to any part of the 1993 Rules other than the 1993 Amendment.)

Notwithstanding the requirement both of Rule 19 of the 1993 Rules and of Rule 23.0 of the 1988 Rules that members be notified of any amendments which affect his/her benefit entitlement under the Scheme, there was no notification issued to members following adoption of the 1993 Rules. There was an (undated) “Explanatory Booklet”. But, since this booklet defines “Normal Retirement Date” as 65 for men and 60 for women the booklet must have pre-dated the 1993 Rules and indeed the 1991 Amendment and is therefore irrelevant for the purposes of notification. There was also another undated booklet directed to new entrants which is a “Guide to Benefits” under the Scheme. The booklet referred to Normal Retirement date for both men and women being age 65, explained the two classes of member which the 1993 Rules effectively created, stated that with the Company’s consent, members could “retire from age 60 with an immediate pension” and noted that there would be a reduction in benefits in cases of retirement prior to age 60. But the evidence does not establish when and to whom this booklet was sent.

I must first decide the issue of fact whether (as maintained by the Claimant) the Trustees and Employer when making the 1991 Amendment had the intention to require the Employer’s consent for early retirement between the ages of 60 and 65 without any deduction of pension entitlement. For this purpose it is necessary to evaluate the evidence of Mr Young and Mr Dewhurst against the background of the documentation to which I have referred as well as the other available evidence. The contemporary documentation points clearly and unequivocally to the absence of any requirement for consent.

Mr Young was a chartered accountant who became finance director of the Employer and in the 1980s a member of the main board. Mr Young was in the 1990s Vice President of Bekaert and from late 1991 Chief Executive of the TWIL Group. Mr Young became a trustee of the Scheme in the mid 1980s and Chairman of the Trustees in 1992 and he remained a trustee until his retirement in 1997. Mr Dewhurst was employed as pensions manager at TWIL between the 27^th April 1987 until his retirement on the 31^st April 2000 and was a trustee of the Scheme from the 1^st December 1991 until the 1^st October 1996.

I regret to say that I found neither Mr Young or Mr Dewhurst a satisfactory witness. Their evidence related to events some fifteen years or so ago and disclosed a marked inclination to a reconstruction of events as (in retrospect) they would wish them to have been. In retrospect, as seen shortly after the 1991 Amendment was adopted, the 1991 Amendment was perceived by the Employer to have been too generous and expensive and (embarrassed by their failure to appreciate this at the time that the 1991 Amendment was made) to right the balance Mr Young and Mr Dewhurst have sought to persuade the court, and have perhaps persuaded themselves, that the Employer’s consent to retirement between the ages of 60 and 65 was required if the pension was not to be reduced.

Mr Young was living and working in Belgium for a period of about a year until October 1991 and during this period was not involved in the Trustees’ decision-making. He gave no satisfactory explanation why he did not disclose this fact in his witness statement. His evidence was in large part advocacy of the interest of the Company on the issues raised and he was unwilling to concede any fact which might support the claim of the Beneficiaries unless and until forced to do so. He sought to make a series of dubious propositions none of which he could sensibly maintain:

“flexible” means “manageable from the point of TWIL, having in mind the need to be able to have a measure of control over the workforce.

“option” referred to the choice of TWIL and not the employee.

women would not be “worse off” because TWIL would probably give them permission to leave early, though there was no guarantee.

I found Mr Dewhurst an even less satisfactory witness. In his proposals for the 1991 Amendment and in the 1991 Amendment itself and the Announcement, Mr Dewhurst made no reference to any provision requiring the consent of the Company to such early retirement. The dispensation with this requirement was not a matter of any real moment, because for economic reasons male members were unlikely to be able to afford or wish to retire before the age of 65 when they became entitled to their state pensions unless they were dismissed in which case they would be likely to become entitled to compensation in the form of a redundancy package or the like, and if they were dismissed the Company was anyway likely to give any necessary consent. Only after the 1991 Amendment was made did Mr Dewhurst have second thoughts. It is apparent that it was only on second thoughts thereafter that Mr Dewhurst adopted and promulgated the argument that the 1991 Amendment was never intended to and did not dispense with the need for the Company’s consent for any such retirement (of male members) if it was not to result in a deduction from pension entitlement and that the 1991 Amendment required the imposition of the requirement of consent for early retirement by female members.

The first expression of that point of view was in the statement at the meeting of the TWIL Group Pension Fund held on the 20^th August 1991 to which I have already referred which (as recorded in the minutes) reads as follows:

Mr Dewhurst says that in the course of a series of “road-shows” he explained to employees the 1991 Amendment and in the course of such explanations referred to the requirement for consent. I cannot accept that he told the employees, and in particular female employees, of the requirement for consent (in particular) for two reasons. The first is that he says that this statement produced no adverse reaction. It is scarcely credible that this statement on his part, if clearly made, would not have produced an adverse reaction, most particularly from female members. The second is that an employee Mr Ferris recalls no such explanation at the meetings which he attended.

Mr Dewhurst in his evidence has maintained that there was at all times agreement within the Company that the requirement for consent in the unamended rules should continue in full force. In his witness statement he stated (in paragraph 9) that throughout all the discussions surrounding the changes in 1991 the requirement for the Company’s consent was “taken as read” (paragraph 9) and that: “Everyone understood implicitly that there was a need for consent”. Indeed in a witness statement prepared for other purposes and disclosed on discovery he made a statement to like effect adding: “Indeed I believe that is why there is no mention of Company consent being removed”. Yet under cross-examination he said (unconvincingly) that these statement were wrong and that there was explicit agreement to this effect, though this is not reflected in any contemporaneous document. I do not accept that there was any such agreement or understanding.

To support his contention that the requirement for consent was never withdrawn, Mr Dewhurst told me that the withdrawal would have been a “huge” issue. This is clearly not correct, as is clear from the evidence of Mr Young. Mr Dewhurst was totally unable to explain how: (1) silence on the topic of consent (and there was silence) could result in the imposition of the requirement of consent on female members whilst not affecting the status quo in respect of male members; or 22) how the imposition of the condition requiring consent on female members could not leave them “worse off”.

It is unfortunate that the Claimant, if it wished to raise the issue of what the Trustees and Employer really intended, did not call any evidence on this issue from any director or trustee other than Mr Young and Mr Dewhurst and called no evidence from their advisers at the time GRE and Godwins. The evidence that the Claimant did call was quite inadequate for this purpose.

I am firmly of the view that the issue whether or not the 1991 Amendment required the consent of the Employer to early retirement without deduction can only and should be determined on the documents and in particular the primary evidence to which I have referred. It is quite apparent on the documents read in their context that the 1991 Amendment was proposed and adopted to secure equality of treatment by increasing the Normal Retirement Age of women members to that of men i.e. from 60 to 65, but to avoid the breach of contract which this would otherwise occasion and to avoid women members being “worse off” than previously by extending to male members the right previously only enjoyed by female members to retire at any time between the ages of 60 and 65 without any reduction of pension entitlement. The members were being granted an “option” to retire early without any deduction. The “flexible retirement age” (as I have already stated) was an age that was to be flexible at the instance of the employees and not of the Employer. The Announcement announced the 1991 Amendment as an “improvement of benefits”. Any honest and fair announcement would have announced the dilution or withdrawal of benefits if there were any and in particular the withdrawal of the right of female members to retire without any deduction at the age of 60. That is clearly how the Announcement was understood by Mr Robinson and the Beneficiaries and how they were intended to understand it. This was the clear understanding of GRE, though (it appears) not of Godwins.

The concession in respect of male members (reducing the age to 60 at which they could retire without penalty) was plainly not a considered matter of moment: for (as Mr Young confirmed in his evidence) few (if any) members would wish to avail themselves of the right to do so because they would not (in any ordinary case) wish to have to manage on their pensions unless and until supplemented by the state pension which only kicked in at the age of 65. So far as it had any moment, it was part of the purchase price of obtaining the consent of members to the Scheme becoming contributory. There could have been no question of female members voluntarily foregoing the right to retire without reduction at the age of 60. It was recognised that to attempt to do so would constitute a breach of contract and preclude the commendation to female members of the 1991 Amendment that they would not be worse off.

Nothing was even said about the requirement of consent: this is confirmed by the witness statements of Mr Young and Mr Dewhurst, though they unconvincingly seek to explain the silence as referable to the assumption that the condition previously applicable only to male, but not to female, members continued to operate. I do not think that there can have been any such assumption and still less do I accept the evidence given on behalf of the Claimants (most particularly by Mr Dewhurst) that the consent would “continue” to be required.

I hold that the 1991 Amendment did not require the consent of the Employer to retirement between the ages of 60 and 65 of members of either sex to avoid any reduction in pension entitlement. I also hold that for this purpose the 1991 Amendment made no distinction between existing and new members. The 1991 Amendment was effective from the 17^th May 1990, the date of the ECJ Judgment which conferred this pre-existing entitlement of female members on male members.

A more difficult question arises whether the 1991 Amendment conferred a like entitlement on existing deferred members. There are conflicting indications in the contemporary documents. In favour of including such deferred members is the clear intention to cover by the amendment possibilities (b), (c) and (d) which (read together) include deferred members. On the other hand to include within the scope of the 1991 Amendment existing deferred members may seem unduly generous to them; there is a reference in this context to the 4% deduction which was inapplicable to deferred members; and there were references excluding deferred members from benefit in documents after the 1991 Amendment was made. After anxious consideration I have concluded that the 1991 Amendment was intended to benefit existing deferred members as falling within possibilities (b), (c) and (d). This was an act of generosity later regretted and ineffectively sought to be retracted.

The question is raised as to the entitlement to the benefits of the 1991 Amendment of members of the Scheme who were active on or after the 17^th May 1990 but became deferred members thereafter. The answer is provided by the provisions of Schedule 16 of the Social Security Act 1973 which contained the provision then applicable governing preservation of benefits.

Starting with the Social Security Act 1973 Parliament has legislated on a number of occasions to protect the position of early leavers, i.e. members of occupational pension schemes who leave pensionable service before normal pension age. Before then schemes often made extremely ungenerous provision for early leavers. At the time of the 1991 Amendment the relevant legislation was still to be found in the 1973 Act though shortly thereafter it was consolidated in the Pension Schemes Act 1993.

Under paragraph 2(b) in Schedule 16 to the 1973 Act long service benefit is the benefits payable if the members remain in employment until normal pension age; paragraph 6 imposes an obligation to provide short service benefits for certain members; and paragraph 8 provides that there is to be no discrimination between short service and long service beneficiaries. Paragraph 4 of Schedule 16 provided as follows:

In my judgment, if the 1991 Amendment did not extend to such members, these provisions overrode the provisions of the Scheme in this regard and extended the benefit to them. The Employer has argued that the application of the provisions is excluded by paragraph 4(4) and in particular the direction that there is for this purpose to be disregarded any scheme rule making special provisions as to early retirement “on grounds of ill-health or otherwise”. But in my judgment the words “or otherwise” refer to other grounds, and accordingly the subparagraph does not cover the present case where the member does not have to establish any particular grounds before claiming an early retirement pension.

Since I have upheld the Beneficiaries’ case on the 1991 Amendment I must now turn to the series of questions raised as to the validity of the 1993 Amendment. One preliminary point must be made regarding the 1993 Amendment. The 1993 Deed is invalid to the extent that it fails to recognise that under the 1991 Amendment for all service before the 17^th November 1993 both male and female members had (as I have held) the right to retire on an unreduced pension at the age of 60 and this right was protected by the proviso.

The first issue is whether the 1993 Amendment is invalidated by the failure of the Trustees to notify the members of its making. The Trustees were under a duty under Rule 23 of the 1988 Rules to notify beneficiaries of the 1993 Amendment, because according to its terms it affected their entitlement: it conditioned their right to retire between the ages of 60 and 65 without deduction upon obtaining the consent of the Employer.

There are two separate questions. The first is one of fact whether the Trustees complied with the obligation of notification. The obligation was clearly and unequivocally to notify the members of the amendment. Whilst one of the booklets notified new entrants of their rights under the 1993 Rules, I do not think that the booklet adequately notified existing members of the 1993 Amendment to their rights and (quite reasonably) the Beneficiaries were not so alerted to the change. But the second is one of law whether the failure to notify in any way invalidated the 1993 Amendment. In my judgment there is no basis for holding that the breach of duty by the Trustees in failing to notify in any way invalidated the 1993 Amendment. Neither the language of Rule 23 nor any rule of law lends any support to any argument in favour of invalidation. There may be a failure to notify members (or some of them) for any of a variety of reasons. The amendment may be for the member’s advantage or disadvantage. The accident of a failure of notification cannot prejudice the legal effect of the amendment. What it might do is afford to members who are able to show that they have been prejudiced by the failure to communicate, the possible basis for a claim for compensation against the Trustees.

The question is raised whether (assuming that the 1993 Amendment is otherwise valid) having regard to the proviso to Rule 23, which invalidates the consent requirement as regards benefits entitlements accrued prior to the 17^th November 1993 (the date of the 1993 Deed), the consent requirements under the 1993 Deed are valid as regards benefits accruing thereafter (as the Claimant contends) or whether the consent requirements are incapable of severance and wholly invalid (as the Beneficiaries contend).

The Claimant is plainly correct. The 1993 Amendment must be construed as having effect subject to the overriding limitation on the power of amendment contained in the proviso. Questions of severance do not arise, but if they did the principles governing severance in a case such as the present (as the cited authorities establish) lead to the same conclusion. There is no requirement or scope for application of the “blue pencil” test deleting what is objectionable and leaving standing what is unobjectionable. All that is required is that the distinction between what is and what is not objectionable is clear and that the meaning and application of what is unobjectionable is clear.

The final question raised is whether the 1993 Amendment is invalidated by reason of the mistake: (a) on the part of the Trustees when making it; and (b) of the Employer when consenting to it that the 1993 Amendment involved no amendment to the pre-existing rule but merely formalised an amendment to the same effect made by the 1991 Amendment. The Trustees failed to address their minds to the question whether the 1993 Rules should introduce a new rule reversing the 1991 Amendment. There was no consideration given whether the power of amendment should be exercised in this regard. It was not appreciated that the power of amendment in this regard was being exercised at all. The exercise of the power to re-impose the requirement of obtaining the Employer’s consent was a matter of the greatest importance to members and any decision to exercise it called for anxious consideration of the interests and views of the members and the Employer and the impact of any such exercise upon the members and the Employer and the funds of the Scheme. There was no such consideration.

The court is protective against challenges to the exercise by trustees of their discretion. The principle is stated in In re Hastings Bass deceased [1975] Ch 25 at 41 as follows:

Lloyd LJ in his judgment in Sieff v. Fox [2005] 1 WLR 3811 surveyed the cases in which the exercise of discretion by trustees can be challenged on the grounds of a mistake made by trustees. He held that there was a distinction between cases where: (1) the trustees were free whether or not to exercise the discretion, in which case their exercise of the discretion could only be challenged if the trustees but for the mistake would have acted differently; (2) where the trustees were bound to exercise the discretion in which case their exercise of discretion could be challenged if the trustees but for the mistake might have acted differently. On this basis (subject to one further consideration) challenge to the 1993 Amendment must fail, since the Trustees were free whether or not to exercise their discretion to make the 1993 Amendment and on the evidence before me, it cannot be said with any certainty that but for the mistake the Trustees would have acted differently: at the highest the outcome is uncertain. There are substantial grounds for questioning whether this obstacle in the way of challenging a decision by trustees which is established to be applicable in case of traditional trusts should be carried over into Pension Trusts where the members have “bought” their benefits and may have a legitimate expectation that the erroneous decisions of trustees adverse to their interests will not be (for practical purposes) immune from challenge: see e.g. Pension Trusts and Traditional Trusts: Drastically Different Species of Trusts by David Hayton, The Conveyancer and Property Lawyer 2005 page 227. But Lloyd LJ in Sieff v. Fox expressed the view that the same principle applied in case of traditional and pension trusts and (despite my reservations) I think that I should follow his guidance and leave it to an appellate court to review the law. Applying the principle stated by Lloyd LJ I therefore should uphold the 1993 Amendment unless I can hold that the Trustees in making the 1993 Amendment had no intention to exercise their power under Rule 23 of the 1988 Rules at all.

The further consideration to which I referred above relates to the clear distinction between cases where trustees have exercised their discretion and where they have never exercised their discretion at all. In such a case for a successful challenge it is unnecessary to prove that the result was not the intended outcome of the trustees’ decision. There was never a considered decision at all. This accords with the judgment of Mervyn Davies J in Turner v. Turner [1984] Ch 100 when he held a nullity an appointment made by trustees of a discretionary trust on the instructions of the Settler and without appreciating that they had any discretion at all. He stated (at p.111);

After anxious consideration I have concluded that this case falls within In re Hastings Bass and is outside Turner v. Turner. The Trustees clearly intended to exercise their discretionary power to make the 1993 Rules. They made a mistake in thinking that in respect of the 1993 Amendment they were merely formalising the 1991 Amendment, but that does not detract from the fact that they intended to adopt the 1993 Rules and as part of them the 1993 Amendment. Whether their mistake that the 1993 Amendment did not amend but merely formalised the 1991 Amendment brings into play the Hastings Bass principles, and on the application of those principles the 1993 Amendment is valid.