[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Armstrong & Ors v Newcastle Upon Tyne NHS Hospital Trust [2005] EWCA Civ 1608 (21 December 2005) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2005/1608.html Cite as: [2005] EWCA Civ 1608, [2006] IRLR 124

[New search] [Printable RTF version] [Buy ICLR report: [2007] ICR 469] [Buy ICLR report: [2007] ICR 1644] [Help]

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM
The Employment Appeal Tribunal
HH Judge Ansell

		Royal Courts of Justice Strand, London, WC2A 2LL
		21st December 2005

Lady Justice Arden:

1. These are two appeals against the order of The Employment Appeal Tribunal (HHJ Ansell, I. Ezekiel and PRA Jacques) ("the EAT") dated 22 November 2004. The proceedings concern claims to equal pay, and these appeals arise from the determination by the Employment Tribunal of preliminary issues. The appellants are domestic ancillary workers. The respondent ("NHT") is a hospital trust. In the first appeal (the "wrong comparator" appeal"), Mrs Armstrong and her co-appellants contend that the EAT wrongly held that female ancillary workers at NHT hospitals, other than the Royal Victoria Infirmary ("RVI"), could not use, as comparators for equal pay purposes, male ancillary workers at the RVI. In the second appeal (the "genuine material factor" appeal), NHT challenges the rejection by the EAT of its defence that there was no genuine material factor justifying discrimination between male and female ancillary workers at the RVI.
2. The appellants can be divided into two groups: those employed at one of the four hospitals other than the RVI, who are known as the non-RVI claimants, and those employed at the RVI, who are known as the RVI claimants. The first group cannot treat the RVI porters as comparators for the purposes of article 141 unless NHT is a "single source" for the purposes of that article.
3. The equal pay claim covers the first and third periods set out below. NHT has not been the employer throughout those periods. NHT only came into existence on 1 April 1998 and represents the merger of two hospital trusts, whom I can call simply the Freeman trust and the RVI trust. These two trusts had come into existence as a result of the National Health Service and Community Care Act 1990. Starting in April 1991, the trusts each took over two of the four hospitals previously run by the Newcastle Health Authority ("NHA"). The precise mechanics and timing do not matter. For the purposes of this appeal, the period of time from 1988 to date can be divided into the following periods:

a) the period up to 1991, when the claimants and the male comparators were all employed by the NHA;

b) the period between 1 April 1991 to 1 April 1998, when the two trusts were in operation and the non-RVI claimants and the male comparators were employed by different trusts;

c) the period from 1 April 1998 when the trusts were merged so that the claimants and the comparators were all employed by the same employer again, namely NHT.

4. The equal pay claim relates to bonus payments, to which the domestic ancillary staff were entitled until the work was contracted out to private contractors. The appellants say that their work is of equal value to that of the porters at the RVI. The porters at the RVI were also entitled to these bonus payments. The tendering process was known as compulsory competitive tendering or "CCT". In 1985, NHA decided to put the domestic work out to tender and an in-house tender was accepted as a result of which the domestic staff, who were predominantly female, lost their right to bonuses. The portering work, on the other hand, was not put out to tender and as a result the porters, mostly men, retained their right to bonuses. NHT does not accept that the porters at the RVI can be taken as comparators by the non-RVI claimants ("the wrong comparator defence"). In addition, by its genuine material factor defence, it contends that there was a good reason for the distinction found by the Tribunal between the RVI claimants and the male porters at the RVI. I will need to examine both defences below. The Tribunal found against NHT on the genuine material factor issue as it held that some differences in the pattern of bonus payments had not arisen from material factors other than sex.

Legislative Framework

5. Section 1 of the Equal Pay Act 1970 ("the 1970 Act") provides:-

"(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.

(2) An equality clause is a provision which relates to terms (whether concerned with pay or not) of a contract under which a woman is employed (the "woman's contract"), and has the effect that –

(a) where the woman is employed on like work with a man in the same employment –

(i) if (apart from the equality clause) any term of the woman's contract is or becomes less favourable to the woman than a term of a similar kind in the contract under which that man is employed, that term of the woman's contract shall be treated as so modified as not to be less favourable, and

(ii) if (apart from the equality clause) at any time the woman's contract does not include a term corresponding to a term benefiting that man included in the contract under which he is employed, the woman's contract shall be treated as including such a term;

(b) where the woman is employed on work rated as equivalent with that of a man in the same employment –

(i) if (apart from the equality clause) any term of the woman's contract determined by the rating of the work is or becomes less favourable to the woman than a term of a similar kind in the contract under which that man is employed, that term of the woman's contract shall be treated as so modified as not to be less favourable, and

(ii) if (apart from the equality clause) at any time the woman's contract does not include a term corresponding to a term benefiting that man included in the contract under which he is employed and determined by the rating of the work, the woman's contract shall be treated as including such a term

(c) where a woman is employed on work which, not being work in relation to which paragraph (a) or (b) above applies, is in terms of the demands made on her (for instance under which headings as effort, skill and decision), of equal value to that of a man in the same employment –

(i) if (apart from the equality clause) any term of the woman's contract is or becomes less favourable to the woman than a term of a similar kind in the contract under which that man is employed, that term of the woman's contract shall be treated as so modified as not to be less favourable, and

(ii) if (apart from the equality clause ) at any time the woman's contract does not include a term corresponding to a term benefiting that man included in the contract under which he is employed, the woman's contract shall be treated as including such a term.

(3) 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.

(4) A woman is to be regarded as employed on like work with men if, but only if, her work and theirs is of the same or a broadly similar nature, and the differences (if any) between the things she does and the things they do are not of practical importance in relation to terms and conditions of employment; and accordingly in comparing her work with theirs regard shall be had to the frequency or otherwise with which any such differences occur in practice as well as to the nature and extent of the differences.

(5) A woman is to be regarded as employed on work rated as equivalent with that of any men if, but only if, her job and their job have been given an equal value, in terms of the demand made on a worker under various headings (for instance effort, skill, decision), on a study undertaken with a view to evaluating in those terms the jobs to be done by all or any of the employees in an undertaking or group of undertakings, or would have been given an equal value but for the evaluation being made on a system setting different values for men and women on the same demand under any heading.

(6) Subject to the following subsections, for purposes of this section –

(a) "employed" means employed under a contract of service or of apprenticeship or a contract personally to execute any work or labour, and related expressions shall be construed accordingly;

(b) …

(c) two employers are to be treated as associated if one is a company of which the other (directly or indirectly) has control or if both are companies of which a third person (directly or indirectly) has control.

and men shall be treated as in the same employment with a woman if they are men employed by her employer or any associated employer at the same establishment or at establishments in Great Britain which include that one and at which common terms and conditions of employment are observed either generally or for employees of the relevant classes."

6. So the equality clause implied by section 1(1) of the 1970 Act only operates where employees are "in the same employment", and that expression is defined by section 1(6).

Article 141 and "single source"

7. The 1970 Act is designed to give effect to the obligations of the UK under what is now Article 141 of the EC Treaty. We shall see in a moment that Article 141 is not fully implemented in Section 1(6) of the 1970 Act. Article 141 provides:

"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.

2. For the purpose of this Article, 'pay' means the ordinary basic or minimum wage or salary and any other consideration whether in cash or in kind, which the worker receives directly or indirectly, in respect of his employment, from his employer.

Equal pay without discrimination based on sex means:

a) that pay for the same work at piece rates shall be calculated on the basis of the same unit of measurement;

b) that pay for work at time rates shall be the same for the same job…"

8. Article 141 is directly effective, and so employees can rely on it without also invoking the 1970 Act: see Defrenne v Sabena (No. 2) [1976] ICR 547. That is important because the Court of Justice has made it clear that Article 141 is not restricted to cases where the claimants and the comparators have the same employer, but that nonetheless in all cases – even where there is a common employer - differences in pay must be attributable to a "single source" of employment. This was established in Lawrence v Regent Office Care Ltd [2003] ICR 1092, where the employees in question, who were former employees of a local authority who were transferred to companies to whom their work had been contracted out, unsuccessfully sought to establish that other employees of the local authority were comparators, the Court of Justice held:-

"17. There is, in this connection, nothing in the wording of article 141(1) EC to suggest that the applicability of that provision is limited to situations in which men and women work for the same employer. The court has held that the principle established by that article may be invoked before national courts in particular in cases of discrimination arising directly from legislative provisions or collective labour agreements, as well as in cases in which work is carried out in the same establishment or service, whether private or public: see, inter alia, Defrenne [1976] ICR 547, 568, para 40; Macarthys Ltd v Smith (Case 129/79 [1980] ICR 672, 690, para 10, and Jenkins v Kingsgate (Clothing Productions) Ltd (Case 96/80) [1981] ICR 592, 613-614, para 17.

18. However, where, as in the main proceedings here, the differences identified in the pay conditions of workers performing equal work or work of equal value cannot be attributed to a single source, there is no body which is responsible for the inequality and which could restore equal treatment. Such a situation does not come within the scope of article 141(1) EC. The work and the pay of those workers cannot therefore be compared on the basis of that provision.

19. In view of all the of the foregoing, the answer to the first question must be that a situation such as that in the main proceedings, in which the differences identified in the pay conditions of workers of different sex performing equal work or work of equal value cannot be attributed to a single source, does not come within the scope of article 141(1) EC."

9. The decision of the Court of Justice in Lawrence was considered and applied by this Court in Department for Environment, Food and Rural Affairs v Robertson [2005] ICR 750. Here the issue was whether civil servants in one government department could establish that civil servants in another department were comparators. Mummery LJ, with whom Maurice Kay and Gage LJJ agreed, considered what was meant by "attributed to a single source" in the Lawrence case and held as follows:-

"13. In Lawrence the Court of Justice held that, for equal pay proceedings to come within the ambit of article 141(1), the pay differences between workers of different sex performing equal work must be "attributed to a single source". As I understand it, the focus of this rather imprecise approach is on the location of the body responsible for making decisions on levels of pay in the relevant employment or establishment rather than on the identification of the relevant legal source of that decision-making power. The comparator issue does not turn on precise legal analysis or on a comparison of the employment relationships between the workers and their respective employers or, in the case of state workers, on particular constitutional doctrines and arrangements, which condition the nature of the legal relationship between a member state and its civil servants and which are liable to differ from one member state to another…

28. I agree that the issue in Lawrence was different from this case and that the judgement of the Court of Justice does not directly deal with the case of a common employer. It appears, however, to lay down an approach of general application. In my judgment the statements of principle in paras 17 and 18 indicate that something more than just the bare fact of common employment is required for comparability purposes. I reject the submission that in those paragraphs the Court of Justice proceeded on the basis that common employment is sufficient for comparability purposes. I agree with the appeal tribunal [2004] ICR 1289, 1306, para 21, that:

"Lawrence is not authority for the proposition that common employment is sufficient. It is rather authority for the proposition that what underlines the applicability of article 141 is that which is ordinarily exemplified by common employment, namely the existence of a common source, the existence of a central responsibility for terms and conditions. If that is absent, then comparability is not available…

29. The opinion of the Advocate General in Lawrence supports the view that article 141 is addressed generally to those who may be held responsible for unauthorised differences in terms and conditions of employment and that it is not sufficient simply to look at who are the employers of the applicants and the comparators and to proceed to consider "single source" only if they are in not common employment. It is necessary to consider in each case whether the terms and conditions are traceable to one source: see [2003] ICR 1092, paras 30, 37-40, 46, 48-52, 54. I agree with the appeal tribunal (see [2004] ICR 12889, 1304, para 18) that the Court of Justice was setting out a justification in the form of a "principled basis upon which responsibility for difference and discrimination can be pinned" and that the justification is in the "single source" rather than in common employment. The Court of Justice made it clear that it is not necessarily the person with whom the workers have contracts of employment that determines comparability. The relevant body is the one "which is responsible for the inequality and which could restore equal treatment". The body responsible for the state of affairs will often be the same employer of both the applicants and the comparators, but that is not necessarily so. It depends on the circumstances of the particular case as to whether further inquiry may be necessary. If that were not the case, Mr Langstaff's submission would tend to have the extravagant consequence that every civil servant would be entitled to compare himself or herself with any other civil servant of the opposite sex, subject only to objective justification by the employer of differences in pay. That does not seem to be a sensible or practical approach to the preliminary task of identifying appropriate workers in circumstances comparable to the applicants. In my judgment the tribunals below were right to reject it. On my reading of Lawrence [2003] ICR 1092 the approach of EC law is to locate the single source with the body responsible for setting the relevant terms. This is not determined by only addressing the formal legal question of the identity of the employer."

10. Thus, to constitute a single source for the purpose of article 141, it is not enough for the non-RVI claimants to show that they have the same employer as the comparators. They must show that that employer was also the body responsible for setting the terms of both groups of employees. The civil servants in Robertson were unsuccessful. They had a common employer: the Crown. However, the power to negotiate most aspects of their pay was delegated to individual departments and as a result there were significant differences in pay between the various departments.

Decision of the Employment Tribunal

11. The Employment Tribunal (Mr J R Barton, Mr D McGregor and Mrs M Robinson), in its extended reasons dated 22 December 2003 dealt with the wrong comparator defence and the genuine material factor defence as preliminary issues. In the event, the Tribunal accepted the wrong comparator defence applied to non-RVI claimants. Before doing so, the Tribunal made a number of findings of fact particularly in paragraph 8 of its extended reasons, including the following, which featured in the argument before this court on the genuine material factor defence:

"(u) However an analysis of the effective bonus arrangements appended to the applicants closing submissions nevertheless shows that from 1988 onwards they have had a disparately adverse effect on women as shown by Ms Swanson's evidence contained in her second supplemental statement."

12. There is an issue as to which analysis is referred to here. Counsel produced two analyses and appended them to her closing submissions. The first, which I will call analysis A, related to the RVI claimants. I can conveniently set out the relevant part of the analysis here:
STATISTICAL ANALYSIS OF THE EFFECT OF THE ASC STAFF BONUS SCHEME AT THE ROYAL VICTORIA INFIRMARY

Table 1 – Percentages of male and female ancillary staff employed at the RV1 who are/are not in the bonus scheme

Date	% Men in scheme	% Women in scheme	% Men not in scheme	% Women not in scheme	Disparity
04.84	91.3	84.6	8.7	15.4	6.7
04.88	94.1	27.2	5.9	66.9	66.3
04.98	87.7	27.7	12.3	72.3	60
07.01	36.8	12.2	63.2	87.8	24.6
02.02	35.8	11.5	64.2	88.5	24.3

13. On the basis of this analysis, after the introduction of CTT in 1985 the number of RVI claimants who participated in the bonus scheme fell sharply leading to a disparity of over 60% in April 1998.
14. The second analysis, which I will call analysis B, which was also annexed to Counsel's closing submissions before the Tribunal, showed the disparity of effect of the bonus arrangements as between the male and female ancillary staff employed in the same period in all of NHT's hospitals but the disparity on this analysis was only about half of that shown by the table which I have set out above.
15. The tribunal dealt first with the wrong comparator defence. It set out section 1(6) of the 1970 Act and considered the meaning of "establishment". It then held as follows:-

"11. Neither did we find that there were common terms and conditions under section 1(6) Equal Pay Act 1970 in that notwithstanding the merger of the RVI and Freeman Trust in April 1998, the bonus agreements adopted at various hospitals and the individual departments were subject to collective negotiation on a departmental basis and had to be self funding. Once adopted, the bonus schemes constituted "essentially different employment regimes" (Leverton v Clwyd County Council [1989] ICR 33 and British Coal Corporation v Smith [1996] ICR 515). Also staff at the different hospitals had retained different protection provisions.

12. It was also argued for the applicants that they were entitled to rely upon Article 141 of the EC Treaty. We were referred in this respect in particular to Scullard v Knowles [1996] ICR 399 EAT and South Ayrshire v Morton [2002] ICR 956, but accepted the contention that the leading case was Lawrence v Regent Office Care Ltd [2002] IRLR 822 to the effect that a situation "in which the difference is identified in the pay conditions of workers of different sex performing equal work or work of equal value cannot be attributed to a single source" falls outside the scope of Article 141.

13. Accordingly we are not satisfied that the applicants who were non-RVI applicants, were in the same establishment or service as the RVI comparators. "Service" cannot mean an entire industry because Article 141 was not intended to equalise pay across whole industries and the Regent Care case suggested the definition of "service" under Article 141 must be limited to cases where alleged inequality is attributable to and can be restored by a single source.

14. Reliance was placed by the applicants on the background to the introduction of bonus schemes and detailed procedures arising from Whitley Council guidance. However they arose as a result of a large number of local negotiations, the willingness of the NHA to implement a bonus scheme and a staff vote to adopt the offered scheme so that we do not find that they derive directly from the Whitley Council guidance.

15. Accordingly we conclude, as was submitted for the Respondent, that prior to the establishment of the Respondent Trust on 1 April 1998 neither section 1(6) of the Equal Pay Act 1970 nor Article 141 can assist the former employees of the Freeman or NCH Trusts to protect claims prior to 1 April 1998. Prior to that date the RVI, Freeman and NCH employees were not part of the same establishment or service and from 1991/1992 the various Trusts were autonomous employers. There was no one body responsible for the inequalities alleged or which could restore equal treatment as neither the Freeman, the NCH nor the RVI Trust had Authority or power to direct what the others did. All three were free from central control.

16. As to the period between 1 April 1991/1992 and 1 April 1998 the non-RVI applicants conceded that they were not in the same employment as their comparators when employed by the Freeman or NCH Trust but do rely upon Article 141 for this period. We reach the same conclusions namely that the separate Trusts were autonomous and, therefore, there was no single source or body which was responsible for or could remove the alleged inequality.

17. In relation to the period of employment by the NHA, the non-RVI applicants rely upon both Article 141 and section 1(6) of the Equal Pay Act 1970. The same arguments apply as above in relation to Article 141. The applicants rely on Kells v Pilkington [2002] IRLR 693 in arguing that an equality clause modified their contracts of employment and that this would have continued to have effect following establishment at the Freeman and NCH Trusts in 1991 and 1992."

16. There is an issue as to whether in this passage the Tribunal indeed concluded that, after 1 April 1998, the non-RVI claimants were not in the same employment as their RVI comparators for the purposes of article 141 on the grounds that they were not employed in the same establishment and that the terms and conditions of employment varied and represented essentially different employment regimes. This is issue A1 on the wrong comparator appeal (see paragraph 21 below).
17. On the genuine material factor defence, the Tribunal set out the principles as summarised by Lord Nicholls of Birkenhead in Glasgow City Council v Marshall [2000] 1 WLR 333 at 339, where Lord Nicholls held:-

"23. 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 the 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 than an employer who satisfies the third of these requirements [that the reason relied upon is not the difference in sex] 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 effect 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."

18. The tribunal then held as follows:-

"27. However Glasgow City Council v Marshall suggests that where there is any evidence of sex discrimination, the employer is required to show that the difference in pay is objectively justified. Lord Nicholls at p203 says that when justification is in point "a purely historic explanation of the pay difference between sexes is insufficient".

28. Miss Tether also drew reliance in supporting the principle that the employer needs to justify a difference in pay which has a disparately adverse effect on women by the following authorities: Enderby v Frenchay Health Authority [1994] ICR 112; Jamstall Hetsombudsmanne v Orebro Landsting [2000] IRLR 421; Brunnhofer v Bank der Osterriechischen Postparkasse [2001] IRLR 572."

29. Miss Tether relied upon the facts of the present case as being in important respects similar to those which form the backdrop to North Yorkshire County Council v Radcliffe and others [1995] ICR 833. The domestics of the RVI were exclusively or almost exclusively, female and the labour market for domestics in the north east was almost exclusively female. The porters were male. It was considered that one of the reasons why portering was not put out to tender was that as it was not compulsory to do so a more robust industrial reaction might be experienced. We do see the decision not to put out the predominantly male area to tendering as being tainted by sex.

30. The Respondent relies upon the following explanation for difference in treatment, namely:

(i) the porters were not put out to tender because it was not compulsory to do so and the NHA was philosophically opposed to the idea;

(ii) major savings would not be made in this area because the budget for portering was less than in other groups;

(iii) the NHA management was not aware of local companies who might have been interested in tendering in this area;

(iv) opposition to a proposal to put out to tendering an area which are not compulsory was expected to be more voluble."

31. In considering the above reasons given or the tests set out in Glasgow City Council we can accept the first two of the above reasons might have been genuine at the time. We are concerned about the second two as there appears to have been little or no effect to explore the market and the male porters were in employment traditionally more heavily unionised and voluble because we know from our own experience of the local market more males tend to be full time than females and to be unionised. These two alleged justifications amount to significant and relevant factors tainted by sex and strike at the basis of the justification claimed.

32. However, even if the explanation had been genuine at the time the Respondent then relies upon the historical basis for the introduction of bonus schemes and the reducing disparity but on our findings of fact there has been a smaller proportion of women entitled to receive a bonus at all material times since 1988 than the proportion of men. We accept the contention made for the applicants that no witnesses have been called properly to explain why the bonus schemes were being maintained in existence at the time when the applicants' claims were submitted. We accept that the historical factors are therefore not sufficient to justify the maintenance of differentials which have had a disparate impact on women for a period of nearly fifteen years."

19. In the circumstances the tribunal held that the difference in treatment as to bonus was a material difference between the case of the RVI claimants and that of the men. The non-RVI claimants then appealed to the EAT. On the wrong comparator point, the EAT held as follows:-

"33. We have already referred to the Tribunal's conclusions in relation to the post-1998 period set out in paragraphs 11-14 of their Decision. Although the Tribunal in their earlier fact finding in paragraph 8 had not set out in any great detail contractual arrangements post 1998, both parties' final submissions before the Tribunal made reference to whether or not there were substantially comparable terms and conditions across the separate Trusts' hospitals after April 1998 – see paragraph 42 of the Appellants' closing submissions and paragraph 47 of the Respondents' closing submissions. The Tribunal in particular in their conclusion had identified that staff at the different hospitals had retained different protection provisions. Miss Tether submitted that this was a minor disparity and that no reasonable Tribunal should have found that this prevented terms and conditions being considered substantially comparable on a broad basis. Again, however, we are satisfied that the tribunal had the correct test in mind and were entitled on the facts before them to come to the conclusion that there were essentially different employment regimes and we see no reason to interfere with their conclusion. Similar conclusions apply in relation to the Article 141 issue. Again the Tribunal, in our view, focussed on the key issue which related to the single source and came to the conclusion that post-1998, since there was continuing collective negotiation on a departmental basis, those differences could not be attributed to the Respondent as a single source."

20. NHT also appealed to the EAT on the genuine material factor point. The EAT rejected NHT's argument that the tribunal had failed to deal with its defence based on the pattern of bonus payments. The EAT also rejected NHT's argument that the tribunal had erred in its conclusion that the decision not to tender portering services was tainted by sex discrimination. It concluded that the issue of portering services was secondary to the tribunal's main finding in relation to the process of competitive tendering being tainted with sex discrimination arising out of the domestics being part of a labour market which was almost exclusively female. The EAT further concluded at paragraph 43 of its judgment that the Tribunal was "entitled to come to the conclusion that there were no sufficiently cogent reasons as to why the pattern of differentials had not been eradicated".

The "wrong comparator" issue

21. On this aspect of the appeal, we are concerned with the position of the non-RVI claimants and the comparators for the purposes of Article 141 in the third period, i.e. after 1 April 1998. The arguments have raised four issues, which I will call A1 to A4:

A1 did the tribunal hold that the employment of the non-RVI claimants and the comparators was not attributed in that period to a "single source" of employment for the purposes of the Lawrence case and give its reasons for that conclusion?

A2 if so, was that conclusion against the weight of the evidence?

A3 as a matter of law, could NHT successfully contend that there was no "single source" for this purpose once there was evidence that it had taken some part in pay negotiations relating to both groups of workers?

A4 could NHT constitute a "single source" of employment if it only adopted the pay disparity created by former, separate employers, rather than created it itself?

22. The background can be summarised shortly. Before 1991 the NHA was responsible for the differences in the payment of bonuses because the relevant negotiations were conducted pursuant to a national collective agreement which was binding on the NHA, and within a framework laid down by the NHA. A bonus scheme could not be implemented without the approval of the NHA. In the interim period (1991 to 1998) there were separate trusts. However, after 1 April 1998 the hospitals were brought back under the same umbrella.
23. Miss Melanie Tether, for the appellants, submits that there was a single source of employment for the non-RVI claimants and the RVI porters for the purposes of Article 141 after 1 April 1998 and that the Tribunal was in error if it did not so find. On her submission, it is not necessary for the "single source" of employment to have created the disparity. It is sufficient if the employer adopts a pay difference. Here, on Miss Tether's submission, the evidence before the tribunal showed that NHT decided on assuming responsibility for harmonising rates of basic pay for ancillary staff and small differences in holiday entitlement conditions, together with grievance and disciplinary procedures and equality policies, at the various sites. That harmonisation stopped short of an incentive bonus scheme. (Nor did it cover salary protection arrangements but that was a comparatively minor matter). The tribunal failed to mention this evidence. The respondent decided to stop the bonus scheme for new starters but nonetheless determined that the scheme would continue in existence for those entitled. It thus allowed the scheme to "wither on the vine" in due course. The respondent's intervention showed that it could have entered into negotiations over the continuation of the scheme. Miss Tether submits that is a situation which is capable of giving rise to a single source for article 141 purposes and the tribunal made no or no adequate findings on this point.
24. Miss Tether submits that the EAT was wrong to say that collective bargaining after 1 April 1988 was on a departmental basis. There were, on Miss Tether's submission, no findings of fact by the tribunal to support this: the collective negotiations referred to by the tribunal in paragraph 11 of its decision were historical, not ongoing. There had been discussion about new pay structures but those negotiations failed. By leaving the bonus schemes where they were, the respondent assumed responsibility for them.
25. Mr Michael Supperstone QC, for the respondent, submits that although the reasoning by the tribunal was brief, it concluded that the respondent was not a "single source" for any period. Mr Supperstone submits this is clear from paragraph 11 of the tribunal's decision. Mr Supperstone further submits that, as the tribunal made a finding of fact which was plainly open to it, its conclusion on the wrong comparator issue cannot be challenged as perverse. He submits that the position is not unfair. As a matter of policy, therefore, it would be too onerous to impose liability on an employer particularly if there were discrepancies for which the employer was not responsible. Mr Supperstone further submits that the question whether an employer can be liable if he adopts the terms and conditions imposed by his predecessor employer is not covered by the Robertson case. Mr Supperstone submits that an employer can have different bonus schemes in respect of different parts of the group if there are different employment regimes. So he submits a transfer of an undertaking should not lead to a greater burden on the employer.
26. I will take issues A1 to A4 in turn.

Issue A1

27. I agree with Mr. Supperstone's analysis. The relevant paragraphs of the tribunal's decision are paragraphs 11 to 17 set out in paragraph 13 above. The reference to the merger of the RVI and Freeman Trust shows that the Tribunal were addressing the position in the third period. In the next paragraph, the tribunal identified the Lawrence test under Article 141. In paragraph 13, the Tribunal stated their conclusion that the employment of the non-RVI claimants and the comparators was not attributable to a "single source". In paragraphs 14 and 15 the tribunal dealt with the first period. In paragraph 16 the tribunal dealt with the second period. It is clear that in paragraph 17 the tribunal consider the application of Article 141 to the first and second periods. Although it is confusing that the tribunal should have dealt with the various periods in ante-date order, it is tolerably clear from the structure of paragraphs 11 to 17 that the tribunal rejected the non-RVI claimants' case under Article 141 for the third period.
28. In my judgment, the findings and the reasoning for the period after 1 April 1998 are to be found in the first sentence of paragraph 11 of the decision, which deals with the application of section 1(6) to the period after 1 April 1998. The tribunal then considers the jurisprudence on Article 141 in paragraph 12. The essence of the tribunal's reasoning is that there was pay negotiation on a departmental basis, that is, at the level of or below the separate hospitals. When it starts the next paragraph (paragraph 13) with the word "Accordingly" it must in my judgment be stating a conclusion from paragraphs 11 and 12 together. Accordingly, the tribunal gave reasons for its conclusion on the application of Article 141 to the third period. In those paragraphs the tribunal dealt with the question of whether the terms and conditions of employment of the non-RVI claimants and the comparators were attributable to a single source. In this respect, I agree with the EAT and accept the submissions of Mr. Supperstone.

Issue A2

29. The next issue is whether it was perverse of the tribunal to reach this decision in the light of the evidence as to harmonisation of terms of conditions referred to by Miss Tether in her submissions and summarised in paragraph 17 above. Not all the evidence that was before the tribunal is in the bundle. For instance the statement of Mr. Gooden is missing. There are, however, documents in the bundle before us which suggest that new terms and conditions were given only to new employees or employees who took up new posts after 1 April 1998 (see pp 295 to 297). In the absence of other relevant material, I conclude that the tribunal by implication must have concluded that the matters relied on by Miss Tether did not alter their view as to the overall position on the evidence. In my judgment, there was no reason in law why it should not take this course.

Issue A3

30. I now turn to the third issue under this part of the case, which is a question of law as to the application of the "single source" concept. Unlike the situation in the Robertson case, this is not a case where there was no involvement whatever by the employer in the negotiation of terms and conditions at a departmental level, but in my judgment the Robertson case does not turn on a brightline test of that kind. The question is always whether the NHT was responsible for setting the terms of employment of employees whose employment began before 1 April 1998. Therefore the tribunal had to evaluate all the relevant evidence and see which side of the line the case fell. I do not accept Miss Tether's approach that because there was some evidence of harmonisation of the terms and conditions the tribunal could only properly conclude that the respondent had assumed responsibility for the terms and conditions of all the employees for the purposes of the Robertson test.

Issue A4

31. This point is not covered by the Robertson case. As it does not in my judgment require to be decided, I express no view thereon. Accordingly I turn to NHT's appeal on the genuine material factor issue.

The genuine material factor issue

32. 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.
35. Accordingly the question arises as to what degree of disparity is required to be shown by the gender-based comparison in order to trigger the rebuttable presumption of sex discrimination. This question was considered in R v Employment Secretary ex parte Seymour-Smith [1999] ICR 447. At paragraph 60-62 of its decision, the Court of Justice held:

"60. As the court has stated on several occasions, it must be ascertained whether the statistics available indicate that a considerably smaller percentage of women than men is able to satisfy the condition of two years' employment required by the disputed rule. That situation would be evidence of apparent sex discrimination unless the disputed rule were justified by objective factors unrelated to any discrimination based on sex.

61. That could also be the case if the statistical evidence revealed a lesser but persistent and relatively constant disparity over a long period between men and women who satisfy the requirement of two years' employment. It would, however, be for the national court to determine the conclusions to be drawn from such statistics.

62. It is also for the national court to assess whether the statistics concerning the situation of the workforce are valid and can be taken into account, 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: see Enderby v. Frenchay Health Authority (Case C-127/92) [1994] 1.C.R. 112, 161, para. 17 …"

36. At paragraph 65 of its decision, the Court of Justice held:

"A measure adopted by a Member State has a disparate effect as between men and women to such a degree as to amount to indirect discrimination for the purposes of Article 119 of the Treaty, where the statistics available indicate that a considerably smaller percentage of men than women is able to fulfil the requirement imposed by that measure".

37. Accordingly, it is not enough for the tribunal to receive statistics showing an apparent difference in effect between men and women. It must form a view that the impact is considerable. They must also consider whether the considerable impact is purely accidental or whether it has greater significance. However, the statistics can attain the requisite degree of significance not simply by the size of the percentage of the disparity in any given year, but also where the percentage is smaller but persists over a number of years. Thus, when the Seymour-Smith case returned to the House of Lords, the House, by a majority, held that a persistent and constant disparity, which resulted in ten men meeting an impugned condition for every nine women, over a period of about eight years was sufficient to show that the condition had a disparately adverse effect on the women in question: see R v Secretary of State for Employment ex parte Seymour-Smith [2003] 1CR 244.
38. When it comes to step 2 and the question whether a difference in treatment is tainted with sex, a tribunal has to make findings of fact. In North Yorkshire County Council v Ratcliffe [1995] ICR 833, the applicants were three female school catering assistants whose work was rated as of equal value to that of men employed by the same employer at various establishments. Compulsory competitive tendering was introduced. A direct service organisation (a "DSO") was established to provide school meals. The catering assistants were dismissed and re-employed with the DSO at rates of pay which were less than the men but which were designed to enable the DSO to compete with commercial organisations. The tribunal made a finding that the conditions of the catering assistants had been reduced because it was thought necessary to do so to enable them to compete. Competitors only employed women and employed them on less favourable terms than the Council previously employed the catering assistants. The tribunal took the view that while that might have been a material factor it was certainly a material factor due to a difference of sex "arising out of the general perception … that a woman should stay at home to look after the children and if she wants to work it must fit in with that domestic duty and the lack of facilities to enable her, easily, to do otherwise". On that basis the employer had to justify the difference in treatment. The House of Lords held that the tribunal was entitled to make that finding. Lord Slynn expressed the view that "to reduce the women's wages below that of their male comparators was the very kind of discrimination in relation to pay which the Act sought to remove." In other words, once it was shown that the rates of pay for dinner ladies were determined by rates applicable to what the market perceived to be women's work, competitive forces did not, in the circumstances of that case, constitute a defence and objective justification had to be shown.
39. Section 1 of the 1970 Act does not distinguish between direct and indirect forms of discrimination. Direct discrimination would occur where a woman is paid less than a man on account of her sex. Indirect discrimination can occur where the employer's practice does not discriminate on the face of it but, in the way it operates, it is disproportionately disadvantageous to women. It does not matter that the employer has acted without a motive to discriminate.
40. As to objective justification, the employer may rely on a legitimate commercial objective, but he must also show that the maintenance of the practice in question is reasonably necessary for the purpose of achieving that objective: Bilka – Kaufhaus Gmbh v Weber von Hartz [1987] ICR 110. In other words, a test of proportionality applies. However, once the tribunal is satisfied that there was genuine justification, it does not have to consider whether the result was fair. The legislation is designed only to eliminate discrimination which is due to the difference of sex.
41. Part of the challenge to the decision of the tribunal on this point of the case has been directed to the sufficiency of the tribunal's reasons. The approach which this Court should adopt to a "reasons" challenge of this nature was considered by this Court in English v Emery Reimbold & Strick Ltd [2002] 1WLR 2409. As appears from the head note, that case established a number of propositions. The following apply to this case:

(a) a judicial decision which affects the substantive rights of the parties should be reasoned;

(b) a tribunal is not bound to deal with every argument or identify every factor which weighed with it, but it must deal with the issues which are essential to its conclusion;

(c) the decision should enable the parties readily to analyse the tribunal's essential reasoning;

(d) the decision should be read against the background of the evidence and submissions at the hearing before the tribunal.

42. It follows from points (b) and (d) in particular that this Court will be slow to overturn a decision of a tribunal if its essential reasoning can be deduced from what it has said in combination with the arguments and evidence, documentary or oral, placed before it. As Bingham LJ said in Meek v City of Birmingham District Council [1987] 1RLR 250:

"It has on a number of occasions been made plain that the decision of an [Employment] Tribunal is not required to be an elaborate formalistic product of refined legal draftsmanship, but it must contain an outline of the story which has given rise to the complaint and a summary of the Tribunal's basic factual conclusions and a statement of the reasons which have led them to reach the conclusion which they do on those basic facts. The parties are entitled to be told why they have won or lost. There should be sufficient account of the facts and of the reasoning to enable the EAT … to see whether any question of law arises; and it is highly desirable that the decision of an [Employment] Tribunal should give guidance both to employers and trade unions as to practices which should or should not be adopted."

43. It is particularly the case that this Court should look at the arguments and evidence placed before a tribunal where it is dealing with the decision of an employment tribunal, as opposed to a court of record. In the case of an employment tribunal, the reasons are primarily addressed to the parties and will not establish a binding precedent.
44. NHT also challenges the substance of some of the Tribunal's conclusions. Where the conclusion is a finding of fact, the test is whether the Tribunal was entitled to reach the conclusion it reached for the reasons it gave, or, to put it another way, whether the reasons were perverse and incapable of sustaining the conclusion reached. Where the question is one of law, there will be an error of law if the tribunal misdirects itself as to the law. Some conclusions may involve both a finding of fact and an application of the law. An error of law may also occur where the tribunal accepts, or rejects, an argument but its reason for doing so does not in law constitute an answer, or a complete answer, to the argument in question.
45. On this part of the case, the following issues, which I will call B1 to B3, may be formulated on the basis of counsels' submissions:

B1. did the tribunal, in paragraph 8(u) of its decision (set out in paragraph 11 above) make a finding of sex discrimination? If so, was the pool of women a pool consisting of the non-RVI claimants and the RVI claimants or a pool consisting only of RVI claimants?

B2. was the tribunal entitled to find that the decision to put out to CCT the work of the domestic ancillary staff was tainted by sex?

B3. did the tribunal consider NHT's argument that the pattern of bonus payments made to ancillary staff was genuinely due to reasons stemming from the adoption and regulation of the bonus schemes? Did the tribunal consider whether that reason constituted justification for the disparity in treatment?

Issue B1

46. In my judgment in paragraph 8(u) of the tribunal's decision the tribunal was addressing Step 1 in the approach laid down in the Marshall case. As I see it, this sub-paragraph cannot be read as a finding of discrimination. As I have explained, the words "disparately adverse effect" are used in this field to denote the trigger at which a rebuttable presumption of sex discrimination arises. The tribunal dealt with the question whether there was sex discrimination at paragraph 29 of its decision when it concluded: "We do see the decision not to put out the predominantly male area to tendering as being tainted by sex".
47. Mr Supperstone submits that it is not clear whether the pool of women referred to in paragraph 8(u) is the pool of all the women in question or only the RVI claimants. In my judgment, it is the RVI claimants who are referred to in paragraph 8(u). That subparagraph appears in a group of subparagraphs headed CCT, which only applied to the RVI claimants. In those circumstances the tribunal could not logically be referring to analysis B (as defined in paragraph 14 above).
48. Mr Supperstone further submits that the tribunal could not reach the conclusion in paragraph 8(u) even in relation to the RVI claimants because the percentage of employees receiving bonuses was very small. He produced a schedule showing that out of the employees of all the hospitals as at 13 December 2004, there were only 55 staff in the bonus scheme as opposed to a total ancillary services staff of 1,153. On that basis the percentage in the bonus scheme was 4.78%. However, he does not suggest that the relevant percentage for the purposes of ascertaining whether there is a disparately adverse effect was other than that adopted, namely the percentage of women receiving bonuses as a percentage of all the women in the relevant group as against the percentage of the men receiving bonuses as a percentage of their whole group.
49. Mr Supperstone's submissions to the tribunal were to like effect as those made in this court, though the figures before the tribunal were those as at 2002 shown in analysis A not the later figures produced in this court. The reason why the number of employees participating in the bonus scheme had declined was that the scheme was no longer open to new entrants. However, Mr Supperstone does not suggest that the increase in the number of women not entitled to receive a bonus is due to some extraneous factor. The core of his case is that the number of male workers now entitled to bonuses is now statistically insignificant, and that that is sufficient to bar the RVI claimants' claim.
50. Mr Supperstone's submissions about the statistics derive from paragraph 62 of the judgment of the Court of Justice in the Seymour-Smith case, set out above. The Court of Justice directed national courts to consider whether the statistical evidence covered enough individuals. But since it is accepted that the statistical evidence covers all the relevant individuals, the fact that the number of persons disadvantaged is declining and at the time of the tribunal's decision is in absolute terms a small figure cannot destroy the value of the statistical evidence. The point goes to the genuine material factor defence based on the phasing out of the bonus schemes, which I consider below. In view of the substantial period covered by Miss Tether's analysis, and the consistency of the decline which her analysis reveals, the tribunal's finding is not, in my judgment, inconsistent with the principles laid down by the Court of Justice in the Seymour-Smith case which I have quoted above.
51. Mr Supperstone further submits that the tribunal did not consider the need for it to evaluate the statistics in order to reach the conclusion that the effect was disparately adverse for the purposes of the Seymour-Smith case. In my judgment, the figures set out in analysis A, which I have set out in paragraph 12 above, show that the disparity ran over a lengthy period and the resulting percentage was considerable. As I have said, no extraneous reasons were put forward for the disparity.
52. In those circumstances, in my judgment, the tribunal was entitled to reach the conclusion expressed in paragraph 8(u) of its decision. The tribunal's reasoning on this point can in my judgment be deduced by looking at analysis A. Mr. Supperstone's detailed objections to the statistics were set out in his closing submissions to the tribunal, and we must proceed on the basis that the tribunal considered and rejected them.
53. The tribunal made an express finding on disparate adverse impact. In my judgment, this court should not set this finding aside simply because the tribunal made it at the end of a long paragraph stated to contain findings of fact when this was a finding of mixed law and fact.
54. I would not therefore remit the issue of disparate adverse impact to the tribunal.

Issue B2

55. This issue is directed to step 2 in the approach laid down in the Marshall case. Mr Supperstone submits that the tribunal was not entitled to find that the process of tendering was tainted by the difference of sex. He submits that the tribunal did not give adequate reasons for its conclusion that there was sex discrimination. In my judgment, however, its reasons are to be found in the tribunal's decision. I deal separately with the adequacy of its reasons.
56. The core of the tribunal's reasoning appears immediately before its conclusion on this point and it is that the RVI claimants were:

"exclusively or almost exclusively female and the labour market for domestics in the north east was almost exclusively female. The porters were male. It was considered that one of the reasons why portering was not put out to tender was that as it was not compulsory to do so and more robust industrial action might be experienced . . ." (paragraph 29)

57. The passage quoted from paragraph 29 discloses three reasons:

(a) the RVI claimants were female and competitors were likely to employ females;

(b) it was not compulsory to put portering out to tender;

(c) if portering was put out to tender, the opposition from the workforce would be greater.

58. As to (a), paragraph 29 has to be read with the description of the CCT process in paragraph 8(m). The tribunal explained that the factors on which the decision to put a particular service out to tender was based included the size of the saving and "the perceived capability of private contractors to bid for the tender". In other words, one of the reasons why domestic work was put out to tender was that it was perceived that there were private contractors who would tender less than the cost to NHT of employing its existing domestic staff.
59. To that factor has to be linked the tribunal's finding in paragraph 29 that the tenderer's staff would be almost exclusively female.
60. Once those findings are made, then in my judgment, if those are the only relevant findings, it was open to the tribunal to find that the decision to submit domestic ancillary work to CCT was, in the words of section 1(3), due to "the difference of sex". The situation is very similar to that in the Ratcliffe case. In that case, the House of Lords held that a decision to put services out to tender to obtain savings can be tainted by sex where market forces lead to a lowering of pay of the complainants and that lowering of pay is conditioned by the fact that competitors employ mainly women who are prepared to work for less pay. As Lord Slynn, with whom the other members of the House agreed, held in Ratcliffe's case:

"Once it had to compete with others the council was in a very difficult position. On the other hand it is inescapable that when the evaluation had shown that women were being paid less than men for work rated as equivalent then it is impossible to say that the difference in pay was genuinely due to a material factor other than the difference of sex. Unlike the Employment Appeal Tribunal I do not find it necessary to remit this case for further findings nor do I consider that the industrial tribunal so misdirected itself that its findings cannot stand. The fact is that the council re-engaged the women at rates of pay less than those received by their male comparators and no material difference other than the difference of sex has been found to exist between the case of the women and their male comparators"

61. Mr Supperstone submits that the Ratcliffe case turns on its own facts. He points to the finding by the tribunal that in North Yorkshire where the women worked there was a general perception that women should stay at home and that they must effectively accept what was offered if they wanted to work. It is important not to read the decision too narrowly. It was not necessary for the applicants to show that the reason why competitors paid dinner ladies less than the applicants' comparators was because they consciously discriminated against women. As Lord Slynn observed, in the passage I have already cited, to permit employers to rely on market forces as the reason for reducing women's pay below that of their male comparators would be to permit the very kind of discrimination the 1970 Act sought to redress. In equal pay cases, it is necessary at all times to keep in mind the overarching social and economic purposes of the legislation.
62. However, issue (a) in paragraph 57 above does not stand alone. As to (b) and (c), NHT's case was that the decision to put domestic work, but not portering, out to CCT could not be said to be tainted by sex because it was not compulsory to put portering out to CCT, and a hostile reaction from the porters was expected. Mr Supperstone also submits that the women too were unionised and indeed went on strike when CCT was proposed. The point the tribunal was making was one of comparative judgment. The tribunal heard the evidence and, in my judgment, the tribunal was entitled to form a view as to whether the industrial action likely to be taken by the men was greater and more of a threat than that taken by the women. Nevertheless, it is apparent that the tribunal did not ingest the argument that the reasons for selecting domestic work for CCT and not portering was not due to a difference of sex but to these other reasons, which could have been genuine operational reasons. It is apparent from paragraph 31 that the tribunal was minded to accept that those reasons might have been genuine at that time, but it did not go on to consider whether a threat of industrial action or ideological opposition to CCT where it was not compulsory was the real reason for submitting ancillary work to CCT and not submitting portering to CCT. In addition, the final sentence of paragraph 29, which expresses the tribunal's conclusions on this point, proceeds on the basis that it was solely the decision not to put portering to CCT which was in issue, rather than that decision and the decision to put domestic work to CCT. While this last error might in some circumstances have been overlooked, in my judgment it underscores the fact that the tribunal did not undertake the structured approach required by the Marshall case. In those circumstances, I cannot uphold the tribunal's decision as to satisfaction of the requirement for the difference in pay to be due to the difference of sex. Accordingly, I do not accept the view taken by the EAT that these further matters could be treated as secondary to the Tribunal's conclusion on (a).
63. In paragraphs 30 to 31 of the decision, the tribunal went on to deal with the reasons put forward for the difference in its treatment of the RVI claimants and its treatment of the comparators. I will not at this stage examine the first and second reasons as the tribunal accepted that the first two reasons "might" have been genuine at the time. By contrast, the tribunal rejected the third and fourth reasons. I have set these out in paragraph 18 above.
64. The third reason (that the NHA management was not aware of local companies who might have been interested in tendering for the portering) was rejected on the ground that there was little, or no effort to explore the market for putting the portering out to CCT. I accept that this reason does not, as the tribunal states, go to the question whether the decision to put domestic work out to tender was tainted by sex, but it does demonstrate that the NHA's lack of knowledge about local companies who might be interested in tendering for portering might well have led to a train of enquiry as to whether the failure to put portering out to CCT while putting domestic work out to CCT was not genuinely due to a reason other than the difference of sex.
65. The fourth reason put forward by NHT was that opposition to a proposal to put portering out to tender when it was not compulsory to do was expected to be more voluble. In rejecting this the tribunal used its own local knowledge as to unionisation and the fact that the males were more likely to be full time.
66. Mr Supperstone submits that the claimants accepted that, if portering was put out to CCT, NHT would be subject to greater criticism as it was not compulsory to apply CCT to portering. He submits that this was one of the factors which justified the decision to put domestic catering but not portering out to CCT. Moreover Mr Supperstone submits that there was no evidence that the level of opposition would be affected by unionisation or by the fact that portering was predominantly full time. Accordingly, he submits that the tribunal could not rely on its own knowledge about this as it conflicted with the way the case was put. Furthermore, the tribunal should have taken into account that the savings from putting portering out to CCT were likely to be small.
67. I accept these submissions. The tribunal did not consider whether the factors identified in the preceding paragraph were the reasons for the decision to put the domestic work to CCT and, if so, whether the reasons for that decision were on their true analysis not tainted by sex. The inquiry on this could not stop at what I have identified in paragraph 52 above as issue (a). Mr. Supperstone may well be correct when he submits that the tribunal wrongly considered that the Ratcliffe case meant that the requirement for the variation in the contracts of the two groups of employees was sufficiently shown if the wages paid by competitors were lower because it was "women's work". I am satisfied that in any event the tribunal fell into error.
68. What the tribunal appears to have done is to proceed to the question of objective justification without first dealing with all the issues affecting the question whether the variation in pay between the applicants and their comparators was due to the difference in sex. That required it to look at all the factors which drove the decision to put the ancillary work to CCT at the time the decision was taken. It simply looked at the issues which were analogous to those that had arisen in the Ratcliffe case. Only if the tribunal was satisfied that that requirement was met was it necessary to show that the decision was objectively justified.
69. Accordingly, in my judgment, for these reasons, also, the decision of the tribunal, as to whether the employer's decision to place domestic work out to CCT led to a difference in pay which was not due other than to the difference of sex, cannot stand. I deal with the question of the appropriate relief below.

Issue B3

70. The essence of NHT's argument is that the various bonus schemes were a product of local collective bargaining. Moreover, as I have explained, NHT stopped the bonus schemes for new starters for financial reasons. Mr Supperstone calls this argument NHT's broader material factor defence. He submits that the tribunal did not take this broader material factor defence into account. For my part, I do not accept that argument. In the sub-paragraphs leading up to paragraph 8(u) the tribunal set out at length the history of the bonus schemes, for instance at paragraph 8(s) the tribunal noted that, on the creation of the new trust on 1 April 1980, entry into the existing bonus scheme was stopped for new starters and for employees accepting positions or altering the terms and conditions of their employment. The tribunal noted that this was a deliberate policy change as bonus schemes were seen as a financial drain on the new trust and it was intended to phase them out. Likewise, in paragraph 8(t), the tribunal noted that as a result of the new policy the number of staff receiving bonuses was steadily decreasing.
71. There is a separate point as to whether the tribunal's reasons for rejecting this broader material factor defence can be challenged on their merits. The tribunal dealt with this defence in paragraph 32. At the end of that paragraph it refers to the "historical factors" which are a reference back to the history which the tribunal had set out at length at the start of its decision and which I have summarised in the last paragraph. The tribunal had already quoted a passage from the speech of Lord Nicholls in the Marshall case, where he held that "a purely historic explanation of the pay difference between the sexes is insufficient". In paragraph 32, the Tribunal rejected the broader material factor defence on the grounds that the historical factors were not sufficient to justify the "maintenance of differentials which led to a disparate impact on women for a period of nearly fifteen years". In addition, the tribunal accepted Miss Tether's submission that no witnesses had been called properly to explain why the bonus schemes were being maintained in existence at the time when the appellants' claims were being submitted. However, this did not occur until about 2000. In other words, none of NHT's witnesses could speak to the reasons why the bonus schemes were not phased out earlier.
72. The tribunal's conclusions in paragraph 32 are not directed to the question of the difference in sex, but to the question of justification. In effect they amounted to an acceptance of the following submissions in Miss Tether's closing submissions to the tribunal:

"54. The approach taken by the Health Authority was clearly shaped by Department of Health Circular 83/18. It is common ground that although this circular was expressed as a request, it made it mandatory for health service employers to market test catering, laundry and domestic services. Mr Spry accepted that if portering services had been covered by Circular 83/18, the Authority would eventually have put the portering service out to tender. As Messrs Whitfield and Wing pointed out, the policy instituted by Circular 83/18 was itself indirectly discriminatory, since the compulsion to tender was limited to services which were predominantly female …"

62. In the light of the foregoing, it is submitted that the Respondent cannot sustain a defence under section 1(3) of the EPA unless it shows that the differences in bonus which are the subject of these proceedings are objectively justified. In order to satisfy this test, the Respondent would have to show that the payment of incentive bonuses to some staff but not others corresponds to a real business need, is necessary to achieve the objective in question and conforms to the principle of proportionality …

63. The Applicants submit that the factors which are relied on by the Respondent come nowhere near providing an objective justification of the differences in bonus. Indeed, it is conspicuous that paragraph 8.1 of the Respondent's Notice of Appearance states simply that the payment and non-payment of bonuses to ancillary staff "is explained historically" by the factors which are described in the ensuing paragraphs. As Lord Nicholls pointed out in Glasgow City Council, a purely historic explanation of a pay difference is insufficient where justification is in point …

70. It is striking that the Trust has not called any witnesses who could explain why the bonus schemes were being maintained in existence at the time when the Applicants' claims were submitted. Mr Gooden retired from his personnel post in April 2000. Mr Holmes, the Finance Director, commented on the Trust's financial situation but surprisingly could remember nothing of the discussion concerning bonuses which took place at the Board meeting in September 2000.

71. In view of the foregoing, it is submitted that the historical factors which are relied on by the Respondent as explaining the differences in bonus which are at the heart of this case cannot objectively justify the maintenance of differentials which have had a disparate impact on women over a period of nearly 15 years i.e. from the late 1980s to 2002/2003 …

73. Furthermore, historical forces which have exhausted their power are incapable of constituting a material difference between the case of the applicants and their comparators. In Benveniste v University of Southampton [1989] IRLR 122 [Joint Authorities, tab 16], the respondent university sought to defend a pay difference between the applicant and her comparators by showing that she had been appointed at a time of financial constraint. The Court of Appeal held that as the financial constraints which applied at the time of the applicant's appointment came to an end at the end of that academic year, the special factors which justified her lower salary had disappeared. Accordingly, the material difference between her case and that of her comparators had evaporated – see paragraphs 27, 30 and 31."

73. The tribunal did not, however, deal with the broader material factor defence on this point in its entirety. NHT's case was that the policy was to phase bonus schemes out to avoid the financial drain which they caused. The bonus schemes could not simply be withdrawn from those entitled without inviting legal or industrial action. NHT's case was that the financial position of the trust and the increased cost to the trust of paying bonuses to all staff meant that it was not feasible to pay bonuses to all staff. In my judgment, this defence is not based on purely historical factors. It is also based on financial and legal reasons, as well as reasons connected with industrial relations. The tribunal did not express a conclusion on this defence properly understood. It was contested by the applicants who contended that the NHT had failed to prove that any financial pressures were the reason for maintaining the bonus schemes (applicants' closing submissions, paragraph 70). Accordingly, in my judgment, the EAT should have held that the tribunal's decision on this issue could not stand. I deal with the question of the appropriate relief below.
74. Mr Supperstone's further submission is that it was, in any event, not clear from paragraph 32 as at what date the tribunal found that the defence was not made out and this would give rise to difficulties later when remedies were being considered. In my judgment, while on the conclusion reached in the last paragraph this point does not now arise, the answer to it is to be found in the final sentence in paragraph 32 where the tribunal refer to a period "of nearly fifteen years", i.e. from about 1 January 1988. This is consistent with Miss Tether's closing submissions set out above.

Appropriate relief

75. I now turn to consider the appropriate form of relief in the light of my conclusions on the genuine material factor appeal brought by NHT. Since writing this judgment I have had the benefit of reading in draft the judgments of Buxton and Latham LJJ.
76. As this appeal is on a point of law only, the appropriate relief if the appeal is allowed on any issue is to remit that issue for determination by the tribunal, unless this court is satisfied that no tribunal acting reasonably could find for the RVI claimants. Accordingly if, for example, the difference in treatment for bonus scheme purposes between the RVI claimants and their male comparators could not be found to be due to the difference of sex, it would not be right to remit that issue to the tribunal.
77. I consider first whether any and if so which issues relating to discrimination should be remitted to the tribunal. As I see it, the claimants put their case on discrimination to the tribunal in two ways: (i) the tendering of domestic work was discriminatory and (ii) the tendering of domestic work, and not portering work, was discriminatory.
78. The RVI claimants' first way of putting their case is apparent from Miss Tether's closing submissions. At paragraph 50, she submitted:

"50. In the first place, it is common ground that the domestic staff at the RVI lost their entitlement to bonus as a result of the competitive tendering which took place in 1985. It is clear from the House of Lords' decision in North Yorkshire County Council v Ratcliffe and others [1995] ICR 833 [Joint Authorities, tab 15] that a perceived need to compete with an external tenderer does not constitute a difference other than the difference in sex between women and men who are engaged on equal work – see the speech of Lord Slynn at pp836H-837E and pp840-841D.

…53 It is no answer to the point made by Lord Slynn to say that, if services in which the workforce was predominantly male had been exposed to the market, the wages of the men would also have been reduced. The fact is that in this case services were not put out to tender, with the consequence that tendering led to inequalities in pay between domestics who were exclusively or almost exclusively female and porters who were exclusively male."

79. Mr Supperstone responded to this argument at paragraphs 93 to 98 of his closing submissions. He did not submit either to us or to the EAT that the RVI claimants had put their case this way for the first time only in the course of the appeal to the EAT. For the purposes of this argument, the disparity with the porters was simply the result of the discrimination.
80. I would also refer to the quotation from Miss Tether's closing submissions to the tribunal set out in paragraph 72 above. The argument there developed is based not on discrimination resulting from the operation of adverse market factors but on discrimination resulting from the decision to put the domestic work out to CCT rather than the portering work. This argument was developed so as to include a further contention, namely that the maintenance of the bonus schemes for the porters was discriminatory.
81. Both issues in paragraph 77 above were also before the EAT although Miss Tether submitted that the first way of putting the case was her primary contention. In point of fact, the decision not to put portering out to tender was part of the same sequence of events as that of putting the female work out to CCT and therefore naturally formed part of the tribunal's enquiry on both issues.
82. In my judgment, this is not a case where the question whether the difference in treatment was due to the difference of sex could only be decided one way by the tribunal, (viz. against the RVI claimants) because of the commercial and policy reasons which drove the decision to put domestic work, but not portering, out to CCT. It is to be noted that the tribunal reached no concluded view on the first two points in paragraph 30 of their decision, i.e. whether the reasons for the disparity were that it was not compulsory to put portering to CCT and the question of financial savings. Buxton LJ has described the policy in relation to CCT and the handling of bonus schemes in cases where services were put out to tender, as detailed and thought out. However, the mere fact that an employer has a carefully calibrated scheme, for example for assigning employees to different pay grades, does not mean that the implementation of that scheme could not be said to be tainted by sex if it turns out that all (say) the male employees are in the top grade and all the female employees in the lowest grade. The tribunal has to determine whether the non-gender based policy and other considerations in this case were the genuine reasons for the pay disparity, or whether there was simply a witting or unwitting adoption of a practice which subjected female employees to market forces which would depress their pay. I have put to one side for the purposes of this example the fact that the Catering and Laundry Departments were put out to tender but retained their bonuses, but clearly that matter will have to be considered in due course.
83. It is not desirable that I should comment in depth on the validity of the Trust's reasons in paragraph 30. However, it may be difficult for a mere philosophical objection to CCT to have the effect of preventing a difference in the treatment for pay purposes of male and female workers being a difference attributable to a factor other than the difference of sex.
84. Moreover, the evidence available to this court is only a selection of the evidence before the tribunal. That affects both the question of the difference of treatment being due to a factor other than the difference of sex and the defence of objective justification. The thrust of the case which the tribunal accepted in paragraph 32 of its decision was that, although there might have been legitimate reasons for putting the domestic work out to tender originally, the maintenance of the disparity between the two groups of employees was not justifiable. The bonus schemes were being phased out but it is a question of fact, which as I see it this court cannot determine, as to whether that was an adequate and proportionate reason for not equalising the pay of the two groups of employees in some way.
85. For the reasons given above, this case should in my judgment be remitted to the tribunal on both aspects of the question whether the decision to put out the ancillary work to CCT was due to the difference of sex and in addition on the question of objective justification.
86. I consider that this court should make a formal direction to the tribunal to limit evidence on issues remitted to it to that which was originally before it, together with a further version of analysis A (referred to in paragraph 12 above) updated to the latest practicable date. In my judgment, that course is a proportionate and fair way of dealing with such issues.

Disposition

87. For the reasons given above, I would dismiss the wrong comparator appeal and allow the genuine material factor defence appeal. I would make the order for remission indicated above.
88. Before leaving this case, I add these observations. This case illustrates the complexities of equal pay cases. The tribunal has sought to reduce them by its case management directions for hearing the wrong comparator and genuine material factor issues as preliminary issues. That may or may not have been the right course. Each of those issues, however, required detailed examination of the facts and law. A particular difficulty on the second issue is that factors relevant to step 2 in the approach laid down in the Marshall case (dealing with the issue whether the variation in pay was due to the difference of sex) were also relied on by NHT as showing that, if the variation was due to the difference in sex, there was objective justification for the employer's decision. It is little wonder that the tribunal, faced with a considerable volume of evidence and complex legal principles, fell into error. I would express the hope that some way is found of simplifying these important cases for the future. That observation in no way diminishes the importance of these issues being dealt with, as Parliament has determined, by an employment tribunal, given its superior experience in employment and industrial matters.

Lord Justice Latham:

89. I have read in draft the judgments of both Lady Justice Arden and Lord Justice Buxton. I agree with them both in relation to the "wrong comparator" appeal. As to the "genuine material factor" appeal, whilst both are agreed that the decision cannot stand, they are not wholly agreed as to the reasoning, and, differ to the same extent as to the appropriate disposal.
90. In my view, the issue depends essentially on the Tribunal's reasoning in paragraphs 30 and 31 of its determination. Paragraph 30 set out the Tribunal's understanding of the employer's reasons. The first two, namely that the porters were not put out to tender because it was not compulsory to do so and the employers were philosophically opposed to the idea, and that major savings would not be made in this area because the budget for portering was less than in other groups, were accepted by the Tribunal, as was inevitable on the evidence, and clearly formed the core of its justification. As to the third and fourth, the employers were not aware of local companies who might be interested in tendering and that opposition to put out to tendering an area which was not compulsory was expected to be more voluble, were not rejected by the Tribunal, it merely expressed concern. It noted that little or no effort had been made to explore the market and expressed the view from their own experience that more males tend to be in full time employment than females and to be unionised, and stated that the male porters were in employment traditionally more heavily unionised in volume.
91. As to the concern expressed in relation to the third of the justifications it identified, I simply fail to understand the Tribunals conclusion that it was "significant and relevant" that it was a factor tainted by sex. The Tribunal would have been entitled to conclude from the failure to explore the market that it was not a true reason for the decision. But the Tribunal did not say so. As to the assertion that the male porters were in the employment traditionally unionised and voluble based on the Tribunal's experience of the local market, this, as Lord Justice Buxton has pointed out, cannot be supported on the evidence. To come to that conclusion the Tribunal would have had to explore the extent of unionisation within the relevant hospitals of women employees. The experience of the local market was simply an irrelevance to that question.
92. I agree with Lord Justice Buxton that on the material before the Tribunal, it was not possible for it to hold that the decision not to put the portering services out to tender was discriminatory. Indeed my conclusions in relation to this part of the Tribunal's reasoning are such that I had serious doubts as to whether it was necessary to remit any of the issues to the Tribunal on the basis that there could only be one answer as a result to the question as to whether or not the Trust had established that any disparate adverse effect was due to a genuine material factor other than sex. But having considered the matter in the light of the draft judgments of the other two members of this court, I have concluded that it is not open to us to take such a robust approach. I do so with some reluctance. This is a long running saga which should be brought to an end as soon as possible. And I find the attempt to use a policy of phasing out what is clearly an anachronistic bonus scheme as a justification for reinstating it unattractive.
93. Like Buxton LJ and for the same reasons, I do not consider that the Tribunals' finding on disparity can stand. As we are remitting the matter to the Tribunal anyway, I agree with Buxton LJ that this aspect of the case must also be reconsidered by the Tribunal. I therefore agree with his judgment, and his proposed disposal.

Lord Justice Buxton :

94. On the "wrong comparator" appeal I respectfully agree with the conclusion proposed by Arden LJ, and there is nothing that I wish to add. In relation to the "material factor" I have the misfortune to differ from the EAT and, to some extent, from Arden LJ. I therefore need to explain my conclusions on that topic. I hope in so doing to avoid adding to the "legal materials on indirect discrimination and equal pay [that] are increasingly voluminous and incredibly intractable", as characterised by this court in Secretary of State for Trade and Industry v Rutherford (No2) [2005] ICR 119[3].
95. This is a most unhappy case. The originating application of the representative applicant, Mrs Anderson, was filed in May 2001. It complained of decisions originally taken in the mid and late 1980s, affecting an employment that Mrs Anderson had held since 1987. The Employment Tribunal considered the case over a period of seven days in November 2002 and May 2003, and gave its decision in December 2003. It found that the incidence of the bonuses that are at the centre of the case had significantly declined since their original introduction, so that by the operative date of the decision, September 2002, only 111 employees (split about 60:40 between men and women) were receiving bonus. But because those 111 included eighteen RVI porters, who are men, the effect of the Employment Tribunal's decision is to require the reintroduction of bonus payments to some 350 women ancillary workers who do not currently receive them; and, no doubt, as a matter of practical industrial relations, also to the over 100 male workers who do not currently receive bonus. How that structure, conceived in the 1960s, is to be implemented in the world of 2006 remains obscure, at least to me.
96. The EAT heard appeals in October 2004, and with commendable despatch delivered its conclusions in November 2004. This court heard appeals in July 2005, and expresses its strong regret that the difficulties of the case have caused it only to deliver judgment after a period of significant delay. Not the least of the court's problems, and an indication of the difficulties of the proceedings in the court below, was that the only relief that the Trust felt able to seek from this court was a new trial: that is, remission to the Employment Tribunal. I have been persuaded, with very considerable reluctance, that such a remission is required, and that it is not possible, in the light of the way in which the case and the appeals have proceeded, for this court to determine the issue of substance as to whether the tendering process that led to the removal of bonus from the RVI domestics was, as the EAT put it in its §42, tainted with sex discrimination.
97. I first make some observations about the general state of the law that the Employment Tribunal was obliged to apply, and then turn to the development of this case.

Statistics

98. It would be to fight a battle long since lost to point out that the figures relied on in this case, as in other indirect discrimination cases, are just that: raw figures. They have not been subjected to any form of statistical analysis, to determine, for instance, whether the variations in incidence of bonus payments as between men and women exceed the limits of randomness; whether the numbers involved are sufficient for meaningful analysis; and whether other factors alleged to affect the payment of bonuses have a causally more significant impact than does the incidence of gender. Without such analysis the figures, taken on their own, prove nothing. It would appear, from paragraph 60 of the opinion of Mr Advocate-General Cosmas in R v Employment Secretary ex p Seymour-Smith [1999] ICR 447, that the European Commission in its submissions in that case sought to introduce statistical analysis into this chapter of the law; but the Advocate-General characterised that move as American-inspired and not in accordance with the settled case-law of the court, a view that was clearly shared by the Court of Justice itself.
99. As a result, as this court put it in Rutherford (No2) [2005] ICR 119[35],

"It has been left to national courts and tribunals…..to work out from case to case a satisfactory method for assessing whether or not there is disparate adverse impact in the particular case. It is a matter of applying considerations of logic, relevance and common sense to the raw material of the statistical evidence in order to determine the existence or otherwise of the objectionable state of affairs"

100. This lack of a clear theoretical rationale for whatever conclusions are drawn from the figures has caused very great difficulty in deciding cases on a consistent basis and in the light of agreed verbal statements of the rules, as was demonstrated by the disagreement on those points when the Seymour-Smith case reached the House of Lords, [2000] ICR 244. And substantively, because it is unclear what type of factual conclusion can be proved by figures of this sort, they have played only a limited role in indirect discrimination cases, as analysis of the structure of that chapter of the law demonstrates.

A taxonomy of indirect discrimination

101. Some of the multiplicity of citation of which this court complained in Rutherford (No2) may be avoided if, in dealing with a part of the law drawn from or influenced by international rules or conventions, in this case the law of the European Community, we remind ourselves that once the House of Lords has determined the meaning of those rules it is not open to domestic courts to resort to the decisions of the Court of Justice on which the House of Lords based its analysis in order to find a different or wider meaning: see Leeds City Council v Price [2005] 1 WLR 1825, and Commissioner of Police for the Metropolis v Hurst [2005] EWCA Civ 890[25]. In the present case, therefore, the proper approach to cases of indirect discrimination not only starts but also finishes with the guiding passage of the speech of Lord Nicholls of Birkenhead in Glasgow City Council v Marshall [2000] ICR 196, cited by the Employment Tribunal in paragraph 23 of its determination, where his Lordship said:

"The scheme of the Act is that a rebuttal 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 the 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 [that the reason relied upon is not the difference in sex] 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 of indirect. If there is any evidence of sex discrimination, such as evidence that the difference in pay has a disparately adverse effect 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."

102. If further exposition is required, it is to be found in the judgment of this court in Nelson v Carillion [2003] ICR 1256[39], where Simon Brown LJ (as he then was) said:

"It is for the applicant to show on a balance of probabilities a disproportionate adverse impact and thereby establish a prima facie case that she had suffered indirect discrimination."

103. The figures, if they demonstrate a disproportionate or (what seems to be the same thing) disparately adverse effect on female employees, thus open the door to a claim of indirect discrimination, by placing on the employer the burden of proving that the variation in terms between the woman's and the man's contract is "genuinely due to a material factor which is not the difference of sex": Equal Pay Act 1970, section 1(3). How he discharges that burden was the major matter of dispute in this case. However; we must first address the legally prior issue identified in Nelson v Carillion, cited in paragraph 99 above, of whether the applicant has shown on a balance of probabilities a disproportionate adverse impact.

The findings in the tribunals below on disproportionate adverse impact

104. The Employment Tribunal's only reference to this issue came in its paragraph 8(u):

"an analysis of the effective bonus arrangements appended to the applicants' closing submissions nevertheless shows that from 1988 onwards they have had a disparately adverse effect on women as shown by Ms Swanson's evidence contained in her second supplemental statement"

105. However, as Mr Supperstone pointed out, paragraph 8 is stated by the Employment Tribunal to be a list of findings of fact. But, as the citation from Rutherford in § 93 above demonstrates, "disparately adverse impact" is not a matter of fact, or at least not simply one of fact, but a matter of judgement. That is not merely a technical objection, since a finding of disparate adverse impact can only be reached, as Rutherford required, by the application of logic, relevance and commonsense. There is no sign in the Employment Tribunal's reasons that it deployed those considerations, or of how it would have addressed them had it done so.
106. The EAT did not directly confront this problem, but in its §38 it referred to the disparities between men and women identified in the speech of Lord Nicholls in Seymour-Smith, and concluded that the Employment Tribunal "was entitled to come to conclusions that it did based on the Applicants' statistical evidence, showing greater and more persistent disparity than in the Seymour-Smith case". But the percentages in Seymour-Smith were based on the whole of the British workforce. It cannot be enough, in applying the approach required by Rutherford, to take the raw numbers extracted from the context of that case and apply them without more, as some sort of guideline or bench-mark, to reach conclusions that relate only to a limited number of workers in a particular employment. It may be that further exploration of the latter issue would lead to a conclusion that the Employment Tribunal's finding, however inadequately reasoned and expressed, could be supported; and that in that process the outcome in Seymour-Smith was of potential relevance even if it could not be determinative. But, as it stands, the EAT's analysis does not respect the need to assess disparate adverse impact separately from case to case, as this court required in Rutherford.
107. I should also mention, for completeness, that the EAT appears, from their rejection in their §37 of an argument in the contrary sense by Mr Supperstone, to have held that a finding of disparate adverse impact in the case of given arrangements amounted to a finding that those arrangements were indirectly discriminatory. Such a view is inconsistent with the guidance of the House of Lords and of this court, cited in §§ 101 and 102 above. None of the cases mentioned by the EAT in this connexion carry the point, and the only quotation relied on, from §4 of the judgment of Mummery LJ in Rutherford, was not concerned at all with this issue, but rather with explaining why the EAT in that case was not to be criticised for taking 226 paragraphs to address what, absent authority, would be a comparatively straightforward issue. But the EAT do not seem to have treated their view as conclusive of the appeal, as in logic it would be if it were correct, and it is not therefore necessary to pursue that issue further.
108. What does, however, remain is that neither the Employment Tribunal nor the EAT demonstrated adequate grounds for their conclusion that disparate adverse impact existed in this case. If that conclusion cannot stand, then the applicant fails at the first hurdle, as explained by this court in Nelson v Carillion. I am conscious that, although both tribunals' handling of this issue was criticised by the respondents, the appeal was not argued in quite the same manner as set out above; but in view of the fundamental importance of this issue in the approach to complaints of discrimination that is mandated by the House of Lords in Glasgow City Council v Marshall the point cannot be left where it lies. Granted that remission to the Employment Tribunal has to take place in any event, that process should extend to the making of proper findings on the issue of disparate adverse impact, in the light of the guidance quoted in §99 above.
109. If the Employment Tribunal, on reconsideration, concludes that disparate adverse impact is not established, then there is no need for the Trust to go on and prove a genuine material factor defence. However, since that issue remains to be determined I must go on to consider for the purposes of this appeal how the case stands in respect of that defence.

A material factor which is not the difference of sex: the structure of the argument

110. As Lord Nicholls said at the end of the passage from Glasgow City Council v Marshall cited in §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 § 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. I deal with those issues in turn.

The discrimination argument in this case

111. I have not found it at all easy to elucidate how this issue was addressed by the applicants (who bore the burden of establishing an arguable case, even though the employers bore the burden of persuasion); or by the Employment Tribunal. It is necessary to look in some detail at how the point was handled. Arden LJ has set out the history of reorganisation of the hospital workforce and of the contingent effect on bonus payments, but for ease of reference the findings of fact of the Employment Tribunal in its paragraph 8 (m) – (p) and (s)-(t) bear repetition:

"Compulsory Competitive tendering

(m) Compulsory competitive tendering (CCT) was introduced to the NHS in the mid 1980's by Circular HC83/18 (8/Tab7). The circular was effectively mandatory in relation to domestic, laundry and catering services and health authorities were put under considerable pressure to conduct market testing. They were also encouraged to consider testing in other areas. The NHA was philosophically opposed to contracting out, but did consider it as a result of both government pressure and the need to make cost savings. It considered it across the whole range of ancillary services, not simply domestic, catering and laundry. The decision to put one particular service to tender rather than another was based on size of the particular budget and, therefore, the saving that could be made together with the ease of specifying the service required, the perceived capability of private contractors to bid for the tender as well as the need to comply with government policy. This did, however, have the effect that a decision was made to put out to tender domestic, laundry and catering services at the RVI which were services predominantly staffed by women whereas portering, which was predominantly staffed by men, was not put out to tender. This was not necessarily the case throughout the health service. For example, in Bolton and in Bradford portering services were put out to tender.

(n)……Managers drawing up the in-house bids had a judgment to make as to whether it was necessary to adjust terms and conditions of staff in order to be competitive with the private sector. In certain cases, this led to a decision to forego bonus arrangements……

(o) As a result the domestics of the RVI did lose their bonuses when a bid was put in based on the staff losing the entitlement within one year.

(p) The RVI Catering and Laundry Departments won tenders without abandonment of their bonus scheme. Portering was never put out to tender. This was justified on the basis that the budget was limited and the scope for savings small. Also, it was suggested that those responsible were not aware of local private sector companies which offered such services in the Newcastle area. It was not mandatory under HC 83/18. This being the case, opposition to a proposal for competitive tendering was expected to be "more voluble".

(s) On creation of the new Trust on 1 April 1980 entry into the existing bonus schemes was stopped for new starters and for any employees accepting positions or altering the terms and conditions of their employment. This was a deliberate policy change as bonus schemes were seen as a financial drain on the new Trust and it was intended to phase them out. In all of the circumstances, staff transferred to the new Trust with their terms and conditions intact. The new Trust inherited an underlying financial deficit of £9.4 million, mostly attributable to the RVI, and although it has been successful in managing its income and expenditure account thereafter, this has only been achieved through a range of non-recurrent measures and a significant underlying deficit remains and is increasing.

(t) As a result of the new policy, the number of staff receiving bonuses is steadily decreasing. Fifty-six RVI porters (fifty-five male, one female) and fifty-eight RVI caterers (fourteen male, forty four female) were receiving bonuses in April 1988, as against only eighteen (all male), and twenty-four (nine male, fifteen female) respectively as at September 2002. Something in the region of 200 staff were receiving bonuses in April 1998, as against 122 in February 2002 and 111 at the end of September 2002. As a result of the bonus freeze in 1988, the value of the bonus received is also steadily declining."

112. These findings set out a detailed and thought-out policy in relation to the adoption of CCT, and of the handling of the bonus scheme in those cases where services were put out to tender, which on its face has nothing at all to do with the gender of those employees affected by it. As is apparent from §§ 29-31 of their determination, cited below, the Employment Tribunal approached the issue of discrimination through the policy adopted towards the comparators, the RVI porters. The contention, accepted by the Employment Tribunal, was that the decision not to put the portering services out to tender, thus shielding them from the need at least to consider removal of bonus in order to remain competitive, was influenced by the gender of the employees involved. The Employment Tribunal said this:

"29 [Counsel for the applicants] relied upon the facts of the present case as being in important respect similar to those which form the backdrop to North Yorkshire County Council v Radcliffe and others [1995] ICR 833. The domestics of the RVI were exclusively, or almost exclusively, female and the labour market for domestics in the north east was almost exclusively female. The porters were male. It was considered that one of the reasons why portering was not put out to tender was that as it was not compulsory to do so a more robust industrial reaction might be experienced. We do see the decision not to put out the predominantly male area to tendering as being tainted by sex.

30. The respondent relies upon the following explanation for difference in treatment, namely:

(i) the porters were not put out to tender because it was not compulsory to do so and the NHA was philosophically opposed to the idea;

(ii) major savings would not be made in this area because the budget for portering was less than in other groups;

(iii) the NHA management was not aware of local companies who might have been interested in tendering in this area;

(iv) opposition to a proposal to put out to tendering an area which are not compulsory was expected to be more voluble.

31. In considering the above reasons given or the tests set out in Glasgow City Council case we can accept the first two of the above reasons might have been genuine at the time. We are concerned about the second two as there appears to have been little or no effort to explore the market and the male porters were in employment traditionally more heavily unionised and voluble because we know from our own experience of the local market more males tend to be full time than females and to be unionised. These two alleged justifications amount to significant and relevant factors tainted by sex and strike at the basis of the justification claimed."

113. It will be seen from the foregoing that the Employment Tribunal accepted that North Yorkshire County Council v Ratcliffe was relevant to the issue of discrimination. In Ratcliffe the employers were, as in our case, faced with CCT. They saw themselves as compelled to reduce the wages of female employees in order to tender in a market in which the workforce was exclusively female, and for that reason had depressed conditions and protections. That was accordingly a reason for the reduction of the female employees' wages below those of their male comparators; a reason that the Employment Tribunal was entitled to find was attributable to sex, and not merely to the need to retain the business. That same implication was thought by the EAT in our case to apply to the decision to remove the bonus from the RVI domestics, a matter to which we will have to return.
114. But the complaint about the privileged treatment of the male RVI porters could have nothing to do with the factor that concerned the courts in Ratcliffe, of the effect of decisions taken in an exclusively female work-market. Accordingly, and with some difficulty, I am driven to conclude that the references to Ratcliffe can only indicate an extension of the allegation of discrimination from a complaint of negative discrimination, that the porters were not put out to tender thus exposing the female domestics to competition in a gender-unbalanced market; to a further complaint of positive discrimination, that to require the domestics to compete in the market as it then stood was in itself necessarily discriminatory. And that was the view of the EAT, which at its §42 recorded its acceptance of a submission by counsel for the applicants that

"the issue of portering services was secondary to the Tribunal's main finding in relation to the process of competitive tendering being tainted with sex discrimination arising out of the domestics being part of a labour market which was almost exclusively female"

Comparison with the actual exposition of the Employment Tribunal, as set out in §18 above, does make that analysis, and certainly the claim that what might be called the "Ratcliffe" point was the principal issue as to discrimination in the Employment Tribunal, a thing of some difficulty. Nevertheless, no objection appears to be taken before us to the EAT's approach, and it is therefore necessary to consider separately the Employment Tribunal's handling of the portering issue; and the Ratcliffe question.

Discrimination: the failure to put the portering services out to tender

115. Ground II of the Grounds of Appeal to this court contended that the Employment Tribunal's conclusion that the decision not to put the portering services out to tender was tainted by sex discrimination was wrong in law. I agree. Of the reasons for that decision summarised by the Employment Tribunal in its §30, the tribunal accepted the first two, subject to an issue as to persistence over time that I discuss further below. Its rejection of the second two reasons are discussed in the tribunal's § 31. There are difficulties in both the latter cases.
116. First, as to the failure to explore the market, I find it impossible to see how that could demonstrate any gender-related influence, as opposed to incompetence or failure to follow through the stated policy. It would of course be different if the employer, in order to avoid putting a male-dominated service out to tender, had disingenuously claimed that there were no alternatives, or deliberately avoided looking for alternatives; but none of that has ever been suggested. On the evidence, it is impossible to argue that this lapse, if it was one, occurred because the porters were men.
117. The stated reluctance to encounter opposition on the part of the workforce in a case where the arguments for putting the service out to tender were already weak would potentially be attributable to the gender of the employees concerned if it were possible to establish that the porters were significantly more likely to cause trouble than the employees who had made them their comparators, and that that attitude sprang at least in part from their being men. The Employment Tribunal so found, but it did so on the basis of its "experience of the local market", and not on the basis of any evidence as to the workforce with which this case is actually concerned. Indeed, as the employer pointed out, such evidence as there was indicated a significant level of union activity amongst its female employees, one example of which was specifically recorded in the tribunal's §8(o): union activity being the mechanism through which the tribunal thought that opposition would be expressed by male employees.
118. An Employment Tribunal is of course well permitted to rely on its own knowledge in deciding an issue to which that knowledge is relevant. But where the issue is whether the workforce of a particular employer had certain characteristics, the tribunal cannot apply general knowledge and nothing more to that particular case unless it has established that no other source of information is available. That the Employment Tribunal never did. Had it taken that step, it would in my view have been forced to hold that the decision not to put the portering services out to tender was not influenced by the gender of the porters.
119. I would therefore hold that it was not open to the Employment Tribunal, on the evidence before it, to hold that the decision not to put the portering services out to tender was discriminatory.

Discrimination: the Ratcliffe issue

120. It is much less easy to analyse this issue, not least because of the indirect way in which it has entered the appeal, as already indicated. In the light of the history of the point it is hardly surprising that the grounds of appeal and skeleton of the Trust concentrated very largely on the portering issue, and hardly touched at all on this point.
121. Despite the centrality that this issue now claims, it will have been seen from §29 of the Employment Tribunal's determination, set out in §112 above, that the tribunal merely stated that the facts of Ratcliffe were similar to those in our case, and never considered how, if at all, the law expounded in that case related to the issue of discrimination on the facts of our case. The EAT by contrast did accept the relevance of Ratcliffe, as already noted. The EAT does not reason out that conclusion, but it appears that it rests on the fact that the labour market for domestics was almost exclusively female, read in the light of what was said by the House of Lords in Ratcliffe. Those two matters, taken alone, would however only be dispositive if the House of Lords had held in Ratcliffe that to adjust wages or conditions in order to compete in a predominantly female labour market was necessarily discriminatory either as a matter of fact or as a matter of law. But the House of Lords did not so hold. In the central passage of the leading speech of Lord Slynn of Hadley, cited by the EAT in its §42, his Lordship said, [1995] ICR at pp 840 B-E:

[The industrial tribunal was] satisfied that the [employer] had failed to show that the variation between the applicants' contracts and those of their male comparators was due to a material factor which was not the difference of sex. In my opinion it is impossible to say that they were not entitled on the evidence to come to that conclusion. It is obvious that the [employer] reduced the applicants' wages in order to obtain the area contracts and that to obtain the area contracts it had to compete with CCG which, the tribunal found, employed only women and "because of that, employed them on less favourable terms than the [employer] did previously under the NJC agreement"….The fact, if it be a fact, that CCG discriminated against women in respect of pay and that the [employer] had to pay no more than CCG in order to be competitive does not however conclude the issue. The basic question is whether the [employer] paid women less than men for work rated as equivalent. The reason they did so is certainly that they had to compete with CCG. The fact, however, is that they did pay women less than men engaged on work rated as equivalent. The industrial tribunal found and was entitled to find that the council had not shown that this was genuinely due to a material factor other than the difference of sex……The fact is that the [employer] re-engaged the women at rates of pay less than those received by their male comparators and no material difference other than the difference of sex has been found to exist between the case of the women and their male comparators. [emphasis supplied]

122. I venture to draw from those observations the following conclusions relevant to this case.
123. First, Lord Slynn's emphasis on the primacy of the decision of the Employment Tribunal suggests that the EAT in our case should have confined itself to the basis on which the case was approached by the Employment Tribunal. If the question is posed in terms of whether on that basis the Employment Tribunal was entitled to come to the conclusion that it did, the answer is in the negative, because the Employment Tribunal failed to reason out the application of Ratcliffe to the facts of the case before it.
124. Second, Lord Slynn saw the employers as having shown no material difference between the applicants in Ratcliffe and their comparators other than the difference of sex. That is not so in our case. The complex history and effect of the introduction of CCT, fully set out by the Employment Tribunal, indicates that the different decisions or outcomes in relation to bonus payments were the result of a very wide range of factors. That is no doubt why the case originally concentrated on the decisions as to whether the various services should be put out to tender at all, with the resulting claim that it was the treatment of the men rather than of the women that was influenced by gender.
125. Third, we have already noted that Lord Slynn did not regard as conclusive that the employers had to tender in a market staffed largely by women, and against a competitor who discriminated against women. That was despite the Employment Tribunal in that case having received detailed evidence as to the employer's motivation during the tendering process, and having found (as cited by Lord Slynn, [1995] ICR at p 837D) that wage differentials between the sexes arose from :

"the general perception in the United Kingdom, and certainly in North Yorkshire, that a woman should stay at home to look after the children and if she wants to work it must fit in with domestic duty and a lack of facilities to enable her, easily, to do otherwise."

By contrast, the Employment Tribunal and the EAT in our case knew little or nothing about the detailed tendering process or about the market that it was addressing. I appreciate that there were difficulties in recovering what exactly had happened some fifteen years before the hearings, but the employers, as respondents, can hardly be blamed for the delay in bringing the proceedings. And if that history had been explored, by either tribunal, it would have been necessary to reconcile findings of discrimination with the fact that bonuses were retained by the Catering and Laundry departments, found by the Employment Tribunal to have the same dominant characteristic as the domestic service, that they were predominantly staffed by women

126. Given the assumption (which it will have been seen that I accept with some reluctance) that the Ratcliffe argument was on the agenda of the Employment Tribunal, and thus is on the agenda of this court, I am driven to conclude that it was not properly addressed by either tribunal below. I see no alternative to remitting this part of the case to the Employment Tribunal, for it to consider, in the light of the observations of this court, whether on the facts found in the present case the decision to put the domestic services out to tender was, or involved, an act of discrimination on grounds of gender. That consideration will need to bear carefully in mind that, in view of the court's finding as to the decision in respect of the porters not being discriminatory, the decision as to the domestics must be assessed without any reference to the fact that the porters are men.

Failure to adjust the system

127. The Employment Tribunal went on to say, in its § 32, that even if the disparity had been justified when CCT was introduced, the continuation over time of a system whereby more men than women received bonuses, even though that system was in the process of being phased out, was indirectly discriminatory. This conclusion was not fully explained, and I do not think that it was justified. I do not see how it can be said that a failure to deprive the male comparators of part of their income was discriminatory, if the assumption is that their original receipt of that part of their income was not discriminatory. And the same is true of the decision to put the domestics out to tender. If that decision was discriminatory when made, it continued as such thereafter. If it was not discriminatory when made, it could not become such thereafter.

The genuine material factor defence

128. Having found that the difference between the men's contracts and the women's was discriminatory, the Employment Tribunal then needed to go on and consider the Trust's material factor defence, as summarised in §§ 11-13 of the Trust's closing submissions before it. I would say quite shortly that the Employment Tribunal entirely failed to address that issue. The EAT, in its §§ 35-36, accepted a submission by Miss Tether that the Employment Tribunal had at least had those matters in mind, as witness its recitation of some of the relevant facts in the statement of facts in its §8. I cannot agree that that was in any way adequate. The defence needed a careful assessment and judgement, which it did not receive.

Disposal

129. I would accede to the Trust's submission that the case should be remitted to the Employment Tribunal, but on the following terms:

i) No further evidence should be called, save that I agree with Arden LJ that it would be artificial for the Employment Tribunal not to be aware of the current incidence of bonus payments amongst the Trust's employees. I would therefore order that the Trust place before the Employment Tribunal an updated version of "analysis A", as referred to in paragraphs 12 and 86 of the judgment of Arden LJ.

ii) The Employment Tribunal should first consider whether the applicants have demonstrated that the bonus arrangements have a disparate adverse impact on female employees: see §108 above.

iii) If the answer to question (ii) is in the affirmative, the Employment Tribunal should then consider whether the decision to put the domestic services out to tender, or to discontinue the domestic services' bonus scheme, was an act of discrimination on grounds of gender: see §126 above.

iv) If the answer to question (iii) is affirmative, the Employment Tribunal should then consider whether the Trust has established its genuine material factor defence: see §128 above.

130. It will be for the Employment Tribunal to determine the procedure that it adopts to address those issues. It may well think, however, that a prudent first step would be for both parties to submit short skeletons indicating what they consider the answer to the issues should be in the light of the evidence and of the guidance that this court has endeavoured to give.