ON APPEAL FROM The Employment Appeal Tribunal
HH Judge Eady QC
Royal Courts of Justice
Strand, London, WC2A 2LL
Before:
LORD JUSTICE PATTEN
LORD JUSTICE LEWISON
and
LORD JUSTICE UNDERHILL
Between:
MISS T BLACKWOOD | Appellant |
- and - | |
BIRMINGHAM & SOLIHULL MENTAL HEALTH NHS FOUNDATION TRUST | Respondent |
Christopher Milsom and Rachel Barrett (instructed by Bailey Wright & Co) for the Appellant
Edward Pepperall QC and Jonathan Meichen (instructed by Birmingham & Solihull Mental Health NHS Foundation Trust – Legal Department) for the Respondent
Hearing date: 5 May 2016
Judgment Approved
Lord Justice Underhill :
INTRODUCTION
The Appellant claims to have suffered sex discrimination in connection with a work placement which she was offered as part of her studies to become a nurse, and she has brought proceedings in the Employment Tribunal. The issue raised by this appeal is whether the ET had jurisdiction to entertain her claim or whether, as it and the Employment Appeal Tribunal both held, she should have proceeded in the County Court. That depends on whether the claim falls under Part 5 of the Equality Act 2010, which is concerned with discrimination at work, or under Part 6, which is concerned with discrimination in education. The issue is of some general importance because it is a standard part of very many educational courses with a vocational element – most obviously those for teachers and for doctors and nurses – that students undergo training or work experience arranged by their university with another institution such as a school or hospital. It is important that it is clear against whom, and in what forum, a claim can be made if they experience discrimination in the course of that placement.
The Appellant has been represented before us by Mr Christopher Milsom, leading Ms Rachel Barrett, neither of whom appeared below. The Respondent, the Birmingham and Solihull Mental Health NHS Foundation Trust, has been represented by Mr Edward Pepperall QC, leading Mr Jonathan Meichen.
THE FACTS
Nothing turns on the details of the facts in this case, and indeed no findings of fact have been made. The facts as alleged by the Appellant can be very shortly summarised. In September 2008 she started a three-year course at Birmingham City University for a Diploma for Higher Education in mental health nursing. As part of her course she was required to undertake work placements. The course was extended because she had a baby in 2010. In March 2012 the arrangements about work placements were formalised in a written agreement between her and the University. Under the agreement the University would approach NHS bodies to identify clinical placements and the Appellant agreed to accept placements to which she was allocated. In November 2012 she was allocated a place at a unit operated by the Trust. She attended on the first day and discussed with a manager what shift pattern she would be working. She explained that she would have difficulty working nightshifts and weekends because of her childcare responsibilities. The initial indication was that this would not be a problem but within a few days she was told that the placement was being withdrawn, apparently because the Trust believed that she was not prepared to work nights and that that was incompatible with the requirements of her university course and/or of the Nursing and Midwifery Council.
THE LEGISLATION
THE EQUALITY ACT 2010
Discrimination by “employment service-providers”
The Appellant’s claim is brought under section 55 of the 2010 Act, which is headed “Employment Service-Providers”. Section 55 falls under Chapter 1 of Part 5 of the Act (“Employment etc”). The relevant sub-sections for our purposes are (1) and (2), which read:
“(1) A person (an ‘employment service-provider’) concerned with the provision of an employment service must not discriminate against a person—
(a) in the arrangements the service-provider makes for selecting persons to whom to provide, or to whom to offer to provide, the service;
(b) as to the terms on which the service-provider offers to provide the service to the person;
(c) by not offering to provide the service to the person.
(2) An employment service-provider (A) must not, in relation to the provision of an employment service, discriminate against a person (B) –
(a) as to the terms on which A provides the service to B;
(b) by not providing the service to B;
(c) by terminating the provision of the service to B;
(d) by subjecting B to any other detriment.”
Sub-sections (3)-(5) go on to proscribe harassment and victimisation by employment service-providers.
Definition of “employment service-provider”
What constitutes the provision of an “employment service” is covered by section 56. Sub-section (2) reads (so far as material):
“The provision of an employment service includes –
the provision of vocational training;
the provision of vocational guidance;
making arrangements for the provision of vocational training or vocational guidance;
(d)-(i) …”.
“Vocational training” is defined in sub-section (6) as:
“(a) training for employment, or
work experience (including work experience the duration of which is not agreed until after it begins)”.
(I should say for completeness that none of heads (d)-(i) under section 56 (2) is concerned with vocational training: they cover other kinds of “employment services”, such for example as those provided by employment agencies.)
It is common ground in this case that the Trust was an employment service-provider as regards the Appellant because on the work placement it would be providing vocational training within the meaning of section 56 (2) (a) (as glossed in section 56 (6)). It does not matter whether the arrangement in her case is best described as “training for employment” or “work experience” or both: I will simply refer to it as a work placement. It follows that, but for the provisions which I go on to describe, the Trust’s alleged conduct would constitute a breach of section 55 – probably of sub-section (2) (c).
The exception for students
Sub-sections (3)-(5) of section 56 provide for three exceptions, as follows:
“(3) This section does not apply in relation to training or guidance in so far as it is training or guidance in relation to which another provision of this Part applies.
(4) This section does not apply in relation to training or guidance for pupils of a school to which section 85 applies in so far as it is training or guidance to which the responsible body of the school has power to afford access (whether as the responsible body of that school or as the responsible body of any other school at which the training or guidance is provided).
(5) This section does not apply in relation to training or guidance for students of an institution to which section 91 applies in so far as it is training or guidance to which the governing body of the institution has power to afford access.”
This appeal is concerned with sub-section (5), but sub-sections (3) and (4) are relevant in as much as the three appear to form part of the same group of provisions. I set out below the terms of section 91, to which sub-section (5) refers, but for present purposes it is enough to say that it falls under Part 6 of the Act (as also does section 85, referred to in sub-section (4)) and that the institutions to which it applies are universities and other further and higher education institutions – for short I will simply say “universities”. Thus its effect is that section 56 does not apply in relation to training of students of a university where the university “has power to afford access” to the training in question.
It was common ground before us that the intended consequence of disapplying section 56 is also to disapply the substantive provisions of section 55. Although the drafting is clumsy, I accept that that must be the correct reading.
The circumstances which trigger the operation of sub-sections (3)-(5) are all circumstances in which a remedy against discrimination is afforded by other provisions of the Act. In the case of sub-section (3) the only “other provisions” of Part 5 which apply to “training” (none refer to “guidance”) are those proscribing discrimination by the various people or institutions who provide work, such as employers (section 39), partnerships (section 44) or persons making appointments to public offices (section 50): the proscribed acts include discrimination in the way in which access is offered to (among other things) training – see, e.g., section 39 (2) (b), which I set out at para. 12 below. In the case of sub-sections (4) and (5) the two sections to which they refer, sections 85 and 91, are likewise sections which proscribe discrimination by the institutions in question – that is, schools and universities. That being so, it seems reasonably evident that the purpose of all three sub-sections is to prevent overlap – that is, to prevent claims in relation to the same “training or guidance” being brought under more than one provision. The effect of sub-sections (4) and (5) is that any claims arising out of discrimination in relation to training and guidance to which access is afforded by schools or universities would have to be brought under a different Part of the Act, which in turn means that they would have to be brought in the County Court rather than the Employment Tribunal. However, that is not so in the case of sub-section (3), which is designed to prevent overlap with other provisions in the same Part, in respect of which the Employment Tribunal is equally the allocated forum. Accordingly it does not appear that the draftsman was concerned, or in any event concerned only, with forum allocation.
Section 91 falls under Chapter 2 of Part 6 of the Act (“Further and Higher Education”). Sub-sections (1)-(3) and (6)-(8) proscribe various forms of discrimination by institutions “to which this section applies”. I need only set out sub-section (2), which reads as follows:
“The responsible body of such an institution must not discriminate against a student –
(a) in the way it provides education for the student;
(b) in the way it affords the student access to a benefit, facility or service;
(c) by not providing education for the student;
(d) by not affording the student access to a benefit, facility or service;
(e) by excluding the student;
(f) by subjecting the student to any other detriment.”
It should be noted that section 91 (2) follows the same broad pattern as the other provisions of the 2010 Act proscribing discrimination, to which I will refer as the “primary provisions” of the Act. Section 39 (2), for example, which proscribes discrimination by employers against employees, reads:
“An employer (A) must not discriminate against an employee of A's (B)—
(a) as to B's terms of employment;
(b) in the way A affords B access, or by not affording B access, to opportunities for promotion, transfer or training or for receiving any other benefit, facility or service;
(c) by dismissing B;
(d) by subjecting B to any other detriment.”
Specifically, the language of “affording access” to benefits is common to all the primary provisions.
The institutions to which the section applies are specified, so far as concerns England and Wales, by sub-section (10), as follows:
“(a) a university
(b) any other institution within the higher education sector;
(c) an institution within the further education section;
(d) a 16 to 19 Academy.”
Enforcement
“Indirect liability”
I need to set out the provisions of the 2010 Act in relation to liability for acts done by others. As a shorthand I will refer to this as “indirect liability”.
Sections 109 and 110 deal with the liability of employers and principals for the acts of their employees/agents and the liability of those employees/agents themselves. The relevant provisions are as follows:
“109. Liability of employers and principals
(1) Anything done by a person (A) in the course of A’s employment must be treated as also done by the employer.
(2) Anything done by an agent for a principal, with the authority of the principal, must be treated as also done by the principal.
(3) It does not matter whether that thing is done with the employer’s or principal's knowledge or approval.
(4)-(5)…
110. Liability of employees and agents
(1) A person (A) contravenes this section if –
(a) A is an employee or agent.
(b) A does something which, by virtue of section 109 (1) or (2), is treated as having been done by A's employer or principal (as the case may be), and
(c) the doing of that thing by A amounts to a contravention of this Act by the employer or principal (as the case may be).
(2)-(5) …
(6) Part 9 (enforcement) applies to a contravention of this section by A as if it were the contravention mentioned in subsection (1)(c).
(7) …”.
I should spell out the way that those provisions work in cases involving agency relationships. The primary provisions of the Act which proscribe discrimination, such as sections 55 or 91, do so by reference to the relationship of the putative discriminator with the victim – that is, what they proscribe is discrimination by, say, an employer against an employee or, relevantly to the present case, by the person providing the putative victim with an employment service or the university at which he or she is studying. If the person doing the act complained of is not in the relevant relationship they will not be directly caught by the relevant primary provision. However:
(a) The effect of section 109 (2) is that the principal of the person doing the actual act is treated as having done it himself, so that if he is in the relevant relationship, though the agent is not, he is brought within the scope of the primary provision and proceedings can be brought against him in the forum appropriate to that provision.
(b) In such a case the agent who did the act complained of is also rendered liable, by section 110 (1), on the back of the liability of the principal. Section 110 (6) provides that any claim against the agent must be brought in the forum appropriate to the contravention by the principal.
Sections 111 and 112 read (so far as material) as follows:
“111. Instructing, causing or inducing contraventions
(1) A person (A) must not instruct another (B) to do in relation to a third person (C) anything which contravenes Part 3, 4, 5, 6 or 7 or section 108(1) or (2) or 112(1) (a basic contravention).
(2) A person (A) must not cause another (B) to do in relation to a third person (C) anything which is a basic contravention.
(3) A person (A) must not induce another (B) to do in relation to a third person (C) anything which is a basic contravention.
(4)-(9) …”.
(I will as a shorthand refer to all three kinds of act proscribed by section 111 as “inducing” discrimination.)
“112. Aiding contraventions
(1) A person (A) must not knowingly help another (B) to do anything which contravenes Part 3, 4, 5, 6 or 7 or section 108 (1) or (2) or 111 (a basic contravention).
(2)-(6) …”
The effect of sections 111 and 112 is reasonably self-explanatory. I would only draw attention to the fact that A is only liable if B’s act is itself a contravention of one of the primary provisions of the Act. Claims under them must be brought in the forum appropriate to the “basic contravention”: see sections 114 (1) (e) and 120 (1) (b).
THE PRE-2010 LEGISLATION
For reasons which will appear, it is necessary to identify the effect of the predecessor provisions of sections 55-56 as regards the provision of vocational training. These were spread across a variety of different statutes and regulations, but we were referred to the Sex Discrimination Act 1975 by way of example. (Footnote: 1)
Section 14 of the 1975 Act, which fell under Part II (“Employment Field”), was headed “Persons concerned with provision of vocational training”. Sub-section (1) proscribed discrimination against a woman in relation to vocational training by “any person who provides, or makes arrangements for the provision of, facilities for” such training: I need not set out its full terms. However, sub-section (2) provided that sub-section (1) should not apply to (inter alia) “discrimination which is rendered unlawful by … section 22”.
Section 22 of the Act, which fell under Part III (“Discrimination in Other Fields”), proscribed discrimination by “bodies in charge of educational establishments”, including universities. As it stood immediately prior to the enactment of the 2010 Act, sub-section (1) read:
“It is unlawful … for [the “responsible body” of an educational establishment] to discriminate against a woman—
(a)-(b) … or
(c) where she is a pupil of the establishment—
(i) in the way it affords her access to any benefits, facilities or services, or by refusing or deliberately omitting to afford her access to them, or
(ii) by excluding her from the establishment or subjecting her to any other detriment.”
Discrimination in relation to the provision of work placements would plainly have fallen under head (c) (i).
Jurisdiction over claims under Part II of the 1975 Act was given to the Employment Tribunal and over claims under Part III to the County Court: see sections 63 and 66 respectively.
The point that needs to be taken from those provisions is that although the 1975 Act, like the 2010 Act, included an “educational exclusion” in the provisions relating to vocational training, it was differently expressed. The liability of the provider under section 14 was excluded only where and to the extent that the discrimination in question constituted discrimination which was rendered unlawful under the educational provisions: to spell it out, the exclusion depended on whether the acts complained of under section 14 (1) could have given rise to a valid claim under section 22. The criterion was not, as under the 2010 Act, expressed as being whether the university had “power to afford access” to the training in question. The most obvious example of a case where the same acts would be caught by both section 14 (1) (subject to sub-section (2)) and section 22 would be where the university arranged for the provision of vocational training, though there might also be cases where it provided training itself.
THE EU DIRECTIVES
The 2010 Act represents in the relevant respects the UK’s implementation of its obligations under EU anti-discrimination legislation, and I should set out the relevant provisions of that legislation. Discrimination at work is the subject of three EU Directives – 2000/43/EC, which covers racial discrimination (the Race Directive); 2000/78/EC, which covers discrimination on the grounds of religion or belief, disability, age or sexual orientation (generally referred to as the Framework Directive); and 2006/54/EC, which covers sex discrimination (the Equal Treatment Directive). The Equal Treatment Directive is a “recast” version of four earlier directives on sex equality, including 76/207/EC (also referred to as the Equal Treatment Directive).
I take the 2006 Directive first, since the appeal before us is concerned with sex discrimination. Article 1 is headed “Purpose” and reads (so far as material):
“The purpose of this Directive is to ensure the implementation of the principle of equal opportunities and equal treatment of men and women in matters of employment and occupation.
To that end, it contains provisions to implement the principle of equal treatment in relation to:
(a) access to employment, including promotion, and to vocational training;
(b) working conditions, including pay;
(c) occupational social security schemes.”
Chapter 3 is headed “Equal Treatment as regards access to Employment, Vocational Training and Promotion and Working Conditions”. Its principal operative provision is article 14.1, which reads:
“There shall be no direct or indirect discrimination on grounds of sex in the public or private sectors, including public bodies, in relation to:
(a) conditions for access to employment, to self-employment or to occupation, including selection criteria and recruitment conditions, whatever the branch of activity and at all levels of the professional hierarchy, including promotion;
(b) access to all types and to all levels of vocational guidance, vocational training, advanced vocational training and retraining, including practical work experience;
(c) employment and working conditions, including dismissals, as well as pay as provided for in Article 141 of the Treaty;
(d) membership of, and involvement in, an organisation of workers or employers, or any organisation whose members carry on a particular profession, including the benefits provided for by such organisations.”
“Discrimination” includes harassment and victimisation: see article 2.2.
The earlier version of the Equal Treatment Directive was differently structured but to substantially the same effect. Article 4 required that “the principle of equal treatment” (which proscribed both direct and indirect sex discrimination – see article 2) should be applied with regard to “access to all types … of vocational training”; and article 5 required that it be applied with regard to “working conditions”.
The Race Directive and the Framework Directive are differently structured from the Equal Treatment Directive, but all that it is necessary to note is that in both cases article 3.1, which defines the scope of the directive, uses identical language to that of 14.1 of the Equal Treatment Directive.
The scope of articles 4 and 5 of the Equal Treatment Directive in its earlier form – corresponding to heads (b) and (c) under article 14.1 in its current form – was considered by the EAT in Fletcher v Blackpool Fylde & Wyre Hospitals NHS Trust [2005] IRLR 689. The case concerned the entitlement of student midwives who were on training placements to financial support during pregnancy. The claimants were on academic courses at a higher education institute but for about half of their course they were placed for training with the local NHS trust, which was responsible for providing bursaries. It was argued for the trust that the effect of the dichotomy between articles 4 and 5 of the Directive was that the principle of equal treatment only applied to trainees as regards access to their training and not as regards their working conditions during such training. The ET had described it as a “big step” to apply the Directive to working conditions “in a non-employment context”. The EAT rejected that submission. Cox J said, at para. 58 (p. 705):
“Working conditions for vocational trainees like these applicants, regularly carrying out work on the hospital wards alongside qualified colleagues for half of their training, cannot legitimately be said to raise issues in a 'non-employment context'. As Mr Lynch [counsel for the trust] accepted, vocational training encompasses a wide range of activities and some trainees might in fact be workers. It is in our judgment inconceivable that the principle of equal treatment in the Directive was intended to apply to vocational trainees, in relation to access to their posts, but not to working conditions in their posts once they are occupying them. Mr Lynch was unable to identify any other EU legislation, which would presently protect them. In relation to pregnancy, if he were right, this would have the somewhat startling consequence that pregnant workers were intended to be protected throughout employment, including recruitment, working conditions and dismissal, whilst pregnant women undergoing vocational training were to be left without any protection once it had been ensured that they had equal access to that training. Ms Gill [counsel for the claimants] also provided the telling example of a woman who could complain of unequal treatment if a vocational course provider refused to allow her onto a course to become a plumber because of her gender. If Mr Lynch's submissions were correct such a woman who, once in post, was persistently given the worst jobs to carry out because she was a woman and the provider hoped she would leave, would have no complaint at all under the Directive. The 'big step' in our judgment would be to interpret the Directive in such a way as to restrict the protection of vocational trainees in the workplace in this way. We would expect clear words disclosing such an entirely different policy in relation to vocational trainees undergoing training in the workplace, if there was one.”
That decision is not binding on us, but Mr Milsom submitted that it was correct and that Cox J’s reasoning applied equally to the Equal Treatment Directive in its current form. Mr Pepperall did not seek to argue otherwise. I find Cox J’s reasoning persuasive and I am content to proceed on the basis that Fletcher was correctly decided.
THE PROCEEDINGS IN THE ET AND THE EAT
The Appellant started her claim in the Employment Tribunal on 20 February 2013. Initially she brought claims against both the University and the Trust. In part 5 of her claim form she alleged that the withdrawal of her placement because of her inability to work nights for childcare reasons constituted indirect sex discrimination on the part of the Trust. As for the University, it was alleged to be liable on the basis that it had knowingly aided the Trust and/or that the Trust had acted as its agent: she was thus invoking, although they were not expressly identified, sections 109 (2) and 112 of the Act.
In her claim form the Appellant acknowledged that there was a potential issue about the jurisdiction of the Tribunal, saying:
“The claim against the Trust has been brought in the Employment Tribunal on the basis that the Trust provides vocational training and guidance so discrimination allegations/breach of the Equality Act 2010 should be brought in the Employment Tribunal. If either Respondent believes that the case should be brought in the County Court, consideration can be given to the proceedings being stayed, pending a claim in the County Court.”
The Trust in its Response ignored that invitation and pleaded only in response to the substance of the allegation; but the University squarely denied that the Employment Tribunal had jurisdiction. It said:
“10. The Claimant has presented her claim against the First Respondent on the basis that it provides vocational training and guidance in accordance with s. 55 EqA 2010.
11. If the First Respondent accepts that it falls within the scope of s.55 EqA 2010 as an ‘employment service provider’, the Second Respondent relies on s.56 (5) EqA 2010, which applies for the purposes of s.55 EqA 2010, and which states ‘This section does not apply in relation to training or guidance for students of an institution to which section 91 applies in so far as it is training or guidance to which the governing body of the institution has power to afford access’.
12. Therefore, the Second Respondent contends that the Tribunal does not have jurisdiction to consider the Claimant’s claim.”
The Appellant subsequently withdrew the claim against the University.
The claim against the Trust came before an Employment Tribunal sitting at Birmingham, chaired by Employment Judge Kearsley, on 21 October 2013. The Appellant was represented by her solicitor and the Trust by Mr Jonathan Meichen of counsel. At the start of the hearing Mr Meichen submitted that by reason of section 56 (5) the Tribunal had no jurisdiction to entertain the claim. The Appellant’s solicitor was taken by surprise, given that the Trust had not taken the point either in its original Response or at an intervening directions hearing. He was given a short adjournment to consider his response. The Tribunal then proceeded to hear submissions on the jurisdiction issue, without any evidence being called.
The Tribunal’s decision was that the effect of section 56 (5) was that it had no jurisdiction, and it dismissed the claim accordingly. Written reasons were promulgated on 13 November 2013. Its reasoning was commendably succinct. It held that a university was to be treated as having the “power” to afford access to training if in practice it had the ability to do so. In the present case the University plainly had the ability to afford the Appellant access to training with the Trust because it had in fact done so. Accordingly section 55 did not apply in relation to the training so provided. I should add that it expressed its disapproval of the Trust having raised the point so late in the day: I would endorse that criticism, but since the point went to jurisdiction it had to be decided.
The Appellant appealed to the Employment Appeal Tribunal. The appeal was heard by HH Judge Eady QC, sitting alone, on 22 September 2014. The Appellant was represented by Mr Jim Tindal of counsel and the Trust again by Mr Meichen.
Judge Eady delivered judgment on the same day dismissing the appeal. The judgment is now reported at [2015] ICR 308. Since the issues with which we are concerned are issues of pure law which we will have to decide for ourselves in any event I need give only a summary of her reasons, as follows:
(1) She held, like the ET, that the University plainly “had power to afford access” to the training to which the Appellant’s complaint related: see para. 58. Mr Tindal had submitted that that phrase should be construed narrowly, so as to apply only to cases where a university had “the ability to act without the need to obtain the consent of someone else”. She rejected that submission on the basis that “vocational training” was defined so as to include work experience and that in the real world such experience could only be provided with the consent and co-operation of the provider, so that if Mr Tindal’s submission were right sub-section (5) could never bite: see paras. 57-58.
(2) She acknowledged that the phrase “afford access” could reasonably be construed as referring only to the point at which a work placement is arranged – what she called “the point of entry” – or to affording “continuing access”. She inclined to the former interpretation (see para. 56) but she did not believe that the point was decisive because the real question was whether the university had the “power” to afford access (para. 57).
(3) She held that it followed that section 56 (5) applied so as to exclude the operation of section 55. She acknowledged that it was necessary, so far as possible, to construe the Act conformably with EU law, and she seems implicitly to have accepted, having been referred to Fletcher, that the Directives required that students on work placements should be protected against discrimination (see paras. 51-52). But she believed that Part 6 of the 2010 Act would give them substantially the same protection as they would have had under Part 5 (paras. 52 and 59). She accepted at para. 52 Mr Meichen’s case (most fully stated at para. 43) that
“… it was open to [the Appellant] to argue … before the County Court that [the University] was acting as principal and [the Trust] as agent, and thus liability would arise under sections 109-110 of the Equality Act. Or she could seek to fix [the Trust] with potential liability under section 112.”
She rejected a submission by Mr Tindal – not advanced before us – that a right to sue in the County Court constituted “a second-class form of protection” (paras. 54 and 59).
(4) She regarded that conclusion as reflecting Parliament’s intention. As she put it at para. 50 of her judgment:
“Looking at section 56(5), Parliament’s intention is plain: if a university has power in respect of vocational training and guidance undertaken by its students, then that training and guidance should be seen as part of that higher education, and complaints of discrimination should be pursued under the education provisions. In England and Wales that means claims should be brought in the county court, not the Employment Tribunal.”
THE APPEAL
THE APPELLANT’S CASE
The essential issue before us is whether, as both tribunals below have held, section 55 is disapplied by section 56 (5). The fundamental point underlying Mr Milsom’s submissions was that if section 56 (5) had that effect it would mean that students on work placements arranged by a university who were subjected to discrimination by the person with whom they were placed (“the provider”) would have no effective remedy. They would have no claim under section 55 because that would be excluded. But nor would they have any claim under section 91. The provider would not be an institution of the kind to which the section applies, as defined in sub-section (10) (see para. 13 above), so it would not be “primarily” liable. The student could thus only have a remedy by reference to the “indirect liability” provisions. As to that, two routes had been canvassed in the EAT and apparently accepted by Judge Eady. Taking them in turn:
(1) Section 109: agency. If the provider was an agent of the university and the act complained of had been done with the university’s authority, the university would be treated by section 109 (2) as having done that act and would accordingly be liable under section 91; and the provider would also be liable under section 110 (1). But, Mr Milsom submitted, the relationship between a university and a provider of work placements would not typically be a relationship of principal and agent, nor would an act of discrimination by the provider be done with the authority of the university. There might be exceptional cases where, on the particular facts, such authority might be demonstrated, but in the generality of cases there would be no route to a remedy under sections 109-110.
(2) Sections 111 and 112. Even if on the facts of a particular case the university might be said to have induced, or knowingly assisted, the doing by the provider of the act complained of (as indeed was alleged in the claim form in the present case) that goes nowhere for the reason noted at para. 19 above: an inducer/helper cannot be liable under these sections unless the person doing the act complained of is itself liable, and the very problem which these provisions are being invoked to get round is that the Trust is not so liable.
Mr Milsom submitted that it cannot have been the intention of Parliament that students should have no protection against discrimination by work placement providers save where, exceptionally, the provider was acting as the agent of the university.
Mr Milsom reinforced that submission by reference to the legislative history. Under the 1975 Act there would have been no gap such as is produced by the EAT’s decision: the (primary) liability of the provider would only be excluded if the university was liable – see para. 24 above. He submitted that it was inconceivable that Parliament had intended to remove protection against discrimination from students undertaking work experience. He referred to my observation in Rowstock Ltd v Jessemey [2014] EWCA Civ 185, [2014] 1 WLR 3615, that “[a]lthough the [2010] Act is not formally a consolidating statute, its purpose was to re-state, with some clarifications and enhancements where necessary, existing protections against discrimination” (see para. 31). He told us (and Mr Pepperall did not submit otherwise) that there was no indication either in the Explanatory Notes or in Hansard of any intention on the part of Parliament to reduce the protection available to students on work placements.
Mr Milsom further submitted that if the 2010 Act failed to proscribe discrimination against students by the providers of work placements the UK would be in breach of the underlying EU Directives. It was clear from Fletcher that the protection of the Directives extended to the treatment of vocational trainees in the course of their placements: see para. 29 above.
Against that background Mr Milsom submitted that it was necessary to adopt a construction of section 56 (5) that did not deprive students who had suffered discrimination by a provider of employment services in the course of a work placement of the right to bring a claim for discrimination against the provider. He submitted that this could be achieved by the application of ordinary domestic principles of construction, adopting a purposive approach; but if necessary he relied on the special principles governing the construction of statutes which are intended to implement obligations under EU law (the so-called Marleasing principle – see the decision of the ECJ in Marleasing SA v La Comercial Internacional de Alimentación SA (Case C-106/89) [1990] ECR I-4135). He suggested various ways in which section 56 (5) could be construed so as to avoid the gap in protection which he had identified and achieve conformity with the requirements of the Directive. I will return to these later.
DISCUSSION AND CONCLUSION
The starting-point in considering those submissions is to decide whether the EAT’s construction of section 56 (5) would indeed leave a lacuna in protection such as Mr Milsom alleges. In his skeleton argument Mr Pepperall denied that that was so. He submitted that the sub-section gave effect to a logical demarcation between the operation of Parts 5 and 6 of the Act. He submitted that students would have a claim against their university under section 91 (2) (b) or (d) if it discriminated against them by not affording them access to work placements, or in the way in which it did so, since a work placement was plainly a facility or a benefit. He acknowledged that they would not have a primary remedy against the work placement provider. But he contended that in that regard they were in no different position from employees, who likewise have a claim against their employers for discrimination as regards access to training (see section 39 (2) (b)) but who, he submitted, cannot proceed against the training provider; for the latter point he relied on section 56 (3). He also acknowledged that the availability of an indirect liability might be limited, but he said that the position was no different than for other persons protected by the Act.
Those submissions do not meet Mr Milsom’s point. It is no doubt correct that students have a remedy against their university under section 91 if it discriminates against them by not affording them access to work placements, or in the way in which it does so: see sub-section (2). But it seems to me clear that the concept of “affording access” is concerned only with what the university itself does or does not do in enabling the student to enjoy the benefit in question – that is, in the present context, in making and, so far as necessary, maintaining the arrangements under which the student receives vocational training from the third party. The phrase does not in its natural meaning extend to responsibility for acts done by the work placement provider in the course of the placement. To take Cox J’s example of the female trainee plumber on a work placement who is consistently assigned the least attractive tasks, that has nothing to do with the way that her college affords her access to training; likewise where a student nurse is subjected to sexual or racial harassment by a manager. Those are the acts of the work provider, and there is no basis – subject to any indirect liability that might be available on the facts of a particular case – on which they can be caught by the terms of section 91. The relevant provision is plainly that which proscribes discrimination by providers, i.e. section 55; but if section 56 (5) has the meaning found by the EAT a student can have no remedy under that section. If Mr Pepperall is right that the effect of section 56 (3) is that there is a similar gap in the case of employees, that does not make the problem go away; it only means that it is more extensive.
That conclusion does not involve any acceptance that the concept of “affording access” is limited to what Judge Eady calls “the point of entry” (see para. 37 (2) above). On the contrary, and differing respectfully from her tentative view, I think it more natural to regard a university as affording access to a work placement for as long as the student enjoys that placement: the arrangement with the provider is of its nature a continuing one. But it does not follow from that that everything that happens in the course of the placement is part of what the university is doing by way of affording access.
Mr Milsom accordingly gets to first base: the construction of section 56 (5) adopted by the EAT does indeed create a lacuna in protection. The question then is whether it is possible to construe the Act so as to avoid that lacuna. As I have said, Mr Milsom’s primary case was that it was possible by applying ordinary domestic principles of purposive construction to give section 56 (5) a restrictive construction such that students were not prevented from proceeding against the work provider under section 55. He suggested two ways in which that might be achieved. I take them in turn.
First, he repeated Mr Tindal’s submission that the phrase “power to afford access” could be construed as referring only to cases where the university had a unilateral right to require the provider to provide the work placement. This would mean that section 56 (5) had no application where, as here, access to work placements was a matter for agreement between the university and the provider. I agree with Judge Eady (see para. 37 (1) above) that this construction would deprive the sub-section of any “real world” meaning and that it is accordingly hard to accept on the basis of an ordinary domestic approach to construction.
Secondly, he submitted that the concept of power to afford “access” to training should be treated as continuing throughout the duration of the placement. As appears from para. 44 above, I would agree with that; but I do not see how it assists him. Whether a university’s role in affording access to a work placement is to be regarded as limited to the point of entry or as continuing, the problem is that (subject to his first point) it has the “power” to afford access, and I agree with Judge Eady that it is that element in the statutory language which creates the stumbling-block for his submission.
I do not, therefore, think that the Appellant can get home by applying ordinary principles of construction. I turn to Mr Milsom’s alternative case based on the Marleasing principle. The limits of that principle, and the cognate approach under section 3 of the Human Rights Act 1998, have been the subject of a good deal of exposition in the case-law, most authoritatively in the decisions of the House of Lords in Pickstone v Freemans plc [1989] AC 66, Litster v Forth Dry Dock and Engineering Co Ltd [1990] 1 AC 546 (Footnote: 2) and Ghaidan v Godin-Mendoza [2004] UKHL 30, [2004] 2 AC 557. For working purposes, it is sufficient to adopt the summary in the judgment of Sir Andrew Morritt C in Vodafone 2 v Her Majesty's Commissioners of Revenue and Customs [2009] EWCA Civ 446, [2010] Ch 77, at para. 37 (p. 90):
"In summary, the obligation on the English courts to construe domestic legislation consistently with Community law obligations is both broad and far-reaching. In particular:
(a) It is not constrained by conventional rules of construction (per Lord Oliver in Pickstone at 126B).
(b) It does not require ambiguity in the legislative language (per Lord Oliver in Pickstone at 126B; Lord Nicholls in Ghaidan at 32).
(c) It is not an exercise in semantics or linguistics (see Ghaidan per Lord Nicholls at 31 and 35; Lord Steyn at 48-49; Lord Rodger at 110-115).
(d) It permits departure from the strict and literal application of the words which the legislature has elected to use (per Lord Oliver in Litster at 577A; Lord Nicholls in Ghaidan at 31).
(e) It permits the implication of words necessary to comply with Community law obligations (per Lord Templeman in Pickstone at 120H-121A; Lord Oliver in Litster at 577A).
(f) The precise form of the words to be implied does not matter (per Lord Keith in Pickstone at 112D; Lord Rodger in Ghaidan at para 122; Arden LJ in [R (IDT Card Services Ireland Ltd) v Customs and Excise Commissioners [2006] STC 1252] at 114)”.
Sir Andrew continued, at para. 38 (pp. 90-91):
“The only constraints on the broad and far-reaching nature of the interpretative obligation are that:
(a) The meaning should ‘go with the grain of the legislation’ and be ‘compatible with the underlying thrust of the legislation being construed’ (per Lord Nicholls in Ghaidan at 33; Dyson LJ in [Her Majesty's Commissioners of Revenue and Customs v EB Central Services Ltd [2008] EWCA Civ 486] at 81). An interpretation should not be adopted which is inconsistent with a fundamental or cardinal feature of the legislation since this would cross the boundary between interpretation and amendment; (see Ghaidan per Lord Nicholls at 33; Lord Rodger at 110-113; Arden LJ in IDT Card Services at 82 and 113) and
(b) The exercise of the interpretative obligation cannot require the courts to make decisions for which they are not equipped or give rise to important practical repercussions which the court is not equipped to evaluate. (See Ghaidan per Lord Nicholls at 33; Lord Rodger at 115; Arden LJ in IDT Card Services at 113.)”
While reminding us of Sir Andrew Morritt’s “point (f)”, Mr Milsom suggested two alternative ways in which section 56 (5) could be re-drafted so as to preserve the right of a student on a work placement to claim against the provider under section 55. These were as follows:
(1) The phrase “of an institution” should be read as if it said “by an institution”. Thus the sub-section as a whole would read:
“This section does not apply in relation to training or guidance for students by an institution to which section 91 applies in so far as it is training or guidance to which the governing body of the institution has power to afford access.”
(2) The opening words of the sub-section should be altered so that it read:
“This section does not apply to claims in relation to training or guidance for students of an institution to which section 91 applies in so far as it is training or guidance to which the governing body of the institution has power to afford access.”
I am not comfortable with either of those alternatives. The first achieves the desired result, but it does not do so very explicitly, and the second half of the sub-section becomes redundant if not positively confusing. As for the second, I am not sure that I understand how it works. A more radical re-wording seems to me to be preferable in order to achieve the necessary effect, as follows:
“This section does not apply to discrimination in relation to training or guidance for students of an institution to which section 91 applies to the extent that the student is entitled under that section to make a claim as regards that discrimination.”
Such a formulation would not render the sub-section meaningless, though its scope would be reduced as compared with its natural meaning. There is an area of overlap between section 55 and section 91. In arranging for work placements a university is not only affording access to a benefit within the meaning of section 91 (2) (b) but also “making arrangements for the provision of vocational training” within the meaning of section 56 (2) (c); and it may also in some cases itself be a provider of vocational training or guidance within the meaning of section 56 (2) (a) and (b). Accordingly claims for discrimination in the affording of access would be capable, absent provision to the contrary, of being brought by reference to either section. On my proposed construction section 56 (5) would prevent that occurring. This of course reproduces the position as it was under the predecessor legislation: see para. 24 above.
Is such a construction possible on a Marleasing approach ? In so far as we are concerned simply with the substantive question of what protections against discrimination should be afforded to students on work placements, it is in my view plainly not a “fundamental or cardinal feature” of the 2010 Act that their only protection should be against discrimination by the university itself in how it affords access to such placements (with any associated indirect liability), with no protection against discrimination by the provider. I can see no indication that Parliament made a deliberate choice to that effect. On the contrary, it is very unlikely that it did so. The position under the previous law was well-established. The EAT had held in Fletcher that the Directives required that students on work placements be protected against discrimination by the provider; and the predecessor legislation achieved that result (see para. 24 above). It would be remarkable if Parliament intended to remove that protection. As Mr Milsom submitted (see para. 39), it was not the purpose of the 2010 Act to reduce the scope of protection against discrimination.
In so far as there is a difficulty with Mr Milsom’s case, it arises out of the fact that on his approach the student’s remedy falls under Part 5 of the Act (so that the correct forum is the Employment Tribunal). Mr Pepperall argued that the reason for the change in the language as between the old legislation and the 2010 Act must have been to adjust the demarcation line between Part 5 and Part 6 so that all claims in relation to student work placements fell under Part 6 and could be litigated in the same forum. Cases might arise in which a student wished to complain both about discrimination by the university in the way that she was afforded access to such placements and about discrimination by the work provider in the course of the placement – indeed the two complaints might be factually closely related; it would be inconvenient, to put it no higher, if the two claims had to be brought in different forums. Bringing all student work placement claims under Part 6 would avoid that problem. He argued that the allocation of the forum for student work placement claims represents a deliberate legislative choice, and that disregarding it would cross the boundary between interpretation and amendment.
I am not in fact convinced that the draftsman’s intention in choosing the wording of section 56 (5) was make a change of this kind. In the first place, section 56 (5) must be read with section 56 (3). As I have pointed out at para. 10 above, the two sub-sections appear to form part of the same group of provisions, yet in the case of sub-section (3) there is no question of allocation between forums. Secondly, the language of section 56 (5) suggests that the draftsman had specifically in mind the terms in which discrimination is proscribed in section 91 (2) (b) and (d). The language of “affording access” is common to both, and the use of the phrase “training … to which [the university] has power to afford access” – rather than simply “training … to which [it] affords access” – appears to be chosen so as to include the case where the university fails to provide access, which is the case covered under head (d). That raises at least the possibility that the scope of section 56 (5) was intended to track the scope of section 91 (2), i.e. to exclude from section 55 only claims which could have been brought under section 91. I appreciate that this is not the result which the full wording of the sub-section produces: that is why we are here. But it is suggestive of how an error in the drafting might have arisen. (The drafting of the sub-section is of course unsatisfactory in at least one other respect: see para. 9 above.)
However that is debatable ground. I am prepared to confront the possibility that there was a conscious intention on the part of the draftsman to move the demarcation line between work and education claims in the case of student work placements, albeit failing to appreciate that section 91 did not give any protection against discrimination by the provider. If that is indeed the case, I would nevertheless regard it as possible, on a Marleasing approach, to construe section 56 (5) so as to afford students a remedy under Part 5. I do not believe that it can be regarded as a fundamental feature of the legislation that claims should be allocated to one forum rather than another, which is an essentially procedural matter, at the cost of substantive protection against discrimination: on the contrary, such a result would go clean against the grain of the Act. If the choice is between students having no remedy at all against discrimination by the placement provider and a restriction of the effect of section 56 (5) so as to give them such a remedy, albeit in a different forum than the draftsman intended, I believe that it is legitimate to choose the latter.
That is not, however, the end of the matter. In his oral submissions Mr Pepperall argued that there was no need for that stark choice. Even if there was a lacuna in the protection for students on a work placement who were discriminated against by the provider, it was possible to fill it by means of a claim under section 91. That could be done by taking a “benign” approach to the operation of sections 109 and 110. He submitted that it was possible to treat the relationship of a university and a work placement provider in every case as one of principal and agent, and to regard every act of the provider in the course of that relationship as being done with the authority of the university. That would mean that where the provider discriminated against a student its acts would be treated as having been done by the university, which would accordingly be primarily liable under section 91 (2); and the provider would also be liable under section 110 (1). Mr Pepperall accepted that this approach involved some “stretching” of the concept of agency, but he submitted that that was legitimate in a Marleasing context. He submitted that the route via section 91 was preferable to Mr Milsom’s proposed solution based on the emasculation of section 56 (5), because it gave effect to Parliament’s evident intention that cases involving student work placements should fall under Part 6 rather than Part 5.
Well though Mr Pepperall developed the point, I am not persuaded that that approach represents a possible construction of the statute. It is settled law that references to agency concepts in this legislation are to be given their ordinary common law meaning: see Kemeh v Ministry of Defence [2014] EWCA Civ 91, [2014] ICR 625 (in particular para. 70 of the judgment of Lewison LJ (p. 643H) (Footnote: 3)). There was no evidence either below or before us as to the details of the arrangements under which work placements were arranged in the present case: it seems from the authorities to which we were referred (Footnote: 4) that such arrangements between universities and NHS bodies can take many forms. But for the purpose of considering Mr Pepperall’s submission what matters is the general position. Although I do not rule out the possibility that there may be some kinds of arrangement between a university and a provider of work placements under which the provider acts as the agent of the university, it is inconceivable that this will be universally the case, still less that every act of the provider can be done with the authority of the university. On the contrary, in the case of a typical student work placement, say with a school or hospital, while the university will no doubt have agreed certain parameters with the provider about what work will be provided, it will in the nature of things have no involvement in, or control over, the student’s day-to-day working environment. It is hard to see any basis for characterising such a relationship as one of principal and agent as opposed to an arm’s length relationship between two principals, and harder still as one where every act of the provider is to be treated as done with the authority of the university. To insist on such a characterisation in every case, as a means of achieving compliance with the requirements of the EU Directives, would be unacceptably artificial. It would also produce unfair results, since it would mean that the university was treated as automatically liable for acts on the part of the provider for which it could not fairly be regarded as having any responsibility. (Footnote: 5) In my view the scheme under sections 109 and 110 of the 2010 Act depends on the concept of agency being applied in accordance with the ordinary understanding, and a departure from that approach would be inconsistent with a fundamental feature of the legislation.
We were referred in this connection to two decisions of the EAT, which I will take in turn.
In Lana v Positive Action Training in Housing (London) Ltd [2001] UKEAT 245/00/1503, [2001] IRLR 501, the respondent, which appears to have been a company whose business was to arrange training for quantity surveyors, placed the appellant for training/work experience with a company called Walker Management, who terminated the placement when she became pregnant. The EAT (Mr Recorder Langstaff QC presiding) held that the relationship between the respondent and Walker Management as shown by the written agreement into which they entered was one of principal and agent (see paras. 5-7) and that the termination of the placement was an act done with the respondent’s authority. Accordingly, it was liable under section 14 (1) of the 1975 Act for Walker Management’s discrimination against the appellant. I can see nothing in this decision that bears on the question before us. The fact that the relationship between the training company and the training provider in that case was one of principal and agent does not mean that such a relationship can be found in every case where a work placement is arranged for a student by a university.
In Moyhing v Homerton University Hospitals NHS Trust UKEAT/0851/04 the appellant was studying for a degree in nursing at the City University in London, which arranged a work placement with the respondent trust. He brought proceedings in the employment tribunal alleging that he had been unlawfully discriminated against in the course of that placement by not being allowed to perform certain procedures on female patients. The ET declined jurisdiction on the basis that the trust had been acting as the university’s agent and that the effect of section 14 (2) of the 1975 Act in such a case was that the claimant could only claim under section 22 and thus in the County Court. The EAT (Burton P presiding) held that that was not the effect of section 14 (2), and the question whether the trust was the university’s agent accordingly did not arise. That question was, however, discussed at some length on an obiter basis: see paras. 43-50 of the judgment. Burton P was prepared to contemplate that on the facts of the particular case there might be an agency relationship, but he expressed sympathy with the appellant’s submission that the better analysis was that the arrangement for work placements between the university and the trust was an arm’s-length relationship between two principals – what he described as a “joint venture” or “partnership” (see para. 50). That is consistent with the position as I have analysed it above.
It may be helpful after so lengthy a discussion if I summarise what I believe to be the effect of sections 55 and 56, construed so as to give effect to the relevant Directives. The starting-point in any case is to identify the nature of the student’s complaint – that is, whether it is about discriminatory access to a work placement or about discrimination occurring during the placement.
(1) If the claim is about access – either that the university has failed to provide a placement at all or that it has done so in a discriminatory way – it can only be brought under section 91, and thus in the County Court. The primary claim will inevitably be against the university, because it is the university that has the responsibility for the provision of access, and it is hard therefore to see any role for sections 109 and 110; but if the provider has induced or aided that contravention it will be secondarily liable under section 111 or 112 and the student can proceed against it (in the County Court) as well as, or instead of, the university.
(2) If the claim is about discrimination by the provider in the course of the work placement, the provider will typically have done the act complained of as a principal and will thus be primarily liable for that discrimination under section 55, with the forum for any proceedings being the Employment Tribunal. There may be untypical cases where the act was done by the provider as the agent of the university. In those cases both the university and the provider will be liable, by virtue of sections 109 (2) and 110 (1) respectively, but the liability will still arise under section 55, so that the ET will still be the correct forum whether the claimant chooses to proceed against only one of them or against both. The university may of course also in a particular case be liable, depending on the facts, under sections 111 or 112 as having induced or assisted the discrimination. Any such claim will, again, have to be brought in the ET: see sections 114 (1) (e) and 120 (1) (b).
I recognise that on this analysis there may occasionally be cases in which a student wishes to complain both about discrimination in his or her access to a work placement and about discrimination in the course of that placement and that the need to pursue those claims in different forums will involve expense and inconvenience. That is regrettable, but it could only be avoided by denying students the right to bring claims arising out of work placement discrimination at all or by re-writing the Act so as to allow such claims to be brought under Part 6. For the reasons given above neither course is in my view acceptable.
I would accordingly conclude that section 56 (5) did not in this case operate to deprive the ET of jurisdiction to determine the Appellant’s complaint about the withdrawal of her placement by the Trust.
I should add for completeness that there was a further ground of appeal before us, namely that even if the ET’s construction of section 56 (5) was correct it should not have reached any conclusion about whether the University had power to afford the Appellant access to her placement with the Trust without hearing evidence about the arrangements in place between them. That ground does not now arise.
DISPOSAL
65. I would allow the appeal against the decision of the Employment Appeal Tribunal and remit the case to the Employment Tribunal for a determination of the merits of the Appellant’s claim.
Lord Justice Lewison:
66. 1 agree.
Lord Justice Patten:
67. I also agree.