Dobson v North Cumbria Integrated Care NHS Foundation Trust (Sex Discrimination)

THE HONOURABLE MR JUSTICE CHOUDHURY:

Introduction

I shall refer to the parties as "the Claimant" and "the Respondent" as they were below.

This is the second appeal in these proceedings, which arise out of the Claimant's former part-time employment with the Respondent as a community nurse. At the relevant time, in 2016, the Claimant had three children, two of whom were disabled. From about 2008, the Claimant had a fixed working pattern whereby, due to her caring responsibilities for her children, she would only work on Wednesday and Thursday each week. In 2016, the Respondent introduced a Flexible Working policy ("the FW Policy"), pursuant to which the Claimant was asked to work flexibly, including at weekends. The Claimant was clear that she could not accommodate that request, and she was eventually dismissed. The Claimant brought proceedings claiming indirect discrimination - in that an unjustified provision, criterion or practice ("PCP") was applied to her employment - and unfair dismissal. The first hearing of this matter was in 2019, when the Employment Tribunal ("the Tribunal") (Employment Judge Langridge presiding) dismissed her claim ("the First ET Judgment"). The Claimant's appeal against the First ET Judgment was heard by the EAT (Choudhury P sitting with members, Ms Branney and Ms Wilson) in February 2021. The EAT upheld the appeal ( Dobson v North Cumbria Integrated Care NHS Foundation Trust [2021] ICR 1699 (" Dobson 1 ")), deciding that the Tribunal had erred in concluding that there was no evidence of "group disadvantage" to found a claim of indirect discrimination, and that the Tribunal ought to have taken judicial notice of the "childcare disparity", which meant that women bear the greater burden of child care responsibilities than men. The matter was remitted to the same Tribunal to consider, amongst other things, whether the Respondent could show the PCP to be a fair and proportionate means of achieving a legitimate aim.

The remitted issues were considered by the same Tribunal in November 2022, with its judgment being sent to the parties on 11 August 2023 ("the Judgment"). The Tribunal concluded that the PCP was justified and that the claim of indirect sex discrimination failed, as did the claim of unfair dismissal. The Claimant appeals against the Judgment, the principal issue before the EAT being whether the Tribunal erred in its conclusion that the PCP was justified.

The Claimant is once again represented by Mr Sethi KC, assisted by Ms Berry (who appeared below in the Second ET hearing), Ms Balmelli and Mr Rodgers. The Respondent is represented by Mr Brittenden KC, who also appeared below. I am grateful to all Counsel for their helpful written and oral submissions.

Background

As there was no challenge to the findings of background fact, it is convenient to restate the summary of the background set out in Dobson 1 :

The Claimant's eldest child, who was severely disabled, sadly passed away in 2022.

At the second ET Hearing, the Tribunal made detailed additional findings relevant to the remitted issues. These included the following:

The Tribunal's decision

The Tribunal began its conclusions by identifying the question to be considered:

The Tribunal went on to conclude that:

Legal Framework

As in Dobson 1 , the key provision here is s.19 of the Equality Act 2010 ("EqA"), which provides:

Grounds of Appeal

Permission was granted to proceed with the following grounds of appeal:

Grounds 4 (& 7): Failure to consider discriminatory impact of PCP on entire pool

Claimant's submissions

It is further submitted that the Tribunal erred in finding that if the PCP was justified in the Claimant's team, then it was no less likely to be justified in respect of the wider pool. Such 'scaling up' was not appropriate on the evidence available. This was not a case where there could be a "reasoned projection" of the discriminatory effect of the PCP as there was simply insufficient information on which to undertake the required analysis.

The result, submits Mr Sethi, is that the Tribunal failed to apply the EAT's conclusion in Dobson 1 to the effect that group disadvantage was made out. The Tribunal's subsequent analysis in respect of justification could not therefore be regarded as safe, and ought to be set aside, with the EAT substituting its own decision on justification or remitting the matter to a freshly constituted Tribunal.

Respondent's submissions.

Mr Brittenden reminded the EAT that the Tribunal's finding on justification is essentially one of fact (see Dobson 1 at [64]), and that, as such, the Claimant's appeal amounts to a perversity challenge. In any event, submits Mr Brittenden, the Tribunal correctly directed itself on the legal principles and went on quite clearly to consider justification by reference to the impact of the PCP on all community nurses as well as on the Claimant herself, the latter analysis being perfectly legitimate. In this case, whilst group disadvantage was a given (in light of Dobson 1 ), the evidential burden as to the extent of such disadvantage lay with the Claimant and clearly had not been discharged. Even if that were not so, it was not a requirement for the Respondent to undertake a granular analysis of the myriad different ways in which Community Nurses could be disadvantaged, as the particular disadvantage being relied upon by the Claimant was the inability to comply with the PCP at all (which led to her termination) and any such analysis would impose a disproportionate burden on the employer.

As to 'scaling up', it is submitted that this was a perfectly legitimate finding of fact and discloses no error of law.

Grounds 4 (& 7) - Discussion

Guidance on the approach to be taken to the question of justification was provided by the Supreme Court in Chief Constable of West Yorkshire Police v Homer [2012] ICR 704 at [19]-[24]:

From these passages, it can be seen that, in determining whether the PCP in question is justified, the Tribunal has to weigh the real needs of the undertaking against the discriminatory effects of the PCP on "the affected group". The affected group will of course include the Claimant. This is not a straightforward task as was made clear in Hardy & Hansons plc v Lax [2005] ICR 1565 at [32] to [33]:

The Claimant's contention is that the Tribunal failed in its task by focusing on the disadvantage to the Claimant rather than the disadvantage to the group. Mr Sethi made it clear that in making that submission it was not being suggested that the impact on the Claimant was irrelevant. The submission therefore appears to be that the Tribunal placed undue weight or focus on the impact on her as opposed to the impact on the group. Indeed, that is the basis on which permission to pursue this ground of appeal was granted when it was said:

The legal proposition relied upon by the Claimant to establish an error of law is therefore that the Tribunal should not focus "too much" on the disparate impact of the PCP on the Claimant. However, that begs the question, "How much is too much?" Where a Tribunal has, in accordance with the authorities, considered the impact on the affected group, which will necessarily include the claimant, at what point does it become erroneous in law for that analysis to consider one aspect more than the other? The apparent emphasis on the impact on the individual may arise for good reason, for example, the preponderance of the evidence presented being to that effect, or that the evidence of impact on the group addresses a different period of time from that which is relevant. In other cases, where the position of others is not materially different, the evidence of the impact on the individual may reliably inform the Tribunal about the impact on the group. Where the law stipulates no precise level of "focus" to be placed on either the group or the individual, it is a matter of judgment for the Tribunal as to how to approach the available evidence. In the present case, the Tribunal repeatedly noted that the evidence as to the precise statistical evidence about the wider pool of community nurses was lacking. However, there were some concrete items of evidence about them, including the fact that none worked on fixed days and none had had to leave employment as a result of the PCP. There was of course extensive evidence about the Claimant's particular difficulties in complying with the PCP. The Tribunal was entitled, in those circumstances to approach the evidence, such as it was, in the way that it did, with there being more analysis of the impact on the Claimant. Short of that approach being perverse, which is not alleged, no error of law arises.

None of this detracts from what was said about group disadvantage in Dobson 1 . The EAT there concluded that whilst taking judicial notice of the childcare disparity did not invariably result in group disadvantage being made out, the Claimant's case was one where the PCP was "inherently more likely to produce a detrimental effect which disproportionately affected women": Dobson 1 at [51] and [52]. The Tribunal clearly accepted that conclusion and took group disadvantage as read, saying as follows at [15]:

Further, at [16], the Tribunal stated:

These passages rebut Mr Sethi's suggestion that the Tribunal failed to give effect or have regard to the EAT's conclusion on group discrimination. Of course, in determining the question of justification, the Tribunal is required to balance the extent of that disadvantage against the needs of the employer. In doing so, the analysis is not undermined by considering the impact on the individual:

An approach that takes account of the impact on the individual as well as on the affected group more generally is consistent with the legislative provisions and accords with common sense. Section 19, EqA requires both that those sharing the protected characteristic are put to a particular disadvantage and that the Claimant is put to "that disadvantage". Those are the preliminary requirements under the second and third limbs of s.19 (the first being the application of a PCP to a wider group) even before one gets to the fourth requirement under s.19, EqA that justification is shown. It cannot be right that in considering justification, it is necessary for the Tribunal to disregard or pay less attention to the very disadvantage to the Claimant that was found to exist under the third limb of s.19. Given that the disadvantage suffered by the Claimant, namely the childcare disparity, is that to which other women community nurses were also put, it is logical to consider, so far as is possible on the evidence, the impact of the PCP both on the individual and the group. Indeed, the claim of indirect discrimination could not be made out if there was no impact on the individual, irrespective of the impact on the wider group. What is not permissible is an analysis solely of the impact of the PCP on the individual as that would be to "negate the purpose of having a [PCP] and justifying the [PCP] in the particular circumstances of the business".

The passage in Harvey's on which the Claimant places reliance provides:

The highlighted words must be read in the context of the whole passage. As the passage goes on to make clear, it is the rule (or PCP) which must be justified as opposed to particular instances of its application to an individual. Read thus, it becomes clear that the highlighted words (upon which the Claimant relies) are not to be interpreted as meaning that little or no consideration is to be given to the impact on the individual, but rather that the focus should be on the PCP and its effect more generally. That will of course include the impact on the individual.

Mr Sethi submitted that there was a misdirection of law on the part of the Tribunal at [32] of the Judgment where it was said as follows:

It is said that this test in fact derives from a different case, namely R (Tigere) v Secretary of State for Business Innovation and Skills [2016] 1 All ER 191 , and is not apt for the present case, which is concerned with indirect discrimination rather than one involving justification for a contravention of Article 2 of the First Protocol to the European Convention on Human Rights . Mr Sethi is undoubtedly correct that the precise formulation as set out in [32] of the Judgment does not appear in Homer as suggested by the Tribunal. However, no criticism can be made of the first two bullet points, each of which do appear in Homer in a similar form: see the passages cited above. Homer states expressly that a measure may be inappropriate if it goes further than is reasonably necessary in order to achieve a legitimate aim: Homer at [23]. One way of testing whether a measure has gone further than reasonably necessary would be to ask whether a less intrusive measure could have been used. In other words, the third bullet point in [32] of the Judgment is also to be found in Homer in some form.

Mr Sethi's real objection lies, however, with the inclusion of the fourth bullet point and the reference to the balance between the rights of the individual and the interests of the community. That same phrase is used at paragraphs [191] and [228] of the Judgment. However, as Mr Sethi fairly accepted during submissions, the Tribunal's reference to "the community" in its judgment is clearly intended to refer to the Trust and all its employees rather than the wider EU community in any European law sense. As the Tribunal stated at [191]:

Viewed thus, the criticisms of the Tribunal's misquote of Homer collapse into nothing. In any event, as is apparent from that passage at [191], the Tribunal had in mind precisely the question it had to consider and was not focused solely on the impact on the Claimant. Whilst the Tribunal did give extensive consideration to the impact on the Claimant, it also considered the impact on the group as the following passages in the Judgment demonstrate:

These passages, amongst others, indicate that the impact on the group was considered. The information as to the impact on the group could be summarised as follows:

Mr Sethi's complaint is that that is not sufficient and that the Tribunal ought to have gone further and inquired as to the specific challenges faced by community nurses as a result of the PCP. I disagree. The disadvantage relied upon by the Claimant was not that she could comply with the PCP with some difficulty or additional cost, but that she could not comply with it at all to the point where her employment had to be terminated. The inference to be drawn from the findings summarised at [32] above is that all other community nurses could comply with the PCP. That is not to say that there was no disadvantage caused by the PCP: clearly there was as the EAT's finding that there was group disadvantage by reason of the childcare disparity dictates. Some may have had to call in help from other family members or incur the expenditure of additional babysitting. The Tribunal implicitly recognised that such steps would amount to a disadvantage when it said at [236]:

However, the fact that every other community nurse was ultimately able to comply with the PCP is relevant in assessing the "seriousness of the detriment" caused to the group: see Homer at [20] citing from R (Elias) v Secretary of State for Defence [2006] 1 WLR 3213 , at [151]. The conclusion, reasonably to be drawn, is that the detriment was at the lower end of the scale, a conclusion that the Tribunal set out expressly in the remainder of [236]:

It follows that the disadvantage to the group - none of whom were unable to comply, and for whom the disadvantage may therefore be inferred as being less serious than for the Claimant - is also at the "lower end of the scale". The disadvantage experienced by others would be of the same type as that experienced by the Claimant, namely the need to arrange childcare for those days on which they are rostered for work, the difference in the Claimant's case being one of degree, given the particular difficulties she faced with (at the time) two disabled children.

In any case, there was no evidence as to the specific challenges (short of non-compliance leading to dismissal) faced by other community nurses. No such challenges were alleged by the Claimant. Mr Sethi relies upon the suggestion in Essop (citing from the EAT's decision in Essop ) at [29] that:

as imposing some sort of requirement on the Respondent to have monitored the precise effects of the PCP on female staff, and submits that the Respondent's failure to do so and to adduce any evidence in this regard should lead to a conclusion that the burden of justification has not been discharged. However, Essop does not establish any such requirement. Indeed, as Baroness Hale also stated in Essop , "The requirement to justify a PCP should not be seen as placing an unreasonable burden upon respondents". There can be no doubt that a requirement to undertake a forensic analysis across many thousands of employees at the Trust to whom the PCP was applied as to the myriad ways in which inconveniences might have been suffered or costs incurred in each case would place an unreasonable burden on the Respondent. As stated by Eady P in Pitcher v University of Oxford [2022] ICR 338 :

The position might have been otherwise if specific complaint had been made of particular disadvantages suffered other than those suffered by the Claimant, but there was none. It was sufficient in the circumstances of this case for the Tribunal to rely upon the unchallenged evidence summarised at [32] above. No other employee did complain of being unable to comply, nor it seems raise any sort of grievance arising out of the PCP. That evidence, which gave a clear insight into the seriousness of the detriment caused by the PCP, enabled the Tribunal to understand both the quantitative and qualitative effect of the PCP's impact on the group.

I was referred to the decision of Collins J in G v St Gregory's Catholic Science College [2011] EWHC 1452 (Admin) , in which the Court considered whether a school uniform policy that forbade the keeping of hair in "cornrows" was discriminatory. The School relied on the fact that no child other than G had complained about the prohibition on cornrows. As to that matter, Collins J said as follows:

The Court went on to conclude that the indirect discrimination caused by the prohibition on cornrows was not justified. Mr Sethi submits that, similarly in the present case, too much weight should not be attached to the absence of any complaints or grievances about being required to work flexibly. The first point to note is that the decision in St Gregory's was not that the absence of complaints was irrelevant - it was acknowledged to be a "material factor" - but that it was not determinative. Furthermore, the reason that the claimant in that case found it difficult to conform to the uniform policy was that it was his family tradition not to cut his hair and to keep it in cornrows: see St Gregory's at [2]. Others might have kept cornrows out of a fashion choice or to make the hair more manageable. In such circumstances, others may have had available to them a choice to conform as their difficulties were of a different nature to those of the claimant, G. In the present case, as stated above, the difficulties that others in the Trust faced in respect of the PCP to work flexibly were of the same type as those faced by the Claimant, namely the need to arrange childcare, the difference being one of degree. They could not simply choose to do without childcare, but would have sought to make such arrangements as they could in order to comply. In such circumstances, the fact that every other person was able to comply with the PCP is clearly a highly material factor in assessing the seriousness of the detriment that is to be weighed against the Trust's needs. The decision in St Gregory's does not therefore indicate any error on the part of the Tribunal.

I turn then to the second aspect of this ground of appeal which is that the Tribunal erred in 'scaling up' its conclusions on justification in respect of the Claimant's team to the whole of the affected group. The impugned conclusions are set out in the Judgment at [197] and [199], both of which are cited at [31] above. Whilst the Tribunal accepted the Respondent's submission that if the PCP was justified for the Cockermouth team, it is no less likely to be justified in respect of the wider pool, there is nothing in the Judgment to suggest that the Tribunal treated such "scaling up" as determinative in its assessment of proportionality. As the analysis above demonstrates, the Tribunal expressly considered the position of the affected group more widely. As such, any reference to extrapolation from the analysis in respect of the team, which was the analysis undertaken in the First ET Judgment, was merely supportive of the eventual conclusion rather than determinative. This is also demonstrated by the fact that these references to justification in respect of the team appeared under a section of the Judgment with the sub-heading, "Evidential Considerations", in which the Tribunal considered, amongst other matters, the absence of statistical evidence and the Respondent's contention about scaling up. However, the Tribunal's analysis of proportionality proper begins at [202] and/or [203] onwards with the Tribunal reminding itself that it is for the Tribunal to reach its own judgment on the proportionality question. (It would appear that a further intended sub-heading was omitted). Thereafter, the Tribunal made findings as to: the legitimate aim (at [203]); and whether the PCP was rationally connected to the legitimate aim (at [210]); before turning to the "broader aspects of the proportionality issue" from [211] onwards. It is notable that the Tribunal makes no further reference to scaling up at any point after [203].

In any case, even if the Tribunal had relied more heavily upon its findings in respect of the team, there would be no error of law in doing so. The conclusion that justification would be "no less likely" in respect of the wider group would be a reasonable inference to draw from the established facts which are a matter for the Tribunal.

The Tribunal carefully explained why it was appropriate to extrapolate from the data relating to the team to that which existed across the Trust. This included the fact that the predominantly female make-up of the Claimant's team was reflected in the available data as to gender balance across the Trust. Whilst there were no specific figures of the position as at 2016, the Tribunal was able to draw that inference from the available data, including a later survey by the Royal College of Nursing. Furthermore, the Tribunal was firm in its conclusion that the business case "put beyond doubt that the circumstances that affected the claimant were not limited to the team in Cockermouth but affected community nursing across the Trust": Judgment at [199]. This is an example of the sort of "reasoned projection" which the Tribunal is entitled to make based on the available evidence:

Once again, in the absence of a perversity challenge, which is not pursued here, these are matters that are for the Tribunal to determine and with which the EAT will not interfere.

For these reasons, it is my view that there was no error of law on the part of the Tribunal, either in its approach to the evidence or in respect of scaling up, in reaching the conclusion that the PCP in this case was justified within the meaning of s.19(2)(d), EqA. Its conclusions in these respects were largely findings of fact with which the EAT should not interfere. Accordingly, Grounds 4 & 7 of the appeal fail and are dismissed.

Ground 8: Irrelevant Considerations

There are two aspects to this ground of appeal: the first is that the Tribunal wrongly took into account the Claimant's unwillingness to modify her stance that she should work only Wednesdays and Thursdays. Mr Sethi submits that the Tribunal erred in treating the Claimant's responses during the consultation and her failure to make any alternative suggestions as relevant considerations in determining the proportionality of the PCP. Reliance was placed on Homer , in which it was said that it is "necessary to weigh the [employer's] need against the seriousness of the detriment to the disadvantaged group" ( Homer at [20]); and on an extract from Harvey's which refers to the need for the Tribunal to consider the balance between the discriminatory effect of the measure and the legitimate aim ( Harvey's at [338.03]).

I cannot accept that the Claimant's responses during the consultation or her failure to suggest any alternative measures are irrelevant. No authority was cited in support of such a stark proposition. The authorities that Mr Sethi directed me to do not assist, as any objective analysis of the "seriousness" of the discriminatory effect will almost inevitably involve some investigation of the difficulties caused and the extent of the inability to comply with the PCP. That could include, as it did in this case, whether the Claimant's insistence on being unable to comply was reasonable in the circumstances.

Whilst the burden of establishing justification lies with the Respondent, the Tribunal must reach its own view as to whether the PCP is objectively justified. The Tribunal is not bound simply to accept the Claimant's assertions as to the extent of the disadvantage, notwithstanding the starting point that the childcare disparity will mean that there is at least some such disadvantage, just as it was not bound to accept the employer's assertions as to business need. In order to assess the seriousness of the detriment - bearing in mind that no other employee in the group was unable to comply - and whether the PCP was reasonably necessary, the Tribunal was entitled to take account of the Claimant's stance and whether she was able to suggest any alternative.

Such authority as does bear on the issue would appear to be against the Claimant. Baroness Hale in Essop stated as follows at [47]:

Whilst Baroness Hale's observation cannot be read as imposing any sort of legal requirement on claimants to propose alternatives in every case, it is undoubtedly useful guidance, and certainly makes it clear beyond peradventure that whether or not alternatives are suggested may be a material consideration.

Mr Sethi's fall back position is that if such matters are not irrelevant then the Tribunal erred in attaching too much weight to them. However, the weight to be attached to relevant evidential matters is for the Tribunal to determine. Short of an allegation of perversity (which, once again, is not alleged), the Tribunal cannot be said to have erred in law in treating the Claimant's stance as a "salient feature of the overall circumstances" to be weighed in the balance. In this case, as the Tribunal found, the Respondent made extensive efforts to reach an accommodation with the Claimant, but she had remained "intransigent", notwithstanding evidence available to the Tribunal that the occasional weekend or bank holiday working "was in fact manageable" for the Claimant: Judgment at [235]. It was such evidence that led the Tribunal to find as an unchallenged fact that the Claimant "could meet the respondent's requirement to work flexibly, including at some weekends, albeit with difficulty": Judgment at [241]. That was a finding that the Tribunal was entitled to make and which was pertinent to the balancing exercise that it was required to conduct.

The second aspect of this ground of appeal is that the Tribunal wrongly focused on the wrong PCP in that it considered the Claimant's ability to comply with a "diluted" version of the PCP rather than the PCP that was in fact imposed on her and which led to her dismissal.

This contention is without merit. The PCP at the heart of this case was that set out at [72] of the First ET Judgment:

That PCP does not specify the regularity with which community nurses are required to work flexibly or at weekends. There is no suggestion, for example, that community nurses were required to work every weekend. The Tribunal made a further finding of fact in respect of the PCP at [208] of the Judgment:

That of course makes sense: a rigid or fixed approach to the requirement to work flexibly including at weekends would be inconsistent with the desired flexibility inherent in the policy. Within those parameters, a PCP that involved occasional weekend working on notice, is no different in substance from the PCP that community nurses work flexibly including at weekends.

In any case, the Tribunal's consideration of the Respondent's efforts to reach an accommodation with the Claimant cannot be said to be wrong: as is clear from Essop , that is what a reasonable employer will attempt to do: see Essop at [29].

Ultimately, the Tribunal concluded at [241] that the Claimant "could meet the respondent's requirement to work flexibly, including at some weekends." That PCP is not materially different to the one referred to at [52] above. In the absence of any precise stipulation as to the frequency of weekend working, the inclusion of the qualifier, "some", does not change the substance of the requirement.

For these reasons, Ground 8 of the Appeal also fails and is dismissed.

Ground 11: Unfair Dismissal

This ground stands or falls with the grounds relating to indirect discrimination. The Tribunal's finding that the PCP was justified remains undisturbed. As such, Ground 11 also fails and is dismissed.

Conclusion

For the reasons set out above, this appeal fails and is dismissed.