Michael Cowie & Ors v Scottish Fire and Rescue Service

The appeals before us raise the following questions:

Whether the Employment Tribunal (“ET”) was correct in finding the preconditions imposed by the employer for obtaining paid special leave during the coronavirus pandemic amounted to unfavourable treatment, or gave rise to a disadvantage, for the purpose of section 15 and/or section 19 Equality Act 2010 (“EqA”)?

If there was, in principle, a disadvantage for the purposes of section 19 EqA, whether the ET erred in concluding that it had not been shown that the preconditions in issue had placed women at a particular disadvantage?

Having found that the section 15 claims were made out, whether the ET erred in making no award for injury to feelings?

In giving this Judgment, we refer to the parties as the claimants and the respondent, as below. This is the full hearing of the parties’ respective appeals against the Judgment of the Glasgow ET (Employment Judge Hoey sitting with lay members, N Bakshi and N Elliot) following a full merits hearing conducted by cloud video platform (“CVP”) over five days in July 2021, with three further days of deliberations in chambers. By that Judgment, the ET upheld the claimants’ claims of disability discrimination under section 15 EqA but ruled that it was not just and equitable to award any compensation. The ET dismissed the claimants’ claims of indirect sex discrimination.

By EA-2021-SCO-000096, the claimants appeal against the ET’s dismissal of their claims of indirect sex discrimination and the decision to make no award on their section 15 EqA claims. By EA-2021-SCO-000097, the respondent appeals against the ET’s approach to the question of “unfavourable treatment” when determining (in the claimants’ favour) the question of liability for disability discrimination under section 15 EqA.

Given the constraints on judicial resource, it was not possible for the hearing of these appeals to take place in-person in Edinburgh on the date listed (or at any time before December 2022, at the earliest). As the parties did not consent to an in-person hearing in London, having regard to the need to avoid undue delay and in accordance with the overriding objective, I directed that this hearing take place remotely, by CVP. Mr McHugh represented the interests of the claimants on the appeal, as he did before the ET; the respondent appeared before us by Ms Beale, having been represented before the ET by its solicitor.

The respondent is the fire service in Scotland. It is subject to a number of statutory duties in carrying out its emergency fire and rescue functions and has a more general duty, as a public body, to secure best value in the use of public funds. It employs some 8,000 employees: 7,000 operational and 818 non-operational staff; split 86% male and 14% female.

The claimants were all employees of the respondent, based in different locations. They were all members of the Fire Brigade Unions (“FBU”), an independent trade union recognised by the respondent for collective bargaining purposes.

The respondent operated a special leave policy, agreed with the relevant trade unions, which provided guidance relating to the circumstances under which paid or unpaid special leave might be granted. Although special leave was not a contractual entitlement, the policy was adopted as part of the respondent’s commitment to assist employees in balancing their work/life responsibilities, recognising that, from time to time:

Separately, the respondent operated a TOIL policy, negotiated with the relevant trade unions, applicable to all uniformed employees. In the introduction to this policy, it was stated:

The policy detailed the means by which TOIL could be accrued and taken. Most relevantly, TOIL could be accrued through overtime, detached duty, acting-up, or public holiday enhanced duty; it could only be taken with management approval; it was to be cost neutral and avoid generating additional costs; it could only be taken in full shifts or in periods of four hours or more; employees could only hold a maximum of 32 hours TOIL at any given time; it was each employee’s responsibility to apply for TOIL, giving at least 24 hours’ notice.

The coronavirus pandemic had a significant impact upon how the respondent carried out its functions and its arrangements changed, depending upon the prevailing situation. Additional pressure was placed on the respondent to maintain service delivery and crews and, in order to ensure an appropriate response was available for emergency call-outs, some changes to crewing appliances and staffing were implemented (for example, the respondent created “bubbles” where those working the same shifts would, so far as practicable, stay within that group and not mix with other staff; it also reduced the number of staff needed for crewing each appliance). As the ET found, the respondent sought to be as supportive as it could to ensure staff were able to manage personal issues while attending work. If staff were able to work from home, that was accommodated; if they were not, a sympathetic approach was adopted to seek to assist employees balance work and personal issues, such as by agreeing shift or location changes where practical.

In meetings with the FBU in the early months of the pandemic, the respondent advised that there would be interim changes to manage the effects of the pandemic. Specifically, paid special leave would be extended to address circumstances to which it would not normally apply, but TOIL and annual leave would be required to be used first. At around that time (March/April 2020), 52 employees were shielding because of household members and 197 because of their own medical condition (around 250 staff in total). More generally, some 555 staff were working from home, being mainly those working in support roles. There were some 150 staff who were shielding who could not perform their duties, or could only perform a very small part of their duties, from home. Other staff with school-age children and/or who had caring responsibilities also faced difficulties due to the disruption of the arrangements they would otherwise have relied upon to enable them to work their contracted hours.

The respondent duly issued guidance to its staff during the pandemic that addressed these various issues. It confirmed that all of its staff were key workers and should be entitled to send their children to the schools that remained open. It was noted that if a worker had underlying health conditions or was pregnant a discussion should take place with the relevant line manager, who would provide the necessary support, which would include looking into the possibility of home working. For staff with children whose schools had closed, it was advised that they may be able to work from home. Where that was not possible, a discussion was to take place with the relevant manager to agree what hours could be worked to suit the individual’s circumstances; a temporary adjustment to usual core hours could be made to allow weekend or evening working, or shift patterns could be altered. Where the employee could only work fewer hours, the hours they could work would be recorded and a discussion would then take place about using TOIL or accrued annual leave. Alternative working arrangements, including adjustments to shifts, would also be considered for those who had caring responsibilities whose arrangements were disrupted by coronavirus.

Where such staff were unable to work their full contracted hours, it was provided that they could apply for paid special leave but they were required to first use any TOIL and accrued annual leave. Similarly, all staff who had to shield and who could not work from home were required to first use their TOIL and accrued annual leave before being paid special leave.

More generally, after a short period early on in the pandemic when new leave had not been authorised, the guidance advised that the respondent encouraged staff to use their annual leave as it would enable them to take paid time off work so as to have regular breaks to rest and re-energise, and staff were to continue to request leave in the usual way. Where staff were unable to take annual leave during 2020, the respondent offered to buy back any leave in excess of 5.6 weeks (although, in the event, no staff member availed themselves of this offer).

Had staff not been asked to utilise their accrued annual leave and TOIL, the ET found the respondent’s operation would have been adversely affected in a number of ways.

The FBU considered that the respondent’s paid special leave policy during the pandemic amounted to sex and disability discrimination, and, in July 2020, it raised a collective grievance on behalf of its members. The FBU also relied on fair work statements issued by the Scottish government, the Scottish Trades Union Congress and others, which set out the expectations of the signatories to the statements as to how workers were to be treated during the pandemic. Relevantly, it was stated:

The respondent rejected the grievance. It took the view that there was no breach of the fair work principles: no worker was subject to a financial detriment as all leave was paid. Moreover, it was not considered unreasonable to require accrued leave and TOIL to be utilised before paid special leave was granted when those who were required to shield but could work from home had to use annual leave in the usual manner, as did those who continued to attend work. The respondent considered that not applying these rules would place those who were the subject of the grievance in a more favourable position.

The FBU appealed against that ruling but was again unsuccessful.

The proceedings before the ET concerned two sets of claims: (1) a group claim, complaining of unfavourable treatment because of something arising in consequence of disability, under section 15 EqA; and (2) a separate group claim, of indirect sex discrimination, under section 19 EqA. The claims had been conjoined because they stemmed from the same situation and related to the same action by the same respondent. Specifically, both sets of claims arose from the respondent’s policy to require employees to exhaust accrued annual leave and TOIL before being treated as entitled to paid special leave in the context of coronavirus pandemic-related leave and shielding.

It was common ground between the parties that the claimants in this claim were, at all material times, disabled persons for the purposes of the EqA.

It was the claimants’ case that they were subjected to unfavourable treatment in the following respects: (1) having to use their accrued TOIL; (2) being compelled to use annual leave at a time when they did not wish to do so; (3) being compelled to take annual leave while shielding at home; (4) not being able to utilise re-allocated annual leave as would have happened had they been categorised as on annual periods of sick leave.

The ET noted that “unfavourable treatment” is “widely defined and the threshold is low” (ET paragraph 309). The respondent had argued that the treatment in question was in fact the granting of special paid leave (paying for a worker to stay at home), which was “clearly favourable treatment” (ET paragraph 313); the ET ruled that was not the case:

Considering the requirement to use accrued TOIL, the ET characterised this as

In respect of the first matter relied on by the claimants, the ET concluded that depriving them of the TOIL they had earned, and removing the choice and flexibility they had, was unfavourable treatment; it was not something the claimants wanted to do. Moreover, other staff, who had worked the same hours but had chosen money instead of TOIL, were given paid special leave immediately (without taking into account the additional payment they had received for those hours). Requiring the claimants to use accrued TOIL before paying special leave put them at a disadvantage and amounted to unfavourable treatment. (see the ET at paragraphs 315-317)

As for the second, third and fourth matters relied on, the ET concluded that being compelled to use annual leave at a time when the claimants did not wish to use it was also unfavourable. If the option was taking paid special leave and then using accrued annual leave at a later date, as opposed to having to exhaust annual leave first, no claimant would opt for the latter, as it was disadvantageous or unfavourable. Comparing the position of the claimants to those who had used their accrued holidays at an earlier date (pre-pandemic), the ET noted that those other individuals would have received the benefit of their accrued annual leave and then paid special leave; the claimants, however, had to use their accrued leave before receiving paid special leave. That loss of flexibility and choice again amounted to unfavourable treatment (ET paragraphs 318-320).

Returning to the argument that the treatment in question was advantageous and that the claimants were thus seeking more advantageous treatment (to get TOIL and accrued annual leave in addition to receiving paid special leave), the ET disagreed, finding that this case was:

The ET took account of the fact that the respondent had imposed the same requirement on all staff, but did not consider that treating everyone in the same way necessarily meant that the treatment was not unfavourable as “it was possible for a policy of general application to amount to unfavourable treatment” (ET paragraph 330).

The claimants had argued that the “something” arising in consequence of their disability was the requirement to make use of the paid special policy; that only arose because of their respective disabilities, which meant that they were unable to attend work due to government advice. The ET agreed, finding it was self evident that the reason why each of the claimants required to make use of the paid special leave policy was because their disability prevented them from attending work (which could not be done from home) (ET paragraphs 335-336).

Turning to the question of justification, it was not in dispute that the treatment in issue was introduced by the respondent for legitimate aims, but it the ET concluded that the respondent had not demonstrated that the measures were objectively justified. There is no appeal before us against this aspect of the ET’s decision.

The question of remedy had been included in the list of issues for the hearing and the parties had confirmed they were in a position to deal with this (ET paragraph 421). Mr McHugh accepted, however, that there was no material before the ET which dealt with the financial position in relation to the individual claimants. As for possible injury to feelings, the ET considered the case of each claimant in turn (although it only received oral evidence from Mr Cowie), but concluded that the evidence did not provide a basis for an award of compensation under this head. Mr Cowie, for example, was “upset” and believed the treatment was “not very nice” but the ET found the upset he felt was not attributable to the unlawful act of discrimination but because of the consequences of the pandemic more generally, and the evidence pointed to a sense of (justified) grievance rather than to a suffering of any injury to feelings. The ET had not heard evidence from the other claimants and did not consider it was able to assess what, if any, injury to their feelings arose as a result of the discrimination found. More generally, the ET considered that it would be an error of law to award compensation for injury to feelings without hearing from the individual concerned as to how the discrimination had impacted upon them.

On 14 September 2021, those acting for the claimants applied for reconsideration of the ET’s decision on remedy. That application was considered at a further (in-person) hearing on 14 December 2021 but rejected.

It was common ground before the ET that the respondent had applied a provision criterion or practice (“PCP”) in requiring employees, in the context of the coronavirus pandemic, to take annual leave and TOIL before being granted paid special leave. That was a PCP applied to all staff, both men and women, and the ET therefore asked whether it had placed the claimants at a particular disadvantage when compared with persons with whom they did not share the protected characteristic of being a woman (i.e. men).

The claimants had relied on the following three matters as particular disadvantages: (1) having their TOIL removed; (2) preventing them from exercising their accrued TOIL to balance work and family commitments; (3) compelling them to use their annual leave at a time when they did not wish to do so. It was the respondent’s case that, as a matter of principle, the matters relied on could not amount to a particular disadvantage as the entitlement to special paid leave was inherently advantageous.

The ET noted that the claimants were not saying that paid special leave was disadvantageous, rather the conditions that were required to be satisfied in order to obtain that leave were disadvantageous. With regard to the first matter relied on, the ET had no doubt that “having accrued TOIL removed was in principle a disadvantage”; that arose because the claimants had to use it before being paid special leave, rather than having both the benefit of paid special leave and the ability to use TOIL at a later date (ET paragraph 388). The ET did not, however, consider that the evidence established that the claimants were prevented from using accrued TOIL to balance work and family life commitments, concluding that the claimants’ case on the second disadvantage had not been made out (ET paragraph 389). As for the third matter, the ET considered it was self-evident that employees would not want to take their accrued leave if it was possible to have paid special leave instead and to thus retain accrued leave for a later date (ET paragraph 390).

Having found that there were two matters that could, in principle, amount to disadvantages, the ET went on to consider whether there was in fact a group disadvantage; that is, whether women in general were placed at the particular disadvantages of having TOIL removed and being compelled to take accrued leave when they did not wish to do so. The ET noted that no evidence had been led on this issue but considered, nevertheless, whether it was possible to infer the disadvantage from the evidence it had heard or as a matter of logic. It concluded that it was not. There was no evidence that female employees had been more likely to accrue TOIL than male employees; both had accrued similar TOIL balances, which suggested that men had used TOIL as much as women. Accepting that women ordinarily carry greater child care responsibilities (albeit the ET considered it was entirely possible that, during the pandemic, the position might have been more nuanced) did not necessarily mean that more women than men would require to rely on the paid special leave provisions when child care had broken down. Moreover, the paid special leave provisions were relied on not just by those with child care responsibilities: those who were disabled or caring for a disabled person were also required to use it; it could not be said that the majority of those using the policy, for whatever reason, were women rather than men. In the circumstances, the necessary group disadvantage had not been made out (see ET paragraphs 393-404).

In the alternative, however, the ET went on to consider individual disadvantage. Although it had only heard from the claimant Ms McCrone, the ET considered that it was “self evident” that, for those claimants whose TOIL and/or accrued annual leave had been removed, the particular disadvantage had been established (ET paragraph 407).

On this alternative basis, the ET turned to the question of justification, finding that the respondent had not demonstrated that the measures were objectively justified. Again, there is no appeal before us against this aspect of the ET’s decision.

By section 15 EqA, it is provided:

A person (A) discriminates against a disabled person (B) if— (a) A treats B unfavourably because of something arising in consequence of B's disability, and (b) A cannot show that the treatment is a proportionate means of achieving a legitimate aim.”

This appeal concerns the requirement that “A treats B unfavourably”. That expression was at the heart of the appeal in the case of Williams v Trustees of Swansea University Pension & Assurance Scheme [2018] UKSC 65; [2019] ICR 230 SC. Mr Williams was a disabled person, who had over 13 years of pensionable service at the date his employment ended. For the first ten years he had worked full-time but in the last three years of his employment his hours had been reduced for disability-related reasons. At the age of 38, Mr Williams retired for ill-health reasons. Under the ill-health retirement provisions of the pension scheme he was entitled to an enhanced element, calculated on the basis of his actual salary at the date of retirement. Mr Williams complained that this amounted to unfavourable treatment (the calculation being related to the lower salary he was receiving at that time, without having regard to his earlier full-time earnings) because of something (his reduced hours) arising in consequence of his disability.

In his Judgment (with which the other members of the Supreme Court agreed), Lord Carnwath noted that section 15 did not require any comparison of treatment, instead:

In answering those questions in Williams, Lord Carnwath reasoned as follows:

In his Judgment, Lord Carnwath expressed himself as being “substantially in agreement with the reasoning of the Court of Appeal”, where the leading Judgment was given by Bean LJ (see [2018] ICR 233 paragraphs 42-43). Before the Court of Appeal it had been argued (on Mr Williams’ behalf) that he had suffered a disadvantage as compared to those who had a different disability, which might mean that they qualified for ill-health retirement in circumstances where they had not suffered any reduction in their earnings (see the passages set out at paragraph 19 of Lord Carnwath’s Judgment); that, the Court of Appeal reasoned, did not mean that Mr Williams had been placed at a disadvantage. Distinguishing “unfavourable treatment” from detriment (the term in issue in Shamoon v Chief Constable of the Royal Ulster Constabulary [2003] ICR 337 HL); as Bean LJ explained:

In the present case, the ET considered that the matters complained of were not to be seen through the prism of Williams but were closer to the circumstances of Chief Constable of Gwent Police v Parsons and anor UKEAT/0143/18 (His Honour Judge Shanks presiding). In that case, the respondent police force ran a voluntary exit scheme, open to all officers, under which a compensation lump sum was payable. The claimants, who were both disabled for the purposes of the EqA, were entitled at the point of termination of employment to a immediate deferred pension under a separate scheme. The claimants having made successful applications under the voluntary exit scheme, a decision was then taken to apply a cap to the compensation lump sum in the case of employees entitled to an immediate deferred pension. The EAT upheld the ET’s finding that this constituted “unfavourable treatment”, thereby rejecting the Chief Constable’s argument that the effect of the two schemes should be considered together; on this point, the EAT held that the ET had been entitled to consider the two schemes separately given that they had separate purposes and were set up at different times. The EAT further held that Williams was distinguishable because it had there been decided that:

Section 19 EqA defines indirect discrimination, as follows :

It is helpful at this point to refer to the case of Barclays Bank plc v Kapur and others (no. 2) [1995] IRLR 87 CA, determined under the earlier provisions of the Race Relations Act 1976. Although, in relation to indirect discrimination, the decision in Kapur was strictly concerned with the term “detriment”, rather than disadvantage, this case is relied on by the respondent as providing a useful analogy when considering both the concept of “unfavourable treatment” (section 15) and “disadvantage” (section 19)).

In Kapur, the claimants (of Asian origin) were required to waive their pension rights accrued in Kenya on commencing employment in the United Kingdom because they had already been compensated for the loss of those rights through ex gratia payments (see paragraphs 13-14 of the Judgment). Other employees (predominantly of white European origin) were not required to waive their pension rights, but did not receive an ex gratia payment (paragraph 20). The claimants’ claims of direct and indirect race discrimination ultimately failed. In respect of the latter, the Court of Appeal (see paragraph 43) endorsed the view of the EAT (which had allowed the bank’s appeal against the decision of the ET) in holding that the claimants had suffered no detriment:

As for the requirement that the disadvantage be shared by others with the relevant protected characteristic (commonly referred to as group disadvantage), it has been held that a mechanistic approach should not be adopted when assessing the sufficiency of disparity; as the EAT (Cox J presiding) explained in Ministry of Defence v DeBique [2010] IRLR 471:

In the present case, it was not in dispute that there may be cases where the relevant disadvantage is sufficiently notorious that specific evidence (statistical or otherwise) is unnecessary. In relation to the protected characteristic of sex, judicial notice is thus taken of the disadvantages suffered by women arising from the disproportionate burden placed upon them in respect of caring responsibilities. The point was explained in Dobson v North Cumbria Integrated Care NHS Foundation Trust [2021] ICR 1699, EAT (Choudhury P presiding) in relation to child care responsibilities:

The relevance of such an established disparity will, however, depend on the particular circumstances of the case and the EAT in Dobson went on to warn that group disadvantage may not be established in such cases if that is not demonstrated in relation to the particular PCP in question:

When determining whether the PCP had put any particular claimant at the particular disadvantage identified (section 19(2)(c)), the ET had to consider where there was a causal link between the PCP and the disadvantage suffered by the claimant in question (see Essop and ors v Home Office (UK Border Agency) [2017] UKSC 39; [2017] ICR 640 SC, at paragraph 32). Where individual disadvantage has been demonstrated, however, that might provide support for the contention that there is group disadvantage, although the latter is not inextricably linked to the former; see Dobson at paragraph 55, where it was observed:

Where a claim under the EqA is upheld by an ET, a respondent may be ordered to pay compensation (see section 124(2)(b)), which can include an award in respect of injury to feelings. It is common ground before us that a finding of discrimination does not automatically lead to an award for injury to feelings; the injury must be proved, see Ministry of Defence v Cannock and ors [1994] ICR 918 EAT (Morison J at paragraph 4(2)).

That said, in Murray v Powertech (Scotland) Ltd [1992] IRLR 257 EAT (Lord Mayfield presiding) had opined that:

For the claimants it is submitted that the threshold for establishing injury to feelings should not be set too high, consistent with the principles laid down in Her Majesty’s Prison Service v Johnson [1997] IRLR 162 EAT. The respondent counters, however, that the ET was entitled to take into account the fact that this was not a case where the discrimination in question was personally directed at the claimants and drew a legitimate distinction between harbouring a legitimate and principled sense of grievance and suffering injury to feelings (see Moyhing v Barts and London NHS Trust [2006] IRLR 860 EAT at paragraph 27).

By the respondent’s appeal it is contended that, in determining the claim under section 15 EqA, the ET erred in law in: (a) its identification of the relevant treatment; and (b) in finding that the claimants were subjected to unfavourable treatment. In resisting the claimants’ appeal against rejection of their claims of indirect sex discrimination under section 19 EqA, however, the respondent seeks (so far as it is necessary) to uphold the ET’s decision on alternative grounds, namely that it erred in finding (i) that the preconditions for special paid leave could amount to a “particular disadvantage”, and/or (ii) that the individual claimants had suffered that disadvantage. The arguments raised in respect of the terms in issue – “unfavourable” and “disadvantage” – give rise to similar considerations and it is helpful to address both points at this stage. If the respondent’s arguments in these respects are successful, it is common ground that this would be determinative of the appeals.

The respondent submits that the ET erred in law in artificially and impermissibly separating certain of the preconditions for obtaining paid special leave from the (“clearly favourable”; ET paragraph 313) granting of that leave itself:

The claimants complied with the preconditions solely to obtain the advantage of indefinite, paid special leave; considering the preconditions in isolation (as the ET did) left out a vital part of the picture; this was an important point of distinction with Parsons (in Parsons (assuming that case had been correctly determined) the claimants did not accept preconditions in order to obtain an advantage and the “unfavourable treatment” was not intrinsically entwined with the substantial advantage otherwise available to them); in any event, the guidance in Williams (of higher authority) should be followed.

The claimants were only entitled to the advantage of paid special leave because they were disabled and shielding: this was an advantage conferred because of their inability to attend work, which arose from their disability (just as Mr Williams was only entitled to an early pension at the age of 38 because he was incapacitated from work).

The preconditions were not unfavourable to disabled employees when compared with others (not disabled) entitled to that leave. The ET impermissibly compared the claimants to those who had chosen to be paid overtime rather than accrue TOIL, or who had taken accrued leave before the pandemic, but that was a comparison on all fours with that rejected in Williams; the claimants were not subjected to unfavourable treatment simply because others entitled to paid special leave (including other disabled employees) may have been treated more favourably. Similarly, in comparing the treatment to the hypothetical possibility of taking paid special leave while retaining accrued TOIL or leave, the ET wrongly assumed a more favourable situation that did not exist (just as no employee in Kapur was entitled to both the ex gratia compensation and accrued pension rights; and just as Mr Williams could not compare his pension with that of an imaginary non-disabled employee entitled to an immediate pension based on full-time earnings at the age of 38).

The respondent says the ET therefore erred in law in mis-identifying the relevant treatment and in finding that the claimants were subjected to unfavourable treatment; had it correctly identified the relevant treatment, it could only have found that it was advantageous, and thus not unfavourable.

In finding the treatment of the claimant to amount to a disadvantage for the purposes of section 19 EqA, the respondent contends that the ET fell into the same errors as it had when determining the question of unfavourable treatment under section 15.

In addition, the respondent submits that the ET erred in concluding that the individual claimants had been disadvantaged by having accrued TOIL removed or by being compelled to use annual leave when they did not wish to do so. In Ms McCrone’s case, she had not had any accrued TOIL removed and although she had used some of her annual leave to cover child care, given the other alternatives available as a result of the flexible arrangements put in place by the respondent, it was incumbent on the ET to find whether Ms McCrone had in fact been put at the disadvantage/s alleged because of the PCPs. Moreover, as the ET had not heard from the other claimants, it was impossible to explore whether they had in fact been placed at the disadvantages relied on.

It is the claimants’ contention that the respondent’s appeal is contingent on a finding that the ET adopted a flawed approach to the correct identification of “treatment”.

The claimants submit, however, that the ET gave careful consideration to the question of what the relevant treatment was for the purposes of the section 15 claims. In Williams, criticism had been made of the attempt to artificially separate the method of calculation of a pension from the overall award to which the calculation gave rise; the same criticism could not be applied in this case. The ET had been entitled to come to the conclusion that the preconditions for paid special leave could be identified as “treatment” in their own right, rather than looking at the paid special leave scheme as a whole. Moreover, the ET had properly considered the distinction drawn by the EAT in Parsons, between the facts of that case and those of Williams, and had permissibly concluded that neither case was precisely on point, proceeding (correctly) to apply a common-sense interpretation of the statutory wording.

The key difference between the current case and Williams was that in this case the claimants had to exhaust their annual leave and TOIL (the unfavourable treatment) before they were entitled to paid special leave (the favourable treatment); Mr Williams, however, already had access to the award of a pension before the attempt to draw a distinction between that and the calculation of the pension.

The claimants further submit that the ET had correctly identified that the preconditions complained of by the claimants could be identified as “treatment” separate to the granting of paid special leave (something it had expressly acknowledged to be “clearly favourable treatment” (ET paragraph 313)), before then going on to assess whether that treatment was unfavourable. For completeness, it was the claimants’ case that the guidance in Williams at paragraph 27 confirmed the relatively low threshold sufficient to trigger a requirement to justify treatment; in the present case, the application of the paid special leave policy as a whole was unfavourable and the ET’s detailed factual assessment of why this was the case was correct.

Just as the ET had been correct in its finding of unfavourable treatment for section 15 purposes, the claimants contended it had permissibly found that particular disadvantage (for which there was, again, a low threshold) had been shown for the purposes of section 19 EqA (see points made above).

As for individual disadvantage, there had been evidence before the ET that the claimants had been required to take leave while their husbands or partners continued to work (that material having been part of the collective grievance) and, unlike Kapur, this was not a case where the ET had found that they would benefit from a double recovery such that their grievances were unjustified. More specifically, it was not clear that it was part of the respondent’s case below that Ms McCrone could have avoided the particular disadvantage she was found to have suffered. In any event, the ET had found that the claimants had been subjected to these arrangements and the EAT should be reluctant to interfere with that finding.

The claimants’ appeal in this regard centres on the guidance provided in relation to the application of the “child care disparity” in Dobson. It is the claimants’ case that key passages of the ET’s reasoning in relation to group disadvantage demonstrated a flawed approach in relation to how the issue of the child care disparity should have been approached in relation to the PCP in this case. In particular, at paragraph 397, the ET had observed that:

The claimants submit that the ET’s approach was erroneous and that the nature of the PCP adopted by the respondent was one that was (taking into account the child care disparity) inherently more likely to cause detriment (DeBique):

It was not controversial that the coronavirus pandemic created (via closure of schools/nurseries and/or disruption to informal child care arrangements) an additional child care burden on a significant number of families across the UK.

The increased child care burden caused by the pandemic was more likely to fall disproportionately on female employees than their male counterparts (the child care disparity) – there was no basis for considering that the pandemic would do anything other than exacerbate the existing child care disparity.

Due to that increased child care burden, the respondent’s female employees were more likely to need to access the paid special leave policy than its male employees and, therefore, more likely to be required to exhaust their accrued TOIL and/or annual leave, placing them at a disadvantage when compared to their male counterparts

The claimants further submit that the finding by the ET that the individual claimants were placed at the particular disadvantage caused by the PCP in this case undermined the finding that there was no group disadvantage (per the guidance in Dobson at paragraph 55).

For the respondents it is contended that, on a proper reading of the ET’s reasoning, it had acknowledged and taken into account the child care disparity (per Dobson), but had then gone on (correctly) to consider the interrelationship between that disparity and the particular PCP in issue. This was not a case where a finding of group disadvantage would be highly likely to follow from taking judicial notice of the child care disparity and it was entirely appropriate for the ET to interrogate whether, in the circumstances of an unprecedented global pandemic in which employees might be required to take leave to cover child care in their normal working hours, more female than male employees would need to rely on the paid special leave policy, and thus use TOIL or accrued leave.

Having correctly identified and applied the Dobson principles, the ET legitimately concluded that the PCP in this case was not inherently likely to give rise to group disadvantage, and that the claimants had failed to adduce any evidence supporting such a finding. In the absence of such evidence, the ET properly found that group disadvantage was not made out; the claimants’ appeal did not (and could not) challenge that analysis as perverse.

Accepting that a finding of discrimination cannot automatically lead to an award for injury to feelings (per MoD v Cannock), the claimants contend that the decision in Murray v Powertech made clear that the threshold for establishing injury to feelings was not a particularly high one, it was indicated in that case that a bare assertion of injury to feelings might be sufficient for the ET to then go on to consider the degree of injury. That was in line with the principles in HMPS v Johnson: (a) awards should not be too low, as that would diminish respect for the policy of the anti-discrimination legislation; and (b) society has condemned discrimination and awards must ensure that it is seen to be wrong

In this case the ET accepted the evidence of Mr Cowie that it was “not very nice” to have to use his TOIL before accessing paid special leave; it had placed too high a burden on the claimants when determining that this was insufficient to demonstrate injury to feelings as a result of being discriminated against.

For the respondents, it is contended that the ET’s approach gave rise to no error of law (the possibility of there being no award for injury to feelings was allowed in MoD v Cannock) and was consistent with both Murray v Powertech and HMPS v Johnson. In Murray v Powertech, the error was in finding there was no evidence that the claimant (dismissed because she was pregnant) had suffered hurt feelings, in circumstances where she was shocked by her dismissal and had been upset and tearful during the hearing. Even then, however, the case was remitted so that it could be determined whether any award should be made; the EAT did not rule that an award must be made.

In the present case, only Mr Cowie had given evidence of the effect of the discriminatory treatment. The ET analysed that evidence, in the context of the factual matrix it had found, and concluded that any upset was not connected with the discriminatory treatment. As for his view that the treatment was “not very nice”, the ET permissibly concluded that did not evidence an injury to feelings. Even if (which the respondent disputed), the ET erred in its approach to Mr Cowie’s case, there was no basis for saying that any such error affected its findings in relation to the other claimants.

As is common ground, if the respondent is correct in its contention that, properly understood, there was no unfavourable treatment and the claimants suffered no disadvantage (the respondent contending, on the contrary, that the “treatment” was both favourable and advantageous) that would dispose of all matters before us; the claims under both section 15 and section 19 EqA would necessarily fail. Although there can be a danger in attributing the same meaning to different words used in a statutory context, it has not been suggested to us that there is any distinction to be drawn between “unfavourable treatment” (for the purposes of section 15 EqA) and “disadvantage” (as used in section 19 EqA) and we have proceeded on the basis that broadly the same approach can be adopted to both terms (mindful of Lord Carnwath’s observation at paragraph 27 in Williams, that little is likely to be gained by seeking to draw narrow distinctions between “unfavourable” and “analogous concepts such as ‘disadvantage’ or ‘detriment’”). For convenience, however, we have started by considering the position in relation to “unfavourable treatment” under section 15 EqA.

As the ET noted, in Williams it was held that there was a relatively low threshold which would be sufficient to trigger the requirement to justify under section 15. In order to consider whether that threshold has been crossed, however, an ET must first answer two questions of fact: (1) what was the relevant treatment? (2) was it unfavourable to the claimants?, see per Lord Carnwath in Williams at paragraph 12. In this case, the crucial question was that posed at (1), what was the relevant treatment?

The respondent had argued that the treatment in question was in fact the granting of paid special leave (characterised as paying for a worker to stay at home and not to carry out any work). That, the ET held, was “clearly favourable treatment” (ET paragraph 313); it did not, however, consider that was the treatment in issue in this case. Rather, the ET found that the “treatment” for section 15 purposes comprised the acts complained of by the claimants, essentially: having to use their accrued TOIL and annual leave at a time not of their choosing. The issue at the heart of the respondent’s appeal is whether the ET erred in its approach in this regard.

If the requirement to use TOIL and annual leave was properly to be seen as separate from the claimants’ entitlement to paid special leave, we can see how this might have been viewed as more akin to the factual matrix of Parsons than that of Williams (ET paragraph 322), and why the ET would have been entitled to conclude that such treatment was “unfavourable” (we acknowledge that preventing employees from taking TOIL and annual leave at a time other than that of their choosing will generally be “unfavourable”). The question is whether that was the correct way of defining the treatment in the present case. In Parsons, the claimants were complaining of unfavourable treatment in the application of the voluntary exit scheme whereby the compensatory lump sum was reduced (the unfavourable treatment) because of something (the claimants’ entitlement to an immediate deferred pension) arising in consequence of their disabilities. Although the unfavourable treatment arose in the context of a benefit being received by the claimants (the pension), that was not a necessary precondition or consequence of the voluntary exit scheme for which the claimants had applied: the claimants had not accepted that precondition in order to obtain the benefit of the voluntary exit scheme and the unfavourable treatment was not intrinsically entwined with the advantage the claimants would otherwise receive in respect of their pensions. Referring to the separate purposes and histories of the two schemes in issue, the EAT thus concluded that the ET had been entitled to find as a fact that they were not to be considered together (one as a condition or consequence of the other).

Although we can acknowledge that the loss of flexibility and choice in terms of when to take accrued TOIL and annual leave could constitute unfavourable treatment in general terms, the difficulty that arises in the present case is that there was no general requirement on the claimants to use TOIL and/or leave at a time of the respondent’s choosing; rather, the specific requirement to exhaust any accrued TOIL and/or leave arose only when, and to the extent that, the claimants sought to access paid special leave. It would be artificial to consider the requirement to use accrued TOIL and/or annual leave separately from the entitlement to paid special leave because the two were thus inextricably linked.

The error made by the ET was, in our judgement, to effectively approach its assessment of the “treatment” in this case as if this was to be defined by the claimants’ complaint. The claimants may have complained of the preconditions that had been imposed, but the ET was wrong to focus solely on the acts thus identified rather than having regard to the factual matrix it had itself found, which included its findings of fact that the acts complained of by the claimants were “preconditions to obtaining or consequences of paid special leave”, which it had held to be “clearly favourable” (ET paragraph 313). This error, we note, continued through the ET’s assessment of the constituent elements of section 15; hence the “something” arising in consequence of the claimants’ disabilities was not the use of the paid special leave policy (as the ET concluded, accepting the way the claimants had put their case in this regard, see paragraphs 335-336) but the claimants’ inability to attend work. The treatment meted out by the respondent because of that inability to attend work (the “something”) took the form (relevantly) of the paid special leave policy. As was not in dispute, that policy was favourable to the claimants, providing them with an entitlement to paid leave on an indefinite basis. Although the policy was subject to conditions for entitlement (the prior use of accrued TOIL/annual leave) that could not detract from the favourable nature of that treatment.

Viewing the facts of this case with the guidance in Williams in mind, similar points can be made to those identified as relevant in that case. The section 15 claimants were granted an entitlement to paid special leave during the periods of time they were unable to work due to their disabilities; that was an advantage they would otherwise not have enjoyed during those periods of absence. The claimants complained of the conditions of entitlement to that paid special leave but those conditions could not be viewed in isolation from the benefit thus provided: the conditions in question were only applied because the claimants were being granted an entitlement to paid special leave and it would be wholly artificial to separate out the two elements, the benefit and the conditions of accessing that benefit. Moreover, although the ET spent some time comparing the treatment of the claimants to how the position might have been had different terms of entitlement applied (viewed solely in terms of the preconditions imposed for entitlement to paid special leave), that was not the question: no doubt the advantage provided by the paid special leave policy could have been improved by removing the preconditions for entitlement but it did not amount to “unfavourable treatment” by virtue of the fact that it could, hypothetically, have been even more favourable. Similarly, the “clearly favourable” treatment provided by the paid special leave policy did not become “unfavourable” by reason of the fact that some beneficiaries (including those in the same position as the claimants) might not have had to give up any accrued TOIL or annual leave: the preconditions of entitlement were the same for all potential beneficiaries and the claimants suffered no disadvantage because of something arising in consequence of their disabilities.

On the basis of the facts found by the ET, we therefore conclude that this was a case falling within the analysis provided in Williams: the claimants were complaining of the conditions of entitlement to the favourable treatment extended to them under the paid special leave policy. Viewed in that context, there was no unfavourable treatment for the purpose of section 15 EqA. The error made by the ET was to allow the claimants’ complaint to define its assessment of “treatment” and thus to artificially separate out the conditions of entitlement to a benefit from the benefit itself.

Turning to the ET’s decision in relation to the section 19 claims, the same error arises. The PCP was defined as the requirement to use accrued TOIL and/or annual leave before being granted paid special leave, but that was not determinative of the question of “disadvantage”. The PCP only operated in the context of the paid special leave policy – there was no requirement to use accrued TOIL and/or annual leave other than as part of that policy – and the question of disadvantage thus had to be viewed through the operation of the policy as a whole. Having concluded that the provision of paid special leave was “clearly favourable”, the ET could only find that the particular PCP relied on by the claimants amounted to a “disadvantage” if it then separated out the conditions of entitlement to the benefit from the benefit itself. That, however, would be a wholly artificial separation between these two elements of the paid special leave policy. Thus, for essentially the same reasons as we have explained in relation to the section 15 claims, we consider that the ET erred in its approach to the question of “disadvantage” for section 19 purposes.

Given our conclusions on the questions of “unfavourable treatment” and “disadvantage”, the issues raised by the claimants’ appeal do not strictly arise: (1) our conclusion on the section 15 claims means the ET’s liability decision in this regard must be set aside and the remedy appeal is thus rendered academic; (2) the view we have formed in relation to the ET’s approach to the section 19 claims would mean that its decision in this respect would be upheld on that alternative basis, even if the claimants were successful on their group disadvantage appeal. For completeness, however, we have gone on to set out our findings on the claimants’ appeal, below.

Adopting the ET’s analysis of “disadvantage”, we do not consider it can be said that the ET failed to have regard to the issue of the child care disparity identified in Dobson. Indeed, the ET’s reasoning demonstrates a careful consideration of the guidance provided in Dobson, allowing that judicial notice could be taken of the fact that women ordinarily carry greater child care responsibilities than men, but also recognising that it was necessary to consider whether that disparity gave rise to a particular disadvantage given the PCP in issue in this case.

Adopting that approach (not itself criticised as amounting to an error of law) the ET correctly focused on the disadvantages it had found to arise from the PCP; namely, the removal of TOIL and being compelled to use annual leave at a time the claimants did not wish. As the evidence demonstrated that both men and women had accrued similar TOIL balances (ET paragraph 396), there was no particular disadvantage arising simply from the requirement that TOIL be exhausted before employees would be entitled to paid special leave. The ET therefore considered whether those accessing the special leave policy (and thus more likely to face the disadvantages it had found) were disproportionately female; it found, however, that there was no evidential basis for this conclusion.

The claimants identify particular observations made by the ET in this part of its reasoning as demonstrating a flawed approach to the impact of the child care disparity in this regard. We can see the force of these points and it may be that we would have been swayed by the claimants’ submission to the effect that there was no basis for considering that the pandemic would do anything other than exacerbate the existing child care disparity. Adopting a more holistic approach to the ET’s reasoning, however, we consider that the decision reflects a more nuanced assessment of the evidence than the claimants’ criticisms allow; ultimately we consider that this was an assessment of the evidence that fell within the remit of the ET and cannot be susceptible to interference on appeal.

First, we note that the ET’s reasoning was informed by the fact that the paid special leave policy did not apply solely to those who faced child care difficulties but also extended to other categories of employee, most obviously those who were disabled (such as the other claimants in these proceedings) or who were living with others who had to shield. No data had been adduced to demonstrate whether men or women were more likely to use the paid special leave policy, or for what reasons, and the ET permissibly considered that it was possible that the impact of the measure in fact affected men just as much as women (ET paragraph 399).

Secondly, the ET appropriately considered whether the impact of the pandemic might have adversely impacted more women than men given the breakdown in normal child care arrangements that then took place. It was in this regard that the ET made the observations of which the claimants complain. The crucial part of its reasoning, however, demonstrates that it did not reject the claimants’ proposition in this regard but, allowing for the general child care disparity, considered it could not simply assume, absent any evidential basis, that women more than men would then have to use the paid special leave policy and thereby suffer the disadvantages found. Notwithstanding our sympathy with the claimants’ arguments, we cannot say that the ET thereby erred, either in its approach or its conclusion, on this point.

Finally, as for the potential inference that it is said that the ET might have drawn from its findings of individual disadvantage, we note the caution expressed at paragraph 55 of Dobson in this respect. Although, contrary to the respondent’s arguments, we do not consider the ET erred in its approach to the assessment of the individual cases (when considering individual disadvantage at the liability stage, the ET was entitled to take a broad approach to the evidence), neither can it be said that its findings were such that any meaningful conclusions about the group picture might be drawn from any individual disadvantage thus found. The ET’s findings on individual disadvantage did not answer the question of disparate impact in the application of the paid special leave policy, or of the conditions of its application that were the subject of the claimants’ complaint. Equally, the very limited evidence adduced in respect of the individual cases did not permit the ET to further extrapolate as to the impact of the pandemic on men and women in the workplace.

For those reasons, we would dismiss the claimants’ appeal against the ET’s decision on the claims of indirect sex discrimination under section 19 EqA.

The short answer to the claimants’ appeal in this regard is that the ET was entitled to find that they had failed to adduce evidence of any injury to feelings such as would warrant an award under this head. As the claimants accept, a finding of discrimination does not automatically lead to an award for injury to feelings: the injury has to be established on the evidence (MoD v Cannock). While we would not go so far as to say that it could never be appropriate to make such an award absent hearing from the claimant (we can envisage cases where it might not be possible to hear from the claimant but where the evidence would permit the ET to find they had suffered an injury to their feelings that ought to lead to an award under this head), we cannot say the ET erred in taking the view that it would not be right to do so in this case. As in Moyhing, this was not a case where the discrimination found by the ET was personally directed at the claimants and whilst they might have harboured a principled sense of grievance, the ET was entitled to find that this was something other than an injury to feelings.

In relation to the one claimant who did give evidence, Mr Cowie, the ET obviously had the benefit of hearing his evidence and was in the best position to assess what had given rise to his “upset” (the best evidence of any injury to feelings), concluding that related to the consequences of the pandemic rather than his treatment by the respondent. As for his evidence that his treatment was “not very nice”, the ET permissibly concluded that was insufficient to evidence hurt feelings (as opposed to a sense of grievance).

In reaching these conclusions, we do not consider the ET fell into the error identified in Murray v Powertech. In that case, the decision itself recorded various matters that might be considered to evidence injury to feelings; the EAT did not hold that that was bound to be the case but remitted the issue to be determined taking into account those matters. For the claimants it is said that the failure to make an award could, however, be inconsistent with the principles laid down in HMPS v Johnson. We disagree. Where awards are made in discrimination cases it is right that they should not be set at too low a level, such as would diminish respect for the policy of the legislative protection; equally, however, awards that are unjustified on the evidence would not further the legislative aims. As the EAT observed at paragraph 27 HMPS v Johnson:

Where no award for injury to feelings is made because there is no evidence of the relevant hurt, that does not undermine the purpose of the anti-discrimination legislation; it demonstrates only that compensation will not be awarded where there is no evidence of such an injury. The ET in this case was entitled to reach that conclusion.

For the reasons provided:

The respondent’s appeal is allowed and the ET’s decision upholding the section 15 EqA claims is set aside and replaced by a Judgment that those claims be dismissed.

The claimants’ appeal is dismissed.

To the extent it would have been necessary to do so, the ET’s dismissal of the claimants’ claims of indirect sex discrimination would, in any event, have been upheld on the alternative ground that there was no particular disadvantage for the purposes of section 19 EqA.