SIR RALPH KILNER-BRQWN: This is an appeal by eleven women against the unanimous .decision of an Industrial Tribunal held at Manchester on 16th and 17th October 1986. Their applications that. in each case the woman was entitled to equal pay with that of a male comparator because her work was of equal value were all dismissed.
It is yet another case to come before the Employment Appeal Tribunal where industrial tribunals have endeavored to apply the amendments to the original Equal Pay Act 1970. These were introduced as a result of the decision of the European Court of Justice in 1982 that the Act (already amended once by S.8 of the Sex Discrimination Act 1975) did not fully implement Community law because in the absence of job evaluation the law did not provide for comparison of dissimilar jobs to determine whether they were of equal value. The concept of dissimilar work of equal value was provided for by a new sub sub-section (c) of S.l(2) of the Equal Pay Act l970 . Before January 1984 industrial tribunals and the Employment Appeal Tribunal adopted the position that unless the job evaluation study was manifestly suspect, it should normally be accepted without detailed examination provided that it came within the provisions of S.l(5) of the Act. This reads as follows:-
" (5) A woman is to be regarded as employed on work rated as equivalent with that of any men if, but only if, her job and their job have been given an equal value, in terms of the demand made on a worker under various headings (for instance effort, skill, decision), on a study undertaken with a view to evaluating in those terms the jobs to be done by all or any of the employees in an undertaking or group of undertakings, or would have been given an equal value but for the evaluation being made on a system setting different values for men and women on the same demand under any heading".
It is important to bear in mind that this sub-section remains as an integral part of the Act.
Now a job evaluation study has been an accepted feature ';.f industrial and business life for something like sixty years. The techniques and expertise have become more and more sophisticated as the years have gone by. Whereas originally a job evaluation study was designed for the purpose of providing a basis for a grading and pay structure, it has become in the United Kingdom since 29th December 1975 an integral part of the method of assessing comparisons between men and women when determining whether the work can be rated as equivalent and now since January 1984 whether the work, although dissimilar, can be said to be of equal value. The problems for industrial tribunals have been created by the introduction of the new S.2A which is headed "Procedure before tribunal in certain cases'5 but which in fact requires an industrial tribunal to make a close examination of the evidence and material before it in relation to job evaluation. A job evaluation study can now itself be challenged as discriminatory. This new section reads as follows:-
"(1) Where on a complaint or reference made to an industrial tribunal under section 2 above, a dispute arises as to whether any work is of equal value as mentioned in section 1(2)(c) above the tribunal shall not determine that question unless -
(a) it is satisfied that there are no reasonable grounds for determining that the work is of equal value as so mentioned; or
(b) it has required a member of the panel of independent experts to prepare a report with respect to that question and has received that report.
(2) Without prejudice to the generality of paragraph (a) of subsection (1) above, there shall be taken, for the purposes of that paragraph, to be no reasonable grounds for determining that the work of a woman is of equal value as mentioned in section 1(2)(c) above if -
(a) that work and the work of the man in question have been given different values on a .study such as is mentioned in section 1(5} above; and
(b) there are no reasonable grounds for determining that the evaluation contained in the study was (within the meaning of subsection (3) below) made on a system which discriminates on grounds of sex.
(3) An evaluation contained in a study such as is mentioned in section 1(5) above is made on a system which discriminates on grounds of sex where a difference, or coincidence, between values set by that system on different demands under the same or different headings is not justifiable irrespective of the sex of the person on whom those demands are made".. (4) This is not relevant for this case.
These provisions are somewhat convoluted and begin with that bane of grammarians, a double negative. Industrial tribunals have found difficulty in applying them. At first sight they appear to prevent an industrial tribunal from determining the question of work of equal value unless either there is already in existence a job evaluation study as envisaged by S.l(5) of the Act or they have referred the case to an independent expert. However the general view is that the words in S.2A(2) "without prejudice to the generality of paragraph (a) of sub-section (1) above" preserved the power of an industrial tribunal to dismiss an application outright before considering any kind of job evaluation study or report from an independent expert if it seemed to be a hopeless case which would "never get off the ground". In any case other than the obviously hopeless, industrial tribunals in order to determine-whether there are no reasonable grounds for determining that the work of a woman is of equal value have to investigate in detail a job evaluation study if that is relied on by the employer or it may be to consider whether the report of an independent expert satisfies the tribunal one way or the other, if: there is a sufficiently-arguable dispute over the job evaluation study.
In the appeal before us the question for decision was whether or not the job evaluation study which had been carried out was or was not within the provisions of S.l(5) of the Act, and if it was, whether or not the job evaluation system was discriminatory. It did not involve a reference to an independent expert, but the appellants submit that it should have done. However, because of the extremely sophisticated techniques in modern job evaluation, an industrial tribunal and the Appeal Tribunal have to acquire at least a basic understanding of the jargon and the techniques and to develop a degree of expertise in this field.
The primary facts which lie behind this somewhat complicated and difficult case can be shortly stated. The Respondent employers are motor agents who operate from a number of sites as main agents for Ford motor vehicles. In total they employed about 820 people of whom 189 were women. In the year 1985 and the beginning of the year 1986 there were complaints from a number of women that they were being paid less than men 'for comparable jobs and as a result the employers decided to engage a firm of independent consultants to carry out a job evaluation study. The employers sought the cooperation of the two relevant trade unions namely, Transport and General Workers and ACTSS and invited representation by the Equal Opportunities Commission, In the event, only the representatives of TGWU took any part in the exercise. The firm engaged was Inbucon, well known as management consultants with some 60 years of experience and Mr Crane a senior consultant was instructed to carry out the job evaluation study. Various panels were set up under his direction and guidance and the Industrial Tribunal accepted that he was aware of and warned the panels against the dangers of both conscious sex discrimination and unconscious sex discrimination through traditional or ingrained attitudes on the part of members of the panels.
The Industrial Tribunal had to consider a mass of detailed evidence, both oral and documentary and in the Reasons a great deal of this material is referred to. For the purpose of this judgment it is sufficient to deal with the study in broad outline and to identify the crucial matters found by the Tribunal1 as fact or by way of conclusion. Mr Crane decided that the most suitable method for the job evaluation study was one known as direct consensus. The method involved not only participation by employees but other people likely to be affected by the end result, which therefore included in addition representatives of management, supervision and relevant trades unions. The study embraced the shop floor and manual workers (which the tribunal referred to as NJC workers) on the one hand and clerical and administrative workers on the other hand and the system was explained to employees and management at a meeting attended by over 50 persons. For the purpose of obtaining a consensus there was first set up a panel of relevant representatives which identified the nature and application of the system, which then was developed in detail and operated with two panels.
The job evaluation process then proceeded along the following lines for the understanding of which we are indebted more. to counsel than we are to the Reasons of the Industrial Tribunal.
Step 1 Choose the factors
(a) Skill, training and experience;
(b) responsibility;
(c) external contacts;
(d) physical demands;
(e) mental demands;
(f) working conditions.
These six factors were broken down into sub-headings -
Step 2 Choose the jobs which will represent the range of factors
Two panels were set up, one representing the shop floor workers (NJC workers) arid one representing the administrative workers. The NJC panel numbered 17 but with no women as there were no female workers of this category. The administrative panel consisted of 19 men and women, about one third of whom were managerial staff and more than a third were women members of clerical and administrative staff. Each panel picked 37 jobs in their respective area as representative of the jobs available.
Step 3 Produce job descriptions of these jobs
These described the jobs under the six factor headings.
Step 4 Rank the jobs by paired comparison
Each panel was provided with the job descriptions and then began the process of evaluation, presided over and guided by Mr Crane. Each member of the panel was required to compare two jobs at a time on the basis of the job descriptions and to mark a job with two points if it was considered as a whole more important than the other, and. with one point if it was considered of the same value as the other and nil, if it was considered less important. This is known as "paired comparisons".
Step 5 Select bench-mark jobs
Twenty-three of the 74 jobs were selected as "bench-mark" jobs - 11 from the- NJC side and 12 from the staff side.
Step 6 Make paired comparisons of bench-mark jobs
A joint panel of 13 people was set up, selected from the two original panels. It comprised six women staff, three members of management and four male employees. It subjected the 23 bench mark job's to paired comparisons on a whole job basis in the same way as under 4 above and ranked the bench mark jobs in order.
Step 7 Make paired comparisons using factors
The 13 person panel then performed paired comparisons of each job, the first comparison being the overall position and the next the six factors, as set out at (1) above. The two point and one point scoring was used throughout all seven columns in relation to each paired comparison.
Step 8 Carry out multiple regression analysis
Multiple regression analysis was used to determine the percentage contribution made by each factor. This represented the degree of importance given to the different factors - and was used to draw up a factor plan. The result was 43% attributed to the factor of skill/training/experience; 37% to mental demand; 10% to responsibility, 7% to physical demand and working conditions and 3% to external contacts. A retrospective examination of all the job gradings indicated that the coefficients of agreement amongst the panel in relation to the apportionment of the various factors was remarkably similar and consistent, namely 92% and 92%, for the first factor, 94% and 94% for the second, 93% and 89% for the third, 93% and 93% for the fourth amalgamated with the sixth and 92% and 93% for the fifth factor.
Step 9 Decide on grade boundaries
Step 10 Permit the panel to chancre the ranking of the bench-mark jobs
This was to be done on a "felt fair" basis, that is an order in accordance, with the general level of expectation as to the value of jobs.
Step 11 Slot in the remainder of the original 74 jobs
This was done using the factor plan and the job descriptions prepared under (3) above.
Step 12 Slot in all the remaining jobs
There was an appeal mechanism. After jobs had been graded in this fashion any individual affected by a grading had a right of .appeal against the grading attached to his or her job. Then a full job description of the job, under the factor headings, would be carried out, if there was not one already in existence. The appeal committee consisted of two from management, two from NJC workers who were both men and two from the administrative workers, both of whom were women. The objectors stated their case, then withdrew and the committee adjudicated. Only two of the applicants appealed and had their jobs re-examined.
The general effect of the evidence of Mr Crane was that after a very full and detailed study there was produced a job evaluation in which women had. fully participated and which was almost one hundred percent agreed by everyone on the panels. He had from the outset warned everyone to be' careful to avoid any discrimination and to be careful of discriminatory factors such as a traditional bias towards women in some jobs such as telephone operators and against them in others such as work in a warehouse with handling of heavy articles. In his opinion the study was free from any such bias.
Before the Industrial Tribunal evidence on behalf of the Applicants was given by Mr R F Beddoe a highly qualified expert who, after many years in the academic field set up in practice as a management consultant. His basic criticism of Mr Crane's job evaluation study was that it was not an analytical scheme but depended on drawing up a list of jobs in rank order which was a "felt fair" order, which depends upon the general level of expectation as to the value of jobs. Job ranking and paired comparisons of whole jobs are by their nature "felt fair" rather than analytical. In his opinion any paired comparison scheme is more vulnerable to sex discrimination. When these general matters were put to Mr Crane in cross examination he agreed that there was a risk of sex bias in his method but would not have it that there had in fact been any such bias in the way this study had been carried out. He considered that the 23 "benchmark" jobs had to represent all the skills involved in every context and the choice of representative jobs was the firm base of his system. It was Mr Crane's opinion that any danger of incorporating sex discrimination as a result of traditional demarcation was specifically warned against and closely considered at every stage. In. the Reasons of the Industrial Tribunal the various detailed criticisms by Mr Beddoe of the actual way in which the job evaluation was carried out are set out. We do not think that there is anything to be gained by referring to them in full in this judgment because this appeal has not turned on detail, but has been argued on a broad but fundamental approach. No point has been taken that the Industrial Tribunal failed to have regard to the evidence/ save in one respect to which reference will later have to be made. Mr Beddoe was however compelled to admit in cross examination that the core of any analysis has to be based on consideration of relevant and appropriate factors.
In reaching their conclusions the Industrial Tribunal recognised that under S.2A(2) they had to decide first whether the work of a woman and the work of the man in question had been given different values on a job evaluation study. They did not at this point, but had referred early on in the Reasons to the necessity of compliance with S.l(5) in carrying out the study. They got themselves involved at the beginning of their final conclusion with a semantic consideration of what were reasonable grounds, but did in the end come to clear decisions. Thus they said that the applications must be dismissed because in the light of the job evaluation study there were no reasonable grounds for determining that the work was of equal value. They directed themselves in accordance with S.2A(3) that if there was a difference or coincidence between values set by the system on different grounds under the same or different headings they had to be satisfied that the difference was justifiable. In both contexts they recognised that they had to be sure that the study had not been vitiated or tainted either by direct or indirect sex bias or sex discrimination. In this area .they accepted the case made out for his study by Mr Crane. For the Appellants it was submitted that the job evaluation study undertaken by the employer was not a study as described in S.1(5) of the Act an-j did not meet the test set out in S.2(A)(2)(a). It was argued that the Tribunal did not give proper consideration to the S.1(5) test, never set out the terms of S.1(5), never analysed the requirement of that subsection and made only a passing reference to it at an early stage in the Reasons. All that the Industrial Tribunal said was -
"However in this case the respondents made arrangements for a study to be undertaken of all the respondents' various places of business, such a study being .for the purpose Of S.l(5) of the Equal Pay Act 1970. In these circumstances the consideration of the Tribunal is directed to paragraphs 2 (A)(2) and (3) of the Equal Pay Act"..
It was submitted before us that S.l(5) clearly requires a study of the demand made on a worker under various headings (for instance effort, skill, decision) and this has to be an analytical study. Therefore i.t is not sufficient compliance for a job to be assessed under a "whole job" or "felt fair" scheme which is not analytical. Quite apart from the increased likelihood of a sex discrimination which may not be intended or even contemplated, a "whole job" approach does not analyse each job in terms of the demand. The examination of factor by factor and then relating the total to the whole job as at stages 7 and 8 above is merely an exercise in arithmetic.
It was submitted that nowhere in this process was there en analysis of each job in terms of demand. The factors were given weightings arrived at by the multiple regression technique and which were only the result of working back from the ranking of "bench mark"' jobs which in their turn were ranked not by analysis but a ranking which conformed to the general level of expectation within the organisation and that is a classical "felt fair" order of ranking done on a "whole job basis". The argument then proceeded to the submission that there, should have been proper job descriptions for all the jobs and not merely the 74 and the slotting in of the remainder was done merely as decided by management. Moreover from an examination of the final ranking order it can be seen that at least two of the jobs have been changed from their original ranking apparently by management on a purely arbitrary basis. The Industrial Tribunal made no reference to this. This was said to be a serious failure to take into account a material matter of evidence which should have led to a conclusion that the study was not properly analytical. It was argued that the Industrial Tribunal did not adequately examine the job evaluation study to see whether or not it was one which discriminated in any way at all on the grounds of sex. S.2A2(a) and (b) are linked and an examination of difference of value has to be carried out to see whether or not it is in any way discriminatory in its effect. The other main limb of the Appellents' case concerned the way in which the Industrial Tribunal dealt with the question of reasonable grounds and whether or not it is open for an industrial tribunal to make its own judgment without reference to an independent expert. It is convenient to separate this argument and deal with it at a later stage.
The general effect of a full and careful submission by Miss Debra Holliman for the Respondents was that in the end the decision was one of fact for the Industrial Tribunal. The findings of the Industrial Tribunal could only be interfered with if there was a material misdirection or misinterpretation of the law amounting to error and there was no basis for detecting any error in point of law. As to the main thrust of the Appellants' argument, this was' itself based on a complete misconception of what was required by S.l(5) of the Act. The provision merely refers to a study undertaken with a view to evaluating the demand made on a worker under the various headings. Here the Industrial Tribunal found as a fact that this was precisely what the employers had achieved. Nowhere in the Act is there any suggestion that a study has to be "analytical". The use of the term "analytical" has been introduced by academic and professional text book and pamphlet writers and the Act does not remotely suggest that job evaluation studies which have been adopted for many years are invalid unless they are analytical.
Miss Holliman went into the history and the nature of job evaluation in support of her general criticism of the whole basis of the Appellants' case. She would prefer the description "systematic", for that is what the Act envisages. The real point which emerges from all the discussion in the text books and pamphlets and which is envisaged in both S.l(5) and S.2A is that the study must be free from bias or sex discrimination. Here the Industrial Tribunal was well aware of this and concluded firmly that the study as carried out was in no way discriminatory.
Miss Holliman submitted that the various steps in the study were each a logical step in a well designed system which depended upon examination under headings as S.l(5) requires, of factor by factor and each being set against another and double and treble tested. Furthermore the system of appeal provide an opportunity for a second look and for correction of error.
At the end of the argument and after much discussion, we find ourselves divided. Ms Sue Corby considers that the Tribunal misdirected itself. In its decision it failed to set out the test contained in S.l(5) when quoting what it considered to be the relevant- sections and its reference elsewhere to 8.1(5} was inadequate. Even if the Tribunal were satisfied that the Respondents had made arrangements for a study to be undertaken for. the purpose of S.l(5) it should have gone on to consider'whether the study actually met S.l(5). This Tribunal did not do so.
Secondly the minority member takes the view that there was considerable evidence to show that 5.1(5) was not met. The bench mark jobs were initially ranked on a whole job basis and the factors came later. Also the evidence showed that two members of management (i.e. a minority of the job evaluation panel) then moved at least two benchmarks jobs. As a result, panel beater, which scored 63 points was put in the same grade (grade D) as parts front counter sales which scored 40. This was a higher grade than cost office supervisor, which also scored 63 points and car hire receptionist which scored 45 points. These two were both in grade C.
These glaring discrepancies were not addressed by the Industrial Tribunal nor explained satisfactorily to the Employment Appeal Tribunal. In consequence the failure of the Industrial Tribunal to address these points was both an error in law and perverse.
Furthermore the non-benchmark jobs, which included the jobs of the Applicants and the comparators, were fitted into the rank order by management. There were no job descriptions prepared for these jobs; no analysis according to factors. According to the employers' own evidence they were just slotted in on a "felt fair" basis. This is clearly outwith S.l(5) which specifically requires the woman's job and that of her comparator's, to have been evaluated under various headings. The Industrial Tribunal's failure to take this evidence on board is both an error in law and perverse. Section 1(5) is set out in precise and detailed terms. It is a test that the Tribunal had to consider, yet this Tribunal did not address the text explicitly and furthermore it disregarded considerable, undisputed evidence which pointed to the inescapable conclusion that the test had not been met.
The minority member accepts the submission on behalf of the Respondent employers that Parliament does not require a particular method of job evaluation and indeed there is no absolute method. That is so. The minority member however, does not accept that all job evaluation schemes are as bad (or as good) as each other and an employer can adopt any scheme he wants. Clearly this interpretation is totally inconsistent with S.l(5). It was also 'argued that the Appellants could have used the appeals procedure and that if they had done so, a description of their jobs under factor headings would have gone before the panel. The minority member finds this argument essentially irrelevant. The Tribunal has to consider not whether S.1(5) might be met or could be met but whether it has been met. In this case it had not. Also even if the applicants had made use of the appeals procedure, would their male comparators who are not party to these Tribunal proceedings, have done so? This latter point was not canvassed, by the Respondent but, S.1(5) is concerned with what has happened, not with what might have happened.
The other two members, whilst recognising the force of Ms Corby's opinion, feel constrained to take a different view. We are quite satisfied that the reference at the beginning of the Reasons to the necessity of compliance with the terms of S.l(5) of the Act is sufficient indication that the Industrial Tribunal were fully aware that S.1(5) and S.2A have to be taken together and in tandem. We recognise that this point is really only peripheral and is no more than background to the main argument that neither the study itself nor' the acceptance of the study by the Industrial Tribunal was in compliance with the provisions of S.l(5) of the Act. In this regard we, the majority, are not convinced by the submissions made on behalf of the Appellants although we recognise that Mr Clarke of Counsel put forward a strongly argued case and made several points to which in courtesy we should in due course refer.
The matter on which we feel most strongly and on which we accept Miss Holliman's argument is the proposition that any job evaluation study which is not analytical is thereby invalid because it does not comply with the provisions of S.l(5). In this context we adopt and apply the submission to which we have referred earlier in this judgment. The introduction of the term "analytical" by various experts and we regret to say, by the Equal Opportunities Commission is to put a gloss upon the -words of the provision in question. In our judgment this concentration upon the form and nature of a job evaluation study may lead to a failure to identify the real mischief which the Act is designed to meet. The whole purpose of S.1 and S.2A is to give to a woman a right to an equality clause if compared with a man her work is the same or rated as equivalent, or though dissimilar, is of equal value. A job evaluation study is the means to an end, but most important of all, the job evaluation must be free of discrimination. We find ourselves completely in agreement with the Equal Opportunities Commission in the opinion that systems, which are not analytical, but which may be based on a "felt fair" hierarchy or a paired comparison on a "whole job" basis are much more vulnerable to sex discrimination. We also recognise that the fact that the system which involves the participation of women in the evaluation process does .not necessarily mean that the scheme is non-discriminatory. Women may themselves be unconscious parties to a system which maintains a traditional demarcation which may in fact be discriminatory. It therefore becomes necessary for an industrial tribunal to look very closely and in detail at the job evaluation scheme on which the employer relies for a conclusion that the job of the woman is not of equal value with the dissimilar work of the man with whose job the comparison is made. In the opinion of the majority there are clear findings of fact that the operators of the study were constantly reminding all members of the panels to guard against any inbuilt or traditional discrimination. There was no evidence of any such bias in any part of the evaluation process. We consider also that the Industrial Tribunal was fully aware of the dangers, bore them in mind and found as a fact that there was no discriminatory effect in the process at any stage.
The majority acknowledge the criticism that although the scheme properly, in our opinion, used factors in setting the "benchmark" jobs, there was no use of the factors for the jobs that were slotted in. -Having fixed the "benchmarks" the other jobs were slotted in without full evaluation. Moreover the "benchmark" jobs were changed by two of the panel (both from management). Any such change should have been done by the full membership of the panel.
These matters were certainly blemishes, but in the opinion of the majority were not sufficiently serious or material so as to invalidate the scheme as a whole: nor in our judgment was the failure of the Industrial Tribunal to take these matters into account and indeed to ignore completely the action of the two panel members, sufficient to amount to a material misdirection or a perversity of decision. Moreover the right of appeal provided a further check against irregularity.
We turn next to the argument about misunderstanding of S.2A(2)(b) which is more troublesome. The question which her to be answered is whether the effect of the words "no reasonable grounds" in that subsection is that if there is any ground for alleging discrimination which cannot be dismissed as unarguable the tribunal must refer the case to an independent expert. The Industrial Tribunal got themselves involved at the beginning of their final conclusion with a semantic consideration of what were reasonable grounds.
They said:
"We do not accept that the applicants have only to put forward a reasonable argument which is not foredoomed to failure. If an argument is reasonable, it may still not prevail against other reasonable arguments".
The Appellants submitted that when dealing with the question of reasonable grounds the Industrial Tribunal were not only considering the term in a semantic fashion but' misinterpreted and misapplied the provisions of S.2A, In particular the Tribunal erred in law by assuming that they should adjudicate between opposing arguments whereas all that is required of an applicant is to show that there is a reasonable complaint, that is to say reasonable grounds for determining that the study was discriminatory in its effect. As soon as it appears that the study may be discriminatory then the matter must be referred, to an. independent expert. Matters alleged to be discriminatory were set out in the Grounds of Appeal. These were sufficient to establish reasonable grounds of complaint that the job evaluation study was discriminatory in its effect.
Miss Holliman, for the Respondent submitted that throughout the whole of the study everything which could have been done to produce an evaluation of each job free from sex bias or discrimination was in fact done. On the question of the burden of proof on reasonable grounds it was submitted that the argument for the Appellants must be wrong. There has to be an examination by the Tribunal of whether or not there are reasonable grounds for alleging that the work is of equal value and whether or not there has been discrimination. If every case that raises the possibility of discrimination has to be referred to an independent expert the question of whether or not there has been discrimination may never be determined. The independent expert is an expert on job evaluation and not on sex discrimination which is something for the Industrial Tribunal to decide.
The minority member accepts the Appellants' construction of S.2A(2)(b). The Applicant is required only to show there are one or more reasonable grounds for determining that the evaluation was made on-a system which discriminated on grounds of sex. This would seem to be the proper, and common sense construction of the test, which was designed as a preliminary hurdle - to weed out the hopeless cases. In this case the applicant succeeded in showing such grounds by producing evidence of anomalies and discrepancies in the grading.
The majority feel constrained to take a different view.
The Industrial Tribunal took the view that this sub-section requires an industrial tribunal to make an assessment of whether or not there were no reasonable grounds for alleging work of equal value and no reasonable grounds for alleging discrimination on the grounds of sex in the design and execution of the job evaluation study. Everyone seems to agree that it is open to an industrial' tribunal to weed out the hopeless case. The difference of opinion is whether or not an industrial tribunal can go on to determine the case once the case gets beyond the hopeless. In other words, if there may be some defect in the evaluation study or if there may be some element of discrimination in the nature of or execution of the evaluation study, is the tribunal prevented from making its own value judgment of the study and be compelled to refer the matter to an independent expert. The problems which have bedeviled industrial tribunals and the Appeal Tribunal were considered by the Employment Appeal Tribunal in the case of Forex Neptune (Overseas) Ltd v Miller and Others (1987) ICR 170. In that case authoritative guidance was respectfully requested from the Court of Appeal which may yet come to hand. In the .instant case' all three of us are in agreement that some guidance would be welcomed. The majority take the view that this Industrial Tribunal was entitled to adopt the course which it did. .We do not believe that Parliament intended that an- industrial tribunal can never in the first place be allowed to make its own decision on whether or not the alleged grounds are reasonable once a case gets beyond the hopeless and has got to refer every case to an independent expert. We agree with Miss Holliman that the independent expert is not likely to be qualified to deal with the element of discrimination which is always a question of fact for an industrial tribunal to decide. It seems to the two of us to be contrary to common sense to obtain the opinion of an expert on the nature and quality of a job evaluation study and then for an industrial tribunal to adjudicate on both aspects as they must and it may well be, reject the opinion of the expert, as we are reliably informed, some industrial tribunals have been constrained to do.' The majority considers this is not merely a two bites at the cherry situation but a three bites one and does not make sense to us. In the event the majority members of this Appeal Tribunal are firmly of the opinion that there were no errors in point of law and no perversity of decision on the part of the Industrial Tribunal in this case. As the majority decision prevails the appeal must be dismissed.
Appeal dismissed.
Leave to appeal.
