Pregnancy discrimination and zero-hours contracts

The Equality Act 2010 protects women from discrimination “because of the pregnancy”, or because of illness suffered by her as a result of the pregnancy. In employment cases, the relevant section is Section 18. The core protection is against being treated “unfavourably”. It needs to be read in conjunction with Section 39.

Discrimination because of pregnancy is, by all reports, widespread. A recent Equality and Human Rights Commission stated that around one in nine mothers felt that they were either dismissed, made compulsorily redundant (where others in their workplace were not), or treated so poorly that they had to leave their job. Scaled up, that would amount to around 54,000 mothers per year. The link to the reports (both the full report and the summary of key findings) is here.

The Equality Act 2010 refers to “employees”, but “employment” in the Equality Act 2010 has a much wider meaning than “employment” for the purposes of unfair dismissal claims. The key section of the Equality Act 2010 is Section 83. We know from case law (Pimlico Plumbers and anor v Smith [2018] UKSC 29 and Clyde & Co LLP and anor v Bates van Winkelhof [2014] UKSC 32) that “employee” in the Equality Act covers “workers” within the meaning of s 230(3) Employment Rights Act 1996.

That has important consequences for “zero-hours” workers. A “zero-hours” contract is one where the worker does not have an obligation to accept work, and the employer does not have an obligation to provide work. As a result, the worker will generally not be an employee (in the unfair dismissal sense), but is still protected by discrimination law (although Tribunals have sometimes found that the worker is an employee when they are actually at work, although not a worker in between periods of work).

So a worker on a zero-hours contract still has rights, and those rights include the right not to be discriminated against because of pregnancy. That is an important point, because zero-hours contracts may be particularly attractive to expectant mothers who want flexible working arrangements. Employers may be concerned that expectant mothers will require restricted duties, and this will place a financial burden on the business. Can the employer then rely on the fact that the contract is a “zero-hours” contract, and simply choose not to offer work?

The short answer is “no”. Under s 18 of the 2010 Act, the woman is protected against unfavourable treatment. Unfavourable treatment is essentially the same as “detriment”, and the threshold is “relatively low” (Trustees of Swansea University Pension and Assurance Scheme and anor v Williams [2018] UKSC 65, para 27). The test is whether the treatment is “of such a kind that a reasonable worker would or might take the view that in all the circumstances it was to his [or her] detriment” (Shamoon v Chief Constable of the Royal Ulster Constabulary [2003] UKHL 11). If there is a legitimate expectation of work, then it is very likely that a failure to offer work would constitute unfavourable treatment.  That is the first ingredient of a claim under Section 18 of the 2010 Act.

Furthermore, the test for whether the treatment is “because of” pregnancy is relatively easy to meet. It is sufficient that the protected characteristic (here, pregnancy) has a “significant influence” on the reason why: Nagarajan v London Regional Transport [1999] IRLR 572. Moreover, the treatment will still be “because of pregnancy” if the reason is the economic consequences for the employer of employing a pregnant woman (Dekker v VJV Centrum C-177/88) or the health and safety implications of pregnancy, including a prohibition on performing certain work (Bush v Klinikum Neustadt Gmbh and Co Betriebs-KG C-320/01).

Furthermore, unfavourable treatment of a woman because of pregnancy is direct discrimination (in European terms), and so cannot be justified: see Webb v EMO Air Cargo (UK) Ltd C-32/93.

It follows that economic loss or disruption to work caused by pregnancy or absence because of pregnancy-related illness or maternity leave cannot negate pregnancy or maternity discrimination (see, for example, Tele Danmark A/S v Handels- og Kontorfunktionaerernes Forbund i Danmark (C109/00) [2001] E.C.R. I-6993 at para 23; “…the employer has to assume the risk of the economic and organisational consequences of the pregnancy of employees”).

So, if a worker on a zero-hours contract becomes pregnant, and as a result would need to go onto restricted duties (suppose, for instance, that the work has a substantial manual handling component), the employer cannot simply decide not to offer work. To do so would (very likely) constitute unlawful discrimination, and the worker could (in principle) claim for injury to feelings as well as for lost earnings.

If the employer’s concerns are health and safety concerns, then there is a potential statutory defence in Schedule 22 to the 2010 Act. It is not a particularly transparent piece of legislation. But the Explanatory Notes to Schedule 22 make it clear that Schedule 22 permits differential treatment based on sex or pregnancy and maternity at work which is required to comply with laws protecting women who are pregnant, who have recently given birth, or against risks specific to women.

One potentially relevant piece of legislation is the Management of Health and Safety at Work Regulations 1999, which require employers to carry out risk assessments (Regulation 3) and to carry out specific assessments where the workers include women of child-bearing age, if the work might be risky to the health and safety of a new or expectant mother (Regulation 16). If the risk assessment highlights a risk, then the employer is (as always) required to take steps to avoid that risk.

If those steps won’t avoid the risk, then the employer shall (if it is reasonable to do so, and would avoid the risk) alter the woman’s working conditions or hours of work. And it isn’t reasonable to do that, or if altering working conditions or hours won’t remove the risk, then the employer shall suspend the employee from work – subject to Section 67 Employment Rights Act 1996, which gives the employee the right to be offered to be provided with suitable alternative work (if it is available). Refusing suitable alternative work may mean that the employer can refuse to provide paid suspension.

Where employees (in the Employment Rights Act 1996 sense) are involved, it is clear that the suspension must be on full pay if the suspension is on maternity grounds (Section 68 of the 1996 Act). The 1996 Act is also clear that the employee is to be regarded as suspended from work on maternity grounds only if and so long as she continues to be employed, but is not provided with work or does not perform the work she normally performed before the suspension (Section 66 of the 1996 Act). The Explanatory Notes (see above), I think, make the answer clear – giving the example of a care home which cannot lawfully dismiss, but can lawfully suspend, a night-shift worker because she is pregnant.  Regulation 16 MHSWR 1999 implements the Pregnant Workers Directive (Directive 92/85), which addresses “workers”. A failure to carry out a risk assessment will (if the employer is notified of the pregnancy) amount to a detriment: Hardman v Mallon t/a Orchard Lodge Nursing Home [2002] CMLR 59. So the protection offered by Regulation 16 must cover “workers” (in the extended sense) as well as “employees” (in the narrow sense). If suspension is the only option, it is hard to see how a failure to offer paid suspension, following a Regulation 16 risk assessment, could amount to anything other than unfavourable treatment within the meaning of Section 18 Equality Act 2010.

There is also an “occupational requirement” exception in Schedule 9 to the 2010 Act (which is specifically provided for by Article 14(2) of the Recast Equal Treatment Directive). The Explanatory Notes to Schedule 9 give the example of a modelling job which requires someone of a particular race, age or sex. But that “occupational requirement” must be crucial to the post, and not merely an important factor; and the application of that requirement must be proportionate so as to achieve a legitimate aim.

To recap: a pregnant worker on a zero-hours contract is still entitled to ask for work. If there is a legitimate expectation of work – and this does not require a contractual entitlement to work – then it will very likely be unfavourable treatment not to offer the employee work. If a reasonable worker might take the view that it is a detriment, then the threshold for “unfavourable treatment” is met.

If the consequence is that the employer is put to additional cost, then so be it. The law is clear that the employer, not the pregnant worker, is to bear the economic consequences of pregnancy.



Can the Employment Tribunal limit the number of discrimination claims to be determined at any one hearing?

Can the Employment Tribunal limit the number of discrimination claims to be determined at any one hearing?

“Only with great caution”, said the Employment Appeal Tribunal in Tarn v Hughes (and ors) UKEAT/0064/18/DM – and the EAT then went on to give some helpful guidance on case management of discrimination claims.

In Tarn, a claimant GP had brought a large number of complaints of discrimination against her (former) partner GPs.

At the case management stage, the ET had required the claimant to select:

the most recent and serious 10 (maximum) events relied upon as giving rise to the above complaints and on which the Tribunal is required to make findings of fact and determinations

The order went on to state that the claimant was not precluded from relying on the other allegations as background and/or context, or pursuing them at a later hearing.

The underlying claim contained some 31 factual allegations. That, commented the Tribunal, would require it to make 180 separate findings (i.e. for each act, whether it happened, whether it was less favourable treatment, and whether it was because of the protected characteristic).

Although the appeal was pursued on a number of grounds, the crux of the matter is set out at para 25 of Tarn. In particular, HHJ Eady QC commented that

. . . of its nature, a discrimination claim is likely to require an ET to draw inferences from the evidence and from its primary findings of fact; to adopt a fragmented approach to the issues may “have the effect of diminishing any eloquence that the cumulative effects might have [on the determination of causation” (see per Mummery J (as he then was) in Qureshi at page 875H)

The judgment goes on to discuss guidance as to the possibility of selecting sample allegations in discrimination cases drawn from Hendricks and HSBC Asia Holdings BV and another v Gillespie UKEAT/0417/10, before giving a summary at para 28. I won’t reprise the paragraph in its entirety but the essence is this:

  1. The ET has a broad discretion to manage cases justly.
  2. In discrimination claims, the starting point is to identify precisely what claims are made and on what basis (and a list of issues will help).
  3. There will be some cases which are apt for striking out. However, it is not otherwise open to the ET to limit the claims a complainant can pursue – that would be to restrict her access to justice and to potentially deny an effective remedy in a case of unlawful discrimination.
  4. The ET could, in an appropriate case, separate out a sample of complaints or issues to be heard in advance of the remaining allegations.
  5. But that does not mean that such a course should be adopted, “save in those cases where it is clear this would not endanger the just determination of the case – something that might be difficult for the ET to assess at a preliminary stage”.
  6. It is often necessary, in discrimination claims, to look at the whole picture before looking at whether there has been unlawful discrimination in respect of a particular allegation. Asking a claimant to select his or her ten best points may often be inconsistent with the just determination of the claim(s).
  7. Moreover, separate determination of selected allegations or issues may not be the proportionate course in a particular case.

As to the appeal in Tarn itself, the EAT thought that the ET had failed to have regard to critical factors – for instance, given the list of issues, it was clear that much of the evidence given on one issue would inevitably address other factual issues at the same time. The Claimant had been placed in the (potentially) unfair position of having to choose between either continuing to assert her right to pursue all of her claims, or simply relying on them as context for the others. The appeal was therefore allowed.

I think this case is interesting not because of the particular disposal, but because of the general treatment of case management and discrimination claims. That inevitably has an impact on how practitioners should present discrimination claims. There is, very often, a tension between choosing the strongest allegations, thereby providing focus, and choosing a large number of allegations, thereby providing context – assisting the Tribunal to draw crucial inferences from the evidence. I suspect it is now less likely that breadth will be “managed out” at a preliminary hearing.

Unfair dismissal: “expired” warnings and final written warnings

When is it fair to rely on an “expired” warning when dismissing an employee? When is it fair to rely on a final written warning?

The short answer (rather unhelpfully) is “it all depends”. The starting point is the statute:

Section 98(4) of the Employment Rights Act 1996

(4) Where the employer has fulfilled the requirements of subsection (1), the determination of the question whether the dismissal is fair or unfair (having regard to the reason shown by the employer) –

(a) depends on whether in the circumstances (including the size and administrative resources of the employer’s undertaking) the employer acted reasonably or unreasonably in treating it as a sufficient reason for dismissing the employee, and

(b) shall be determined in accordance with equity and the substantial merits of the case.”

Nothing new there. But what about a situation where the final written warning relates to conduct which is unrelated (or dissimilar) to the dismissal conduct (i.e. the conduct which resulted in the dismissal)?

There is helpful guidance given in Davies v Sandwell Metropolitan Borough Council [2013] EWCA Civ 135 at para 22:

22 First, the guiding principle in determining whether a dismissal is fair or unfair in cases where there has been a prior final warning does not originate in the cases, which are but instances of the application of s. 98(4) to particular sets of facts. The broad test laid down in s.98(4) is whether, in the particular case, it was reasonable for the employer to treat the conduct reason, taken together with the circumstance of the final written warning, as sufficient to dismiss the claimant.

(emphasis added)

So, in principle, there is scope for argument: one can imagine a situation in which the final written warning is wholly unrelated to the dismissal conduct (suppose that the final written warning is about productivity on the factory floor, and the dismissal misconduct consists of smoking outside of a designated smoking area).

But case law indicates that final written warnings will very often (or usually) justify dismissal in the event of further misconduct.

In Wincanton Group Plc v Stone & Anor [2013] IRLR 178, [2013] ICR D6, [2012] UKEAT 0011_12_1110, the EAT issued guidance which is relevant both to final written warnings and expired warnings (more on those, later). The relevant passage is at para 37 and is worth citing in full:

[37] We can summarise our view of the law as it stands, for the benefit of Tribunals who may later have to consider the relevance of an earlier warning. A Tribunal must always begin by remembering that it is considering a question of dismissal to which section 98, and in particular section 98(4), applies. Thus the focus, as we have indicated, is upon the reasonableness or otherwise of the employer’s act in treating conduct as a reason for the dismissal. If a Tribunal is not satisfied that the first warning was issued for an oblique motive or was manifestly inappropriate or, put another way, was not issued in good faith nor with prima facie grounds for making it, then the earlier warning will be valid. If it is so satisfied, the earlier warning will not be valid and cannot and should not be relied upon subsequently. Where the earlier warning is valid, then:

(1) The Tribunal should take into account the fact of that warning.

(2) A Tribunal should take into account the fact of any proceedings that may affect the validity of that warning. That will usually be an internal appeal. This case is one in which the internal appeal procedures were exhausted, but an Employment Tribunal was to consider the underlying principles appropriate to the warning. An employer aware of the fact that the validity of a warning is being challenged in other proceedings may be expected to take account of that fact too, and a Tribunal is entitled to give that such weight as it sees appropriate.

(3) It will be going behind a warning to hold that it should not have been issued or issued, for instance, as a final written warning where some lesser category of warning would have been appropriate, unless the Tribunal is satisfied as to the invalidity of the warning.

(4) It is not to go behind a warning to take into account the factual circumstances giving rise to the warning. There may be a considerable difference between the circumstances giving rise to the first warning and those now being considered. Just as a degree of similarity will tend in favour of a more severe penalty, so a degree of dissimilarity may, in appropriate circumstances, tend the other way. There may be some particular feature related to the conduct or to the individual that may contextualise the earlier warning. An employer, and therefore Tribunal should be alert to give proper value to all those matters.

(5) Nor is it wrong for a Tribunal to take account of the employers’ treatment of similar matters relating to others in the employer’s employment, since the treatment of the employees concerned may show that a more serious or a less serious view has been taken by the employer since the warning was given of circumstances of the sort giving rise to the warning, providing, of course, that was taken prior to the dismissal that falls for consideration.

(6) A Tribunal must always remember that it is the employer’s act that is to be considered in the light of section 98(4) and that a final written warning always implies, subject only to the individual terms of a contract, that any misconduct of whatever nature will often and usually be met with dismissal, and it is likely to be by way of exception that that will not occur.

The problem is the phrase “of whatever nature”, which strongly suggests that there is no need for the dismissal conduct and the final written warning to be related.

What about a case where the final written warning post-dates the dismissal conduct? How (one might wonder) can it be fair to sack an employee for misconduct committed whilst on a final written warning, if they had not actually received the final written warning when the dismissal misconduct took place?

In Sweeney v Strathclyde Fire Board UKEATS/0029/13 the Appellant had sought to argue that there is an analogy between “expired” warnings (which should normally be disregarded) and final written warnings which post-date the dismissal conduct (because, in effect, they don’t cover the material time). The Appellant also pointed to the ACAS Code and argued that, since a final written warning is supposed to give an opportunity to improve, it would be unfair to rely on a final written warning in dismissing an employee if the final written warning post-dated the dismissal conduct (and so the employee couldn’t possibly have observed the warning). However, Lady Stacey commented, at para 36 (quoting selectively) that:

. . . While it is correct to argue that a warning is an admonition that tells the employee that future misconduct will have certain consequences, it is in my opinion more than that. It is also a recording of the commission of misconduct in the mind of both employer and employee. Mr Napier submitted that a warning is “Janus like” in that it looks both ways. I accept that submission. I am of the view that the Respondent was entitled to look at the Claimant’s record when deciding on the disposal in the disciplinary procedure relating to the criminal convictions. The Respondent was entitled to take notice of a finding of misconduct which was marked by the imposition of a final written warning . . .

So a final written warning is a powerful device. It will (usually) entitle an employer to dismiss for further misconduct, even where that misconduct is not similar to the conduct for which the employee received a final written warning. And it will (usually) entitle an employer to dismiss for misconduct which comes to the employer’s notice after the final written warning has been issued, even if the dismissal conduct pre-dates the final written warning.

What about expired warnings? Surely an employer should have to disregard expired warnings when deciding on disciplinary action?

In Diosynth Ltd v Thomson [2006] CSIH 5, 2006 S.C. 389 the employer had dismissed an employee for misconduct which would not normally have resulted in dismissal. It had dismissed the employee because it had taken into account an (expired) written warning. The Court in Diosynth (and note that Diosynth was heard in the Court of Appeal) referred to the ACAS Code, which says that warnings should normally be disregarded for disciplinary purposes after a certain period, and said that (at para 26, quoting selectively)

 . . . a warning which remains hanging over an employee’s head for an indefinite period would not normally be consistent with good industrial relations practice. It would be contrary to the spirit of para 15 [of the ACAS Code]

Furthermore (para 27 and 28, again quoting selectively)

[27] . . . the relevant warning was not stated to remain in force for an indefinite period but, according to the letter of 20 July 2000, was to stay on the respondent’s record for 12 months, a period which had expired before the acts of misconduct took place. Nevertheless, in regarding the warning as tipping the balance in favour of dismissal, the appellants acted as if it remained in force beyond the expiry of the 12-month period . . .

[28] The respondent was entitled to assume that the warning letter meant what it said, and that it would cease to have effect after one year. In seeking to extend the effect of the warning beyond that period the appellants, in our view, acted unreasonably. We therefore agree with the conclusion of the Employment Appeal Tribunal that the respondent was unfairly dismissed.

Diosynth remains good law, but its application has been clearly delimited by a later case, that of Airbus UK Ltd v Webb [2008] EWCA Civ 49, [2008] ICR 561, in which Mummery LJ observed, in short, that Diosynth does not establish any rigid principle of law; it does not establish that one can never rely on an expired warning. He commented, at para 73, that

73 The Diosynth case was addressing a different issue than that which arose in the tribunal in this case. As Lord Philip pointed out in para 27, on the facts of that case, the position of the employer was that the expired warning tipped “the balance in favour of dismissal” as the other factors taken together would not have justified dismissal . . . the expired warning . . . was the principal reason for the dismissal. As the warning had ceased to have effect, it was not reasonable for the employer to rely on it as the principal reason for the dismissal


[72] Diosynth is not authority for the general proposition of law that the misconduct, in respect of which a final warning was given, but has expired, can never be taken into account . . . it did not decide that the earlier misconduct and the expired warning are irrelevant circumstances of the case or are irrelevant to the equity and substantial merits of the case. It did not decide that the dismissal is necessarily unfair if account is taken of the expired warning . . .


[46] . . . I am persuaded that it is open to a tribunal to find that a dismissal for misconduct is fair, even though the employer, in his response to the reason for which the employee is dismissed, has taken account of the employee’s previous similar misconduct, which was the subject of an expired final warning.

One might wonder whether Airbus does leave at least some of Diosynth of potential use to claimant lawyers. Specifically, it does suggest (para 73, quoted above) that where an expired warning is the principal reason for the dismissal, it may not be reasonable for the employer to rely on it as such.

But the last word has to go to David Richards LJ in Airbus (again quoting selectively):

83 Common notions of fairness require that people should not go back on their word, to the detriment of others. This is particularly the case with those in authority, such as employers . . .

84 There cannot, however, be exact prescription where fairness is the statutory test . . .

Which is precisely what we should expect, given the wording of the statute.

To summarise:

  1. There is no principle of law which prevents employers from relying on expired warnings.
  2. There is no principle of law which prevents employers from relying (in dismissal proceedings) on a final written warning which post-dates the dismissal conduct.
  3. There is no principle of law which requires that the dismissal conduct and the final written warning conduct be the same (or even similar).
  4. A final written warning (almost) always implies that any misconduct of whatever nature will often and usually be met with dismissal, and it is likely to be by way of exception that that will not occur (Wincanton at para 37(6) again)
  5. Nonetheless, the test remains the broad test set out in s 98(4): whether the dismissal is fair in the circumstances, having regard to equity and the substantial merits of the case.

Quantifying how much an employer can withhold from a teacher who lawfully goes on strike

To oversimplify: Hartley & ors v King Edward VI College [2017] UKSC 39 dealt with a case where teachers had lawfully gone on strike, and where their employer had withheld salary apportioned, pro rata, over their actual working year. The teachers were attached to a sixth form college which, unlike state-maintained secondary schools, did not have an express term about the rate of deduction (in secondary schools, this is 1/365). Since weekends amount to 5/7 of the year, this meant that that each day of strike action attracted a deduction of 1/260 of salary. Was the employer entitled to do this?

“No”, said the Supreme Court.

The issues in the appeal were as follows. The Apportionment Act 1870 says that salary is to be considered as accruing from day to day, and is apportionable in respect of time accordingly (ss 2 and 5 read together). s 7 of the Act says that it does not apply to cases “in which  it is or shall be expressly stipulated that no apportionment shall take place.” Did the appellants’ contracts of employment provide, expressly or by necessary implication, for their salary to be paid pro rata so that 1870 Act did not apply? What did “from day to day” mean in the context of s 2 of the 1870 Act? And what is the correct construction of s 7 of the 1870 Act?

The Court found that there was no provision in the contracts of employment for the salary to be paid pro rata. The 1870 Act therefore applied. The central question was the application of s 2 to the contracts of employment. The Court noted that teachers’ contracts provide for a substantial portion of “undirected working”, i.e. time spent doing the work which is necessary to maintain the on-the-clock classroom work, and that teachers will invariably also work evenings, weekends, holidays, etc., on an ad hoc basis to cope with the workload.

Given this, the difficulty with the employer’s figure of 1/260 was that,

it makes no sense to choose a calculation of 1/260 of the annual salary, which assumes only week day working . . . [a]lthough . . . a case might perhaps be made for some othe rfigure, the only alternative figure put forward during the argument was 1/365 (para 29)

And although 1/365 might give some odd results, “that is almost always true of deeming provisions.” The correct interpretation of s 2 Apportionment Act 1870 is that “from day to day” means “from [calendar] day to [calendar] day”. (The Court of Appeal, in contrast, had concluded that the Act was aimed at ensuring an entitlement to such portion of salary as was referable to the period of service – but that didn’t require payment to accrue at an even rate).

Lastly, what about section 7? On its face, s 7 said that there must be an express provision in the contract which has the effect of disapplying the statutory formula so that “no apportionment shall take place” (para 38). And that was the correct interpretation.

So, absent an express provision in the contract, the principle of equal daily accrual will be the obvious principle to adopt. But (para 41):

In any case the precise figure will depend upon the true construction of the particular contract . . . [a] critical feature of the instant case which leads to a figure of 1/365 is that the contracts are annual contracts. If the contracts were not annual contracts the position would be very different . . .

This case, of course, is not as wide-ranging as the headline suggests, but it does provide extremely helpful guidance in relation to apportionment – not only in relation to teachers, but also in relation to other workers, particularly professionals, whose work will inevitably involve some duties which form part of regular working hours, and some which do not.

Supreme Court clarifies (and simplifies) law on indirect discrimination

The recent decision in Essop v Home Office and Naeem v Secretary of State for Justice [2017] UKSC 27 brings some much-needed clarity and simplicity to the law on indirect discrimination.

Employment law is a classic example of law which should be simple and accessible, in particular to laypeople, but which is, in practice, horrendously complex. Discrimination law is no exception.

In Essop, Lady Hale started out by noting that “discrimination ought to be an easy concept, although proving it may be harder. But we do not live in an ideal world and the concepts are not easy . . .”

Indirect discrimination requires that a provision, criterion or practice places persons with whom the alleged victim shares their protected characteristic at a particular disadvantage when compared with persons with whom they do not share that characteristic.

The principal issue of law raised in Essop was whether proving indirect discrimination requires that the reason for the disadvantage suffered by the group be established. The similar issue raised in Naeem was whether the reason for the disadvantage suffered by Mr Naeem had to be related to his protected characteristic of religion or race.

The judgment in Essop provides an interesting discussion of how discrimination law has evolved (significantly, protections have become wider over time, rather than narrower), and contrasts direct discrimination (which expressly requires a casual link between the characteristic and the treatment) with indirect discrimination (which requires a causal link between the PCP and the particular disadvantage). The discussion of course goes further (paras 23 – 34, if anyone is interested).

But the take-home message is at para 33:

“In order to succeed in an indirect discrimination claim, it is not necessary to establish the reason for the particular disadvantage to which the group is put. The essential element is a causal connection between the PCP and the disadvantage suffered, not only by the group, but also by the individual.”

That, thought Lady Hale, did not unduly disadvantage the Respondent, who could always (in principle) rely on objective justification.

A helpful decision, and a useful reference point for indirect discrimination claims in future.

(As a postscript:

Naeem doesn’t really change anything. In so far as it is relevant, a summary is as follows:

In Naeem, the reason why the PCP (related to a specific pay scale) put the group (Muslim chaplains) at a disadvantage was known: it depended on length of service and Muslim chaplains had, on average, shorter lengths of service than Christian chaplains. So the PCP put the group at a particular disadvantage; Mr Naeem suffered that particular disadvantage and so the question became whether the PCP (specifically, whether the steps being taken to move towards a new pay system) could be justified as “a proportionate means of achieving a legitimate aim”. But that was a factual question: “. . . if it was not fully explored before the Employment Tribunal it is not for the EAT or [the Supreme Court] to do so.”)

Workers, employers, and whistleblowing: s 43K Employment Rights Act 1996 and Dr Day

The case of Day v Health Education England [2017] EWCA Civ 329 (available on Bailii here) has attracted a fair bit of media attention.

Dr Day’s case was that he had been subjected to a number of significant detriments by Health Education England (“HEE”) as a result of having made protected disclosures – specifically (it was alleged) relating to staffing problems which impacted the safety of patients. His claim was struck out at a preliminary hearing, because HEE argued – successfully – that they could not (as a matter of law) be liable to him under whistleblowing legislation. If that argument is correct, then – in principle – a junior doctor (i.e. any doctor below the level of consultant) has no recourse for whistleblowing detriment against HEE. That is, obviously, significant.

The case in some more detail is as follows. Although Dr Day was not directly employed by HEE, he claimed that he was covered by s 43K ERA 1996. s 43K(1)(a) provides that

“For the purposes of this Part “worker” includes an individual who is not a worker as defined by section 230(3) but who— ”
(a) works or worked for a person in circumstances in which—
(i) he is or was introduced or supplied to do that work by a third person, and
(ii) the terms on which he is or was engaged to do the work are or were in practice substantially determined not by him but by the person for whom he works or worked, by the third person or by both of them…..

(s 230(3) gives a general definition of “worker” – it was common ground that Dr Day’s employer, under s 230(3), was Lewisham and Greenwich NHS Trust)

and there is an extended definition of “employer” in s 43K(2)(a)

“For the purposes of this Part “employer” includes—

(a) in relation to a worker falling within paragraph (a) of subsection (1), the person who substantially determines or determined the terms on which he is or was engaged . . .

At first instance, HEE took the preliminary point that they were not caught by s 43K. The ET sided with HEE: they looked at the relationship between Dr Day and the Trust, on the one hand, and Dr Day and HEE, on the other, and took the view that Dr Day’s relationship with HEE was a training relationship which ran parallel to the employment relationship (with the Trust) but did not materially determine its terms. The EAT thought that this was a conclusion which the ET was entitled to reach, and which displayed no error of law. There was a further point raised in the EAT, being that (in terms) a s 230(3) worker could not also be a s 43K worker.

There were two issues on appeal. First: is someone who is a s 230(3) worker thereby precluded from being a s 43K worker (in Dr Day’s case: whether his employment with the Trust prevented him from being a worker for HEE)? Second: if the answer to that question is “no”, then can HEE be said to be substantially determining “the terms on which he is or was engaged to do the work” so as to engage s 43K? Counsel for Dr Day, James Laddie QC, argued that the ET approached this question wrongly: the ET had asked itself which of the two bodies (HEE or the Trust) played the greater role in determining the terms of engagement, and therefore failed to appreciate that both might have substantially determined those terms.

The Court of Appeal noted that someone who is a s 230(3) worker cannot be excluded from also being a s 43K worker simply on that basis, despite what the EAT had said, because one might be a s 230(3) worker in one job (e.g. as a florist during the daytime) but a s 43K worker in another (e.g. as a delivery driver contracted by an agency during the evenings). So words had to be “read in” to the statute. The Court of Appeal observed that

“. . . where, as here, some words need to be read into the provision because a literal construction cannot be what Parliament intended, then in my view the court should read in such words as maximise the protection whilst remaining true to the language of the statute. In my judgment the words which both the appellant and intervener suggest should be inserted better achieve that objective.”

those words being as follows:

“”worker ” includes an individual who as against a given respondent is not a worker as defined by section 230(3).”

So on the first ground of appeal, Dr Day succeeded: he could, in principle, be a s 43K worker for HEE notwithstanding that he was a s 230(3) worker for the Trust.

Moreover, Dr Day also succeeded in arguing that the Tribunal had asked the wrong question in relation to s 43K: it had asked itself, for instance, about “the body” which substantially determined the terms of engagement (and had therefore assumed that there could be only one body which met this test).

Dr Day further argued that the ET would have been bound to find in his favour on the s 43K point – i.e. that it would have been bound to find that HEE did substantially determine his terms of engagement. But the Court of Appeal did not accept this argument, and the matter has been remitted to the ET.

Whether or not the ET will find in his favour is another matter. The judgment is not precedent for any point of principle about the relationship between junior doctors and Health Education England. There is an open question as to whether junior doctors have (any or any adequate) whistleblowing protection as against HEE. That is an unhappy state of affairs. HEE has now generated a document which purports to confer equivalent protection on whistleblowers to that afforded by the ERA 1996, albeit this time by a contractual route. As James Laddie QC somewhat puckishly notes, “[w]hy HEE is simultaneously contesting the question of whether it falls within s 43K in Dr Day’s case yet purporting to confer effective protection via another route is not a matter upon which I am able to comment.” And whether or not the protection is effective is open to debate.

For further reading: the collated judgments in Dr Day’s case can be found here; James Laddie QC’s opinion on the state of whistleblowing protection in relation to junior doctors is here; the BMA / HEE opinion can be found here, with the BMA’s overall view and guidance here.