IBM United Kingdom Pensions Trust Ltd v IBM United Kingdom Holdings Trust Ltd & Ors

This judgment is supplemental to my judgment handed down on 12 October 2012 and should be read with it. Terms defined in that judgment have the same meanings in this judgment. References to paragraph numbers in that judgment appear in this judgment as follows “Judgment [x]”.

For reasons given in Judgment [521] – [523], I did not decide the issue concerning IBM’s Imperial duties. I gave the opportunity (see Judgment [524]) for that issue to be the subject of further written submissions if the parties were unable to resolve the issue in the light of the conclusions which I did reach on the issues argued before me. The parties have made written submissions in accordance with directions which I made after the hand-down. This judgment is given on the basis of those submissions without any additional oral argument. I have received written submissions from Mr Evans on behalf of the Trust Company and from Mr Simmonds and Mr Goldsmith on behalf of IBM.

Before coming to the substance of the matter, I need to rule on what is open for argument. There are three areas of contention:

In Judgment [524] I stated that I would

“decide the issue whether, following rectification, IBM is bound to give consent to amendments to reflect the Preservation Requirements as they stood before the Pensions Act 2004 in respect of service before 6 April 2005.”

I can now see that, in an attempt to make clear what I was intending, I have produced ambiguity. One of the arguments now available to IBM – albeit a fall-back argument – is that, even if the Trust Company is right in its submission that there is a breach of the Imperial duty, there should be a remedy only in relation to service prior to 6 April 2005. I had anticipated, without articulating, that argument when writing what I did. My intent was to allow the Trust Company to argue that the IBM has to give effect to the Preservation Requirements as they stood prior to 6 April 2005 and as they would have applied (necessarily in relation to service before that date) before that date; I did not intend the Trust Company to be prevented from arguing that effect should be given to the Preservation Requirements as they stood prior to date. I will proceed on that basis.

The Trust Company has included in its submissions an argument which is different from the issue concerning IBM’s Imperial duty. It is an argument based on the equitable maxim “equity sees as done that which ought to be done”. This is an argument which I did not address in my judgment and it is not one which falls within the scope of Judgment [524]. Mr Simmonds says that I should not now entertain it having delivered a final judgment reserving only one point; the argument was not pursued at the hearing and should not now be capable of resurrection otherwise the principle that judgments bring finality would be undermined.

Mr Evans is no doubt correct to say that the point is not an entirely new point and identifies occasions when it was mentioned:

What I actually said about confining submissions was this:

“Before you go on, I have got a massive amount of material. I have read it all, it's very helpful, but to pretend I'm on top of it all, it would be dishonest. On the other hand, I don't want you to repeat in your oral submissions stuff that is in there which, in effect, means, apart from new material which Mr Sawyer is going to take me to, you really can take the overview very, very quickly. I have got an overview from having read it, and I'm a bit sceptical about what you can add at all. I have got detailed submissions on preservation. I don't know what Mr Stallworthy is going to say, but again, I'm not going to find it very helpful to have a slightly different spin on what's in writing. I have got enough material to have to deal with, without having it put a different way,”

Mr Evans’ submission would give a reader, unfamiliar with the case, the impression that the equitable maxim issue had been identified and argued in the written opening and adopted in the written closing without the Court needing more. If that was the view taken about this issue – justifying no mention at all being made of it in oral closing – then I venture to suggest that, had the same approach been taken to all of the other points which were raised, the oral submissions on behalf of the Trust Company would have been a very great deal shorter than they actually were. Be that as it may, the point, having been identified in the pleading and in the written opening, was not pursued. It would be idle to pretend that what was said in the written opening was any sort of reasoned argument with only a mention of this being the “appropriate territory” for the application of the maxim. Nowhere in oral submissions was the decision in HR Trustees v Wembley mentioned; I was not even given references within the decision which might help me. I was not referred to the cases which gave rise to the maxim or even taken to leading textbooks such as Snell.

In those circumstances, I did not say anything about the point in my judgment. I do not consider that it was my duty to do and I make no apology for not doing so. Were it not for the fact that I invited further argument about IBM’s Imperial duty, it is very difficult to believe that the Trust Company would now be seeking to raise the point, saying that I had failed to deal with a relevant issue which had been raised in argument. That would be to attempt to re-open my decision. Further, if it is the Trust Company’s view that I should have dealt with the point, the time to raise that with me was when suggesting corrections, in the usual way, to my draft judgment.

There is one other point I should deal with. Mr Evans says that the argument relates directly to the issue identified in Judgment [524] (namely whether IBM is bound to give its consent to a rule amendment to reflect the Preservation Requirements prior to 6 April 2005). He says this is the occasion on which the Court could and should consider it. If he is suggesting that my invitation for further submissions went beyond submissions relation to the scope of IBM’s Imperial duties, I disagree. What I said in Judgement [524] was directed, and clearly directed, only at Imperial duties. It would be wrong to take my identification of the issue out of context and to read it as covering every possible argument which could be raised that IBM is bound to consent to a relevant amendment. The time for raising and pursuing those other arguments was at trial.

My conclusion is that the Trust Company is not entitled to pursue its argument based on the equitable maxim.

Mr Simmonds submits that I should not consider any further argument on the meaning of Clause 13. He says that I rehearsed the arguments and decided, in Judgment [513], not to decide the point. That is correct in the sense that I did not need to decide the point in the context in which I had been addressing it. It will be remembered that, in Judgment [508], I concluded that there is no subsisting power under Clauses 12 or Clause 13 to amend the C Plan or to give an undertaking in relation to it so as to provide a deferred member with a right to an unreduced pension before age 63. But I went on, in Judgment [509] to [512], to consider what the position would be if my conclusion in Judgment [508] was wrong. Since the true meaning of Clause 13 was a difficult issue, I determined in the end not to decide it since a decision would not affect the result – although had I decided it in favour of IBM, it would have provided an alternative reason for rejecting the Trust Company’s contractual case.

The position is now different. Part of the Trust Company’s argument on the scope of IBM’s Imperial duties rests on the proposition that Clause 13 casts an obligation on IBM to consent to an amendment to bring the C Plan into conformity with the Preservation Requirements; in particular, if the C Plan had been rectified in, say 1993, rather than in 2012, IBM would have been obliged, pursuant to Clause 13, to give its consent to the amendment which the Preservation Requirements required the Trust Company to effect. The argument is that, by failing now to consent to an amendment to give effect to what should have happened years ago, if the C Plan had been in its correct, rectified, form, IBM is in breach of its Imperial duties. It is clear to me that the Trust Company should be entitled to run its argument on Clause 13 in order to lay the ground for its case on the Imperial duties. My decision in the main judgment not to decide the point in a different context has no impact on what should be allowed now.

As well as a dispute about the application of the law to the facts of the present case, there is a dispute about precisely what the scope of the Imperial duties actually is. Mr Simmonds submits that it is uncontroversial to say that, when considering an employer’s Imperial duties, the test is not whether a particular course of action on the employer’s part would be unreasonable but rather the test is whether a decision to take such a course of action would be irrational or perverse. He can find support for that in the Prudential case [2011] PLR 239 where Newey J reviewed the authorities and in the decision of Burton J (referred to by Newey J at [135] of his judgment) in Clark v Nomura International plc [2000] IRLR 766. That latter case is perhaps the high point, expressly deciding, in the context of an award of a discretionary bonus, that the test was one of “irrationality or perversity” that is to say that “no reasonable employer would have exercised his discretion in this way”.

He is either right or wrong to identify the test in that way, but I fear that he is not correct in saying the matter is uncontroversial. It is hotly challenged by Mr Evans who says that it is not merely controversial but obviously wrong. Mr Evans starts with Imperial Tobacco itself where the duty was stated (see [1991] 1 WLR 589 at 597G) in these quite general terms:

“employers will not, without reasonable and proper cause, conduct themselves in a manner calculated or likely to destroy or seriously damage the relationship of confidence between employer and employee”

Things have, however, moved on since then. As Newey J points out in the Prudential case at [141], “it would make no sense to freeze-frame the duty of trust and confidence as it appeared at the date of Browne-Wilkinson V-C’s decision”. Mr Evanspoints out that the assessment is objective and highly fact specific, relying on what was said by Lord Nicholls in Malik v BCCI SA [1998] AC 20 at 35.He submits that the focus must always be on whether the conduct of the employer is such that it might destroy or seriously damage the relationship of trust and confidence. Indeed, irrationality and perversity is an inappropriate test, he would say, because it is not a sufficient condition. There can be cases where there is no breach of the Imperial duty even though the exercise of a discretion may be irrational and perverse: that could be so where the decision relates to a comparatively trivial matter where the irrational and perverse decision does not result in that serious damage to the employment or pensions relationship. Nor is it a necessary condition: Mr Evans relies on Hillsdown plc v Pensions Ombudsman [1997] 1 All ER 862 where, he submits, Knox J found there to be a breach of the Imperial duties notwithstanding that the employer’s conduct was not characterised as capricious, perverse or irrational. I do not propose to go into a detailed analysis of precisely what Knox J did or did not decide, remarking only that there were elements of breach of trust on the part of the trustee in making the relevant transfer.

Nonetheless, four things can be said with confidence in the light of Newey J’s judgment. First, the exercise of a discretion such as the discretion in the present case vested in IBM (whether or not to consent to an amendment pursuant to Clause 13) requires “a genuine and rational as opposed to an empty or irrational, exercise of the discretion”. In other words, as on Mr Simmonds’ test, IBM must consent, or be treated as consenting, if to refuse consent would be irrational or perverse. Secondly, the correct test is not one of fairness. Browne-Wilkinson V-C himself rejected that test and Newey J reached the same conclusion (see for instance at [142] of his judgment). Thus, as Newey J puts it, assessing whether a decision is irrational or perverse is not to be equated with the application of an objective standard of reasonableness. Thirdly, whatever the test is, it is a “severe” one: see [132] of the judgment, citing Hale LJ in Gregory v Hertfordshire CC [2000] IRLR 703. What Lady Hale meant by severe was that the conduct must be such to destroy or seriously damage the relationship so perhaps that takes one no further. Fourthly, the test, whatever it is, is objective. The observations made by Newey J in the context of the objective and subjective approaches (starting at [151] of his judgment) focus on the state of mind of the employer. That is obviously correct when looking at the “calculated to” damage the relationship. It is of less obvious relevance in relation to the “likely to” limb of the test. Here, it is perhaps the state of mind of the employee or pension scheme beneficiary which is more relevant. But as with the employer, I consider that an objective test must be applied (assuming that Mr Evans’ test, rather than Mr Simmonds’ test, is to be adopted since if Mr Simmonds’ test is correct, it is hard to see how the state of mind of the employee or beneficiary could come into the equation).

Given those considerations, it is no more easy for Mr Evans to categorise Mr Simmonds’ test as “obviously wrong” than it is for Mr Simmonds to categorise his own test as “uncontroversial”. There is force in Mr Simmonds’ submission about the correct test given the focus of Newey J on irrationality and perversity with “otherwise in breach of the obligation of good faith” appearing only as a sweep-up reference in a number of places (eg [183] and [187] of the judgment). Further, it might be said that, if an employer is not acting irrationally or perversely, he is acting within the range of decisions which an employer can reasonably make. And if that is so, it could not, so the argument would run, be possible objectively for the decision to be seen as one which could seriously damage the employment relationship. On this argument, it would be unreasonable, objectively, for an employee or beneficiary to treat the employer’s decision as destroying or significantly damaging the relationship. Hillsdown may not provide a fatal objection to that argument.

Mr Evans starts with the proposition that an order for rectification has fully retrospective effect. It follows that active members of the C plan have always had a right to an unreduced pension from age 60. This is the effect of the relation back of the order for rectification. Reliance is placed on Munt v Beasley [2006] EWCA Civ 370. This case concerned rectification of a lease which conferred on Mr Munt the right to possession of a loft and defeated Mr Beasley’s claim for trespass.

And so in the present case, and taking into account IBM’s obligations under Clause 13 and the Preservation Requirements as they stood between 14 December 1983 and 6 April 2005, compliance with the Preservation Requirements required IBM up until 6 April 2005 to confer on deferred C Plan members the right to an unreduced pension from age 60. Clause 13 precluded IBM from refusing consent. Reliance is placed on the analysis in Judgment [511].

The counterargument suggested in Judgment [513] should be rejected for two reasons:

As a result of my decision to order rectification in relation to active members, the rules of the C Plan have all times since 1983 conferred a right to an unreduced pension from age 60 for active members. The effect of the Preservation Requirements until 6 April 2005 was that the Trust Company was obliged, under section 132 Pension Schemes Act 1993 to take steps to confer on deferred C Plan members the same right to an unreduced pension from age 60. The effect of Clause 13 was that IBM could not lawfully refuse consent to a rule amendment whose effect was limited to that which was necessary to achieve conformity with the Preservation Requirements. It follows that IBM could have been compelled to consent to a rule amendment prior to 6 April 2005.

The right to short service benefit was provided for by Parliament; and the C Plan was established by IBM on the clearly expressed basis that it would comply with the Preservation Requirements. The Trust Company’s request is that IBM agree to an amendment the sole purpose and effect of which is to secure for members the rights which at all times from 1983 to 2005 Parliament had decided they should have. It is a request for consent to an amendment which would achieve the result that IBM’s professional advisers knew in 1983 was required if it was intended to confer a right to flexible retirement on active members. I would add, however, that the decision-makers did not know of this requirement.

IBM is required to consent to the amendment. Any other result would effectively allow IBM to take advantage of its own wrong or error in designing and implementing a benefit structure in direct conflict with the Preservation Requirements. The relationship of trust and confidence between IBM and C Plan members would now unquestionably be seriously damaged by IBM in

That is a summary of Mr Evans’ further submissions. Mr Simmonds provided his further submissions in response, which I summarise in the next section of this judgment. Mr Evans has provided reply submissions: I do not propose to deal with those separately but will cover the points which he makes, so far as necessary, in the Discussion section of this judgment.

Basing himself on the test identified by him as relevant (and which I have already discussed, without so far resolving, above) and on the (correct) proposition that IBM is entitled to make decisions having regard to its own financial interests, Mr Simmonds makes the following submissions.

To begin with, a challenge is made to two propositions put forward by Mr Evans namely (i) that prior to 6 April 2005, IBM was under an obligation to take steps to confer or to consent to the conferral on deferred members the same right as those enjoyed by active members to an unreduced pension at age 60 and (ii) that the result contended for by IBM would “effectively allow IBM to take advantage of its own wrong”.

So far as concerns (i), Clause 13 did not impose such an obligation on IBM. The submissions made at trial (summarised in Judgment [512]) are maintained: the effect of Clause 13 was not to impose such an obligation. In support of that it is argued as follows:

Mr Simmonds submits that I should not, in any event, now determine the true construction of Clause 13 having decided not to do so in my main judgment. I have already dealt with, and rejected, that submission.

So far as concerns (ii) (taking advantage of own wrong), there was and is no “wrong” of which advantage can be taken. Clause 13 did not have the effect for which the Trust Company contends. Even if it did have that effect, IBM’s obligation is only triggered by a request for consent to an amendment necessary to achieve conformity with the Preservation Requirements as they stood at the time of the request. No request was made before 6 April 2005 and the Current Trust Deed and Rules conform, so that no request can now properly be made. The only party with an obligation under the Preservation Requirements was the Trust Company; no obligation could have arisen on the part of IBM unless and until a request to consent to an amendment had been made. Therefore, as a matter of fact, there has been no “wrong” on the part of IBM.

In any case, the Trust Company is taking a one-sided view of a counterfactual situation. As a result, it is mischaracterising who, if anyone, can be saidto be taking advantage of another’s wrong or error. Thus it is the Trust Company (on behalf of the members) which is seeking to take advantage of an error (shared by the Trust Company itself). The Trust Company’s argument ignores what might have happened – indeed what is likely to have happened – from 1983 onwards in the event that it had come to light at an earlier juncture that the trust deed and rules were liable to be rectified as I have held. That would, presumably, have resulted in the Trust Company seeking to comply with the Preservation Requirements. Compliance would have resulted in (i) increased cost and (ii) the creation of a benefit structure including a right for deferred members to take an unreduced pension at age 60 which IBM had never intended. It is therefore quite possible that IBM would have taken steps to bring about a termination of future accrual under the rectified Trust Deed and Rules (then compliant with the Preservation Requirements) earlier than in fact occurred. In ignoring this aspect of the counter-factual situation (ie a situation in which the error is treated as having been identified and rectified), the Trust Company is seeking to take advantage of an error (in which it shared) by seeking to draw advantage from a state of affairs that, in all likelihood, would not have continued to subsist if the parties had not been in error.

According to Mr Simmonds, IBM’s calculations show that the additional cost of the amendment proposed by the Trust Company would be between £175m and £240m. The Trust Company does not object to reference being made to that figure. Mr Simmonds, correctly, points out that IBM is entitled to take into account its own financial interests in considering whether it can properly refuse its consent to an amendment. But I am bound to say that I do not gain much assistance from the reference to that figure. It is obviously a not insignificant sum of money but I have no idea at all of the impact of that cost on IBM’s business or of the seriousness of that cost on its balance sheet or profit and loss account. I do not even know over what period IBM would have to contribute that sum. If it is over 10 years, for instance, I have no idea what impact the payment of, say, £17.5m to £24m in the first year, and varying amounts thereafter, would have. I can accept that the figure is large enough to preclude the Trust Company from saying that the financial impact is irrelevant or even not at all important; but I cannot, in this action, assume that payment of this amount is somehow critical to IBM’s future business.

Mr Simmonds correctly identifies that it is common ground that IBM was not, and could not have been, in breach of any of its Imperial duties prior to the making of a request to consent to an amendment which took place after receipt of my main judgment. He says that not only is there no subsisting statutory requirement to underpin the request (so that it is not a proper request under Clause 13) but also “the cost to IBM would be huge (and would provide no countervailing benefit for either IBM’s current employees or its shareholders)”. Compliance with the Imperial duties is to be judged in the light of circumstances actually obtaining when the duties are alleged to have been breached (a proposition which I think is clearly correct); it cannot possibly be said that a refusal by IBM to give consent would be unreasonable, let alone perverse or irrational. Accordingly, IBM is not in breach of those duties by reason of its refusal to consent to the Trust Company’s request.

So far as Clause 13 is concerned, I do not consider that IBM would have been entitled under the similar provision contained in the 1990 Trust Deed and Rules (see Clause 14 Schedule C) simply to refuse its consent to any amendment whatsoever. In deciding whether to consent to a particular amendment requested by the Trust Company, IBM would have been subject to Imperial duties. It might have had good reasons for refusing the particular amendment sought by the Trust Company. For instance, the amendment sought might have gone beyond what was strictly necessary to comply with the Preservation Requirements; or there may have been different ways of achieving compliance (there might, for instance, be disagreement about what alternatives might be provided under section 73 Pensions Schemes Act 1993) the with IBM refusing to agree to the amendment proposed by the Trust Company but with the Trust Company refusing to agree to a counterproposal put forward by IBM. Any impasse could then be resolved by the OPB. I do not consider that IBM could have simply refused to consent to a proposed amendment without perhaps even giving it due consideration and then successfully argued that it was not in breach of its Imperial duties because the matter could be resolved by the OPB. Its conduct in irrationally and perversely refusing to properly consider whether to agree to the amendment, rejecting the proposal out of hand, would be destructive of the relationship of trust and confidence whether or not there was ultimate recourse to the OPB.

I have just said that the amendment might have gone beyond what was strictly necessary. Mr Simmonds suggests that it cannot have been the purpose of the consent requirement to ensure that amendments are limited to those necessary to achieve compliance with the relevant regulatory requirement. It was not, I accept, the only purpose, but it could be at least one of the purposes. Mr Simmonds says that that is not so since the power of the Trust Company only extended to making necessary amendments in the first place without the consent requirement. I disagree with Mr Simmonds’ analysis. It is possible to envisage occasions where a dispute arises between the Trust Company and IBM about what is necessary; without the consent requirement, the Trust Company would be able to effect its proposed amendment leaving IBM to challenge the amendment in the court. Further, there might be occasions (as already mentioned) where compliance with the Preservation Requirements could be achieved in more than one way; without the consent requirement, the Trust Company would be able to impose its preferred amendment. The possibility of resolution of the dispute by the OPB does not, I consider, detract from the conclusion that at least one purpose of Clause 13 is to ensure that amendments are limited to necessary amendments.

That conclusion makes it unnecessary to consider the dispute about whether the meaning of the provision was different in the Trust Deed and Rules post-dating the abolition of the OPB from its previous meaning since there is no difference in meaning.

As to Mr Evans’ “relation back” argument in respect of an order for rectification or, indeed, a right to obtain rectification vindicated by an eventual order, that may be relevant to the equitable maxim issue (an issue which I have decided he is not entitled to raise) but it is not, so it seems to me, relevant to the scope of IBM’s Imperial duties. If the Imperial duties compel IBM now to give its consent to the amendment requested by the Trust Company, it is not because of a technical effect of an order for rectification but because IBM would be damaging the relationship of trust and confidence by refusing to provide what would, on the Trust Company’s case, have been the right of deferred members if the C Plan had been compliant with the Preservation Requirements from the beginning. According to the Trust Company’s case, that right would have been to payment of a deferred pension unreduced at age 60.

The Trust Company's approach seems to elide the underlying error made in 1983 and the mistake which is the relevant mistake for the purposes of rectification. The underlying error was the failure by the decision-makers to take account of the Preservation Requirements because their effect was not drawn to their attention; that is different from the mistake which has led to the successful rectification claim, namely the failure to include flexible retirement in the 1983 Trust Deed and Rules. If the error had not occurred – if the decision-makers had been fully advised – the mistake would not have occurred either. It is pure speculation what IBM would then have done in relation to the implementation of the C Plan. It would have been, at least to some extent, “back to the drawing board” because “powers reserved” would surely have necessitated a referral back to Armonk.

It is therefore speculation precisely what benefits anyone would have received if the error had been spotted in 1983. Indeed, it is speculation what the benefits would have been if the error had been spotted later, whether before or after the OPB had been abolished. It was at least a possibility – Mr Simmonds suggests likelihood, but that is speculation – that IBM would have terminated further accrual on the basis of flexible retirement.

But what is not speculation is that neither the Trust Company nor IBM intended the C Plan to provide for payment of an unreduced deferred pension at age 60 or understood that it would do so: that the C Plan would have had to provide for that benefit if it provided for flexible retirement was a consequence, unforeseen by the Trust Company and by IBM, of the lack of understanding of the Preservation Requirement on the part of the decision-makers.

In my judgment, whatever test is to be applied in establishing the scope of IBM’s Imperial duties, there is no breach of those duties on the part of IBM in refusing consent to an amendment which was not, at the time of the recent request to make it, required to give effect to the Preservation Requirements as they stood at the date of that request. It is not unreasonable, in my judgment, for IBM to refuse to give its consent; it is, on any view, not irrational or perverse for it to refuse to do so; nor on an objective view could it be said, I consider, that such a refusal would destroy or seriously damage the relationship of trust and confidence between IBM and the beneficiaries of the C Plan. To refuse its consent will be to allow effect to be given to the intention of IBM and the Trust Company when the C Plan was introduced; it will be to give effect to the announcements to staff made before the commencement of the C Plan; and it will be to give effect to the various iterations of the Trust Deed and Rules and the handbooks. There is no-one – or at least no-one has been identified – who believed that they would be entitled to payment of an unreduced deferred pension payable at age 60 if they fell into the category of a deferred pensioner at that age.

I reject Mr Evans’ submission that the IBM would be in breach of its Imperial duties by refusing now to consent to something which, on one view, it might have been compelled to consent to in, say, 2005 before the change in the law. It is arguable that IBM could have been compelled to consent at that time to an amendment such as that now sought by the Trust Company; but even if that is so – I do not need to decide – it would have been because IBM would then have been under an obligation (whether contractual or pursuant to its Imperial duties) to bring about compliance with the Preservation Requirements as they then stood. The change in the law, however, changed the landscape in a fundamental way; IBM was no longer shut out from relying on the original error as a valid reason for refusing to formalise, by way of amendment, the unforeseen consequence of that error.

I do not consider that IBM can, in any sense, be said to be taking advantage of its own wrong or error. Mr Evans accepts that there has certainly been no wrong in the sense of a wrongful refusal in the past to consent to a rule amendment (ie prior to the request by the Trust Company to consent to the proposed amendment), a proposition which has never formed part of Mr Evans’ case. He also accepts that there was no wrong in the sense of contravention on the part of IBM of any duty which had arisen in the past: there was no statutory duty to achieve compliance and no breach of Clause 13 (or its predecessor provisions) in the absence of a request by the Trust Company to consent to an amendment. The “wrong” or “error” was, he says, the implementation and maintaining of a benefit structure which did not comply with the Preservation Requirements. He therefore makes no criticism of IBM for its failure to achieve compliance in the past. But what he does say is that IBM now wishes to refuse to put right the errors of the past by taking advantage of the adventitious effect of subsequent legislation, a refusal which would seriously damage the relationship of trust and confidence in issue.

That, in my view, is to look at matters the wrong way. To repeat the point in slightly different words: the intention of both IBM and the Trust Company in 1983 was to provide for flexible retirement and for payment of an unreduced deferred pension only at age 63. The decision-makers were in error in failing to appreciate that it was not possible to achieve both of those results consistently with the Preservation Requirements. The consequence of the C Plan providing for flexible retirement is that it must comply with the Preservation Requirements. It does so, since there is currently no statutory obligation resting on either IBM or the Trust Company to amend the C Plan to provide for the payment of unreduced deferred pensions at age 60. Nor is there any subsisting contractual obligation on IBM to consent to such an amendment. In my judgment, there is no breach of IBM’s Imperial duties either, in its now refusing to consent to an amendment to provide more than that which (i) it originally intended should be provided (ii) the Trust Company originally intended should be provided and (iii) the members were told consistently from 1983 onwards would be provided. The fact that, in the past, legislative provisions would have required provision of more, does not mean that more must now be provided when neither statute nor contract requires it.

There is one final matter which I would mention. As held in the main judgment, Clause 13 is not now available to bring about compliance with the Preservation Requirements as they stood in the past. It is not, therefore, easy to see how that power could nonetheless be invoked even if IBM was under some sort of obligation, pursuant to its Imperial duties, to do whatever it can to bring about such compliance. The Clause 13 power is simply not available so that IBM cannot be in breach of any duty by refusing to consent to the exercise of that power. Such compliance would have to be brought about in some other way; the only way, so far as I can see, is through an exercise of the general amendment power in Clause 1 Part II of the Current Trust Deed and Rules. The exercise of that power requires the involvement of the Actuary. In the light of my conclusions on the matter of principle, this somewhat peripheral point does not arise, but I mention it so that it is not overlooked should this matter go further.

In my judgment, IBM is not required by its Imperial duties to consent to the amendment which the Trust Company has requested. Nor, I should add for completeness, can the Trust Company make the amendment without that consent.