J McAllister v Commissioners for HMRC

This appeal relates to the approach to be adopted to the assessment of objective justification for the purposes of section 15 Equality Act 2010 (" EqA "). There is also a cross-appeal which addresses the question of unfavourable treatment in a section 15 claim.

In this judgment, I refer to the parties as the claimant and respondent, as below. This is the full hearing of the claimant's appeal and the respondent's cross-appeal against a reserved judgment of the Manchester Employment Tribunal (Employment Judge Hoey, sitting with lay members Mr Haydock and Mrs Humphreys on 6-10 January and 12 March 2020, with four further days of deliberations; "the ET"), sent to the parties on 11 May 2020. By that judgment, the ET rejected the claimant's claims that his dismissal was unfair and amounted to an act of disability discrimination contrary to section 15 EqA , but upheld his section 15 EqA claim in respect of the 50% reduction to his payment under the Civil Service Compensation Scheme (“CSCS”), although stating there was no breach of section 15 arising from the subsequent decision on appeal to reduce the payment by 20%.

The claimant has permission to appeal against both the ET's decisions under section 15 EqA and against its decision on unfair dismissal, to the extent that any errors in its reasoning on the section 15 claim on dismissal impact on that decision. For its part, the respondent contends that the ET's decisions should be upheld save that it cross-appeals the finding that there was any unfavourable treatment for section 15 purposes in respect of the calculation of the payment made to the claimant under the CSCS.

Mr Campion of counsel appears for the claimant on this appeal, as he did below. Before the ET, the respondent was represented by counsel but not by Mr Purchase QC, who now represents its interests.

The respondent is a public body responsible for the collection of tax in the United Kingdom. The claimant started his employment with the respondent on 3 May 2011. At the material time, he was working as an administrative officer in the respondent's Personal Tax Operations Unit ("PT Ops") in Manchester. The respondent had around 6,000 staff based in Manchester in a number of departments including PT Ops, Fraud Investigation Services, Risk Investigation and Talent. There were shared resources but the departments were different in terms of job roles, recruitment, culture and skills.

The claimant's contract of employment provided that he would be entitled to sick pay at a full rate of pay for six months and then half pay for a further six. It was also stated that poor attendance would be managed in accordance with the respondent's attendance management procedure, which could potentially lead to dismissal, albeit managers were to treat each case on its own facts.

Before the ET, it was accepted that, by the time of the relevant decisions in these proceedings, the claimant was a disabled person for the purposes of the EqA by reason of his suffering from anxiety and depression. It seems that the claimant experienced his first episode of anxiety and depression in 2013, being then signed off work for around two months. From December 2016 to March 2017, he was again absent by reason of stress. On 11 January 2017, an Occupational Health report was obtained, which stated that he was reporting ongoing stress, anxiety and depression triggered by work issues. A further report was obtained on 13 March 2017 and the claimant was able to return to work on a phased basis on 18 March 2017.

The claimant's level of absence meant that he was invited to a formal attendance meeting on 18 April 2017 and was issued with a first written improvement warning, which he did not appeal. Adjustments were put in place for the claimant at work, which he confirmed were satisfactory.

Thereafter, the claimant had various further short periods of absence unrelated to anxiety and depression. On 16 November 2017, he met with his line manager, Ms Williams, to discuss his sickness absences and was reminded of the trigger points under the attendance management procedure. After yet further absences, the claimant was invited to a meeting at which it was likely he would be given a final written warning. He was, however, unable to make that meeting as, from 29 May to 15 July 2018, he was absent from work for a more sustained period with a condition that saw him admitted to hospital, although again this was not an absence that was stress-related or connected to any earlier period of sick leave. During this period of absence, the claimant's managers kept in contact, again keeping him informed as to how the attendance management procedure was to be applied.

Attending a meeting with Ms Williams on 25 June 2018, the claimant asked whether he would be able to attend summer camp that year (this was a paid but voluntary activity that the claimant had undertaken each year for the previous four years). Concerned that this might not be in his best interest given his health issues, Ms Williams responded, "no for the moment."

On 29 June 2018, the claimant attended a sickness absence meeting, where he confirmed that there was nothing further that the respondent could do to facilitate his attendance. The claimant was advised that if the time came when the respondent considered it could no longer support his absences his case would be referred to a decision-maker to determine the appropriate action; it was confirmed, however, that that stage had not been reached and measures would be put in place to support his return to work on a phased basis, with additional adjustments.

On his return to work on 16 July 2018, the claimant met with Ms Williams and raised the issue about his attendance at summer camp. Ms Williams said she had reached a decision that this request should be refused as she did not feel the claimant would physically be able to attend and was worried about the impact on his health. The claimant took that decision badly and responded in a bad-tempered way. It was his case that, effectively as a result of how he had been treated in relation to that issue, he suffered a breakdown, which set in train the events that ultimately led to his dismissal.

On 18 July 2018, the claimant was late for work. He reported having been crying for most of the night and said he planned to resign. He was advised to speak with a wellbeing advocate and not to make any hasty decisions. The claimant left work early that day.

On 19 July 2018, the claimant attended a meeting with Ms Williams to update his fit-for-work plan. Although the plan was agreed, the claimant indicated his disagreement with Ms Williams's decision regarding the summer camp, saying he would be going in any event and that he was resigning, throwing his work ID on the table as he said this and banging the door on his way out of the meeting. Attending his GP that afternoon, the claimant was signed off work until 15 August 2018, having been diagnosed with mixed anxiety and depressive disorder. Later that evening, the claimant emailed Ms Williams to apologise for his earlier behaviour. The next day, the claimant called into work to say he had had a breakdown and did not intend to resign. It was agreed his resignation would be treated as withdrawn.

On 1 August 2018, Ms Williams held a formal meeting with the claimant, accompanied by his union representative, under the managing attendance policy. It was explained that meetings would take place each month to determine whether the respondent could continue to support the claimant's absence.

On 9 August 2018, an Occupational Health report was obtained, which confirmed the reasons for the claimant's recent period of absence and stated that the claimant believed he had had a breakdown and that he was reporting feeling very low in mood and was struggling to sleep. Looking at the prognosis, it was stated that, once the claimant was engaged with counselling and his mediation had started to work (some four to six weeks after commencement), there was no reason why he should not provide regular and effective service, albeit he was seeking a move to a different department if that could be accommodated. Although Ms Williams tried to contact the claimant to discuss this report, he did not feel able to discuss matters and, on 14 August 2018, texted to say that his mood was really low and he did not want to talk to anyone.

On 15 August 2018, the claimant spoke to Ms Williams to say that, although his GP's note had expired, he would not be returning to work as he was not fit to do so. Ms Williams advised that the business could no longer sustain the claimant's absence and his case would be referred to a decision-maker to consider next steps, which could include dismissal. Subsequently taking HR advice, however, Ms Williams decided to delay any referral to a decision-maker pending consideration as to whether ill-health retirement might be an option. She also considered that more time should be allowed to see how the claimant's position progressed given the advice relating to his medication and therapy. Ms Williams accordingly advised the claimant of her revised view on 22 August 2018, when he informed her that he had a further four-week fit note from his GP extending his absence to 14 September 2018.

A continuous absence meeting was held with the claimant and his union representative on 5 September 2018. This was to explore whether a successful return to work could be achieved within a reasonable period of time. The claimant explained that his moods could change quickly and he would be seeing his doctor when his fit note was due to expire. Other than changing departments, the claimant was unable to suggest any further reasonable adjustments; the respondent made clear that it would be for the claimant to apply should he wish to transfer roles.

Ms Williams continued to try to keep in touch with the claimant but he was not always able to answer her calls. When she managed to speak to him on 17 September 2018, he said he was not sure he would be able to return to work and was not then fit to work in any capacity. Ms Williams advised the claimant that the matter would be referred to a decision-maker with a recommendation of dismissal as the business could no longer support his absence. When this was confirmed to the claimant in a further call on 19 September 2018, he responded that he was "not surprised".

During September 2018, the claimant was asked to agree a stress risk assessment drafted by Ms Williams. There were, however, disagreements between them relating to the claimant's conduct at the earlier meeting regarding his attendance at summer camp. More generally, there were various occasions during this period when the claimant did not answer Ms Williams's calls and failed to complete or return the documentation he was sent relevant to an application for workplace injury benefit. As the ET found, this was largely due to the claimant's mental health issues and the effects of the medication he was prescribed.

On 4 October 2018, the decision-maker, Mr Khan, wrote to the claimant to invite him to a meeting to discuss whether he should be dismissed or his absence supported. This meeting ultimately took place on 19 October 2018 at the claimant's home; the claimant chose not to have a union representative present. It was common ground that, at the end of the sustained improvement period under the attendance management policy, the claimant's absence would total 99 days, increasing to 107 days by the end of the month. The claimant explained he had just started his third course of new medication and had attended 9 of the proposed 12 therapy sessions. He stated he had had a breakdown in the office on 19 July 2018 and was not ready to return. He considered the office was "a bad place to work" and wanted to move to a different directorate as PT Ops was "a horrible place to work". In cross-examination at trial, the claimant accepted he was still struggling with sleep and mood swings at this time and would not have been fit to return to work even if a new role had been offered to him.

On 30 October 2018, the claimant again contacted his GP and was declared unfit for work until 12 November 2018.

Having considered the information before him, Mr Khan concluded that the claimant's absence could no longer be supported and he should be given notice of dismissal. This decision was communicated to the claimant by a letter of 2 November 2018, giving notice of dismissal to take effect on 28 December 2018.

In explaining why he had reached his decision, Mr Khan acknowledged there was no specific evidence or data to consider; he was only able to estimate the impact of the claimant's absence using his experience. There was no backfilling of the claimant's role and other employees were dealing with the same amount of business as before; customers were likely to have to wait longer to have their queries dealt with and, as the number of calls to the claimant's team had remained the same, those left would have had to take up the slack. Mr Khan concluded that the claimant’s absence impacted upon productivity and morale and he noted that Ms Williams was having to spend a considerable amount of time managing and supporting the claimant rather than supporting the rest of the team. He also took into account that all reasonable adjustments had been exhausted and there was no business case to support a transfer to another team (not least as there were a limited number of roles and recruitment was competitive, and other teams had different ways of working, which included customer contact, which the claimant had highlighted as a difficulty); in any event, the claimant would need to be ready to return to work, which he was not.

The claimant appealed against Mr Khan's decision but his appeal, heard by a Ms Foxley, was dismissed after a hearing on 6 December 2018, at which the claimant was accompanied by his union representative.

As the claimant had been dismissed for reasons of capability, Ms Williams completed an application in respect of the CSCS. The CSCS, which is made under section 1 of the Superannuation Act 1972 , provides for the payment of compensation in circumstances in which a civil servant is dismissed for what is called "inefficiency" and "the employing department decides that payment of compensation would be appropriate". The respondent had published guidance as to how it would calculate any payment it chose to make under the CSCS, which made clear that such payments are considered where employees are dismissed for "unsatisfactory attendance" and that:

There is a discretion whether to award 100% of the sum fixed under the CSCS or a lesser proportion and the guidance provides a table showing what is needed for 100%, 75%, 50%, 25%, and 0% of the relevant sum. Relevantly, 100% is payable where the employee has cooperated with all measures to improve attendance and has been proactive in seeking solutions, kept in touch throughout their absence, shown a positive attitude and full commitment to work and cooperated with Occupational Health and followed all advice; 75% is payable where the employee has done most of what is needed for 100%; around 50% is payable where there has been some proactivity and the employee has kept in touch some of the time and a fair amount of commitment has been shown and a fair number of reasonable adjustments have been achieved.

In the claimant's case, Ms Williams recommended an award of 50% on the basis that, although he had cooperated to some extent with efforts to support his return to work, he had not always been available for agreed contact and had had to be chased for completed forms. He had also been late to work during his phased return and his behaviour had been disruptive during the four days when he did return, although he had subsequently apologised for his behaviour. The 50% recommendation was approved.

The claimant appealed against that decision and, on 28 March 2019, the Civil Service Appeal Board ("the CSAB") decided to increase the award from 50% to 80% because it was decided that the rationale for the 50% payment missed any assessment of the impact of the claimant's proven mental health condition on the situation in which he found himself, which had impacted upon his ability to cope with the events facing him.

It was common ground before the ET that the claimant's dismissal was unfavourable treatment. As the ET was satisfied that the claimant's later, disability-related absences from work had led to his dismissal, the unfavourable treatment was accordingly due to something (the claimant's absences) arising in consequence of his disability. The respondent had said, however, that any unfavourable treatment was a proportionate means of achieving a legitimate aim, namely: (i) to ensure staff were capable of demonstrating satisfactory attendance and a good standard of attendance (comprising the aims of the maintenance of a fair, effective and transparent sickness management regime and efficient use of resources), (ii) to provide a good customer service, and (iii) to apply the respondent's policies fairly and consistently.

The ET accepted these were legitimate aims which were legal and not discriminatory in themselves. It further accepted that the aims were rationally connected to the dismissal such that dismissal was potentially capable of achieving those aims.

Turning to the question of whether the aims were proportionally achieved by dismissal, the ET took into account the discriminatory effect of the treatment (the loss of the claimant's job and career with the respondent) as against the respondent's reasons for applying the aims, considering whether the aims could be achieved by less discriminatory means such as by the issuing of a warning.

Undertaking this exercise in respect of each of the aims relied on, the ET accepted that the first matters identified (ensuring staff were capable of demonstrating satisfactory attendance and a good standard of attendance, comprising the aims of maintenance of a fair, effective and transparent sickness management regime and efficient use of resources) were interrelated. The ET noted that the claimant had been absent for a lengthy period of time (he had 245 days of absence over 23 periods 2016 until his dismissal, each for different reasons and periods, continuous and intermittent), that the respondent had issued a warning in 2017, and that it had sought to keep in touch and secure his return to work. This was not a case where any breach of sections 20 and 21 EqA had been suggested and the ET concluded as follows:

As for the aim of providing a good customer service, described by the respondent as "maintaining proper standards of service", the ET noted that there was little evidence before it as to the precise impact of the claimant's absence on customer service; indeed, it was the respondent's case that the nature of the work was such that such evidence was not available, but that those taking the decision to dismiss and on the appeal were experienced officers who knew what the impact of a person carrying out the claimant's role would be in circumstances where there was a team of around 10 or 11, in which there were no other staff on long-term absence. The claimant's role was not filled during his absence and the remaining staff were required to deal with the same amount of calls and business as they would have been when the claimant was present. In the circumstances, the ET accepted that the impact upon the respondent's business needed no evidence, in particular given Ms Williams's view that the claimant's absence impacted upon productivity and morale and that the respondent was unable to provide customers with the same level of service when the claimant was not at work to carry out his tasks. All that said, the ET considered it was unable to assess the precise impact the claimant's absence had on the respondent's ability to maintain customer service levels given the lack of evidence in this regard and it therefore concluded that the respondent had not shown that dismissal was a proportionate means of achieving the aim of customer service by itself.

Turning, then, to the third and subsidiary aim relied on - to apply the respondent's policies fairly and consistently - the ET accepted that the respondent had sought to be fair and consistent and had provided the claimant with a fair opportunity to improve his attendance. For similar reasons as those provided in relation to the first aim, the ET concluded that dismissal was a proportionate means of achieving this subsidiary aim.

The ET also found that the claimant's dismissal by reason of capability had been fair. Consideration had been given to what the claimant had said had caused his breakdown in July 2018 and to whether it was reasonable to wait any longer before deciding whether to dismiss. The ET found the decisions reached on these issues were within the reasonable band of responses. A reasonable employer might have waited longer, given the further change in the claimant's medicine, and/or sought a further medical report but it was also open to a reasonable employer to decide not to do so given the time that had already passed. The claimant had complained that further warning should have been given before he was dismissed but Ms Williams had believed that she could not issue a final warning when the claimant was absent and the ET concluded there was no requirement to issue warnings prior to dismissal in cases of continuous absence under the respondent's policy; the respondent had reasonably considered the claimant's absence to be continuous and had not acted unfairly in this respect. The claimant had also argued that the dismissal process should have been paused pending the outcome of his application for injury benefit, but this was not something of which the dismissing officer was aware and the claimant had not brought it to his attention or asked for the process to be paused in this respect. Although it had been known at the appeal stage, the ET did not consider that it was required for the process to be paused while that application was being considered and, in any event, given that it was not known when the application would be determined, it had been within the range of reasonable responses for the respondent not to pause the process in this case. The ET was satisfied there were no alternatives that the respondent had unreasonably failed to consider; it had adopted a reasonable process and reached a decision that was within the band of reasonable responses. The dismissal had not been unfair.

Considering the CSCS payment, the ET rejected the respondent's argument that this was analogous to the position considered by the Supreme Court in Williams v Trustees of Swansea University Pension and Assurance Scheme [2018] IRLR 306. It considered the key distinctions were twofold: (i) the fact that the entitlement in Williams (the pension) was only available to those who were permanently unable to work and were therefore disabled, and (ii) the rules as to the entitlement to the benefit were the specific rules of the pension scheme in question.

In the present case, entitlement to a compensation payment under the CSCS was available to all those who were dismissed by reason of capability, which would include disabled persons but would also extend to those who did not satisfy the definition of "disabled" under the EqA . Moreover, the applicable rules in this case differed from those in Williams , giving the respondent a discretion to determine whether to pay 100% or a lesser sum. It was that exercise of discretion that was the subject of complaint in these proceedings, it being the claimant's case that the respondent had reduced the sum that would otherwise have been payable because of something arising in consequence of the claimant's disability. Whilst in Williams there had been nothing intrinsically unfavourable or disadvantageous about the pension, in this case the ET concluded, "It is clearly intrinsically unfavourable or disadvantageous to reduce the compensation payment."

The reasons given for reducing the claimant's CSCS payment had been his not answering calls, his delay in returning relevant forms, disruptive behaviour and turning up late for work. The ET found, however, that:

Turning to the question of justification, the ET accepted that distributing the fund fairly and economically in line with the applicable guidance constituted a legitimate aim, which corresponded to a real need of the respondent to ensure public funds were used properly in line with the guidance. It was further satisfied that the measure applied (the reduction of compensation) was capable of achieving that aim. The ET did not accept, however, that the aim was proportionately achieved by reducing the payment by 50%, finding that the decision of the CSAB on the claimant’s appeal had been correct in this regard. In reaching this conclusion, the ET explained:

In its formal judgment on this aspect of the claim, the ET explained its decision in the following terms:

The claimant contends that the ET's rejection of the respondent's case as to the customer service aim (aim 2) due to lack of evidence should have been fatal to the question of objective justification; the ET had wrongly accepted the aim of achieving a satisfactory attendance and achieving a good standard of attendance (aim 1), which: (i) was a vacuous aspiration rather than a "real need"; (ii) could not justify dismissing the claimant without some underlying real need such as impact on customer service; and (iii) meant there was a lack of evidence of the reasonable needs of the respondent to weigh against "the immense impact dismissal had upon the Claimant", in particular given the ET's findings under the heading of the unfair dismissal claim that other employers might reasonably have adopted a different, less discriminatory course. Further, to the extent that this aim included applying sickness absence procedures consistently (aim 3), which was seen as a subsidiary aim, this was not of itself a legitimate aim that could form the basis of a fair dismissal as it was inherently discriminatory, most likely to disadvantage disabled workers, and would itself need to be justified based on a legitimate aim. The ET had thus erred in placing weight on the question of consistency of treatment.

Although there are different statutory tests to be applied under section 15 EqA and section 98 Employment Rights Act 1996 ("the ERA"), it would almost inevitably not be within the band of reasonable responses for an employer to effect a discriminatory dismissal: see per Underhill LJ at paragraphs 53-54 O'Brien v Bolton St Catherine's Academy [2017] ICR 737 CA.

As for the ET's approach to the reduction in the payment made to the claimant under the CSCS, while it had correctly determined that the 50% reduction was not proportionate, it had then erred in its approach to causation and remedy, wrongly asking itself whether the later decision taken by the CSAB had been proportionate. Discrimination compensation was to be calculated in the same way as damages for a tort. The correct question would have been to ask what award the respondent would have made in the absence of discrimination. More than that, the decision of the CSAB was not before the ET; it was independent of the respondent and no claim had been made or could have been made in respect of its decision.

For the respondent, it is contended that the ET had correctly directed itself as to the relevant legal principles; the claimant was impermissibly seeking to re-argue conclusions with which he disagreed (see Sullivan v Bury Street Capital Ltd [2021] EWCA Civ 1694, in particular paragraphs 39 and 42). Specifically, the ET had been entitled to find that dismissal was justified on the basis of the first aim identified and it was not open to the claimant to assert, contrary to the ET's findings of fact, that the aim was vacuous and did not correspond to a real need. In any event, it could not be said to be wrong for the ET to find it was legitimate for the respondent to aim for its employees to maintain a good standard of attendance: see Ruiz Conejero v Ferroser Servicios Auxiliares [2018] IRLR 383 CJEU. Moreover, although the ET had not accepted the customer service aim (aim 2) justified dismissal of the claimant of itself, it had accepted that there was an impact on customer service (paragraph 355 ET judgment). Not only was there evidence to support that finding (for example, at paragraphs 123, 126 and 158 of the ET’s judgment) but the ET was entitled to consider, as it did, that such impact was obvious. Similarly, the ET had accepted that there was also an impact on the respondent's undertaking more broadly, on morale and on the need to manage the claimant's absence.

As for the subsidiary aim of applying the respondent's policies fairly and consistently, the ET was entitled to find this was a legitimate aim: see per Lord Bingham at paragraph 30 R (on the application of O'Brien) v Independent Assessor [2007] 2 AC 312 and, in the context of an unfair dismissal claim, paragraphs 11 and 12 Post Office v Fennell [1981] IRLR 221. In any event, as the ET had recognised, under the attendance management policy, each case was to be treated on its own facts: see ET judgment paragraph 9. That the discretion should be exercised in a fair and consistent manner (the legitimate aim) did not mean that everyone must be treated the same or that reasonable adjustments should not be made for disabled people or people with particular disabilities. Equally, there was no error in the ET's finding that a different reasonable employer might not have dismissed at that stage. The question whether it was proportionate to take a decision at the particular time it was taken was a matter for the ET as a tribunal of fact.

Accepting that the ET had been wrong to state in its formal judgment that it had made a finding in respect of the CSAB decision, which was not before it, its reasoning demonstrated that it had not in fact trespassed on any issue of causation or remedy. Its finding that an award of 80% of the maximum available would be proportionate was a finding that was open to it. It was thereby finding that an award of 80% rather than 50% was a potentially less adverse means of achieving the legitimate aim. If alternative, less discriminatory possibilities are obvious or are placed before an ET, as was the case here, then it would be incumbent upon it to consider whether such a possibility would have been a more proportionate means of achieving the legitimate aim.

It is the respondent's case that the ET erred in finding the calculation of the efficiency payment under the CSCS amounted to unfavourable treatment: if a claimant is paid a benefit to which he is entitled as a result of his disability and, because of the method of calculating the amount to be paid, he is paid less than he would otherwise have been paid for a reason arising in consequence of that disability, it is not permissible to separate out the method of calculation or to base a claim on a reduction in a sum which would otherwise have been paid (see Williams, in particular at paragraph 28). The ET had failed to refer to paragraph 28 Williams and had erroneously proceeded on the basis that the treatment in question in this case was a reduction of the compensation, which was not the correct characterisation of the award made under the CSCS, albeit the respondent had accepted there had been a reduction in its case before the ET and acknowledged it was raising this point for the first time on appeal, arguing that it should be permitted to do so as it would not require any further evidence to simply raise a point of construction on the documentation.

The ET had further erred in distinguishing Williams on the basis that the benefit in that case could only be awarded to disabled persons. That was immaterial given that the claimant had only received the award because of his disability. In any event, to the extent it was possible for a non-disabled person to be awarded an efficiency payment, it was unlikely that such an entitlement would be allowed given that such payments were designed to compensate persons unable to sustain adequate attendance because of an underlying health condition. In Williams, the ill-health pension was awarded on the basis that the recipient was likely to be permanently incapable of efficiently discharging the duties of his post with the university or in relation to any comparable post.

The ET had also wrongly distinguished Williams on the basis that the method of calculation in the present case involved an element of discretion. That was a point immaterial to the Supreme Court's reasoning and gave rise to no principled point of distinction, not least as discrimination can arise from the application of set criteria as much as from decisions made in individual cases: see, for example, R (on the application of E) v Governing Body of JFS & Anor [2010] 2 AC 728 paragraphs 62-64.

For the claimant, it is submitted that the ET had correctly concluded (see paragraph 324) that the entitlement to the compensation payment in this case included both disabled persons and those who did not satisfy the legal definition of "disability". The respondent had a discretion to pay compensation under CSCS "where staff depart on inefficiency grounds". It was not restricted to those disabled under the EqA. At its highest, the respondent's case before the ET was that "the nature of the CSCS is that invariably most, if not all, beneficiaries will be disabled within the meaning of the EqA". In Williams, it was decided that Mr Williams had not suffered unfavourable treatment because the relevant treatment was the award of a pension that would not have been received if he had not been disabled. In contrast, in the present case, the relevant treatment was the act of reducing the claimant's payment for reasons that arose in consequence of the claimant's disability. The reasons for reducing the payment were, as the ET found, in part because of something arising in consequence of the claimant's disability: see the approach of the EAT in Chief Constable of Gwent Police v Parsons & Roberts UKEAT/0143/18. Similarly, the ET had been entitled to find that the compensation paid at 50% of the maximum was a reduction to that which the claimant would have received had it not been for the matters arising in consequence of his disability. This had been described as a reduction before the ET by both parties and it was not open to the respondent to now seek to change its position on appeal: see Secretary of State for Health v Rance [2007] IRLR 665 (EAT).

In any event, whether described as an act reducing the payment or otherwise as a decision to pay the claimant less under the CSCS due to something arising in consequence of his disability, the respondent's conduct was easily sufficient to cross the relatively low threshold for the treatment to put him at a disadvantage: see paragraph 5.7 EHRC Statutory Code.

By section 15 EqA, it is provided relevantly:

As for whether the particular treatment amounts to a proportionate means of achieving a legitimate aim (objective justification), the approach that an ET will need to adopt was summarised at paragraph 10 in MacCulloch v Imperial Chemical Industries plc [2008] ICR 1334 (EAT) as follows:

In referring at subparagraph (4) to the range of reasonable responses test, the EAT in MacCulloch was referring to the approach that is to be adopted by an ET in a claim of unfair dismissal. Under section 94 of the ERA, an employee has the right not to be unfairly dismissed. Pursuant to section 98 ERA, it is for the employer to show the reason for the dismissal and that it is a reason that is capable of being fair. Where the employer has done so, under section 98(4), it is provided that the determination of whether the dismissal is fair or unfair (having regard to the reasons shown by the employer):

In determining the question posed by section 98(4), the ET is required to ask itself whether the decision to dismiss fell within the band of reasonable responses open to a reasonable employer (Foley v Post Office [2000] ICR 1292-3 CA). The questions thus posed under section 15 EqA (objective justification) and under section 98(4) ERA (fairness) are subject to separate tests, albeit in some factual situations they may have similar effect (see per Underhill LJ at paragraphs 53-54 O'Brien v Bolton St Catherine's Academy [2017] EWCA Civ 145; [2017] ICR 737 CA and per Sales LJ (as he then was) at paragraph 55 York City Council v Grosset [2018] EWCA Civ 1105; [2018] ICR 1492 CA).

Returning to the approach to be adopted when determining a question of objective justification, in Hardys & Hansons plc v Lax [2005] ICR 1565; [2015] IRLR 726, Pill LJ provided the following guidance:

When determining whether the measure complained of was reasonably necessary, however, the ET is required to keep in mind that this does not mean there can only be one course open to the employer; reasonable necessity allows that there may be more than one option which would constitute a proportionate means of achieving the legitimate aim in question, see Health and Safety Executive v Cadman [2005] ICR 1546 CA at paragraph 31.

As for the more specific questions raised by this appeal, it is relevant to recognise that a requirement that employees maintain a certain level of attendance at work in order not to be subject to warning and dismissal is likely to disadvantage those disabled workers whose disability results in more frequent and perhaps longer absences. Moreover, it should be acknowledged that that disadvantage will not be eliminated by both disabled and non-disabled workers being treated equally under a sickness absence policy if the requirements bite harder on disabled workers (or on a category of disabled workers) than on non-disabled workers. It is, of course, in recognition of this context that section 15 EqA does not require a comparative approach, but simply that the claimant is able to establish unfavourable treatment (and see the observations made by Elias LJ in Griffiths v Secretary of State for Work and Pensions [2017] ICR 160 CA). That such a requirement might disadvantage a disabled employee does not, however, necessarily mean that it would not be capable of objective justification; as the CJEU acknowledged in Ruiz Conejero: "Combating absenteeism at work may be regarded as a legitimate aim." The question will, of course, be for the ET on the evidence and on the facts of the particular case before it.

Applying the language of the statutory protection afforded by section 15 EqA, it is thus necessary to determine first what is the relevant treatment and to then ask whether that was unfavourable to the claimant. In Trustees of Swansea University Pension and Assurance Scheme v Williams [2019] 1 WLR 93, the Supreme Court had to answer those questions in the context of an ill-health pension where benefits were calculated by reference to the salary of the individual at the time of the termination of employment; that disadvantaged the claimant in that case because, as a result of his particular disability, his hours and, therefore, his salary had been reduced. The Supreme Court held, however, that this could not properly be regarded as unfavourable treatment; as Lord Carnwath explained, at paragraph 28:

At paragraph 27, Lord Carnwath expressed his broad agreement with the Court of Appeal's analysis in that case, including (as set out at paragraph 20):

There is no dispute, pursuant to sections 119 and 124 EqA, that damages for unlawful discrimination are to be calculated in the same way as damages in tort. As was held in Abbey National v Chagger [2010] ICR 397 CA at paragraph 57, it is necessary to ask what would have occurred had there been no unlawful discrimination.

Although the EAT has a discretion to permit a new point of law to be argued on appeal, the circumstances in which it may do are limited, as helpfully summarised by HHJ McMullen QC at paragraph 50 Secretary of State for Health v Rance. Such circumstances may include those cases where the EAT is in possession of all the material necessary to dispose of the matter in issue without recourse to a further hearing.

In addressing an appeal against an ET's finding on the issue of objective justification, in Hardys & Hansons plc v Lax, Pill LJ provided the following further guidance (see paragraph 34):

More generally, when considering the reasoning provided by the ET, the EAT will keep in mind that the standard required is one of adequacy, not perfection: see per Lewis LJ in Sullivan v Bury Street Capital Ltd at paragraph 42, and the helpful guidance provided by Popplewell LJ in DPP Law Ltd v Greenberg [2021] EWCA Civ 672 at paragraphs 57-58.

Between 2016 and his dismissal, the claimant had been absent from work for a total of 245 days on 23 different occasions. Indeed, at the time of this dismissal, he had been continuously absent since 29 May 2018, save for three days between 16-18 July 2018. An Occupational Health report of 9 August 2018 had said the claimant should be fit to return to work within two to four weeks, but that did not happen. The respondent had initially considered referring the claimant's case to a decision-maker when he confirmed he was not fit to return to work at the end of his then fit note on 15 August 2018, but had determined he should be given additional time. His case was only referred to a decision-maker on 19 September 2018, after the claimant had confirmed he was still not fit to return to work in any capacity. At the time the decision to dismiss was taken. and at the date of his appeal, the claimant was still not fit to return to work in any capacity and was unable to say when he would be. This was so notwithstanding that, by the time of his appeal, he had completed his therapy and had tried three different types of medication. The ET was satisfied that the respondent had provided the claimant with support and adjustments over the material period, during which time he had expressed dissatisfaction with the respondent, threatened to resign and behaved in a way that (as the claimant accepted) might be viewed as somewhat aggressive and petulant, and at times had also been difficult to contact.

Although the ET was prepared to accept that other employers might reasonably have allowed yet further time before making a decision on the claimant's future employment - in particular to undertake yet further investigation, to obtain a further medical report or to provide for more time for the claimant to trial a further change in his mediation - it did not consider it was outside the band of reasonable responses for this employer to take the approach that it did. Simply applying the test required under section 98(4) ERA, there could be no challenge to the ET's conclusion thus reached on the claimant's complaint of unfair dismissal.

A different test is, however, imposed under section 15 EqA. In determining whether the decision made by the respondent was justified, the ET was not asking whether that decision fell within a band of reasonable responses (Hardys & Hansons plc v Lax) but had to be satisfied, carrying out the requisite critical evaluation of the means adopted, that it corresponded to a real need on the part of the undertaking, was appropriate to achieving that need and reasonably necessary to that end ( Bilka-Kaufhas GmbH v Weber Von Hartz ). In undertaking that assessment, the ET was entitled to consider whether the need in question might have been achieved by less discriminatory means even if those other options were not considered by the respondent at the time; the test, after all, is one of objective justification.

As the respondent has pointed out, however, although a different test is imposed by section 15 EqA, a finding that there might have been more than one option open to the employer does not mean that the particular course adopted in the case under consideration might not still demonstrate the requisite reasonable necessity (and see Health and Safety Executive v Cadman). Inevitably, it will be a question of fact and degree, and Parliament has vested responsibility for determining this issue in any particular case in the ET. The role of the appellate tribunal is to scrutinise carefully the manner in which the ET has reached its decision; it is not to substitute its own view for that reached by the first instance tribunal.

The starting point for the ET, for these purposes, was to determine whether the respondent had established any legitimate aims and, if so, what. Here, the ET accepted that there were aims that corresponded to a real need on the part of the respondent: (i) to ensure staff were capable of demonstrating satisfactory attendance and a good standard of attendance, that aim comprising the aims of the maintenance of a fair, effective and transparent sickness management regime and the efficient use of resources, (ii) to provide a good customer service; and (iii) to apply the respondent's policies fairly and consistently (albeit this third aim was seen as subsidiary to the first). The claimant's first objection is, however, that the first aim could only constitute a legitimate aim if supported by the second: if the respondent could not establish any adverse impact on customer service as a result of the claimant's absence, it could not demonstrate it had a real need to ensure staff were capable of demonstrating satisfactory attendance and a good standard of attendance.

I am not persuaded that this is a challenge that identifies any error on the part of the ET. First, the ET did not find that there was no impact on customer service as a result of the claimant's absence, it just did not consider the evidence to be sufficient to demonstrate that that impact would justify the claimant's dismissal (the respondent was unable to establish proportionality in this context). Secondly, even if there had been no impact on customer service, it was open to the ET to find (as it did) that the claimant's absence gave rise to other adverse consequences, both on the morale of the team and, potentially relatedly, in terms of the management time and energy that had to be devoted to managing that absence. Third, the ET was also entitled to conclude (as it did) that it was legitimate for the respondent, in particular as a public body, to seek to maintain satisfactory attendance on the part of its employees in any event, so as to ensure an efficient use of resources.

Having permissibly found as a fact that the first aim was legitimate, and thus corresponded to a real need on the part of the respondent, the ET turned to consider whether the measure adopted to achieve this aim - in this case, the dismissal of the claimant - was proportionate (that is, appropriate and reasonably necessary) given its discriminatory impact. In this regard, it is clear that the ET accepted that there was an impact on the respondent arising from the claimant's absence and the management resources this required (see ET judgment paragraph 347). I was initially concerned that the ET had not then demonstrated how it had measured that impact so as to carry out the required balancing exercise against the discriminatory impact on the claimant. Reading the ET's reasoning as a whole, however, I am bound to take into account the findings it recorded relevant to impact when addressing the evidence relating to the second aim (the impact on productivity and morale; see the ET judgment at paragraph 355) and to set its conclusions against the background of its earlier findings of fact, which amply demonstrate the very real impact that the claimant's absence had (in particular, on management time and on team morale) in terms of the respondent's efficient use of resources. Adopting a holistic approach to the ET's reasoning, as I am required to do, I cannot say that it failed to weigh the impact of the claimant's absence when carrying out the balancing exercise required of it.

Moreover, when carrying out its assessment, the ET plainly looked back to the questions that it had considered in respect of the unfair dismissal claim and the different choices that had been available to the respondent. When thus determining that it was reasonably necessary for the respondent to decide that the claimant should be dismissed, the ET again considered whether the aim in question might appropriately have been achieved by a less discriminatory measure (waiting longer, subjecting the claimant to a further written warning and so on); it did not fail to take into account the possibly less discriminatory options that might have been considered, but addressed those in its evaluation under section 15 EqA, finding that these would not have been appropriate to achieve the aims of ensuring satisfactory attendance and a good standard of attendance, comprising the maintenance of a fair, effective and transparent sickness management scheme and an efficient use of resources.

As for the subsidiary third aim, I do not consider it was wrong of the ET to find it was a legitimate aim for the respondent to apply its policies fairly and consistently. Given the particular policy in question - which emphasised the need to deal with each case on its own facts, and allowed for different treatment for those who were disabled - the fair and consistent application of that policy was not necessarily discriminatory. Seen as part and parcel of the first aim, the ET was entitled to find that this, too, would have justified the Claimant's dismissal.

Given my findings on the ET's approach to the section 15 question of objective justification, it must follow that I equally see no basis of challenge to the ET's conclusion on unfair dismissal. I therefore dismiss this first ground of appeal.

Addressing the second basis of challenge pursued by the claimant, I can deal with this more shortly. As is common ground, to the extent that the ET purported to make a determination in relation to the decision of the CSAB, that was wrong. The CSAB is independent of the respondent and its decision was not before the ET. To the extent that the ET's judgment recorded that it had made a determination in this regard, that was in error and would have needed to be set aside.

That, however, is not the substance of the claimant's challenge. It is his case that, in finding that the 80% payment ultimately allowed by the CSAB under the CSCS was proportionate, the ET effectively predetermined the issue of remedy.

I do not consider that is correct. As the respondent has observed, at the liability stage, the claimant had fairly raised the question whether the original 50% payment under the CSCS was proportionate given that it took into account, at least in part, behaviour by the claimant that arose as a consequence of his disability. It was the claimant's case that a less discriminatory measure would have been to make a 100% award in this case.

Having found that reducing a CSCS award to something below 100% would be a justified means of achieving the aims of distributing the fund fairly and economically in line with the applicable guidance, the ET appropriately turned to consider whether there was a less discriminatory means of achieving that aim on the facts of this case. It had found that the lesser payment was in part informed by factors arising from the claimant's disability but concluded that this might proportionately be recognised by a payment at 80% of the total possible award. It was on that basis that the ET determined that the original decision made by the respondent had amounted to unlawful discrimination. As it recognised (see paragraph 389 of its judgment), the actual remedy that might result would need to be determined at a further hearing.

Although I canvassed with Mr Campion during oral argument whether it might be said that the ET's reference to the CSAB's decision had wrongly tainted its reasoning (bringing in the possibility of an 80% payment when that was not a relevant part of the evidence before it), that was not the way he put the claimant's case. His submission was that the ET's error had been to make a decision on remedy when only issues of liability were before it.

I can see that it may well be that the ET's finding as to what might have been a more proportionate level of payment may feed into any subsequent consideration of remedy, but finding that a payment of 80% of the total would be a more proportionate means of achieving the aim in question is not the same as finding that is the level of payment that the respondent would determine to make once aware of the discrimination arising from its original decision (that is, if it were to make the decision absent the discrimination as found). In any event, the position that arises is not dissimilar to the situation that occurs quite often in, for example, the unfair dismissal context where the ET's findings in relation to liability largely set the scene for subsequent findings that may, for example, go to a Polkey reduction or a finding of contributory fault relevant to remedy. To the extent that the claimant seeks to contest that a different sum would need to be awarded at a remedy stage, that would still be open to him to make good that submission, although that would need to take into account the findings of fact made by the ET at the liability stage. For completeness, I should also observe that the ET's judgment does not address any question of injury to feelings that might arise from the original decision.

Given the decision I have reached on the respondent's cross-appeal, I do not need to make a formal decision on the claimant's appeal but, were I to be wrong about the cross-appeal, then I would have allowed the appeal against the ET's formal judgment relating to the decision of the CSAB, but otherwise would have dismissed this second ground of appeal.

The issue raised by the cross-appeal is defined by the answers given to the questions identified by the Supreme Court in Williams: first, what is the relevant treatment? and, second, was it unfavourable to the claimant?

In this case, the relevant treatment was the payment to the claimant under the CSCS. That was a payment made because the claimant had been dismissed for unsatisfactory attendance. The objective of the payment is clear: it is:

In this case, the ET had found that the claimant's dismissal for unsatisfactory attendance was related to his disability. Although his previous absences had included periods of non-disability-related leave, the ET was clear that the reason for his dismissal was in consequence of his disability.

It was, moreover, the claimant's underlying health condition, arising from his disability, that gave rise to his entitlement under the CSCS. As such, the relevant treatment - being treated as entitled to a payment under the CSCS - was not unfavourable treatment; if anything, it was more favourable than would have been the position if the claimant had been dismissed for a reason other than his disability. It is possible, although unlikely, that an entitlement under the CSCS might arise in the case of somebody who would not meet the statutory definition of a disabled person for the purposes of the EqA, but that was not this case where, on the facts found by the ET, the claimant's entitlement arose solely by reason of his disability. That, in my judgement, puts this case on all fours with Williams and to conclude otherwise, as the ET did, would be to make the error of artificially separating out the entitlement to the award (the relevant treatment, which did not constitute a disadvantage), from the calculation of that award; the latter would not have arisen but for the initial entitlement.

The point can be made good by considering if the claimant had been entitled to a payment under the CSCS by reason of a different type of disability, perhaps a physical disability rather than one relating to his mental health, albeit one that had led to their unsatisfactory attendance at work. In such circumstances, there is no reason for thinking that the matters which led to the reduction of the claimant’s award would have arisen; in such circumstances, because of his disability, the claimant would have been entitled to 100% of the award. As in Williams, that points to a situation where the treatment in issue is not itself unfavourable treatment; the argument only arises within that context, in the particular calculation of the payment.

The position was different in the case of Chief Constable of Gwent Police v Parsons & Roberts. In that case, the relevant treatment was an entitlement to a payment under the Police Voluntary Exit Scheme (analogous to a redundancy scheme). The claimants’ access to that scheme had not been found to be related to the fact that they were disabled. The subsequent reduction of benefits under the scheme had, however, arisen as a consequence of something arising from the claimants’ disabilities. That reduction thus amounted to unfavourable treatment for section 15 purposes because it fell to be viewed within the context of an entitlement to a benefit to which the claimants had been able to access regardless of disability.

In the circumstances, I do not need to address the subsidiary question as to whether the ET was right to regard the calculation in this case as a reduction; I am satisfied that it erred in failing to see this as a case in which there was no unfavourable treatment for the same reasons as identified in Williams and I find that the distinctions that the ET purported to draw were not material to the reasoning of the Supreme Court in that case. First, whether or not the benefit in Williams could only be awarded to disabled persons, would be immaterial given that the claimant in the present case (as was the position for Mr Williams) only received the award because of his disability. Secondly, as the respondent has pointed out, even if it were theoretically possible for a non-disabled person to be awarded a payment under the CSCS, that would be factually highly unlikely, given that such payments are designed to compensate persons unable to sustain adequate attendance because of an underlying health condition. Thirdly, I cannot see that it is a point of valid distinction to refer to the aspect of discretion that arises in relation to a payment made under the CSCS. As the respondent submits, that cannot give rise to a valid way of distinguishing the approach taken in Williams. In any event, such discretion as was exercised in this case, in relation to the calculation of the payment, was subject to the particular guidance and rules provided.

For all those reasons, I therefore allow the respondent's cross-appeal.