At the Tribunal
Judgment delivered on 24th July 1995
THE HONOURABLE MR JUSTICE TUCKER
MRS T MARSLAND
MR T C THOMAS CBE
Transcript of Proceedings
JUDGMENT
Revised
APPEARANCES
For the Appellant MR R MCMANUS
(of Counsel)
The Treasury Solicitor
Room 449
Lacon House
Theobalds Road
London WC1X 8RY
For the Respondents MR P FORTUNE
(of Counsel)
Adams Blair Cox
29/31 Guildhall Walk
Portsmouth
PO1 2RY
MR JUSTICE TUCKER: We have had before us 6 appeals from decisions of Industrial Tribunals relating to servicewomen who were dismissed because of their pregnancy. Such dismissals are and were unlawful, as the employers, the Ministry of Defence, conceded. The question for decision by each of the Industrial Tribunals was that of compensation. In two of the appeals the Appellants are the Ministry of Defence, and in the remaining four appeals the Appellants are the ex-servicewomen. Two of the appeals, those of Mrs Collins and Mrs Dawson raised discrete and separate issues and we have already delivered our judgments in those two cases.
Such cases have been considered by the Appeal Tribunal on many occasions, notably in MOD v Cannock and Others [1994] ICR 918, where the Employment Appeal Tribunal sought to give general guidance to Industrial Tribunals on their approach to such cases. Although the guidance was obiter to the decisions, it was given by an experienced Appeal Tribunal after careful submissions by distinguished leading Counsel. While each application must be viewed individually, it is to be hoped that guidance such as this will if possible be followed, since it is highly desirable that there should be certainty in these matters. Some Chairmen of Tribunals have questioned some aspects of those guidelines.
The principal ground of each of these appeals is that the decisions of the Industrial Tribunals were perverse in the assessments they made of the chances that the applicants would have returned to work, and as to the duration of their service.
It was contended on behalf of the two Respondents to the Ministry of Defence appeals that it is impermissible to raise the question of perversity without having obtained the Chairman's Notes of Evidence; and that the appeals on this ground should not be entertained. It is correctly pointed out that in the two appeals by them the Ministry of Defence made no attempt to obtain these Notes until a very later stage. It was only shortly before the Appeals were due to be heard, that any application for the Notes was made. It was dealt with by the President of the Employment Appeal Tribunal, who refused the application, on the grounds that it was too late, and that he was not persuaded that they were necessary for the fair disposal of the appeals. A similar application was made by the Appellant Servicewoman in another appeal. Again the President refused it, on the ground that he did not consider that the Notes would be helpful.
In this context we have been referred to a number of authorities. The first of these was the decision of the Court of Appeal in Martin v MBS Fastenings (Glynwed) Distribution Ltd. [1983] IRLR 198. In that case Sir John Donaldson MR said this:-
"It is very important, and sometimes difficult, to remember that where a right of appeal is confined to questions of law, the appellate Tribunal must loyally accept the findings of fact with which it is presented and where, as can happen from time to time, it is convinced that it would have reached a different conclusion of fact, it must resist the strong temptation to treat what are in truth findings of fact as holdings of law or mixed findings of fact and law. The correct approach involves a recognition that Parliament has constituted the Industrial Tribunal the only Tribunal of fact and that conclusions of fact must be accepted unless it is apparent that on the evidence, no reasonable Tribunal could have reached them. If such be the case, and happily it is a rarity, the Tribunal, which is to be assumed to be a reasonable Tribunal, must have misdirected itself in law and the Employment Appeal Tribunal will be entitled to intervene."
and later said:-
"It is also submitted that the Tribunal's findings of fact were perverse. This involves the proposition that on the evidence no reasonable Tribunal could have reached the same conclusion. But neither we nor the Employment Appeal Tribunal had any note of the evidence which the Tribunal heard. It is no part of the duty of a Tribunal setting out its reasons to record all the evidence. In practice, in telling the story, the Tribunal will often advert to parts of the evidence, but no court having an appellated jurisdiction limited to question of law is entitled to assume that this is the totality of the evidence. If it is intended to appeal upon the ground that there was not evidence to support the Tribunal's findings, the appellant must take the necessary steps to obtain a note of the evidence."
We were also referred to another decision of the Court of Appeal in Piggott Bros. & Co. v Jackson [1991] IRLR 309. In that case Lord Donaldson MR again referred to the difficulties of raising a question of perversity without reference to the notes of evidence. He said this is paragraph 13:-
"I fail to understand how if an appeal is based upon, or includes, an allegation that the Industrial Tribunal's decision was perverse, it is possible to contemplate allowing the appeal without having access to all the evidence bearing on the alleged perversity."
And in paragraph 17 Lord Donaldson, having referred to the exposition of May LJ in Neale V Hereford and Worcester CC [1986] IRLR 168 at page 173, went on to say this:
"Nevertheless, it is an approach which is not without its perils. A finding of fact which is unsupported by any evidence clearly involves an error of law. the Tribunal cannot have directed itself, as it should, that findings of fact need some evidence to support them. The danger in the approach of May LJ is that an appellate court can very easily persuade itself that, as it would certainly not have reached the same conclusion, the Tribunal which did so was `certainly wrong'. Furthermore, the more dogmatic the temperament of the judges concerned, the more likely they are to take this view. However, this is a classic non sequitur. It does not matter whether, with whatever degree of certainty, the appellated court considers that it would have reached a different conclusions. What matters is whether the decision under appeal was a permissable option. To answer that question in the negative in the context of employment law, the EAT will almost always have to be able to identify a finding of fact which was unsupported by any evidence or a clear self-misdirection in law by the Industrial Tribunal. if it cannot do this, it should re-examine with the greatest care its preliminary conclusion that the decision under appeal was not a permissable option and had to be characterised as `perverse'.
The most recent authority on perversity to which we were referred is Stewart v Cleveland Guest (Engineering) Ltd [1994] IRLR 440, a decision of the Employment Appeal Tribunal presided over by the President. At paragraph 33 Mummery J. said this:-
"Whenever an appeal is based on the perversity ground, this Tribunal must be extremely cautious not to conclude that the decision of the Industrial Tribunal is flawed because the Appeal Tribunal would have reached a different conclusion on the evidence or thinks that another Industrial Tribunal would have reached a different conclusion on the evidence. An appeal should not be allowed on this ground simply because the Employment Appeal Tribunal disagrees with the Industrial Tribunal as to the justice of the result, the merits of the case or the interpretation of the facts. This Tribunal should only interfere with the decision of the Industrial Tribunal where the conclusion of that Tribunal on the evidence before it is `irrational', `offends reason', `is certainly wrong' or `is not a permissible option' or `is fundamentally wrong' or `is outrageous' or `makes absolutely no sense' or `flies in the face of properly informed logic'. This variety of phraseology is taken from a number of well-known cases which describe the circumstances in which this Tribunal (and higher courts) have characterised perversity. The result is that it is rare or exceptional for an appeal to succeed on the ground of perversity. The reason why it is a heavy burden to discharge is that it had been recognised by those with wide experience and practical wisdom that there are many factual situations arising in the field of industrial relations, including sex discrimination, in which different conclusions may be reached by difference tribunals, all within the realm of reasonableness. It is an area in which there may be no `right answer'. The consequence of this approach also approved in cases of high authority, is that it is not appropriate or fruitful to subject the language of the decision of the Industrial Tribunal to `meticulous criticism' or `detailed analysis' or to trawl through it with a fine-tooth comb'. What matters is the substance of the Tribunal's decision looked at `broadly and fairly' to see if the reasons given for the decision are sufficiently expressed to inform the parties as to why they won or lost the case and to enable their advisers to identify an error of law that may have occurred in reaching the conclusion. Viewed in that way, the decision of the Industrial Tribunal is not perverse."
We have borne these decision well in mind when considering these appeals. In some of these case we might not have arrived at the same conclusions as the Industrial Tribunal did. But that is not sufficient to cause us to interfere. We have only done so where we feel compelled to do so because we have been persuaded that the Tribunal's conclusion is plainly wrong or irrational.
We have felt able to deal with all the cases without having the Chairman's Notes of Evidence; and to remit one case for re-hearing despite not having seen them, for reasons which will appear. However, in future cases where perversity is alleged, those who advise the Appellants ought to consider in good time whether Notice of Evidence should be applied for, and should make any necessary applications well in advance of the hearing. It is no good making applications the week before the Employment Appeal Tribunal has a list of 6 or more such appeals to deal with. Such belated applications are likely to receive very little sympathy.
The first piece of general guidance offered by the Appeal Tribunal in Cannock was that the Industrial Tribunal should assess compensation in these cases by making an estimate of the chances of the events occurring which are relevant to the evaluation of the award. Such matters are likely to be, first and obviously, the likelihood of the return to service in the armed forces after a period of maternity leave; second, the prospects of promotion; and third, the possibility of continuing in the services for the whole period of the engagement.
That this is the correct approach has not been doubted before us. It has long been so in the field of personal injury litigation (see e.g. Mallett v McMonagle [1970] AC166, and Davies v Taylor [1974] AC 207). It is clearly the appropriate approach in cases such as the present, and we respectfully endorse the views to this effect contained in the judgment of the Appeal Tribunal in Cannock.
The criticisms made are not as to the adoption of this approach, but as to the methods by which Industrial Tribunals carried it out. In particular, it is contended that undue effect was given in some cases to the evidence given by the individual applicants as to what would have occurred but for the dismissals, and too little or no effect to statistical evidence as to what has happened in other cases.
The evidence of an ex-servicewoman as to whether she would have returned is obviously relevant, and is something to which an Industrial Tribunal would attach importance. But as was said in Cannock (page 951 B-C):
"... it is merely one piece of the relevant material, although it is evidence of a self-serving nature."
There is other material which is normally available, and which was available in some
of the present cases. We refer to statistical material, showing that the percentage of servicewomen who, having been given the option to return to work after childbirth was 46%. As the Employment Appeal Tribunal said in Cannock that statistical evidence is likely to be a good starting point, not only on the question of a return to work, but also on the question of length of service. Such evidence should not be overlooked or ignored, as the Ministry of Defence claim happened in some of the cases.
However, if an Industrial Tribunal carries out the assessment in a proper way, taking all relevant considerations into account, and disregarding matters that are irrelevant, there should be no reduction made from the assessment of compensation which follows from that exercise, simply because the amount seems excessive. If it is established that an individual has suffered a substantial loss, then it is the duty of the Tribunal to award that loss as compensations so as to ensure that the loss and damage actually sustained as a result of the discriminatory dismissal is made good in full in accordance with applicable national rules. (See Marshall v Southampton and S.W. Hants H.A. (Teaching) No. 2 Case C-271/91 [1993] ICR 893 at page 932). When the Employment Appeal Tribunal in Cannock enjoined the Industrial Tribunals to have a due sense of proportion, they were not suggesting that that principle should be overlooked, but that Industrial Tribunals should make sure when calculating the total award, that it was "a sensible and just reflection of the chances which have been assessed" (page 950H). This was sound advice to the Industrial Tribunals to keep their feet on the ground, and to bear in mind that they are not dealing with people who have been incapacitated by grave injury from following any employment at all. The sense of proportion to which reference was made, should apply to the assessment of the percentage to be adopted as a reflection of the likelihood of a material event occurring.
It has been acknowledged before us (and could not have been disputed) that the Employment Appeal Tribunal is entitled to lay down guidelines for the assistance of the Industrial Tribunals. In another context, the Employment Appeal Tribunal has made it clear that guidance which it seeks to give is not to be regarded as if it were a rule of law or an Act of Parliament. (See Williams v Compair Maxam Ltd [1982] IRLR 83 and Rolls Royce Motors Ltd v Dewhurst [1985] IRLE 184). Such guidance need not be rigidly adopted, and it does not necessarily follow that if an Industrial Tribunal chooses not to do so, its decision will be categorised as perverse. Nevertheless, as we have said, it is to be hoped that such guidelines as we were given in Cannock will be followed, in order that there should be consistency of approach among Industrial Tribunals. In general, it is desirable that guidelines should not be departed from unless there are exception circumstances for doing so.
We now turn to the individual cases.
MR JUSTICE TUCKER: This is an appeal by the Ministry of Defence from a decision of an Industrial Tribunal held at London (South) on 20th June and 15th August 1994. The dates are significant, because the decision in Cannock was not delivered until 29th July 1994. However, the decision was considered by the Tribunal, who must have been one of the first to do so. They were told by Mr Fortune, who also represented the Respondent before us, that the guidelines given by the Employment Appeal Tribunal were not binding on them. Nevertheless, although expressing doubts, about it, the Tribunal did follow and apply those guidelines, without derogating from their position as a Tribunal of fact. The Appellants submit that what they describe as the open doubting of Cannock influenced the Tribunal's minds when it came to the question of assessing chances.
What the Tribunal decided was that there was 100% chance that the Respondent would have resumed here career, that there was only a 10% chance that she might have left the Royal Navy before the expiry of 22 years, and that there was a 4 in 7 chance that she would have been promoted to CPO in 1988.
Mrs Bristow was born in July 1952, so she is now 42. She came from a family with a service background. She joined the WRNS in 1970 when she was just 18. She trained in Radar and became a leading Wren in 1972. She extended her periods of service to, eventually 22 years. In 1978 she married a sailor. He had been married before and had three children, whom he had to maintain.
In 1982, at the age of 30, Mrs Bristow became pregnant. She was discharged from the Royal Navy in November of that year. She was very upset about it. She did her best to find other employment. In due course she accompanied her husband on a posting to Belgium. He left the Royal Navy in 1988. They lived first in North Wales and later in Sheffield.
The Tribunal set out their conclusion, and the reasons for them, in Paragraph 19 of their decision. It is clear to us, as we have already said, that the Tribunal faithfully applied Cannock guidelines and adopted the "chance" approach. We do not find any substance in the criticism that they allowed their minds to be affected by the doubts which they had earlier expressed. In any event, there was evidence before them, given by the Respondent, which entitled them to find as they did.
The first Ground of Appeal which was pursued before us, was that in assessing the chance of return to work at 100%, the Tribunal came to a decision which no reasonable Tribunal could have come to, in that it failed to have regard to the relevant statistical evidence, to the fact that the Respondent did not do any full time work for five years after her dismissal, or to the fact that she appeared to prefer to follow her husband rather than pursue a career. Similar criticisms are made of the Tribunal's other assessments.
The Tribunal was undoubtedly referred to the Ministry of Defence statistics, which we have referred to in connection with an earlier appeal. Those statistics were criticised by the Respondent's Counsel, Mr Fortune, because he said that there was no evidence in the present case that 90% of service people obtained work. The Tribunal (at paragraph 15) agreed with that. They relied, as they were entitled to do, on their own experience.
In at least one other case which we have had to consider, a witness was called to support the statistical evidence, and to proffer himself for cross-examination upon it. That was not done in the present one. While it may be open to parties to present documentary evidence of this kind to a Tribunal, as is often done, it is unlikely to carry as much (if any) weight as if it had been supported by oral testimony. We suggest that those who seek to rely on such evidence would be well advised to call a witness to produce it, and who can assist the Tribunal in its interpretation.
In any event, Mr Fortune, having told us that the statistical evidence was before the Tribunal, submitted that it is not incumbent upon them to make specific reference to it in their decision, and that where the decision is silent about a particular issue, we cannot infer perversity without more.
In our opinion this is correct. We do not think it right to say that this Tribunal was perverse in not paying regard to the statistical evidence. This is another case where the Tribunal saw and heard the witness, and clearly preferred her oral evidence and their own experience to the documentary statistics. In our opinion the Tribunal were in this case entitled to focus on the individual Respondent's actual history and what she said she would have done.
Mr Fortune submitted that the Tribunal were entitled to make a 100% finding that the Respondent would have been ready, willing and able to resume her career in the WRNS, since this was almost entirely a question of fact, and in this we think he is also correct.
As for the finding that there was only a 10% chance that Mrs Bristow might have left the Royal Navy before the expiry of 22 years, we think that there is a force in Mr Fortune's point that she would have only have had to stay on a few years to complete her service, and obtain an immediate pension, after her husband left the services in 1988.
However, there are two matters in this decision which do cause us some concern.
The first arises from comments contained in Paragraph 12 of the decision, dealing with the question of the Respondent's duty to mitigate her loss. After a reference to the case of Banco de Portugal v Waterlow [1932] AC 452, the Tribunal go on to say this:
"The approach in the employment field is that where an employer by his unlawful discrimination has put an employee in the position of having to seek new employment in a difficult market or in circumstances where she is disadvantaged (whether by reason of having a baby or care for) or some other problem the steps which the employee takes to mitigate her loss are not to be weighed too precisely at the instance of the employer."
The Appellants submitted that the Tribunal were wrong to refer to the disadvantage of having a baby to care for, since they said it was clear from Cannock at page 953G that:
"... Tribunals should approach the question of mitigation on the hypothesis that the applicant was fully fit for onerous employment duties 6 months after the birth of the first child."
An applicant who has been dismissed by reason of pregnancy may have good reasons for not looking for other work, but that does not mean that she can "stay at home with the child" and recover compensation. She is complaining of the loss of employment, often the loss of a career. After the six-month period she is expected to be in the job market actively looking for work and applying for jobs and, if she is not, then however understandable her behaviour might be, she cannot recover damages for loss of employment thereafter.
The Appellants submitted that the Industrial Tribunal's approach was wrongly based on consideration of gender.
When the Industrial Tribunal returned to the question of mitigation of loss at Paragraph 20 of their decision, they found that there had been a failure to mitigate only at the point where the Respondent had left full-time work to follow her husband to Sheffield in September 1991. The Appellants submitted that this finding was perverse in the light of the evidence that:
(a) the Applicant did not even seek employment until October 1983,
(b) that the Applicant only did part-time employment in the period up to November 1987, 5 years after the dismissal (paragraphs 4-6, Bristow page 14),
(c) that the Applicant chose to go to Belgium with her husband in July 1984 (paragraph 6, Bristow page 14) which would have clearly put her at a disadvantage in the labour market.
It was submitted that this finding was also reached without reference to the statistics in Cannock that 90% of service personnel are able to obtain employment after having left.
Mr Fortune submitted that the comments in Paragraph 12 were not gender based, but were simply examples of the Industrial Tribunal having regard to the labour market and other relevant factors such as the Respondent's age, but we do not agree with him about this.
We think the Industrial Tribunal were wrong in law to refer to the disadvantages of having a child to care for, and that they failed to take relevant evidence into account in determining the stage at which the Respondent failed to mitigate her loss, and that the whole question of mitigation should be reconsidered.
The second point which troubles us arises from Paragraph 4 of the decision dealing with the calculation. We are aware of the fact that the actual calculation was left to the parties, but in our opinion the Industrial Tribunal were in error in the way they set out the method of calculation. The deduction of the 10% discount was in the wrong order.
In our opinion, the point at which the discount should be applied is before the deduction is made for failure to mitigate which should be made in full, and without a discount. We agree with the Appellants' submissions that mitigation is not a question of loss of a chance, but one of fact. The figure is definite where there has been mitigation in fact, and can be certainly calculated where there has not been, but should have been, mitigation. This is the course which was adopted by the Industrial Tribunal at Exeter in another appeal before us, that of Mrs V Bennett v MOD.
We take the view that the position here is analogous to the situation which the Employment Appeal Tribunal had to consider in Derwent Coachworks v Kirby [1994] IRLR 639. We are glad to learn that at least one Chairman of the Industrial Tribunals takes the same view. In Malseed v MOD, a decision of the Industrial Tribunal sitting at Bristol on 19th April 1995 the Tribunal said this at Paragraph 4 of their decision:
"Our conclusion is that the principle of the Derwent Coachworks case should be applied in the present case. In making the original award we were assessing the applicant's chances, and that calculation produces a figure. It was against that figure we intended that any wages earned in mitigation should be set."
We have no doubt that that is the position so far as the deduction for failure to mitigate ins concerned. We leave open the question whether the same applies to deductions of child care costs.
In our opinion the Tribunal's conclusions on mitigation and calculation are flawed. We allow the appeal to the extent that we remit the case to a differently constituted Industrial Tribunal for reconsideration of these matters.
Finally, there is in this case a Cross-Appeal by Mrs Bristow from the decision in relation to the calculation of pension loss set out in paragraph 23 of the decision. The Industrial Tribunal say this:
"In the absence of evidence as to the employer's contribution this should be calculated in accordance with the Industrial Tribunals Compensation for Loss or Pensions Rights guidelines published by HMSO."
The grounds of appeal are:
1. The guidelines make no provision for calculation of pension loss where the compensation exceeds the Statutory limit of (£11,000).
2. The appropriate method of calculation where an immediate pension is payable and where there is no statutory limit on the level of compensation is by the value method.
Mr Fortune submitted that in this passage the IT gave a decision which they were not invited to give, and was one as to which they had heard no evidence. He observed that the HMSO guidelines were drawn up when the statutory limit or cap was £11,000, and that they are inapplicable to today's situation. He did not ask us to say what the alternative ruling out to have been Mr McManus submitted that in the absence of actuarial evidence the Tribunal were fully entitled to adopt the HMSO Guidelines, and that such a course was sanctioned in Cannock at page 955.
However, we think that this is a matter on which some further argument should be addressed to the Industrial Tribunal, and we remit this point also for further consideration.