United Kingdom Employment Appeal Tribunal

APPEARANCES

 

For the Appellant	MR D MATOVU (of Counsel) Instructed by: Messrs ASB Law Solicitors Innovis House 108 High Street Crawley West Sussex RH10 1AS
For the Respondent	MRS J KERR (of Counsel) Instructed by: Messrs OH Parson & Partners Solicitors 3rd Floor Sovereign House 212-224 Shaftesbury Avenue London WC2H 8PR


 


SUMMARY

Sex Discrimination: Indirect

On the issue of disparate adverse impact, Tribunal correct in taking as a pool for comparison women with childcare responsibilities who suffered detriment by reason of fixed rotational shifts.

HIS HONOUR JUDGE ANSELL

1. This has been an appeal against part of a Reading Employment Tribunals reserved judgment, promulgated on 1 September 2006 (as confirmed by a further review decision of 1 December 2006), whereby the Tribunal upheld the claim of unlawful indirect sex discrimination against the Appellant employer. There is no appeal or cross-appeal in respect of other claims that were dismissed by the Tribunal, including claims relating to direct sex discrimination, flexible working and less favourable treatment of part-time workers.
2. This appeal was set down for a full hearing by order of 20 April 2007, following a preliminary hearing presided over by Mr Justice Underhill. By this appeal, the Appellant wishes to challenge the correctness of the Tribunal's decision on disparate adverse impact as well as on justification.

Background Facts

3. The Appellant provides ground handling services including check-in and loading for various airlines using Heathrow. The Respondent started work in 1992, employed by British Midland, who were taken over by the Appellant in March 2001. The Respondent worked as a supervisor. Her second child was born in March 2001, and she returned to work in January 2002. She informed the customer care manager that she had a problem working early shifts and could also not make childcare arrangements for Wednesdays and Fridays, and she was told that if she had particular problems she could swap days with a colleague or alternatively be re-rostered.
4. A general pattern of work was to work one day from 1300 to 2300 and the next day from 1300 to 2030 and then have four days off. Marimar Torres was another part-time supervisor, working from 1000 to 2000 and 0530 to 1300, also with four days off. On August 2003 the Respondent went on maternity leave and her third child was born in September.
5. In the summer of 2004, the Appellant carried out a restructuring exercise and created three full-time Duty Controller positions, whose main function was to oversee the smooth handling of the contracts handled in Terminals 1 and 2. There were also separate Duty Controllers for the contracts in Terminal 3. The work was for three days on and three days off, covering a total shift from 0930 to 2130, with suitable breaks. The Respondent on her return from maternity leave applied for the position and following an interview on 1 September 2004, she was informed by a letter of 13 September that she had been successful. However matters were put on hold because she had been offered a temporary position for six months as Duty Manager within the Human Resources team covering for a colleague on maternity leave.
6. Marimar Torres had also applied for a job, and though she and the Respondent only wanted to work part-time (although strictly not a job share), it was clear that their hours would require co-ordinating, and the notes of the Respondent's interview make it clear that various proposals were discussed. The Tribunal found that the reality of the Appellant's position as at 1  September 2004 was that if the Respondent and her colleague, Marimar Torres, were offered the position of Duty Controller then they would share the shifts between them. Indeed it appears that Marimar Torres's initial agreement with the Respondent was that the Respondent could work the first part of the shift and Marimar would cover the afternoons from 1530 to 2130.
7. However once Marimar had accepted the position, she then advised management that she also wanted to work morning shifts due to personal reasons. Whilst the Respondent was on secondment, Marimar did agree to work from 1430 to 2130 and worked these shifts for 6 months then changed to 3 days at 0930 starts and 3 days at 1430 starts until September 2005 which was when the Respondent was due to return to the position of Duty Controller, following her return from secondment.
8. The Appellant wrote to the Respondent on 13 September 2004 confirming that she had been successful in her application and that she would be required to work a shift of three days on and three days off. But

"as coverage is required from 930 to 2130, it has been decided that working three half days would be more appropriate to the operation."

9. The Respondent was due to return as Duty Controller on 19 September 2005 and prior to her return discussions took place as to the Respondent's hours in the light of the situation arising from Marimar's withdrawing her agreement to work evening shifts. On 12 July 2005, the Respondent met with Mike Cathrall together with his assistant to discuss the Respondent's return to work as Duty Controller. The Respondent explained that due to child care she could only cover between 0930 and 1530, and the Tribunal found that it was at this meeting that the Respondent was told that she would have to work rotational shifts. Various permutations were suggested by the employer including a suggestion of working two fixed days with one at the weekend.
10. At a further meeting on 4 August that suggestion appears to have been accepted by the Respondent. However Mr Cathrall appears to have said that that option was no longer available. Marimar had been willing to do rotational shifts which would mean working the morning shift from 0930 to 1530 on the first three days, then three days off, then an afternoon shift from 1530 to 2130, the suggestion clearly being that the Respondent should work a similar pattern. The Respondent was also offered other positions at that meeting which would involve a demotion.
11. On 8 August the Respondent made an application by letter for flexible working. Her current working pattern in the Duty Management secondment position allowed her to work fixed days Tuesdays and Thursdays from 0900 to 1700, and a Saturday or Sunday as required. In her application the Respondent set out in detail her four options for flexible working, which generally involved morning shifts on fixed dates during the week, with a longer shift on a weekend day.
12. On 18 August, Mike Cathrall indicated to the Respondent that the proposals were rejected because it would not fit in with the preferences of the other staff and it would also mean that the Duty Controller position in the department would not be filled for certain operational periods. A formal rejection was sent in a letter of 22 August. The Respondent appealed that decision and in the meantime further discussions took place but they were unsuccessful. The Tribunal, recorded at paragraph 3.17, that it was the Appellant's case that it was implicit from the time that the offer of the new job was made to the Respondent that she would be required to work on a rotational roster basis. The Tribunal however was satisfied from the evidence that there was no condition that she would have to work on that basis at the time the job offer was made to her, and that the condition was sought subsequently by the Appellant as a result of Marimar not being prepared to continuously cover the afternoon/evening shift.

Indirect Discrimination – Sex Discrimination Act 1975, Section 1(2)(b)

13. As indicated above the Respondent was successful in her claim under this provision. The first issue raised in this appeal was to clarify which version of this provision applied to this case. Both parties agree that the correct form of section 1(2)(b) that applied to this case was the version substituted by the Sex Discrimination (Indirect Discrimination and Burden of Proof) Regulations 2001 as from 12 October 2001, namely:

"A person discriminated against a woman if— […]

(b) he applies to her a provision, criterion or practice which he applies or would apply equally to a man, but—

(i) which is such that it would be to the detriment of a considerably larger proportion of women than of men, and

(ii) which he cannot show to be justifiable irrespective of the sex of the person to whom it is applied, and

(iii) which is to her detriment."

14. The above version is to distinguish it from the later version of the Employment Equality (Sex Discrimination) Regulations 2005 as from 1 October 2005, which provides as follows:

"(2) In any circumstances relevant for the purposes of a provision to which this subsection applies, a person discriminates against a woman if –

(a) on the ground of her sex, he treats her less favourably than he treats or would treat a man, or

b) he applies to her a provision, criterion or practice which he applies or would apply equally to a man, but –

(i) which puts or would put women at a particular disadvantage when compared with men,

(ii) which puts her at that disadvantage, and

(iii) which he cannot show to be a proportionate means of achieving a legitimate aim."

15. On behalf of the Appellant, Mr Matovu's first complaint was that both in the original decision and in the review decision, the Tribunal appeared to have confused the two different versions. In the original decision, having set out the correct statutory provision in paragraph 4.5, the Tribunal set out its conclusion in paragraph 4.7 as follows:-

"Does that provision, criterion or practice put women at a particular disadvantage? The answer is inevitably yes. The Tribunal is entitled to take into account that women are more likely to be primary carers of children and as such less likely to be able to meet the requirement to work different hours on different days of the week. The Respondents applied the PCP to those working in Customer Services and Ramp Services. The Claimant worked in Customer Services (256 staff). The Respondent argues that because in Customer Services out of 80 part-time workers, 6 out of 18 (33%) were accommodated on flexible hours and 15 out of 62 women (24%) were accommodated on flexible hours that indicated a higher proportion of part-time men would appear to have been disadvantaged. In considering Customer Services, it is relevant to consider the number of women who requested variations to fixed rotational shifts in order to deal with child care responsibilities as opposed to men requesting variations for that reason. There was no evidence of men in Customer Services having made application for variation of fixed rotating rosters to accommodate child care responsibilities whereas there was evidence that, in addition to the Claimant, Nabila Awan had her application for variation rejected in like manner to that of the Claimant's. The flexibility enjoyed by Marimar del Rio was withdrawn in April 2006 and that Iwona Wylewinska had her request for flexible working turned down. That was direct evidence which was not challenged, that the PCP put women at a particular disadvantage. Did it put the Claimant at a disadvantage? On the evidence heard the Tribunal are entirely satisfied that it did."

Mr Matovu highlighted the Tribunal's use of the word "disadvantage" at both the beginning and the end of that paragraph. He also pointed to a similar error in paragraph 6 of the Review decision, where the Tribunal dealing with the issue said this:-

"The direct evidence before the Tribunal, referred to in the Judgment (4.7) was of women who requested variations to fix rotational shifts in order to deal with childcare responsibilities were put at a particular disadvantage because those applications were refused. This is clear evidence of disparate impact."

Miss Kerr responded that the Appellant's complaint that the Tribunal may have become confused in relation to the statutory provisions was not raised in the application for review, nor did it appear in the original Notice of Appeal. At the preliminary hearing the complaint was raised that the Tribunal had misdirected itself in applying the wrong statutory wording.

16. Mrs Kerr argued that notwithstanding the Tribunal's use of the word "disadvantage", they clearly had the correct approach in mind, having set out the correct statutory provision at paragraph 4.5. Further she submitted that the Tribunal's analysis of the facts in paragraph 4.7 dealt with the issue of whether the Respondent had established disproportionate or disparate adverse impact even though they had unfortunately interchanged the words "disadvantage" and "detriment." We agree with her submissions on this point, and are satisfied that the Tribunal certainly approached their task on the correct basis.

Disparate Adverse Impact – The Pool

17. Both parties agree that the combined effect of Section 1(2)(b) of the Sexual Discrimination Act 1975, together with burden of proof provisions in Section 63(A) require a complainant to establish disproportionate adverse impact on the balance of probabilities and thereby establish a prima facie case that she had suffered indirect discrimination. Further, merely to raise a credible suggestion that were the relevant, valid and significant statistics provided they might establish disproportionate impact, was not sufficient for the complainant's purposes and imposed no further burden of explanation on the employer; accordingly, it was for the applicant to provide necessary statistics to show on the balance of probabilities a disproportionate impact, and thereby establish a prima facie case that she had suffered indirect discrimination See Nelson v Carillion Services Ltd [2003] ICR 1256.
18. It was further agreed between the parties that the relevant provision, criterion or practice (PCP) for the purposes of the SDA 1975, was the imposition of a fixed rotational shift. There was a query raised within the review application as to whether the Tribunal had understood the full meaning of this phrase. However the Tribunal in paragraph 5 of the review decision indicated that they did understand what was meant by the phrase, namely a shift pattern "where employees work with a fixed number of days on and a fixed number of days off. In particular it sets out the specific hours on those days on which the employee is required to work."
19. Mr Matovu complained that the Tribunal's decision process in this aspect of the case was flawed. Rather than identifying a core of workers, to which the relevant comparison could be made between men and women's disparate treatment, the Tribunal chose to decide this aspect of the case by identifying in addition to the Claimant four women who had allegedly been put at a disadvantage by the imposition of the PCP, to which they added intrinsic evidence that women were more likely to be primary carers of children, and as such less likely to be able to meet the requirement to work different hours on different days of the week.
20. Mr Matovu submitted that the correct pool for comparison in the case should have comprised all those to whom the relevant PCP was applied, i.e. all those to whom the imposition of fixed rotational shifts applied, namely those in the Customer Services division to which the Respondent was attached, and also those in the Ramp Services division.
21. The Tribunal in their review admit that they left out of consideration all those in the Ramp Services division, on the ground that the were not undertaking the same work as those in the Customer Services division and their roles were different. They claimed in the review decision that they had looked at the totality of the Customer Services staff, both full and part-time, but again only highlighted the number of women who had requested variations of fixed rotational shifts in order to deal with childcare responsibilities, as opposed to men requesting variations for that purpose. They rejected the evidence presented by the Appellant in relation to part-time workers, that a higher proportion of men were accommodated on flexible hours, indicating therefore that a higher proportion of part time men would appear to have been disadvantaged.
22. The Respondent's statement before the Tribunal also contained some evidence of statistics as to the number of female and part-time workers employed by the Appellant, and there was also a schedule before the Tribunal, again not referred to in their decision, showing a list of 25 women who were mothers with children under the age of six, of whom five worked set days and hours and a duty manager, Helen Lilley, who was stated to work not on a rotational basis.
23. Mr Matovu submitted that the statistical evidence relied on by the Tribunal was far too limited, concentrating exclusively on the disadvantaged rather than the advantaged and backed up with the impermissible "intrinsic" evidence, from the Tribunal's own experience. He took us firstly to London Underground Ltd v Edwards (No. 1) [1995] IRLR 355. In that case Mrs Edwards, a single parent with a young child, was employed as a train operator – the rostering arrangements allowed her to be home in the morning and the evening to attend to her son. Her employer sought to introduce a new flexible shift system under which duties were to begin very early in the morning and were to include Sundays, although it was possible for employees to change shifts so as to avoid early and late work. The trade off was a longer shift for the same money and Mrs Edwards was not prepared to work the new system. An Industrial Tribunal upheld her complaint of unlawful sex discrimination because a considerably smaller portion of female single parents than male single parents could comply with the rostering conditions or requirements. The Employment Appeals Tribunal found that the Industrial Tribunal had erred in law in finding that the employer's new rostering arrangements had an indirect discriminatory effect, because a considerably smaller portion of female single parents than male single parents could comply with their requirements to work non-social or extended hours. A pool which consisted of only those train operators who were single parents was the wrong pool for comparison. The correct pool for comparison was that of all train operators to whom the new rostering arrangements applied and the proper question was whether the requirement relating to availability was such that a considerably smaller portion of women qualified to be train operators than of men so qualified could comply with it.
24. At paragraph 14 of the decision Mr Justice Mummery presiding said this:

"We agree with the submission made by Mr Bean on behalf of London Underground that the correct question under s.1(1)(b)(i) and s.5(3) is whether the condition or requirement of availability for rostering was such that a considerably smaller proportion of women qualified to be train operators than of men so qualified could comply with it. Instead, the tribunal applied the following test in paragraph 5 of the decision when they stated:

'…In our view the condition or requirement in question was such that a considerably smaller proportion of female single parents than male single parents could comply with it…'

We agree with Mr Bean that it was wrong in law to select 'single parent train operators' as the pool for comparison in relation to the question of indirect discrimination. In Jones v University of Manchester [1993] IRLR 218 at paragraph 46, Ralph Gibson LJ commented on the proper application of s.1(1) as follows:

'In order to compare the proportion of women who can comply with the requirement with the proportion of men who can comply with it, it is necessary to determine the relevant total. In my judgment, the relevant total is the number of men and women referred to in the subsection, i.e. those men and women to whom the person - in this case the employer - applies or would apply to the requirement. In this case, that means all men and women graduates with the relevant experience. I do not accept that the relevant total is all men and women: the employer would have no occasion to apply the requirement to any men or women other than those who are able to comply with the requirements of the advertisement other than the requirement in question.'

The advertisement which had been answered by the complainant was for the post of a careers adviser at Manchester University and stated that the person to be appointed would be 'a graduate, preferably age 27-35 years.' The complainant was 46, outside the preferred age range. She was not selected for interview."

And later at paragraph 17:

"On the same reasoning, the industrial tribunal erred in law in having regard to a 'pool' which consisted of only those train operators who were single parents, a subdivision not warranted by the statutory provisions. The pool consisted of train operators, male and female, to whom the new rostering arrangements were applied."

25. Mr Matovu then took us to the case of Cheshire and Wirral NHS Trust v Abbott [2006] ICR 1267. The claimants in an equal pay case were members of domestic staff employed at two hospitals, operated by the employer. The claimants brought proceedings in relation to the payment of a bonus and compared themselves only with the porters, who received a bonus, but the employer maintained at the Employment Tribunal that the appropriate comparison was with the porters and also caterers, some of whom were also paid a bonus. The Tribunal held that the claimants had the right to choose their comparator group and rejected the employer's further argument that the porters were too small a group to be a statistically significant comparison. The Court of Appeal held that whilst it was for the claimants to identify their comparator group, the Tribunal had to ensure that the chosen group was not an artificial or arbitrary one, and that as a matter of logic it was not appropriate to confine a comparator pool to one solely consisting of porters rather than the porters and caterers who also received a bonus, since in principle the comparison should be made between the disadvantaged group and the advantaged group. Further, as this is a matter of statistics, comparison with a larger group is likely to produce a more reliable result, so long as the group shared the relevant characteristics and could be seen to be doing work of equal value. The large numbers were less important where there was an identified difference in pay or benefits.
26. Finally, on this aspect of the case Mr Matovu took us to a recent decision of the Employment Appeals Tribunal, Faulkner v Chief Constable of Hampshire Constabulary (UKEAT/0505/05), with His Honour Judge McMullen QC presiding. That case concerned a policy preventing police officers in a partnership working together in a supervisor/subordinate role. The Employment Tribunal had found the pool for comparison was the group of police officers who had an existing partnership since this focused entirely on the group disadvantaged by the policy. (A group on the evidence made up of six of the 3,802 police officers. This pool was not contended by either party.) The Employment Tribunal ruled that the pool should have been the whole of the Respondents' work force - police officers and support staff alike – since the policy applied across the board.
27. Mr Matovu contended that the Respondent had failed to produce any adequate statistical information for the Tribunal and indeed had focussed entirely on the group they argued was at a detriment. In the Respondent's ET1, the claim for indirect discrimination had been pleaded in a very general way:-

"…the Claimant also believes that the requirement for fixed rotating shifts is indirectly discriminatory, as it detrimentally affects part-time workers, who are predominantly female, often with childcare commitments, compared to the full-time male employees without justification."

28. In her closing submission to the Tribunal, Miss Kerr dealing with indirect sex discrimination, had said this:-

"Women are more likely to be primary carers of children and as such less likely to be able to meet the requirement to work different hours on different days of the week. ...London Underground Ltd v Edwards [1998] IRLR 364). It is not disputed that in addition to the Claimant, Maria Dolores Corral was extremely stressed about the effect of post-restructuring rosters; that Nabila Awan had her application for variation rejected in like manner to the Claimant's; that the flexibility enjoyed by Marimar del Rio was withdrawn in April 2006, and that Iwona Wylewinska had her request for flexible working turned down."

29. Further Mr Matovu submitted that the correct pool for comparison in the case comprised all members of staff to whom the relevant PCP was applied, i.e. all those to whom the imposition of a fixed rotational system applied. Not only those working in the Customer Services division but also those in the Ramp Services division.
30. The Tribunal in their review decision admitted that they had left out of consideration all those in the Ramp Services division on the grounds that they were not undertaking the same work as those in the Customer Services division and their roles were different. Mr Matovu then reminded us of the provisions of Section 5(3) of the SDA which provided so far as is relevant

"a comparison of the cases of persons of different sex under Section 1(1) or 2(…) must be such that the relevant circumstances in the one case are the same not materially different in the other."

He contended that the different working roles were not a material difference.

31. In Shamoon v Chief Constable of the RUC [2003] ICR 337, the House of Lords had to consider the operation of Section 5(3), albeit in relation to a direct sex discrimination claim. The Court was considering the relevant provisions in the Sex Discrimination (Northern Ireland) Order 1976 which were in all material the same respects as in the 1975 Act. At paragraph 9, Lord Nicholls said this:

"The present case is a good example. The relevant provisions in the Sex Discrimination (Northern Ireland) Order 1976 are in all material respects the same as those in the 1975 Act which, for ease of discussion, I have so far referred to. Chief Inspector Shamoon claimed she was treated less favourably than two male chief inspectors. Unlike her they retained their counselling responsibilities. Is this comparing like with like? Prima facie it is not. She had been the subject of complaints and of representations by Police Federation representatives, the male chief inspectors had not. This might be the reason why she was treated as she was. That might explain why she was relieved of her responsibilities and they were not. But whether this factual difference between their positions was in truth a material difference is an issue which cannot be resolved without determining why she was treated as she was. It might be that the reason she was relieved of her counselling responsibilities had nothing do with the complaints and representations. If that were so, then a comparison between her and the two male chief inspectors may well be comparing like with like, because in that even the difference between her and her two male colleagues would be an immaterial difference."

32. Later at paragraphs 48 and 49, Lord Hope dealt with the issue in this way:-

"The structure of article 3(1)(a), on the other hand is quite different. What it requires is proof that the alleged discriminator treated the woman less favourably in the ground of her sex. The words "than in those circumstances" which define the scope of the comparison required in article 6 are absent. Instead the direction which is set out in article 7 must be applied. It provides that the comparison must be such that "the relevant circumstances" in the one case are the same, or not materially different, in the other. The reason why article 7 does not say that the basis of comparison which it lays down must be applied to a comparison under article 6 is because article 6 itself sets out the basis for the comparison which it requires. It does so by directing attention to the circumstances mentioned in the opening words of the article. This is a narrower basis than that required of article 3 by the phrase "the relevant circumstances" in article 7. … The phrase directs attention to all the circumstances which are relevant to the way in which the woman has been treated. Moreover, there is no need to break this test down into two parts. It is, as Lord Nicholls has said, in essence a single question. The comparison which is invited by the words "treats her less favourably" lies at the heart of the entire sub-paragraph.

…Whichever approach is adopted, one must not lose sight of the fact that article 3(1) must be read as a whole and that it must be read together with article 7. A comparison of the cases of persons of a different sex under article 3(1)(a) must therefore be such that all the circumstances which are relevant to the way they were treated in the one case are the same, or not materially different, in the other."

33. And finally, Lord Rodger, at paragraph 134 and again at 136:-

"134 Parliament has not spelled out what constitute 'the relevant circumstances' in article 7. The meaning of the words must therefore be deduced from the context. They occur in a provision prescribing how anyone administering the Order is to determine whether an alleged discriminator treated a woman less favourably than a man on the ground of her sex under article 3(1). So 'the relevant circumstances' must be those circumstances that are relevant for that purpose. In Nagarajan v London Regional Transport [1999] ICR 877, 884E-F, Lord Nicholls pointed out that in every case of alleged discrimination it is necessary to inquire why the complainant receiver less favourable treatment. As he went on to say, save in obvious cases, this will call for some consideration of the mental processed of the alleged discriminator in order to identify the grounds of his decision. I accordingly infer that "the relevant circumstances" in article 7 are those which the alleged discriminator takes into account when deciding to treat the woman as he does or when deciding to treat the man as he treats, or would treat him. …

136 The relevant circumstances for the proposed of article 7 cannot be confined, however, to those which the alleged discriminator takes into account when deciding to the treat the woman as he does. … If, faced with a man in the same circumstances, the employer took into account his problems in looking after his mother and therefore decided to reprimand rather than dismiss him, then those problems would be a relevant circumstance in his case. Therefore, for the purpose of the comparison required by article 7, those problems would be among the relevant circumstances, even though the employer had not taken the same problems into account in deciding how to treat the woman. Indeed it is precisely because the employer took account of those problems in the case of the man but ignored them in the case of the woman that he treated her less favourably by dismissing her. If he did this on the ground of her sex, then she will have a valid claim under article 8(2)(b). Article 7 allows for this when it says that the relevant circumstances "in the one case" are to be the same, or not materially different, "in the other". The comparison runs, and is intended to run, in both directions. So circumstances which the alleged discriminator takes into account or would take into account in the case of the male comparator are relevant if they were also present, though not taken into account, in the case of the woman."

34. Mr Matovu submitted that the difference in work or roles carried out by those in the two departments was an immaterial difference for comparison purposes, with regard to the imposition of fixed rotational shifts, which applied equally to all those employed in both departments. In deciding to apply the fixed rotational shifts to the Respondent, and when deciding to reject her application for flexible working, nothing really turned directly on the specific nature of her work or her role as such, rather it had more to do with the size of the team she was employed in and the fact that coverage was required operationally from 0930 to 2130.
35. Further, even within the Customer Service team there were differing roles, yet the Tribunal considered the whole team could be considered for the purposes of the comparative pool. He also relied on the Cheshire decision as set out above, where it was held out as not appropriate to simply confine the pool to one consisting solely of porters, rather than porters and caterers, since they both received the bonus. He argued that there was no logical reason for excluding those from the Ramp Services division from the comparator group, otherwise the Tribunal's only course should have been to confine the group to those only undertaking the same work in the role of the Respondent, namely Duty Controllers employed at Heathrow Airport, all of whom were female. Further, he argued insofar as there had been evidence introduced into the case, to suggest that there was a difference between the two divisions, this was only because of the relevance in relation to the part-time worker's claim, where the regulations required that the Respondent had to identify a comparable full-time worker, who was "engaged in the same or broadly similar work." (See Regulation 2(4)(a)(2) of the Part-Time Worker Regulations). He argued that the Tribunal further compounded the error by going on to consider an even narrower poll, a subdivision of the Customer Services division, and only those within that department who had requested variations to fix rotational shifts in order to deal with childcare responsibilities, and in making that comparison appears to have totally disregarded the unchallenged evidence as to the eighty part-time workers within the Customer Services division, produced by the Appellant, indicating that more part-time men than women needed to avoid the imposition of fixed rotational shifts.
36. Finally, on this aspect he argued that the Tribunal could not legitimately then bridge the gap by seeking to rely on their own general knowledge and expertise, to look outside the pool for comparison, and take into account the fact that nationally more working women than working men have direct childcare responsibilities. In London Underground Ltd v Edwards (no. 2) [1998] ILR 364, the second Industrial Tribunal found that 100 per cent of the 2,023 male train operators could comply with the new rostering conditions, but Mrs Edwards was the only one of the 21 females train operators who complained that she could not comply with the new arrangements, and the proportion of women who could comply was 95.2 per cent (20 out of 21). However the Tribunal concluded that taking into account the number of male train operators as compared to very few female train operators, and also added to the fact that it was most people's common knowledge that females were more likely to be single parents or caring for a child than males, therefore it was clear that this was a condition or requirement that a considerably small number of females could comply with. The Court of Appeal approved the Industrial Tribunal's approach, and held that the Tribunal did not err in finding that the requirement bore disproportionately between men and women even though only one woman was affected by it. They also held the Tribunal was entitled to use its general knowledge and expertise to look outside the pool for comparison and take into account the national figure that ten times as many women as men were single parents that take care of children. Mr Matovu referred us in particular to a passage in the judgment of Brown LJ, at paragraph 42:-

"There is no evidence before us as to what proportion of the working population as a whole are single parents who would be unable to comply with the appellants' roster system and, even if there were, that is not the group which it is agreed must be used here to make necessary comparisons. […] To my mind the sole relevance of the 10:1 statistic here is to preclude the appellants from contending that is fortuitous to find the women in the group to any degree disadvantaged. It would, for example, have been highly relevant had by chance a man in the group been affected, but no woman, and the man then claimed. As it is, and given the appellants do not contend that there is anything fortuitous about the figures here, the 10:1 statistic seems to me to be of essentially neutral effect."

37. The Court of Appeal did not appear to have been referred to an earlier European case of John O'Flynn v Adjudication Officer, Case C-237/94. The issue concerned whether provisions regarding freedom of movement for workers within the European Community precluded a rule of the member state (UK) which made grant of a payment to cover funeral expenses incurred by a migrant worker, subject to the condition that burial or cremation took place within the territory of that member state.
38. At paragraph 20 of the Court's decision it stated as follows:-

"It follows more before going case law, that, unless objects are justified and proportionate to its aim, the provision of national law must be regarded as indirectly discriminatory if it is intrinsically liable to affect migrant workers more than national workers and if there is a consequent risk that it will place the former at a particular disadvantage"

39. The O'Flynn case was referred to in a more recent Employment Tribunal case in Scotland, Ministry of Defence (Royal Navy) v MacMillan EATS/0003/04, with the Honourable Lord Johnston presiding. The cause of the dispute was a rejection by the Navy of an application by an employee that she should be allowed to go on working part time at the base, which would be contrary to a policy as regards Royal Navy personnel on the full strength list that they should not be allowed to work part-time but must work full-time. Mr Matovu took us to various passages in the decision as follows:-

"It was the fundamental submission of the Advocate General that there had to be a pool, properly defined by the evidence, in order to determine the issue of less favourable treatment, and, indeed, detriment with regard to the issue as between proportions of men and women. […]

We recognise the importance of Mr Napier's submission that a pool is not necessary in essential terms to establish disparate treatment, although, in many cases, it may be the only way of so doing. We say not necessary because the case of O'Flynn, to our mind, is a clear example of where, to use the word "intrinsic", the very circumstances surrounding the case support the relevant conclusion. The determination by the British Government only to pay the relevant allowance to workers being buried within the United Kingdom, is bound to have a disparate effect on migrant workers by reason of the simple proposition, that, in most cases, there would be an understandable wish on behalf of the family of the deceased, for the person to be buried in his or her country of origin. There does not need to be evidence to establish that fact and O'Flynn is, therefore, totally understandable. In most of the other cases that we have considered, however, there has to be identified a core of workers, whether called pool or not, to which the relevant comparison can be made as between men and women's disparate treatment. This is clear from Barry and both Edwards cases. The case of Chew seems to us to be somewhat unusual but can again support that position.

The general conclusion we therefore draw, is that for disparate treatment to be established in the form of indirect discrimination as between men and women in the context of the relevant provision, the evidence must support it and lead directly to the necessary inference.

In this respect, we do not find the approach by the Tribunal to the issue of a pool to be coherent, or even consistent. At one point reference was made to the UK workforce as a whole and at another to RN personnel. […]

However, that does not mean that evidence is not required, and, we therefore accept the submission of the Advocate General, that the failure on the part of the respondent as applicant, to lead any evidence of any nature which would suggest other women in full-time employment in the Navy were suffering the same or equivalent detriments that she claims, leaves her in a category of one as a matter of evidence, and, not, therefore, creating a situation which enables even an inference of discrimination to be drawn such as could be done in O'Flynn. We can only speculate as to how the applicant might have been able to surmount this hurdle, but, at the very least, evidence of women working within the Navy, claiming or said to be suffering disparate treatment by reference to detriment, would have been necessary. Even then, the question of proportionality in the sense of whether it was a greater proportion of men than women would require to be investigated. We do not think it is appropriate, as the Tribunal on one view appears to have done, simply to assume that full-time working mothers suffered disparate treatment from men, per se."

40. Mr Matovu submitted that the intrinsic evidence that the Tribunal sought to rely on, namely that nationally more working women than working men have direct child care responsibilities, could not in the absence of statistical information be sufficient to establish disproportionate adverse impact, contrary to the position in O'Flynn where the very circumstances of the case supported the relevant conclusion. He submitted that, at very best, intrinsic evidence could be used to support a prima facie case of disproportionate adverse impact, made out on the statistics, in the sense that it would preclude the employers from contending that it was fortuitous to find the women in the group to any degree disadvantaged. He also argued that the Tribunal in this case could not even have regard to a large discrepancy in numbers between male and female operators making up the pool, as in Edwards (no. 2), where the discrepancy was there taken as indicating without the need for specific evidence that it was either difficult or unattractive for women to work as train operators in any event. In our case, women outnumbered in the Customer Services division and the vast majority of part-time workers, 62 out of 80, were also female.
41. In her response Mrs Kerr argued that it was open to the Tribunal to consider simply the 256 employees in the Customer Service and not those employees in the Ramp Services department. In cross-examination the Respondent was invited to accept that the Ramp Service operators "do a completely different job from her." She reminded us of paragraph 4.2 of the Tribunal's decision: the Tribunal found that there was a more relaxed approach adopted in the Ramp Services section, and in paragraph 4.3 the Tribunal accepted the Claimant's evidence that Ramp Services agents and Ramp Services Supervisors were not comparable to her. Mrs Kerr submitted that on the evidence the Tribunal had found a material difference in circumstances, and accordingly the Tribunal were entitled to exclude Ramp Services from the pool. We agree with her submissions on this aspect. The Claimant was not seeking to be artificial or arbitrary in seeking to restrict the pool to customer services and indeed we have seen a note of the Respondent's cross-examination in the course of which it was suggested to her that Ramp Services agents do a completely different job. It seems to us that there was material available to the Tribunal that entitled them to come to the decision they did regarding Ramp Services.
42. On the more general issue of whether there was a disparate adverse impact, Mrs Kerr invited us to consider carefully the effect of Edwards (no. 2), which in particular emphasised the need for a Tribunal to be flexible when considering whether or not there had been indirect discrimination in a manner consistent with the statutory purpose underlying the provisions of the SDA as well as the ability of the Tribunal members to use their own intrinsic knowledge. In the Edwards case Potter LJ dealt with these matter at paragraphs 23 and 25 of the judgment, as follows:-

"In my view, there is a dual statutory purpose underlying the provisions of s.1(1)(b) and in particular the necessity under subparagraph (i) to show that the proportion of women who can comply with a given requirement or condition is 'considerably smaller' than the proportion of men who can comply with it. The first is to prescribe as the threshold for intervention a situation in which there exists a substantial and not merely marginal discriminatory effect (disparate impact) as between men and women, so that it can be clearly demonstrated that a prima facie case of (indirect) discrimination exists, sufficient to require the employer to justify the application of the condition or requirement in question: see subparagraph (ii). The second is to ensure that a tribunal charged with deciding whether or not the requirement is discriminatory may be confident that its disparate impact is inherent in the application of the requirement or condition and is not simply the product of unreliable statistics or fortuitous circumstance. Since the disparate impact question will require to be resolved in an infinite number of different employment situations, well but by no means comprehensively exemplified in the arguments of Mr Allen, an area of flexibility (or margin of appreciation), is necessarily applicable to the question of whether a particular percentage is to be regarded as 'substantially smaller' in any given case.

In my view, Mr Bean was incorrect in that last respect. An industrial tribunal does not sit in blinkers. Its members are selected in order to have a degree of knowledge and expertise in the industrial field generally. The high preponderance of single mothers having care of a child is a matter of common knowledge. Even if the 'statistic', i.e. the precise ratio referred to is less well known, it was in any event apparently discussed at the hearing before the industrial tribunal without doubt or reservation on either side. It thus seems clear to me that, when considering as a basis for their decision the reliability of the figures with which they were presented, the industrial tribunal were entitled to take the view that the percentage difference represented a minimum rather than a maximum so far as discriminatory effect was concerned."

43. Further, Lord Justice Brown dealt with the issues from paragraphs 45 onwards:-

"None of the authorities cited to us seemed to me to bear directly on this point. Certainly none of them appears to decide clearly that, unless a considerable proportion of the women in the group are unable to comply with the requirement in question, it matters not how that requirement affects the men in the group. Sometimes in the judgments phrases are used which might be thought to support one approach rather than the other. For example, in Jones v University of Manchester [1993] IRLR 218 at 226, 54 Ralph Gibson LJ stated:

'Parliament enacted that indirect discrimination against women could be found to exist if the proportion of women who could comply with the requirement is considerably smaller than the proportion of men who could comply: it did not put the ground of proof on the proportions within a particular group of people who could not comply with the requirement.'

Conversely, in R v Secretary of State for Employment ex parte Seymour-Smith [1995] IRLR 464 at 475, 90, Neill LJ giving the giving the judgment of the court said:

'In these circumstances, we would accept the submission made on behalf of the Secretary of State that before a presumption of indirect discrimination on the ground of sex arises there must be a considerable difference in the number or percentage of one sex in the advantaged or disadvantaged group as against the other sex and not simply a difference which is more than de minimis' (emphasis added).

I repeat, none of the cases appears to decide the present point.

As to how percentages of this sort have been regarded in the past, one finds only limited assistance, and indeed some inconsistency, in the cases. Whereas, for example, in Staffordshire County Council v Black [1995] IRLR 234, the EAT (presided over, be it noted, by Morison J) upheld the industrial tribunal's finding that 89.5% was not 'considerably smaller' than 97% — stating that 'the figures speak for themselves' in Greater Manchester Police Authority v Lea [1990] IRLR 375, the EAT (presided over by Knox J) upheld the industrial tribunal's finding that 95.3% was 'considerably smaller' than 99.4%.

I can state my conclusions really quite shortly. Given that this legislation is concerned essentially to contrast the impact of a given requirement or condition as between men and women rather than as between the women in the group, it would seem to me wrong to ignore entirely the striking fact here that not a single man was disadvantaged by this requirement despite the vast preponderance of men within the group. Looked at in the round, this requirement clearly bore disproportionately as between men and women, even though only one woman was affected by it. Had there been an equal number of women drivers to male drivers and the same 5% proportion of them been affected, i.e. 100, Mr Bean's argument would remain the same, namely that too large a proportion of women were able to comply with the requirement to leave room for a finding that such proportion was 'considerably smaller' than the proportion of men who could comply. It is not an argument I am ultimately prepared to accept. The approach to s.1(1)(b)(i) must, I conclude, be more flexible than this argument allows. Parliament has not, be it noted, chosen to stipulate, as it could, just what difference in proportions would be sufficient. Once, then, one departs from the purely mechanistic approach contended for by the appellants, and has regard to other facts besides merely a comparison between 95% and 100%, the respondent's argument becomes compelling: no other fact could be more relevant than that, whereas 5% of the women were disadvantaged, not one of 2,023 men was. That further consideration in my judgment supports the industrial tribunal's findings here."

44. Mrs Kerr submitted that the intrinsic evidence in this case was powerful - the Tribunal were entitled to take into account the fact that women were more likely to be primary carers of children and as such, less likely to be able to meet the requirement to work different hours on different days of the week each week, and thus the imposition of fixed rotational shifts would be intrinsically liable to effect more women than men. She submitted that this evidence taken by itself satisfied the requirements set out in the O'Flynn case. In addition, the evidence from the three other women referred to in paragraph 4.7 of the decision was powerful and persuasive. Two of the three were witnesses before the Tribunal on behalf of the Respondent, and their evidence was substantially accepted by the Appellant. A third woman had her circumstances described by the Respondent in the course of her evidence - that evidence was also unchallenged. In contrast the Appellant's case was not supported by any employee witnesses, nor did the Appellant's witnesses dispute that it would inevitably be more difficult for those who were primary carers of children to comply with the requirement.
45. We agree with Mrs Kerr: a flexible approach is required, with the ability to use both statistical information and at times intrinsic knowledge, comparing at times the advantaged or the disadvantaged - it is about the language of comparison. The Tribunal had the direct evidence before them, which they described as clear evidence of disparate impact, together with the fact that not a single man appeared to be disadvantaged by the requirement. To this they coupled their general knowledge and expertise, and in our view rightly ended up with a view that there was strong prima facie evidence. They rejected the Appellant's contention that it created a subdivision of the core – namely part-time workers. In any event, Mrs Kerr argues that this evidence was flawed, since the fact that someone had not been accommodated on flexible working did not mean that they could comply with the new shift system - the Respondent being a case in point. In any event, she argued that the PCP was not applied to those who were accommodated on flexible working, since if they were permitted to work flexibly, they were not having imposed upon them the new fixed rotational shift system. Again we are satisfied that the Appellant's statistical evidence did not help the Tribunal in their analysis of whether women with childcare responsibilities were being disadvantaged by the fixed rotational shifts.

Justification

46. The application for review did not seek to criticise the Tribunal's overall approach to justification, but highlighted the fact that, in cross-examination, Mr Cathrall was specifically asked whether he considered the Duty Controllers in Terminal 3. His response was,

"There was no scope for them to move from Air India… I considered it at the time I was considering the [Claimant's flexible working] application, but they (i.e. the Duty Controllers in Terminal 3) are fully occupied."

47. The Appellant contended in the review that they had in fact considered the Duty Controllers in Terminal 3 and did all they could to explore the possibility of accommodating the Respondent on the hours she wanted, i.e. mornings only.
48. In their written submissions on the review, the Appellant did contend that the Tribunal had failed to address the critical point about the operational needs of the business, namely the need to fulfil its contractual obligations to the airlines in Terminals 1 and 2 such as Iberia. This involved a need to cover the peak period between 0930 and 2130, providing full coverage for a 12-hour shift, and in particular a need to cover the afternoons, especially as two-thirds of the flights were during that period. The Appellant also referred to the Claimant's witness statement which acknowledged that both Customer Service and Ramp Services departments worked rotational rosters, and needed to cover operations from 0500 to 2300, and that even if the part-time Duty Controller position was not offered as a job share, there was a need to co-ordinate the hours to cover the twelve-hour shifts. These issues were further considered by the Tribunal in the review decision in paragraphs 7 and 8.
49. Mr Matovu firstly reminded us of the guidance in dealing with this issue set out in Hardys & Hansons plc v Lax [2005] IRLR 726, in particular paragraphs 32-34 where Pill LJ, said this:-

"Section 1(2)(b)(ii) requires the employer to show that the proposal is justifiable irrespective of the sex of the person to whom it is applied. It must be objectively justifiable (Barry) and I accept that the word 'necessary' used in Bilka is to be qualified by the word 'reasonably'. That qualification does not, however, permit the margin of discretion or range of reasonable responses for which the appellants contend. The presence of the word 'reasonably' reflects the presence and applicability of the principle of proportionality. The employer does not have to demonstrate that no other proposal is possible. The employer has to show that the proposal, in this case for a full-time appointment, is justified objectively notwithstanding its discriminatory effect. The principle of proportionality requires the tribunal to take into account the reasonable needs of the business. But it has to make its own judgment, upon a fair and detailed analysis of the working practices and business considerations involved, as to whether the proposal is reasonably necessary. I reject the appellants' submission (apparently accepted by the EAT) that, when reaching its conclusion, the employment tribunal needs to consider only whether or not it is satisfied that the employer's views are within the range of views reasonable in the particular circumstances.

The statute requires the employment tribunal to make judgments upon systems of work, their feasibility or otherwise, the practical problems which may or may not arise from job sharing in a particular business, and the economic impact, in a competitive world, which the restrictions impose upon the employer's freedom of action. The effect of the judgment of the employment tribunal may be profound both for the business and for the employees involved. This is an appraisal requiring considerable skill and insight. As this court has recognised in Allonby and in Cadman, a critical evaluation is required and is required to be demonstrated in the reasoning of the tribunal. In considering whether the employment tribunal has adequately performed its duty, appellate courts must keep in mind, as did this court in Allonby and in Cadman, the respect due to the conclusions of the fact finding tribunal and the importance of not overturning a sound decision because there are imperfections in presentation. Equally, the statutory task is such that, just as the employment tribunal must conduct a critical evaluation of the scheme in question, so must the appellate court consider critically whether the employment tribunal has understood and applied the evidence and has assessed fairly the employer's attempts at justification.

The power and duty of the employment tribunal to pass judgment on the employer's attempt at justification must be accompanied by a power and duty in the appellate courts to scrutinise carefully the manner in which its decision has been reached. The risk of superficiality is revealed in the cases cited and, in this field, a broader understanding of the needs of business will be required than in most other situations in which tribunals are called upon to make decisions."

50. He argued that the Tribunal had fallen into the trap of carrying out very much the same evaluation exercise in relation to assessing the feasibility of accommodating the Respondent's request for flexible working, rather than focusing on the PCP itself, which had to be justified. The Appellant's evidence had been directed on the operation needs of the business in order to fulfil the contractual obligations to the airlines which it served, and both in the decision and the review he contends that the Tribunal failed to assess properly or at all the Appellant's attempts at justification and failed to show that the Tribunal conducted a proper critical evaluation of the operational needs of the business.
51. In response Mrs Kerr argues that the Appellant had failed to advance an evidential case of any substance whatsoever in relation to justification. In particular no evidence was presented which could permit the Tribunal to make a finding in the Appellant's favour on this issue. She reminded us that in Mr Matovu's closing submissions to the Tribunal, the issue of justification was dealt with in one brief paragraph, raising the issue of operational needs, but not adding to it with any detail. She contended it was not a matter which was pursued at length or in any detail before the Tribunal. In cross-examination, Mr Cathrall could not explain why he felt it appropriate to discuss the possibility of shift changes with the Duty Controllers in Terminals 1 and 2, but not in Terminal 3. On the issue of operational requirements, she contended that the evidence demonstrated that it was Marimar Torres' insistence that she no longer wanted to work all late shifts that caused the Appellant to impose the fixed rotational system upon the Respondent. The Respondent had never alleged that part of the shifts should be left uncovered, but rather the issue was the shift requirements imposed on each individual employee. She contended that the Appellant's witnesses had insisted that their ability to organise rosters was unaffected by scrapping the old system, and she argued that the only potential operational requirement that the Appellant relied on would have been the difficulty of arranging flexible rosters, but this was ruled out by the Appellant's witnesses themselves. Had the Appellant provided evidence that they had discussed with the controllers in Terminal 3 their willingness to change shifts and been met with refusal, then as the Tribunal indicated in the review decision, that would have supported the Appellant's case.
52. We are satisfied that the Tribunal properly analysed such evidence that was put before them on the issue of justification by the employers. In the absence of clear evidence as to the difficulties of arranging flexible rosters, in particular the evidence regarding the Duty Controllers in Terminal 3, we are quite satisfied that the Tribunal were entitled come to the view that the PCP had not been objectively justified. For these reasons, we dismiss this appeal.