England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE EMPLOYMENT APPEAL TRIBUNAL
THE HON MR JUSTICE UNDERHILL (PRESIDENT)
UKEAT/0123/10/LA
BAILII: [2011] UKEAT 0123_10_1803

		Royal Courts of Justice Strand, London, WC2A 2LL
		06/12/2012

Lord Justice Mummery :

Introduction

1. In June 2008 nine women (the Claimants) employed by the Audit Commission began this case under the Equal Pay Act 1970, as amended (the 1970 Act). Since then they have been made redundant. Their jobs ended. Their case goes on. If, at some point, it is referred to Luxembourg for a ruling, it will probably not end before about 2016.
2. In the judgments of the Employment Tribunal (ET), in which the Audit Commission failed, and of the Employment Appeal Tribunal (EAT), in which the Audit Commission succeeded, the conclusions reached on each of the two issues that determine the outcome of the case differed:-

(1) Is this a case of prima facie sex discrimination? If so,

(2) Is the discrimination objectively justified?

General remarks

3. I will set the facts of the case and the law applicable to them in a wider scene with some preliminary remarks on the current spate of equal pay claims in the tribunals and of appeals generated by them. This is an opportunity to pause and take stock of what is going on in the workplace, in the tribunals and in the courts.

Specialist tribunals

4. Two specialist tribunals in disagreement is a sign that the appeal to this non-specialist court is not likely to be straightforward. The 1970 Act is not easy to interpret or to apply. I am not alone in thinking that the cases now being brought under it are more difficult than they were in the past: the evidence is more elaborate, the arguments are longer and more intricate and the accumulating judgments also grow longer and more complicated.
5. Over 30 years of litigation have not eradicated unjustified pay discrimination against both women and men. (Unfairness in pay levels persists, but does not, of course, have anything to do with this legislation, this case or this court.) Equal pay litigation in the ETs has now reached almost epidemic proportions. Unlike most of the recent test cases involving high value and/or multiple claims, this is not a big case: the sums involved, though large enough to matter to the parties, are relatively modest. Nor would I call it a complicated case. Most equal pay cases are more complex on the facts and the law. Yet I have found this case (now in its third, but perhaps not final round) far more troublesome than most and, on reflection, my views about how this appeal should be decided have changed. The case turns mainly on its facts. The points of principle in play matter and not just to these parties, for there are many more cases to come.

Litigation and/or negotiation?

6. As the relevant events all happened before 1 October 2010, the case falls under the 1970 Act. Expensive and unpredictable equal pay cases dragging on for years and years with differing results at ascending levels of decision will now continue in the tribunals and courts under the Equality Act 2010.
7. Why do employees and their employers optimistically invest so much time, effort and money in the uncertainties of contests of this kind? The option of negotiation is often available. Constructive and skilful negotiations conducted in good faith can produce fairer, more realistic and more enduring benefits all round. They also avoid the long-term damage that litigation does to relationships between people who, for their common good, have to work together. Perhaps the question is a pointless one, because there are as many answers as there are cases. Perhaps the question is naïve, because negotiations are in fact as pointless as the question. Behind the question is a concern about avoidable litigation and its detrimental effects on the links in the indispensable "Human Chain" that transmits the "continuities and solidarities" of labour and life (Seamus Heaney Human Chain at p.18).

Legislative aim

8. The long-term goal of the equal pay legislation, which has been in force since 1975, is not, of course, interminable litigation between waged workers and their employers about their rights. They all have other things to do and to spend their money on. The aim is the elimination of sex discrimination against women and against men in matters of pay. Putting that uncontroversial aim into practice is taking a very long time indeed, which is not surprising as the whole set up involves, indeed requires, the clashing of rights not just between employer and employee, but also as between groups of employees. The fact that the rights are qualified, not absolute, has not deterred trips to the tribunals and confrontation in the courts, which have demonstrated that they are not necessarily the best places in which to put an end to the injustices of discrimination in the workplace.

Direct and indirect discrimination in pay

9. Direct sex discrimination in pay is not really a problem. Contested cases rarely occur in practice if the discrimination is direct, even brazen, and cannot possibly be justified under any circumstances.
10. Claims based on indirect sex discrimination are a different matter. They pose special problems of legal analysis, evidence and proof. The legal interpretation and sensible application of the relevant legislation, both domestic and European, to cases of alleged indirect discrimination are most severely tested in cases like the present, which involve distinctions that are highly debatable on the facts: is this a case of (a) lawful non-discriminatory pay differences; or (b) unlawful discriminatory pay differences; or (c) discriminatory pay differences that are lawful, because they are justified?
11. It is a truism, better repeated than left unsaid, that not every difference in pay between men and women is proof of unlawful sex discrimination. It is a useful tip about a trap to avoid. It does not really tell us much else. There is plenty of general guidance in the authorities on how to draw the line between what is lawful and what is unlawful. We are learning about that with the benefit of regularly updated instruction and from the educational experience of trial and error. Where to draw the line in the particular case is a topic on which expert (and non-expert) opinions are almost bound to differ in some cases. This is one of them.
12. The twists and turns of the arguments in discrimination decisions are fraught with danger and difficulty. The risk of unfortunate unintended consequences is ever-present. Removing or reversing unjustified indirect discrimination in the pay of one sex can impact on the pay of the other sex and can even result in unjustified indirect discrimination against them: see, for example, the discussion in Homer v. Chief Constable of West Yorkshire Police [2012] UKSC 15 at [26], [30] and [36] (Homer).

Levelling playing fields and reversals of fortune

13. Reversals of fortune are regularly experienced by litigants in equal pay cases in the tribunals and courts. They are not usually explained by lack of experience, intelligence or understanding on the part of those entrusted with the task of finding sensible solutions that are both within the framework of the legislation and, preferably, workable in practice. The situations presented to the tribunals can sometimes hover on the verge of non-justiciability. Considerable demands are made on judicial skills of conscientious and objective judicial assessment of the detailed evidence and argument. No case can be decided judicially by the mindless application of decisions in other cases, or by the mechanical operation of an inflexible doctrine, or by easygoing leanings in favour of one side or the other.
14. A theory of indirect discrimination only has to be stated, as it was, for example, in Homer's Case at [17], to alert the trained legal mind to a range of potential problems about the definition of basic terms and about acceptable modes of proof available in concrete cases:-

"The law of indirect discrimination is an attempt to level the playing field by subjecting to scrutiny requirements which look neutral on their face but in reality work to the comparative disadvantage of people with a particular protected characteristic."

15. The wise words of Edmund Burke encapsulate the problem of putting that kind of thing into practice, whether in the workplace or in the tribunals and courts:-

"…circumstances are infinite and infinitely combined. Circumstances alter cases and practicality governs conclusions."

16. The architects of equal pay law understood that the procedures and non- specialist personnel in "ordinary" litigation in the civil courts were not suited to legal disputes that involve levelling playing fields, scrutinising the neutrality of requirements for their "reality" and sussing out and assessing situations of comparative economic disadvantage. Hence the designated (though non-exclusive) jurisdiction of the ET and of the EAT, in which most of the cases begin and end. The tribunals have built up an impressive body of specialist expertise supplemented by relevant professional skills.
17. The professional advisers and skilled advocates practising in this area are indispensable for the help they give to tribunals, which have to cope with a heavy, demanding and increasing case load of equal pay claims. The overall standards of competence are, in my view, as high as can be found in any other area of legal practice in our judicial system. The decisions that they help to mould are usually of a high standard and command the respect of appellate bodies.
18. This court is indebted to all counsel and their supporting teams for the help which they have given to this court. Neither leading counsel appeared in the ET. The Claimants have only had the benefit of leading counsel in this court.

The appeal

The issues

19. There are two main issues on the appeal. There is no obvious answer to either of them :-

(1) Did the ET err in law in finding that the Claimants had established a prima facie case of indirect sex discrimination in pay and that the Audit Commission had not rebutted it? If not,

(2) Did the ET err in law in finding that the Audit Commission had failed to establish that the pay discrimination was objectively justified as being for a legitimate aim pursued by proportionate means?

20. The case for the Audit Commission, which is an independent public body responsible for ensuring that public money is spent efficiently in various areas, including housing, is that the difference in pay (a) was one of a genuine material factor within s.1(3) of the 1970 Act with no "taint of sex" (that strange expression is more redolent of an advocate's rhetoric than of dispassionate legal analysis, but, as it has become accepted usage, I will continue to use it); and, if discriminatory, (b) was justified by the legitimate aim of a pay protection policy that was proportionate.
21. The case for the Claimants is that this is an instance of (a) indirect sex discrimination resulting from a significant pay disparity tainted by sex and therefore not attributable to a genuine material factor within s.1(3); and (b) no legitimate aim proportionately pursued.

Outcomes of ET and EAT judgments

22. In the ET the Claimants succeeded for reasons set out in the judgment sent to the parties on 3 December 2010. The ET found that there was indirect sex discrimination in the matter of pay. It could not find any objective justification for it.
23. By order dated 18 March 2011 the EAT allowed the Audit Commission's appeal on both issues and dismissed the claims.
24. Which tribunal got it right? That is what we have to decide. Unusually, the EAT (Underhill P. presiding) granted the Claimants' application for permission to appeal on 11 April 2011, saying:-

"Although we were reasonably confident about our decision, the points are important to the parties and possibly more generally; and we are not prepared to say that the contrary is unarguable."

25. Of one thing I am reasonably confident: there is no satisfactory alternative to setting out the facts and the law yet again, with summaries of the judgments of the ET and the EAT and of the comprehensive submissions of the parties on this appeal. I will attempt to keep everything as simple as possible and to avoid legal gobbledegook so that those affected by the decision can understand it without having to use an interpreter. The Rule of Law does not have much meaning for ordinary people, if a legal qualification is required before an area of law as basic as this can be understood by those for whom the law was made.

Factual background

26. There has never been any serious dispute in this case about the main facts or about the applicable law. The disagreement, as in most other indirect discrimination cases, is in how the general legislative provisions should be applied to the particular facts.
27. The law of the United Kingdom must comply with EC Treaty provisions and implement directives on equal pay. The legislation is of the wide-open-textured type, which the tribunals and courts have to fit to a wide range of individual situations. That brings with it more unpredictability, more litigation, more appeals, more waiting for finality and more legal costs. There is no escape from those consequences in equal pay litigation. In the long-term interests of the community and of continuing relationships that are more likely to flourish in compromise than in conflict, perhaps, as I have already suggested, there should be more give and take on both sides: it is not always necessary to vindicate rights in a tribunal or in a court in order to get justice.

Pre-1 October 2007:IIOs and SIIOs

28. The events go back to 2004 when the Audit Commission took a transfer of the functions of the Housing Corporation in the inspection of social housing.
29. From then down to 1 October 2007, which is the key date in this case, there were two kinds of administrative support jobs in the Audit Commission's Housing Inspectorate: Inspection and Information Officer (IIO) and Senior Inspection and Information Officer (SIIO). The pay grade (FL4) was the same. Within the pay grade there was a pay scale of 23 incremental points from 22 to 45. Progression up the grade depended on a combination of the entry point, length of service and performance history. The difference between IIOs and SIIOs was that the job of SIIO had a higher level of responsibility. It was the more senior role. IIO was the more junior role.
30. The position at the end of September 2007 was that there were 11 IIOs. All of them were women. Some of them had been transferred from the Housing Corporation, where they had been described as "inspection assistants." On that transfer they were placed at a point in the Audit Commission's pay scale that preserved their then current pay.
31. There were 5 SIIOs: 3 of them were men working full time, 2 of them were women job sharing. All of the SIIOs had been transferred from the Housing Corporation, where they had been described as "programme managers" and had long service. On the transfer to the Audit Commission they were placed at a point in the scale that protected their existing pay. They owed their position at the top of the scale to the point that they had reached in the Housing Corporation.

Post-1 October 2007:amalgamation of roles and ISOs

32. With effect from 1 October 2007 structural changes were made by amalgamation of roles into a new administrative grade. Like other organisations the Audit Commission faced changing demands. To meet them, changes were made in the job structure in order to reduce the numbers employed in a particular operation. That was done by amalgamating the roles of employees so that the job of those performing a senior role was amalgamated with the job of those performing a junior role. In this process the organisation wants to retain the best of its existing staff for the amalgamated jobs and so it invites applications for the amalgamated jobs from those in both the junior and senior roles and appoints those it considers best.
33. A new administrative job of Inspection Support Officer (ISO) replaced the old jobs of IIO and SIIO. 11 ISO positions were created. The amalgamated ISO role was substantially closer to the IIO role than the SIIO role, but it incorporated elements of the SIIO role. The nine women, who had all been IIOs, are the Claimants. Two men, who had been SIIOs, were appointed ISOs. They are the comparators.
34. The overall pay grade was still FL4. The ISO position was placed towards the lower end of the scale. The Audit Commission's policy in the case of employees being moved to another job in the same grade was that employees should retain their previous position on the incremental scale, bringing with them the level of earnings, which incorporated disparities.

Summary

35. A summary of the main facts and points not in serious dispute will narrow and clarify the areas of dispute:-

(1) The nine Claimants, who are all women, and the two comparators, who are both men, were employed on "like work" within the meaning of s. 1(2)(a) of the 1970 Act from 1 October 2007 to the date of the presentation of the claims to the ET in 2008. The equal pay claims were confined to the period from 1 October 2007.

(2) The Claimants and their comparators constituted the appropriate pool for comparison for the purposes of determining the question whether there had been prima facie indirect sex discrimination as between them.

(3) The comparators were paid more than the Claimants were paid. As at 1 October 2007 the salary of the highest paid Claimant was £31,577 and the salary of the lower paid comparator was £36,663. The salary of the lowest paid claimant was £26,332 and the salary of the higher paid comparator was £37,506.

(4) The mere existence of a difference in pay between the female Claimants and the male comparators does not, without more, prove discrimination or shift the burden to the Audit Commission to provide objective justification. The ET, as the fact finding tribunal, has first to determine whether there is a prima facie case of indirect sex discrimination to justify.

(5) The explanation for the disparity in pay was the amalgamation of two jobs at two different seniority levels: SIIO being the senior job, and IIO being the junior job were amalgamated into a single post of ISO. That explanation was genuine, and was not a sham or pretence. The explanation for the disparity was also to be found in the application of the Audit Commission's practice of maintaining an employee's existing pay point in such circumstances.

(6) The Claimants were all former IIOs. The comparators were both former SIIOs. There was an overlap of about 90% between the former IIO role and the ISO role. The SIIO role was much less well-matched to the new ISO role and the former IIOs had to spend considerable amounts of time training the comparators to do large parts of the ISO job.

(7) The Audit Commission had been alerted both by the Claimants and by one of its Human Resources advisers, Ms Poyntz, that the inclusion of both the IIOs and the SIIOs in the pool for appointment to the ISO amalgamated position might lead to an unlawful inequality in pay.

(8) There was a gender pay gap of 17.2 % between the average pay of men and women within the Audit Commission, which had been highlighted in an equal pay review in 2007. At the grade of the Claimants and their comparators, women were, on average, earning 8.3% less than men.

(9) As the law presently stands the Claimants bear the burden of showing prima facie indirect sex discrimination, but it is for the Audit Commission both (a) to show that the disparity in pay is due to a genuine material factor that is not the difference of sex and (b) to justify any prima facie sex discrimination.

36. I would mention two other aspects of the case, as lack of clarity about them could lead to serious confusion.
37. In the first place the Claimants' case is not one of direct sex discrimination. They are not saying that the reason why they were paid less was that they were women. Their case is one of alleged indirect sex discrimination since 1 October 2007 arising from disparate adverse impact on the pay of the Claimants effectively caused by the combination of the amalgamation of roles into the ISO role and the application of the Commission's pay protection policy.
38. In the second place, and as emphasised in the submissions of the Audit Commission, the differences in pay between the Claimants and the comparators prior to 1 October 2007 were not the result of discriminatory factors. (It is agreed that the ISOs do "like work"; it is not agreed or found that the IIOs and the SIIOs did "like work.")

The law

39. The key provision of the 1970 Act is in s.1 ("Requirement of equal treatment for men and women in same employment"). Under s.1(1) an equality clause is deemed to be included in a contract of employment. Under s.1(2) the effect of the inclusion of the equality clause is that, in the case of like work, the contract is treated as so modified that the pay of the claimant is equalised with that of the comparator, unless the employer's defence in subsection (3) applies. That is the key provision in this case.
40. The defence supplied under subsection (3) is commonly called "the GMF defence"(Genuine Material Factor) and is relied on by the Audit Commission:-

"(3) An equality clause falling within subsection (2)(a),(b) or (c) above 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-

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) [immaterial]"

41. The submissions on the GMF defence have highlighted a number of points in the authorities.
42. First and foremost, an explanation from the employer for a difference in pay, which genuinely explains the difference and is not "the difference of sex", dispenses with the need for objective justification: there is simply no sex discrimination in the pay which has to be justified: Glasgow CC v. Marshall [2000] ICR 196; Strathclyde Regional Council v. Wallace [1998] ICR 205.
43. Secondly, Lord Nicholls in Glasgow CC at pp 202F-203C, in a passage of remarkable clarity and concision, summarised the scheme of the 1970 Act :-

"…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." The 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."

44. Thus undisputed inequality in pay between the Claimants and the comparators gives rise to a rebuttable presumption of discrimination on the grounds of sex. It is then for the Audit Commission to rebut that by the GMF defence showing that the inequality in pay is due to a genuine material factor which is not a difference of sex. If that defence is established, that is the end of the case and the claims are dismissed.
45. Thirdly, whether a difference in pay has a "disparately adverse impact on women" referred to by Lord Nicholls so as to be evidence of the presence of sex discrimination is "primarily a question of fact": Glasgow CC at 205C; Grundy v. British Airways PLC [2008] IRLR 74 at [31]. There is no statutory provision prescribing how an ET should investigate whether such a prima facie case of indirect sex discrimination has arisen. When indirect discrimination is relied on statistics of gender composition of the relevant groups used need to be examined. It is for the ET, as the tribunal of fact, to make the relevant assessment: Home Office v. Bailey [2005] ICR 1057 at [20]-[21].
46. Fourthly, the concept of indirect discrimination under the 1970 Act, read together with European equal pay legislation and case law, is broader than that which applies under the Sex Discrimination Act 1975, as amended: Ministry of Defence v. Armstrong [2004] IRLR 672 at [36] and at [46] where Mrs Justice Cox explained the operation of the "causative link" test applied by s.1(3) in an oft-cited passage:-

"46. …the principles to be applied in determining the s.1(3) defence, in our judgment, involve the tribunal focusing on substance, rather than the form and on the result, rather than the route taken to arrive at it. We agree with Mr Linden's submission that, in approaching these issues, technicalities should be eschewed. The fundamental question for the tribunal is whether there is a causative link between the applicant's sex and the fact that she is paid less than the true value of her job as reflected in the pay of her named comparator. This link may be established in a variety of different ways depending on the facts of the case. It may arise, for example, as a result of job segregation or from pay structures or pay practices which disadvantage women because they are likely to have shorter service or to work less hours than men, due to historical discrimination or disadvantage, or because of the traditional social role of women and their family responsibilities."

47. Fifthly, the tribunals must not fall into the trap of thinking that, provided that there is an explanation for the difference in pay that is not directly discriminatory, that will suffice to show that the pay practice in question is not sex-tainted: Gibson v. Sheffield CC [2010] ICR 708 at [58]. As I have already emphasised this case is not being brought as one of direct discrimination.
48. Sixthly, in an indirect discrimination case it is necessary for the ET to examine with great care the contention of an employer that a particular disadvantage has nothing to do with the sex of the employee. The explanation must not only be genuine: the ET must also be satisfied that the difference in treatment is not in any way related to the difference of sex. Where the disadvantaged group is heavily dominated by women and the group of advantaged comparators is heavily dominated by men the inference of "sex taint" will be readily drawn: it will be difficult for the employer to prove its absence. Where the work of the disadvantaged group has historically been done by women and the work of the advantaged group has historically been done by men there is a basis for inferring a legacy of sex tainted attitudes, such as that women do not need to earn as much as men. Those are not matters of direct discrimination. The issue is whether the disadvantage complained about is indirectly causally linked to gender: see Gibson at [68]-[71].
49. Seventhly, as already indicated but is worth repeating, the mere existence of a difference in pay between Claimants and the comparators does not amount to indirect sex discrimination that requires justification: Redcar & Cleveland BC v. Bainbridge [2009] ICR 133; Gibson v. Sheffield CC [2010] ICR 708; and Bury MBC v. Hamilton [2010] ICR 665 at[17]. The pay difference gives rise to the rebuttable presumption to which the employer can respond with the GMF defence but only if there is no taint of sex in the factors on which he relies. Where the pay practice has produced an adverse impact on women over a long period and where the statistics are convincing, because they are significant, not fortuitous and not short term, it will generally be difficult for the employer to show that the adverse impact had nothing to do with sex: see Gibson at [71]; Redcar at [59].
50. Eighthly, indirect discrimination can arise in different ways. It can arise from the application of an apparently neutral express or implied provision, criterion or practice (PCP) putting persons of one sex in general or the claimant at a particular disadvantage, which could not be objectively justified. It can also arise in a case of the Enderby kind (Enderby v. Frenchay HA [1994] ICR 112), in which women are disproportionately disadvantaged, such as where the disparity of pay cannot be explained by the application of a PCP, but the statistics show significant differences in pay in two jobs of like work or of equal value and one job is done almost exclusively by women and the other job is done predominantly by men: see Bury MBC v. Hamilton [2011] ICR 655 at [14]-[17]; and Development Skills Scotland v. Buchanan [2011] EqLR 955 at [26] and [43].
51. Finally, the distinction between the PCP cases and the Enderby kind of case are not always capable of clear definition or consistent practical application and may be one of form rather than substance. Usually the disparity of pay between two work groups will reflect the fact that they do different work. There may be features of the work of the advantaged work group, which could be elevated into a requirement or condition. There is no justification for imposing a higher threshold for satisfying the test of prima facie discrimination in the disparity of pay case than in the PCP case: Home Office v. Bailey at [29]-[30].
52. The Audit Commission denies that there is prima facie indirect sex discrimination of any kind. It contends that it was for the Claimants to prove prima facie indirect discrimination and the validity and import of statistics relied on. It accepts that it is for it to establish the GMF defence without taint of sex and, if that fails, it is for it to establish objective justification for the discriminatory difference in pay by showing that the difference is to achieve a legitimate aim and that the means used are proportionate.
53. In that connection the Audit Commission relies on its pay protection policy. There is no dispute that it may be legitimate, for example, in relation to job evaluation schemes, to have pay protection arrangements that implement proportionate steps for the protection of employees against reductions in pay and to keep pay static on the basis that others will catch up as a result of annual pay increases not awarded to the higher paid group. It is recognised that it may not always be possible to achieve that result: Middlesbrough BC v. Surtees [2007] ICR 1644 at [10].

ET judgment

54. The ET hearing lasted for five days. In its detailed judgment it made findings of fact and set out a summary of the law. The facts were not appealed and the self-directions of law in the summary in [6] to [11] were not criticised on appeal. The arguments on appeal focused on [12],which set out in nine sub-paragraphs the conclusions it had reached on the issues.

A. Indirect sex discrimination

55. The first issue before the ET under s.1(3) of the 1970 Act was whether the variation in the pay of the Claimants and their comparators was "genuinely due to a material factor which was not the difference of sex."
56. As mentioned above it was common ground that there was a genuine and material explanation for the difference in pay i.e. the amalgamation of the two jobs SIIO and IIO into the single ISO post and the application of the practice of the Audit Commission of maintaining an employee's existing pay point. That was the "effective cause of the pay differential" between the Claimants and their comparators and it was the genuine reason, not a sham.
57. The first point of dispute was whether there was prima facie indirect sex discrimination, because the factors of amalgamation and pay protection combined to have a disparate adverse effect on women and, in particular, on the Claimants, or were otherwise sex tainted/gender related so as to deprive the Commission of the benefit of the GMF defence. If so, the second point of dispute was whether the Commission had established that the indirect sex discrimination was objectively justified.
58. In finding for the Claimants on the first point the ET held that, when compared with their male comparators, the Claimants were disadvantaged by the amalgamation and by the Audit Commission's pay protection practice. The Audit Commission failed to prove that there was no taint of sex discrimination in the pay disparity. The statistics were sufficient, stark, not fortuitous or short term and they were gender related. Further, the ET found that the IIO role might be described as "women's work" in the sense that the statistics demonstrated that it had, for a number of years, been an administrative role carried out exclusively by women, being a role that fitted in with family responsibilities. The agreed pool, which was not too small to be statistically insignificant, and the statistics clearly indicated disparate impact. The Claimants do not contend and do not seek to support the ET's conclusion that the work carried out by the SIIOs was "men's work." They say that it was not part of the ET's essential reasoning or of the Claimants case.
59. Thus, in the result, the ET found that the Claimants, on comparison with the male comparators, were at a relative or comparative particular disadvantage by the amalgamation and the application of the pay protection policy. It also found that the Commission failed to prove that there was no taint of sex discrimination in the pay disparity.
60. The ET considered the pre-1 October 2007 position, pointing out that in 2004 there had been a justifiable distinction between the IIO role and the SIIO role: they were doing different work. But in October 2007 there was a deliberate decision to include the SIIOs in the ring fence for the ISO role, which was very largely (90%) an IIO role. That decision created a disparity between men and women doing the same work: it was not fortuitous. It was a deliberate decision taken by the Commission when it was well aware of the suggestion of its own HR department that such a policy had equal pay implications.
61. As the Commission had not rebutted the taint of sex, the ET found indirect sex discrimination and therefore moved to consideration of the defence of objective justification.

B. Objective justification

62. On the second point, the ET held that the Audit Commission had not made out the defence of objective justification by proving that amalgamation of the two roles and protecting existing pay points was a proportionate means of achieving a legitimate end.

Legitimate aim

63. The ET rejected the Audit Commission's contentions that it had legitimate aims of preventing the Claimants or their comparators suffering a reduction in pay and retaining their skills and experience. The ET found that the consequence of pay protection was that there were 11 employees carrying out "like work"; all of the women received considerably lower pay than the men were paid; the disparity would widen over time; and the Commission went ahead, despite the warning from Mrs Poyntz.
64. The ET pointed out that protecting the Claimants against a reduction in pay did not remedy the disparate adverse impact on them when their position was compared to that of the comparators. Maintaining everybody's existing pay had a discriminatory effect. There was nothing to prevent the Claimants' pay from being increased to the same level as that of the comparators.
65. As regards the retention of skills and experience, the ET found that there were less discriminatory ways of achieving those aims. The ET weighed up the aim advanced against the discriminatory effects and said that the more serious the disparate impact the more cogent must be the objective justification. It held that the disparity was very significant indeed, being at least £5,000 a year in some cases and more than £10,000 a year in others.
66. In its consideration of proportionality, the ET considered other ways of carrying out the restructure. The ET considered that the discriminatory effect of amalgamation and pay protection could have been eliminated by assimilation of the IIOs into the ISO post. The Commission ought to have looked at the advantages of assimilation more closely. Alternatively, the discriminatory effects may have been reduced by the red circling of the pay of those on higher pay points. The ET concluded that the Commission had not considered whether its decision was justified at the time, but only "after the event." It had failed to anticipate it.
67. On the proportionality issue the Commission's arguments had to be particularly strong, if justification was to be made out, in view of the extent of the disparity and its potential increase over time. They were not strong in the light of the reasonable alternatives available to it.

EAT judgment

Jurisdiction of EAT

68. The EAT only had jurisdiction to set aside the ET's decision if a question of law arose from it, such as a misdirection of law, or a misapplication of the law to the facts agreed or found, or a perverse decision that no reasonable ET properly directing itself could have reached. In considering the EAT judgment this court must examine it for a statement of the question of law which it considered arose from the ET judgment and what error of law it considered that the ET had made in reaching its decision. It was not open to the EAT to disregard the ET's findings of fact or to set its decision aside simply on the ground that it did not agree with it.

A. Indirect sex discrimination

69. The EAT stated that it was common ground that the "difference between the pay of the IIOs and the SIIOs pre-amalgamation was not discriminatory" and that the ET had gone on to find that the differential was indirectly discriminatory and that objective justification had not been proved. In allowing the appeal on both points the EAT discussed the two distinct types of indirect discrimination.
70. First, the type of discrimination resulting from the application of a neutral PCP putting persons of one sex at a particular disadvantage compared with persons of the other sex and which is not objectively justified by a legitimate aim pursued by proportionate means.
71. Secondly, the type of discrimination where significant statistics demonstrate a disparity in pay between predominantly female and predominantly male groups of workers carrying out like work or work of equal value and where such prima facie indirect discrimination is not justified. That is called the Enderby type of discrimination after the leading case of that name in the Court of Justice.
72. The EAT concluded that the ET was "not entitled" to make a finding of prima facie indirect discrimination on either the PCP basis or the Enderby basis. The EAT considered that the differential in the present case was the result of the application of a PCP rather than being an Enderby type of case, which it found confused the analysis; and that the PCP was the Audit Commission's pay protection policy which applied equally to and benefited both men and women. The EAT considered it necessary to see whether the IIO's role was "women's work" and whether the SIIO role was "men's work" and concluded that the latter was not men's work, there being a "very frail statistical base". The EAT then focused on whether women had benefited from the pay protection policy and inquired about "the reason why" men should be proportionately advantaged by such a policy and held that there was no inherent reason why they should be advantaged.
73. The EAT concluded in a key passage that the immediate cause of the disparity in pay was nothing to do with gender:-

"24. In neither respect therefore do we believe that the Tribunal was entitled to find a prima facie case of indirect discrimination. Standing back from the detailed analysis, that seems to us the right result. It is not difficult to see why at first blush a situation where two men are paid substantially more than eleven women doing work of equal value should have set equal pay alarm bells ringing, and we are not surprised that Ms Poyntz gave the advice that she did. But it is in fact perfectly clear that the immediate cause of the disparity has nothing to do with gender but simply reflects the fact that all the employees affected by the amalgamation came into a new role preserving their previous remuneration. If there was reason to believe that those differences in previous remuneration were themselves discriminatory, that would be a different matter; but there is not. No such finding was made by the Tribunal and even if it were sought to press into service in relation to the pre-October 2007 period its conclusions apparently based on an Enderby analysis, notwithstanding what we say in para.18, those conclusions are in our view flawed:see para 22 above…"

74. It followed that the issue of objective justification did not arise. The EAT addressed it, in case they were wrong on the prima facie discrimination issue.

B. Objective justification

75. On the issue of objective justification the EAT concluded that protecting the pay of employees affected by restructuring was, in principle, a legitimate aim and that it was a legitimate aim in this case. It found the discussion in the ET judgment "not entirely easy to follow" [27] and concluded that, if the ET intended to find that the Commission had not established that its pay protection policy reflected a legitimate aim, "it was plainly wrong." That would seem to be in substance a conclusion that the ET's decision on this point was perverse.
76. The EAT went on to say that the pay protection policy was legitimate both as a simple matter of fairness to the employees concerned and also because, if their pay were reduced, it would be difficult to retain them and the employer would be likely to be prejudiced by losing their services. The EAT distinguished the local authority "bonus cases", in which the payments which it was sought to protect had been discriminatorily withheld from the claimants. In this case, the EAT said, there was nothing discriminatory about the differential which the Commission sought to protect by the practice ([28]).
77. As for proportionality, the EAT said that it was hard to see why the practice of the pay protection policy, which did not incorporate past discrimination, was not a proportionate way of achieving the aim. The extent of protection matched precisely the loss which the employee would otherwise suffer and had nothing to do with being a man and a woman. The differences in pay were not due to the difference in sex, but to the different rates which they were non-discriminatorily paid before they were involuntarily moved into the ISO post (see [30]).
78. The suggested alternatives of assimilation and red circling were then considered by the EAT as other proportionate means of achieving the legitimate purpose. It rejected the ET's analysis, saying that whether a disparity could be justified could not be tested by asking whether some different step might have been taken, which would have avoided the disparity.
79. As for assimilation, the EAT accepted that if the Commission had decided simply to give the new ISO roles to the existing IIOs, leaving the SIIOs redundant, no equal pay problem would have arisen as between the ISOs because they would all have been women unless and until a man was appointed an ISO. The EAT did not see how that was relevant to the question whether the resulting disparity could be justified.
80. As for red circling, the EAT recognised that it was good practice and prudent to phase out pay protection arrangements over time and agreed that the arrangement proposed would, in principle, over the years have reduced the extent of the disparity between the Claimants' pay and the comparators' pay, but it would have been very slow and the fact that the option was available did not mean that it was disproportionate for the Commission to prefer a more complete and straightforward form of protection. The EAT repeated that this was a case where the protected rate of pay was not in itself discriminatory in any way and that the position would have been different if the differential protected was itself the product of past discrimination. An arrangement that did not incorporate phasing out was capable of justification.
81. In discussing the assimilation point, the EAT also considered and accepted two arguments advanced by the Commission to it that it was a justifiable objective to want to choose the highest calibre people from both the jobs affected and that the exclusion of the SIIOs in order to avoid equal pay issues would have constituted unlawful direct sex discrimination against the ex-SIIOs, who would have been deprived of the chance of a job as an ISO on the basis that some of them were men.

Claimants' submissions

82. In her excellent submissions Miss Jane McNeill QC submitted that the ET's conclusions on both points were correct, raised no question of law which justified setting its decision aside and should be re-instated. The error of law was on the part of the EAT in overturning the findings of the ET in circumstances where there was no self misdirection of law by the ET and it was not said that the ET's conclusions were perverse.

A. Prima facie indirect sex discrimination

83. Miss McNeill submitted that, whether the Claimants' case was approached as a PCP type case or an Enderby type case, the key question was whether there was a "causative link" between (a) the fact that the Claimants were women and (b) the fact that they were paid less than the male comparators.
84. The ET proceeded on the basis that it was for the Claimants to make out, by reference to a gender-based comparison, that the amalgamation of the IIOs and the SIIOs and the practice for the protection of pay had a disproportionate impact on the women in the pool and/or was gender related. On that point Miss McNeill expressly reserved the Claimants' position that, if the case went further, they wished to argue that the law as laid down in Nelson v. Carillion Services Ltd [2003] ICR 1256 at [26] and [36] was wrong in requiring that, in an indirect discrimination case, the burden of proving disproportionate adverse impact always lies on the claimant and that, unless and until that is established, the ET could not, even without explanation from the employer, conclude that there had been any unlawful discrimination. She would wish to contend that that the Audit Commission bore all the burden of proving all aspects of the GMF defence in circumstances where it had been agreed that the Claimants were paid less than their male comparators.
85. In the present state of the law it was for the Audit Commission to prove that the disparity in pay was not attributable to a difference in gender, and, if not, to establish that the difference in pay was justified: Armstrong v. Newcastle upon Tyne NHS Hospital Trust [2006] IRLR 124 at [32].
86. Taking that approach Miss McNeill submitted that the Audit Commission had failed to prove that the disparity in pay was not tainted by sex. She highlighted a number of points in the ET's findings as showing that there was a clear causative link between the fact that the Claimants were paid less than the comparators and that they had all been employed in the "women's work" role of IIO.
87. First, the statistics, which were stark and not fortuitous or short term, were that the Claimants were paid at a much lower rate than the male comparators, one Claimant being paid as much as 30% less.
88. Secondly, those statistics were very strong evidence of prima facie indirect discrimination and made it very difficult for the Audit Commission to show that the pay differential was not related to the difference in sex: see Redcar and Cleveland BC v. Bainbridge [2009] ICR 133 at [59].
89. Thirdly, the ET found that the IIO role had, since 2004, been populated entirely by female employees and was "women's work" that fitted in with family responsibilities.

Errors in EAT judgment

90. Turning to the EAT's decision to overturn the ET on this point, Miss McNeill said that it was the EAT judgment that was flawed by errors.
91. First, the EAT erred in stating that the statistical evidence was "very frail."
92. Secondly, the EAT considered it relevant that the practice of pay protection applied to all employees equally. But, as has been shown in many of the authorities, that does not prevent the application of the PCP from having a disparate impact and from being discriminatory.
93. Thirdly, the EAT considered that the application of the practice benefited the Claimants, because it protected their pay against reduction. But that is not relevant to whether the practice was discriminatory. The question is not whether the Claimants were advantaged, but whether there was a comparative disadvantage when their position was compared with that of the comparators. There was a comparative disadvantage to the Claimants on the facts of this case.
94. Fourthly, the EAT concluded that there was "no reason why" men should be disproportionately advantaged by the pay protection practice. The "reason why" test relied on by the EAT was irrelevant to a case of indirect discrimination, as distinct from cases of direct discrimination. The EAT had introduced a need for the Claimants to establish "a gender-related reason" why they received less pay than the comparators, when the correct requirement was that it was for the Audit Commission to show that there was no causative link between gender and disparity in pay.
95. Fifthly, the EAT had placed considerable importance on the fact that it was not alleged that the disparity in the remuneration of the posts pre-1 October 2007 was indirectly discriminatory and that "all the employees affected by the amalgamation came into the new role preserving their previous remuneration": see [25]. That statement involved an erroneous legal analysis. It did not follow from the absence of indirect discrimination before restructure by amalgamation, when the employees were not engaged on "like work", that there would be no discrimination after restructure when their duties and responsibilities were adjusted and they were employed on "like work".

B. Objective justification

96. On the second issue in the case Miss McNeill submitted that ET had made no error of law in its analysis of objective justification or in the conclusion reached by it. The ET correctly directed itself on the principles to be applied.
97. Miss McNeill accepts that pay protection may amount to a legitimate aim, but says that there are different kinds of pay protection, some easier to justify than others. The ET was entitled to find that there was no legitimate aim in this case.
98. As for proportionality, the findings of the ET proceeded on the basis that there was a legitimate aim and consisted of consideration of alternative approaches to test proportionality. The ET reached a conclusion that was open to it in the context of inevitable, long term and widening discriminatory disparity in pay.

Errors in judgment of EAT

99. Miss McNeill contended that the EAT should not have overturned the ET's decision in the absence of an error of approach or of a decision that was perverse. The EAT had wrongly substituted its own view for that of the ET which it was not entitled to do: Bainbridge at [169] and [170].
100. Instead of considering whether the pay protection practice was a legitimate aim in this case, it had considered the position of such a practice in principle and made a number of errors in its approach.
101. First, the EAT repeatedly stated that there was nothing discriminatory about the pay differential that the Audit Commission was seeking to protect. However, the necessary premise of considering objective justification was that the pay differential was prima facie indirectly discriminatory.
102. Secondly, the EAT was influenced by the absence of any case for historic indirect discrimination pre-1 October 2007. However, that was irrelevant to the analysis of objective justification post-1 October 2007 when the jobs were amalgamated and the Claimants and the comparators were engaged on "like work". It does not follow from an absence of historic discrimination in relation to pay and an absence of discrimination in relation to selection that a case of justification of the pay disparity was made out.
103. Thirdly, the EAT gave no consideration to the finding of the ET that the Audit Commission's justification was all "after the event". But it did give consideration to arguments on assimilation that were not made before the ET.

Audit Commission's submissions

104. Mr Christopher Jeans QC submitted that the ET erred in law on both issues for the reasons given by the EAT for setting the ET decision aside.
105. The ET wrongly held that the higher pay of the comparators had to be justified. The true position was that the reason for the difference in pay was not the difference of sex and there was no case of adverse impact by reference to "men's work", or on the basis of statistics that required justification.
106. On justification the ET misdirected itself as to legitimate aim and proportionality and reached an unsustainable conclusion.

A. Prima facie indirect sex discrimination

107. Mr Jeans submitted that there were errors of law in the ET judgment on the need for justification.
108. The reason for the differential in pay was that the comparators had previously been employed in a more senior SIIO role. That was a genuine reason, not a gender reason. The statistics did not establish a case requiring justification, whether the case was put on the Enderby basis, or on the PCP basis, or some other independent basis. It was absurd to describe the SIIO's role as "men's work." The numbers were in themselves wholly insufficient for the purposes of reaching a statistical conclusion. The ET erred in law in treating the statistics as significant and valid, in failing to disregard them as fortuitous and in categorising the work of the SIIOs as male work in contradistinction to the work of the IIOs.
109. The material placed before the ET was "wholly insufficient" to discharge the onus on the Claimants to show that there was a statistically demonstrated sex bias requiring justification.

B. Objective justification

110. Mr Jeans contended that the case for objective justification was made out for the reasons given by the EAT in its judgment. Pay protection was a legitimate aim; the pay protected was not discriminatory; the Commission was entitled to select the best people for the ISO role; and there had been no discrimination in the selection. The criticism of "after the event" justification was misconceived, as it is permissible, though it might carry less weight.
111. As for the alternatives of assimilation and red circling, he submitted that assimilation would have contradicted the Commission's very aim of appointing the best people to the new role of IIO; and the suggestion of red circling, as a marking time arrangement, would not serve the legitimate policy of keeping the best people and could involve the Commission in liability for breaches of contract and discrimination.

Discussion and conclusions

112. The main difficulty that I have had with this appeal is in identifying the error of law in the judgment of the ET that entitled the EAT (a) to set aside the ET's decision in favour of the Claimants on the issue of indirect sex discrimination or, if wrong on that, (b) to substitute a decision on objective justification in favour of the Audit Commission.
113. The EAT did not say in terms that the decision of the ET on either issue was perverse. The ET identified the relevant law. It gave itself a proper self direction, which was not criticised by the EAT or in the submissions to this court. The only basis left on which the EAT could interfere with the ET decision was misapplication of the law to the facts in a way that it was "not entitled to" on the issue of indirect sex discrimination and in the conclusion that the ET was "plainly wrong" on the justification issue.

ET decision: question of law

114. Many of the submissions naturally concentrated on the judgment of the EAT from which the Claimants' appeal is brought. However, the jurisdiction of this court, like that of the EAT, ultimately depends on whether a question of law arises from the ET decision rather than from the EAT decision. If no question of law arose from the ET decision, the EAT had no jurisdiction to set it aside and this court would have to allow the appeal, set aside the order of the EAT and restore the order of the ET.
115. There have been criticisms of that approach on the ground, for example, that it sidelines or demotes the intervening judgment of the specialist appeal tribunal. In my view, there is no basis for that criticism. This court places great importance on the EAT's judgments and very frequently agrees with their conclusions on whether or not a question of law arose from the ET decision and with the EAT's analysis of the issues arising on the first level of appeal.
116. The important point is that the issues which this court has to decide should not be skewed by concentrating too much on the EAT judgment so as to divert attention away from the foundation judgment of the ET, in which the findings of fact and initial assessments were made.

A. Prima facie indirect discrimination:

117. I have been persuaded by Miss McNeill QC that no question of law arises from the decision of the ET on the issue of indirect sex discrimination. On the facts and matters before it the ET concluded that there was a prima facie case of indirect sex discrimination in the form of a disproportionate adverse impact on the Claimants by reason of the amalgamation of the IIO and SIIO roles and the application of the Commission's pay protection policy.
118. Although, after 1 October 2007 the Claimants and the comparators were engaged on "like work" as ISOs, the male comparators were paid between £5,000 and £10,000 per annum more than the Claimants. Add to that the findings of the ET, which they were entitled to make on the material before them, that the gender statistics were stark and not fortuitous or short term. Add to that the fact that the statistics reflected the situation prior to 1 October 2007 that all the Claimants had been employed in a position as IIOs, which had been undertaken exclusively by women for a number of years and can fairly be called "women's work."
119. The ET did not arrive at its decision by erroneously relying simply on disparity of pay between the Claimants and the comparators as in itself proof of indirect sex discrimination. Nor do I think that there are grounds for finding an error of law in the way that the ET approached the matter of indirect discrimination via the application of the PCP or Enderby. The relevant unchallenged findings were based on relevant evidence and speak for themselves.
120. As for the criticisms of the ET decision by the EAT and by Mr Jeans, I do not think that the ET's conclusion on indirect discrimination was flawed by the fact that the Commission applied the PCP in the form of the pay protection policy equally to all employees: the true question, which was correctly appreciated by the ET, was whether the application of it had a disparate adverse impact on the Claimants, which the ET found it did. The fact that the Claimants benefited from the application of the policy in the protection of their pay was beside the point, when considering whether the application of the policy has a disproportionate adverse impact on them as compared with men.
121. As for the Enderby approach and the statistics, the EAT concluded and Mr Jeans submitted that the statistics did not establish a prima facie case of indirect discrimination, because the make-up of the comparator group was not the result of indirect sex discrimination: SIIO work pre-1 October 2007 was not stereotyped as "men's work", but rather the fortuitous result of a non-discriminatory selection process. However, I think that Miss McNeill is correct in saying that that is not the point. The Claimants do not rely on establishing that the SIIO was "men's work" but on whether the IIO was "women's work" which the ET was entitled to find that it was on the evidence.
122. The EAT considered that it was relevant that there was no "reason why" the comparators should be disproportionately advantaged by the pay protection policy. However, I agree with Miss McNeill that it would not have been appropriate in an indirect discrimination case, as distinct from a direct discrimination case, for the ET to make that inquiry. The point was that the Claimants were disadvantaged by reason of having done lower paid women's work prior to the amalgamation.
123. That brings me to the important point of the EAT's reliance on the lack of any historic discrimination prior to 1 October 2007 and of any sex link to the difference in pay in that period. I agree that the cause of the difference in the make-up of the Claimants' group and the comparator group was not the result of historic sex discrimination. There was no historic discrimination, because the women and the men were not engaged on like work in the historic period pre-1 October 2007. But I am satisfied that the lack of historic discrimination is irrelevant. The claim is only in respect of the period post-1 October 2007. The Claimants do not rely on the continuation of historic discrimination. They rely on the combined effect of the amalgamation and the application of the pay protection policy after 1 October 2007 and the fact that, after that date, the Claimants and the comparators were doing "like work" but with a significant difference in pay.
124. The decisive point for me on the indirect discrimination issue is in appreciating the significance of the effect of the changes made on 1 October 2007. Before then there were differences in pay, but there was no discrimination, because the IIOs and the SIIOs were not employed on like work. If that had continued, the Claimants would have not had an equal pay claim. But it did not continue. After that date the Claimants and the comparators were all doing like work as ISOs, but for significantly disparate pay. The application of the pay protection policy to all the employees in that changed situation ensured that that difference in pay situation continued and even increased. There was, in those circumstances a rebuttable presumption of discrimination, which the Commission was unable to rebut by reliance on the GMF defence, because the presence of disparate adverse impact on women doing like work was itself evidence of the taint of sex.
125. In conclusion, there was no error of law in the ET's conclusion that this was a case of indirect sex discrimination and that the Commission had failed to establish the GMF defence because of the taint of sex in the form of the obvious disparate adverse impact on the Claimants.

B. Objective justification: the ET

126. The justification issue is more problematical. There are serious question marks over the treatment of it in both the judgments of the ET and the EAT.
127. The relevant legal principles are not in question. As was said by Lady Hale in Homer at [19], "the approach to the justification of what would otherwise be indirect discrimination is well settled. A provision, criterion or practice is justified if the employer can show that it is a proportionate means of achieving a legitimate end". The aim can "encompass a real need on the part of the employer's business." The issue of justification must be approached by the ET in a suitably structured way, which means that the right questions must be addressed: see [26].
128. In R(Elias) v.Secretary of State of Defence [2006] EWCA Civ 1293, [2006] 1 WLR 3213 at [151] it was said that:-

"…the objective of the measure in question must correspond to a real need and the means used must be appropriate with a view to achieving the objective and be necessary to that end. So it is necessary to weigh the need against the seriousness of the detriment to the disadvantaged group."

129. In Homer it was explained at [20] that "…it is not enough that a reasonable employer might think the criterion justified. The tribunal itself has to weigh the real needs of the undertaking against the discriminatory effects of the requirement."
130. At [22] in the same case Lady Hale added that "To be proportionate, a measure has to be both an appropriate means of achieving the legitimate aim and (reasonably) necessary in order to do so." "A measure may be appropriate to achieving the aim but go further than is (reasonably) necessary in order to do so and thus be disproportionate."[23]. The availability of non-discriminatory alternatives is relevant: see [25].
131. Miss McNeill has persuaded me that the EAT decision on objective justification cannot stand. Although the EAT made justifiable criticisms of the way in which the ET dealt with this issue, it then proceeded to substitute its own decision for that of the ET, because it did not agree with it.
132. In my judgment, there are flaws in the way in which the EAT itself approached the issue of objective justification. It appears to have done so on the basis that there was nothing discriminatory about the pay difference that the Commission were seeking to protect and that there was no discriminatory pay difference before 1 October 2007, but that ignores the fact that the stage of justification is only reached, if there is indirect discrimination which has to be justified. In this part of its judgment it seems to me, with the greatest respect, that the EAT was backtracking to the prior issue of indirect discrimination in pay and saying, contrary to the basis on which it was considering the whole question of objective justification, that there was no indirect discrimination in pay.
133. I have concluded that there is no alternative to remission of the justification issue to the same ET for reconsideration.
134. The case is close to the debatable line dividing those cases of equal pay in which the employer is liable from those in which there is no liability. I have hesitated for too long about which side of the line this appeal falls. That is both a tribute to the quality of the decisions at both levels and to the submissions from both sides. It is also a reflection of the difficulty of the case.
135. If I had been sitting in the ET, I might have reached a different decision and found for the Audit Commission. But I was not sitting in the ET nor was the EAT. I agree with the EAT that the ET have not approached the justification issue in the way required by the authorities, but I have not been persuaded that the EAT was entitled to reach the conclusion that the indirect sex discrimination was objectively justified.
136. It is necessary to explore the right questions in a suitably structured way: to re-consider the aim advanced by the Commission, to consider whether it was legitimate, to consider whether the means were proportionate and whether there were less discriminatory and more proportionate alternative ways of proceeding with the restructuring.
137. It follows that I would set aside the order of the EAT, allow this appeal on the indirect discrimination issue, and remit the case to the same ET on the issue of objective justification.
138. Postscript. Draft judgments were sent to Counsel with a view to handing them down on 26 July 2012. The court received and granted a request by counsel for the appellants to make further submissions on the issue of justification in the light of the case of Homer and the conclusions reached by Lewison LJ and Sir Mark Waller. Written submissions were received. They have been considered by the court. It was decided that it was unnecessary to hold an oral hearing, as none of us was persuaded to alter the conclusions we had reached on that issue. The issue does not merit more detailed discussion at this level than is already contained in our judgments. If further examination of the law on objective justification is required, it should be in the Supreme Court which decided Homer.

Lord Justice Lewison :


 


139. Mummery LJ has enormous experience of employment and discrimination cases. I do not. So it is with great diffidence that I explain why I respectfully disagree with him about the ultimate fate of this appeal.
140. I agree entirely that the issue for this court is whether the ET made an error of law sufficient to entitle the EAT to overturn the ET's decision and substitute its own. In addition to the facts that Mummery LJ has set out, I would like to add a few more:

a. At the date when the new post of ISO was created, the evaluation of the job placed it in the bottom third of pay scale F4. It attracted a score which placed it within points 22-30 on the pay scale.

b. Both the claimants and the comparators were transferred at their existing pay points, because the new job was on the same grade (F4) as their old ones. The claimants were between pay points 27 and 38 on the F4 grade; and the comparators were at pay points 44 and 45.

c. The respondent decided to award additional pay to those ISOs towards the bottom of the pay scale up to a minimum pay point of 32. This meant that four of the claimants received pay increases, while the remaining ISOs (both male and female) stayed on their existing pay points.

141. I agree with Mummery LJ that (whatever one may think about the reliability of the statistics on which the ET relied), there is no error of law in their conclusion that there was indirect discrimination.
142. It is the question of objective justification that is far more difficult. There were two aims on which the respondent relied as amounting to legitimate aims. The first was to stop the claimants and the comparators from suffering a reduction in pay. The second (allied to the first) was to prevent the loss of skills and experience. The ET appears to have accepted them as being aims; but to have rejected them as being legitimate. Although it is not entirely clear from their reasoning, they seem to have accepted the claimants' submission that "it cannot be right to have a legitimate aim that is contrary to the equal pay legislation". They added also that it appeared to them that the aim was contrary to the respondent's own gender and diversity policy. However, as the EAT correctly pointed out, whether the aims were contrary to the equal pay legislation and/or the respondent's gender and diversity policy was the very question at issue. The ET's characterisation of the aims begged the very question it was asked to decide. In my judgment, therefore, the ET did not ask itself the right question. I agree with the EAT that the ET was wrong in law in rejecting the respondent's aims as legitimate ones. As the EAT put it (§ 28):

"In our view if the Tribunal did indeed intend to find that the Commission had not established that its pay protection policy reflected a legitimate aim it was plainly wrong. In our judgment protecting the pay of employees affected by a re-structuring is plainly in principle a legitimate aim. It is legitimate both as a simple matter of fairness to the employees concerned and also because if their pay were reduced it would be difficult to retain them and the employer would be likely to be prejudiced by losing their services. That aim does not of course require any payment to employees who would not suffer a reduction, nor does it require any greater payment than is necessary to make up for the loss that that particular employee would suffer: it applies to the potential losers only, and only to the extent of their loss."

143. In Homer (to which Mummery LJ refers) Lady Hale said (§ 25):

""Grandfather clauses" preserving the existing status and seniority, with attendant benefits, of existing employees are not at all uncommon when salary structures are revised. So it is relevant to ask whether such a clause could have represented a more proportionate means of achieving the legitimate aims of the organisation."

144. This, to my mind, reinforces the conclusion that the EAT reached. What the respondent did in our case was to implement a grandfather clause. The respondent's policy in this respect applied equally to men and women. But the respondent in fact went further; because as mentioned those newly appointed ISOs whose pay was lower than pay point 32 had their pay increased. As it happens, they were all women. Moreover, the increase was to a level which the job itself did not warrant.
145. Despite having rejected the respondent's aims as legitimate, the ET went on to consider the question of proportionality. This was clearly a difficult task given that it did not think that the aims were legitimate. Under this head it asked itself whether there were less discriminatory means of achieving those aims. It is important to stress that the question here is whether the means under consideration would achieve the employer's legitimate aims; not whether the means would achieve different aims: Blackburn v Chief Constable of West Midlands [2009] IRLR 135 (§ 25).
146. The first possibility that the ET considered was appointing on an appropriate pay point, and then "red circling" the pay of those on higher pay points. By "red circling" the ET meant freezing the pay of those on higher pay points (i.e. not even awarding cost of living increases). On the facts, this would have meant freezing the pay of both the claimants (apart from one of them) and the comparators. There are a number of other problems with this suggestion:

a. The claimants' complaint was that there was an immediate pay differential on appointment to the new role in October 2007. The ET's suggestion would have preserved the pay differential, with the result that it was not a solution that would have dealt with the problem it perceived;

b. The ET seems to have contemplated that the pay freeze would have applied to the comparators, but not to the claimants. To have frozen the men's pay and not the women's might itself have amounted to discrimination;

c. The ET gave no consideration to the question whether a pay freeze for some employees would have enabled the respondent to achieve its legitimate aim of retaining skills and experience. In short the ET did not ask itself the right question.

147. The second possibility (which the ET ultimately preferred) was that of assimilating the jobs of IIO and ISO and making the former SIIOs redundant. In my judgment this suggestion was seriously (and legally) flawed for the reasons given by the EAT (§ 32):

"First, the view of the project team that it wanted to choose the highest-calibre people from both the jobs affected is on the face of it a justifiable objective, and the Tribunal gives no reason to the contrary. Secondly, if the Commission had decided not to take the ring-fencing option at least partly in order to ensure that there were no male ISOs, that would, it seems to us, have constituted unlawful (direct) discrimination against the ex-SIIOs, who would have been deprived of the chance of a job as an ISO on the basis that some of them were men."

148. In my judgment, therefore, the EAT were entitled to overturn the decision of the ET on this point. Having reached that point, the next question is whether the EAT was entitled to substitute its own view, or whether it should have remitted the case to the ET.
149. In a case where further facts need to be found, or where even if the ET had correctly directed itself in law it might still have come to the same conclusion, the correct course would be to remit the case to the ET. But in the present case the claimants did not suggest what further facts needed to be found. The EAT worked on the basis of the facts as found by the ET. As Mr Jeans QC pointed out, recent observations in this court have referred to the remission of a case as a "last resort": Buckland v Bournemouth University [2010] IRLR 445 (§ 58). Once the two alternative means canvassed by the ET have been seen to be flawed, there was no other means left, except the one that the respondent adopted. Lady Hale has given a steer about the acceptability of a "grandfather clause". I would accept Mr Jeans' submission that the EAT was well equipped to assess whether any issue meriting remission arose and the EAT's decision to substitute conclusions was (at the very least) within its discretion as the specialist appellate body.
150. In respectful disagreement with Mummery LJ, therefore, I would dismiss the appeal.
151. Postscript. When our judgments were circulated in draft the Appellants asked for permission to make further submissions about the effect of Homer. Both the Appellants and the Respondent have now made full written submissions. I have read and considered those submissions; but they do not cause me to alter the judgment as originally circulated.

Sir Mark Waller:

152. There is no disagreement between Mummery LJ and Lewison LJ on the question whether the ET made an error of law on the first issue whether there was indirect discrimination. Furthermore there is no disagreement between them on the question whether the ET made an error of law entitling the EAT to interfere with the ET's decision on objective justification. I agree with both judgments on those aspects.
153. It is as to the result which flows from the answers to those questions on which there is a difference and on which I am placed in the invidious position of having the last word at least in this court. It is a word that having regard to the delay which has occurred up until now I must express quickly and shortly.
154. Mummery LJ is of the view that there are flaws in the way the EAT approached the issue of objective justification . He says

"In my judgment, there are flaws in the way in which the EAT itself approached the issue of objective justification. It appears to have done so on the basis that there was nothing discriminatory about the pay difference that the Commission were seeking to protect and that there was no discriminatory pay difference before 1 October 2007, but that ignores the fact that the stage of justification is only reached, if there is indirect discrimination which has to be justified. In this part of its judgment it seems to me, with the greatest respect, that the EAT was backtracking to the prior issue of indirect discrimination in pay and saying, contrary to the basis on which it was considering the whole question of objective justification, that there was no indirect discrimination in pay."

155. That leads him to the conclusion that the matter should be remitted to the ET for them to rule on that question.
156. Lewison LJ says that the question is whether the EAT was entitled to substitute its own view, or whether the case should be remitted to the ET. He is against remission pointing out that remission should only be ordered as a "last resort" and where further facts need to be found. He concludes that "once the two alternative means canvassed by the ET have been seen to be flawed there was no other means left except the one that the respondent adopted." In so saying he is as I understand it either rejecting Mummery LJ's criticisms of the approach of the EAT or alternatively holding that even if there was some justification for the criticism, the EAT were right in other respects in which they criticised the ET judgment and it follows that there is only one answer.
157. With respect to Mummery LJ it seems to me important if conceivably possible not to remit this matter to the ET.
158. My view put shortly is as follows. The ET did make errors of law on this aspect. The EAT were therefore entitled to interfere. The criticisms of the EAT's approach as identified by Mummery LJ do have some validity but do not go to certain critical matters on which they were entitled to substitute their own views for those of the ET. This court is entitled to accept those views and should do so rather than remit the matter to the ET. I therefore agree with Lewison LJ that this appeal should be dismissed.
159. Postscript: Following circulation of the judgments in draft the parties were invited to make written submissions on Homer an authority cited by Lewison LJ. He has seen no reason to alter his judgment in the light of those submissions and I see no reason to alter mine.