Griffin v Plymouth Hospital NHS Trust

Neutral Citation Number: [2014] EWCA Civ 1240 IN THE COURT OF APPEAL (CIVIL DIVISION) ON APPEAL FROM The Employment Appeal Tribunal HH Judge McMullen QC and two lay members UKEAT055412LA Case No: A2/2013/2977 Royal Courts of Justice Strand, London, WC2A 2LL Date: Friday 19 th September 2014 Before : LADY JUSTICE GLOSTER LORD JUSTICE UNDERHILL and LORD JUSTICE FLOYD - - - - - - - - - - - - - - - - - - - - - Between : SHARAN GRIFFIN - and - PLYMOUTH HOSPITAL NHS TRUST Appellant Respondent - - - - - - - - - - - - - - - - - - - - - (Transcript of the Handed Down Judgment of WordWave International Limited A Merrill Communications Company 165 Fleet Street, London EC4A 2DY Tel No: 020 7404 1400, Fax No: 020 7831 8838 Official Shorthand Writers to the Court) - - - - - - - - - - - - - - - - - - - - - Mr Joel Donovan QC and Mr Christopher Milsom (instructed by Mishcon De Reya) for the Appellant Ms Elizabeth Cunningham (instructed by Bevan Brittan LLP) for the Respondent Hearing dates : 12th June 2014 - - - - - - - - - - - - - - - - - - - - - Judgment Lord Justice Underhill : INTRODUCTION 1. 2. 3. 4. 5. 6. The Claimant in these proceedings, the Appellant before us, is aged 39. She was employed by the Respondent Trust from 2000 until her resignation with effect from 30 September 2009. She was a specialist clinical technician in bone densitometry at Derriford Hospital in Plymouth. She was at the date of the termination of her employment on a salary of £32,753 p.a. She became ill in 2007 and was diagnosed in 2008 as suffering from systemic lupus erythematosus (“SLE”). Her eventual resignation was on the basis that the Trust had made wholly inadequate efforts to enable her to return to work in a way which would accommodate her condition. The Claimant brought proceedings in the Employment Tribunal for (constructive) unfair dismissal and for disability discrimination on the basis of a breach by the Trust of its duty under section 4A of the Disability Discrimination Act 1995 to make reasonable adjustments. That claim was upheld by a Tribunal sitting at Plymouth, chaired by Employment Judge Carstairs. Following a separate remedy hearing on 15 December 2010, the same Tribunal on 5 April 2011 awarded the Claimant compensation amounting to £105,643.01 (including an element by way of grossing-up and an uplift under section 31 (3) of the Employment Act 2002). The Claimant believed that the element in the award in respect of financial loss was too low. She appealed to the Employment Appeal Tribunal. On 18 November 2011 a Tribunal chaired by Supperstone J allowed her appeal and remitted the case to the Employment Tribunal to make further findings and to reconsider its award on the basis of those findings: I will refer to this as “the Supperstone appeal”. Following a further hearing on 17 May 2012 the original Tribunal, by a Judgment and Reasons sent to the parties on 8 August 2012, increased the award to £166,595 1, the element for financial loss being increased to £104,762. The Claimant believed that that figure remained too low. She appealed again to the EAT. On 26 September 2013 a Tribunal chaired by HH Judge McMullen QC dismissed her appeal: I will refer to this as “the McMullen appeal”. She appeals against that decision with the permission of Rimer LJ. The Trust has at all times been represented by Ms Elizabeth Cunningham of counsel. The Claimant’s representation has changed from time to time. At the original liability hearing she was represented by Dr Nicola Isaacs of counsel. At the first remedy hearing she was represented by Mr Tim Pullen of counsel. On the Supperstone appeal a representative from the Free Representation Unit appeared for her. At the second remedy hearing she was unrepresented. On the McMullen appeal she was represented by Mr Christopher Milsom of counsel. Mr Milsom again appears for her before us, led by Mr Joel Donovan QC. Mr Donovan and Mr Milsom have appeared pro bono and deserve the thanks of the Court. In the actual judgment by an oversight the wrong total was given; but the error has been corrected. THE FINANCIAL LOSS AWARD 7. In the first remedy decision the Employment Tribunal made two findings of fact which were fundamental to its assessment of the financial loss which the Claimant had suffered. First, it found that notwithstanding her illness she would if the Trust had made reasonable adjustments have been able to continue in employment with it indefinitely, but only on the basis of a 25-hour week. Secondly, it found that, in the events which had happened, it would not now be possible for her to return to the Trust and that, since it was not reasonable to expect her to leave the Plymouth area and the Trust was the only employer in that area with a need for her specialist skills, she would not be able to find other employment as a clinical technician. On the basis of those two findings its assessment of her loss was as follows. Past loss of earnings 8. The Claimant started to look for alternative employment in January 2010, and the Employment Tribunal at the first remedy hearing had a good deal of evidence about what work she was capable of doing, what jobs were available (both in the NHS and elsewhere) in the Plymouth area, and her efforts to find employment, which had not up to that point been successful. The Tribunal accepted that she had made reasonable efforts to mitigate her loss and that she should recover the entirety of her loss of earnings in the period up to date of hearing (on the basis of a 25-hour week). That part of the award is not controversial and I need say no more about it. Future loss of earnings 9. The Tribunal considered the issue of future loss of earnings at paras. 5.3.2-5 of its original remedy Reasons. After referring to various factors affecting the assessment it held that she was “likely to obtain suitable alternative employment at 25 hours per week in a year’s time”2; and it awarded one year’s loss of earnings, being £15,201.48, on that basis. At the risk of spelling out the obvious, that is not a finding that it was more probable than not that the Claimant would find a job after precisely one year. Rather, it is an estimate, made on the assumption that the Claimant continued to make reasonable efforts to mitigate her loss, of the mid-point of the probabilities. In Wardle v Crédit Agricole Corporate and Investment Bank [2011] ICR 1290, at para. 52 (pp. 1306-7) Elias LJ said: “In the normal case, if a tribunal assesses that the employee is likely to get an equivalent job by a specific date, that will encompass the possibility that he might be lucky and secure the job earlier, in which case he will receive more in compensation than his actual loss, or he might be unlucky and find the job later than predicted, in which case he will receive less than his actual loss. The tribunal’s best estimate ought in principle to provide the appropriate compensation. The various outcomes are factored into the conclusion. In practice the speculative nature of the exercise means that the tribunal’s prediction will There was some discussion before us about what precise date “a year’s time” connoted. It was suggested it was 27 January 2012, being one year from the date of the Tribunal’s meeting in chambers; but in fact I think it must be a year following the hearing, i.e. 15 December 2011. However, nothing significant turns on this. rarely be accurate. But it is the best solution which the law, seeking finality at the point where the court awards compensation, can provide.” It is, however, convenient to refer to it, as the Tribunal did, as the date on which it was likely that she would obtain employment. The Tribunal made no finding about the nature of the job that the Claimant was likely to obtain. Nor, surprisingly, did it make any express finding about the level of remuneration. But since no award for future loss of earnings was made beyond that first year the necessary inference was that her new employment would be on no less favourable terms than her job with the Trust. The Supperstone Tribunal held that the Employment Tribunal had erred in failing to consider what level of remuneration the Claimant might receive in her new employment. It dealt with the point succinctly, saying only: 10. 11. “6. ... We are not satisfied that the Tribunal gave proper consideration to, and made proper findings of fact, in relation to any continuing loss that the Appellant may suffer after she obtains suitable alternative employment, which the Tribunal found would be in a year of the Tribunal hearing. 7. In particular, the Tribunal made no finding, as to the salary that the Appellant would be earning, for the purposes of considering whether there is any continuing loss and calculating any such sum. In our view, this issue should be remitted to the same Employment Tribunal for determination.” The relevant part of the order reads: “… that the matter be remitted to the same Employment Tribunal to review the Tribunal’s decision on (1) continuing loss of earnings …”. 12. On the remitted hearing the Employment Tribunal accepted a submission from Ms Cunningham that (as recorded at para. 3.1 of the Reasons): “… the Appeal Tribunal had not directed that the Tribunal should rehear the matter. Instead, the matter referred back by the Appeal Tribunal should be re-argued and any new decision should be reached as if made on the day of the judgment referred back to the Tribunal.” Accordingly it heard no further evidence and proceeded on the basis of the evidence and submissions at the first remedy hearing, together with further written and oral submissions from the Claimant and Ms Cunningham3. That course involved it The Claimant’s written submissions were of remarkably high quality for a non-lawyer. We were told that they were the work of her husband, who is also not a lawyer, though drawing on the submissions of Mr Pullen at the first hearing (which were themselves admirably clear and comprehensive). refusing to consider some further documents which the Claimant wanted to adduce; but it will be more convenient if I return to that point later. 13. The delivery of the Tribunal’s decision took an unusual course. The Employment Judge intended to give the judgment and reasons orally at the conclusion of the hearing. However the Claimant intervened in the course of his doing so to complain that the Tribunal had overlooked an important piece of evidence: again, I will return to the details later. But the intervention led to the Tribunal deciding to reserve its decision, which was in the event only sent to the parties some three months later. 14. The Tribunal’s findings on the remitted issue are at paras. 3.2-3.6 of the Reasons. At para. 3.2 it said: “The Tribunal noted that the claimant had applied for positions with pro-rata salaries of between £11,000 and £22,000. The Tribunal was satisfied, at the time of its original remedy decision, that there was no satisfactory evidence that there were any other positions outside that range which were suitable for the claimant.” It went on to repeat its previous findings that the Claimant could not return to her previous job or to any technical role; but it said that she “did have transferable skills of administrative and clerical natures”. It continued: “3.3. The Tribunal had no evidence before it as to how the claimant could obtain promotion in an administrative role but concluded that the claimant’s ability would assist in that respect. The claimant is very intelligent and capable and is also a determined individual who, the Tribunal would expect, would impress a future employer. 3.4. The Tribunal would have expected her to obtain work as already set out in the Tribunal’s original decision. However with regard to the market at the time the Tribunal is of the view that the claimant would have gone for any available job and not only jobs paying anything comparable to what she earned with the respondent. 3.5. The Tribunal has concluded, in the light of the above, that it accepts the submissions made by Mr Pullen at the original remedy hearing, adopted by the claimant, that she would have obtained a job at £18,000 pro rata. However, because of the claimant’s qualities, the Tribunal is of the view that she would have progressed so that she would have been able to move to management and increase her salary to £25,0000 pro rata after five years and then to £30,000 pro rata after a further five years. 3.6. The Tribunal has also concluded that the claimant’s career progression will be such that after a further two years she will achieve parity with her salary with the respondent.” In short, it found that the Claimant would gradually return to her previous level of earnings, by rising through an administrative role into management; but that that would take twelve years, and its “stepped” approach was intended to reflect that gradual process. The figures are worked out at para. 4.2 of the Reasons and produce a total of £43,196.51. 15. There are two further paragraphs – 3.7 and 3.8 – addressing the point made by the Claimant in her interjection during the oral reasons, but it is more convenient to set these out later (see para. 35 below). Pension Loss 16. The Claimant was a member of the NHS pension scheme. This was a final salary scheme of the kind now rarely found outside the public sector. The loss of her employment plainly meant a diminution in her rights under the scheme. Both parties in their submissions for the first hearing referred, as is standard, to the third (2003) edition of the booklet Compensation for Loss of Pension Rights – Employment Tribunals prepared by a committee of Employment Tribunal chairmen (as they then were) and the Government Actuary (“the Guidance”). The Guidance recommends that tribunals use one of two alternative approaches to the calculation of pension loss – “the substantial loss approach” and “the simplified approach”: I will explain the differences between the two approaches later. The Claimant contended for the use of the former and put in a report from an actuary, Mr Lockett, on the basis of which Mr Pullen submitted that the loss should be quantified at at £97,527. The Trust contended for use of the latter. 17. The Tribunal elected for the simplified approach. I need not set out its reasons at this stage. It calculated the Claimant’s loss at £32,827.69. The details of that calculation are not in issue, but I should note that at para. 5.5.3 it said: “The Tribunal, taking into account changes to the pension legislation in the coming year, has concluded that it would be just and equitable to make an award based on the claimant not having a final salary or defined benefit pension for four years … .” Although it is not quite expressed that way, I think this must be intended as a finding that the Claimant would become a member of a final salary scheme after four years. 18. 19. On the Supperstone appeal the Claimant argued that the Tribunal should have adopted the substantial loss approach. The EAT rejected that submission, but it held that the question of which approach was appropriate would need to be revisited by the Employment Tribunal in the light of its decision on the question of future loss of earnings. In its further remedy decision the Employment Tribunal duly reconsidered the issue of pension loss as directed by the Supperstone tribunal, but it decided to stick with the simplified approach, and the award in this regard was accordingly unchanged: again, I will give its reasons later. THE ISSUES 20. On the McMullen appeal the Claimant challenged the decisions of the Employment Tribunal (a) to limit her future loss of earnings to a period of twelve years and (b) to use the simplified approach to the assessment of pension loss. Those remain the areas in issue before us. I need not set out the reasoning of the EAT in dismissing her challenge: what we are ultimately concerned with is the reasoning of the Employment Tribunal. I will consider the issues under those two heads. Grounds 1-6 relate to future loss and grounds 7 and 8 to pension loss. There are also two more general grounds – that the Tribunal “failed to provide an effective remedy for a European tort” (ground 9) and that its decision on remedy was perverse (ground 10). But Mr Donovan in his oral submissions sensibly conceded that the former point could not succeed if his other submissions did not; and although he made no formal concession on the perversity ground, both in the skeleton argument and in his oral submissions he was at pains to emphasise that his case depended on the other specific grounds. I need accordingly say nothing more about grounds 9 and 10. (A) THE PERIOD OF LOSS 21. (1) 22. 23. The grounds under this head are of two kinds. First, the Claimant challenges the Tribunal’s self-direction that it should make its decision on the remitted issue as at the date of the first hearing and thus should exclude any evidence of subsequent developments (see para. 12 above): this is the subject of ground 1. Secondly, she contends that even on that basis it was not open to the Tribunal to find that the Claimant would regain her previous level of earnings after twelve years: this is the subject of grounds 2-6. THE DATE OF ASSESSMENT/FRESH EVIDENCE By the date of the remitted hearing in May 2012 several months had passed since the date which the Tribunal had taken as the point at which the Claimant would or should have obtained employment, but she had not done so. In mid-2011 she had accepted advice from her Job Centre adviser to try to obtain an NVQ in Business Administration. For that purpose it was necessary that she have a paid or voluntary work placement; and in default of paid work she applied for and obtained a role as a volunteer administration assistant in the Plymouth office of Shelter, beginning in August 2011. It was anticipated that it would take a year to obtain the NVQ. The Claimant applied in advance of the remitted hearing to put in evidence three documents – one from the prospective NVQ provider and two from Shelter – which established those facts: I will refer to these as “the Shelter documents”. (She had in fact made a similar application4 to the EAT in advance of the Supperstone hearing; but the Trust had objected and the application had been refused on the papers.) At the start of the remitted hearing the Tribunal decided that the Shelter documents should not be admitted, and they were formally removed from the bundle. The Reasons do not as such refer to the application or, therefore, give reasons for the Tribunal’s decision; but Ms Cunningham’s submission as recorded in para. 3.1 of the The application was not quite identical because she had applied to put in not only the Shelter documents but also a short witness statement explaining the context: this was not referred to when the application was made to the Employment Tribunal, but it says nothing that could not be inferred from the documents themselves. 24. Reasons (see para. 12 above) was directed to the issue of the admissibility of the Shelter documents, and the Tribunal’s acceptance of that submission was decisive of the application. It is the Claimant’s case that the Tribunal was wrong to limit itself to the evidence that had been adduced at the first hearing and that it should have admitted her evidence, embodied in the Shelter documents, that she had not obtained paid employment within the one-year period which it had adopted first time round and would in practice be unable to do so until she had obtained her NVQ. That meant at the very least that her period of total loss of earnings should be substantially extended, but it might also have affected the Tribunal’s assessment of whether (and if so when) she would ever recover her pre-dismissal level of earnings. Mr Donovan relied on the well-known dictum of Harman LJ in Curwen v James [1963] 1 WLR 748 that “the court should never speculate when it knows” (see at p. 753). He also referred us to the decision of the EAT, Langstaff J presiding, in NCP Services Ltd. v Topliss UKEAT/0147/09.5 In that case an Employment Tribunal had made a flawed assessment of the future loss of earnings in the case of an employee who had been unfairly dismissed, and the case was remitted (to the same Tribunal) to perform the task afresh. Langstaff J made it clear that on the remitted hearing the Employment Tribunal should make its fresh assessment as at the date of the remitted hearing. He observed, at para. 41: “The passage of time will inevitably have shone light as a matter of fact upon what has in truth been the real loss after 11 December 2008 [being the date of the first decision]. It may be, for instance, that the position is that the employee will have obtained a better paid job within a few months. It may be that he has obtained a less well-paid job in which he is still employed. It may be that he is still unemployed.” The EAT held that it would be artificial and unreal to ignore such evidence. 25. Ms Cunningham submitted that, while the Topliss approach might be correct in a case where the assessment as a whole had been remitted, the present case was different because the Supperstone Tribunal had remitted only a specific question: that question did not involve any reconsideration of the Employment Tribunal’s finding as to the date at which the Claimant would or should obtain paid employment, but only of the level of earnings that she could expect as from that date. In those circumstances there was no justification for departing from the established approach in English law that a claimant’s compensation is assessed once for all at the date of assessment. 26. 27. In order to choose between those submissions it is necessary to identify with some care the applicable principles. In my view it is necessary to distinguish between two different situations. The first situation is where the court or tribunal at first instance is conducting its primary assessment of compensation. It is well established that in such a case it must In fact the point had not been argued, and Langstaff J made his observations, for the guidance of the tribunal on remittal, ex improviso and without reference to authority: indeed he gave counsel the opportunity following his judgment to draw his attention to any contrary authority. As will appear, however, what he said was in my view entirely consistent with principle and authority. have regard to all the facts known to it at the date of assessment. The best-known statement of this principle is in the speech of Lord Macnaghten in Bwlfa and Merthyr Dare Steam Collieries (1891) v Pontypridd Waterworks Co [1903] AC 426. That was a case where part of a coalmine had been closed by a statutory undertaker under powers which entitled the owner to compensation for lost production. The price of coal rose in the period following the closure and the issue was whether the arbitrator should fix compensation by reference to the price as predicted, or reasonably predictable, as at the closure date or by reference to the price as it was subsequently known to have been. Lord Macnaghten said, at p. 431: “... the arbitrator's duty is to determine the amount of compensation payable. In order to enable him to come to a just and true conclusion it is his duty, I think, to avail himself of all information at hand at the time of making his award which may be laid before him. Why should he listen to conjecture on a matter which has become an accomplished fact? Why should he guess when he can calculate? With the light before him, why should he shut his eyes and grope in the dark?” In Curwen v James Harman LJ relied on that passage in support of the statement on which Mr Donovan relies (see para. 24 above)6. In my view the “Bwlfa principle” must apply just as much to a case where the court or tribunal is re-considering the issue of damages as a result of a remittal by an appellate tribunal as where it is considering it first time round: as a result of the order for remittal the primary assessment of compensation has not been concluded. I accordingly agree entirely with the approach enunciated by Langstaff J in NCP v Topliss. The second situation is where an appellate court is asked to admit evidence of events occurring subsequent to the primary assessment of compensation. The essential difference from the first situation is that a valid final award has been made. As Ms Cunningham submitted, the general approach in such a case is that, since damages have to be assessed once and for all, it would be contrary to the important interest in the finality of litigation that either party should be entitled to re-open a final award in respect of (what was at the date of assessment) future loss simply because things had turned out differently from expected. However, the authorities show that that approach is not applied with absolute rigour. The cases are fully reviewed in McGregor on Damages, 19th ed., at paras. 38-045 to 38-054, and I need not analyse them in detail here. Their effect is that the court will in some, but only some, cases be prepared to allow an appeal based on subsequent events which undermine the basis on which the original award was made.7 Curwen v James (above) was a case of this kind. As to how cases where evidence of such events will be admitted are to be distinguished from those where it will not, the best guidance appears in the speech of Lord Wilberforce in Mitchell v Mulholland [1971] AC 666, at pp. 679-680. After As it happens, Curwen v James seems to me to have involved the rather different situation which I discuss in the next paragraph; but nothing turns on this. I would add, since the point was raised in argument before us, that although the reported cases are concerned with appeals, it seems to me that the same principles – if they can be called that – should apply in the case of an application to the Employment Tribunal to exercise its power to reconsider an award under rules 70-73 of the 2013 Rules of Procedure (the successor to the power of “review” under the 2004 Rules). 28. 7 referring to the reasoning in Curwen v James, Jenkins v Richard Thomas & Baldwins Ltd [1966] 1 WLR 476 and Murphy v Stone-Wallwork (Charlton) Ltd [1969] 1 WLR 1023, he says: “These cases are useful as instances, but they cannot be generalised into a formula. I do not think that, in the end, much more can usefully be said than, in the words of my noble and learned friend, Lord Pearson, that the matter is one of discretion and degree (Murphy [1969] 1 W. L. R. 1023, 1036). Negatively, fresh evidence ought not to be admitted when it bears upon matters falling within the field or area of uncertainty, in which the trial judge's estimate has previously been made. Positively, it may be admitted if some basic assumptions, common to both sides, have clearly been falsified by subsequent events, particularly if this has happened by the act of the defendant. Positively, too, it may be expected that courts will allow fresh evidence when to refuse it would affront common sense, or a sense of justice. All these are only non-exhaustive indications; the application of them, and their like, must be left to the Court of Appeal. The exceptional character of cases in which fresh evidence is allowed is fully recognised by that court.” (In most of these cases the appeal was brought, though generally on some other ground, within the time allowed by the Rules; but in Jenkins v Richard Thomas time for appealing was extended by almost a year to permit an appeal based on the introduction of evidence of subsequent events.) 29. At first sight the present case might appear to fall into the first category: this was a remitted hearing. But it is not so straightforward. As Ms Cunningham pointed out, the Supperstone Tribunal’s order remitted the case to the Employment Tribunal “to review [its] decision on … continuing loss of earnings”. It is clear from the judgment that the phrase “continuing loss of earnings” refers to the question of the rate of the Claimant’s remuneration in the employment which the tribunal had found that she was likely to obtain after a year: there had been no challenge to the Tribunal’s finding as to the date at which she would obtain that job. The Tribunal was not being directed to reconsider its decision in any other respect. 30. In those circumstances it seems to me that the position about the evidence of subsequent events which the Claimant sought to adduce is as follows: (1) (2) To the extent that that evidence was relevant to the task which the Employment Tribunal had been required by the EAT to perform – that is, to the assessment of the Claimant’s likely level of earnings in her new employment – it would have to be admitted, as of right, on the Bwlfa principle. If, however, the Tribunal was to be asked to admit the evidence for a different purpose, namely so as to revisit its decision as to the likely date on which the Claimant would obtain new employment, it would be in a position analogous to that of an appellate court, as considered at para. 28 above: although the assessment of compensation was properly before it, that was only as a result of a remittal limited to a different point. The decision whether to allow the 31. 32. Claimant to take advantage of the, adventitious, opportunity to adduce fresh evidence was a matter for its discretion, to be exercised in accordance with the guidance given by Lord Wilberforce in Mitchell v Mulholland. I should therefore consider whether the Shelter documents should have been admitted on either of those bases. As to the first basis, I cannot see that the Shelter documents, or what emerged from them, added anything of significance, so far as the remitted question was concerned, to the material which the Tribunal already had. All that they established was (a) that the Claimant had not yet obtained employment and (b) that she was gaining administrative experience at Shelter and hoped to obtain an NVQ in business administration. That added nothing specific about the level of job that she might in due course hope to obtain or her likely rate of pay. The Tribunal already had a wealth of information, placed before it at the first hearing, about the kind of work that she might be able to do; and it was on the basis of that material that it made the finding that her likely employment would be in an administrative, eventually leading to management, role. As to the second basis, it was never in fact the Claimant’s case before the Tribunal that the Shelter documents should be admitted in order to challenge the finding first time round as to her likely date of employment. Although the application itself did not identify the purpose for which she wanted to rely on them, her written submissions, which proceeded on the basis that the documents would be admitted, only clearly refer to them in support of a submission as to the rate of pay that she would receive in any future employment. She says, at para. 51: “… the Tribunal … may find [the Shelter documents] of some assistance. These show that the Claimant has been accepted as a volunteer at Shelter. The Claimant believes that there will be a further period of at least 18 months while she gains experience before she finds paid employment, and suggests that having previously applied for a post at Shelter at a salary of £17,000 pro rata that this would be her likely future salary.” The consistent focus of the submissions is on the loss that the Claimant would suffer “after she obtains suitable alternative employment”, as it is put in her introductory paragraph on this point: there is nowhere any submission that the date which the Tribunal had fixed for when that would occur should be reconsidered. That probably reflects her (correct) understanding of the scope of the remittal ordered by the Supperstone Tribunal; but in my view it is hard to criticise the Employment Tribunal for not undertaking an exercise which it was never asked to undertake – whether on the basis of the Shelter documents or indeed simply of the Claimant’s statement that she was still unemployed. 33. In any event, however, even if the Tribunal had been asked to exercise its discretion to revisit its finding as to the likely date at which the Claimant would find employment, I do not believe that it would have been right for it to accede to that application. Its initial finding that she would find paid work after a year seems to me a good example of what Lord Wilberforce in Mitchell v Mulholland called “matters falling within the field or area of uncertainty in which the trial judge’s estimate has previously been made”. The likely date of employment was indeed, as I have emphasised at para. 9 above, an estimate, and it was in practice bound to be shown in the event to be either too optimistic or too pessimistic, albeit hopefully not by too exorbitant an amount. This is an inevitable consequence of our once-for-all system of assessment. It would of course on this basis equally have been impossible for the Trust to apply to re-open the issue if the Claimant had unexpectedly obtained fresh employment at her original level of remuneration after six months.8 (b) THE TWELVE-YEAR PERIOD 34. To recap, what we are concerned with here is the Tribunal’s decision at the remitted hearing that the Claimant would over a period of twelve years gradually return to her previous level of earnings. That finding is challenged on five grounds (nos. 2-6), which I take in turn. Ground 2 35. This ground of appeal focuses on paras. 3.7-3.8 of the Reasons. These follow on directly from the passage in which the Tribunal finds that the Claimant will be able to achieve her previous level of earnings within twelve years. They read as follows: “3.7. During the course of further discussion as to the approach to be taken, the claimant highlighted the fact that a medical report before the Tribunal at the first remedy hearing indicated that the claimant would not be able to take on a management position because of the stress involved. The only report the Tribunal had, so far as it can recall, is the report of Dr Reckert of 5 October 2010. On the last page, the second paragraph states: “Stress, poor social support and psychological distress are associated with mental and physical health of SLE patients. It is very difficult to measure these phases of disease activity but I do think in this case we do have objective factors such as the loss of weight, the incapacity to walk, unsupported, findings including an elevation in her LFT’s and PV in August 2010, loss of voice during the time of the phase 3 hearing and the neuro-psychiatric symptoms”. 3.8 Furthermore, the Tribunal has noted from paragraph 51 of the claimant’s submission that she has been accepted as a volunteer at Shelter. That organisation helps people in the most desperate situations and it is, therefore, likely that it is a stressful occupation. This suggests that the claimant is able to deal with stress so that there is no reason why she should not be able to achieve a level in another occupation commensurate with the position she occupied with the respondent.” I should add, as a matter of caution in case the question comes up in a different case, that I can conceive of a situation in which evidence of subsequent events had to be admitted for purpose A and it was then, at least arguably, impracticable not to admit it also for purpose B, even if it would not have been admissible if that had been the sole purpose. But that is not the case here. 36. 37. 38. It is clear that these two paragraphs represent the Tribunal’s response to the Claimant’s intervention during the oral reasons referred to at para. 13 above: that led to the “further discussion” referred to at the beginning of para. 3.7. The paragraphs under section 3 down to para. 3.6 represent in substance what the Tribunal had said orally up to the moment that the Claimant interrupted. The gist of the Claimant’s intervention was, as the Tribunal says, that its finding that she would in due course be able to take on a management role disregarded the medical evidence about the effect of stress on her condition. What Mr Donovan says on this appeal is that in the two paragraphs which I have quoted the Tribunal did not deal properly with that point. He has essentially two criticisms which I consider in turn. The first criticism is that the Tribunal failed to have regard to the full medical evidence before it and wrongly referred only to the report of Dr Reckert dated 5 October 2010. I should accordingly start by identifying what medical evidence was in fact available. For the purpose of the liability hearing the Claimant put in evidence a number of letters from the rheumatologists treating her for her SLE. The first letters are to her GP and are concerned with the diagnosis and management of her condition. There are then two to the Trust about its impact on her employment. The most recent, from her consultant Dr. Reckert, was dated 12 May 2010, and was sent to the Claimant’s solicitors in response to a request for “a medical report explaining the nature and consequences of [her] condition”. For the purpose of the first remedy hearing the Claimant put in a further letter from Dr. Reckert dated 5 October 2010 intended to address (as she recites at the beginning of the letter) “the effect that the Trust’s conduct as well as the Tribunal procedures have had on [the Claimant’s] health”: this is the report to which the Tribunal refers. In it Dr. Reckert expressed the opinion that the Claimant’s condition had been severely exacerbated by the stresses engendered by the Trust’s conduct and by the litigation process including the passage quoted by the Tribunal at para. 3.7 of its Reasons. The report concludes: “Due to the ongoing tribunal process and its impact on Sharan’s disease activity, it is not possible for us to make a judgement on how effective the current treatment is for her. I am quite sure that if she is no longer experiencing the pressure of Tribunal process we will find a lower activity of the disease and we will hopefully then be able to see if this combination therapy will lead to a remission of her disease.” Dr. Reckert also provided a short further letter dated 3 December 2010 identifying difficulties about her returning to work with the Trust. 39. At the remitted remedies hearing all the material referred to above was before the Tribunal, and its effect was summarised, by reference to several of the letters, in the introductory sections of the Claimant’s written submissions. There was no relevant evidence from the Trust. No witnesses were called. 40. Mr Donovan submits that against that background it was plainly wrong of the Tribunal to say that the only evidence addressing the question of the effects of stress 41. 42. on the Claimant’s condition was Dr. Reckert’s report of 5 October 2010: the earlier letters were equally relevant. Mr Donovan did not argue – as the Claimant had apparently claimed in her intervention – that any of the letters or reports said in terms that the stress involved would make it impossible for her to take a management role; but he did submit that the earlier evidence raised a serious question whether that was so and that the Tribunal wholly failed to consider that evidence. The Tribunal’s language in para. 3.7 of the Reasons is unfortunate. There was no question of it having to “recall” what evidence was before it. As noted above, the Claimant’s written submissions identified the medical evidence on which she relied, all of which was in the bundles for the liability or remedy hearings; and that evidence was not limited to Dr. Reckert’s report of 5 October 2010. I suspect that what the Tribunal meant was that that report was the only report specifically produced for the purpose of the remedy hearing; but that did not necessarily mean that the earlier evidence could be disregarded. However, I am not persuaded that the way the Tribunal expressed itself reflects any error of substance. The broad point that the Claimant intervened to make was that her illness would preclude her taking any managerial role. That point was not even potentially relevant to the liability issue, and none of the letters put in evidence at the liability hearing is directed to it. There are, particularly in Dr. Reckert’s report of 12 May 2010, some general observations to the effect that stress can lead to a flare-up in the symptoms of SLE, and it is these which are quoted in the Claimant’s written submissions for the hearing; but they are not related to any consideration of the kind of work that she might be able to do in future. If the Tribunal had made an express reference to that report, or any of the earlier letters, it would not have added in substance to what Dr Reckert said in her later report, which contained very similar observations about the effect of stress. The fact is that there is nothing in any of the medical evidence that would justify the conclusion that the stress that might be associated with a managerial role meant that the Claimant would be incapable of fulfilling such a role. None of the evidence goes further than to establish that stress can produce a flare-up of the symptoms of SLE; and that the stresses associated with the litigation appeared to be having that effect. 43. Mr Donovan’s second point relates to para. 3.8 of the Reasons. He submitted that the Tribunal was quite wrong to make any point based on the Claimant’s work at Shelter, about which she had herself been refused permission to adduce evidence. This was not only unfair in principle but had led to a serious mistake because in fact the Claimant’s work as a volunteer was wholly administrative and she had no involvement with “people in the most desperate situations”: accordingly no conclusions could be drawn about her ability to cope with stress. 44. 45. This criticism of the Tribunal is in my view justified. If it was going to rely on the Claimant’s work at Shelter as an indication of her ability to withstand stress it ought to have put the point to her and given her the chance to comment: if it had done so, the probability is that it would have been recognised to be a non-point. Again, however, I do not believe that this error vitiates the Tribunal’s reasoning as a whole. The essence of that reasoning is that its judgment of the Claimant’s attitude and abilities justified the conclusion that she would eventually regain her previous level of earnings: see paras. 3.2-3.6. That was not based in any way on her work for Shelter, which was (potentially) relevant only in the context of her intervention based on the medical evidence. But even in that context it was not central to the Tribunal’s reasoning. The essential point is simply that there was nothing in the medical evidence to justify the conclusion that the Claimant would be unable to cope with a managerial role: see above. The observations about her work for Shelter – preceded by a “furthermore” – were in truth no more than an ill-judged makeweight. Grounds 3-5 46. 47. 48. Under the heading of these three grounds the Claimant makes a number of particular criticisms of the Tribunal’s adoption of a twelve-year period of loss. On analysis, and as developed by Mr Donovan orally, they boil down to three points, which do not precisely correspond to the pleaded grounds. I take them in turn. First, and most generally, Mr Donovan emphasised the difficulties of the Claimant’s predicament as established by the Tribunal’s core findings summarised at para. 7 above. She had had a secure and well-paid specialist job in which she could, but for the Trust’s conduct, have expected to continue until retirement. It was, he submitted, simply unrealistic to believe that she could within ten or twelve years return to the same level of earnings and job security – unable as she was to find other work in her specialist field or to leave the Plymouth area. She was also, by reason of her condition, only capable of part-time work. The only realistic finding was that she would suffer a career-long loss of earnings. In finding otherwise the Tribunal was simply building castles in the air, on the basis of no evidence save its own assessment that she would find an administrative job which would in due course lead to a management role. That is in substance a submission that the Tribunal’s finding was perverse and/or inadequately reasoned. I cannot accept this. The kind of exercise on which the Tribunal was, necessarily, engaged was bound to be to a considerable extent an exercise in speculation, based on its assessment of the Claimant herself – that is, her attitude and abilities – and of the local job market. That is an exercise which an employment tribunal – and, particularly, a tribunal including lay members – is much better qualified to perform than this Court, which has never seen the Claimant and has no experience of local conditions. In the present case the Tribunal had at the first remedy hearing had before it very extensive evidence about the kinds of work that might be available to the Claimant. I see no basis on which we could hold that it was not in a position to make the finding which it did. As for the adequacy of its reasoning, the judgments involved are necessarily broad-brush and do not admit of much detailed justification. The Tribunal made it clear that its expectation that the Claimant would be able in due course to proceed to a managerial role was based on its assessment of her as “intelligent … capable … and determined” (see Reasons, para. 3.3). I do not think it was necessary for it to say more. 49. That, however, brings me to Mr Donovan’s second submission, which is that it was wrong for the Tribunal to put such weight on its assessment of the Claimant’s abilities and attitude, which was inevitably largely based on her performance as a witness and (latterly) advocate. He accepted that it was entitled to give that factor some weight, but not as much as he says that it evidently did. This submission is, with all respect, hopeless. If the challenge is simply to the weight given to an admittedly relevant factor, no question of law arises – at least unless the Tribunal’s conclusion about the Claimant’s abilities as an employee can be characterised as perverse, which Mr Donovan rightly did not argue. 50. Mr Donovan’s third point was that the Tribunal ignored the possibility of any “upside” loss – that is, that over the period in question the Claimant might have progressed within the Trust to some more senior and better-remunerated role. It is right that that is a possibility which must always be borne in mind; but it must be based on evidence. Our attention was not drawn to any evidence that the Claimant, had she remained at the Trust, might have moved to a more senior role; and no submission to this effect was made by her or on her behalf. That is not, perhaps, very surprising given the nature of her role as a specialist technician. In those circumstances the Tribunal cannot be criticised for not taking this possibility specifically into account. Ground 6 51. The ground here is that the Tribunal erred in law in not making use of the Ogden tables in calculating the Claimant’s future loss. But Mr Donovan accepted that that was dependent on acceptance of his submission that the Claimant had suffered a career-long loss. Since, for the reasons given, I believe that the Tribunal was entitled to find that she had not (subject only to the question of pension loss, which I consider below), the point does not arise. OVERVIEW ON PERIOD OF LOSS 52. 53. The Employment Tribunal had to make a judgment about whether the Claimant would eventually regain her previous level of earnings (on the basis of a 25-hour week) and, if so, about what period or periods of loss best represented the mid-point of the probabilities as to when that would occur. I see nothing outlandish in its conclusion that she would eventually achieve a managerial job which would allow her to earn at that level; nor can I see anything wrong in its estimate that this would occur gradually over a twelve-year period. At the risk of repeating myself, this can only ever be a very broad exercise; but the Tribunal seems to me to have approached it sensibly. I have every sympathy with the Claimant in having the misfortune to suffer from the debilitating disease that she does. But the Tribunal was not obliged to take the pessimistic, if not indeed rather demeaning, view that that disease prevented her, as a woman of ability and determination, ever again undertaking a job involving some degree of responsibility. There was nothing in the evidence to prevent it concluding that she would indeed in due course be able to cope with such a role, particularly once she was freed from the toils of litigation. It may, understandably, not seem that way to the Claimant at present, but I hope and expect that with time her outlook will change. (B) PENSION LOSS THE TWO APPROACHES 54. The only issue before us is whether the Tribunal erred in law by preferring the simplified approach to the substantial loss approach. The two approaches are introduced and fully described in chapters 4-8 of the Guidance, and they can only be 55. 56. properly understood by a careful reading of those chapters, together with the memorandum of the Government Actuary at Appendix 2. For the purpose of this appeal I need only to draw attention to some features of the two approaches; and what follows should not be treated as an attempt at a comprehensive summary or used as a short-cut in cases where different issues arise. I should, by way of laying the ground, make two points, which are obvious but of fundamental importance, about the pension loss claim in this case. First, the pension rights which the Claimant enjoyed as an employee of the Trust were under a final salary scheme. The essential feature of such a scheme is that a member becomes entitled on reaching retirement age to a defined level of benefits calculated by reference to length of service and final salary (i.e. the level of salary being earned either on retirement or at some earlier termination). This is in contrast to the position under a money purchase scheme, where the employee’s entitlement is to have the employer make contributions, at a prescribed rate (typically a percentage of salary), into a scheme. In the case of a final salary scheme the employee has no entitlement to the contributions as such: the entitlement is to the benefits, and how the scheme is funded is a matter for the employer. 57. Secondly, in the case of a final salary scheme the loss suffered by a dismissed employee is of two kinds: (1) Accrued rights. The years of service which the employee has already accrued at the date of dismissal will earn him or her a deferred pension on reaching retirement age, but they will almost certainly do so to a lesser extent than if the employment had continued up to that point (or indeed to some intermediate date) because the final salary which forms the multiplicand will almost certainly be less. The Guidance refers to this as “loss of enhancement”. (2) Future rights. If the employment had continued the employee would of course have accrued further years of service and the multiplier would thus have increased. To anticipate, it is only this element in the loss which is affected by the choice between the two approaches; but, as will appear, how the Tribunal treated the claim for loss of enhancement of accrued rights is relevant to the argument before us. The Simplified Approach 58. Para. 4.11 of the Guidance summarises the simplified approach as follows: “The simplified approach is set out in Chapters 5, 6 and 7. It involves three stages – (a) in the case of a final salary scheme, the loss of the enhancement to the pension already accrued because of the increase of salary which would have occurred had the applicant not been dismissed, (b) in all cases, the loss of rights accruing up to the hearing and (c) the loss of future pension rights. These last two elements are calculated on the assumption that the contribution made by the employer to the fund during the period will equate to the value of the pension (attributable to the employer) that would have accrued. ....” 59. The important point to note is that the loss in respect of the rights that would have accrued but for the dismissal – stages (b) and (c) – is measured simply by reference to the loss of the employer’s contributions; and it is that which is the essential difference between this approach and the substantial loss approach. As the Guidance acknowledges (see para. 6.3), in the case of a final salary scheme measuring loss of future rights by reference to loss of contributions is “not technically correct”: that is because, as noted above, there is no direct correlation between contributions and benefits. But it is adopted in the interests of simplicity – whence the label. The loss of enhancement claim with which stage (a) is concerned is not amenable to rough- and-ready treatment of this kind and has to be calculated by reference to actuarial tables (see Appendix 4 to the Guidance): these are of the same character as those used in the substantial loss approach as described below. The Substantial Loss Approach 60. Para. 4.12 summarises the substantial loss approach as follows: “The substantial loss approach, by contrast, uses actuarial tables comparable to the Ogden Tables to assess the current capitalised value of the pension rights which would have accrued up to retirement. There may be cases where the tribunal decides that a person will return to a job at a comparable salary, but will never get a comparable pension see Bentwood Bros (Manchester) Ltd v Shepherd [2003] IRLR 364. In such cases the substantial loss approach may be needed even where the future loss of earnings is for a short period. But it must be remembered that loss of pension rights is the loss of a fringe benefit and may be compensated by an increase in salary in new employment.” The approach is described in detail in chapter 8. The steps required are as follows. 61. The first step is to perform the core calculation described in para. 8.3, which reads: “The calculation required is: Loss of future pension rights = A minus B minus C where: A = value of prospective final salary pension rights up to normal retirement age in former employment (if he or she had not been dismissed) B = value of accrued final salary pension rights to date of dismissal from former employment [i.e. the “deferred pension”] C = value of prospective final salary pension rights to normal retirement age in new employment. C will of course be zero if it is found that the applicant will probably not obtain further pensionable employment or if he or she has joined a money purchase scheme in the new employment. In that case, see 8.11 below.” Each of elements A-C is calculated by reference to actuarial tables provided at Appendices 5 and 6 to the Guidance: I need not go into the details, save to emphasise that (necessarily) the tables value the rights in question as at normal retirement age. 62. The next step derives from the fact that the calculation of element A assumes, in accordance with the tables used, that the claimant would have remained in the same employment, and thus an active member of the scheme, up to retirement age. But it would not normally be right to treat that assumption as a certainty. Accordingly, para. 8.4 of the Guidance recommends that some discount be made. It reads: “Once these figures have been calculated, the tribunal has a further decision to make as to the amounts of any withdrawal factors. The Tables work on the basis that the applicant would have remained in his previous employment until retirement, subject to the usual risks of mortality and disability. However, it is recognised that people leave even the most stable employment for a variety of reasons. As with the Ogden Tables, no Tables are available to assist the tribunal in making this deduction. It will vary with the age, status, work record and health of the applicant and with the perceived future viability of the respondent’s business.” Thus, while the substantial loss approach is sometimes described as assuming a “whole-career loss”, that is an over-simplification: the methodological starting-point is indeed to assume such a loss, but the provision for a discount reflects a recognition that unless qualified to some extent that would be likely to lead to over-compensation. 63. The third step only arises in some cases and is not, strictly speaking, a step in the calculation of the pension loss itself. It derives from para. 8.11 of the Guidance (which is referred to at the end of para. 8.3 which I have set out above). That reads as follows: “If the applicant loses a job with a final salary pension scheme and obtains one with a money purchase scheme or signs up to a stakeholder pension, the loss is calculated as in 8.3 but only A minus B. There is no need to worry about any loss of employer pension contributions in the new job because those contributions have already been factored into the A minus B calculation. When assessing loss of earnings, however, it will be appropriate to take account of any employer contributions in the new job in order to ascertain whether there is a continuing loss of earnings or not. Thus the comparison will be the difference between net earnings in the old job (ignoring employer payroll contributions) and net earnings plus any employer pension contributions in the new.” The relevant part is in the second half of the paragraph. What it means is that, where a claimant has lost a job with a final salary scheme but has obtained a new job where the employer makes contributions to a money purchase scheme, those contributions are not brought in so as to reduce the compensation for loss of pension rights; but they must be taken into account in assessing her loss of earnings. It is only a further “step” in the sense that the tribunal must check that that has been done. 64. I should note that the approach recommended in para. 8.11 was questioned by the EAT in Network Rail Infrastructure Ltd v Booth (unreported, 22.6.06) UKEAT/0071/06. At para. 15 Elias P makes the point that it would be more natural to deduct the contributions in question from the pension loss than take them into account in computing the loss of earnings – though he also observes that the difference should only affect the final award in cases where there is no continuing loss of earnings. I respectfully agree, and indeed I note that Mr Lockett followed that approach in his report; but at this stage I am simply setting out the steps required by the Guidance. The Guidance as to which Approach to Use 65. The Guidance acknowledges that the choice between the two approaches may make a substantial difference to the compensation recovered for loss of pension rights (see para. 4.10). How to decide which is the more appropriate in a particular case is addressed at paras. 4.13-4.14, which read as follows: “4.13 Experience suggests that the simplified approach will be appropriate in most cases. Tribunals have been reluctant to embark on assessment of whole career loss because of the uncertainties of employment in modern economic conditions. In general terms the substantial loss approach may be chosen in cases where the person dismissed has been in the respondent’s employment for a considerable time, where the employment was of a stable nature and unlikely to be affected by the economic cycle and where the person dismissed has reached an age where he is less likely to be looking for new pastures. The decision will, however, always depend on the particular facts of the case. 4.14 More particularly, we suggest that the substantial loss approach is appropriate in the following circumstances: (a) when the applicant has found permanent new employment by the time of the hearing and assuming no specific uncertainties about the continuation of the lost job such as a supervening redundancy a few months after dismissal; further, the tribunal has found that the applicant is not likely to move on to better paid employment in due course; (b) when the applicant has not found permanent new employment and the tribunal is satisfied on the balance of probabilities that he or she will not find new employment before State Pension age (usually confined to cases of significant disability where the applicant will find considerable difficulty in the job market); (c) when the applicant has not found new employment but the tribunal is satisfied that the applicant will find alternative employment (which it values, for example, with the help of employment consultants) and is required then to value all losses to retirement and beyond before reducing the total loss by the percentage chance that the applicant would not have continued to retirement in the lost career. See Ministry of Defence v Cannock and Others [1994] ICR 918 et al. subject to our comment below. The simplified approach becomes inappropriate in these cases because there is a quantifiable continuing loss which can be assessed using the pensions data and Tables 1 to 4 of Appendices 5 and 6. These tables use factors similar to those in the Ogden Tables for personal injury and fatal accident cases. Although tables for pension loss are included in those tables, the tables in this booklet use some different assumptions to those underlying the Ogden Tables (see Appendix 2).” 66. 67. I think I should spell out the thinking behind paras. 4.13 and 4.14 of the Guidance, which is not perhaps as explicitly articulated as it might be. Para. 4.13 is intended to identify the general considerations affecting the choice of approach. It identifies three factors which point in favour of the use of the substantial loss approach – the length of time that the claimant has been employed; the “stability” of the employment; and whether the employee “has reached an age where he is less likely to be looking for new pastures”. What those three factors have in common is that they all increase the likelihood that the employee would, but for the dismissal, still have been an active member of the scheme at retirement. The importance of that is that it tends to justify the use of a method that starts – as element A in the equation – with an assumption of “whole-career loss” (albeit modified by the use of a discount as described at para. 62 above). The contrast is with the case where the employee would probably have changed jobs, and thus have left the scheme, before retirement age anyway, as a result of “the uncertainties of employment in modern economic conditions”. In such a case he or she would have suffered, perhaps only a year or two later, precisely the same kind of loss as is being claimed for in the proceedings; and it is more appropriate simply to award lost contributions up to that date, as per the simplified approach, rather than embarking on the exercise of valuing rights on retirement which would almost certainly never have accrued and then applying a massive “finger-in-the air” discount. The question is whether the uncertainties that would have to be reflected in such a discount are so great that they undermine the point of assessing the hypothetical whole-career loss in the first place. Whether that is so in any particular case is a matter for the judgment of the tribunal. The observation at the beginning of para. 4.13 that the simplified approach “will be appropriate in most cases” is because, so the authors believe, experience shows that in most cases the relevant uncertainties are indeed too great. 68. Para. 4.14 is introduced by the phrase “more particularly”; but in fact the three particular cases which it proceeds to identify are not primarily concerned with the question discussed in para. 4.13 – that is, the uncertainties affecting what would have happened if the claimant had not been dismissed – but with a different kind of uncertainty, namely that affecting the claimant’s actual future in terms of fresh employment. Thus the common feature of examples (a) and (b) is that that future is reasonably fixed – in the first case because the claimant has obtained stable new employment from which he or she is unlikely to move on, and in the second because he or she will not obtain other employment. Example (c) is a little less clear-cut; but the essential point appears to be that the Tribunal has made a finding as to the date and terms of hypothetical fresh employment. In all three cases there is said to be a “quantifiable continuing loss which can be assessed using … [the tables in] … Appendices 5 and 6”. These uncertainties are potentially relevant to “element C”, but I confess to finding the relationship between paras. 4.13 and 4.14 a little opaque. However, the broad theme of both is unquestionably the need to have a sufficiently firm basis to make the assumptions required by the substantial loss approach. 69. That analysis is consistent with an observation made in the earlier part of chapter 4 of the Guidance. Para. 4.7 begins as follows: “The key choice to be made by the tribunal is whether to look at the whole career loss to retirement which can then be discounted to allow for the eventuality that the applicant would not have remained in the employment throughout, or to look only to the next few years and assume that by that time he will have obtained comparable employment either with a similar pension scheme or a higher salary to compensate. Tribunals have tended to find in many cases that the applicant would obtain comparable employment within a fairly short period, ranging from 3 months to 2 years.” That is essentially the same point as underlies paras. 4.13 and 4.14. THE TRIBUNAL’S REASONS 70. In its original remedy decision the Employment Tribunal began by referring to paras. 4.13-4.14 of the Guidance. It said (at para. 5.5): “Having considered those paragraphs at some length, the Tribunal has noted that the substantial loss approach may be chosen in cases where the employee has been in the respondent’s employment “for a considerable time”. The claimant was employed for ten years by the respondent. However, the Tribunal has understood that the expression “considerable time” denotes a time which brings an employee much closer to retirement; so, for example, in the case of the claimant who first started working for the employer when she was 24 years of age, if her employment had been brought to an end when she was nearer to 50. On that basis, the Tribunal has concluded that the appropriate approach for loss of pension rights is the simplified approach.” 71. As noted above, on the Supperstone appeal the EAT was not prepared to say that that approach was necessarily wrong; but it said that the issue might have to be reconsidered in the light of what the Tribunal found about the Claimant’s continuing loss. As to that, in its second remedy decision the Tribunal said only this (at para. 3.9): “Because the Tribunal is not, therefore, considering ongoing loss until retirement, the Tribunal confirms its previous decision that the Tribunal should adopt the simplified approach for loss of pension rights for the reasons given at paragraph 5.5 of its original decision. It is, therefore, not appropriate to use the Ogden tables in calculating pension loss.” The “therefore” evidently refers to the conclusion in the previous paragraphs that the Claimant would regain her previous level of earnings after twelve years. (The reference to using “the Ogden tables” is not strictly correct, but the tables provided in the Guidance for use on the substantial loss approach are of the same character as the Ogden tables and that is obviously what the Tribunal meant.) DISCUSSION The Tribunal’s Original Reasons 72. 73. The Tribunal’s reasons for rejecting the substantial loss approach first time round appear from para. 5.5 of the Reasons to have turned entirely on its interpretation of the reference in para. 4.13 of the Guidance to whether the claimant has been in the respondent’s employment “for a considerable time”. It said that that factor did not apply where, as here, the claimant was still a long way from retirement. In my view that approach failed to engage with the more general question to which para. 4.13 is directed, as explained above, namely whether the Claimant was likely to stay the course up to retirement date. The Tribunal does not identify or address that question at all; and the question of the Claimant’s age could not be treated by itself as decisive of it. No doubt in some cases the fact that a claimant was aged only 34 when dismissed might be a very strong reason for holding that the chances of her remaining in the relevant scheme until retirement were simply too remote to justify the use of the substantial loss approach; but each case turns on its particular circumstances, and it is still necessary to consider the underlying question. The Claimant’s case, and her evidence, was that, as an employee with a specialist skill for which the principal (if not indeed the only) market was in the NHS, she was likely to remain in the NHS for the whole of her career, even though she was only 34; and that she would not, in the language of the Guidance, seek “new pastures”. That was all the more so because she suffered from a medical condition that made her cautious about embarking on major change. None of that is considered by the Tribunal at all. 74. In my view, therefore, even if one focuses exclusively on para. 5.5 of the Reasons the Tribunal misdirected itself. But that conclusion is reinforced by reference to what the Tribunal says elsewhere. As noted at para. 59 above, the simplified approach requires the use of actuarial tables in calculating the “loss of enhancement” element in the claim. Those tables necessarily proceed on the assumption that the claimant would have remained in the scheme until retirement; and the Guidance requires tribunals to apply a “withdrawal factor” to the resulting figure to reflect the uncertainties affecting that assumption. Para. 5.15 reads: “The figure obtained by applying the multiplier should be reduced if appropriate by a percentage representing the likelihood that the applicant would have lost his job before retirement for reasons other than unfair dismissal or discrimination, such as a fair dismissal, redundancy, leaving voluntarily etc. The 1980 paper set out a table of such deductions called the “withdrawal factor”, but, as stated in the previous guidelines, we remain of the view that any such figures are inappropriate and that it is best to leave this percentage to the discretion of the Tribunal.” The Tribunal, having elected for the simplified approach, had to perform that exercise. In that connection it said this, at para. 5.5.1: “The respondent submits that the claimant was likely to leave the respondent’s employment before retirement so this figure should be discounted by 50%. The claimant had said that she would not have left the respondent’s employment. The Tribunal accepts the claimant’s evidence of her intention in that her medical condition was worsening so that she was very limited geographically where she could work and was clearly limited in the work that she could do. Accordingly, the Tribunal accepts Mr Pullen’s suggestion that the discount should be 20%.” In my view that finding, albeit made for a different purpose, is inconsistent with a finding that the uncertainties about whether the Claimant would have stayed the course but for her dismissal are so great that the substantial loss approach was inappropriate. 75. I am supported in my conclusion on this part of the case by the decision of the EAT in the Network Rail case referred to above. The claimant in that case was a 36-year old engineer who had worked for the respondent for 14 years where she was a member of a final salary scheme. The Employment Tribunal held that she would recover her previous level of earnings after 2½ years but that in any future employment her only pension entitlement was likely to take the form of a money purchase scheme. It applied the substantial loss approach. The EAT upheld that decision, endorsing (at para. 23) counsel’s submission (at para. 22) that it was important that “… the Employee in this case was, as the Tribunal found, employed in a very specialist industry in which there was effectively only the one Employer, namely the Respondent. She had spent all her working life in that industry and there was every expectation that she would have remained there, but for the dismissal. … [She] falls fairly and squarely within paragraph 4.13.” The Tribunal’s Reasons on Remittal 76. When the Tribunal reconsidered the question second time round, following the Supperstone appeal, it said (see para. 70 above) that the simplified loss approach was appropriate because its finding that the Claimant would regain her pre-dismissal earnings after twelve years meant that there was “no ongoing loss until retirement”. Although it said that that confirmed its original reasoning, it seems to me that in fact the reason stated is different: it depends not on a conclusion about what would have happened but for the dismissal but on its findings about what would happen in the actual future. But that reasoning also seems to me to be flawed. The fact that the Claimant would eventually regain her pre-dismissal earnings is irrelevant: what matters is her pension rights. As pointed out at para. 4.12 of the Guidance (founding on the decision of this Court in Bentwood Bros (Manchester) Ltd v Shepherd [2003] IRLR 364), a claimant may return to a job at a comparable salary but never get a comparable pension. 77. In that connection I should refer to the Tribunal’s finding noted at para. 17 above that the Claimant would find employment with the benefit of a final salary scheme after four years. It does not appear to have relied on this finding as a reason for preferring the simplified approach; and nor did Ms Cunningham. They were right not to do so. The fact that a claimant may in due course obtain other employment under which she enjoys membership of a final salary scheme is not in any way inconsistent with use of the substantial loss approach: after all, element C in the calculation is specifically directed at such a case. However, Mr Donovan submitted that the finding was in any event unsustainable, and since the issue will be relevant at a later stage I think I should consider that submission. 78. Mr Donovan accepted that the Tribunal was entitled to find that the Claimant would within four years be in an employment under which her employer contributed to a pension scheme; but he submitted that it was not open to it on the evidence to find that that would be a final salary scheme. Mr Lockett had at para. 3.4 of his expert report (which had not been challenged) expressed the opinion that: “It is unlikely that any future employer would offer a final salary scheme as most of these are now closed to new entrants, so we have had to assume that any future scheme would be a money purchase scheme.” The Claimant’s witness statement said the same. Mr Pullen’s submissions had been made on that basis first time round; and the Claimant had in her written submissions on the second occasion expressly relied on Mr Lockett’s opinion. The Tribunal was wrong to make its original finding, and wrong not to reconsider it when it had the opportunity. 79. I can see no answer to that submission. The Tribunal gave no reasons for its original finding beyond saying that it had taken into account “changes in the pension legislation in the coming year”. That is apparently a reference to the provisions of the Pensions Act 2008, which require employers to arrange for the automatic enrolment of “jobholders” in a qualifying pension scheme; but a qualifying scheme will not necessarily be a final salary scheme. If there was some basis for rejecting Mr Lockett’s opinion it was incumbent on the Tribunal to say what it was, but it did not. CONCLUSION ON PENSION LOSS 80. Ms Cunningham submitted that the choice of which approach to adopt was a matter for the Tribunal’s discretion, with which this Court should not interfere. For the reasons given above, I cannot accept that submission. In my judgment not only did the Tribunal misdirect itself in the reasons which it gave for choosing the simplified approach but if it had directed itself correctly the only conclusion open to it, in the particular circumstances of the case and in the light of its other findings, was that it should follow the substantial loss approach. The case falls squarely within para. 4.13 of the Guidance for the reasons given at paras. 70-71 above. It also, to the extent that this adds anything, falls within the terms of example (c) in para. 4.14: the Tribunal has found that the Claimant will find alternative employment but there remain losses up to retirement, namely in respect of pension rights, which it is required to value. There is “a quantifiable continuing loss which can be assessed using [the tables in the Guidance]”. 81. I add one cautionary point. The Guidance was rightly described by Elias P in Network Rail as “extremely valuable”, and it and its predecessors have been used by industrial and employment tribunals on countless occasions and to good effect. In this appeal neither party has questioned its terms, and the issue has been presented purely on the basis of how they should be understood. But it should not be assumed that that will be the correct approach in every case. The Guidance has no statutory force and the recommendations in it are not gospel. Indeed in Network Rail itself the EAT questioned one aspect of the recommendations about the substantial loss approach: see para. 64 above. But there is a more particular reason for its application to be considered critically. There have been a number of important changes in pension law and practice since the current edition of the Guidance was published in 2003, and others are imminent: the extent to which its recommendations on particular points remain valid will increasingly need to be carefully considered. I would very much hope that HMCTS and/or the Judicial College may give priority to producing an updated version. DISPOSAL 82. 83. I would accordingly dismiss the appeal as regards the award for loss of earnings but allow it as regards the award for pension loss and remit the case to the Tribunal to assess that loss using the substantial loss approach. I have considered whether the remittal should be to a different Tribunal, but I do not believe that that is necessary. It would be difficult for a new Tribunal to have to undertake the exercise on the basis of findings made by the previous Tribunal, and the hearing would be likely to take longer and cost more. I see no reason to doubt that the Tribunal would conscientiously apply the substantial loss approach. Having said that, I would very much hope that the parties may be able to resolve the case without a further hearing. In that context I would make the following short observations, though they are not to be taken as binding because we were not directly addressed about how the substantial loss approach should be applied if it were to be taken. (1) As regards the first of the three steps identified at paras. 61-63 above – that is, the “A–B–C” calculation – Mr Lockett has quantified elements A and B in his two reports. No contrary evidence was lodged by the Trust, nor was Mr Lockett cross- examined (the Trust’s position being that the simplified approach should be adopted). In those circumstances it may be that his calculation of elements A and B cannot now be challenged. As regards element C, if, as I have held above, there was no basis for the Tribunal’s finding that the Claimant would in due course become a member of a final salary scheme, this has, according to para. 8.3 of the Guidance, a nil value (subject to what I say at (3) below). (2) As regards the second step, it would seem hard for either party to challenge the application for this purpose of the 20% discount applied by the Tribunal for the purpose of stage (a) in the simplified loss calculation, since the essential exercise is the same whichever approach is adopted. The 20% figure was proposed by Mr Pullen on behalf of the Claimant and was not the subject of any cross-appeal by the Trust. (3) As regards the third step, findings will be needed as to the pension contributions which the employer would be likely to make in the Claimant’s predicted new employment9; but I would hope that that could be the subject of agreement now that the Tribunal has made findings as to her future level of earnings. The contributions so found would then have to be taken into account as mitigating the Claimant’s loss. As discussed above, para. 8.11 of the Guidance says that that should be done as part of the loss of earnings calculation; but regard has to be had to the observations of the EAT in Network Rail referred to at para. 64 above. It must be in both parties’ interests to reach a compromise if at all possible rather than expend further costs on this long drawn out litigation, and I would strongly urge them to do so. Lord Justice Floyd: 84. I agree. Lady Justice Gloster: I also agree. 85. 86. Mr Lockett did in fact provide some figures, but they were based on the Claimant’s more pessimistic case as to her future earning capacity and require reconsideration on the basis of the Tribunal’s findings.