Equal Opportunities Commission v Secretary of State for Trade & Industry

The Equal Opportunities Commission (“EOC”) is the body authorised by s53(1) of the Sex Discrimination Act 1975 (“the 1975 Act”):

“(a)

to work towards the elimination of discrimination

(aa) to work towards the elimination of harassment that is contrary to … [the 1975] Act

(b)

to promote equality of opportunity between men and women generally …

(c)

to keep under review the working of [the 1975] Act and the Equal Pay Act 1970 and, when they are so required by the Secretary of State or otherwise think it necessary, draw up and submit to the Secretary of State proposals for amending it.”

The EOC contends that the Government has failed in certain respects to comply with its obligation as a Member State in the manner in which it implemented Directive 2002/73/EC of the European Parliament and of the Council of 23 September 2002, amending Council Directive 76/2007/EEC, (“the Directive”) in the form of the Employment Equality [Sex Discrimination] Regulations 2005 (“the Regulations”) which, when they came into force on 1 October 2005, introduced amendments to the 1975 Act. There was a great deal of consultation and communication between the Government and the EOC prior to the legislation being enacted, but the EOC was not able to convince the Government as to some aspects, nor to persuade it to reconsider once the legislation was enacted, in response to a detailed protocol letter before action dated 15 November 2005 sent to the Secretary of State for Trade and Industry. Responsibility for discrimination legislation now vests in the Secretary of State for Communities and Local Government, but proceedings are brought by the EOC as Claimant, for whom Miss Dinah Rose QC has appeared with Ms Karon Monaghan, against the Secretary of State for Trade and Industry, as progenitor of the legislation, represented by Mr David Pannick QC and Mr Gerard Clarke.

There have been six such aspects of the legislation which the Claimant has pursued in these proceedings, although, as will be seen, the first two, and to an extent the fourth, have blended together in argument. They can be summarised as follows:

I shall turn to deal with each of these six aspects in turn. I shall not at this stage analyse fully what it is that the Court is seeking to achieve and/or what remedies are available, unless and until I have identified whether there are any of the defects or failures pointed out by Miss Rose, of what they consist and what their impact is. It is common ground that regression is not permitted, as set out in paragraph 3(v) above. It is also common ground that Richards J, as he then was, in R (Amicus – MSF) v Secretary of State for Trade and Industry and Others [2004] IRLR 430 concisely and cogently summarised the relevant general principles which govern an investigation by the Court into whether there has been lawful and adequate implementation of Directives. The cornerstone is that which he set out in paragraph 47:

“Although Member States are free to choose how a Directive is implemented, they must adopt in their national legal systems all the measures necessary to ensure that the Directive is fully effective, in accordance with the objective which it pursues: Von Colson v Land Nordrhein – Westfalen [1984] ECR 1891 at 1906-7, paragraphs 15 and 18. It is inherent in Article 249 EC, and is clear from Von Colson and later authorities, that a Member State is not required to copy out the exact wording of the Directive. It has considerable flexibility in implementation, provided that the requisite result is achieved.”

There are other issues to be addressed, namely certainty, clarity, purposive interpretation and the effect on third parties, to which I will return.

I turn to deal with each of these aspects in turn.

“On the grounds of”

The argument which ranged widely in correspondence and in skeleton arguments slimmed down considerably in the light of Mr Pannick’s response to Miss Rose’s oral submissions. Those submissions were that “on the grounds of sex” is the appropriate definition for discrimination, where causation or ratiocination is required, but that harassment is distinct, and cannot be defined in the same way, dependent as it is simply on a connection or association with sex. The second and more explicit form of harassment set out in s4A(1)(b), relating to “unwanted conduct of a sexual nature …” with the proscribed purpose or effect (“sexual harassment”) is set alongside the definition, in s4A(1)(a), of the less explicit unwanted conduct in relation to sex, whose prohibition was also required by the Directive.

She accepted that, if s4A does, as she submits, incorrectly define the concept of harassment in the sphere of sex, then the equivalent provisions of the other discrimination statutes or regulations (s3A of the Race Relations Act 1976, s3B of the Disability Discrimination Act 1995, paragraph 5 of the Employment Equality (Religion or Belief) Regulations 2003 and paragraph 5 of the Employment Equality (Sexual Orientation) Regulations 2003) are also unlawfully formulated as not implementing the relevant Directives (Council Directive 2000/43/EC and Council Directive 2000/78/EC, in both of which harassment (as contrasted with discrimination) has the identical definition to that in the Directive.

She submits that what has happened is that the wording aptly relevant to discrimination has been repeated in the UK statute in relation to harassment, when the concepts are, and are intended to be, distinct. In the 1975 Act, direct discrimination is defined in s1(1):

“In any circumstances relevant for the purposes of any provision of this Act … 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.”

Apart from the addition of a comma in s4A(1)(a), set out in paragraph 3(i) above, – which appears to make no substantive difference - the wording is identical, and yet the concepts are intended to be distinct, and the importance of causation to the concept of discrimination has been emphasised in Nagarajan and Shamoon, as discussed above. Miss Rose however refers not only to the definition of harassment in the Directive, also set out in paragraph 3(i) above, and to be contrasted with the Directive’s definition of direct discrimination in the same Article, which, as does the UK statute, founds it solidly on causation – “direct discrimination [is] where one person is treated less favourably on grounds of sex than another is, has been or would be treated in a comparable situation” – but also in Recital (8) of the Directive:

“Harassment related to the sex of a person and sexual harassment are contrary to the principle of equal treatment between women and men.”

The difference in practice between direct discrimination on grounds of sex and harassment related to sex is illustrated by Miss Rose by reference to examples drawn from a number of cases, all of them of course decided by reference to direct discrimination under s1, Porcelli v Strathclyde Regional Council [1996] ICR 564, Brumfitt v Ministry of Defence [2005] IRLR 4, Kettle Produce Ltd v Ward (EAT unreported 8 November 2006 EATS/0016/06) and B v A (EAT unreported 9 January 2007 EAT/0450/06). If, she submits, conduct cannot be shown to have been discriminatory, in that the reason for the conduct cannot be shown to have been on grounds of sex, it should still be capable of being shown, if it is otherwise unwanted conduct with the relevant purpose or effect, to have been harassment, if it related to sex.

Thus the training officer in Brumfitt was found, by dint of the generally unpleasant nature of his language and the fact that the audience was of mixed sexes, not to have discriminated against the claimant on grounds of sex. Given that the Tribunal decided that the claimant had been exposed to language which was “offensive and humiliating to her as a woman”, it appears likely that she would have succeeded in a claim in respect of unwanted conduct related to her sex. Similarly, by reference to the facts of B v A, a claimant, who was unfairly treated on the grounds of jealousy because of her conduct with another man, may not be entitled to claim discrimination on grounds of sex, but would appear likely to be able to succeed in a claim for harassment by reference to unwanted conduct related to her sex. Again, by analogy from the facts in Kettle, a manager barging into the ladies toilet, when he would be likely to have similarly barged into a men’s toilet, may not render his employer liable for discrimination on grounds of sex, but such conduct would be likely to be conduct related to sex.

It is in this context that her argument on this aspect blends into that on the second point (paragraph 3(ii) above), whereby conduct directed towards a third party – whether a man or a woman – in the presence of the complainant (and thus plainly not harassment of the complainant on the grounds of her sex) could and should amount to harassment as intended by the Directive if such conduct related to sex, and had the effect, if not the purpose of e.g. creating an intimidating or humiliating or offensive environment for her.

None of these examples of unwanted conduct could (though Miss Rose submits they plainly should) come within the definition of s4A, so firmly rooted as it is in causation, particularly given the identical wording in s1, so often defined and constrained by the courts. It would be no answer even to seek to construe the words on the grounds of her sex widely, so as to assimilate it to the statutorily wider definition to be found in s3A of the Disability Discrimination Act, where it is discrimination if a person treats another less favourably “for a reason which relates to the disabled person’s disability”: albeit wider, this is still causation-based, as would be expected of a complaint of discrimination, as opposed to the basis for the distinct complaint of harassment.

Mr Pannick did not in the end dispute that the qualification for harassment was, in his words, associative rather than causative, although he emphasised that the examples that Miss Rose used were, as he put it, on the margins, i.e. effectively exceptional cases. He had pointed to what he contended were variations in the wording in the various directives, which suggested that there was not the clear distinction between discrimination and harassment, for which Ms Rose contended. There was some force in his submission in this regard by reference to the draft proposal for the Council Directive, published in Brussels on 7 June 2000, with an Explanatory Memorandum, to which I will refer below. However, so far as concerned his trawl of the existing directives, there did not seem to me to be anything to support his proposition. It is true that occasionally the words “based on” are used instead of “on the ground of”, but in the context of the definition of discrimination, and always carrying the implication of causation.

He pointed to Article 1.2.6 in the Directive:

“Member States may provide, as regards access to employment including the training leading thereto, that a difference of treatment which is based upon a characteristic related to sex shall not constitute discrimination where … such a characteristic constitutes a genuine and determining occupational requirement …”

However it is clear that discrimination, as opposed to harassment, is there being referred to, and that “based on” implies causation, and the use of the words “characteristic related to sex” are being used in a manner such as, rather as in relation to s3A of the Disability Discrimination Act referred to above, to expand the causation.

He also pointed to Article 1.2.3:

“Harassment and sexual harassment within the meaning of this Directive shall be deemed to be discrimination on the grounds of sex and therefore prohibited.”

This was in fact relied upon by both sides. I am satisfied that it means, and Mr Pannick did not in the event vigorously contend against this, that harassment, however itself defined, is here to be deemed to be discrimination and hence prohibited, but that the provision does not thereby assimilate the two different definitions.

The same can be applied to Recital (8), which I have partially cited above, and which I recite again in full:

“Harassment related to the sex of a person and sexual harassment are contrary to the principle of equal treatment between women and men: it is therefore appropriate to define such concepts and to prohibit such forms of discrimination. To this end it must be emphasised that these forms of discrimination occur not only in the workplace but also in the context of access to employment …”

Once again, this is relied upon by both sides. Miss Rose refers to the emphasis which is thus placed upon careful definition, which then follows in Article 1.2.7, and I am again satisfied that this Recital simply means that harassment, as defined, is, like direct and indeed indirect discrimination, as defined, a form of discrimination which is to be prohibited.

Mr Pannick did seek, in his skeleton argument, to contend that (paragraph 20) “there is no material difference between the two terms” and (paragraph 31) that “it is wrong to draw a subtle (or indeed any) discrimination between “on the ground of “ and “related to””. In this context, he referred to the draft Commission proposal with attached Explanatory Memorandum to which I have referred in paragraph 13 above. The Explanatory Memorandum was plainly drafted by reference to the definition then appearing in the European Commission’s Code of Practice, as referred to in paragraph 2.1 of the Memorandum, at paragraph 18: “sexual harassment means unwanted conduct of a sexual nature, or other conduct based on sex affecting the dignity of women and men at work”. This was not the definition eventually adopted. The draft Directive attached was also not in the same form as was eventually approved by the European Parliament and Council. The draft proposal, in the form in which it was at the time of the Explanatory Memorandum, appears not to have had a definition section, but only an equivalent deeming section to that in Article 1.2.3, albeit a deeming provision which appears to have been also an operative provision:

“1a. Sexual harassment shall be deemed to be discrimination on the grounds of sex at the workplace when an unwanted conduct related to sex takes place with the purposes or effect of …”

Miss Rose relied upon the fact that there was the change between the draft proposal and the eventual Directive, resulting in the careful definitions which I have recited.

Mr Pannick effectively did not pursue this case that there is no material difference in the definition, in his oral submissions. His case was that the words “on the ground that” could be construed broadly and purposively so as to mean “related to”, i.e. so as to exclude the need for causation. There is, as Miss Rose points out, a limit to the role of purposive construction. She pointed out the further passage in Richards J’s judgment in Amicus, by reference to the decision of the European Court in Commission v France [1997] ECR 1-1489, at 1-1501 paragraph 15, when he summarises:

“Member States’ broad discretion to the manner of implementation is circumscribed not just by the need to achieve the required result but also by the principle of legal certainty.”

At paragraph 56, he palliates this to an extent by saying that the further authorities of the European Court there cited:

“… show that if the legal position is not sufficiently precise and clear, as where there has been no specific implementing measure or only an inadequate implementing measure, a Member State cannot plug the gap by reliance on the obligation of the national courts to interpret national law compatibly with Community obligations. It does not follow, however, that where detailed implementing measures have been adopted, it is impermissible to have regard to the interpretative obligation of the national court when determining whether the national measures comply with the Directive and are lawful.”

Miss Rose relies on the fact that the wording in s1(1) is, subject to the comma, identical and has been so carefully addressed and construed, but must now, if Mr Pannick is right, be construed, albeit it identical, as meaning something different in s4A. In any event, she submits, it is impossible to construe “on the ground of her sex he engages in unwanted conduct” other than by reference to causation. Whether it is that a person has engaged on the ground of a complainant’s sex in certain conduct, or whether he engaged in certain conduct on the ground of the complainant’s sex, on either basis it is plain that the Court must address “the reason why”.

Mr Pannick sought to make a further, alternative submission in order to meet this argument. He submitted that purposive construction could include a reading of the section which involved notionally moving the words “on the ground of her sex” along the sentence, so that that subsection could be construed as meaning “he engages in unwanted conduct that has the purpose or effect on the ground of her sex of violating … creating …”. This, he submitted, would have the desired effect of eliminating the objectionable causative approach. He submitted that this kind of purposive interpretation would accord with that carried out by the House of Lords when they were required to reconstrue the plain words of s1(1) in Webb (No 2).

In Webb (No 1) the European Court had concluded that, by reference to the combination of s1 and s5(3), discrimination on grounds of pregnancy or maternity leave, which could alone fall within s1, was being made subject to a requirement for a male comparator. I have already set out s1(1) in paragraph 8 above. S5(3) read at that time as follows:

“A comparison of the cases of persons of different sex or marital status under s1(1) or 3(1) must be such that the relevant circumstances in the one case are the same, or not materially different, in the other.”

The House of Lords, driven by the European Court’s judgment to eliminate any question of a comparator, felt able to do so by delimiting the definition of the words “relevant circumstances” (see Lord Keith at 1026H-1027B). So in Litster v Forth Dry Dock Co Ltd [1991] AC 546 the House of Lords implied into the Transfer of Undertaking Regulations additional words to achieve the intended effect of the relevant Council Directive. Commissioner of European Communities v United Kingdom [1997] 1 ECR 2649 [1997] AER(EC) 461 was a case where the Commission brought a pre-emptive strike in respect of legislation which, albeit not tested in the English courts, the Commission argued could not be construed compatibly with the Directive: although the relevant provision was in very different, and seemingly inconsistent, terms, the European Court found that the Commission had not proved its case. Mr Pannick heavily relied upon the principle in Marleasing SA v La Comercial Internacional de Alimentacion SA [1990] ECR 1-4135, particularly at paragraph 8 of the judgment, that “in applying national law … the national court called upon to interpret it is required to do so, as far as possible, in the light of the wording and the purpose of the directive in order to achieve the result pursued by the latter and thereby comply with the third paragraph of Article 189 of the Treaty”.

Miss Rose submitted the following:

Miss Rose formulated a proposed amendment, for which she sought leave in reply, in response to Mr Pannick’s submission, which read as follows:

“It is irrational for the Secretary of State to make or to maintain regulations for the purposes of implementing the Directive which he concedes to be incompatible with the Directive on their ordinary construction, and which he concedes can be read compatibly with the Directive only by a strange construction which would not be evident to, amongst others, employers or employees reading the legislation.”

I did not rule on that amendment, which was simply left on the table, because it appeared to me that, if she was right in that submission, then the proper course would be to conclude that such purposive interpretation could not and should not save the legislation, rather than to address any question of Wednesbury unreasonableness.

On the one hand, Miss Rose did not submit that I ought to be the more stringent in my approach to the legislation simply because this was an action brought to test that legislation, in which the rights of an individual party was not at risk: obviously it would be the more tempting to try to salvage the legislation if the rights of one or more claimants hung upon it, but she accepted, and indeed affirmed, that there was only one correct way to interpret or construe the legislation, and a different answer could not be given in that regard dependent upon the nature of the proceedings in which the issue had to be resolved.

On the other hand, although Mr Pannick did urge that the Court should be slow to conclude that legislation was unlawful, particularly where it would leave a lacuna, he accepted both that, if the legislation were unlawful and had to be re-enacted, it would and could be done by secondary legislation, i.e. the passage of amending regulations, rather than the need for the full panoply of Parliamentary procedure and also that there would be, if I were not persuaded by his submissions, an option for me, whether coupled with declaratory relief or otherwise, to adjourn the proceedings, having made my decision clear, to allow the Government to bring forward fresh Regulations, without the present ones having been struck down.

I shall return to my decision on this question later in my judgment, so as to be able to draw together all the various aspects and make a decision on them all at the same time.

On the ground of her sex

There was not in the event dispute between the parties, by reference to the example in paragraph 24(iii) above, that there could be harassment of a woman if the effect of denigratory conduct, directed towards another party – perhaps a man – related to sex, but not of a sexual nature, had the effect of creating a humiliating or offensive environment for her. This would not be available on a straightforward reading of s4A, as it would not be conduct on the ground of her sex. ‘Reading down’, if available, would be necessary.

The objective test

Miss Rose, in her skeleton argument, launched an attack on the allegedly inappropriate introduction of an objective test by s4A(2) of the 1975 Act, set out in paragraph 3(iii) above.

However, it is clear that, as Richards J expounded, the Member State has “considerable flexibility in implementation” (paragraph 47 of Amicus). Article 249 of the EU Treaty itself expresses this:

“A Directive shall be binding, as to the result to be achieved, upon each Member State to which it is addressed, but shall leave to the national authorities the choice of form and method.”

Mr Pannick put the Defendant’s case on this aspect forcefully in his skeleton argument:

“54.

The Secretary of State contends that the European Commission’s Code of Practice supports the test adopted by the … Regulations. The Code indicates at paragraph 2 that conduct is unacceptable if it is “unwanted, unreasonable and offensive to the recipient”. This supports a test which requires the conduct complained of to be objectively unreasonable as well as being offensive to the recipient.

55.

It may well be that in real, as opposed to hypothetical, cases, tribunals will be very slow to find that the perception of a complainant is unreasonable, but it would be wrong to elevate the complainant’s perception to the status of the sole and determinative issue in the case. If this were done, tribunals would face the invidious task of having to determine whether a claimed perception was genuine.”

Miss Rose had relied upon dicta of Morison P in Reed & Bull Information Systems Ltd v Steadman [1999] IRLR 299, which supported the proposition of an objective test. However, even in that judgment, he made it clear that there would be the kind of difficulties of proof as to whether conduct was unwelcome as are referred to by Mr Pannick above. In any event, I agree with Mr Pannick that the later decision of the EAT per Holland J in Driskel v Peninsula Business Services Ltd [2000] IRLR 151 is to be preferred, in which he gave very helpful guidance at 155, paragraph 12(d)(3):

“(3)

The ultimate judgment, sexual discrimination or no, reflects an objective assessment by the tribunal of all the facts. That said, amongst the factors to be considered are the applicant’s subjective perception of that which is the subject of complaint and the understanding, motive and intention of the alleged discriminator. Thus, the act complained of may be so obviously detrimental, that, disadvantageous (see Insitu [1995] IRLR 4) to the applicant as a woman by intimidating her [or] undermining her dignity at work, that the lack of any contemporaneous complaint by her is of little of no significance. By contrast she may complain of one or more matters which if taken individually may not objectively signify much, if anything, in terms of detriment. Then a contemporaneous indication of sensitivity on her part becomes obviously material as does the evidence of the alleged discriminator as to his perception. That which in isolation may not amount to discriminatory detriment may become such if persisted in notwithstanding objection, vocal or apparent. … By contrast the facts may simply disclose hypersensitivity on the part of the applicant to conduct which was reasonably not perceived by the alleged discriminator as being to her detriment – no finding of discrimination can then follow.”

In those circumstances I am satisfied that, under the old pre-s4A law of discrimination, the test was properly to be regarded as objective in accordance with the decision of Holland J. Thus no question of regression arises.

On further consideration, Miss Rose, both in opening submissions and in reply, was prepared to concede that s4A(3) could not be said to be unlawful if, as she submitted, it fell to be interpreted in accordance with the submissions of Mr Pannick as set out in his skeleton argument above. Mr Pannick confirms that that is the inevitable result, as he submits, of the very wording of s4A. In those circumstances this challenge is no longer pursued and fails.

Third Party Liability

Again the difference between the parties narrowed in the course of the hearing. The Defendant’s position was made clear, in a Fact Sheet on harassment published by the Women and Equality Unit of the new Department of Communities and Local Government, in paragraphs 37 and 38 of the Defendant’s Summary Grounds and in Mr Pannick’s skeleton argument. In the Fact Sheet, published in October 2006, it was stated at page 4:

“Nevertheless, it might be argued, on appropriate facts, that an employer should take steps to protect an employee from third-party actions, which provide an offensive working environment for employees, in respect of which the employer might have some degree of control, and in such a case liability might arise if an employer fails to do so. So, on appropriate facts, the harassment provisions in the [1975 Act] might be interpreted so that where an employer knowingly fails to protect an employee from, for example, repetitive harassment by a customer or a supplier, the employer is “subjecting the employee to harassment”.”

Similar points are made in paragraph 70 of Mr Pannick’s skeleton:

“Adopting this approach does not necessarily exclude the possibility that an employer could be held liable on appropriate facts for the conduct of, for example, a supplier or customer (or, more accurately, held liable for the violation of dignity or unwelcome working environment brought about by such conduct). It might be the case that an employer could be held liable for failing to take action where there is a continuing course of offensive conduct, which the employer knows of but does nothing to safeguard against. The employer could be responsible for failing to act, albeit not responsible for the third party’s actions in themselves. By contrast, fixing an employer with liability arising from a single act by a third party could go too far.”

Miss Rose’s primary submission was that the 1975 Act failed to implement the Directive by imposing third party liability. She accepted however the difficulty of such a submission. She referred, in regard to its general reference to the workplace, to Recital 8 of the Directive, which I have already cited in full at paragraph 16 above, and also to Recital (9):

“In this context, employers and those responsible for vocational training should be encouraged to take measures to combat all forms of sexual discrimination and, in particular, to take preventive measures against harassment and sexual harassment in the workplace, in accordance with national legislation and practice.”

There is an exhortatory Article 1.2.5:

“Member States shall encourage, in accordance with national law, collective agreements or practice, employers and those responsible for access to vocational training to take measures to prevent all forms of discrimination on grounds of sex, in particular harassment and sexual harassment at the workplace.”

She accepted however that there is nothing explicit, or even arguably implicit, in any of the Articles requiring a Member State to impose vicarious liability on an employer, or indeed liability for negligent failure to take steps, such as were expressly found by the House of Lords in Pearce not to exist in the present discrimination legislation. It became apparent in the course of her submissions that her real best position was that which she soon adopted, namely that, whereas the position taken by the Defendant, by reference to the passages in the Fact Sheet and the skeleton referred to above, would, at any rate from her point of view arguably, constitute adequate compliance with the Directive, yet on the face of paragraph 4A, as it stands, it is impossible to see how such a result could be achieved. This once again returns me to consideration of the respective arguments under the first aspect above. So long as s4A is to be framed in terms of unwanted conduct engaged in on the ground of her sex by the employer, it seems difficult, if not impossible, to see how an employer could be held liable simply for even knowing failure to take steps to prevent harassment by others. If, by reference to the disapproved authority of De Vere Hotels, it could have been shown that the employers knew of continuing and/or regular objectionable conduct by Mr Manning, and failed to take any steps to prevent it, it could be said that they were thereby themselves indulging in unwanted conduct (including omission) in relation to sex, with the consequent upsetting effect on the claimant waitress. However, it would seem very difficult to be able to say that such knowing failure on their part would amount to unwanted conduct by the employers on the ground of her sex. However, the result of adopting the associative rather than causative approach to harassment, either by a purposive and transliterative construction such as is urged by Mr Pannick or by its replacement by wording more compatible with Article 1.2.2, as urged by Miss Rose, would resolve the problem. Hence once again the issue, so far as it survives in the abbreviated form which alone Miss Rose now pursues, is bound up with my conclusion on the first aspect.

In the light of Mr Pannick’s clarification referred to in paragraph 3(v) above, s3A is to be the only vehicle for the bringing of a claim for sex discrimination in the case of pregnancy/maternity leave, so that if such section offends against European Law, by reference to its imposition of a requirement of a comparator, it would constitute regression. The only issue accordingly is whether it does.

It is common ground, as set out in paragraph 3(i) above, that Webb No 1 rendered it unlawful to require a male comparator in such circumstance. On the face of the passage from the European Court’s judgment in Gillespie at paragraph 17, from which I have quoted in paragraph 3(v) above, it seems, at any rate in relation to a maternity leave claim, that there is no distinction to be made between the use of a male and female comparator. Nevertheless:

Mr Pannick submits that all that s3A does is to emphasise the importance of the guidance given by Mummery LJ, and to support the approach which the Court is any event adopting. The Defendant addresses the examples given by Miss Rose as follows:

If, contrary to Mr Pannick’s primary contentions, s3A did not comply with European Law as it stands, then he submitted that it can be read down, so as to exclude the requirement for a comparator, as the House of Lords achieved in Webb No 2.

Miss Rose pointed out that in Madarassy itself Mummery LJ only puts the comparator test forward as of evidential assistance, against a background that comparators are in fact not required. It can be seen, she submits, that it may be a way of testing whether the treatment, the less favourable treatment, of a pregnant employee is by virtue of her pregnancy rather than for some other reason. However, she submits, it must not be elevated into a requirement. She emphasises the special position occupied by pregnant women and those on maternity leave, as spelt out by the European Court in Gillespie at paragraph 17, set out in paragraph 3(v) above. She further points to paragraph 16 of Gillespie, where the Court said:

“16.

It is well settled that discrimination involves the application of different rules to comparable situations or the application of the same rules to different situations.”

She submits that, as a result, there can be discrimination of a pregnant woman if she is not dealt with the same as if she were not pregnant, but also if she is dealt with the same as if she were not pregnant: and the special protection is not limited to the context of a lawful requirement for a risk appraisal pursuant to the Pregnant Workers Directive, as indeed the second Fact Sheet seems to recognise. It is not possible to restrict the less favourable treatment to a comparator test. The less favourable aspect is simply that she is receiving less favourable treatment by reason of her pregnancy, i.e. the treatment could be treatment which must be special because she is pregnant, when if she were not pregnant the same treatment would not be unfavourable treatment. In the event, Mr Pannick appeared to accept that the argument which was postulated by the Defendant in the Summary Grounds with regard to the toilet break would in fact fail if it fell to be tested within the words of s3A as they stand. In any event, the pregnant woman could not be compared with a woman with a pathological condition (paragraph 25 of the judgment in Webb No 1) relating to her bladder, and so, if she needs to be compared with herself not pregnant and without a bladder problem, i.e. with no need for frequent visits to the toilet, she would fail on the words of s3A.

With regard to reading down, Miss Rose repeated the submissions that she made with regard to the same suggestion in relation to the first aspect relating to harassment. The suggestion was simply that s3A should be read as if it said that a person would discriminate against a woman if - so far as concerns pregnancy - a person treats her less favourably (or Miss Rose would prefer unfavourably) for a reason related to pregnancy. This would be a much wider exercise than the implication of words as in Litster or than occurred in Webb No 2, when, although no doubt the result, that is of excluding reference to a comparator, was the same, all that in fact occurred was a (very strained) definition of the words “relevant circumstances”.

I refer to the summary in paragraph 3(vi)(a) above. Mr Pannick did submit that s6A(1) did not exclude from the protection of a claim for discrimination the recovery of benefits during the compulsory period which were not derived from contractual terms and conditions i.e. it entitled a claim in respect of a discretionary bonus such as that in Denda remained available, because, on a proper construction, s6A(1) only extended to exclude benefit from contractual terms and conditions. However, not only because contractual terms and conditions relating to remuneration would be subject to the Equal Pay Act 1970 (and thus, by reference to s8(5) of the 1975 Act, not be subject to the 1975 Act) but, in any event, it is likely that s6A(1) would be construed similarly to the provisions of s71 of the 1996 Act relating to maternity leave itself, whereby (subject to the exclusion, in subsection 71(5)(b), of terms and conditions about remuneration) terms and conditions of employment are by s71(5)(a) defined, for maternity leave purposes, as including “matters connected with an employee’s employment whether or not they arise under her contract of employment”. The second Fact Sheet states at page 9:

“Discretionary bonuses are covered by the SDA, whereas contractual bonuses fall within the EPA.

Whether discretionary or contractual, a bonus should be paid for work done or attendance prior to maternity leave, and following the woman’s return, and should include payment for the two-week period of compulsory maternity leave (see EPA section 1(2)(e).”

In the circumstances, in order to avoid regression, Mr Pannick is left only to assert that, whereas “ordinary maternity leave” is defined by s6A(7) to include “compulsory maternity leave”, where those words are used in s6A(1) they should be construed so as to exclude compulsory maternity leave: alternatively an implicit exception to the exception should be read in, so as to preserve a Denda claim for loss of discretionary bonus during the compulsory maternity leave period.

Again I refer to the summary of this point at paragraph 3(vi)(b), where I have set out subsections (iii) and (iv) of s6A of the 1995 Act. A number of examples is set out in subsection 6A(4) of cases where discrimination can still be claimed in respect of conduct during the additional maternity leave period; but otherwise a distinction is drawn between the ordinary leave period, when there is no limit on the kind of discriminatory claim that can be made, and the additional period.

Examples have been given by Miss Rose, some of which are argued by Mr Pannick to be capable of being fitted in within the present exclusive list of permitted claims. Thus, a claim for inadequate recognition of length of service might fall within s6A(4)(i), and a complaint of lack of consultation in relation to redundancy might fall within s6A(4)(i)(ii).

However, at page 10 of the second Fact Sheet there is advice given as to “examples of benefits that continue to apply during ordinary maternity leave and additional maternity leave”; and it is said that it may be discriminatory to fail to consult a woman, not only about pending redundancies, as to which Mr Pannick has made the suggestion above, but also with regard to “a reorganisation or changes effecting the woman’s job or working conditions” to fail “to inform a woman about pay rises, bonuses, promotion or vacancies”, and to fail to appraise, or give an annual assessment to, a woman, for a reason relating to her pregnancy or her absence on maternity leave. If that advice is right, it is difficult to see how it is to be reconciled with the express exclusion of such conduct as being actionable discrimination during the additional leave period by s6A(4). Mr Pannick indicates the probable provenance of the subsection as being paragraph 17 of the Maternity and Parental Leave etc Regulations 1999, where all the specific inclusions in the subsection, save the reference to membership of a pension scheme, can be found as being benefits preserved during the additional leave period. However, it is one thing to have a regulation which preserves certain rights, which may well be construed as not being exhaustive, and another to have a subsection which, as here, excludes any rights save for those specifically referred to.

Mr Pannick’s answer is effectively to disavow the content of the second Fact Sheet in this regard, and to contend that it is not in fact unlawful to exclude certain benefits, or claims for discrimination in respect of denial of benefits, during the additional, as opposed to the ordinary, leave period.

Miss Rose referred, as set out in paragraph 3(vi)(b) above, to the European Court judgment in Sass. That decision involves somewhat unusual facts, in that maternity leave periods were longer in East Germany prior to reunification than they were in West Germany, and the European Court concluded that it was discriminatory to seek to limit the claimant to the minimum period of 14 weeks laid down by the Pregnant Workers Directive in terms of calculating her seniority, as opposed to making allowance for the full period that she had been entitled, under East German state law, to take.

Mr Pannick sought to place reliance on an earlier decision of the European Court in Boyle and Others v Equal Opportunities Commission [1999] ICR 360, where it was held that it was not discriminatory to make special provisions in relation to an additional period of maternity leave which was not imposed statutorily, but was applied by an individual employer (see paragraph 71 of that judgment). Boyle was distinguished by the Court in Sass, by reference to the fact that the additional period of maternity leave in that case was granted by legislation:

“47.

As regards the judgment in Boyle and others it must be held that, contrary to the contentions of the Commission, that judgment does not in any way prejudge the answer to the question referred in the present case since Boyle concerned additional leave granted by an employer rather than statutory leave.”

Subject only to one caveat, which was left to the National Court, as will appear, the conclusion was that there could and should be no difference between the ordinary period and the additional period, if such was granted statutorily. The caveat appears from paragraph 58 of the judgment:

“Accordingly, if a national court reaches the conclusion that the maternity leave provided for … is such statutory leave intended to protect women who have given birth, the whole of that leave must be counted towards the qualifying period to be completed in order to be classified in a higher salaried grade, to prevent a woman who has taken such leave from being placed in a worse position because of her pregnancy and her maternity leave, than a male colleague who started work in the former GDR on the same day as she did.”

It is not suggested by Mr Pannick that the caveat can here be relied on, i.e. it is accepted that the intention of the additional leave is the same as that of the ordinary leave, namely further to protect women who have given birth. His submission is that I should not follow Sass, because it is a decision of the First Chamber of the European Court, but rather follow Boyle, which was a decision of the Grand Chamber. I am entirely satisfied, applying ordinary principles of English jurisdiction, that it cannot be argued that the First Chamber were acting per incuriam, where they so carefully addressed and distinguished Boyle. In any event, I conclude, however, that I am as bound by s3(2) of the European Communities Act 1972, as amended by s2 of the European Communities (Amendment) Act 1986, to follow the decision of the First Chamber as that of the Grand Chamber.

Accordingly, given that I conclude that Sass bindingly enshrines European Law, to deprive a claimant of the opportunity to claim discrimination in relation to the kind of conduct that is set out in the second Fact Sheet would be regressive.

I have referred in paragraph 19 above to Miss Rose’s emphasis, by reference to the words of Richards J in Amicus, on the need for certainty which, she states, goes hand in hand with the need for clarity. It was clarity which was put in the forefront of the purposes of this legislation by the UK Government:

The inadequacies which have been identified by Miss Rose relate substantially to the definition of harassment and of discrimination during pregnancy or maternity leave, which, by reference to Recital (8), the Directive intended to address expressly.

It is obviously not just open to a national court, but its obligation, to set out to construe statutes and regulations passed by the Member State so as to render them compliant with a relevant Directive. But, by reference to the aspects which have been in issue at this hearing (save for the objective test point summarised in paragraph 3(iii) above, and addressed by me in paragraphs 28 to 33) both individually, and in any event collectively, I do not consider that to do so is appropriate – by virtue of the extent of reading down/transposition which would be required to be considered in order to render them compliant: or possible - because I am not persuaded that even such extreme application of the Marleasing principle would in any event be effective: or sensible – because of the need for clarity and certainty, and comprehensibility, by employees and employers alike.

On the one hand I am not influenced, as Miss Rose submitted I should not be influenced (see paragraph 26 above), by the fact that at this early stage in the implementation of the legislation these defects, all but admitted by Mr Pannick, could be easily put right by fresh Regulations, but, on the other hand, by virtue of what I have referred to as looking at the matters collectively, I am all the less persuaded that it is a sensible approach to leave these sections in place, to be interpreted (subject to any appeal) in the light of my judgment, when there are in fact three different sections of the Act which all need to be addressed (quite apart from the apparent knock-on effect of statutorily incorporating a correct definition of harassment, referred to in paragraph 7 above).

I am satisfied that:

I propose to deliver this judgment without making an order at this stage in respect of the relevant sections and/or the Regulations which led to them, but to give the Defendant the opportunity to consider the effect of my conclusions. I shall hear Counsel on the form of order.