This appeal related only to the calculation of pension loss. The claimant had been a member of a pension scheme and the calculations were complicated due to the nature of the scheme. The ET adopted a bespoke approach, saying that neither the booklet 'Compensation for Loss of Pension Rights (2003)', nor the approaches contended for by the parties, appeared to address the loss over two periods. The ET declined to discount the pension loss figure to represent the chance the claimant would obtain work as a teacher and thereby gain access to a final salary pension scheme again. However, they discounted it for accelerated payment and for the vicissitudes of life and the risk that employment might have ended sooner. Finally they grossed up the figure to take account of the fact that the claimant would have to pay tax on the sum received. The claimant appealed against the discounts applied to the figure, and the respondent appealed against the decision not to discount to represent the chance of joining a new pension scheme and the grossing up calculation.

The EAT allowed the cross-appeal against discounting for accelerated payment - the multiplicands had already taken this into account. They dismissed the respondent's appeal against the chance the claimant would join an equivalent pension scheme in the future but allowed it in respect of the grossing up calculation because the pension loss had been calculated as a gross figure and a gross figure should not be grossed up. The EAT emphasised the two ways in which the calculation could be done: 1) calculate the gross figure (which would be rough and ready) and not gross up; or 2) calculate the net figure and gross up this figure to account for the tax payable by the claimant. The EAT preferred the second method which was the one more generally adopted.

____________

Appeal No. UKEAT/0027/15/BA

EMPLOYMENT APPEAL TRIBUNAL

FLEETBANK HOUSE, 2-6 SALISBURY SQUARE, LONDON EC4Y 8AE

At the Tribunal

On 1 June 2015

Judgment handed down on 9 September 2015

Before

HIS HONOUR JUDGE DAVID RICHARDSON

(SITTING ALONE)

CHIEF CONSTABLE OF NORTHUMBRIA POLICE (APPELLANT)

ERICHSEN (RESPONDENT)

Transcript of Proceedings

JUDGMENT

For the Appellant
MS HOLLY STOUT (of Counsel)
Instructed by:
Samuel Phillips Law Firm
Gibb Chambers
52 Westgate Road
Newcastle upon Tyne
NE1 5XU

For the Respondent
MR JACK FEENY (of Counsel)
Instructed by:
Pattinson & Brewer Solicitors
1 Bridge Street
York
YO1 6WD

DISABILITY DISCRIMINATION - Compensation

Assessment of pension loss

1. The Employment Tribunal did not err in law in declining to discount the award of pension loss to take into account the chance that the Claimant would obtain work as a teacher and thereby gain access to a final salary pension scheme again. It was entitled to find that there was no realistic chance that the Claimant would obtain such work. Abbey National plc v Chagger [2010] ICR 397, Thornett v Scope [2007] ICR 236 and [Eversheds Legal Services v De Belin ]()[2011] ICR 1137 considered.

1. The Employment Tribunal erred in law in the way in which it calculated the Claimant's pension loss from the age of 52 to 60. Chief Constable of the Metropolitan Police v Gardner UKEAT/0174/11 and [Griffin v Plymouth Hospital NHS Trust ]()[2014] EWCA Civ 1240 considered.

1. The Employment Tribunal erred in law by discounting the Claimant's pension loss from the age of 60 onwards for accelerated payment: it had already adopted tables for the multiplier which took accelerated payment into account.

1. The Employment Tribunal erred in law in "grossing up" the award of compensation for loss of pension when it had used gross figures for assessing that loss. However, it was correct to proceed on the basis that the award of compensation for loss of pension would be taxed under Part 6, Chapter 3 of the Income Tax (Earnings and Pensions) Act 2003: [Yorkshire Housing Limited v Cuerden ]()[2010] UKEAT/0397/09 followed.

1. The Respondent sought to "renew" at the Full Hearing a ground of appeal which had been rejected under Rule 3(7). It had made no application under Rule 3(10) and was long out of time for doing so. It was not entitled to renew a ground of appeal in this way.

1. By its Judgment dated 26 June 2014 the Employment Tribunal sitting in Newcastle (Employment Judge Wade presiding) awarded to Mr Erichsen ("the Claimant") compensation for disability discrimination in the sum of £258,551 to be paid by the Chief Constable of Northumbria Police ("the Respondent"). Of that award some £126,877 related to pension loss, and there was an additional sum to represent the tax which Claimant would have to pay in respect of the award of compensation for pension loss.

1. Two grounds of appeal and two grounds of cross-appeal were permitted to proceed to the Full Hearing of this appeal. They all concern aspects of pension loss: (1) whether the ET ought to have discounted its award for the chance that the Claimant might have made good his final salary pension loss from another source; (2) whether the ET erred in law in its assessment of pension loss for the period between age 52 and 60; (3) whether the effect of the award in respect of this period was to "double discount" for accelerated payment; (4) whether the award should have been "grossed up" for tax.

1. In addition to her two grounds of appeal Ms Holly Stout on behalf of the Respondent sought permission to amend to add another ground of appeal, also relating to the assessment of pension loss. She also sought what she describes as "permission to renew" a ground which was found by Lewis J to have no substance.

1. Section 119(2) of the Equality Act 2010, which applies where the civil courts have jurisdiction to adjudicate upon contraventions of the Act, provides that the County Court in England and Wales or the sheriff in Scotland has power to grant any remedy which could be granted by the High Court in proceedings in tort. Section 124(2)(b), which applies where Employment Tribunals have jurisdiction to adjudicate upon contraventions of the Act, provides that an Employment Tribunal may order the Respondent to pay compensation to the Claimant. Section 124(6) provides that the amount of compensation which may be awarded corresponds to the amount which could be awarded by the county court under section 119. The effect of these provisions is that Employment Tribunals are required to adopt the tortious measure of damages when awarding compensation for unlawful discrimination.

1. The Claimant was born on 10 August 1969. He entered police service on 1 December 1991 at the age of 22. On 31 December 2010 he became ill with stress and depression. He would have been fit to return to his work by February 2012 except for six acts of unlawful disability discrimination on the part of the Respondent. By reason of those acts of discrimination he resigned from police service on 1 February 2012 and was in very much worse health than he otherwise would have been for a period from 18 August 2011 until February 2013.

1. Following his resignation the Claimant brought proceedings for disability discrimination. By a Judgment dated 16 October 2013 the ET upheld most of his claims. The Remedy Hearing took place in June 2014. By this time the Claimant had made a good recovery. He had started a 3-year degree course in archaeology. He was doing bar work at a public house in his spare time. He had a "vague aspiration" (the ET's words) to become a teacher.

1. The ET awarded him £10,000 for injury to feelings and a separate £5,000 for the exacerbation of his psychiatric condition. It found that if he had acted reasonably to mitigate his loss he should have been able to earn the minimum wage by February 2013 and increased his earnings yet further by August 2013. These findings enabled the parties to reach the figure of £61,244 for past and future loss of earnings. He did not claim loss of earnings beyond the summer of 2015; but it was his case that he would never be able to mitigate his loss of pension benefits.

1. The ET described the police pension plan as containing a "rare benefit". It is by virtually any standards an exceptionally generous scheme, no doubt crafted to take into account the exigencies of police service.

1. The plan is a final salary scheme. During the first 20 years of service there is a one sixtieth accrual rate. Then for years 21 to 30 there is a two sixtieth accrual rate. The "rare benefit" to which the ET referred is the entitlement of a police officer to retire with a pension of two thirds of final salary after 30 years' service. The pension is thereafter index-linked.

1. There is built into the scheme a considerable incentive to complete 30 years service. If an officer does not remain in the scheme for 30 years, but becomes an "early leaver", he or she cannot access unreduced pension benefits until age 60.

1. In these circumstances it is not surprising that (as the ET found) most police officers complete their 30 years and the vast majority then retire and take their pension. They can of course work elsewhere as well as drawing their pension.

1. Applying this to the Claimant's case, if he had not resigned he would have been able to retire on 1 December 2021 at the age of 52 on his full pension. He would, of course, also have been able to work. By reason of his resignation he could not take a pension without reduction until the age of 60, which would be 10 August 2029. Even then it would be less than it would otherwise have been.

1. So there were two periods of pension loss to consider. First, there was the period from age 52 to 60 (about 7 years 8 months) during which time he would not be able to take his pension at all (except with an actuarial reduction). Second, there was the period from 60 onwards, when he would be able to take his pension, but it would have been less than if his service had continued beyond his date of resignation.

1. This brings me conveniently to the Respondent's first ground of appeal. While the ET found that the Claimant ought to obtain alternative employment, largely though not entirely mitigating his loss of earnings, the ET did not make any allowance, in percentage terms or otherwise, for the possibility that he might make up some or all of his final salary pension loss. The substantial loss approach (as to which, see further below) required a value to be placed upon prospective final salary pension rights to normal retirement age in new employment. The ET valued this at nil.

1. On behalf of the Respondent Ms Stout submitted that the ET erred in law in failing to make this allowance. Her argument ran as follows. Compensation in discrimination cases is assessed on the tortious basis: to put the victim in the position in which he would have been but for the unlawful conduct. Where there is a possibility that a particular factor (such as the chance of supervening unemployment or ill-health unrelated to the discrimination) would in any event have affected the Claimant's loss, the tortious measure involves assessing the loss of a chance rather than applying a balance of probabilities test. The ET did not apply this legal approach to the question whether the Claimant would have secured access to a pension scheme and thereby mitigated his pension loss. There was a chance that the Claimant would have become a teacher and if so he would have secured access to the Teacher's Pension Scheme. The ET therefore ought to have reduced his pension loss by an appropriate percentage. She took me to Abbey National plc v Chagger [2010] ICR 397, Thornett v Scope [2007] ICR 236 and Eversheds Legal Services v De Belin [2011] ICR 1137 in support of this approach.

1. On behalf of the Claimant Mr Feeny accepted Ms Stout's legal propositions with the caveat that any contingency or possibility must be realistic or significant before an award is discounted by reason of it. He submitted that the ET found there was no significant chance that the Claimant would be able to reduce his pension loss by employment as a teacher. He took me to relevant passages in the Reasons of the ET. He submitted also that it was for the Respondent to place before the ET evidence to establish the existence of the chance and the value of any likely benefits; the Respondent did not do so.

1. I prefer the submissions of Mr Feeny on this point. My reasons are as follows.

1. When civil courts and tribunals are required to make findings of fact as to what took place in the past they make their findings upon the balance of probabilities. When, however, they are required to assess what would have happened in the past upon a contingency, or what may happen in the future, it is well established that they must not apply a "balance of probabilities" test. Their task is to make an assessment of relevant chances and factor that assessment into the calculation. The question is not whether an event is or would be more likely than not to occur.

1. The Employment Tribunal's task will involve a degree of speculation. As Underhill P put it in De Belin - in a phrase which, once learned, is difficult to forget - "'speculative' is not a dyslogistic term in this field" (paragraph 45). In employment cases the question often arises in the context of dismissal: would the employee have been dismissed at some stage absent the unfair dismissal or discriminatory dismissal for which compensation is to be awarded? It has long been established in unfair dismissal law that this question is to be assessed by an assessment of the chances: see Polkey v AE Dayton Services [1988] ICR 142. Chagger, Thornett and De Belin all involved this question, and the authorities - which include an important passage in Software 2000 Limited v Andrews [2007] ICR 825 - are helpfully gathered together in De Belin. Although the principle often arises in this context in employment law it is a general principle of the measure of damages in tort.

1. I would only draw attention to three points in an area which is now well travelled.

1. Firstly, application of this principle does not require an Employment Tribunal to factor in every imaginable possibility, however remote. Common sense comes into play. The Employment Tribunal is expected to make an assessment of realistic chances. This concept is captured in different ways in the cases: "realistic prospect" in Chagger (paragraph 62), "a sufficiently substantial chance" in De Belin (paragraph 47), "a risk which a fact-finding Tribunal could not ignore" in Thornett (paragraph 35). If a realistic prospect exists the Employment Tribunal must make its best assessment of the chance, taking into account any material and plausible evidence it has from any source, even if that involves speculation.

1. Secondly, application of the principle may involve the application of a percentage, upwards or downwards (as in many Polkey cases), to reflect the Employment Tribunal's assessment of the chance, but it is important to appreciate that the application of a percentage is not the only way of reflecting a chance. Suppose the question is whether an employee would have continued in a particular job for a number of years. There may be a number of realistic chances bearing on that assessment. The best way - indeed often the only practicable way - of expressing the result of the assessment will be to say that the employee would have remained in employment for X years, taking into account the various chances. This way of expressing the result by choosing a mid point or tipping point can look rather like an assessment on the balance of probability, and occasionally Employment Tribunals in my experience wrongly use this form of language to express their conclusion. But as long as the Employment Tribunal actually takes into account the realistic chances either way, it is applying the law correctly.

1. Thirdly, the question for the Employment Appeal Tribunal is whether the Employment Tribunal applied the law correctly. If it did so, the Employment Appeal Tribunal will not interfere with the result.

1. In this case I have no doubt that the ET found there was no realistic chance that the Claimant would find employment with a pension plan. The ET said in paragraph 67:

"67. At what point do we consider that the claimant will find an equivalently remunerated job (Issue 2.2)? We do not consider the claimant will find an equivalently remunerated job prior to retirement at 52 (or indeed 60). It will be apparent from the findings we have made about the Force pension scheme that it is almost a certainty he will not be able to replace a final salary pension scheme with an accrual rate of 1/30th nor earn the capital needed to save to replace such a fringe benefit."

1. The ET found that the Claimant, by taking a degree course and doing part-time bar work, had failed to mitigate his loss. It said:

"68. … In our judgment there has been a wholesale failure to mitigate the lost earnings, albeit we accept there is little to be done about the pension loss given the rare nature of the scheme."

1. Further, when dealing with the Claimant's prospects of obtaining comparable employment, it said:

"70. … We have made our findings in relation to the national minimum wage and later, more remunerative employment on the basis of our industrial knowledge, which on the part of the lay members of this Tribunal is considerable in relation to the local and wider job market. We think it is right to draw on that knowledge in the circumstances. We have rejected the respondent's case that the claimant could have achieved a mean salary in excess of £30,000, in view of the facts set out at paragraphs 33 to 36 above. None of the jobs suggested as being attainable for the claimant included either the advertisement with the essential requirements, nor the qualifications required for the role. The claimant is without qualifications entirely apart from CSEs and his experience and training as a police officer. The latter plainly qualifies him for many things, but not necessarily the jobs suggested or other available jobs in the local market and certainly not at mean salaries in excess of £30,000."

1. To my mind it is plain from these passages that the ET did not think there was a realistic chance that the Claimant would become a teacher. It had mentioned his apparent desire to become a teacher only in passing, commenting that he had no concrete plans in place. His only qualifications were CSEs; he was doing bar work in his partner's business; his choice of undergraduate degree was archaeology.

1. I see no reason to suppose that the ET at this point in its reasoning had a "balance of probability" test in mind. It would have been easy enough for the ET to say that on the balance of probabilities he would not have replaced his final salary scheme. It did not do so. It said that it was "almost a certainty" that he would not do so. This, to my mind, was the ET's way of saying that there was no realistic chance that he would do so. I consider that the ET had the correct test in mind. Accordingly I reject this ground of appeal.

1. In the Respondent's original Notice of Appeal there was a ground (ground 1) which sought to argue that the ET had erred in law by applying a "balance of probabilities" test rather than a "loss of chance" test to other aspects of its assessment of future loss. It was in particular argued that the ET had applied the wrong legal test when considering whether the Claimant's employment with the police would have terminated prior to his likely retirement age of 52. Ground 1 was rejected at the sift stage by Lewis J who commented as follows:

"Grounds 1 and 2 contend that the tribunal failed to assess the chance that the claimant's service would have ended had there been no unlawful discrimination and failed to apply that percentage reduction to the award of compensation. The two matters relied upon as giving rise to a chance that there would have been dismissal are the receipt of housing allowance to which the claimant was not entitled and one occasion when he neglected his duty.

The tribunal directed itself to the relevant law and cited Abbey National plc v Chaggar [2010] ICR 397. It considered that there was no prospect that either the housing allowance issue or the neglect of duty issue would have led to the claimant's dismissal: see paragraphs 74 and 75 of the remedies decision. The tribunal adopted the correct approach and its decision is not perverse. Grounds 1 and 2 do not disclose any reasonable grounds of appeal. No action need be taken in relation to grounds 1 and 2 of the notice of appeal."

1. Ms Stout sought to renew this argument. She wished to argue that Lewis J was wrong to reject ground 1 at the sift stage and that the ET erred in law in concluding that the Claimant's employment with the police would have terminated prior to age 52.

1. The procedure which the EAT operates at the sift stage is governed by Rule 3 of the Employment Appeal Tribunal Rules 1993. The relevant Rules are Rule 3(7), (7A) and (10) which provide as follows:

"(7) Where it appears to a judge or the Registrar that a notice of appeal ….

(a) discloses no reasonable grounds for bringing the appeal; or

(b) is an abuse of the Appeal Tribunal's process or is otherwise likely to obstruct the just disposal of proceedings,

he shall notify the Appellant …. accordingly informing him of the reasons for his opinion and, subject to paragraph (10), no further action shall be taken on the notice of appeal …

(7A) In paragraphs (7) … reference to a notice of appeal … includes reference to part of a notice of appeal …

(10) Subject to paragraph (7ZA), where notification has been given under paragraph (7) and within 28 days of the date the notification was sent, an appellant … expresses dissatisfaction in writing with the reasons given by the judge or Registrar for his opinion, he is entitled to have the matter heard before a judge who shall make a direction as to whether any further action should be taken on the notice of appeal …"

1. The decision of Lewis J to reject ground 1 was taken under Rule 3(7) and (7A). The closing words of Rule 3(7) are entirely clear. Subject to any application under Rule 3(10) no further action may be taken on that ground. In this case there was no application under Rule 3(10). It is not permissible simply to renew the ground at the Full Hearing of the appeal.

1. Although an extension of time can be granted for making an application under Rule 3(10), the Employment Appeal Tribunal does not readily grant an extension of time. It applies the same strict principles as it applies to the institution of an appeal: see Echendu v William Morrison Supermarkets Plc [2008] UKEAT/1675/07 (20 June 2008). In this case, as Ms Stout acknowledged, there was no good excuse justifying an extension of time; and I do not think there is any other factor which would justify me in granting an extension of time.

1. Ms Stout applied in the alternative for permission to amend; but since the ground of appeal is in the original Notice of Appeal the difficulty she faces is not that she requires an amendment: her difficulty is that the ground has been addressed and disposed of.

1. In these circumstances it is not necessary to consider the ET's reasoning on this question any further. I would, however, point out for the avoidance of doubt that it did reduce pension loss by 30% avowedly to take account of the risk of employment ending sooner and the vicissitudes of life.

1. The remaining grounds of appeal and cross-appeal concern the method of calculation adopted by the ET in respect of pension loss. It is now convenient to summarise the ET's conclusions on the question of pension loss and to explain the issues between the parties on this appeal.

1. The ET quoted at length from the decision of the EAT in Chief Constable of the Metropolitan Police v Gardner UKEAT/0174/11, a case to which I will return later in this judgment.

1. The ET separated out its calculation of pension loss into two periods. The first period related to ages 52 to 60. The second period related to 60 onwards. The parties do not object in principle to this separation.

1. The ET calculated the loss from the age of 60 onwards using what is known as the "substantial loss approach". This approach is derived from the third edition of a booklet known as "Compensation for Loss of Pension Rights" (2003) - a document prepared by a committee of what were then Chairmen of Employment Tribunals, the Government Actuary and a member of his department. A valuable summary of this approach and the thinking which underlies it will be found in Griffin v Plymouth Hospital NHS Trust [2014] EWCA Civ 1240 at paragraphs 54 to 68 (Underhill LJ). The parties do not object in principle to the adoption of this approach.

1. I should mention that the ET expressed some misgivings about adopting the substantial loss approach. It noted that the 2003 Booklet "contained actuarial assumptions about life expectancy for men, investment returns and commutation which might or might not still be current". While acknowledging that the guidance was written "in different financial times" it considered it would not be right to depart from the approach in the Booklet.

1. Applying the substantial loss approach, the ET reached a figure of £188,811, using tables 5 and 6 in the 2003 Booklet. It made a reduction of 30% for the vicissitudes of life and the risk that employment might have ended sooner. It made a further reduction for accelerated payment, taking a 2.5% discount rate. This brought its figure to £90,799. Mr Feeny challenges the further reduction for accelerated payment on the basis that the multiplier for the substantial loss approach already takes account of accelerated payment.

1. One of the reasons for separating pension loss into two periods was that the guidance in the 2003 Booklet lent itself to use for the period over the age of 60 (where the calculations were conventional and made use of available actuarial tables), whereas it did not lend itself to use for the period from age 52 to 60. The guidance involved the use of a multiplicand and multiplier. The multiplicands for the two periods would be entirely different and the tables in the 2003 Booklet provided no ready assistance in calculating a multiplier for a man aged 42 in respect of loss accruing over a period in the future representing his loss from age 52 to 60.

1. In respect of this period the ET started with a figure of £60,000. Once again it made a reduction of 30% for the vicissitudes of life and the risk that employment might have ended sooner. Once again it made a reduction for accelerated payment, taking a 2.5% discount rate. This brought its figure to £36,078. Mr Feeny challenges this award. He says that the figure of £60,000 was too low, and that in any event the ET's bespoke approach had already taken into account accelerated payment.

1. The ET's calculation of pension loss appears to have taken gross figures without making any deduction for the tax which the Claimant would have paid on his pension when received. Its total award for loss of earnings and pension loss was £188,390. The ET grossed up its award to allow for the tax that would be payable on the award: the figures were calculated by the parties' representatives. Its eventual figure for loss of earnings and pension loss was therefore £243,373. I am not concerned on this appeal with the loss of earnings award; but it is apparent from the figures that a substantial element of the grossing up related to the award for pension loss. Ms Stout argues that this grossing up was wrong. I now turn to the grounds in detail.

1. It will be recalled that the award of loss from age 52 to age 60 was to cover a period of 7 years 8 months when the Claimant would be unable to take his pension at all, whereas (but for the Respondent's unlawful conduct) he would have been able to take in his pension in full and also work. The Claimant's annual pension loss over this period was put at £24,534. This figure appears to have been accepted by the ET.

1. It was the case for the Claimant that there should be a straight line calculation for the period from age 52 to age 60. It was not conceded (at least in the Claimant's schedule and written submissions) that there should be a discount for vicissitudes of life, risk that employment would have ended sooner and accelerated payment. One would, however, expect such discounts. Even with discounts, however, one might expect a figure of the order of £100,000.

1. The Respondent's calculations prior to the hearing had not differentiated between the period from age 52 to age 60 and the subsequent period. Its calculations had not adopted the substantial loss approach from the 2003 Booklet. In a witness statement by Ms Milling, a payroll manager, it had provided figures based on crystallised benefit values at 1 February 2012 and 2 December 2021 and on the approach of the HMRC to the valuation of pension pots for tax purposes.

1. Having said that it was applying the substantial loss approach to the period after the age of 60, for which it awarded £188,811, the ET continued as follows:

"84. … it seems to us that there is a considerable risk in awarding compensation on the straight line Mr Feeny suggests, which is neither adopted in the 2003 Booklet, the Ogden tables or anywhere else. It potentially produces a windfall for the claimant: he would receive now a sum which on Mr Feeny's figures and before reduction would amount to some £410,000, which, if invested now, would put the claimant in a considerably better position than he would have been absent discrimination; he would have the benefit of cash sums immediately which would be payable, on his case, in tranches over thirty years or more.

85. For these reasons, and on the basis that neither the Booklet, nor the approaches contended for by the parties, appear to address [loss over both periods] which we are satisfied will be sustained by the claimant, we have adopted a bespoke approach to calculate [loss from 52 to 60]. We add here we would rather not be in this position, but we cannot see a just alternative without expert actuarial advice, which neither of the parties have sought, notwithstanding the unusual circumstances. We have assumed a life expectancy for someone in the claimant's circumstances of age 79, in accordance with the tables; his approximate cash pension loss over the period from the ending of his employment to that age then, will be 9 x £24,000 … followed by 19 x £7000 (the difference between his accrued pension and pension had he remained in service) … That cash loss is £216,000 plus £133,000. I caveat this by saying that we are prepared to be corrected on our maths, but that produces a total cash loss over 28 years of £349,000. If Mr Feeny's approach was to be adopted (the substantial loss approach for [age 60 onwards] and the cash approach [for age 52 to 60], the claimant would be overcompensated in our view (receiving £410,000 now for a future cash loss commencing in 2021 of £349,000).

86. In relation to [loss for age 52 to 60] then, we consider we have to do something different and bespoke in these very unusual circumstances. We have decided that the right approach, bearing in mind principles of accelerated receipt, upon which we have yet to hear submissions, is to average the £349,000 cash loss over 28 years, producing an average annual loss of pension benefit of £12,500. What sum would we need to award to the claimant to compensate him for the loss of a prospective annual pension of £12,500 over 29 years, commencing in 2021? We consider the just multiplier for that is that used by HMRC to value pension benefits generally, giving a total sum of £250,000. There then falls to be deducted £190,000 (the sum to be awarded under the substantial loss approach). That seems to us to strike the right balance in addressing a highly unusual state of affairs and loss to be sustained by the claimant as a result of acts of discrimination. We note, as Mr Gibson says, that the ultimate figure is not a million miles away from the CETV approach he advocated, but we have satisfied ourselves that the claimant is not prejudiced by the simple application of an approach which is neither recommended by the 2003 Booklet, nor appears to take account of the acute and accelerated losses to be sustained if an officer leaves service in the last third of the common thirty year term."

1. This was the ET's reasoning for awarding £60,000 in respect of the period from 52 to 60. As I have already explained, it then discounted it further both for vicissitudes of life and for accelerated payment, awarding in the end just £36,078 in respect of the period from 52 to 60.

1. Mr Feeny submitted that the ET fell into error in the way it calculated the loss for the period from 52 to 60. Its task was to assess the amount which, paid in 2014, would replace a loss of £24,534.67 annually between December 2021 and August 2029. It was no part of its task to "strike a balance": this was an error of law. Moreover it adopted a 20 year multiplier which it derived from HMRC; but it had already rejected this multiplier earlier in its reasoning. Moreover the result was perverse. The sum of £60,000 was plainly inadequate to compensate for loss of £24,534.67 per annum starting in just 7 years' time. He demonstrated this by the use of the Ogden Tables approved for use in personal injury cases, using Table 28 to establish a multiplier for the period and table 27 to establish a discount for accelerated receipt. He calculated (though I think some details of the calculation would require revision, and the figure did not allow for vicissitudes of life) that a figure of £149,000 would be required, adopting 2.5% rates.

1. Ms Stout submitted that the ET had explained its reasoning sufficiently in paragraphs 84 to 86 of its Reasons; that its reasoning was "cogent, intelligible and appropriate"; and that it therefore did not commit any error of law. The 2003 Booklet did not provide any ready means of calculating the loss from age 52 to age 60. If the ET adopted the substantial loss approach for the period from age 60 onwards it was therefore bound to find a "bespoke" solution for the earlier period. It was entitled to use HMRC figures: the law did not require any one solution to be adopted.

1. I start by considering the role of the EAT. Since there is an appeal only on a question of law, the EAT must not upset an assessment of pension loss unless it is founded on incorrect legal principles, is not properly reasoned, or is perverse.

1. The role of the ET was to award the Claimant full and fair compensation for his pension loss. The law did not require it to adopt a particular method of assessment so long as the method it chose was calculated to provide full and fair compensation. It was entitled to adopt the methodology contained in the 2003 Booklet or in the Ogden Tables which are now customarily used in personal injury claims, but it was not bound to adopt either of these methodologies: see Gardner at paragraph 83 and Griffin at paragraph 81. Gardner, however, sounds a cautionary note:

"84. If it does depart from such an approach, however, it must do so for good reason and must say what its reasons are. Any such reasons require to be cogent and intelligible and appropriate. A Tribunal will only ever stray from such recognised approaches for such carefully articulated good reasons, and will do so at potential peril to the acceptability of its decision. It is not a course which we would recommend, except where it is plain that the interests of justice require it."

1. In this case the ET started from the proposition that the Claimant's loss from the age of 60 would be properly compensated by the substantial loss approach. The parties do not quarrel with this proposition; and the ET was plainly entitled in law to reach this conclusion. The substantial loss approach is rather more generous than the Ogden Table approach, not least because it takes into account that over the years wages have tended to outstrip inflation: see Gardner at paragraph 86. This has, of course, not been the picture for the last few years; but the ET was looking to the future when it is hoped normality will return. On the other hand the 2003 Booklet adopts a discount rate of 2.5% which it is not required by law to adopt and which is arguably too high in current conditions. At all events, the ET's decision to adopt the substantial loss approach, while not inevitable, was certainly lawful.

1. The ET then had to grapple with the problem that the 2003 Booklet contained no tables that readily enabled it to calculate loss from age 52 to age 60. With respect to the ET, this should not have been a great problem. It would not have been difficult to calculate the loss for a period of 7 years 8 months and discount by one means or another to allow for accelerated payment of that loss. It could, for example, have done so by taking as its starting point tables 28 and 27 of the Ogden Tables (7th edition, 2011). This, together with a discount for the vicissitudes of life, mortality and the risk of earlier job loss, would have produced a satisfactory figure. It would, however, have been significantly higher than the figure awarded.

1. The effect of the ET's reasoning in paragraph 87 was, however, to abandon the substantial loss approach altogether and replace it with a multiplicand based on its own judgment and a multiplier derived from HMRC usage for a purpose entirely different from assessment of tortious loss. I do not think the ET acknowledged that it had abandoned the substantial loss approach, for it referred to it again in its summary and was careful to leave the actual figure for the years after 60 untouched. But, because it regarded its figures as a cap within which the whole award had to be contained, the practical effect was to abandon the substantial loss approach.

1. I do not think the ET's approach in paragraph 87 can be described as cogent, intelligible and appropriate as a means of awarding full and fair compensation. Once granted that it had found the substantial loss approach to be appropriate for the years after age 60, an overall multiplier of 20 was plainly too low: it added only 2.5 to the multiplier which it had derived from appendix 5 to the 2003 Booklet. Moreover averaging out the multiplicand to produce a figure of £13,500 was unsatisfactory because it did not recognise that the highest level of loss was in the earliest years. I cannot see any rational basis upon which the ET could have thought that a loss of £24,534 per year for 7 years 8 months could be compensated by a figure of £60,000 before taking account of vicissitudes or accelerated payment.

1. The Claimant's cross-appeal will therefore be allowed on this ground: the ET erred in law in its assessment of pension loss in respect of the years from age 52 to age 60.

1. I then turn to Mr Feeny's submission that the ET was wrong to discount the award in respect of loss over the age of 60 for accelerated payment. His submission is a short one: accelerated payment is already taken into account by the relevant tables in the 2003 Booklet.

1. This submission appears to me to be correct. I cannot find an explicit statement in the 2003 Booklet that accelerated payment is taken into account in the relevant tables (which are tables 5 and 6). However the tables are intended to serve a similar function to the Ogden Tables: see page 43. The Ogden Tables take accelerated payment into account: see paragraph 3 to the Introduction to the 7th edition. There would be a remarkable lacuna in the 2003 Booklet's explanation of the substantial loss approach if the tables did not take accelerated payment into account. Accordingly I uphold the cross-appeal on this ground also.

1. I have already concluded that in respect of the period prior to the age of 60 the appeal will be allowed. Whatever method is used to assess loss in respect of this period will of course have to take into account accelerated payment.

1. The ET did not say whether the figures which it took for the Claimant's annual loss of pension were gross or net of tax. Although I do not think Mr Feeny quite conceded this point, it is to my mind plain that the figures which it took were gross; there is no reason to think that any tax calculation was undertaken at any stage prior to their inclusion in the Claimant's schedule, and the Claimant undertook no tax calculation. The point appears to have passed unnoticed and unchallenged at the hearing.

1. The ET, as I have explained, grossed up its award to allow for tax which would be payable on the compensation for pension loss.

1. The legal context for the parties' submissions on this question is well known: I will state it briefly.

1. In an action for damages in tort (or, for that matter, for wrongful dismissal) the calculation of the Claimant's compensation in respect of loss of earnings, past and prospective, must take into account the tax which would have been payable upon them. See British Transport Commission v Gourley [1956] AC 185. There is no reason for the approach to be any different as regards pension loss.

1. It is not in dispute that in this case the Claimant would have paid tax on his pension income when received. The starting point is accordingly that his pension loss should be calculated using figures net of tax. This, of course, involves making some broad estimation of future taxation - a process with which personal injury and employment lawyers and forensic accountants are familiar.

1. A complication arises, however, if the award of compensation would itself be taxable. If the loss of earning or pension has been assessed net of tax, but tax would be payable on the award of compensation, the Claimant will not be awarded full compensation.

1. There are two ways of dealing with this complication. Authority can be found for both of them.

1. The first is a rough and ready way: it is to assess the loss using gross figures, on the basis that the additional amount included in the assessment of loss will in a rough and ready way compensate the Claimant for the tax he will pay on the award of compensation.

1. The second is a more principled way: it is to assess the loss using net figures, but then to add an allowance to the award of compensation for the tax which will be payable on it. This process is known as "grossing up". Ms Stout submitted to me that this was the better approach, and the one more generally adopted: she referred to Harvey on Industrial Relations Law at B11-27 and the cases there cited. I agree with her. But this does depend on the parties putting forward the appropriate calculations to the tribunal.

1. Against this background, I can turn to the submissions of the parties.

1. Ms Stout's first submission was that, since the pension loss was assessed gross, there was no basis for grossing up the award as well. In principle I can see no answer to this submission. As we have seen, the general rule is that loss of earnings and pension must be assessed net of tax. The only rationale for using gross figures is the one which I have already outlined: it can be a rough and ready way of taking into account tax payable on the award. But if that is the rationale for taking gross figures there is no foundation for "grossing up" the award as well.

1. Mr Feeny, as I have said, did not necessarily accept that the pension loss was assessed gross: I have found that it was. Otherwise I do not think he put forward any justification for grossing up the award after using gross figures for loss. He did, however, say that the point had not been taken below.

1. It would appear to be the case that no objection was taken to grossing up below (although I note that this point is itself not taken in the Respondent's Answer). The Employment Appeal Tribunal has power to permit a point to be argued on appeal even if it was not taken below. That power is closely circumscribed: the authorities are discussed and summarised in Secretary of State for Health v Rance [2007] IRLR 665 at paragraph 46 to 50. In this case I am satisfied that the circumstances are such that I should allow the point to be taken. The Claimant's schedule should have been prepared using net figures for pension loss and (if it was desired to claim grossing up) should have dealt with grossing up. It did not. The Claimant should not have put forward gross figures in the Schedule of Loss if it was desired also to claim "grossing up" of the award. Once the point is identified it is an obvious one: it was an error of law to award loss of pension gross of tax and then gross up the award for tax. It can be corrected in a straightforward way which does not require a further hearing by adopting the "rough and ready" approach which I have identified.

1. Ms Stout's second submission was, however, more fundamental. It was that there is no need to gross up the award in any case because it would not be taxable. I think it is fair to say that she pursued this ground with less enthusiasm.

1. The argument derives some support from a passage in the 2003 Guidelines. This passage, which is found in the Government Actuary's memorandum of explanation at Appendix 2, reads as follows:

"Awards made by Employment Tribunals for loss of pension rights are tax free; however, tax is payable on the income and gains arising through any investment of the award, although it may be possible to defray the extent of this by investing in suitable tax efficient vehicles. After discussion with the ET chairmen on the working party, it was agreed that some allowance for the possible effects of tax being payable on the proceeds arising from investment of the compensation award should be made in determining the financial assumptions to be used for the rates of return net of earnings and net of revaluation of deferred pension."

If this passage were correct - so that awards made by Employment Tribunals for loss of pension rights were tax free - there would be no reason to gross up the award even if pension loss had been calculated net.

1. I do not think, however, that it represents the current law. That part of an award for unlawful discrimination which is a "payment … received directly or indirectly in consideration of or in consequence of , or otherwise in connection with … the termination of a person's employment" is taxable in accordance with Part 6, Chapter 3 of the Income Tax (Earnings and Pensions) Act 2003. Thus in Walker v Adams [2003] STC 269 an award for loss of earnings and pension rights was held to be taxable on similar wording in the Income and Corporation Taxes Act 1988. I note from Simon's Tax Cases (E4.816 under the heading "compensation for discrimination") that HMRC in practice applies the law in this way.

1. The decision of the EAT in Yorkshire Housing Limited v Cuerden [2010] UKEAT/ 0397/09 (Judge Peter Clark presiding) proceeds on that basis, and I have no doubt that it was correct. In that case the EAT also dealt with an argument that the award was exempt from tax by reason of section 407, which creates an exception to the charge under Chapter 3 for a payment "under a tax-exempt pension scheme". The EAT held that this provision was concerned with payments to a beneficiary out of a pension scheme not with a Tribunal award of compensation on termination of employment: see paragraph 31. The Respondent's Notice of Appeal argued that this decision was incorrect: I have no doubt that the decision was correct.

1. Ms Stout put forward a third reason (which was not in the Notice of Appeal even in a vestigial form, and would have required permission to appeal, and which was not argued below) why the pension loss should not have been grossed up. She submitted that it will be open to the Claimant, on receiving his award for pension loss, to invest it in pension schemes again and obtain tax relief upon it. She took me through provisions of the taxing legislation in order to demonstrate that this was the case. Therefore, she submitted, there was no reason to gross up for taxation. I would not allow this point to be taken. Generally speaking it is a matter for an individual how he invests his award, so that what he will actually do with it is irrelevant: see for example Wells v Wells [1999] 1 AC 345 at 371C, 373H and 394H-395A. In any event the Claimant lost his rights in a pension scheme in the public sector. It is not self evident that he should be required to invest his award in quite different types of scheme in the private sector; there was before the ET and there is before me no evidence of the advantages and disadvantages of doing so.

1. For these reasons the appeal will be allowed in part and the cross-appeal will be allowed.

1. In some respects it is only a matter of arithmetic to give effect to the result of the appeal and the cross-appeal. This seems to me to be the position in respect of the grounds relating to accelerated payment in relation to the period over the age of 60 and in relation to taxation. The parties are asked to agree the figures for inclusion in my order or to explain to me any differences that they have.

1. The position is not quite so straightforward in respect of loss for the period from the age of 52 to 60. While I believe it should not be difficult for the parties to agree a figure based on the Ogden Tables I have mentioned, I do not think I am in a position to decide a figure unless the parties agree to my doing so: see [Jafri v Lincoln College ]()[2014] ICR 920 at paragraph 21 (Laws LJ) and paragraphs 45 to 47 (Underhill LJ). The matter will accordingly have to be remitted unless the parties agree a figure themselves or agree to my doing so (which I would, as presently advised, be prepared to do after exchange of written submissions). The parties are given a period of 21 days from the handing down of this Judgment to do the work required by paragraph 81 of this Judgment and to decide how to address the question in this paragraph - i.e. by agreeing a figure, by agreeing that I should decide the figure, or inviting me to remit the matter, and if so making submissions to whom it should be remitted.