MIDDLESBROUGH BOROUGH COUNCIL v MS M SURTEES & OTHERS [2007] UKEAT 0077_07_1707

EMPLOYMENT APPEAL TRIBUNAL
BAILII case number: [2007] UKEAT 0077_07_1707Case No Appeal No. UKEAT/0077/07Venue 58 VICTORIA EMBANKMENT, LONDON EC4Y 0DS
Between
MIDDLESBROUGH BOROUGH COUNCILAPPELLANTMS M SURTEES & OTHERSRESPONDENT
Before
Mr Justice EliasDate 17 July 2007
[1]APPEARANCES For the Appellant MR CHRISTOPHER JEANS (One of Her Majesty's Counsel) and MS JANE WOODWARK of Counsel Instructed by: Middlesbrough Borough Council Legal Services PO Box 99A Municipal Buildings MIDDLESBROUGH TS1 2QQ For the Respondent MR ROBIN ALLEN (One of Her Majesty's Counsel) and MS CLAIRE McCANN of Counsel Instructed by: Messrs Stefan Cross Solicitors Buddle House Buddle Road NEWCASTLE-UPON-TYNE NE4 8AW SUMMARY EQUAL PAY ACT Material factor defence European law Certain employees of the Council claimed equal pay with respect to their chosen comparators. In some cases the claim related to a period before a new job evaluation scheme had been introduced. As part of the implementation of the job evaluation scheme, the employers introduced a protected pay arrangement designed to cushion those in receipt of higher pay when the new scheme was introduced from suffering an immediate and significant drop in pay as a result of the operation of the scheme. The claimants submitted that they should also receive the benefits of the protected pay arrangements, even though they were not in receipt of the higher pay when the scheme was introduced, because they would have received it had the equality clause been applied to them at the right time. The Employment Tribunal acceded to that argument and rejected the Council's justification defence. The EAT upheld the Council's appeal and found justification established. There were two further categories of claim in which the Council contended that the difference in pay was caused by a non-sex tainted collective bargaining. The Employment Tribunal rejected that claim, and the EAT rejected the appeal. Observations on the circumstances in which prima facie indirect discrimination arises, so as to trigger the need to establish objective justification. THE HONOURABLE MR JUSTICE ELIAS (PRESIDENT) This is another appeal from the decision of an Employment Tribunal sitting in the North East of England dealing with an aspect of the Equal Pay litigation which has been launched against local authorities and health authorities in that region. The general background[2]We first sketch out the general context in which these claims arise. We deal with the specific facts material to each issue when we address those claims.[3]For many decades there were different bargaining arrangements carried out nationally for three distinct groups of local authority employees. Manual workers were governed by a national collective agreement known as "the White Book". Administrative, professional, technical and clerical workers ("APTC") had terms and conditions governed by a national agreement known as "the Purple Book". Craft workers were governed by an agreement known as "the Red Book". Different unions were involved in the respective national negotiating bodies.[4]As Mr Jeans QC, counsel for the Council, pointed out in the course of his submissions, the existence of the separate negotiations inevitably meant that different terms and conditions existed as between the separate groups. There were variations in the length of the working week; the amount of pay; the way pay was made up; the intervals between which payment was made; the level of pension contributions; and other matters. Taking one particular feature of some relevance in this case, the White Book envisaged bonus schemes, the Red Book provided for minimum bonuses, and the Purple Book did not allow for bonus schemes at all.[5]White Book employees became subject to a job evaluation scheme in 1987. This was relatively comprehensive, unlike the other two categories which had no significant job evaluation schemes in place at all at that time.[6]In 1997 a collective agreement was reached nationally which began the process of unifying the terms and governing structures of the manual workers and the APTC workers. This was known as the "single status" agreement and, following another colour transformation, the document in which the agreement was identified is known as "the Green Book". This assigned previous APTC and manual jobs to a single pay spine. Certain key national provisions were to apply to the relevant local authority employees generally, whilst other national provisions could be modified by local negotiation.[7]An important provision in the Agreement provided for a new job evaluation scheme to be carried out embracing all manual and APTC jobs. These schemes were to be implemented locally under procedures involving representative panels of unions and employers. Pending this, however, the Green Book expressly provided that various provisions of the White and Purple Books should remain in force until superseded by local reviews. In particular, it was specifically provided that:
"existing local bonus schemes/arrangements are not affected by this agreement. The NJC has agreed however to establish a joint technical working group to examine bonus and associated issues with particular reference to equal pay."
[8]We were told that the working group was established and identified certain potential legal risks, arising from the operation of equal pay law, in the existing arrangements.[9]After much negotiation and various delays, a job evaluation scheme in Middlesbrough covering virtually all jobs was established and came into effect on 1 April 2005. A vitally important aspect of these negotiations concerned the provision of protected pay. Inevitably, where job evaluation schemes are negotiated there will be some whose pay increases as a consequence of the re-grading and some whose pay is reduced.[10]It is accepted throughout industry that in general it is legitimate to seek to protect, so far as is possible and proportionate, the interests of those who lose out following the introduction of a new job evaluation scheme so that they do not suffer significant and immediate reductions in pay. In many cases the hope and expectation is that, pending equality being achieved, their pay will remain static without actually having to be reduced, on the basis that the others will gradually catch up as a result of annual pay increases not awarded to the higher paid group. However, it is not always possible to achieve that result.[11]In this case the negotiations about protection were conducted under the guidance of ACAS. In contrast to many schemes of this kind, the protection here was relatively modest. There was full protection for the first year, but it was reduced to 75% of the difference between old and new pay in the second year, falling to 50% in the third year. In the fourth year the protection would only apply to a loss of pay exceeding £2,000 per annum, and thereafter there was no protection at all. That agreement was finally reached in February 2005, less than two months before the job evaluation scheme took effect. Letters were sent out to individual employees informing them of how these provisions would affect them.[12]At the time these arrangements were made, there were already a number of equal pay claims on foot. These all related to the pre-April 2005 terms. None of these claims had been conceded or resolved, however, before the new job evaluation scheme took effect. Some of the claims have since succeeded whilst others have failed. The law .[13]We start by setting out the relevant statutory background, and a brief summary of the legal principles relating to equal pay.[14]The starting point is Article 141 on the Treaty Establishing the European Community (formerly Article 119). This is as follows:
"1. Each Member State shall ensure that the principle of equal pay for male and female workers for equal work or work of equal value is applied."
The implications of that principle are then set out in the remainder of Article 141, read with the Equal Pay Directive , ( Council Directive 75/117 ) whose purpose is to implement and supplement Article 141.[15]The implementation of Article 141 into domestic law is achieved by the Equal Pay Act 1970 . Section 1, so far as is material, is as follows:
"(1) If the terms of a contract under which a woman is employed at an establishment in Great Britain do not include (directly or by reference to a collective agreement or otherwise) an equality clause they shall be deemed to include one."
[16]Section 1(2) provides that the equality clause applies whenever the claimant and her chosen comparator are employed on like work, which is when work is of the same or a broadly similar nature; or on work that has been rated as equivalent under a non-discriminatory job evaluation scheme carried out by the employer; or on work which has been rated as of equal value in terms of the demands made on her work. We shall use the term 'equal work' to encompass these three sets of circumstances.[17]The effect of the equality clause is to ensure that the contractual terms under which the woman is employed are no less favourable than those under which the man is employed.[18]Section 1(3) is highly material to this case. It provides a defence to an equal pay claim where the employer can show – the onus being on him – that there is a genuine non-sex reason for the pay difference: "An equality clause shall not operate in relation to a variation between the woman's contract and the man's contract if the employer proves that the variation is genuinely due to a material factor which is not the difference of sex and that factor—(a) in the case of an equality clause falling within subsection (2)(a) or (b) above, must be a material difference between the woman's case and the man's; and(b) in the case of an equality clause falling within subsection (2)(c) above, may be such a material difference." (The slightly different formulations in (a) and (b) are historical accidents arising from the gradual way in which European law was implemented into domestic legislation. The differences are not material. The employer will in all cases have to prove that the difference in pay is genuinely due to a material factor other than sex ("the GMF defence").)[19]There are certain differences between the scope of Article 141 and the Equal Pay Act . The principal one is that Article 141 relates only to pay although that concept has been very widely construed by the European Court and will include both contractual and non-contractual aspects of remuneration. By contrast, the domestic legislation is not limited to pay but applies to all contractual terms and conditions of employment. Non-contractual aspects have to be dealt with under the Sex Discrimination Act 1975 rather than the Equal Pay Act as a matter of domestic law.[20]There are certain undisputed legal principles which provide the context in which this appeal must be determined:(1) Article 141 can be directly relied upon in the courts of the United Kingdom by a woman wishing to claim equal pay with an appropriate male comparator. This was established by the ECJ in the seminal case of Defrenne v Sabena [1976] ICR 547 .(2) It follows that the duty of the court is to interpret the relevant provisions of the Equal Pay Act , which is the domestic implementation of Article 141, fully in compliance with that Article. Any inconsistency must be resolved in favour of community law. Lord Nicholls of Birkenhead expressed the position as follows in Autologic plc v IRC [2006] 1AC 118 at paras 16-17:
"Where such an inconsistency exists the statutory provision is to be read and take effect as though the statute had enacted that the offending provision was to be without prejudice to the directly enforceable community rights of persons having the benefits of such rights. That is the effect of section 2 of the European Communities Act [1972] …. 17 ….Accordingly, if an inconsistency with directly enforceable community law exists, formal statutory requirements must where necessary be disapplied or moulded to the extent needed to enable those requirements to be applied in a manner consistent with community law." (3) Article 141 forms part of the social objectives of the community. As the European Court of Justice expressed it, in Deutsche Telecom AG v Lille Schröeder [Case C-50/96): "the right not to be discriminated against on grounds of sex is one of the fundamental human rights, whose observance the court has a duty to ensure". (4) The obligation to construe the domestic law in accordance with the community law is of particular relevance when analysing the scope of s1(3). The reason is that the scope of discrimination in European law extends beyond direct discrimination, i.e. the case where the woman is paid less because she is a woman, and embraces also indirect discrimination. That is not an entirely straightforward concept, not least because there are different definitions of indirect discrimination, even in community law itself. (See for example the different definitions in the Burden of Proof Directive and the Sex Discrimination Directive . We discuss later in this judgment the various ways in which indirect discrimination can arise.) The essential point, however, for current purposes is that indirect discrimination occurs where a pay practice in some way has a disparate impact on women or subjects them as a group to a particular disadvantage when compared with men. That is generally, but not necessarily, established by a statistical analysis of the relevant pools. The fundamental difference between direct and indirect discrimination is that (save arguably in very exceptional cases) direct discrimination can never be justified, whereas indirect discrimination can. An employer whose pay practices involve paying women less than men because of their sex is necessarily acting unlawfully; one whose practices have the effect of adversely impacting on women may be able to demonstrate that there is an objective justification for this difference in pay. (5) The test which the courts must apply when prima facie indirect discrimination exists and the employer has to show objective justification is now well established. The classic definition was given by the European Court of Justice in Bilka-Kaufhaus Gmbh v Weber von Hartz [1987] ICR 110 and has been regularly repeated since. In the very recent decision of Cadman v Health and Safety Executive [2006] ICR 1623 , the European Court summarised it thus ... para 33 ): "
The justification given must be based on a legitimate objective. The means chosen to achieve that objective must be appropriate and necessary for that purpose". This involves the application of a principle of proportionality, as Lord Nicholls recognised in Barry v Midland Bank [1999] ICR 859 , 870E. Lord Nicholls described how he understood the principle would operate: "In other words, the ground relied upon as justification must be of sufficient importance for a national court to regard this as overriding the disparate impact of the difference in treatment, either in whole or in part. The more serious the disparate impact on women, or men as the case may be, the more cogent must be the objective justification. There seems to be no particular criteria to which the national court should have regard when assessing the weight of the justification relied upon." (6) In Rainey v Greater Glasgow Health Board [1987] 1 AC 224, the House of Lords considered the scope of the GMF defence in s1(3) of the Act where prima facie indirect discrimination exists and objective justification needs to be established. It adopted the approach of the European Court in Bilka . Lord Keith, giving the only reasoned speech in their Lordship's House, confirmed that there was no difference between European and domestic law on this matter. The structured analysis of equal pay cases.[21]The scheme of the Equal Pay Act was expounded by Lord Nicholls in Glasgow City Council v Marshall [2000] UKHL 5 ; [2000] ICR 196 at 202 as follows:
"The scheme of the Act is that a rebuttable presumption of sex discrimination arises once the gender-based comparison shows that a woman, doing like work or work rated as equivalent or work of equal value to that of a man, is being paid or treated less favourably than the man. The variation between her contract and the man's contract is presumed to be due to the difference of sex. The burden passes to the employer to show that the explanation for the variation is not tainted with sex. In order to discharge this burden the employer must satisfy the tribunal on several matters. First, that the proffered explanation, or reason, is genuine, and not a sham or pretence. Second, that the less favourable treatment is due to this reason. The factor relied upon must be the cause of the disparity. In this regard, and in this sense, the factor must be a 'material' factor, that is, a significant and relevant factor. Third, that the reason is not 'the difference of sex'. This phrase is apt to embrace any form of sex discrimination, whether direct or indirect. Fourth, that the factor relied upon is or, in a case within section 1(2)(c), may be a 'material' difference, that is, a significant and relevant difference, between the woman's case and the man's case. When section 1 is thus analysed, it is apparent that an employer who satisfies the third of these requirements is under no obligation to prove a 'good' reason for the pay disparity. In order to fulfil the third requirement he must prove the absence of sex discrimination, direct or indirect. If there is any evidence of sex discrimination, such as evidence that the difference in pay has a disparately adverse impact on women, the employer will be called upon to satisfy the tribunal that the difference in pay is objectively justifiable . But if the employer proves the absence of sex discrimination he is not obliged to justify the pay disparity."
(emphasis added.)[22]In short, once the work is established to be of equal value, the employer must satisfy the tribunal that any difference in pay is genuinely due to a material factor other than sex. Absent any indirect discrimination, that is all that he has to establish (subject to a potential qualification referred to in the next paragraph). However, where there is prima facie indirect discrimination, then objective justification is required. By "prima facie indirect discrimination" we mean the situation which triggers the obligation objectively to justify the difference. When that point is reached is one of the issues in this case.[23]We should note that there is an argument, which has not in the event been relied upon in this case, that as a consequence of a recent decision of the European Court of Justice in Brunhoffer v Bank der Osterreichischen Postsparkasse AG [2001] IRLR 571 , the approach adopted in Marshall is no longer good law. It is said that the effect of Brunhoffer is that Article 141 requires an employer objectively to justify a difference in pay even where there is no hint of sex discrimination. In other words, the mere fact that a woman is making a comparison with a man whose job is of equal value itself automatically requires that any difference in pay is objectively justified, even though the employer shows that the difference has nothing to do with sex If the argument is correct, equal pay has broken loose from its moorings in discrimination law. The EAT analysed that argument very fully in the case of Villalba v Merrill Lynch [2006] IRLR 437 and decisively rejected it. Mr Allen has not sought to challenge that case before us but reserves the right to do so if necessary in a higher court. When does prima facie discrimination arise?[24]An issue which has implications in this case is the question when prima facie indirect discrimination arises. If the pay arrangements have a sufficiently disparate adverse impact on women, is that sufficient to give rise to prima facie indirect discrimination, so that they will be unlawful unless objectively justified? Or can the employer avoid the need objectively to justify if he can show that, even although the arrangements have that disparate impact, they are not in any way tainted by sex?[25]To put the point another way, does the fact that there is adverse disparate impact, at least if it is sufficiently striking, of itself establish an irrebuttable presumption of prima facie sex discrimination requiring objective justification? Or can the employer rebut the presumption by satisfying the tribunal that notwithstanding the adverse impact, in fact the variation in pay has nothing to do with sex at all?[26]In answering that question, there has been some focus on how the words we have italicised in Lord Nicholls' judgment (para.21 above) are to be construed. Was Lord Nicholls intending to say that that prima facie indirect discrimination arises whenever there is evidence that the arrangements have an adverse impact on women; or was he only intending to provide an example of the kind of case where prima facie indirect discrimination will frequently, but not inevitably, arise?[27]Mr Jeans submits that both principle and authority require that it should be the latter. As to authority, he notes that in the last sentence of the quote from Lord Nicholls reproduced above, his Lordship expressly says that the question of objective justification does not arise if the employer establishes no sex discrimination. He further relies upon the decision of the Court of Appeal in Armstrong v Newcastle Upon Tyne NHS Hospital Trust [2006] IRLR 124 . In that case the Court analysed the speech of Lord Nicholls in Marshall . Lady Justice Arden formulated the following principles after referring to the same passage of Lord Nicholls' judgment as we have identified:
"32 The genuine material factor issue I have set out in paragraph 17 above the well-known passage from the speech of Lord Nicholls in the Marshall case. (The remainder of the House agreed with his speech.) That passage sets out a step by step guide to proving a genuine material factor defence. For the purposes of this appeal, the steps can be summarised as follows: (1) the complainant must produce a gender-based comparison showing that women doing like work, or work rated as equivalent or work of equal value to that of men, are being paid or treated less favourably than men. If the complainant can produce a gender-based comparison of this kind, a rebuttable presumption of sex discrimination arises. (2) the employer must then show that the variation between the woman's contract and the man's contract is not tainted with sex, that is, that it is genuinely due to a material factor which is not the difference of sex. To do this, the employer must show each of the following matters: (a) that the explanation for the variation is genuine, (b) that the more favourable treatment of the man is due to that reason, and (c) that the reason is not the difference of sex. (3) if, but only if, the employer cannot show that the reason was not due to the difference of sex, he must show objective justification for the disparity between the woman's contract and the man's contract. 33 It follows from the Marshall case that there is no need for an employer to provide justification for a disparity unless the disparity is due to sex discrimination. Miss Tether does not submit any different principle applies by virtue of Article 141. 34 In the Marshall case, Lord Nicholls uses the words 'disparately adverse effect'. He held that evidence that a difference in pay had such an effect on women could be evidence of sex discrimination. He did not, however, hold that the mere fact that there was a disparately adverse effect was itself sex discrimination. In the Marshall case, Lord Nicholls used the phrase 'disparately adverse effect' to denote the trigger at which the rebuttable presumption of sex discrimination would arise under step 1 mentioned above."
[28]We pause to note that there are two qualifications or corrections which, at least on one reading of these paragraphs, would have to be made to this analysis - as we believe Mr Jeans accepted - although neither is material in this case. The first is that if the disparate treatment is the result of direct discrimination on grounds of sex, and not merely indirect, then there is no scope for justification under s1(3).[29]The second is that establishing a "disparately adverse impact", or showing that women are paid less favourably than men, is not the trigger at which the rebuttable presumption of discrimination arises. That plainly arises immediately a woman shows that she is doing equal work to a man but is paid less, as the first sentence from Lord Nicholls' judgment in the extract reproduced above makes clear. The concept of adverse impact is only of any relevance in indirect discrimination claims. (In fairness, it may be that by stating that this analysis was "for the purposes of this appeal", Arden LJ was merely intending her analysis to be applied to indirect discrimination claims, so that the references to a rebuttable presumption of sex discrimination are intended merely to refer to a rebuttable presumption of indirect sex discrimination).[30]Lord Justice Buxton in terms accepted this analysis. He said this (para 110):
"As Lord Nicholls said at the end of the passage from Glasgow City Council v Marshall cited in paragraph 101 above, if the employer proves the absence of sex discrimination he is not obliged to justify the pay disparity. That is the basis of the step by step approach explained by Arden LJ in her paragraph 32 above. Once disparate adverse impact has been established, the burden passes to the employer in respect of two issues. First, that the difference between the man's and the woman's contract is not discriminatory, in the sense of being attributable to a difference of gender. Second, if the employer cannot show that the difference in treatment was not attributable to a difference of gender he must then demonstrate that there was nonetheless an objective justification for the difference between the woman's and the man's contract."
[31]Again, Buxton LJ could not have been intending to deal with direct discrimination and his observations about when the burden shifts should be limited to an analysis of indirect discrimination only. Furthermore, the reference to the difference in pay being "attributable to gender" must, we think, be a reference not to direct sex discrimination, but to the difference being "sex tainted." However, Buxton LJ plainly understands Lord Nicholls to be saying that the obligation to establish objective justification is not established by the mere fact of disparate impact; it only arises where there is both disparate impact and the employer has failed to discharge the burden of showing that the difference has nothing to do with sex.[32]Mr Jeans submits that this is wholly in accordance with principle. An employer should not be held to have discriminated on grounds of sex - which is what the Equal Pay legislation is all about - if he has fully satisfied a tribunal that he has not done so. That is what the House of Lords has said in Marshall and again in Strathclyde Regional Council v Wallace [1998] ICR 205 . The mere fact that the pay arrangements have a disparate impact on women will justify a court or tribunal scrutinising with great care the employer's explanation for the difference in treatment to ensure that it is not sex tainted. But if the employer can discharge that burden, that is the end of the matter. No question of objective justification then arises.[33]Mr Jeans further submits that in any event the EAT is bound by this interpretation of Lord Nicholls' judgment given in Armstrong and must apply it. That was the view of another division of the EAT (Wilkie J presiding) in South Tyneside Metropolitan Borough Council v Anderson and others UKEAT/0684/05 ZT when it chose to follow the approach in Armstrong in preference to some inconsistent and obiter observations of mine in Villalba v Merrill Lynch [2006] IRLR 437 , para 113.[34]In Villalba I suggested that the approach in Armstrong appeared to be inconsistent with the decision of the European Court of Justice in Enderby v Frenchay Health Authority [1994] ICR 112 and that at least where the disparate impact was of a nature identified in that case, that would establish an irrebuttable presumption of prima facie indirect discrimination which would need to be justified to be lawful. Enderby was not expressly considered in the Armstrong case.[35]Furthermore, Mr Jeans submits that even if the EAT considers that the decision was reached per incuriam , because of a failure to consider Enderby , it is not for the EAT to adopt that analysis. Only the Court of Appeal can do so, and even then only in very limited circumstances: see Duke v Reliance Systems Ltd [1988] 1 QB 108 (CA).[36]The claimants say that Mr Jeans' argument is wrong and inconsistent with established principle. The fact that the pay arrangements impact adversely on women will, at least in a sufficiently clear case, of itself constitute prima facie indirect discrimination requiring justification. The employer cannot escape liability even if he can show that the difference in pay is not sex tainted. The statistics, if sufficiently powerful, establish a sufficiently strong likelihood of sex discrimination.[37]Mr Allen QC, counsel for the claimants, says that the analysis in Armstrong is wrong, plainly misrepresents what Lord Nicholls said in Marshall (even though the judges purported to be following it), and in any event is inconsistent with clearly established principles of European law laid down in Enderby , not referred to by the Court of Appeal, and other cases. He relies on the reservations about the decision expressed by the EAT in the Villalba case and submits that they accurately encapsulate European law.[38]In Enderby the claimant was a speech therapist who compared herself with certain clinical psychologists and pharmacists. She claimed that at her level of seniority she was paid significantly less than her comparators. Her profession was overwhelmingly female, whilst clinical psychologists were predominantly male and pharmacists relatively equal as between men and women. The National Health Service, the employer of all the staff, relied upon the fact that the pay had been fixed by different collective bargaining mechanisms, and that the industrial tribunal had found as a fact that there was no discrimination within the bargaining units, i.e. no evidence that in fixing pay within each group, the bargainers had been discriminating either directly or indirectly. In these circumstances, the employers contended that the employers had discharged the burden of proving that the variation in pay had nothing to do with sex.[39]The European Court of Justice disagreed. Its explanation was as follows:
"13. It is normally for the person alleging facts in support of a claim to adduce proof of such facts. Thus, in principle, the burden of proving the existence of sex discrimination as to pay lies with the worker who, believing himself to be the victim of such discrimination, brings legal proceedings against his employer with a view to removing the discrimination. 14. However it is clear from the case law of the court that the onus may shift when that is necessary to avoid depriving workers who appear to be the victims of discrimination of any effective means of enforcing the principle of equal pay. Accordingly, when a measure distinguishing between employees on the basis of their hours of work has in practice an adverse impact on substantially more members of one or other sex, that measure must be regarded as contrary to the objective pursued by Article 119 of the Treaty, unless the employer shows that it is based on objectively justified factors unrelated to any discrimination on grounds of sex (judgments in case 170/84 Bilka-Kaufhaus [1986] IRLR 317 , at paragraph 31; case C-33/89 Kowalska [1990] IRLR 447 , at paragraph 16; and case C184/89 Nimz [1991] IRLR 222, at paragraph 15). Similarly, where an undertaking applies a system of pay which is wholly lacking in transparency, it is for the employer to prove that his practice in the matter of wages is not discriminatory, if a female worker establishes, in relation to a relatively large number of employees, that the average pay for women is less than that for men (judgment in case 109/88 Danfoss [1989] IRLR 532, at paragraph 16). 15. In this case, as both the FHA and the United Kingdom observe, the circumstances are not exactly the same as in the cases just mentioned. First, it is not a question of de facto discrimination arising from a particular sort of arrangement such as may apply, for example, in the case of part-time workers. Secondly, there can be no complaint that the employer has applied a system of pay wholly lacking in transparency since the rates of pay of NHS speech therapists and pharmacists are decided by regular collective bargaining processes in which there is no evidence of discrimination as regards either of those two professions. 16. However, if the pay of speech therapists is significantly lower than that of pharmacists and if the former are almost exclusively women while the latter are predominantly men, there is a prima facie case of sex discrimination, at least where the two jobs in question are of equal value and the statistics describing that situation are valid. 17. It is for the national court to assess whether it may take into account those statistics, that is to say, whether they cover enough individuals, whether they illustrate purely fortuitous or short-term phenomena, and whether, in general, they appear to be significant. 18. Where there is a prima facie case of discrimination, it is for the employer to show that there are objective reasons for the difference in pay. Workers would be unable to enforce the principle of equal pay before national courts if evidence of a prima facie case of discrimination did not shift to the employer the onus of showing that the pay differential is not in fact discriminatory (see, by analogy, the judgment in Danfoss , cited above, at paragraph 13). 19. In these circumstances, the answer to the first question is that, where significant statistics disclose an appreciable difference in pay between two jobs of equal value, one of which is carried out almost exclusively by women and the other predominantly by men, Article 119 of the Treaty requires the employer to show that that difference is based on objectively justified factors unrelated to any discrimination on grounds of sex."
[40]Mr Allen submits that it is plain from this decision, and in particular paragraph 16, that at least where the statistics are of a kind therein described, there is indeed an irrebuttable presumption of prima facie sex discrimination; it is assumed that the pay practice is sex tainted and the employer must show that the variation in pay is objectively justified. In Enderby the employers demonstrated that the processes of collective bargaining were not sex tainted and yet prima facie discrimination i.e. unlawful discrimination in the absence of objective justification, was held to arise from the statistics alone.[41]Mr Allen contends that this is precisely what Lord Nicholls is likely to have had in mind in the controversial passage in his speech in Marshall . (Moreover, it is to be noted that whilst the Court in Enderby itself inferred prima facie discrimination only from the particularly striking statistics demonstrating adverse impact in that case, namely the fact that the groups were predominantly male and the overwhelmingly female respectively, the Court of Appeal has subsequently held that the statistical evidence required to establish discrimination in an Enderby situation is no different to that required to establish it in a case where the disparate impact is the result of the criterion chosen by the employer to determine pay: Bailey v Home Office [2005] EWCA Civ 327 ; [2005] ICR 1057 .[42]Mr Allen also relies on the following passage of the ECJ in the Cadman case:
"It is apparent from settled case law that Article 141 EC, like its predecessor Article 119 of the EEC Treaty (which became Article 119 of the EC Treaty – Articles 117 to 120 of the EC Treaty have been replaced by Articles 136 EC to 143 EC), must be interpreted as meaning that whenever there is evidence of discrimination, it is for the employer to prove that the practice at issue is justified by objective factors unrelated to any discrimination based on sex (see, to that effect, inter alia, Danfoss , paragraphs 22 and 23; case C-33/89 Kowalska [1990] IRLR 447, paragraph 16; Hill and Stapleton , paragraph 43, and joined cases C-4/02 and C-5/02 Schonheit and Becher [2003] IRLR 983, paragraph 71)…."
[43]Mr Jeans submits that the crucial consideration which Enderby makes clear is that the statistics must appear to be "significant". He contends that once the employer has demonstrated that the reason for the variation in pay is wholly free from any sex taint, then the statistics are meaningless and not significant. They can be of no assistance in inferring discrimination which has already been disproved. Whatever the impact of the pay arrangements, they tell you nothing about potential sex discrimination. The essential question when considering a GMF defence is whether the reason for the variation in pay is sex tainted, and if the employer demonstrates that it is not, then the fact that the pay arrangements have a disparate impact is of no relevance.[44]As to Cadman , he submits that it is consistent with his submissions. The ECJ is saying that there is a need to justify only where there is evidence of discrimination. But where the employer has rebutted such discrimination, there is no need to justify. Conclusion and Discussion.[45]We have no doubt that we are bound by the decision in Armstrong . It is not for the Employment Appeal Tribunal to act on the assumption that the Court of Appeal has misunderstood European law. Nor, in a case where that Court has sought to explain the effect of a decision of the House of Lords, at least in circumstances where that is part of the essential reasoning of the Court, is it open to the Employment Appeal Tribunal to say that the Court of Appeal has misinterpreted it and to adopt its own preferred construction of their Lordships' speeches. Whether right or wrong, the principle of precedent requires a loyal adherence to what is clearly the ratio of the decision of the Court of Appeal.[46]In any event, after a careful consideration of the relevant arguments, we consider that in principle the analysis of indirect discrimination adopted by the Court of Appeal in Armstrong is correct. The EAT went too far in Villalba in saying that where the arrangements have a sufficiently strong disparate impact there is always an irrebuttable presumption of prima facie indirect sex discrimination, although in many cases - and particularly situations akin to Enderby itself - the presumption will be extremely difficult to rebut in practice. But logically it ought in principle to be open to an employer to show that even although there is disparate adverse impact, that it is not in any way related to any act of the employer which is sex tainted, and thereby avoid the need to establish justification.[47]It is true that Enderby does on its face appear to be a case where statistics themselves established a prima facie case of sex discrimination. However, it is necessary to look at that case in context. The non-discriminatory explanation for the difference in pay was the fact that the independent processes of collective bargaining had led to different terms and conditions. Neither process taken independently identified discrimination. But that would not meet the possibility that with regard to each of the bargaining arrangements there were stereotypical assumptions as to the appropriate pay for what were historically perceived to be male and female jobs. Indeed, para 39 of the opinion of the Advocate General supports this analysis:
"The structure of the reasoning for both direct discrimination and indirect discrimination is comparable with regard to the evidentiary aspect of the proceedings in as much as a rebuttable presumption of discrimination can be raised, in one case by means of a specific comparison and, in the other, by a comparison of groups, which places the onus on the employer to adduce evidence in rebuttal of that presumption or to produce a justification."
This presupposes that the statistics may be rebutted by proof that there is no discrimination, although the Advocate General did not consider that this was achieved merely be showing that the bargaining processes, taken separately, did not disclose discrimination.[48]The purpose of the equal pay legislation - both domestic and European - is to eliminate discrimination on grounds of sex. It is not to correct pay differentials for other reasons. If the employer adduces evidence which satisfies the tribunal that the variation in pay is not for sex reasons, that objective has been satisfied and there is no reason in policy or logic for requiring a non-sex based distinction to be objectively justified. The implications of this analysis .[49]We accept, therefore, Mr Jeans' submission that proof of a non-sex based reason will be a complete answer to any discrimination claim, direct or indirect. At the same time, it is important to bear in mind the purpose of the legislation and in particular the fact that there are structural reasons causing unequal pay. There has historically been much stereotyping of jobs with assumptions being made both about what work is suitable for men and women and what pay is appropriate for these jobs. This has led to much de facto job segregation (which is not to suggest that this is a deliberate or intended policy of employers, or that they have in any way formally limited women's access to the predominantly male jobs.)[50]That history continues to leave its mark on pay structures. Tribunals must be alive to the very real possibility that where there is adverse impact, identified where necessary by sufficiently cogent statistics, that may be the result of factors which are sex tainted. We agree with the observation of Cox J in the EAT in Ministry of Defence v Armstrong [2004] IRLR 672 , para.42, that tribunals should not apply a formulaic approach to issues of sex discrimination; what matters is whether the tribunal is satisfied in any particular case that the evidence discloses a pay difference which is related to the difference of sex.[51]It may be helpful to consider what we perceive to be the practical significance of this analysis, assuming it to be correct. First, in all cases of what might be termed "classic" examples of indirect discrimination the very criterion which the employer chooses to differentiate pay scales will, because of the position of women in society, itself impact adversely on women rather than men (or vice versa). A traditional example is where full timers are paid more than part timers as in the Bilka case. The criterion itself - the distinction between full and part time - is the very factor that causes the disparate impact. In such cases the pay arrangements are inevitably tainted by sex - it is the direct consequence of the employer's pay criterion - and plainly the obligation to justify arises. That was not of course the Enderby situation, as the Court of Justice in terms recognised.[52]Second, in many cases the women will be claiming that the pay arrangements adversely impact upon them as a group, but there is no obvious feature which causes the differentiation. It is not possible to point to a specific factor which would be likely to cause the women to fall into the lower paid group. However where disadvantage, typically gleaned from the statistics, is sufficiently striking, it may be justified to draw the inference that the difference in pay reflects traditional attitudes about what is appropriate male and female work and pay, even though no obvious discriminatory factor is identified.[53]Classically the employer will rely upon the fact that the result is the outcome of different negotiating structures, as in Enderby . In practice it will be extremely difficult for the employer to demonstrate that the differences are not sex tainted, as Enderby itself shows. As we have said, the fact that jobs are determined by separate negotiating structures does not of itself show that the bargainers have not, for example, been adopting stereotyped assumptions about the appropriate rates for the jobs; and in practice the employer will be likely to find that it is very difficult to prove otherwise.[54]Even then, it is possible to envisage how exceptionally the charge of sex discrimination may be discharged. For example, it may be shown that a particular group of workers (group A) has always been paid less than another group (group B) even although the jobs are of equal value. If both groups were originally predominantly male, but group B has over time become mainly female (such as might well be the case with lawyers or academics, at least in certain fields), a tribunal might readily be satisfied that despite the current adverse effect, there is no proper basis for inferring prima facie discrimination, whether based on historical stereotyping or otherwise. The factors leading to the difference in pay may be long established but the history suggests that they do not have their roots in sex discrimination but have operated independently of the sex of the job holders.[55]A third situation is where the employer identifies some particular and specific factor which he submits causes the difference in pay but which is applied only to the predominantly male group. The factor does not create the two pools but it is applied to only one of them. In those circumstances it will be sex tainted unless the employer can show - the onus being on him - that notwithstanding that the factor has been applied so as to benefit only the male group, there are non-discriminatory reasons why that is so. This may involve not merely focusing on why the men receive more but also why similar opportunities to earn the higher pay were not afforded to the women. For example, where the difference results from a bonus arrangement, the employer might show that the bonus scheme was offered to both groups and the predominantly female group chose not to adopt it. That might demonstrate that there was no sex taint in the application of the scheme and therefore nothing to justify. The issues in this case .[56]We turn to the issues in the case and the particular facts bearing upon them. For the purposes of understanding the issues, it is necessary to distinguish three categories of claim each of which has given rise to a distinct legal issue in the appeal. It is potentially material that none of the claims were either determined or conceded prior to the coming into force of the new job evaluation scheme and the protected pay arrangements.[57]First, in three cases female carers who had been rated under the White Book job evaluation scheme compared themselves with males who had been rated as equivalent with them under that scheme but who were in receipt of certain bonus payments denied to the women. Two other claimants, community support workers covered by the Purple Book negotiations, were also claiming equal value with the same comparators and also alleged that they were entitled to the bonuses. In a preliminary hearing on this and various other preliminary matters, the Tribunal had to consider whether the additional bonus payment received by the predominantly male workers achieved a legitimate objective or whether, in truth, the productivity pay was in effect simply a disguised form of basic pay.[58]The Tribunal found that in some cases it was the former and in some the latter. They found that in all cases where bonus schemes had been introduced, the initial intent had been to reward productivity and that had for some time been achieved, but in relation to two categories of comparator, gardeners and street sweepers, the Tribunal found that the workers' performance had ceased to be effectively monitored. The situation had crystallised whereby the payments were made automatically, irrespective of performance. For these workers, the bonuses were no longer a function of productivity; they were in effect treated by everyone as part of the basic pay applicable to these groups.[59]By contrast, the bonuses paid to refuse collectors and road workers genuinely rewarded productivity and did not form part of basic pay. (There was one exception to this, a comparator who was a road worker called Mr Willoughby; he was a special case and the Tribunal found that his productivity bonus was not justified.)[60]In the light of these conclusions, the Tribunal later held that the Council's GMF defence failed with respect to those comparators whose bonuses no longer represented genuine productivity arrangements. It is now conceded that the White Book female claimants whose jobs were rated as equivalent to these comparators paid this additional unjustified bonus are entitled to receive the same pay.[61]The Purple Book claimants are in a similar position. They will also be entitled to the bonus but subject to two qualifications: they must establish that their jobs are of equal value as they claim (this merely being assumed for the purposes of the preliminary issue); and there is potentially a separate collective bargaining defence applicable to them. We discuss this below.[62]However, these two groups of claimants are not simply claiming that they were entitled to the same pay for the relevant period for which they were performing work of equal value. In addition they are claiming the benefit of the protected pay arrangements. Their argument is that if they had been receiving equal pay when they ought to have been i.e. prior to the introduction of the Green Book job evaluation scheme, then they would have been in receipt of higher pay when pay protection was introduced and would, therefore, have had the benefit of those arrangements if their pay was thereafter reduced in the job evaluation exercise.[63]The Council does not accept that the application of the protected pay provisions results in a difference in pay which is sex tainted, because the Tribunal found that the pay protection arrangements did not adversely impact on women. Further, it is submitted that even if there was any disparate impact, it was objectively justified for a number of reasons, including the fact that the purpose of the scheme is to prevent, or at least minimise, any reduction in actual pay received. The first issue therefore is whether excluding these claimants from the scope of the protected pay arrangements was unlawfully discriminatory.[64]The second issue arises only with respect to the APTC workers who have compared themselves with former White Book employees with respect to the period prior to the implementation of the Green Book job evaluation scheme. The Tribunal considered these claims with respect to a particular comparator, Mr James, who was a gardener whose pay under the White Book arrangements included a bonus which had become part of basic pay.[65]The Council submitted that even if the bonus could not be justified on the basis that it reflected a genuine productivity arrangement, there was a separate defence available to it, namely that the variation in pay was the result of the different arrangements which had historically fixed the pay of the manual and APTC workers.[66]The Council contended that the separate negotiations were a genuine explanation of the difference in treatment which had nothing to do with sex. There was evidence to show they were not sex tainted. This provided a complete answer to these equal pay claims, notwithstanding that the claimant group was predominantly female and the comparator group male.[67]The third issue also raises a collective bargaining defence, but in rather different circumstances. The APTC claimants also sought to compare themselves with a Red Book craftsman, Mr Fenny, who was an auto mechanic. They alleged that their jobs were of equal value with his. As with the gardeners, the Tribunal did not accept that his bonus was a genuine productivity payment. The Council again also sought to rely on the fact that there were different bargaining arrangements for Red Book and Purple Book workers. They contended that overall this had led to broadly equivalent terms and conditions when taken across the board, and that again this demonstrated an absence of sex discrimination.[68]There is, however, a difference of some potential significance between the Red Book comparator, Mr Fenny, and the White Book comparator, Mr James. Whilst the proportion of women covered by the White Book pay arrangements was broadly similar to the proportion of women covered by the APTC bargaining arrangements, that is not so in this comparison. The Red Book craft negotiations dealt almost exclusively with men, whilst the Purple Book staff were very largely women. The Tribunal's conclusions in summary .[69]The Tribunal heard both the different collective bargaining and the pay protection defences, together with other matters not the subject of appeal, in eight days in November 2006. We are told that the precise comparators changed during the course of the hearing but they were finally identified before the Tribunal gave its reasons.[70]As to the pay protection, the Tribunal made two findings. The first was that there was no disparate impact on women in the operation of the pay protection itself. The Tribunal noted that there were two groups of employees eligible for pay protection; the first were those who suffered a loss of bonus and were almost exclusively male; the second were those who suffered a loss of pay on the re-evaluation of jobs; they were predominantly female. They analysed the statistics both in terms of the numbers of men and women adversely affected and the proportions.[71]Although proportionately more men than women benefited from the protected pay arrangements, the Tribunal concluded that the statistics were not sufficiently detailed to justify the conclusion that there was disparate impact. Accordingly the Tribunal held that there was no need to show that the difference in treatment was justified. It was only necessary for the employer to show that the element of sex discrimination was such that the reason for the difference could properly be described as being a genuine material factor other than sex within the meaning of s1(3). Their finding that there was no disparate impact is the subject of a cross appeal.[72]The Tribunal's second finding was that the reason for the difference in treatment was so closely related to the sex of the claimants that the presumption of sex discrimination was not discharged. In reaching that conclusion the Tribunal relied heavily upon the decision of this Tribunal (Elias P presiding) in Redcar and Cleveland Borough Council v Bainbridge (No.1) [2007] IRLR 91 . The contentions of the parties in part involve a consideration of whether the effect of that decision was properly understood by the Tribunal.[73]The collective bargaining defences were also rejected. The Tribunal dealt separately with the White book and Red Book comparators. In each case they found that the reason for the difference in pay was sex tainted. Since the Council had not sought to justify it, it followed that it was unlawful. The pay protection .[74]As we have said, the Tribunal's conclusion on protected pay was heavily influenced by the decision of the EAT in Redcar and Cleveland Borough Council v Bainbridge (No.1) [2007] IRLR 91 . The facts were very similar save that in that case the employer had conceded, prior to the coming into force of the pay protection scheme, that certain of the claimants were entitled to equal pay with their chosen comparators. The comparators received pay protection but the claimants did not. This was because the claimants were not in receipt of the higher pay at the date when the new job evaluation scheme with the protected pay scheme came into force. The employers contended that since the purpose was to protect pay actually received, there was a genuine material difference other than sex which caused the difference in pay.[75]The Employment Appeal Tribunal rejected this argument with respect to those particular claimants whose case had been conceded. There were various other claimants whose cases had not been determined or conceded at the date when the protected pay came into force, but since it was not argued that they should be treated any differently, the Employment Appeal Tribunal held that they too should be entitled to the benefit of pay protection also. The EAT found that in the particular circumstances the discrimination was so closely related to the sex of the claimants that it could not properly be said to be a material difference other than sex.[76]They put weight on the earlier decision of the EAT in Snoxell v Vauxhall Motors [1977] IRLR 123. In that case the employers had provided pay protection (then known as "red circling") to a group of men who were in a job grade not open to women. There was no time limitation on the arrangement. It was conceded that the women claimants were doing like work. The Court found that the employers had not rebutted the presumption of sex discrimination. (The distinction between direct and indirect discrimination in the context of s1(3) defences had not at that date been drawn). Phillips J, giving the judgment of the Court, said this:
"Putting these arguments side by side it can be seen that the solution depends upon whether, in analysing the history of the difference in treat ment of Miss Snoxell and Mrs. Davies on the one hand and the red circle male inspectors on the other, one stops at the moment of the formation of the circle or looks further back to see why Miss Snoxell and Mrs. Davies were not within it. The arguments presented to us have, not surprisingly, considered questions of causation, and it has been said that the inability of Miss Snoxell and Mrs. Davies to join the red circle was, or was not, the effective cause of the current variation in the terms of their contracts of employment. It seems to us that this earlier discrimination can be said to be an effective cause of the current variation. But we would put the matter more broadly. The onus of proof under section 1 (3) is on the employer and it is a heavy one. Intention, and motive, are irrelevant; and we would say that an employer can never establish in the terms of section 1 (3) that the variation between the woman's contract and the man's contract is genuinely due to a material difference (other than the difference of sex) between her case and his when it can be seen that past sex discrimi nation has contributed to the variation. To allow such an answer would, we think, be contrary to the spirit and intent of the Equal Pay Act 1970, construed and interpreted in the manner we have already explained. It is true that the original discrimination occurred before December 29, 1975, and accordingly was not then unlawful; nonetheless it cannot have been the intention of the Act to permit the perpetuation of the effects of earlier discrimination."
[77]In Redcar the EAT considered that the position of the women whose cases had been conceded was sufficiently analogous to the women in Snoxell to lead to a similar conclusion:
"Discussion. 158. There is no doubt that the reason for the difference in treatment resulting from the operation of pay protection was the fact that protection is given only to those actually in receipt of higher pay at the time when the scheme was implemented. The reason why the successful women claimants were not given the same pay protection was not, therefore, specifically because of their sex. It was because the pay they were in fact receiving was not being reduced, and mitigating measures were not needed for them. That, however, does not of itself preclude the GMF being sex-tainted. In Snoxell only the male group had their pay protected. It was accepted that the employer was not seeking deliberately to protect the pay of that group because they were male, but that was the inevitable effect of the way in which pay protection was applied, and furthermore it would have been obvious at the time. It is true that there was no time limit on the pay protection arrangements in Snoxell, but the reasoning of the EAT in that case suggests that they would have found the GMF to be sex tainted even if there had been. 159. Not without some hesitation, we have concluded that the position here falls within the principle enunciated in Snoxell. The tribunal was entitled to find that the GMF was sex tainted and could not be relied upon. Sex discrimination was a direct and significant cause of the difference in treatment. In our view this can be seen most clearly perhaps with regard to those whose equal pay claims had been conceded prior to the protected pay being implemented. By fixing on whether the employee was actually in receipt of pay, rather than being entitled to it, the effect was to exclude that predominantly female group. There was an inextricable link, apparent to the employer at the time pay protection was implemented, between the sex of these workers and the reason why they were not in receipt of pay protection. 160. It is true that since they were not actually receiving pay the need for cushioning was not present. They would presumably have arranged their financial commitments on the basis of what they were actually receiving. But this in turn was only because their claims to equal pay, which the employer had already conceded were justified, had not been implemented. Mr Cavanagh accepted that had they been in receipt of their lawful pay by the date of implementation then there would be no answer to their claim to pay protection. The council can surely not pray in aid its own failure to implement equality as a justification for defeating it. It would frustrate the fundamental principle of equality to deny them benefits which, as everyone accepted, they were legally entitled to receive. Had the employers corrected the inequality when they ought to have done, then there can be no doubt that these women would have been in exactly the same situation as the comparators. They would have had the benefit of the pay protection. (We recognise that had the true cost been appreciated then the council may have structured a different scheme with lesser pay protection. To that extent the claimants may receive more than they would have done had their pay been equalised earlier. But this is no more than speculation.)"
[78]The EAT then noted that no basis had been advanced for treating the other groups differently. It was noted, however, that the case turned to some extent on its own facts, and that there may be situations where giving pay protection to prolong historical discrimination might be lawful.[79]The Employment Tribunal focused heavily on this analysis. In particular, they sought to analyse the extent to which the facts of this case did or did not correlate with those in Bainbridge :
"The similarities are obvious. The pay protection scheme was to all intents and purposes the same. The loser group entitled to pay protection was the same and the claimant groups who seek a remedy are the same or very similar. There is very similar evidence of what the President called "historic discrimination"
. The men continued to receive, albeit at a reduced level, discriminatory payments as the Tribunal has now found. Had these claims been heard and determined prior to the implementation of the pay protection arrangements there would have been no answer to the claim to pay protection."[80]The Tribunal also observed that the Council must have realised that "there was a considerable risk" that parts of the pay structure under scrutiny would betray discrimination, and that there were communications between them and the adjoining Redcar Council where claims had succeeded. Nor had the Council even assessed the potential cost of extending the scheme to those who had lodged claims at the date the scheme came into force. The Tribunal's conclusion was that:
"We do not accept that the respondent had no reason to suppose, when pay protection was implemented, that it would have discriminatory effects. Applying the Bilka test we consider that a period of protection of previously discriminatory pay limited in time in order to phase it out will have been in itself a legitimate aim. It may be that it was a reasonably necessary means of achieving the objective of phasing out pay, but it has not been shown that it is an appropriate and proportionate means of achieving that objective when the size of the disadvantaged group and the cost of including or excluding that group had not been calculated. In short, this defence was tainted by sex discrimination and, to the extent that it was, it was not justified."
[81]Mr Jeans criticises the decision on a number of grounds. First, he strongly criticises the Tribunal for focusing so sharply on the Redcar case. He submits that they ought to have gone back to basic legal principles and analysed the facts in accordance with the statutory provisions. He also submits that the Tribunal was not justified in speculating about the likelihood of claims succeeding and said that there was no evidence of any collaboration or communication between the Middlesbrough and Redcar Councils.[82]Mr Jeans says that had established principles of discrimination law been applied, they yield a clear and inevitable answer. The evidence was that more women than men benefited from the scheme, therefore it did not demonstrate any disparate impact or other disadvantage to women and so there was no need objectively to justify. Furthermore, there was plainly no direct discrimination between men and women and therefore there could be no breach of the Equal Pay Act .[83]Even if there were a need to justify the scheme, that hurdle was clearly surmounted here. He pointed to the fact that the pay protection was an important aspect of the arrangements made to implement the principle of equality for the future; indeed, without it, the agreement over job evaluation would probably not have been reached. In addition there were legal problems involved in reducing the pay of any employees. He also emphasised the need to protect the income stream of those suffering a reduction as a consequence of the new arrangements, and the fact that the scheme was made under the auspices of ACAS.[84]Mr Allen cross appeals against the Tribunal's finding that there was no disparate impact in the operation of the protected pay arrangements. He says that the Tribunal's conclusion in the light of the statistics was extraordinary. Whilst in simple numerical terms there may have been more women than men affected, when considered proportionately far fewer women than men benefited, as the statistics clearly demonstrated. Accordingly there was disparate impact and therefore an obligation to justify[85]Both these arguments focus on the protected pay arrangements as implemented, whether for the purpose of seeking to justify them or to show that they had a disparate impact. In our judgment they miss the point. The question is not whether the scheme itself, in so far as it was applied at the material time, operated in a discriminatory way or could be justified. There is no doubt that it could be justified. Even if there were disparate impact, Mr Allen concedes that the principle of pay protection could be justified provided it is proportionate. So his cross appeal, which seeks to establish that the pay protection arrangements as introduced adversely impacted on women, goes nowhere, even if correct. (We think it probably is correct but given the irrelevance of the issue we will not lengthen this judgment with an analysis why.)[86]The difference which has to be justified here, as the Employment Tribunal clearly recognised, is not extending the pay benefit of the scheme to those who, albeit unbeknown to the employer at the time when the scheme was introduced, were in fact also subsequently found to be entitled to equal pay when the scheme was introduced. That group was overwhelmingly female, and not simply fortuitously so. Since there was sex tainting, the obligation objectively to justify did apply to them. However, the justification which then has to be addressed is not the justification for having a pay protection arrangement at all; it is the justification for not extending that arrangement to those women who subsequently established their equal pay rights.[87]The relevant question, therefore, is whether it infringes the principle of equal pay to fail to apply the scheme retrospectively to those persons, predominantly female, who it was later found were entitled to receive equal pay but did not receive it even before the protected pay scheme came into force. Mr Jeans does not answer this claim by demonstrating that the scheme itself was introduced and implemented in a non-discriminatory fashion.[88]However, Mr Jeans submits that even focusing on the right question, the Tribunal erred in law. Considerations of policy, principle and authority all justified the approach adopted by the Council.[89]As to policy, he points out that in practice, when an employer introduces a new job evaluation scheme, there will almost inevitably be cases where, having been alerted to the fact that jobs are now rated as equivalent, some women will take the view that their jobs must have been of equal value at an earlier stage. If the claimants are correct, not only will that yield them back pay, but in addition the benefit of any protected pay arrangements. This would, he submits, be a very considerable disincentive indeed for employers introducing job evaluation schemes. The experienced lay members of this Tribunal have no doubt that he is right about that, and we recognised the potentially adverse implications of a case like this succeeding in Bainbridge . At the same time, although policy may inform the law, it cannot determine it.[90]On the question of principle, although much of Mr Jeans' argument was directed at the evidence adduced which provided justification for the scheme itself rather than for not applying it retrospectively to successful claimants, many of the factors he relied on also support the latter justification. In particular he focused on the purpose of the scheme being to cushion employees from a drop in income; and he contended that the Tribunal was wholly unrealistic about the problems of costing an indeterminate number of claims, only some of which might succeed, and with no touchstone for assessing their value. He also criticised the Tribunal in this context for assuming, without any evidence to support it, that there were discussions between the Council and Redcar. He submits that had the Tribunal properly identified the legal principles involved, they would inevitably have concluded that the Council's stance was justified.[91]Finally, on authority, Mr Jeans relies upon the case of Farthing v Ministry of Defence [1980] IRLR 402, a decision of the Court of Appeal, in which (unusually) women were paid more than men as a consequence of procedures which were adopted specifically with a view to eliminating discrimination against women with respect to pay. Mr Jeans says that similarly in this case the purpose of this protected pay was to move towards the implementation of a principle of equal pay and there was nothing discriminatory about it.[92]He also submits that his submission is entirely in accord with Bainbridge . He reserves the right to argue before a higher court that Bainbridge was wrongly decided and further submits that even if that is not so, this case is distinguishable and the Tribunal erred in not so finding. The key feature of Bainbridge , he submits, is that the women were already entitled to equal pay at the date when the pay protection was introduced. To the knowledge of the employer, they were being denied their clear and unambiguous equal pay rights even before the pay protection scheme was introduced. Had they received that at the relevant time then they would have been in receipt of the protected pay.[93]That is not the situation here; there had been no case either conceded or determined when the pay protection came into force. He distinguishes the case of Snoxell v Vauxhall Motors by saying that that turned on the fact that there was historic direct discrimination. He contends that on the assumption that Bainbridge is correct, the proper boundary for the cases where pay protection will fall within or without s1(3) should be where there is either direct or admitted discrimination. Where there are merely claims that the old arrangements were discriminatory, it should be a defence that the scheme is limited to those in receipt of the higher pay when it is introduced.[94]Mr Allen submits, and we agree, that essentially the Tribunal did focus on the right question, namely whether it was justified not to extend the protected pay arrangements. He contends that they found, as they were entitled to do on the evidence, that the Council must have appreciated that at least some of the claims would be likely to be successful. They must have recognised that the scheme once implemented would be likely to have some discriminatory effects. Thereafter, it was for the Council to make an assessment of the potential discriminatory effect of their arrangements and to seek to strike a balance between those discriminatory effects and the legitimate objectives of the scheme.[95]The Tribunal concluded that they had not done that, not least because they had not sought to assess the cost implications of applying the arrangements to women as and when they succeeded in their claims. Accordingly, Mr Allen submits that there was evidence to support their conclusion and no failure properly to apply the appropriate legal principles. In reality this was really a perversity appeal dressed in a different and more beguiling guise.[96]Mr Allen further submits that the Tribunal was right to say that the logic of Bainbridge compelled the conclusion that the defence of justification failed and Farthing has no relevance to this case since it did not concern historic discrimination. If the Council's argument were correct, it would permit them to perpetuate the effects of unlawful discrimination, contrary to Article 141 and in breach of the fundamental principle of equality. Discussion and Conclusion .[97]We recognise that this case is not determined by the Bainbridge decision. A significant feature of that case was that the employers knew when the scheme was implemented that they were not paying at least some women the pay due to them, in breach of the equality clause. Had they implemented the pay immediately they accepted that they had been acting unlawfully, the women would have had the benefit of pay protection. In those circumstances sex was "a direct and significant cause of the difference in treatment" (para.159) The difference could not therefore be explained by a difference other than sex within the meaning of s1(3).[98]The issue of justification did not then arise. Even if it did, it was bound to fail since the Council were in effect relying on their own failure to correct an inequality which they knew existed as the justification for defeating the claim. Although the reason for not including the claimants was that they were not in receipt of the higher pay when the scheme was introduced, that was only because the employers had failed to make that payment knowing they were entitled to it. The case was analogous to Snoxell which involved the continuation of direct sex discrimination. The other claimants in that case effectively succeeded on the back of the women whose equal pay claims had been conceded.[99]In this case the scheme when implemented was not, to the actual knowledge of the Council, exacerbating existing discrimination. The Council was not structuring an arrangement in a way which it knew would perpetuate such discrimination. Payments were made to those in receipt of the higher pay - and these included all those who to the Council's knowledge were entitled to the higher pay - when the scheme was introduced. That was not direct discrimination nor anything analogous to it.[100]However, we have no doubt that the Tribunal was entitled to find that the Council must have realised that some of the outstanding claims at least were likely to succeed. (We think it was of no great materiality whether the Council were involved in discussions with Redcar or not, or whether there was evidence to that effect, although we would be surprised if they were not). Since the claimants were overwhelmingly female, the decision to limit the benefits to those who to the knowledge of the Council were in receipt of the higher pay when the scheme was introduced required objective justification.[101]Is the scheme justified? Mr Jeans submits that Farthing is binding on us and demonstrates that it is justified. We do not accept that submission. The EAT considered that case in Bainbridge (paras 145-147). We will not repeat that analysis here. Suffice it to say that we agree with Mr Allen that Farthing , which was not an historic discrimination case, is of little relevance in the analysis of this case. It does indeed show that pay protection can be justified in an appropriate case, but that is not disputed.[102]So we must look at the matter shorn of authority and bearing in mind that we are reviewing the decision of the Tribunal. An important preliminary question, which lies at the heart of this issue, is whether it can, as a matter of law, ever be proportionate to deny a woman a benefit she would have had if she had received her equal pay at the appropriate time. Is the employer necessarily prevented from relying on his own failure to provide equal pay when he ought to have done?[103]We have come to the conclusion that he is not. The Council is not knowingly and deliberately dragging his heels in failing to implement pay rises which it knows are due to the women. Moreover, the claimant is not being denied the appropriate pay she ought to have received for the job. That is achieved by granting back pay and it is not suggested that there could be any justification for denying her that. She is being denied a payment which was contingent on the employee receiving the higher pay at the material time, which she was not doing. (Further if the Council had had to structure the scheme to enable the claimants to receive protected pay, no doubt the effect of including them would have meant that the scheme would have had to provide less protection than was in fact afforded to employees.) That adversely impacted on these women but we see no reason in principle why such adverse impact should not be capable in an appropriate case of constituting a justification for the difference in treatment. Accordingly, in our view the withholding of protected pay does not inevitably involve a failure to give full effect to Article 141, as Mr Allen submits; that provision itself recognises that some forms of sex tainted discrimination can be justified.[104]We find the arguments on justification in this case very finely balanced. It is plainly a legitimate objective to protect the salary stream and to distinguish between two employees on that basis. For example, redundancy payments are in part designed to protect against precisely that consequence and therefore fixing the compensation by reference to final salary is justified even if some women are adversely affected by that scheme: see Barry v Midland Bank [1999] ICR 859 (HL). It is also legitimate to have as an objective the introduction of a job evaluation scheme which will eliminate discriminatory pay for the future. If this would be undermined by an extension of the scheme, that is also a highly relevant consideration potentially constituting justification.[105]Are the means proportionate to the legitimate aim? An unusual feature in a case of this nature is that it is impossible to know precisely what the discriminatory effect of the scheme - or more accurately, not extending the scheme to this group - will be. We accept the submission that it was in this case, and will usually be, impossible to predict in advance the potential liability of equal pay claims may be, both because there is no knowing how many claims may come out of the woodwork after the scheme is introduced, and because it can be virtually impossible to make any realistic assessment of the prospects of the claims or their potential value. Any assessment will perforce be highly speculative.[106]That makes the proportionality assessment extremely difficult. Indeed, Mr Jeans submits, and we accept, that it is would have necessarily undermined the Council's ability to introduce a scheme of this nature, which was crucial to the creation of the job evaluation scheme, if the Council had had to make allowance in costing the scheme for a wholly unknown, and at that stage unknowable, potential liability. Inevitably a conservative approach to the costing of schemes would be adopted, which might well offer too little to the affected staff to secure the changes at all.[107]The Tribunal placed considerable emphasis on the failure to carry out that exercise but without identifying this intrinsic difficulty. Moreover, in its consideration of the justification of the arrangement, whilst it recognised the cushioning effect, it did not place any emphasis upon it at all when analysing whether the payments were justified. For these reasons we would accept that its decision cannot stand.[108]In our judgment the Council was justified in not applying the scheme to those claimants who might succeed after its introduction in establishing the right to equal pay. We consider given that the purpose of the scheme was to cushion employees from the potentially disastrous effects of a sudden drop in pay, the Council was entitled to take the view that it should limit the benefit to those actually in that group and to exclude all others even if some of them ought to have been in the group. Unless the pay was actually being received, there was nothing to protect. We think that is itself sufficient justification, but it is reinforced by the fact that the need to reach a protected pay arrangement, with the agreement of the unions, was crucial to the making of the job evaluation scheme. Any assessment of future costing would inevitably be highly speculative and would undermine the ability to obtain agreement for the scheme.[109]We have considered whether we should remit the matter for the Tribunal to consider the matter afresh. However, it seems to us that in substance the matters we have identified will be true of all schemes of this nature. In truth the answer is not likely to be fact sensitive, at least not where large employers are involved. In every case the justification based on the need to cushion the reduction in pay actually received will apply. So, to a greater or lesser extent, will the risk of a large but unknown number of potential claims which would scupper the ability to reach agreement.[110]Accordingly we uphold the appeal on this point. We have sympathy for the Tribunal, however, not least because in focusing on cost as they did, they placed weight on a hypothetical example of a potentially lawful scheme which I gave when delivering the judgment of the EAT in the Bainbridge case. In hindsight, that would have been better omitted from the judgment. Different collective bargaining .[111]We turn to the collective bargaining defence. It is necessary to consider the two cases- the comparisons with Mr James and Mr Fenny respectively - separately for reasons we have given. In the case of the White Book comparator, Mr James, it is conceded that the proportion of women covered by the White Book negotiations is broadly similar - indeed, slightly larger - than the proportion covered by the former Purple Book negotiations. Both groups consist mainly of women. That is not so with the Red Book comparator, Mr Fenny. There the male workers covered by that bargaining group are overwhelmingly male. Mr James as comparator.[112]A preliminary issue here is whether there is prima facie indirect discrimination. Mr Jeans submits that the statistics are sufficient to demonstrate that there is no adverse impact on women. It follows that it is enough for the Council to show that the difference in pay is genuinely causally due to a difference of sex without having to go further and establish objective justification. That is so even if, contrary to the argument we have accepted, the fact of disparate impact is without more sufficient to establish prima facie indirect discrimination. He submits that the difference here was the different collective bargaining regimes which were not sex tainted.[113]Mr Allen contends that this is not the appropriate comparison at all. Rather the analysis should be directed at the proportion of women who are in the particular category of worker constituting the claimants, and the proportion of women in the category constituting the chosen comparator i.e. comparing care assistants with gardeners. On that comparison the disparate impact was striking, and this was of itself sufficient to require justification.[114]Alternatively he says that in any event, it was plain that the bargaining process here was sex tainted and required justification on that account. He was critical of the Tribunal for failing to consider justification. However, they did not need to do so. Mr Jeans had not sought to suggest that the difference of pay was objectively justified; his case was simply that the cause of the difference was collective bargaining which was free from any sex taint.[115]For reasons we have already given, in our view Mr Jeans ought to succeed if he can show that the difference in pay was not sex tainted, even if Mr Allen is correct that there is a disparate impact on the female group. Since objective justification is not advanced - and we surmise could not have been since the productivity benefits justifying the difference in pay were no longer being achieved - this is the only material issue. The Tribunal's decision[116]The Council contended that the difference between the pay of Mr James and the APTC comparators could be wholly explained solely by the fact that his terms were fixed by different bargaining arrangements. The Tribunal noted that the basic pay of Mr James was fixed by the national negotiations. His bonus was, however, fixed locally as the national agreement prescribed. As we have said, the Tribunal found that his bonus no longer provided any legitimate productivity benefits and was to be treated as part of the basic pay.[117]The Tribunal considered various authorities, some of which we analyse below, and concluded that the contention failed. It did so somewhat summarily, observing that:
"In the case of Mr James we consider that there is clear evidence that the reason for any difference in pay is tainted by sex and is not explicable in terms of separate collective bargaining. His basic pay is negotiated under the same national bargaining arrangements as the claimants' pay."
[118]Mr Jeans submits that this analysis betrays numerous errors of law. First, although basic pay was fixed in common negotiations, the bonus was only fixed locally for the gardeners because the Green Book expressly provided that bonuses should be preserved pending the full introduction of single status. It was that bargaining history which explained the difference in pay.[119]Second, there was clear evidence that the different bargaining histories brought about a number of different results and it was not suggested that the bargaining arrangements themselves were discriminatory in any way, or that there was discrimination with respect to access to any of the job groups. It would be absurd if the separate bargaining could not be relied upon, reflecting the priorities of the different groups of workers.[120]Third, the Tribunal has failed to give any explanation at all for its cryptic conclusion that the reason for the difference in pay was tainted by sex. He submits that there is simply no evidence to justify that conclusion; alternatively, the Tribunal has not given reasons for that conclusion and therefore the decision does not meet the requirement as to reasons set down in Meek v City of Birmingham District Council [1987] IRLR 250 .[121]Mr Allen submits that, read fairly and in context, the Tribunal's conclusion is manifestly sustainable. The Council, he submits, is seeking to run a perversity challenge in another guise. There was undisputed evidence that both groups had been subject to the same national bargaining since the introduction of the Green Book, and therefore the Tribunal's analysis on this point was flawless. Moreover, even if the adverse impact on the claimant group was itself insufficient to amount to indirect discrimination requiring justification, the sex taint in the bargaining processes was obvious. Bonuses were only paid to the predominantly male workgroups. It was no accident that they were male; this was the result of historical stereotyping. The collective bargaining was not free from sex taint. Discussion .[122]We accept Mr Jeans' submission that the Tribunal was wrong to say that because the same national bargaining applies to both groups and bonuses are now negotiated locally, that negates the Council's argument. The origin of the fact that the comparators were in receipt of the bonus and the APTC workers were not lies in the different national bargaining arrangements, and the current Green Book expressly preserved bonus payments pending the job evaluation scheme being introduced. It therefore carried forward the historical effect of different bargaining structures. If they were not discriminatory, they could not have become so because of their continuation after the introduction of the Green Book scheme, even though the pay of the comparator is now the product of both national and local negotiations.[123]There is no doubt that in an appropriate case the fact that pay is determined by different collective bargaining arrangements which are not themselves operated in a sexually discriminatory way will constitute a genuine reason other than sex which explains the difference in pay: see Reed Packaging v Boozer [1988] ICR 391. We do not understand Mr Allen to dissent from this.[124]An obvious example is where there are two different groups, of similar proportions by gender, but one group earns less than the other. If a woman in the less well paid group compares herself with a male carrying out equal work in the other group, the difference bargaining arrangements will explain, in the sense of cause, the difference in pay and will constitute a genuine factor other than sex within the meaning of s1(3).[125]The position will be otherwise if there is a marked difference in the sex balance in the different groups, as Enderby shows. There is then evidence from which a tribunal could conclude that the process is sex tainted unless the employer can demonstrate otherwise, but the fact of separate bargaining will not of itself be likely to disprove the possibility of discrimination. That is an explanation but not, in general, a justification.[126]In this case the two bargaining groups were not marked by sex imbalance. So in principle different collective bargaining arrangements were capable of providing a non sex-based explanation for the difference in pay. We would accept, for example, that they could in principle explain why bonuses were payable to workers in the one group but not the other.[127]However, for the difference in pay to be shown to be free of sex taint, the employer must show that there is no sex discrimination in the operation of the bargaining arrangements considered separately. In our judgment, that is not so here. There has been discrimination in the operation of the bonus payments by reference to workers covered by the White Book. That much is now conceded.[128]The effect of the national and local agreements treated together is that, considering the White Book employees as a single entity, there has been unlawful sex discrimination in the operation of the bargaining processes applicable to them. Predominantly male groups have been in receipt of a bonus not payable to female groups. The premise that the difference in pay between the applicant and comparator group is the consequence of non-discriminatory collective bargaining is therefore not made out.[129]The logic of Mr Jeans' submission appears to be that whilst the White Book claimants can successfully allege that their comparators have been receiving discriminatory payments by receiving unjustified bonuses paid to them for reasons connected with their sex, Purple Book claimants cannot. Effectively they are barred from crossing collective bargaining boundaries. This is inconsistent with principle and also with the Enderby decision itself.[130]In short, in our view once the Tribunal had found that the bonus differential could not be demonstrated to be free of sex discrimination, the separate collective bargaining argument was doomed to fail. It would explain the fact that bonuses were paid to one group and not in the other, but the fixing of the comparator's pay could not then be free of sex discrimination, and the fact that the claimants pay was separately negotiated could not alter that fact.[131]Although the Tribunal's conclusions on this point were extremely brief, the substance of their conclusion was that the difference in pay could not be explained by non-discriminatory processes. We agree. The Red Book comparison; Mr Fenny[132]As with the Mr James, the bonus of Mr Fenny, the Red Book comparator, is locally negotiated, although it is expressed as a percentage of the nationally agreed rate. Again, the Tribunal found that the monitoring and assessment which had originally been part of the particular bonus scheme to which he was subject had ceased to operate, and the bonus was simply paid as a fixed element of pay. A distinction between the White Book and Red Book comparators is that there were at all material times separate national and local negotiations for the Red and Purple Books.[133]In addition there was plainly a disparate impact on women even taking the proportions covered by the bargaining arrangements as a whole. However, given our conclusion on the proper analysis for determining whether there is prima facie indirect discrimination, it is still open to the Council to demonstrate that the difference in pay was not sex tainted so that no question of justification arose.[134]An argument which was advanced to the Tribunal with respect to this comparison, and was repeated before us, was broadly as follows. The bonus was paid pursuant to a separate bargaining arrangement. True it is that the craft group was predominantly male and the APTC group predominantly female. That showed that the Enderby condition was satisfied. Enderby might in an appropriate case justify an inference of sex discrimination. However, this is not such an appropriate case.[135]In Enderby there were striking differences in pay with respect to persons who, on the assumption made in that case, were doing work of equal value. In the absence of clear evidence demonstrating that this was not sex tainted, the only proper inference was that prima facie indirect discrimination was established. But where there is no such striking difference in terms and conditions then it is not legitimate to draw the same inference. That was the situation here. A careful comparison of the benefits to claimant and comparator under the two schemes would demonstrate that there was no significant shortfall in benefits to the Purple Book claimants and therefore no legitimate basis for inferring sex discrimination.[136]In short, the inference of indirect discrimination is a function of both the statistical disparity in the gender of the relevant workgroups and the disparity in terms and conditions. It is not legitimate to focus simply on the former. If the disparity in terms is weak and arguably non-existent, it is simply not justified to draw the inference of discrimination at all. The difference in pay is then entirely consistent with non-discriminatory collective bargaining arrangements; such arrangements would be likely to produce an outcome where there were differences in terms and conditions, but difference did not necessarily establish discrimination.[137]The Tribunal rejected the submission that the difference in pay resulted from non-discriminatory collective bargaining:
"In the case of Mr Fenny, there is clear evidence of an overwhelmingly male workforce covered by the NJC and, probably in consequence of that, exclusively male representation in the negotiating body. The disadvantaged group not in receipt of any bonus at all was predominantly female and we do not consider that the supposed advantages accruing to the Purple Book employees arising from their separately negotiated terms were equal in value to the terms enjoyed by Mr Fenny's work group. The existence of separate work groups facilitates discrimination and could perpetuate it, or lead to its reintroduction at a later date. At this stage we have to act on the provision or assumption of equality of value."
[138]Mr Jeans again criticises this analysis on a number of grounds. He says that the Tribunal singularly failed to address his argument adequately. They failed to assess in any real depth the benefits accruing to the Purple Book workers when compared with the Red Book workers, and their conclusion that the advantages were not broadly equal was an impossible conclusion in the absence of a careful analysis of benefits and burdens. In any event it was wholly inadequately reasoned and did not comply with the requirements as to reasons in Meek . He asserts that the observation that the negotiating body is overwhelmingly male is an irrelevant consideration and does not begin to show discrimination.[139]Finally, he says that whilst the Tribunal is correct to say that separate bargaining arrangements for different groups might perpetuate discrimination, the question is whether it has done so. The possibility that it might does not show that it did in this case.[140]Mr Allen accepts that the reasoning of the Tribunal is not altogether satisfactory. He accepted in oral argument, we think wisely, that it was unjustified for the Tribunal to treat as evidence of sex tainting the fact that the negotiating body itself consisted of predominantly male representatives. This would not, in our view, of itself show, or even tend to show, that they thereby subconsciously discriminated in favour of their own sex.[141]Indeed, we envisage that there would be powerful criticisms of a tribunal which made the assumption, without any evidence that, say, a black representative would favour others of his own colour, or a Jew or Muslim would favour others of the same religion. Even if it is true as a generalisation that different sexes, races or religions favour their own - and we know of no evidence to support this - it would be unacceptable stereotyping to assume that all negotiators act in the same way without evidence that they have done so.[142]However, Mr Allen submits that the other reasons are valid and sustainable, for much the same reason as with Mr James. Moreover, he submits that in essence the case can be analysed in much the same way as the James comparison (although it is true that the Tribunal did not adopt that easy path). The difference in pay was sex tainted because it was only the predominantly male groups which could earn the bonus. Furthermore, it was open to the Tribunal on the evidence to conclude that even looking at the terms as a whole, the Green Book claimants were significantly less favourably treated than their Red Book comparator and therefore Enderby was directly in point.[143]We agree that the reasoning of the Tribunal is not altogether satisfactory, but we accept Mr Allen's submission that in essence this argument fails for the same reason as that advanced with respect to Mr James. The payment of the bonus was directed at male only groups. The fact that there were different negotiating groups does not preclude a legitimate inference that the payment of a bonus to craft workers, which had become crystallised into part of their basic pay, was connected with their sex and therefore required justification. Since none was put forward, the claim was bound to fail. The comparable benefits argument .[144]Strictly this relieves us of the obligation to deal with the wider aspects of Mr Jeans' argument. We see the force of the argument that an overall assessment of benefits and burdens should be carried out before inferring discrimination merely on the basis that an overwhelmingly male group has some benefits which a predominantly female group does not have and vice versa. It is plainly the case that there is far less justification to infer discrimination where an alleged difference in pay is very small than where it is potentially substantial. Small differences might result from certain stereotypical assumptions about the nature of the work but they might equally be explained by the different outcome of the bargaining process.[145]Indeed one might suppose that if the female group were slightly better paid than the overwhelmingly male group, then an inference of discrimination against the male group would surely not be justified because traditionally men's jobs have not been paid less favourably than women's.[146]However, it seems to us that there are two difficulties in particular which make this submission erroneous in law. The first is that the burden of proof is on the employer and they have to discharge that burden. This requires that they demonstrate that the difference is not sex tainted or that if it is, it is objectively justified. Establishing that there are good reasons to suppose that it is not, or may not be sex tainted is not enough.[147]Second, the argument really seems to come down to the claim that even if the woman is disadvantaged in pay, this can be compensated by countervailing advantages, an approach specifically rejected by the House of Lords in Hayward v Cammel Laird Shipbuilders Ltd. no 2 [1988]A.C.894. The focus, at least in domestic law, must be on each term of the contract. In so far as there is evidence that the particular term as to pay is no less favourable, that goes to the question whether there has been a breach of the equal pay provision at all. Compensating terms do not provide an objective justification if there has been.[148]For these various reasons, therefore, the arguments based on submission that the different negotiating structures have led to different results without any taint of sex discrimination fails. Conclusion .[149]The appeal succeeds with regard to the Tribunal's finding that the pay protection policy should be extended to those women who, after its introduction, were found to have been discriminated against in their pay in relation to the period before the policy was implemented. On a proper analysis, the employers were justified in not extending the protection to that group.[150]However, the Tribunal was right to reject the Council's contention that the difference in pay was justified by different collective bargaining arrangements. The appeal with respect to those aspects of the decision therefore fails.