Sterling Insurance Trustees Ltd v Sterling Insurance Group Ltd

MR JUSTICE NUGEE: I have before me a Part 8 claim form raising a question of construction of a definitive trust deed of a pension scheme known as the Sterling Insurance Pension Scheme.

There is only one question, and both counsel (rightly in my view) have described it as a short point of construction. The question arises on the power of amendment in the definitive deed, and I will read that straight away. Under the heading “Amendment of trust deed and rules” it provides:

“The Trustees shall have power with the consent of the Principal Employer by deed or written instrument to alter modify or add to all or any of the provisions of the Trust Deed or the Rules except that no such alteration, modification or addition shall operate so as to substantially reduce in aggregate the value (as to which the decision of the Trustees acting on the advice of the Actuary shall be final) of the benefits accrued due in respect of any Member up to the date of such alteration, modification or addition.

The Trustees shall notify in writing each Member affected by any such alteration modification or addition.”

The facts are not in dispute. The scheme was established by an interim trust deed dated 23 July 1996. That interim deed established the scheme with effect from 1 August 1996. In the usual way it contained an undertaking by the Principal Employer and Trustees to execute the Definitive Trust Deed within 24 months, which in the usual way was not complied with, and it did not set out the benefits to be provided under the scheme but referred to an announcement which had been circulated to every employee and provided that the Trustees should operate the scheme in accordance with such announcement until the execution of the definitive deed.

It contained an amendment power in clause 5 which was in rather shorter form than that subsequently inserted into the definitive deed in these terms:

“Amendment.

The Trustees shall have the power with the consent of the Principal Employer by deed or written instrument to alter, modify or add to all or any of the provisions of this Deed or the Definitive Trust Deed or the Rules except that no such alteration modification or addition shall operate so as to prejudice the approval of the Scheme under Chapter I.”

The definitive deed was executed on 7 September 1998 and took effect from 1 August 1999. It contained in schedule 1 the trust deed of the scheme and in schedule 2 the rules of the scheme. I need not refer to very many provisions of the definitive deed. I have already referred to the most significant one which is the amendment power which falls to be construed. I was also referred to clause 14.1 which provides as follows:

“The Principal Employer may at any time by giving one month’s advance notice in writing to the Trustees terminate the Scheme. Upon such notice taking effect the liability of the Employers to contribute shall be terminated except in respect of any contributions due before the date on which the notice takes effect.”

The rules set out the benefits and I can summarise those by saying that it is what was at that stage a conventional final salary scheme under which the primary benefit was a pension at normal retirement date, in this case the member’s 63^rd birthday of N/60^ths or, as it is expressed in this scheme, an amount equal to 1/720^th of the member’s Final Pensionable Salary for each complete month of his pensionable service. The Final Pensionable Salary of a member means his highest Pensionable Salary on any 1 January in the last five years before his exit date. There is a definition of “Pensionable Salary” which refers to the member’s basic salary. The exit date is defined as follows: “Exit Date of a Member means (subject to Rules 8 and 9) the date he retires from or leaves Service or, if earlier, the date of his death”. “Service” means service with the Employers, and the Employers are defined as the Principal Employer and the other Employers participating in the Scheme. The practical effect of that is that in the normal way (for a final pensionable salary scheme) each member earns or accrues during his service a right to a future pension measured by reference to his total pensionable service and his final pensionable salary (that is, the salary at the date that he leaves the service of the employers). That is conventionally referred to as the final salary link.

On 25 January 2000 that part of the scheme, the final salary scheme, was closed to new members by deed of variation, which split the scheme into two sections, a final salary section which continued the scheme as it was immediately before 1 January 2000, and a money purchase section, and provided that all persons becoming members of the scheme after 31 December 1999 should, except with the consent of the Principal Employer, become members of the money purchase section. I am not concerned with the money purchase section.

On 1 September 2001 there was a deed of variation which made a number of amendments. One of the amendments to which I was referred was made with effect from 1 September 2001 and introduced a new definition of Final Pensionable Salary as follows:

“Final Pensionable Salary of a Member means:

(i)

in respect of Pensionable Service prior to 1 September 2001, his highest Pensionable Salary on any 1 January in the last five years before his Exit Date; and

(ii)

in respect of Pensionable Service on or after 1 September 2001, the average of the highest three consecutive figures for his Pensionable Salary on any 1 January in the last seven years before his Exit Date.”

Finally, on 31 December 2004 a deed was introduced which closed the final salary scheme to future accrual. That provided that the trust deed and rules should be amended with effect from 1 January 2005, and one of the amendments was as follows:

“Add at the end of Rule 4:

Where the Member’s Exit Date is after 31 December 2004, his Pensionable Service and his Contracted-out Service shall cease at the end of 31 December 2004, and his Final Pensionable Salary shall be determined as at the end of 31 December 2004 …”

The practical effect of that, if it was effective, was to break the final salary link so that not only did the members who were then in service cease to accrue any further pensionable service (a part of the amendment which is not suggested to be ineffective) but also their final pensionable salary was calculated as at the end of 31 December 2004 and not at the later date (if they continued in service) when they left the service of the employers. It is that which is challenged in these proceedings as being ineffective, the argument being that the proviso to the amendment power prevented an amendment which broke the final salary link.

It is common ground between counsel that, at any rate at the level of the High Court, a proviso to an amendment power in a pension scheme which prevents reduction in a member’s accrued benefits (using that phrase just on its own) is one which does prevent the breaking of the final salary link. That was decided by Newey J in a case called Briggs v Gleeds [2014] EWHC 1178 (Ch) in which, as it happens, the counsel who appear before me, Mr Jonathan Evans QC (who appears before me for the trustees) and Mr Paul Newman QC (who appears before me for the employer), both appeared. Newey J was concerned in that case with a large number of questions but one of the questions was a proviso to an amendment power which is set out in his judgment at paragraph [114] and was in these terms:

“The principal employer and the trustees may jointly from time to time without the consent of the members by deed alter cancel modify or add to any of the rules or provisions of this deed, provided that no such alteration cancellation modification or addition shall be such as would prejudice or impair the benefits accrued in respect of membership up to that time.”

His decision is given at paragraph [128], in which he decided that there was no good reason to take “accrued” in that proviso to have a narrower meaning than the word “secured” in a similar context and that Millett J in the Courage case (which I will come to in a moment) considered that benefits that had been “secured” included the “prospective entitlement to pensions based upon final salary”.

The Courage case, under the name Re Courage Group's Pension Schemes, dates back to 1986 and was reported in the All England Reports at [1987] 1 AER 528 and also reported in the Weekly Law Reports at [1987] 1 WLR 495. Curiously, the relevant passage is slightly different in the two versions, and this is all explained in Newey J’s judgment where he accepted that the judgment in the All England report is the more authoritative version. As set out in the All England report at 543f, what Millett J said was this (having referred to the various provisos to the rule-amending power in the various schemes before him):

“'Accrued pensions' is defined in the rules [that is in relation to one of the schemes] to mean pensions based on salary at the relevant date. There was some dispute whether 'benefits already secured by past contributions' [which was the phrase used in the proviso in one of the other schemes] means the same thing, or includes the prospective entitlement to pensions based on final salary. In the absence of express definition, I see no reason to exclude any benefit to which a member is prospectively entitled if he continues in the same employment and which has been acquired by past contributions, and no reason to assume that he has retired from such employment on the date of the employer's secession when he has not. The contrary argument places a meaning on 'secured' which is not justified.”

In the Weekly Law Reports version the last sentence has the words “and accrued” after the word “secured” but it does seem, as set out in paragraph [117] of Newey J’s judgment in Briggs v Gleeds, that Millett J had second thoughts about that and took out the words “and accrued”. Nevertheless, as I have said, in Briggs v Gleeds Newey J decided that there was no reason for taking “accrued” to have a narrower meaning than “secured”, and, as I have said, both counsel before me accept that at this level I should follow Newey J and that, had the proviso in this case simply referred to benefits accrued up to the date of amendment, it would follow that it did protect the final salary link and that the 2004 amendment was ineffective.

Mr Newman has made it clear that he reserves the right to challenge that in the Court of Appeal and as far as I am aware this question of what Millett J said in Courage and whether it applies to “accrued” as well as to “secured” has not been considered at the Court of Appeal level and it may be that one day it will have to be, but I will proceed on the basis that had the proviso in this case simply referred to the benefits accrued in respect of any member up to the date of such alteration, modification or addition, it would protect the final salary link.

Mr Evans (who, although he appears for the trustees, has in order to save costs taken on the burden of arguing the case on behalf of those who are interested in challenging the break of the final salary link) submits that the proviso in this case should be given a similar meaning and that in effect the court should ignore the word “due” which appears in the phrase “the benefits accrued due in respect of any member” or should give it a meaning which does not in any way impact on the force of what I have said is common ground would be the meaning attributable to the phrase “the benefits accrued” if it stood alone.

Mr Newman (who appears for the employer and for any members in whose interest it is to argue against that) says that that is wrong and it is his submission that “benefits accrued due” means benefits that have become payable. The basis on which he says that is that the phrase “accrued due” has a well-known and well-established natural and ordinary meaning (well-known to lawyers at any event) and that it refers to payments which have become payable.

I accept that the word “due” in its primary meaning means something that is due and owing and payable and I accept that in its natural and ordinary meaning the phrase “accrued due” is regularly used by lawyers to mean a payment which has fallen due for payment, by contrast with the phrase “accruing due” which refers to a payment which has not yet fallen due for payment. Mr Newman showed me a number of examples, not of judges construing the phrase “accrued due” in any particular context but simply of using the phrase “accrued due” as evidently having that meaning. One is in fact in the judgment of Newey J in Briggs v Gleeds itself at paragraph [128(i)] in which he said as follows:

“Mr Newman did not suggest, and it cannot in any event be the case, that a benefit must be payable at once for it to have "accrued" for the purposes of clause 5 of the 1979 Definitive Deed. There is, in other words, no question of the proviso contained in the clause operating only where a benefit has accrued due.”

Mr Evans accepted that in that context he must have meant by the phrase “accrued due” “has fallen due for payment”.

Another example is in the judgment of Mance LJ in a case called Fisher v Harrison [2003] EWCA Civ 1047. This was a case in the Court of Appeal reported in [2003] Pens LR 293 which concerned the impact of section 91 of the Pensions Act 1995 on a purported settlement of some litigation in which the defendants, who were entitled to benefits under a pension scheme, submitted to a consent order under which they consented to the “benefits and proceeds of the B.J. Harrison Limited Retirement Benefits Scheme … being paid out to the solicitors for the first and second plaintiffs/claimants”, and one of the questions that was argued was whether that caught only benefits already due or caught future benefits. That appears from paragraph [28] of Mance LJ’s judgment in which he said there were at least three possible constructions that might be put on the concept of ‘benefits and proceeds’: “(1) they are limited to sums absolutely due by 30^th May 2000”, and then he goes on to set out what the others are, and at paragraph [30], having said that Mr McCarroll (who appeared for one of the parties) submitted before the court that that was the correct analysis, said:

“The verbal force of Mr McCarroll's submission that the consent order should be confined to payments already accrued due lies in the consent which it recites to the benefits and payments ‘being paid out’ …”

It is again perfectly clear that he was there using the phrase “already accrued due” to mean the same as sums absolutely due by 20 May 2000; that is, those which have become due and payable.

Finally, I was also referred by Mr Newman to a case called In re Jenkins (1915) 1 Ch 46 in which Sargant J was considering a forfeiture clause in a will under which income had been left to the testator’s son with a direction that the same should only be paid to him “so long as he shall not attempt to assign or charge the same or do or suffer any act whereby the same may become vested in or payable to any other person; And in the event of a determination of his life interest by any means in his lifetime I direct that the income which otherwise would have been payable to him may be applied by my trustees … in their discretion”.

There were various sums in issue but one of the sums in issue represented an apportioned part of income which had not been received at the date of a mortgage which the life tenant entered into in which he assigned his interest to his sister by way of mortgage, and was received by the trustees after the date of mortgage. The sum of £254.6s.2d represented the apportioned part of the sum later received which was attributable to the period up to the date of the mortgage, and the question was whether that was caught by the forfeiture clause, and Sargant J said at page 51:

“It seems to me impossible to say that the sum of £254.6s.2d, representing the apportioned part of the income down to the date of the mortgage, was or represented any sort of income which had become due or was ready in the hands of the trustees. The very careful language of the clause seems to me to be directed to prevent income being payable to the beneficiary after he has attempted to assign or charge it, and although the words are not sufficient as was decided in In Re Sampson, to prevent his receiving income which was payable and in the hands of the trustees, or income which had actually accrued due, yet it is a very different question when you have to consider whether the trustees are to look to each instalment of income as it becomes due and to consider what are the different periods during which the respective parts of the instalment have been accruing.

The effect of the clause, I think, is to prevent the destination of the income being finally determined until the time when it has actually accrued due, or, in other words, become payable, and to direct the trustees, when dealing with the income, to fix their attention upon the moment of time when the instalment of income has either reached their hands or become payable.”

So he held that the entirety of that sum went under the forfeiture clause and none of it went to the life tenant. It is again perfectly apparent, and I accept, that when Sargant J referred to “actually accrued due” he meant the same as when he says “become payable”. The two were treated by him as similar.

I therefore entirely accept Mr Newman’s submission that the natural and ordinary meaning of the phrase “accrued due” is that it refers to matters that have actually become payable. In the case of certain types of income such as for example rent payable under a lease which is payable quarterly in arrears, it means that a gale of rent which is payable on 24 June has not “accrued due” at any time before 24 June but is “accruing due” but at any time after 25 June has “accrued due”.

In relation to pension schemes the practical effect of that construction would be to give the proviso a very limited effect, as I think Mr Newman in effect accepted. It would mean that the scope of the protection of the proviso would be limited to sums which had fallen due and payable to a member but had not actually been paid. Mr Newman gave an example in his argument of a member who became entitled to a pension on attaining his 65^th birthday and thereby became entitled to payment of a pension, but where for ease of administration the payment date for pensions under the scheme was set at the first of each month. Assume a member who attained 65 on 20 January but whose pension was not paid until 1 February. Between 20 January and the payment on 1 February, the pension, he says, would be a payment which had accrued due but had not been paid. In a well-run scheme one would not anticipate that there would be many examples of payments which had accrued due in that sense but had not been paid.

The way in which final salary schemes work, as is well-understood by those who practise in this field, is that the member by joining the scheme receives a pension which builds up during his pensionable service and the future pension, which is necessarily contingent until the member actually comes to receive his pension by attaining normal pension age or taking early retirement under the rules of the scheme, is not only a future payment but is a contingent payment, but nevertheless it is attributable to the service which he has already given while in pensionable service under the scheme, it being a commonplace that pensions are a species of remuneration for service and indeed a form of deferred pay.

It is in that sense in which it is common to speak of final salary benefits accruing during pensionable service, and that is why if the proviso to an amendment power provides that benefits accrued are protected it is to be regarded as protecting the benefits which a member has built up by his pensionable service to the date of the amendment including (according to the consistent jurisprudence adopted by Millett J in Courage and various other cases on the meaning of the word “secured” and by Newey J on the meaning of the word “accrued” in Briggs v Gleeds) the link to final salary.

It can be seen that the practical consequences of the two constructions are therefore radically different. Neither construction, it seems to me, is entirely free from difficulty. The problem with Mr Evans’s construction is that, as forcefully submitted by Mr Newman, it takes the phrase “accrued due” (which has as I have accepted a natural and ordinary meaning, a well-known ordinary meaning among lawyers) and effectively gives it an entirely different meaning as if the word “due” was not there at all. Indeed, Mr Evans said that his construction either required taking the word “due” and giving it a different meaning (he suggested that it should be given the meaning “to” so that the phrase “benefits accrued due in respect of any member” could be read as “benefits accrued to [and then he had to add the word ‘or’] in respect of any member”) or that it should simply be taken out as a mistake within the principle exemplified by Lord Hoffman in Chartbrook Ltd v Persimmon Homes Ltd [2009] 1 AC 1101, a well-known case on the limits of the court’s power under the guise of construction to correct mistakes. So there are undoubted difficulties with Mr Evans’s construction and it remains hard to see why the draftsman on his construction should have used the word “due” at all.

On the other hand, I do not think that Mr Newman’s construction is itself free from difficulties. The purpose of the clause is evidently to provide a protection for members. That is the purpose of these provisos to powers of amendment. It does seem to me that at first blush it is rather unlikely that the purpose of the clause was intended to protect only payments which had fallen due but not been paid. The pension scheme is a trust and the purpose of a trust is to make payments to beneficiaries. It is the duty of the trustees to make payments to beneficiaries when they are due, and one would have thought it unlikely that the draftsman would have gone to the trouble of protecting a payment which the trustees should already have paid in putting forward this proviso. Indeed, I regard it as rather doubtful whether even in the absence of a proviso it would be a proper use of the amendment power to deprive a beneficiary of a payment which had already fallen due. It is difficult to envisage the circumstances in which trustees, in whom the power is vested (although they need the consent of the principal employer), could take away from a beneficiary a payment which had become his own absolute property and which they should according to the trust have already paid to him.

Moreover, as pointed out by Mr Evans, by the time that the definitive deed was drafted, which was in 1998, section 67 of the Pensions Act 1995 had been in force for some 17 months, coming into force in April 1997. Section 67 of the Pensions Act 1995 imposed statutory restrictions on exercising an amendment power which would affect any entitlement or accrued right of a member, and although at that stage there had been no authoritative exposition of what “entitlement” meant – it was subsequently decided by the Court of Appeal in Aon Trust Corporation v KPMG [2005] EWCA Civ 1004 at [181] per Jonathan Parker LJ that “entitlement” refers to a pension already in payment – it was already apparent, it seems to me, in 1998 that section 67 would (either through the means of protecting entitlements or at the very least through the protection of accrued rights which were defined by section 124(2) of the Pensions Act 1995 as being the rights which had accrued to or in respect of him at the time to future benefits under the scheme, his accrued rights to be determined as if he had opted immediately before that time, to terminate that service, thereby not including a final salary link) in any event prevent an amendment which would have the effect of taking away a member’s right to a payment which had already arisen.

It does seem to me to be a matter of some oddity that on Mr Newman’s construction the draftsman must be assumed to have put in place a proviso to the power of amendment which had no practical effect at all because the effect of section 67 was in any event to prevent such an amendment. Mr Newman suggested that it might have been designed by the draftsman to guard against the possibility of section 67 being subsequently repealed. That seems to me to be prima facie unlikely. The possibility of section 67 being repealed or modified might well have occurred to a draftsman but it seems to me very unlikely that any replacement of section 67 would have been thought to be such as to enable trustees to take away payments which had already arisen.

Mr Newman suggested that it was possible that the draftsman unthinkingly took the precedent from a previous draft. That would be speculation, but one thing one can tell from the form of the deed and rules is that whoever drafted these deed and rules (and they have obviously been professionally drawn) was aware of the Pensions Act 1995 because there are various references to provisions in the 1995 Act scattered through the deed. I do regard the very limited effect of the proviso on Mr Newman’s construction, and the fact that it is difficult to see what in practical terms the draftsman was seeking to achieve by it, as an oddity on his construction. Indeed, although Mr Newman submitted quite rightly that the court is reluctant to adopt a construction which has the effect that words selected by the draftsman are to be ignored as otiose and are to be treated as if they were not there, it being a salutary principle that every word should be given effect to if possible, it is also a salutary principle that one does not assume that a whole provision is of no practical effect. Yet the practical consequence of Mr Newman’s construction is that the entire proviso has no practical effect, at any rate so long as section 67 remains in its current form or in anything approaching it.

There are other difficulties with Mr Newman’s construction. One of them is this. The protection of payments that have already accrued due would only require the trustees to identify what payments had already accrued due. That might or might not require the assistance of an administrator or an actuary to identify what the benefits were. But the proviso requires the trustees to form a view as to whether the alteration substantially reduces in aggregate the value of the benefits accrued due “as to which the decision of the Trustees acting on the advice of the Actuary shall be final”. This contains, it seems to me, a number of indications that what the draftsman had in mind was a package of benefits which might require some valuation.

Firstly, there is the reference to the value of the benefits and the advice of the actuary. It is not at all obvious to me the circumstances in which it would be necessary, once the trustees had identified what payments were due, that the payments would need to be valued. Once a payment has crystallised and become due for payment, its value is whatever the payment is. The very reference to a value and an actuarial value of benefits tends to suggest that what the draftsman had in mind was future and contingent benefits, because, as is well-known, the current value of future and contingent benefits under a pension scheme is something that requires actuarial expertise and is something that requires judgments to be made, unlike the value of a current payment which is currently due.

Secondly, there is the reference to “in aggregate”. Although this was not referred to in argument, it does seem to me that that too is a slight pointer in favour of the draftsman having envisaged a package of benefits. In the normal way a member accrues a number of potential future benefits, both his own pension and contingent pensions for dependants and the like, and it is a standard practice for actuaries to value the entirety of the package of benefits. If it is unlikely that the draftsman had in contemplation a payment which had fallen due but not been paid, it is even more unlikely that the draftsman had in contemplation a number of benefits that needed to be both valued and to have their value aggregated, because although one can see circumstances in which a draftsman could envisage that a member had not been paid a benefit, it is rather less likely that the draftsman was envisaging a package of benefits, a number of which had not been paid and were due and payable.

Thirdly, the reference to substantial reduction in the value again tends, it seems to me, to point away from the draftsman having envisaged a payment which had already fallen due. As I have already said, I regard it as rather unlikely that trustees could properly take away at all the value of benefits which had already become payable and which they should have paid, but it is obvious that the draftsman has envisaged that the benefits which are protected by this proviso are capable of being reduced as long as the reduction is not substantial. That makes sense if what one is valuing is a number of uncertain future and contingent benefits on which actuarial assumptions have to be made because it is not possible to be precise as to the current value of a package at that time. It does not, it seems to me, make anything like as much sense in relation to a payment which has crystallised and become due because, as I have already said, once a payment has become due its quantum is precise and knowable, and I do not see the circumstances in which a draftsman might think it appropriate to protect such a payment but nevertheless allow it to be reduced as long as the reduction was not substantial.

All these features of the clause do suggest to my mind that neither construction can be seen to be the natural and ordinary meaning of the words “benefits accrued due” when taken not just in isolation but when the whole phrase, the whole proviso, is read together. In those circumstances the court is I think obliged to have regard to whether it should on the one hand impose a construction on this clause which respects the natural and ordinary meaning of the words “accrued due” but which has all the curious consequences that I have referred to and which seems to sit very uneasily with the remaining wording of the clause, or whether it should effectively treat the “due” as having been included by mistake or as having no force at all and assimilate the proviso to what it would otherwise mean if the word “due” were not there, which as I say is common ground.

I was referred to what Lord Hoffman s aid in Chartbrook, a very familiar case, as to the ability of the court to correct mistakes by construction. I am not going to read it all because it is quite a long passage and it is very well known, but he does undoubtedly say that there are circumstances (although it requires a strong case and we do not easily accept that people have made linguistic mistakes) where the context and background drives a court to the conclusion that something has gone wrong with the language. He uses a number of other phrases in the judgment that construing a document in accordance (in that case) with the ordinary rules of syntax “makes no commercial sense” or that the language is “arbitrary and irrational”, but he does make it clear that if the court is satisfied that a mistake has been made and that it is clear what the mistake is, that there is no limit to the amount of red ink or verbal rearrangement or correction which the court is allowed.

I was referred by Mr Newman to two cases which are comments or glosses on that, one in the Court of Appeal called Pink Floyd Music Ltd & Anor v EMI Records Ltd [2010] EWCA Civ 1429 where Lord Neuberger (then Master of the Rolls) said at paragraph [20]:

“… as Lord Hoffmann also made clear in Investors Compensation [1998] 1 WLR 896, there is a difference between cases of ambiguity, which may result in giving the words a meaning they can naturally bear, even if it is not their prima facie most natural meaning, and cases of mistake, which may result from concluding that the parties made a mistake and used the wrong words or syntax. However, he emphasised the court does ‘not readily accept that people have made mistakes in formal documents’ - Chartbrook [2009] 1 AC 1101, para 23. He also pointed out in paragraph 20, that, as the court, and therefore the notional reasonable person, cannot take into account the antecedent negotiations, the fact that the natural meaning of the words appears to produce ‘a bad bargain’ for one of the parties or an ‘unduly favourable’ result for another, is not enough to justify the conclusion that something has gone wrong. One is normally looking for an outcome which is ‘arbitrary’ or ‘irrational’, before a mistake argument will run.

21.

Accordingly, before the court can be satisfied that something has gone wrong, the court has to be satisfied both that there has been ‘a clear mistake’ and that it is clear ‘what correction ought to be made’ …”

I do not regard that passage from Pink Floyd as doing anything more than summarising what Lord Hoffman himself said in Chartbrook. I do not regard it as imposing any further limitations on the Chartbrook principle.

The other passage to which I was referred by Mr Newman is in the recent decision of the Supreme Court in Marley v Rawlings [2014] UKSC 2, which was a case where a husband and wife had each executed each other’s wills. In that case an argument was put forward that the will could be corrected by interpretation, and Lord Neuberger, again, now the President of the Supreme Court, had some comments about that at paragraphs [37] to [41] (again, I am not going to read it all out). It is fair to say that he expressed some reservations about the approach of Lord Hoffman. He described at paragraph [37] the sentence “if one would nevertheless conclude from the background that something must have gone wrong with the language, the law does not require judges to attribute to the parties an intention which they plainly could not have had” as “controversial”; he referred to an extrajudicial article by Sir Richard Buxton suggesting that Lord Hoffman’s approach was inconsistent with previously established principles and a comment on that in Lewison on The Interpretation of Contracts, describing it as “making out a powerful case”; and at paragraph [41] he referred to the point as a “difficult point”, that being whether a particular approach should be regarded as interpretation or rectification. But he did go on to say that it was not necessary to decide it upon the appeal:

“Interpretation was not the basis upon which the courts below decided this case and it was not the ground upon which Mr Ham primarily relied. Furthermore, and no doubt because of those points, only limited argument was directed to the issue of whether the issue was one of interpretation or of rectification. For the reasons developed below, I consider that this appeal succeeds on the ground of rectification, so I shall proceed on the basis that it fails on interpretation.”

I fully accept that those comments in Marley v Rawlings express some reservations as to Lord Hoffman’s approach in Chartbrook, but it seems to me that sitting at first instance I am bound by what Lord Hoffman said in Chartbrook (with which the other members of the House of Lords agreed). Unless and until the Supreme Court has departed from it, it seems to me that I should follow what Lord Hoffman says.

That therefore squarely raises the question whether I am satisfied that something has gone wrong with the language and that it can and should be corrected, which requires as I say a strong case in which the result is not just one which is unduly favourable to one side or the other but is one which can be regarded as making no commercial sense and is arbitrary and irrational. I have come to the conclusion that this is a case in which the inclusion of the word “due” can only be attributed to a mistake. That is not because I start with a predisposition as to what provisos to amendment powers should provide, it is primarily because the inclusion of the word “due” and its meaning, giving it its normal and natural and ordinary meaning which was so clearly expounded by Mr Newman, does not fit with the remainder of the proviso in the way that I have sought to explain.

The purpose of these provisos to amendment powers is undoubtedly to protect the benefits of members. As construed by Mr Newman, it would have a very limited effect, as I have sought to explain, and one which seems to be almost entirely redundant given section 67 of the Pensions Act 1995, but more significantly it does not fit with the remainder of the language of the clause and it does seem to me to fall into that category of case where the court can and should conclude that that was not the draftsman’s intention when looking at the words that the draftsman used as a whole.

I therefore construe the proviso in the way submitted by Mr Evans, that is, as if the word “due” was not there or as if it simply meant the benefits accrued for a member, it being common ground that on that reading it does protect the final salary link, with the result that the question that I am asked in the claim form, which is:

“Whether, having regard to the true construction of [that clause], clause 4 of the Deed of Variation relating to the Scheme dated 31 December 2004 validly amended the governing provisions of the Scheme so as to provide that members’ benefits calculated by reference to their Pensionable Salary (as defined) as at the end of 31 December 2004, rather than by reference to their Pensionable Salary as at the earlier of the retirement, the leaving service (as defined) or the death of the member,”

The answer that I give to that is No. Clause 4 did not validly amend the governing provisions of the scheme in that respect.