Mr A Fasano v Reckitt Benckiser Group plc and Reckitt Benckiser Health Ltd: 3306340/2020

EMPLOYMENT TRIBUNALS

Case No 3306340/2020

Between

Mr A FasanoClaimant

Reckitt Benckiser Group plc and Reckitt Benckiser Health LtdRespondent

Before

Employment Judge Anstis

Date 29 December 2021

JUDGMENT

The claimant’s claim of indirect age discrimination is dismissed.

REASONS

[1]The claimant was employed by the second respondent (or its predecessors) from 1 June 1997. By the time of his retirement on 30 June 2019 he was employed as Chief Supply Officer. His role was one level below a board director.

[2]The first respondent is a quoted company and is the ultimate parent company of the second respondent. The second respondent is 100% owned by the first respondent, albeit with several entities in between them in the chain of ownership.

[3]The claimant’s claim is of post-employment indirect age discrimination. Page 1 of 52

[4]His employment ended under the terms of a “retirement agreement” dated 7 September 2018. This included provision for a period of garden leave with the claimant’s employment eventually ending on 30 June 2019. For the purposes of his claim the important element of this retirement agreement is that he was to be treated as a “good leaver” under the respondent’s 2017 long term incentive plan (“LTIP”). The LTIP is offered to senior employees within the first respondent’s group and gives employees who participate the opportunity to receive stock and stock options in the first respondent on the attainment of performance conditions, with those performance conditions being assessed on a group-wide rather than individual basis.

[5]In the period September 2019 to May 2020 the claimant was 58 years old. His claim is based on being a member of a group of people aged 57 or above.

[6]On 18 September 2019 the remuneration committee (“Remco”) of the first respondent decided to amend the performance conditions that attached to the 2017 LTIP. The effect of this was that the relevant awards would vest at at least 50%, compared to the 0% that had been previously projected. This relaxation of the performance conditions only applied to “those who are current group employees as at 18 September 2019” and thus did not include the claimant, whose employment had ended on 30 June 2019. It is said by the claimant that this decision amounted to post-employment indirect age discrimination.

[7]The claimant’s claim in the employment tribunal is expressly limited to a claim of indirect age discrimination, with the claimant reserving the right to bring a separate breach of contract claim in the High Court. We therefore do not intend in this decision to address whether there has been any breach of contract in respect of the actions of the respondents, and our decision is not intended to affect or prejudice any later claim of breach of contract in respect of the 2017

The issues

[8]The parties agree that to determine whether the claimant has been the subject of unlawful age discrimination, we must address the following issues (omitting any question in relation to remedy): “1. This is a claim for:(1) indirect age discrimination contrary to sections 19 and 39 of the Equality Act 2010 (“EqA”) against the Second Respondent;

(2) accessory liability contrary to sections 110(2) and/or 112 EqA against the First Respondent. Indirect discrimination Page 2 of 52 2. Did the Respondents apply to the Claimant and/or refuse to amend the following provision, criterion or practice (“PCP”): “requiring LTIP participants (who were not already bad leavers) to be employed as of 18 September 2019 (or, in the alternative, May 2020) in order for them to benefit from the amended performance condition” 3. If so, was any such PCP applied to those of the Claimant’s age group (those aged 57 or older the Claimant was 58 at the time of the decision) as well as to those in other age groups? 4. Did any such PCP put or would it put those aged 57 or older at a disadvantage as compared to those under 57 years of age? 5. Did any such PCP put the Claimant at the particular disadvantage as he could not meet the PCP and, as a result, his LTIP award lapsed? 6. Can the Respondents show that any such PCP was objectively justified as a proportionate means of achieving the legitimate aim of retaining and incentivising existing employees? 7. Without prejudice to the burden of proof which is on the Respondent, the Claimant contends that the PCP is not objectively justified having regard to: (i) the treatment of those who left in the period between 18 September 2019 and May 2020 and were allowed to benefit from the amended performance condition: and (ii) that Mr Rob de Groot benefitted from the amended performance condition. The Respondents 8. If the alleged PCP was applied to the Claimant by the First Respondent, was the First Respondent acting as an agent of the Second Respondent with the Second Respondent’s authority?

[9]If so, is the Second Respondent liable as the Claimant’s (former) employer pursuant to s 39 EqA and/or pursuant to section 109(2) EqA?

[10]Is the First Respondent liable:(1) pursuant to section 110(2) EqA?

(2) pursuant to section 112 EqA? The Claimant contends that the First Respondent knowingly helped the Second Respondent to discriminate. That is denied by the First Respondent. Time Page 3 of 52

[11]Was the claim (presented on 3 July 2020) and/or the amended claim (presented on 7 October 2020) presented in time? In particular:(1) When did time start to run? In particular:

a. Is the PCP complained of properly characterised as an act or omission?

b. If an act, when did it occur? Was any PCP conduct extending over a period which is to be treated pursuant to section 123(3) EqA as being done at the end of the period? If so, when was the end of the period in question - the Claimant says there was a continuing act to May 2020?

c. If an omission, when should it be taken to have occurred having regard to sections 123(3) and (4) EqA?

d. Did time run from when the Claimant became aware of the disadvantage and/or when the Respondent’s refused to reconsider the rule in April and May 2020?

(2) Was the claim brought within time having regard to the time limits?

(3) If not, would it be just and equitable to extend the primary time limit? The Claimant contends that he only became aware of the decision in April 2020 and acted with reasonable expedition from then. That is not accepted by the Respondents.” 9. The respondents’ closing submissions addressed these points by reference to five headings: the agency issue, the PCP issue, the disadvantage issue, the justification issue and the time issue. We will adopt that format in our later statement of the law and our discussion and conclusions. The hearing 10. The hearing took place by CVP with all parties attending remotely. It took place from 4 - 7 October and then on 11 & 13 October 2021. It was listed to determine liability only. The claimant was represented by Mrs L Banerjee of counsel and the respondents were represented by Mr S Forshaw of counsel. 11. At the outset of the hearing we addressed an application by the claimant for a rule 50 order, which has been issued separately. We took the remainder of the first day to read into the case.

The hearing

[12]On the second day of the hearing we heard the claimant’s evidence, which continued through to the middle of the third day. For the remainder of that day and the whole of the fourth day we heard evidence from Simon Barron, the respondents’ Group Head of Reward. On the fifth day we heard evidence from Mary Harris, a non-executive director of the first respondent and chair of the Page 4 of 52 Remco. With the agreement of the parties, they exchanged written submissions and had the opportunity for up to 90 minutes each of oral submissions on the sixth and final day of the hearing. After the hearing

[13]At the end of the hearing we set a date for a provisional remedy hearing to take place (if required) on 24 & 25 March 2022. We invited the parties to agree and submit directions for that hearing, which they later did. We informed the parties that we intended to meet for discussion in chambers on 11 & 12 November 2021. As we were unable to conclude our deliberations then, we met again in chambers on 10 December 2021. During the course of those discussions we reached these unanimous conclusions. B. THE FACTS The claimant and the respondents

[14]The first respondent is a publicly quoted company and part of the FTSE100. It has its head office in Slough, but its group operates on a worldwide basis. At the time we are concerned with it had two trading divisions: health and hygiene/home. Through those divisions it produces and sells many well-known brands internationally.

[15]The first respondent has a board of directors. The only executive directors are the Chief Executive Officer and the Chief Financial Officer. The Remco is a subcommittee of the first respondent’s board. It is chaired by Mary Harris. Although not described as such it appears that Simon Barron, in his capacity as Chief Reward Officer, acted in a role akin to secretary to the Remco, and was also one of several advisors that the Remco relied on. He was not himself a member of the Remco. As may be expected of a publicly quoted company, the Remco had to operate within a complex regulatory environment, encompassing restrictions on what it could do and (in some cases) requirements to report on its activities and decisions.

[16]Below the board is the Executive Committee (or “EC”) which is chaired by the CEO.

[17]The claimant had enjoyed a long and successful career within the first respondent’s group. At the time of his retirement he was employed by the second respondent as Chief Supply Officer. When actively at work (as opposed to being on garden leave) he was a member of the EC.

[18]For its internal purposes, the first respondent designated some managers to be within its “Top 40” (or “T40”) and “Top 400” (or “T400”) groups. In principle these were the 40 and 400 most senior managers within the business. The LTIP was typically a benefit offered to only those in the T400 group. There were another Page 5 of 52 60 or so “high performance middle managers” who may be entitled to some limited participation in the LTIP. Elements of pay

[19]For T400 employees their remuneration package had three main elements: - Basic salary - The Annual Performance Plan (“APP”) and - The LTIP.

[20]The APP was a conventional annual bonus scheme based on individual performance targets. The LTIP

[21]While on one level simple and prescriptive, the LTIP scheme also appears to have given the first respondent considerable flexibility and discretion in its operation.

[22]The first respondent has operated LTIP schemes on much the same basis for several years. An employee is allocated a certain number of shares and/or share options, which vest depending on the performance of the first respondent over the next three years. The criteria that dictates how much of any award vests is the “performance condition”. The employee must remain employed at the relevant time for the shares and/or share options to vest. Subject to conditions that we will come on to, an employee who left employment forfeited any unvested or contingent entitlements under the LTIP.

[23]The “performance condition” for the first respondent’s LTIP was invariably based on “earnings per share” (or “EPS”). In that way it was said that its senior managers’ financial interests were aligned with the interests of its shareholders. The performance condition would typically be a percentage improvement in EPS.

[24]Since a new LTIP is issued every year, employees with more than a few years’ service will have several LTIPs running simultaneously. An LTIP which is currently live is known as a “in-flight“ LTIP. The 2017 LTIP related to performance in the period 2017-19. The claimant was notified of his possible entitlements under the 2017 LTIP, and the performance condition, in December 2016. The first respondent’s financial year ends in December. Vesting under an LTIP typically occurs on the announcement of the first respondent’s annual results at its AGM in May. For the 2017 LTIP this was on 12 May 2020. As will appear below, a concession under the 2017 LTIP (as originally drafted) treated awards for employees who had left the group as vested at the end of the financial year (i.e. December 2019) rather than the following May. Page 6 of 52

[25]The default position was that on leaving employment an employee would forfeit any unvested entitlements under the LTIP. That was known as being a “bad leaver”. However, there was also a separate category of “good leavers” who were entitled to leave employment and retain some entitlements under the

[26]The operation of the LTIP scheme would seem to be simply a matter of mathematical calculation against financial figures, which would then determine how much of any award the employee receives. In practice, it is more complicated than this.

[27]First, the amount of shares and/or share options is set individually for each employee. For those at the very top of the organisation the potential earnings under the LTIP scheme would form by far the most important element of their remuneration. They were granted more stock and/or stock options than less senior employees. We were told that for the vast bulk of employees who participated in the LTIP scheme the potential awards under the scheme were far less significant as a proportion of their total remuneration package.

[28]Second, there could be one-off or ad hoc grants of stock and/or stock options to employees during an LTIP scheme. We heard that this would be done if, for instance, the first respondent wanted to recruit a new employee who would, on leaving their former employer, forfeit entitlements under their former employer’s LTIP or other bonus scheme. The grant of stock and/or stock options under the first respondent’s LTIP would be designed to compensate them for what they may have lost under their previous employer’s scheme.

[29]Third, for at least some employees under some LTIP schemes there was the possibility of “re-testing” the awards. This was effectively a “second chance”. If the awards had not fully vested there would be a further calculation to assess whether the performance conditions had been met in the subsequent performance period. This rule did not apply to the claimant.

[30]Fourth, the Remco always retained a discretion to vary the performance condition, even for an in-flight LTIP.

[31]Fifth, the Remco could, if it wished, designate an employee who left the business as a “good leaver”

[32]Six, various restrictions on the grants (including a requirement for senior managers to maintain a minimum shareholding in the first respondent) meant that in practice the shares and/or share options could seldom be fully cashed in or liquidated immediately on vesting.

[33]The practical administration of the LTIP was handled by Computershare, an external contractor. The claimant’s contract, and relevant provisions of the 2017 LTIP Page 7 of 52

[34]The claimant’s contract of employment (dated 19 February 2009) says at clause 5, under the heading “Employee Benefits”: “The Executive shall be entitled to participate in retirement, welfare benefits, incentive compensation, perquisite and other plans and arrangements of the Company applicable to its senior executives as in effect and on such terms as are prescribed by the Company in its sole discretion from time to time. Such participation includes any awards granted under the “Long Term Incentive Plan” of a Group Company.”

[35]An “amendment to employment agreement” of the same date provides: “Long Term Incentives: You will be eligible for consideration for awards of share options and restricted shares in line with Company plans. The actual size of any award for any year is based on performance, and the other provisions of the relevant Company share plan(s). You will not be eligible for an award if at the date on which grants are awarded your employment has terminated, or you are under notice of termination of your employment …”

[36]A note dated November 2016 gives the amount of shares and share options the claimant may receive under the 2017 LTIP, but beyond saying that those awards are subject to the terms of the LTIP scheme (which is not in dispute) does not appear to impose any other entitlements or conditions in respect of the 2017 LTIP.

[37]These are the relevant provisions of the 2017 LTIP: “ “Committee” means [the Remco] “Company” means Reckitt Benckiser Group plc … 1.4 Performance Conditions When granting an Award, the Company may make its Vesting conditional on the satisfaction of one or more conditions determined by the Committee at least one of which must be linked to the performance of the Company. Performance Conditions must be objective and specified at the Award Date and may provide that an Award will lapse if the Performance Conditions are not satisfied. The Company, with the consent of the Committee, may waive or change the Performance Conditions in accordance with its terms or if anything happens which causes the Company reasonably to consider it appropriate, provided that any changed Performance Conditions will be no more difficult to satisfy. … Page 8 of 524.1 Timing of Vesting Subject to rules 6 (Leaving the Group) … an Award shall Vest on the latest of the following:4.1.1 the date on which the Committee makes its determination under rule

4.1.2 in the case of Awards with no Performance Condition, the Vesting Date;

4.1.3 the date the Committee decides that any other condition (rule 1.5) are satisfied or waived; and

4.1.4 the date on which any Dealing Restrictions which prevent Vesting on the dates specified above cease to apply. 4.2 Determination of Performance Conditions and other conditions As soon as reasonably practicable after the end of the Performance Period, the Committee will determine whether and to what extent any Performance Conditions … have been satisfied and how many Shares Vest for each Award. To the extent that any Performance Conditions or other conditions are not satisfied, the Award lapses. … 6. Leaving the Group

4.2 of the extent to which any Performance Conditions are satisfied or waived;

6.1 General rule on leaving employment An Award which has Vested will not lapse on the date the Participant ceases to be an employee. An Award which has not Vested will lapse on the date the Participant ceases to be an employee unless rule

6.2 applies. 6.2 Leaving in exceptional circumstances – unvested Awards 6.2.1 If a Participant ceases to be an employee of any Member of the Group for any of the reasons set out below, then his Awards which have not Vested will Vest as described in [a separate rule] and lapse as to the balance. The reasons are:

(i) ill-health, injury or permanent disability, established to the satisfaction of the Company;

(ii) the Participant’s employing company ceasing to be under the Control of the Company; Page 9 of 52

(iii) a transfer of the undertaking, or the part of the undertaking, in which the Participant works to a person which is neither under the Control of the Company nor a Member of the Group; or

(iv) redundancy; or

(v) retirement with the agreement of the Company;

(vi) any other reason, at the discretion of the Committee.”

[38]The rules we have taken those extracts from are the rules of the 2015 plan, as amended on 27 July 2016. We understand that it is those rules that applied to the 2017 LTIP.

[39]The 2017 LTIP grant contained the following additional provision: “For the purposes of Rule 6.1 only (Leaving the Group), an Award will be treated as a Vested Award from the end of the financial year of the performance period.”

[40]Although these words are not used in the LTIP, an employee who left subject to clause6.1 was referred to as a “bad leaver” and an employee who left subject to clause

6.2 was referred to as a “good leaver”.

[41]We have not set out the vesting rules that apply to good leavers, but it is not in dispute that bad leavers immediately forfeited all unvested entitlements under an LTIP whereas good leavers retained some entitlements to vesting of share and share options under the LTIP, despite having left employment. The parties appeared to agree that good leavers retain the same entitlements as those who remain in employment, subject only to a pro-rata reduction for the amount of time in the performance period that they were not employed. The way in which this was to operate for the claimant was set out in detail in his retirement agreement.

[42]Very broadly, the category of good leavers comprised those who had not (except for retirees and the possibility of resignation due to ill-health) resigned voluntarily and who had not (except for redundancy) been dismissed. Bad leavers comprised those who had resigned or been dismissed for reasons other than ill-health or redundancy.

[43]Another distinction between good leavers and bad leavers was that whether someone left as a “good leaver” was largely under the control of the first respondent. That was particularly so in the final two categories – retirement (which had to be explicitly with the consent of the first respondent) and “any other reason”, which was entirely under the control of the Remco. Proposed amendments to the LTIP - 2018 Page 10 of 52

[44]The first respondent prided itself on having a “performance culture”, exemplified by the LTIPs’ focus on an increase in earnings per share. Up to around 2018 the performance conditions of the LTIPs had typically been fully met, resulting in full vesting under the LTIPs and high financial rewards for senior managers.

[45]On 1 January 2018 the first respondent had implemented the reorganisation of its group into the current two divisions: health and hygiene & home. This reorganisation was known as “RB 2.0”. It had been thought at the time that this may be a precursor to a formal split of the group into two separate companies.

[46]By mid-2018 it had been recognised within the first respondent that on then current projections the LTIPs would no longer fully vest, and in future years may not vest at all.

[47]The minutes of the Remco meeting on 12 July 2018 record that: “The Committee requested management make a proposal at the next meeting to address potential near-term retention issues in light of the projected LTIP vesting.”

[48]On 31 August 2018 Mr Barron submitted to Ms Harris (for the Remco) what he called a “proposal regarding retention in context of LTIP vesting”. The “executive summary” of his paper projects the current LTIPs vesting at around 50% for 2018 and then 0% for 2019 and 2020. It notes that: “This compares to a history in RB of maximum vesting and significant LTIP values for participants. The potential for reduced vesting over the next three years is therefore likely to have significant retention and motivation impact, which this paper seeks to address.”

[49]Mr Barron’s proposal at this point is that no changes are made to the entitlements of the EC, but that for the T40 and T400 (excluding the EC) adjustments are made on how the performance condition is to be calculated, together with an additional one-off grant of further stock and options.

[50]We note that the Remco requested this proposal “to address … retention issues”, and that Mr Barron then frames his proposal in terms of “retention” and “motivation”.

[51]On 7 September 2018 the claimant entered into his retirement agreement with the second respondent. His retirement is to take effect on 30 June 2019. He is to be on garden leave following a “transition period” expected to last from midNovember 2018 to the end of January 2019.

[52]The Remco met on 17 September 2018 to consider (amongst other things) Mr Barron’s proposal. It endorsed Mr Barron’s proposal, citing “potential near-term Page 11 of 52 retention issues”, but not mentioning “motivation”. The same meeting “approved the leaver treatment for [the claimant]”.

[53]On 2 November 2018 a “Group Principal & Emerging Risk Report” included the following as a “risk profile update”: “Talent – [risk] increased to “highly likely” in light of depleted bench strength and the impact of poor business performance on bonus payments.”

[54]The following also appears in the report: “Post the RB 2.0 reorganisation, there is a risk that RB cannot implement its strategies and meet objectives as a result of increasing numbers of management leaving the business who cannot be readily replaced by equally high-calibre experienced / qualified candidates. The remuneration strategy is under increased scrutiny and pressure and any move from a high variable to higher fixed cost model could trigger a change in culture. The current year continues to be challenging and there is ongoing risk that great talent (T400 and beyond) will seek opportunities elsewhere. This is likely to be exacerbated if bonus targets are not achieved.”

[55]Following a meeting of the EC on 15 November 2018, on 23 November 2018 the first respondent’s then CEO, Rakesh Kapoor, wrote to Ms Harris suggesting some changes to the Remco’s adjustments to the LTIPs. He said that his proposed changes would “keep the [T40] incentivised on EPS growth”. Ms Harris discussed this with Mr Barron. Mr Barron reported back to Mr Kapoor that Ms Harris “is ok with the proposal and supportive”. Later the same day Ms Harris contacts her colleagues on the Remco recommending approval of Mr Kapoor’s changes. In that email she talks of approving “the previous retention plan”. She says “we all agreed to the previous retention recommendation as we all recognise that we are in a period of uncertainty”. She says “this proposal has clear benefits in terms of … retention and is broadly equivalent in cost … to the previous retention plan that was approved.” She says “we need to lock them [presumably senior management] in, and at present I see no alternative given where we are.” Later that day she writes to Mr Kapoor saying: “I have now received approval from all the Remco … Clearly we would all prefer that performance meant we could pay out under our schemes. However we all recognised that in current circumstances, ensuring retention at this time is critical and are supportive of this revised proposal.” Page 12 of 52

[56]The following day Mr Kapoor reported back to the EC saying that the Remco has approved the changes, saying that he had had “an intense discussion around retention and motivation of our people” with Ms Harris.

[57]A couple of days later, on 27 November 2018, the CFO, Adrian Hennah, wrote to the EC in the following terms: “EC Colleagues, Apologies that I missed most of the EC call last week. An outside commitment. In particular the discussion on LTIP transitional arrangements. I think you arrived at a poor solution to a big problem. The problem is clear. Two critical years for the Group, delivering RB2.0 from a position today of some weakness. And an LTIP programme that means our top 500 managers will receive no LTIP pay-out during those two years (barring a substantial depreciation in sterling). A major retention issue. Which must be addressed. The proposal to address the problem by tweaking the flawed LTIP programme that created the mess does not seem optimum. Does it make sense that this year’s pay-out will be near 100% - is this a measure of real Group performance over the last three years? What does this do to performance culture? Does it make sense that the next two year’s pay-outs are likely to hover around threshold, with the biggest driver of the outcome the level of sterling; indeed the outcome of Brexit? What does this do for performance culture? We have been staring at this issue for months. The transition period to a new LTIP programme. I hope our advisers have put forward alternatives. One I would offer is a variant on a form of LTIP used by some companies. An enhanced APP with a large proportion mandatorily converted to shares. One new programme in each of the two transitional years. Explained externally as “handling a retention problem as a transition to a new LTIP arrangement”. With targets set relevant to each year. Constant exchange rates, so no sterling complication. Targets metrics as for current APP. Very aligned to shareholder value creation. Possibly with also an across the board “bump” (say 20%) to the award level this year “to say thank you for all the efforts in delivering the first stage of RB2.0”. Page 13 of 52 I am sure there are better alternatives too.”

[58]Mr Kapoor replies to the whole EC saying “We are all engaged with the issue of retention and I believe the Remco has agreed a very generous arrangement for a the vast majority of the top management for vesting in May 2019.” At this point the claimant is still an active participant in the EC and is receiving these communications.

[59]On 12 December 2018 Mr Barron writes to Mr Kapoor and Mr Hennah talking of discussions that Ms Harris had had concerning “LTIP vesting and retention”.

[60]By 20 December 2018 the first respondent had received news of an unexpectedly favourable resolution of a tax dispute in the US. The effect of this, and other accounting matters, was expected to result in payment under that year’s LTIP. Mr Hennah proposed to the Remco that the agreed changes to the LTIP (which had not been announced) should not apply, but that there should be a further review of likely vesting across 2019 and 2020, aiming at a vesting level of 60-65%, which he said was “supportive of the retention and incentives needed in the business”. A section headed “objectives” includes the following: “… this places us at heightened risk of loss of staff. And with less incentive for performance than is desirable. And with greater than usual exposure to the implications of both. With this context we have developed the following objectives for the next 2-3 years for assessing desirable changes to remuneration programmes, including potentially in-flight LTIP plans: 1. To retain the talent we need to retain. 2. To provide proper performance stretch. 3. To deliver consensus profit in 2018; and a balanced impact on profit in subsequent years. 4. To comply with the contractual and other commitments to our staff. And with commitments to our shareholders. 5. RB2.0 resilience.”

[61]The document goes on to talk of specific measures proposed to achieve those objectives, including various APP enhancements, “targeted enhancement for certain T400 and T40 staff where retention is a particular concern” and “adjusting the vesting level for the LTIP awards maturing at end 2019 and end 2020 such that, based upon delivery of the plan shared with the Board in November, the cost to the Group would equate to vesting at c. 60-65%.” Page 14 of 52

[62]On 22 December 2018 Ms Harris wrote to Mr Hennah to say that the Remco approved this approach in principle, subject to development of more detailed proposals.

[63]Thus although changes to vesting and the operation of the LTIP scheme for 2018 had been approved by the Remco, they were never implemented as the surprise tax resolution meant that for 2018 the LTIP would vest at a satisfactory level without the need for any changes to be made. Further discussions were awaited on changes for 2019 and beyond. Proposed amendments to the LTIP - 2019

[64]In mid-January 2019 it was announced that the then CEO was to stand down later in the year.

[65]On 6 February 2019 Mr Barron presented to the EC a paper on proposals he was expecting to make to the Remco. By this time the claimant was on garden leave and was no longer participating in EC meetings.

[66]The paper noted that projected vesting for 2017-19 and 2018-20 was zero. It says “the key objectives are retention of key talent and ensuring employees are incentivised appropriately”. It considers whether to make changes across the board, or only in respect of particular employees, recommending a “middle way” targeting various levels of LTIP vesting to particular categories of employees. It says “with retention being the most important consideration and having previously identified this will be targeted and selective, the proposal is to have no performance conditions. Vesting will be based on continuity of employment to vesting date.”

[67]The minutes of a subsequent EC meeting on 11 March 2019 record the EC’s endorsement of “an additional ‘special’ award that would apply if current awards do not vest”. This was described as “critical given need to retain top talent”.

[68]In June 2019 it was announced that Laxman Narasimhan would succeed Rakesh Kapoor as CEO of the first respondent. Mr Narasimhan was an external candidate and was new to the business.

[69]The claimant’s employment ended on 30 June 2019.

[70]On 15 July 2019 the EC is noted as having discussed the possibility of a “further restricted share allocation” based on performance “to avoid the likely scenario of further losses of top talent”.

[71]On 25 July 2019 the Remco met, with Mr Barron’s manager (the Chief Human Resources Officer) giving the Remco a “verbal update on the work to date regarding a broad based retention plan in light of the projected zero vesting of the 2017-2019 and 2018-2020 LTIPs.” The notes record that “the current thinking of the Executive Committee is to waive the performance conditions on Page 15 of 52 a proportion of the awards in order to give some level of vesting for participants”. The “rationale for this proposal” was said to be “a desire to retain senior managers through [the period of change the company is going through]”. The reference to a period of change must be a reference to the handover period in late-summer 2019 between the outgoing CEO Mr Kapoor and the incoming CEO Mr Narasimhan, along with any plans that the new CEO may have for developing and extending the RB 2.0 plans.

[72]On 11 September 2019 Mr Barron submitted to Ms Harris (for the Remco) a paper which he said “sets out the management proposal in respect of the inflight LTIP awards. As discussed at the last meeting, the proposal is for vesting of 50% in respect of the 2017 LTIP (vesting May 20) and the 2018 LTIP (vesting May 21).”

[73]That paper includes the following under the heading “context and objectives”: “one of [the new CEO’s] key findings … is the current fragility of the organisation, particularly amongst the senior leadership … the next two years will be equally challenging … Taken together, and alongside APP payments being lower than prior years, this gives us a high retention risk, especially amongst senior leadership. At RB the cultural imperative is to focus on performance. However, given the context it is proposed that the principal objective of any change to the 2017-2019 and 2018-2020 LTIP cycles is in retention. We still have APP for performance stretch, which also makes up a substantial proportion of the remuneration package …”

[74]The “proposal” includes: “- Waive the performance conditions on 50% of the LTIP awards for the 2017-2019 and 2018-2020 performance cycles. - Include all participants that remain employed at the date that the Remuneration Committee determine to make the change, including Executive Committee members but excluding Executive Directors (there being a rational and objective basis for this exclusion, namely that it would be contrary to the Remuneration Policy for Executive Directors). … - The 2017-2019 LTIP would ordinarily vest if a participant is employed to 31 December 2019 and would be released in May 2020, with the 2018-2020 LTIP vesting 31 December 2020 and released May 2021. However, it is proposed that for retention Page 16 of 52 purposes, the waiver of performance conditions will be subject to remaining employed to May 2020 and May 2021, respectively.”

[75]A further element to the proposal is “to make additional awards to selected participants, for further retention”.

[76]Ms Harris forwards this on to her Remco colleagues, saying “we have discussed this topic many times and even approved previous versions”. She says, “clearly we have a retention issue, and this is probably our best option to stabilise the organisation at the moment.” She talks of taking advice on whether the timing on vesting can be extended beyond May 2020 (something that had come up in discussion with a colleague on the Remco). The final amendment to the LTIP

[77]The minutes of the Remco meeting on 18 September 2019 record the following (they are quoted in full insofar as they relate to the “in-flight LTIPs”: “The Chair directed the meeting to the paper (circulated in advance of the meeting) titled “In-flight LTIP”, which was taken as read. The CEO was invited to set out for the Committee his findings since joining RB, the recommendations being made and the rationale. The recommendation to waive the performance conditions in respect of 50% of outstanding 2017-2019 LTIP and 2018-2020 LTIP (in respect of both share awards, and options) was discussed at length. In particular, the Committee considered rule 1.4 of the Reckitt Benckiser Group 2015 Long Term Incentive Plan and noted the Company, with the Committee’s consent, may “waive or change the Performance Condition […] if anything happens which causes the Company reasonably to consider it appropriate, provided that any changed Performance Conditions will be no more difficult to satisfy”. After due consideration, the Committee was satisfied: - the current retention issues faced by the organisation meant waiver or change of the performance conditions attaching to the 2017-2019 LTIP (collectively, the “Performance Conditions”) was appropriate; and - the recommendation would not cause the Performance Conditions to be more difficult to satisfy. The Committee agreed that they would review the 2018-2020 LTIP next year but that in the meantime they wish to maintain a substantial level of provisions to have flexibility, if the Committee considers it appropriate in light of circumstances at the time, to make adjustments to allow for a suitable level of vesting of these awards. The Committee believes that Page 17 of 52 maintaining provision for up to about 50% vesting of these awards is prudent and appropriate. The Committee discussed whether the recommendation to waive the Performance Conditions in respect of 50% of outstanding 2017-2019 LTIP was appropriate, given that if performance recovered, participants could receive vesting at more than 50% for potentially very modest performance. After due discussion, the Committee agreed waiver was not the best way forward, and instead it would be preferable to change the vesting schedule under the Performance Conditions as follows: Current ESP vesting Amended EPS vesting schedule schedule 3-year EPS Proportion of 3-year EPS Proportion of compound award compound award growth vesting growth vesting <6% 0% <7.5% 50% 6% - 10% 20% - 100% 7.5% - 10% 50% - 100% (straight-line (straight-line basis) basis) >10% 100% >10% 100% The Committee noted this approach would deliver no more than 50% vesting if EPS achievement remained modest (and no more than if the original EPS targets were retained), and would not cause the Performance Conditions to be more difficult to satisfy. The Committee requested that in making this change the ability for “re-testing” on the 2017-2019 awards also be removed, such that the awards that do not vest will lapse in May 2020. The Committee noted the recommendation excluded the Company’s executive directors, but included all other participants (including executive committee members) who are in employment at the time the waiver/change is to be effected. After due discussion, the Committee confirmed it agreed with the recommended approach, on the basis that all participants should be treated consistently, save that: Page 18 of 52 - exclusion of executive directors was required due to the need to be consistent with the Company’s shareholder-approved directors’ remuneration policy; and - exclusion of former employees (being good leavers) was appropriate, given the purpose of the waiver/change is to support employee retention. The Committee noted the recommendation that 2017-2019 LTIP only be allowed to vest subject to participants remaining in employment until the normal vesting date of May 2020 (subject to normal ‘good leaver’ exceptions), and, given the retention focus, supported this approach. The Committee noted and discussed the merits of the recommendations regarding additional retention awards to selected employees. The CEO highlighted that these would be made in the 2020 LTIP cycle as part of pay review. After due and careful consideration, IT WAS ACCORDINGLY RESOLVED THAT the Committee consents to the Performance Conditions attaching to 2017-2019 LTIP (held by all participants who are current group employees as at 18 September 2019 and who are not executive directors of the Company), in respect of both share awards and options, being amended pursuant to rule 1.4 of the Reckitt Benckiser Group 2015 Long Term Incentive Plan with effect from 18 September 2019 such that: 1. The EPS performance targets and vesting schedule is changed as follows: Current ESP vesting Amended EPS vesting schedule schedule 3-year EPS Proportion of 3-year EPS Proportion of compound award compound award growth vesting growth vesting <6% 0% <7.5% 50% 6% - 10% 20% - 100% 7.5% - 10% 50% - 100% (straight-line (straight-line basis) basis) >10% 100% >10% 100% and Page 19 of 52 2. the following term is deleted “For the purposes of Rule 6.1 only (Leaving the Group), an Award will be treated as a Vested Award from the end of the ﬁnancial year of the performance period.” and 3. Where relevant, the following sentence is deleted: "If, following the satisfaction of an EPS Growth Condition, the EPS Compound Annual Growth in any subsequent Financial Year (up to and including Financial Year 2021) reaches a level of EPS Compound Annual Growth for the three-year period then ended which meets a threshold which is higher than the EPS Compound Annual Growth threshold achieved in satisfying the EPS Growth Condition in the previous year, the Participant shall become entitled to the additional proportion(s) of the Share Option/Award Shares." and accordingly any unvested portion of the award shall lapse no later than 31 May 2020.”

[78]The effect of the changes was: 1. For those who met the conditions of the amended LTIP there would be a minimum vesting of 50% of their award, as opposed to a minimum vesting of 0% under the original scheme. There were consequential amendments allowing for vesting over 50%, but it does not seem to ever have been in contemplation that those higher targets would be met. The essence of this change was a relaxation of the performance condition so that everyone who qualified under the amended scheme achieved a 50% vesting of their awards. 2. The consequence of the second change is not immediately obvious, but it was described in an private internal briefing document as having the following effect: “… you will have to remain with RB until the vesting date at the AGM in May 2020 to receive these shares and options. As with the normal annual LTIP grant, the Rules of the LTIP will govern leaver provisions. If you resign then the shares will lapse. If you leave as a good leaver on or after 18 September 2019, the treatment of your shares and options will be set out in your leaver agreement; the usual practice is that you will receive a pro-rated amount of the 2017-2019 LTIP up to your leaving date.” 3. The opportunity for “re-testing” was removed from those it applied for.

[79]The changes only applied to those who were employees as of 18 September 2019. For everyone else (including executive directors) the scheme continued Page 20 of 52 unamended. As the claimant was not an employee as of 18 September 2019 he continued to be subject only to the unamended scheme.

[80]The original (unamended) performance condition was not met, and the claimant received no award under the 2017 LTIP. Those who were subject to the amended scheme and otherwise met its requirements would receive at least 50% of their award.

[81]The private internal briefing document referred to above appeared to anticipate complaints from people in the claimant’s position, and includes the following passage: “Q. I have already left as a good leaver why am I being treated differently than others? A. The intention of this change to recognise the expected future contribution of participants over the next 8-9 months as well as the hard work and resilience over recent years. As such, the Committee determined that the change will only apply to employees still employed at the date they made the change, and who remain employed to the vesting date in May 2020. Employees who have already left have already been informed of the treatment of their LTIPs in their leaver agreement.”

[82]The change was announced by the CEO in an email to affected employees on 9 October 2019. The message of the document can be summed up in a short extract: “What this means for you is that if you remain employed to the vesting date in May 2020 you will receive vesting of at least half of the shares and options awarded to you under the 2017-2019 LTIP.”

[83]The claimant was no longer an employee and did not receive this communication. Subsequent events Rob de Groot

[84]The claimant draws attention to the position of Rob de Groot. Mr de Groot was a long-serving employee of the first respondent’s group, and at the time of the claimant’s retirement was the President of the Hygiene and Home division. The claimant says that Mr de Groot was “the strongest internal candidate for the CEO role”, although it is not clear whether he ever put himself forward as a candidate for the role.

[85]In December 2019 it was announced that Mr de Groot was to leave the first respondent’s group in February 2020. In fact he remained an employee until 31 Page 21 of 52 May 2020 on the terms of a retirement agreement. His is entry number 9 on the list of leavers agreements set out below. As the claimant points out, a consequence of Mr de Groot’s leavers agreement was that he benefitted from the amended LTIP.

[86]Irrespective of the legal position, the claimant is aggrieved that Mr de Groot was able to benefit from the amended performance condition whereas he was not. The claimant considers that both of them should have been considered under the amended performance condition. We can understand the claimant feeling that way, but in this case we must apply the law, not any more general concepts of fairness between colleagues (or ex-colleagues). The claimant

[87]In late 2019 and early 2020 the claimant was in Milan, at a time when Covid-19 was first emerging and Lombardy was particularly badly hit. He was caught up in one of the very earliest Covid-19 lockdowns in Europe. Although no longer an employee he stayed in touch with the first respondent’s Chief Human Resources Officer on a personal basis. She was at the time in her last few months of employment. The claimant had no knowledge of the amendments to the LTIP at that time.

[88]He maintained an interest in the first respondent, and in April 2020 he saw from the first respondent’s annual report that the executive directors had not received any awards under the LTIP. This is what would be expected given the original terms of the scheme. Due to regulatory requirements the amended performance condition and other changes had not applied to the two executive directors.

[89]It was said for the respondents that it would have been clear from a reading of the annual report that the performance conditions had been relaxed for others – but it certainly was not clear to us that this was the case, and we consider that it would have taken an expert eye to glean this information from the annual report. It is not something that is explicitly stated in the report and appears (if evident at all) to require a detailed reading of the small print.

[90]He says that he remembered the 2018 discussions about amendment to the LTIP and raised them with the outgoing CHRO shortly after publication of the annual report. She told him about the amendments and, when he asked, about how these arrangements had affected Mr de Groot.

[91]Shortly after this, the claimant contacted Mr Barron. He arranged a call with him for 9 April 2020. Mr Barron reported this call to the outgoing and incoming CHRO the same day. He said: “[the claimant] called me today … Page 22 of 52 He “had heard” that Laxman sent an all employee comms saying that vesting of the LTIP was going to be 50% and he wanted to check if it applied to him. I told him that it only applies to active employees at September 2019, when the Remco decision was taken, that remain employed to May 2020 vesting and as such his award is subject to the performance conditions on his original award, which have unfortunately not vested.”

[92]The claimant appears to have gone on to ask for special consideration to be given to his case, including specific consideration by the CEO. Mr Barron’s note to the CHROs continues: “He wants to know that the ruling applies to him specifically and consciously a decision we have taken that he doesn’t get it, rather than a general rule he has got caught up in. He believes the spirit of the retirement agreement is that he should get the vesting. He has asked for a response from me that confirms either way. As a reminder … the exclusion of good leavers who left before the decision was made was a conscious decision – the Committee determined it was appropriate to exclude them as the purpose of the change was to support retention. I therefore assume that we are not minded to give an exception to [the claimant] after the fact. I suggest that I draft a response to him that he is not eligible …”

[93]Mr Barron concludes by asking the CHROs to check with the CEO whether an exception is to be made.

[94]On 13 April 2020 Mr Barron sent an email to the claimant saying: “As you raised when we spoke, the Remuneration Committee made changes to the performance conditions that apply to a group of participants. The purpose of this change was to support retention of employees and so they decided that the change in performance conditions only apply to active employees at the date they made the decision in September 2019. Given the retention focus they also determined that the award would only vest to participants remaining in employment until the vesting date in May 2020. Therefore, as discussed, the performance conditions that apply to your award are those set out on your award certificate, requiring EPS growth in excess of 6% p.a. for vesting, which have not been met. … I am sorry that this is not better news, but I hope that you understand the intention behind the Remuneration Committee’s actions.” Page 23 of 52

[95]On 23 April 2020 the claimant took the question up directly with Ms Harris, but to no avail. He undertook a period of ACAS early conciliation between 22 and 26 May 2020, before submitting his tribunal claim on 3 July 2020. It was originally submitted as a claim against the first respondent only, with the claim against the second respondent being submitted (by way of an application to amend his claim) on 7 October 2020. The amendments requested by the claimant were permitted at the case management hearing on 24 February 2021. Statistics on those affected by the amendment Generally

[96]We were provided by the respondents with two tables setting out the ages of those who have or have not been affected by this change. They are set out below in their entirely. In each case, the 58 year old referred to under heading 1.2 is the claimant. The second respondent only

[97]The first was included as an appendix to Mr Barron’s statement, and included the second respondent’s employees only (it being the respondents’ case that this was the relevant “pool”): LTIP Age Data for Employees of Reckitt Benckiser Health Limited (ages as at 18 September 2019) 1 Ages of good leavers prior to 18 September 2019 1.2 Ages of good leavers from (and including) 1 January 2019 and prior to 18 September 2019 58, 46 43, 41, 44, 47, 38, 43, 40, 49, 43, 51 3 Ages of good leavers in the period between 18 September 2019 and 12 May 2020 49, 45, 32 4 Ages of bad leavers in the period between 18 September 2019 and 12 May 2020 49, 38, 43 Page 24 of 52 5 Ages of those in employment at May 2020 53, 54, 51, 43, 44, 48, 46, 48, 43, 46, 40, 40, 49, 43, 64, 35, 38, 43, 41, 42, 46, 50, 43, 45, 43, 36, 38, 51, 45, 41, 35, 47, 41, 40, 48, 42, 48, 61, 46, 53, 54, 43, 48, 49, 51, 44, 51 The first respondent’s group

[98]The second table covered the entirety of those participating in the LTIP – i.e. the whole of the first respondent’s group. This had been amended and we understand the final form of the table to be that at p1274 of the bundle. We have retained the deletions to show where the amendments were made. We have amended the numbering of the headings to match with those in Mr Barron’s statement. LTIP Age Data (ages as at 18 September 2019) 1. Ages of good leavers prior to 18 September 2019: 40, 45, 40, 49, 64, 40, 52, 45, 37, 56, 54, 43, 41, 55, 60, 40, 45, 43, 44, 44, 45, 59, 54, 53, 46, 45, 60, 43, 51, 51, 51, 63, 49, 52, 42, 42, 52, 40, 46, 53, 57, 42, 48, 59, 61, 39, 56, 55, 41, 53, 43, 49, 45, 49, 54, 38 1.2 Ages of good leavers from (and including) 1 January 2019 and prior to 18 September 2019 58, 61, 47, 43, 48, 56, 57, 49, 50, 43, 57, 51, 46, 36, 53, 39, 46, 42, 43, 54, 40, 47, 49, 39 47, 37, 36, 43, 41, 43, 48, 41, 45, 44, 45, 38, 47, 38, 48, 43, 44, 36, 53, 47, 54, 48, 40, 49, 43, 51, 38, 46, 40, 50, 47, 38, 49, 46, 41, 53, 53, 44, 50, 37, 46, 42, 51, 51, 48, 45, 53, 43, 43, 40, 41 3 Ages of good leavers in the period between 18 September 2019 and 12 May 2020 50, 46, 42, 49, 46, 37, 51, 45, 39, 32, 50, 44 4 Ages of bad leavers in the period between 18 September 2019 and 12 May 2020 49, 49, 48, 46, 43, 44, 41, 41, 37, 50, 38, 41, 42, 37, 38 5 Ages of those in employment at May 2020 Page 25 of 52 50, 52, 48, 45, 49, 53, 43, 45, 57, 38, 44, 62, 47, 41, 54, 47, 46, 51, 432, 52, 43, 58, 50, 45, 52, 40, 44, 44, 51, 53, 43, 48, 41, 46, 59, 48, 42, 53, 50, 58, 44, 43, 54, 50, 47, 46, 40, 35, 63, 40, 48, 46, 49, 43, 64, 50, 38, 41, 52, 50, 46, 47, 35, 40, 38, 38, 44, 43, 52, 50, 39, 38, 46, 48, 56, 35, 43, 48, 36, 50, 63, 41, 38, 41, 43, 42, 46, 44, 57, 54, 50, 41, 40, 43, 50, 45, 50, 44, 46, 48, 47, 42, 42, 51, 52, 38, 41, 48, 51, 48, 37, 454, 51, 51, 43, 36, 46, 36, 36, 46, 42, 38, 42, 51, 45, 43, 48, 43, 46, 37, 46, 41, 39, 48, 57, 42, 56, 43, 35, 41, 47, 40, 41, 49, 47, 45, 41, 43, 53, 38, 44, 42, 45, 49, 47, 61, 42, 40, 44, 55, 57, 40, 39, 55, 51, 37, 42, 54, 52, 49, 41, 37, 41, 42, 40, 43, 40, 47, 46, 48, 39, 36, 45, 45, 49, 43, 41, 36, 44, 62, 60, 60, 43, 51, 48, 43, 40, 42, 44, 51, 41, 48, 45, 46, 34, 61, 36, 51, 42, 40, 46, 47, 53, 36, 38, 44, 45, 44, 54, 36, 52, 51, 53, 41, 40, 50, 43, 48, 49, 51, 49, 48, 50, 44, 42, 47, 54, 51, 49, 45, 43, 38, 38, 54, 43, 52 Leaver agreements

[99]A further category of potentially relevant employees had been identified during in preparation for the hearing. These were those who had left under “leaver agreements” providing for them to be treated as “good leavers”, in a similar way to the claimant. The table set out below appears at pages 772-3 of the tribunal bundle. Leaver Date of Agreement Agreed Agreed termination Number commencement date garden leave 1. 10 April 2019

1 June 2019

31 October 2019 2. 26 April 2019

1 July 2019

31 December 2019 3. 24 July 2019

30 May 2019

30 November 2019 4. 7 August 2019

8 September 2019

19 January 2020 5. 10 September 2019

12 October 2019

31 December 2019 6. 16 September 2019 n/a 1 November 2019 7. 28 October 2019 n/a 28 October 2019 8. 21 November 2019 n/a 1 March 2020 9. 26 November 2019

1 March 2020

31 May 2020 10. 19 December 2019 n/a 31 January 2020 Page 26 of 52 11. 14 January 2020

28 October 2019

2 February 2020 12. 23 January 2020

1 March 2020

1 April 2020 13. 5 March 2020 n/a 14 April 2020

[100]Each of those individuals were “good leavers” who were in employment as of 18 September 2019. As such, they all benefitted from the amended performance condition. The claimant pointed out that the first six of these benefitted from the amended performance condition despite already having (as at 18 September 2019) entered into agreements providing for their employment to end. C. THE LAW The basis of

The claim

[101]The PCP alleged is said by the claimant to be a PCP applied to him by both respondents (para 17 of his amended particulars of claim). The application of the PCP is said to be an act of indirect age discrimination by both respondents.

[102]The second respondent was the claimant’s employer. The second respondent is said to liable for the application of the PCP on that basis, with the first respondent being liable as its agent and/or on the basis that it knowingly helped the second respondent breach the Equality Act 2010 (“EqA”) (para 23 of his amended particulars of claim).

[103]If there was any unlawful discrimination, it occurred after the end of the claimant’s employment and so (although not referred to as such in his amended particulars of claim) must be founded on section 108(1) of the EqA: “A person(a) (A) must not discriminate against another

(b) (B) if: (a) the discrimination arises out of and is closely connected to a relationship which used to exist between them, and (b) conduct of a description constituting the discrimination would, if it occurred during the relationship, contravene this Act.

[104]It is not disputed that if there is discrimination in this case it “arises out of and is closely connected to a relationship which used to exist between [the relevant respondent and that claimant]”

[105]Indirect discrimination is dealt with by s19 of the EqA: Page 27 of 52 “(1) A person(a) (A) discriminates against another

(b) (B) if A applies to B a provision, criterion or practice which is discriminatory in relation to a relevant protected characteristic of B's. (2) For the purposes of subsection (1), a provision, criterion or practice is discriminatory in relation to a relevant protected characteristic of B's if: (a) A applies, or would apply, it to persons with whom B does not share the characteristic, (b) it puts, or would put, persons with whom B shares the characteristic at a particular disadvantage when compared with persons with whom B does not share it,

(d) A cannot show it to be a proportionate means of achieving a legitimate aim.”

[106]Section 39 renders employers liable for discrimination in the following way: “(1) An employer(a) (A) must not discriminate against a person (B): (a) in the arrangements A makes for deciding to whom to offer employment;

(b) as to the terms on which A offers B employment;

(c) by not offering B employment. (2) An employer (A) must not discriminate against an employee of A's (B): (a) as to B's terms of employment; (b) in the way A affords B access, or by not affording B access, to opportunities for promotion, transfer or training or for receiving any other benefit, facility or service; …

(d) by subjecting B to any other detriment.”

[107]Liability for the acts of agents arises under s109(2): “Anything done by an agent for a principal, with the authority of the principal, must be treated as also done by the principal.”

[108]Liability as an agent arises under s110(1): Page 28 of 52 “A person(a) (A) contravenes this section if: (a) A is an … agent,

(b) A does something which, by virtue of section 109 … (2), is treated as having been done by A's … principal (as the case may be), and

[109]Liability of those knowingly helping with a contravention of the EqA arises under s112(1): “A person(a) (A) must not knowingly help another

(b) (B) to do anything which contravenes Part 3, 4, 5, 6 or 7 or section 108(1) or (2) or 111.”

[110]“Knowingly” in this context is addressed in Allaway v Reilly

[2007] IRLR 864

in the following way: “If a fellow employee does an act in the course of his employment which has the effect of discriminating against the claimant employee on grounds of sex and that is a result which can be concluded to have been within his knowledge at the time he carried out the act in question, the requirements of s.42(1) are met. Discrimination does not have to be what he intended, nor must it have been his motive. It is enough that, on the evidence, the conclusion can be drawn that discrimination as the probable outcome was within the scope of his knowledge at that time. It would not need to be in the forefront of his mind, nor would he need to have specifically addressed his mind to it. It must be enough if, in all the circumstances, it can properly be concluded that it was within the knowledge that was possessed by the alleged discriminator. Inevitably, whether or not it can be concluded that the alleged discriminator did an act “knowingly” is going to depend on the facts and circumstances of each particular case in which it is arguable that, on the available material, that is something which can be established.” Agency

[111]Both parties agree that agency for the purposes of the EqA is to be understood in the same way as the common law concept of agency (see, for instance, Kemeh v Ministry of Defence

[2014] ICR 625

and Blackwood v Birmingham & Solihull Mental Health NHS Foundation Trust

[2018] IRLR 878

). There was, however, considerable difference of emphasis on how that definition should be applied in this case.

[112]Both parties look to Bowstead & Reynolds on Agency for the definition of agency. The basic definition is set out in the first paragraph of the book: Page 29 of 52 “Agency is the ﬁduciary relationship which exists between two persons, one of whom expressly or impliedly manifests assent that the other should act on his behalf so as to affect his legal relations with third parties, and the other of whom similarly manifests assent so to act or so acts pursuant to the manifestation.”

[113]At para 1-004: “The mere fact that one person does something in order to benefit another … does not make the former the agent of the latter.”

[114]And in relation to “control” (para 1-018): “It is common to regard control by the principal as a deﬁning characteristic of agency ... In agency in general however it plays a more limited role. … Agents will often not accept control by their principals as to the manner in which they act, and some will only accept instructions to act in accordance with usages of their own market. Others may be authorised only to do speciﬁc things. In many such situations the principal’s only control lies in his power to revoke the authority, a power which agency law assumes that he has at all times. It might seem therefore that control is not a signiﬁcant feature of the internal relationship, except in so far as the relationship by deﬁnition posits a person, the principal, giving authority, and the agent’s duty to obey instructions if the latter wishes to continue as agent. Nevertheless, if the principal gives up all control of his supposed agent the relationship is only doubtfully one of agency. The idea of control may also be relevant where it is contended that a company is an agent of its parent. For this reason the idea requires mention; but … the idea of control does not seem sufﬁciently important to be inserted into the formal deﬁnition [of agency].” PCPs, pools and disadvantage

[115]The PCP alleged is: “requiring LTIP participants (who were not already bad leavers) to be employed as of 18 September 2019 (or, in the alternative, May 2020) in order for them to benefit from the amended performance condition”

[116]It is for the claimant to show that this PCP existed, was applied by the respondent(s) and put people aged 57 or over at a particular disadvantage compared with those aged under 57. In this context, the word “particular” is redundant, and what needs to be shown is simply a disadvantage, not any special form of disadvantage (McNeil v Revenue and Customs Commissioners Page 30 of 52

[2019] EWCA Civ 1112

). There is no longer the requirement in the former discrimination statutes to show that the proportion of people aged 57 or over who can comply with the PCP is “considerably smaller” than those aged under 57 who can comply.

[117]There are several ways in which the “particular disadvantage” can be demonstrated. One of them, and the method relied upon by the claimant in this case, is using statistics. Statistics are not necessary to show the disadvantage, but they can be used to do so. If the statistics demonstrate particular disadvantage, it is not necessary to demonstrate that that disadvantage arises specifically because of the protected characteristic (Essop v Home Office

[2017] ICR 640

[118]There is no need for everyone with the protected characteristic to be disadvantaged. “All that is needed is a particular disadvantage when compared with other people who do not share the characteristic in question.” (Homer v Chief Constable of West Yorkshire Police

[2012] UKSC 15

[119]To consider statistics, a “pool” of affected individuals must be identified. The identification of the pool follows “as a matter of logic” from the PCP (Dobson v North Cumbria Integrated Care NHS Trust

[2021] IRLR 729

). Different pools may be available to the tribunal. If so, the tribunal must “select … the pool which they consider will realistically and effectively test the particular allegations before them” (Ministry of Defence v DeBique

[2010] IRLR 471

[120]Under the EHRC Code of Practice, para 4.18, “the pool should consist of the group which the provision, criterion or practice affects (or would affect) either positively and negatively, while excluding workers who are not affected by it, either positively or negatively”.

[121]Caution is raised in Perera v Civil Service Commission

[1982] IRLR 147

about the reliability of small samples in showing particular disadvantage. Similarly, para 4.13 of the ECHR Code of Practice talks of statistical analysis possibly being inappropriate where “the numbers of people are too small to allow for a statistically significant comparison”. In R (Seymour-Smith) v Secretary of State for Employment (No. 2)

[2000] ICR 244

tribunals were cautioned on their response to statistics that “cover a short period or … present an uneven picture”.

[122]At L[321.03] of Harvey on Industrial Relations and Employment Law, the point is put this way: “the [tribunal] must assess whether any apparent disadvantage is real and linked to the PCP or ephemeral or some kind of fluke”. Justification

[123]It is for the respondents to show that any PCP that puts those aged over 57 (and put the claimant) at a disadvantage is objectively justified. This means that it must be a proportionate means of achieving a legitimate aim. Page 31 of 52

[124]Justification is to be assessed on an “objective” basis. There is no “range of reasonable responses” that applies. The tribunal must assess for itself whether the PCP is justified, not whether the respondents acted in good faith or themselves thought that the PCP was justified. As cited by the claimant: “It is not enough that a reasonable employer might think the criterion justified. The tribunal itself has to weight the real needs of the undertaking against the discriminatory effects of the requirement” (Hardy & Hansons plc v Lax

[2005] EWCA Civ 846

[125]The aims relied upon by the respondents are retention and incentivisation of staff. There is no dispute that those can amount to legitimate aims. However that “is only the beginning of the story. It is still necessary to inquire whether it is in fact the aim being pursued” (Seldon v Clarkson, Wright and Jakes

[2012] ICR 716

(para 59)).

[126]At para 50 of Seldon, the following appears: “the measure in question must be both appropriate to achieve its legitimate aim or aims and necessary in order to do so ... The gravity of the effect upon the employees discriminated against has to be weighed against the importance of the legitimate aims in assessing the necessity of the particular measure chosen.”

[127]The respondents referred to para 29 of Essop, in which Lady Hale said: “… it is always open to the respondent to show that his PCP is justified ... Some reluctance to reach this point can be detected in the cases, yet there should not be. There is no finding of unlawful discrimination until all four elements of the definition are met. The requirement to justify a PCP should not be seen as placing an unreasonable burden upon respondents. Nor should it be seen as casting some sort of shadow or stigma upon them. There is no shame in it. There may well be very good reasons for the PCP in question …”

[128]Much of this was summarised by Elias J in MacCulloch v ICI

[2008] IRLR 846

, EAT, where he set out four legal principles regarding justification (which have since been approved by the Court of Appeal in Lockwood v DWP

[2013] IRLR 941

): ''(1) The burden of proof is on the respondent to establish justification: see Starmer v British Airways

[2005] IRLR 862

at [31]. (2) The classic test was set out in Bilka-Kaufhaus GmbH v Weber Von Hartz (case 170/84)

[1984] IRLR 317

in the context of indirect sex discrimination. The ECJ said that the court or tribunal must be satisfied that the measures must “correspond to a real need … are appropriate with a view to achieving the objectives pursued Page 32 of 52 and are necessary to that end” (paragraph 36). This involves the application of the proportionality principle, which is the language used in reg. 3 itself. It has subsequently been emphasised that the reference to “necessary” means “reasonably necessary”: see Rainey v Greater Glasgow Health Board (HL)

[1987] IRLR 26

per Lord Keith of Kinkel at pp.30–31. (3) The principle of proportionality requires an objective balance to be struck between the discriminatory effect of the measure and the needs of the undertaking. The more serious the disparate adverse impact, the more cogent must be the justification for it: Hardys & Hansons plc v Lax

[2005] IRLR 726

per Pill LJ at paragraphs [19]– [34], Thomas LJ at [54]–[55] and Gage LJ at [60]. (4) It is for the employment tribunal to weigh the reasonable needs of the undertaking against the discriminatory effect of the employer's measure and to make its own assessment of whether the former outweigh the latter. There is no “range of reasonable response” test in this context: Hardys & Hansons plc v Lax

[2005] IRLR 726

, CA.''

[129]The respondents also cited three authorities on the “drawing of lines” - such as cut-off dates, and how this is to be approached by the tribunal.

[130]First, there is R (Gurung) v The Secretary of State for Defence

[2008] EWHC 1496

, cited by the respondents in the following terms: “Ouseley J noted that the drawing of lines (where those who fall on one side of the line get a benefit and those who fall on the other side of the line do not) may give rise to age disparities: “but that is a weak starting point for an assertion of indirect discrimination on age grounds” which would be defeated where there was a “rational basis for the selection of the date as at which the changes were made.”

[131]Second, “anomalies at the margins” should not trouble a tribunal. It is inevitable that if a cut-off date or age is set there may only be a difference of a day or so between those who are advantaged and those who are disadvantaged. In R (Carson) v Secretary of State for Work and Pensions

[2006] 1 AC 173

Lord Hoffmann said: “Mr Gill emphasised that the twenty fifth birthday was a very arbitrary line. There could be no relevant difference between a person the day before and the day after his or her birthday. That is true, but a line must be drawn somewhere. All that is necessary is that it should reflect a difference between the substantial majority of the people on either side of the line. If one wants to analyse the question pedantically, a person one day under 25 is in an analogous, indeed virtually identical, situation to a person aged 25 but there is an objective justification for such Page 33 of 52 discrimination, namely the need for legal certainty and a workable rule …”

[132]Finally on this point, in Air Products v Cockram

[2018] IRLR 755

(itself a case dealing with an LTIP) Bean LJ said: “Pointing … to the discrepancy between an employee of 54 who has unvested LTIPs on retirement and a 55-year-old who does not is not a convincing argument. Bright line rules are a common feature of benefits payable on retirement. And as Baroness Hale JSC observed in Seldon, ‘where it is justified to have a general rule then the existence of that rule will usually justify the treatment which results from it.’”

[133]We also note the following extracts from the EHRC Code of Practice, cited by the claimant (paras 4.26, 4.28 and 4.29): “… it is up to the employer to produce evidence to support their assertion that it is justified. Generalisations will not be sufficient to provide justification … The aim of the PCP should be legal, should not be discriminatory in itself and must represent a real objective consideration. … an employer solely aiming to reduce costs cannot expect to satisfy the test.” Time limits

[134]Time limits in discrimination claims are dealt with at section 123 of the Equality Act 2010: “(1) Subject to [extensions for early conciliation] proceedings … may not be brought after the end of:(a) the period of 3 months starting with the date of the act to which the complaint relates, or

(b) such other period as the employment tribunal thinks just and equitable. … (3) For the purposes of this section: (a) conduct extending over a period is to be treated as done at the end of the period; (b) failure to do something is to be treated as occurring when the person in question decided on it. Page 34 of 52 (4) In the absence of evidence to the contrary, a person (P) is to be taken to decide on failure to do something: (a) when P does an act inconsistent with doing it, or (b) if P does no inconsistent act, on the expiry of the period in which P might reasonably have been expected to do it.”

[135]We will set out in our discussions and conclusions the rival contentions of the parties as regards time limits. D. DISCUSSION AND CONCLUSIONS Agency

[136]The question of agency (or knowingly helping with discrimination) is linked to the later argument by the respondents that the PCP (if it existed at all) was not applied by either respondent to the claimant.

[137]It is necessary to consider this point because of the corporate structure. It is agreed that the second respondent was the claimant’s employer, but the first respondent was the body that operated the LTIP, and the Remco (which decided on the change to the LTIP) was established by and part of the first respondent, not the second respondent.

[138]The parties agree that the concept of agency in discrimination law is to be interpreted in line with the common law understanding of agency. However, they differ in how that is to be applied in this case, and whether the first respondent can be considered to be the agent of the second respondent in respect of the changes to the LTIP.

[139]The claimant starts from the proposition that a purposive approach is required: “in the employment context … legal principles such as agency … should be approached purposively to give effect to the requirement that those, the subject of discrimination, should have ready access to remedies” (Remploy Limited v Campbell & London Borough of Redbridge

UKEAT/0550/12

[140]The respondents rely on a strict approach to the need for “control” in an agency relationship, and question how it can be said that a subsidiary controls the actions of its parent company. They also point out that a group structure does not imply that one group company is the agent of another (“situations where there is such agency would be required to be proved by normal criteria” (para 1-030 of Bowstead and Reynolds)). They point out that to the extent agency is ever mentioned in any of the documentation it is the second respondent that is said to be the first respondent’s agent (in the claimant’s retirement agreement), not the other way around. The LTIP was also explicitly stated not to be part of a contract of employment. There was no obligation on the second respondent Page 35 of 52 to provide LTIP benefits – any arrangements for provision of such benefits by the second respondent was entirely discretionary.

[141]By contrast, the claimant says that it is consent, not control, that is at the heart of the agency relationship – that is, consent on the part of the principal to the agent acting on their behalf and consent on the part of the agent to act in that manner. There is no need for any express agreement for one party to act as agent of another. The claimant says that the second respondent consented to the first respondent providing and administering the LTIP scheme. The amendment to his contract says that he can participate in his employer’s LTIP scheme. There has never been any LTIP scheme other than that operated by the first respondent, so the second respondent must have delegated operation of the scheme to the first respondent in order to fulfil its obligations.

[142]We accept the claimant’s argument that a purposive approach is appropriate when considering the question of agency in this context. It is striking that if the respondents are right that there is no agency (or knowing help) then there could be no liability in these circumstances even for the most outrageous discrimination. What if, for instance, the first respondent had decided that the LTIP should only be available to male employees? On the respondents’ argument, neither respondent would be liable for such blatant discrimination and an employee subject to such discrimination would have no remedy. That would be a surprising result, and more so when it is common for parent companies to operate benefit schemes that apply to employees of their subsidiary companies, particularly where entitlements to shares are concerned.

[143]The claimant’s contract of employment says: “The Executive shall be entitled to participate in retirement, welfare benefits, incentive compensation, perquisite and other plans and arrangements of the Company applicable to its senior executives as in effect and on such terms as are prescribed by the Company in its sole discretion from time to time. Such participation includes any awards granted under the “Long Term Incentive Plan” of a Group Company.”

[144]The claimant thus has an entitlement to participation in benefit schemes (including the first respondent’s LTIP) “on such terms as are prescribed by the [second respondent] in its sole discretion from time to time”.

[145]In November 2016 as in all relevant previous years he was in fact granted an award under the first respondent’s LTIP. It is not necessary for us to decide whether this was a contractual entitlement as such, but we note that it did appear to form part of the package of benefits with which the second respondent sought to ensure (amongst other things) that he continued to find them an attractive employer to work for, and that it appears almost unthinkable that he would not have received a grant for so long as the scheme continued to exist. Page 36 of 52

[146]It appears reasonably clear that in providing the LTIP scheme the first respondent is acting on behalf of the second respondent in providing a benefit to the second respondent’s employees. The second respondent is the one that benefits from this by (at least in on the respondents’ argument) retaining and incentivising employees. The first respondent benefits indirectly through what must be intended as greater profitability throughout its group. The question that follows is whether this is done, strictly speaking, through the means of agency as “the mere fact that one person does something in order to benefit another … does not make the former the agent of the latter.”

[147]We accept that the concept of “control” is not of great significance in this context. The claimant is correct to emphasise that it is “consent” that is the more important factor, and here there clearly is that consent from the second respondent to the first respondent operating the LTIP scheme and, most likely, other benefit schemes on behalf of it.

[148]Looking at the definition of agency from the first paragraph of Bowstead & Reynolds, the second respondent has assented to the first respondent acting on its behalf in the provision of the LTIP, and there appears to be no doubt that the first respondent also assents to acting on its behalf. As far as we can tell the first respondent has very few direct employees – possibly only the two executive directors. The entire basis and purpose of the LTIPs is that they are offered by the first respondent to the employees of its subsidiary companies, such as the second respondent. It was not argued on behalf of the respondents that the relationship (if it exists) is not a fiduciary one. The question that remains is whether this is intended to affect the second respondent’s legal relations with third parties.

[149]We have said before that we will not trespass on the question of whether the LTIP is a contractual entitlement. That may need to be determined in different proceedings. However, it seems to us that in this situation the first respondent’s handling of the LTIPs is liable to affect the second respondent’s legal relationship with third parties – its employees such as the claimant. What if, for instance, the first respondent had decided to entirely withdraw the an LTIP during its currency for an entirely arbitrary reason? Even without addressing the direct question of whether this would be a breach of contract, it is not difficult to see it being a breach of (or contributing to) a breach of the duty of trust and confidence owed by the second respondent to is employees. Other legal duties may well be engaged, depending on the facts of the particular case.

[150]We have concluded that the LTIP scheme was offered by the first respondent to the second respondent’s employees as agent for the second respondent. As noted above, we draw this conclusion from the common law definition of agency, but it also seems to us that this must be the right conclusion if a properly purposive approach is taken to the definition of agency. To find otherwise would leave a large gap in protection against unlawful discrimination, Page 37 of 52 where the action in question is taken by another company in a respondent’s group.

[151]The outcome of this is that if the application of the PCP did amount to unlawful indirect age discrimination, then:a. The second respondent is treated as having carried out the action of its agent, the first respondent, by virtue of s109(2), and

b. The first respondent is also liable for that discrimination by virtue of s110(1). Knowing help

[152]As for “knowing help”, we do not see that that applies in this case. The claimant’s submission was that: “whilst the discriminatory impact of the decision was not at the fore of the [Remco]’s mind, nor was it their intention, by dint of being aware of the need to consider potential discrimination, the [Remco] would have been aware that discrimination was a possible outcome, their failure to consider any such discriminatory impact and/or to assess mitigating such impact means that it becomes probable. As such, it was within the First Respondent’s scope of knowledge at that time of the decision.”

[153]The logic of this appears to be that knowing that indirect discrimination was a “possible outcome” of a decision (as it would be with almost any meaningful decision made in relation to employee benefits), and not investigating any further renders discrimination a “probable outcome” of the decision. That cannot be right. What is required for someone to “knowingly help” another with discrimination is that “discrimination as the probable outcome was within the scope of [the party’s] knowledge at that time”. We do not see anything in the evidence before us to suggest that the first respondent (or Remco) considered discrimination to be a “probable outcome” of the Remco’s amendments to the LTIP. There is no liability by way of “knowing help” in this case. The PCP

[154]The respondents have criticised the claimant’s formulation of the PCP, calling it “confused”. They say there is no need to refer to either good leavers or bad leavers in the PCP. Nothing in the decision of 18 September 2019 changed the underlying rules in relation to good leavers and bad leavers. They point out the anomalous position of the two executive directors. The executive directors were LTIP participants but would not benefit from the amended performance condition even if they were in employment on 18 September 2019 and May 2020. Page 38 of 52

[155]We do not think that these arguments substantially undermine the formulation of the PCP set out by the claimant. Perhaps it was not necessary to refer explicitly to bad leavers in the PCP, but fundamentally the PCP is made out. It is clear from the resolution of the Remco that participants would have to be employed as of 18 September 2019 in order to benefit from the amended performance condition and (in general) would also have to be employed as at May 2020. There are some exceptions to that, but those limited exceptions, such as the executive directors, do not undermine the formulation of the PCP set out by the claimant. The claimant has shown that that PCP exists.

[156]In their closing arguments, the respondents focussed more on the question of whether the PCP can be said to have been applied by either of the respondents. That is closely linked to the arguments on agency that we have discussed above.

[157]Having found that the first respondent was an agent of the second respondent in offering the LTIP scheme (and subsequently amending it) we do not see that there can be any doubt that the PCP in question was applied to the claimant. He was a participant in the LTIP, subject to its rules, and then was subject to the PCP when the LTIP was amended. Disadvantage

[158]The claimant relies on statistics to demonstrate particular disadvantage. The first thing to be addressed in looking at statistics is the formation of the relevant “pool”. We will do so in accordance with para 4.18 of the EHRC code of practice, referred to above.

[159]The statistics we have come from the respondents. They are broken down into the following categorise as against the first and second respondents: 1.2 Ages of good leavers from (and including) 1 January 2019 and prior to 18 September 2019 3 Ages of good leavers in the period between 18 September 2019 and 12 May 2020 4 Ages of bad leavers in the period between 18 September 2019 and 12 May 2020 5 Ages of those in employment at May 2020

[160]Of these, categories 1.1 and 2 were never affected by the PCP one way or another. They had no (or no further) entitlement to awards under the 2017 LTIP and their position is not affected one way or the other by the PCP. Page 39 of 52

[161]Category 1.2 represents the disadvantaged category. The claimant falls within this category. These employees would have benefitted from the change to the performance requirements but for the additional requirement to have been in in employment at 18 September 2019.

[162]Categories 3, 4 and 5 represent the advantaged categories. As at the time the PCP was imposed they were (prospectively) entitled to benefit from the change to the performance requirements. Those in category 4 did not in fact benefit, but the advantage they gained was the entitlement to LTIPs on the amended performance condition, and the fact that intervening events meant that they did receive does not mean that they did not have this advantage in the first place.

[163]The next question is whether the correct “pool” encompasses only employees of the second respondent who were affected by the change, or whether the pool should include all those (no matter what group company they were employed by) who were affected by the change in the first respondent’s scheme.

[164]Our view is that the correct pool is limited to the employees of the second respondent who were affected by the change. This is because the liability of the first respondent (if there is any) is secondary to and depends on the liability of the second respondent. It is said that the first respondent is liable either as agent of or for knowingly assisting the second respondent’s discrimination. In both cases it depends on the second respondent having discriminated against the claimant, so in looking at the “pool” our focus must be on the employees of the second respondent. If there is no indirect discrimination by the second respondent we do not see how the first respondent can be liable either as agent or for knowingly assisting such (non-existent) discrimination.

[165]However, as follows, the “pool” of employees affected by the first respondent’s decision is not irrelevant. It provides a useful check as to whether (given the limited numbers of the second respondent’s employees affected by the decision) there is any disadvantage for those aged 57 and over.

[166]Looking first at the formal pool – the second respondent’s employees:a. There are 55 in the total pool of advantaged and disadvantaged employees, of whom 52 are under 57 and 3 are 57 and over.

b. In total, 53 people were advantaged and 2 were disadvantaged. Of the 53 people who were advantaged, 51 were under 57 and 2 were 57 and over. Of the 2 who were disadvantaged, 1 was under 57 and 1 was 57 and over (that was the claimant himself).

[167]Thus 51 out of 52 people aged under 57 were advantaged and 1 out of 52 people aged under 57 were disadvantaged. That amounts to 98% advantaged and 2% disadvantaged. Page 40 of 52 168. 2 out of 3 people aged 57 and over were advantaged and 1 out of 3 people aged 57 and over were disadvantaged: 67% advantaged and 33% disadvantaged.

[169]That would appear on the face of it to show a clear disadvantage for those who are aged 57 and over, but we must be cautious about this particularly as the claimant himself was the only person aged over 57 who was disadvantaged, and the overall number of those aged over 57 is relatively small. That is not necessarily a problem (see, e.g. Edwards v London Underground Limited (No. 2)

[1998] IRLR 364

) but warrants further consideration in case the apparent disadvantage is “ephemeral or some sort of fluke”.

[170]A useful check can be provided by looking at the broader population across the entire group affected by the first respondent’s decision. This is not the formation of an alternative pool – it is a check to see whether the differential identified in the (small) pool we are dealing with is a reliable indicator that there is substantial disadvantage for those aged 57 and over. Across the group as a whole:a. There are 294 in the total pool of advantaged and disadvantaged employees, of whom 275 are under 57 and 19 are 57 and over.

b. In total, 270 people were advantaged and 24 were disadvantaged. Of the 270 people who were advantaged, 255 were under 57 and 15 were 57 and over. Of the 24 who were disadvantaged, 20 were under 57 and 4 were 57 and over (again this final group included the claimant).

[171]Thus 255 out of 275 people aged under 57 were advantaged and 20 out of 275 people aged 57 and over were disadvantaged: 94% and 6%. 172. 20 out of 24 aged 57 and over were advantaged and 4 out of 24 people aged 57 and over were disadvantaged: 83% and 17%.

[173]There is a broad theme across both the pool (the second respondent’s employees) and the wider group that those aged 57 and over were worse off (or were less likely to benefit) than those aged under 57.

[174]In table form: Second respondent only Group employees <57 57+ <57 57+ Count: % Count % Count % Count % Advantaged 51 98% 2 67% 255 94% 15 83% Page 41 of 52 Disadvantaged 1 2% 1 33% 20 6% 4 17%

[175]Converting this to ratios: - for the second respondent’s employees only, someone aged 57 or over is 0.68 times as likely to be advantaged as someone under 57, but 17 times as likely to be disadvantaged as someone under 57, and - across the whole group, someone aged 57 or over is 0.88 times likely to be advantaged as someone under 57, and roughly 3 times as likely to be disadvantaged.

[176]As for whether it is the “advantaged” or “disadvantaged” groups that count, the respondents urged us to take the “advantage led approach” referred to by Lord Walker in Rutherford v Secretary of State for Trade and Industry (No. 2)

[2006] ICR 785

– that is, assessing whether there has been a particular disadvantage by reference to the ratio of advantaged rather than disadvantaged employees. The relevant passage of the judgment is para 67: “I do not express the view that some element of disadvantage-led analysis may not be appropriate in some cases. But it must be recognised that there is a difficulty here: the more extreme the majority of the advantaged in both pools, the more difficult it is, with any intellectual consistency, to pay much attention to the result of a disadvantage-led approach. However I can imagine some (perhaps improbable) cases in which a disadvantage-led approach would serve as an alert to the likelihood of objectionable discrimination. If (in a pool of one thousand persons) the advantaged 95% were split equally between men and women, but the disadvantaged 5% were all women, the very strong disparity of disadvantage would, I think, make it a special case, and the fact that the percentages of the advantaged were not greatly different (100% men and 90.5% women) would not be decisive.”

[177]Lord Walker’s statement appears to suggest that disadvantage can be assessed by reference to either an advantage-led or disadvantage-led approach, or perhaps a combination of the two (to avoid the double-negative: “some element of disadvantage-led analysis may … be appropriate in some cases”).

[178]Looking at the position overall, rather than adopting either an exclusively advantage-led or disadvantage-led approach, is the best way of addressing the facts of this case.

[179]For our pool, the disadvantage-led approach produces the remarkable result that people aged 57 and over are 17 times more likely to be disadvantaged than Page 42 of 52 those aged under 57 – but this is entirely dependent on the claimant’s personal position, so may not be an accurate account of the particular disadvantage. We can see that the disadvantage persists, although to a lesser degree, when we look at those disadvantaged across the whole group. An advantage-led approach still shows that those aged 57 and over are less likely to be advantaged than those aged under 57.

[180]We also bear in mind that all that is required under the Equality Act is a “disadvantage” – we are not assessing the “considerably smaller” requirement that applied under the former discrimination statutes and for cases such as Rutherford.

[181]Our conclusion is that overall, these statistics in relation to the pool of the second respondent’s employees who were affected by the PCP do show that those aged 57 or over are at a particular disadvantage on the application of the PCP, compared to those aged under 57.

[182]There is no dispute that the claimant himself suffered that disadvantage. Justification Incentivisation as a legitimate aim

[183]In their submissions, the respondents typically speak of “retention and incentivisation” in one breath, as if they are the same thing. However, we consider they are separate concepts that require separate consideration.

[184]We will start our consideration of justification by looking at incentivisation as a legitimate aim. We understand “incentivisation” in this context to mean motivating employees to work hard and perform well.

[185]The LTIP is, of course, a “Long Term Incentive Plan”, so incentivisation must be an aspect of the LTIP. However, it is the PCP that must be justified, not the

[186]In the internal documents speaking of changes to the LTIP, “incentivisation” or “motivation” are occasionally, but seldom, mentioned. The “incentivisation” element of the LTIP can only be that individual hard work feeds through into higher earnings per share and therefore higher awards under the LTIP.

[187]The difficulty in this case is that the only reason why these changes were contemplated at all was that it was projected that the existing targets under the LTIP would not be met. There is no suggestion in the respondents’ papers that a last-minute group effort would enable the targets to be met. While the amended performance criteria granted by the PCP included the prospect of a greater than 50% award for achieving 7.5% or more on EPS, there is no suggestion in the paperwork that this was ever capable of being achieved or would motivate staff for one final effort before the end of the LTIP. We mention Page 43 of 52 “one final effort” since the PCP was put in place three months before the end of a three year performance period, and the target EPS was measured on compounded basis over the period of the grant. Short of some windfall like the favourable US tax settlement the previous year, there was nothing that anyone could have realistically done or that could have happened to make any difference to whether the original or revised targets were achieved.

[188]The essence of the PCP was that employees would receive awards they did not otherwise qualify for on condition only that they remained in employment. As we shall see, that may well assist retention, but it does nothing for incentivisation. There was no realistic prospect of receiving anything higher than the 50% award. The PCP simply gave everyone who remained employed a flat-rate award. There was nothing there to incentivise them to work harder.

[189]The respondents say that the cut off date of 18 September 2019 was required as former employees cannot be incentivised. That may be so, but in this case for the reasons given above we find that the changes to the LTIP were not aimed at incentivisation, so we do not see that the cut off date can assist with incentivisation either. If incentivisation is not the aim of the scheme changes, it cannot be (and was not) the aim of the cut off date for those changes.

[190]We therefore reject “incentivisation” as a legitimate aim. It was not the aim pursued by the respondents in the application of the PCP. Retention as a legitimate aim

[191]The position is different as regards retention. The concept of retention is a thread that flows through all of the respondents’ discussions and internal materials concerning amendments to the LTIP. Reviewing our account of the facts set out above, the discussions the respondents had about changes to the LTIP start (in summer 2018) with an invitation from the Remco to management to “make a proposal … to address potential near-term retention issues”, and finish with the minute recording the amendment (on 18 September 2019) stating that “the current retention issues faced by the organisation meant waiver or change of the performance conditions … was appropriate”.

[192]Of course, it is for us to assess the legitimate aim. The respondents mentioning it many times does not make it the legitimate aim that was being pursued, but it is striking quite how frequently this is mentioned in internal documents that the respondents would not have expected to have to put before a court or tribunal.

[193]The claimant criticised this on the basis that the respondents seem to have made no effort to quantify or examine whether they really had a retention problem. The claimant suggests that HR data should be readily available showing whether there was or was not a trend towards more senior executives voluntarily leaving. This was never investigated by the Remco or any other managers to see whether any retention problem was real or simply imaginary. Page 44 of 52

[194]There is more that could have been done by the respondents in terms of quantitative analysis of the retention problem and, later, the effect of the PCP on retention – but it is not necessary for them to prove their legitimate aim using statistics. We note that there were several factors that had come together that would give rise to significant concerns about retention. In the first place, the LTIP scheme which had previously given full awards was no longer going to do so. Beyond that, there were the twin uncertainties of what was to occur with RB 2.0 (whether the business would be fully split into two) and the introduction of a new chief executive. These uncertainties would be particularly felt at the most senior end of the business – the T40 and T400 – where individuals would be bound to be concerned about their roles within any redesigned business, and/or what view the new chief executive would take of their part of the business and their role within it. We also note, of course, that the primary purpose of any retention measures would be to prevent a retention problem arising, not to try to recover the situation after the problem has arisen. Those who have left cannot be retained by retention measures implemented after they have left. It is proper to put in place measures ahead of and in anticipation of a retention problem – not to wait until a retention problem is demonstrably occurring before taking measures.

[195]In those circumstances, we do not consider that any quantitative analysis by the respondents of the problem is necessary for us to conclude that this was the legitimate aim pursued by the PCP. Retention was the essence of the PCP. Putting it bluntly, individuals were getting valuable awards they would not otherwise be entitled to, and in return they were doing nothing more than remaining with the business for a particular period of time. The obligation to remain with the business for a particular period of time was reinforced by the extension of vesting to May 2020. That the amended performance condition only applied to those who were employed at 18 September 2019 was entirely consistent with that. Those who had already left the business could not be retained. As we shall see when discussing proportionality, there were a small number of employees who benefited despite them already being under notice – their position (and its effect on the legitimate aim) is dealt with below primarily as a matter of proportionality.

[196]Everything in the materials points to retention being the aim pursued by the respondents in the PCP. We also accept that given the circumstances the respondents had a real (as opposed to imaginary) retention problem that was either actually occurring at the time or looming in the near future (or a combination of the two).

[197]We conclude that retention was a legitimate aim and was the aim being pursued by the respondents in the application of the PCP. Proportionality

[198]For a measure to be proportionate: Page 45 of 52 “the measure in question must be both appropriate to achieve its legitimate aim or aims and necessary in order to do so ... The gravity of the effect upon the employees discriminated against has to be weighed against the importance of the legitimate aims in assessing the necessity of the particular measure chosen.”

[199]The PCP is : “requiring LTIP participants (who were not already bad leavers) to be employed as of 18 September 2019 (or, in the alternative, May 2020) in order for them to benefit from the amended performance condition”

[200]The respondents say that the answer to this is obvious. You cannot retain staff who have already left. If your goal is retention of staff than there must be a cutoff date and choosing the date of the decision was the rational way of doing that. Any amendment that did not have such a cut-off date would result in the advantaging of many staff (including the claimant) who had already left and could not be retained.

[201]We note that the minute of the relevant Remco decision itself spells out the justification for limiting the amended performance condition to those who were employed at the date of the decision: “exclusion of former employees (being good leavers) was appropriate, given the purpose of the waiver/change is to support employee retention”.

[202]If, as we accept it was, the aim of the PCP was retention, it must be the case that it only applies to those who were employees as at the date of the change. They were the only employees who could be retained. The claimant’s argument that the amended performance criteria should apply to those such as him who had already left but had retained rights under the LTIP scheme does not do anything for retention of staff.

[203]It is for the respondents to justify the PCP, not for the claimant to demonstrate that there is no justification, but we note that the claimant has two broad criticisms of the respondents’ attempts to justify the PCP.

[204]First, he says that if retention was the aim there were other, more effective, appropriate and less discriminatory, ways of achieving that. These are individually focussed measures, such as individual mentoring and coaching for career development, and individually targeted financial incentives, such as specific bonuses or LTIP awards on an individual basis. For this to have been done would have required the respondent to carry out some sort of assessment of who they were really in danger of losing – that is, who was a “flight risk” - and put in place individual measures to retain those individuals.

[205]This is a difficult contention. In the first place, it has never been the respondents’ position that they only wanted to retain some individuals and were not bothered Page 46 of 52 about others leaving. Perhaps amongst the T40 and T400 there were some who the respondents would be more or less concerned about leaving, but that has never been the way the respondents have put things. In their closing submissions, it is put this way (emphasis from the original): “[the respondents] wished to keep all of the senior management in place”. That is consistent with the evidence we heard as to the respondents’ intentions.

[206]The respondents go on to point out that if this were to be done, there is nothing to suggest that it would have been less discriminatory and, above all, that the claimant would still not have benefitted from it. None of the individual measures he suggested would ever have been applied to former employees.

[207]We also consider that it is very difficult to see how this could practically have been done. The assessment of who was or was not likely to leave, and what measures could be put in place to stop them leaving, could only be based on the subjective opinion of their manager. It seems to us to be completely impractical for the first respondent to have worked through the entire T400 population and put in place bespoke measures for each individual a manager had assessed as a “flight risk”. If this could have been done, it would have brought with it considerable risk of favouritism and a greater discriminatory effect than the PCP adopted by the respondents.

[208]Second, he points to the fact that the way in which the amendments to the LTIP were implemented meant that some individuals benefitted despite them not actually being retained by the business.

[209]The most striking instance of this is in the six individuals who had already signed leavers agreements (but remained employed) as at 18 September 2019. They were already serving notice and were always bound to leave employment, yet, by virtue of being still employed as at 18 September 2019 and leaving as good leavers, they would benefit from the amended performance condition. In most of these cases the individual was on garden leave and not actively in work as at 18 September 2019, yet they would still benefit from the amended performance condition.

[210]Those six are a difficult category for the respondents. While the Remco were keen to limit the amendment to those who were employed as of 18 September 2019 they do not seem to have contemplated that there would be a category of employees who would be serving notice at that date, could not be retained, but would still benefit from the amended performance condition.

[211]We note, however, that the logical conclusion of the claimant’s arguments on this point is not that he should have benefitted from the amended performance condition, but that these individuals should not have benefitted from the amended performance condition. No measure targeted at retention of employees could ever benefit the claimant, who had already left the business. Page 47 of 52

[212]The seven who signed leaver agreements after 18 September 2019 are in a different position. The PCP gave the opportunity to retain them, but in fact they were not retained. As the respondents say, no retention measure can be 100% correct. The real issue is with the six who benefitted from the amended performance criteria despite there never being any prospect of them being retained.

[213]The respondents say that we should not be concerned with those at the margins (6 out of around 400 original participates or around 270 ongoing participants). Instead, we should focus on “the substantial majority of the people on either side of the line”.

[214]The claimant draws attention to the comments of Eady J in Pitcher v University of Oxford EA-2019-000638-RN at para 103: “The appropriateness of a measure may also be undermined by the inclusion of exceptions that are inconsistent with the aim in question”.

[215]Both of those propositions are correct. Odd exceptions and anomalies do not mean that the PCP is not capable of being a proportionate means of achieving a legitimate aim, but too many exceptions or anomalies cast doubt on whether the PCP is really in pursuance of the claimed legitimate aim, or whether, if it is in pursuance of the claimed legitimate aim, it is proportionate.

[216]What we have here is a category of staff who have simply been overlooked by the Remco. They had not contemplated that there would be individuals who would be serving out notice and who retained rights under the LTIP. Perhaps they should have done so, but in these circumstances that omission does not substantially undermine either the legitimate aim of retention or the justification of the PCP. As previously, we note that the claimant’s position is that the amended performance condition should have applied irrespective of whether or not the person with entitlements under the LTIP remained employed or not. That would only create a further anomalous category of individuals and further undermine the stated aim of retention. These six are “marginal” cases and do not affect the overall aim or the proportionality of the PCP, any more than, say, the exclusion of the two executive directors undermines the overall aim or proportionality of the PCP.

[217]Having found that the respondents were pursuing a legitimate aim of retention we are satisfied that the PCP was a proportionate means of achieving that aim. The amended performance condition had to be subject to some sort of qualifying or cut-off date to be effective as a retention tool and to avoid unnecessary payments to those who were no longer in a position to be retained. The date the decision was made was the most appropriate qualifying or cut-off date that could be applied, and the aim and proportionality of the PCP is not substantially undermined by the fact that there was a small anomalous category of employees who benefitted from the amended performance criteria despite not being in a position to be retained. Page 48 of 52 Time

[218]The respondents describe the claimant’s claim as being “manifestly out of time”. They say that the claimant first became aware that others had been subject to the amended performance condition in April 2020. He had consulted his lawyers by 12 May 2020 at the latest, but it took until 3 July 2020 for the first allegation of age discrimination to be made – that was the date his original ET1 was issued. The claim only reached its current form (against the first and second respondents) on 7 October 2020. We broadly accept that chronology.

[219]The respondents say that if there was discrimination it was a one-off act or omission occurring on 18 September 2019, so that the claimant’s claim is around ten months out of time, and time should not be extended.

[220]The claimant says that the respondents concealed the decision of 18 September 2019 from him. He did not learn of it until he took his own steps to investigate the position in April 2020. It was only much later that he learned that there were good leavers after 18 September 2019 who had been subject to the amended performance condition. The PCP was only applied to the claimant on 6 May 2020 – the point at which others benefitted from the amended performance condition. The PCP was a continuing rule from September 2019 through to May 2020. The respondents’ decision in April 2020 that no exception would be made for the claimant amounted to an act of discrimination in itself or part of a continuing act. All of this meant that the original claim had been submitted within time. The later amendment was “fundamentally the same as the initial claim” and the cause of action remained the same. It was therefore 3 July 2020 that was the relevant date for limitation purposes, not 7 October 2020. If the claim was out of time, time should be extended on a just and equitable basis.

[221]Questions of time limits arising on amendment may often be expected to be addressed in the decision which permitted the amendment. This was not done at the hearing which decided on the amendment in this case. We note that the employment judge on that occasion found that the amendment “introduced new heads of claim and two new causes of action against [the first respondent]”, found that “the claims appear to be significantly out of time” but “[did] not agree that the claimant has little hope of persuading a tribunal to exercise discretion to extend time in all the circumstances of the case …”.

[222]Our findings on the question of whether the claim is brought within the primary time limit are as follows:a. The claimant’s claim is about a PCP that was applied to him in 18 September 2019 at the point at which the Remco made amendments to the LTIP. The complaint is about the PCP, not about the outcome of the PCP, which would only have manifested itself later (either Dec 2019 or May 2020). Page 49 of 52

b. The PCP was an act, rather than an omission. The PCP alleged is described in positive terms as a requirement for participants to be employed as of a particular date, not as an omission or failure to address the position of those who had already left as good leavers.

c. The act occurred on 18 September 2019 by the decision of the Remco to make changes to the LTIP. This was not a continuing act, it was a oneoff decision to apply a PCP.

d. The claimant cannot restart time by making a request for the PCP to be disapplied to him. That follows from the claimant’s definition of the PCP. As pointed out by the respondents, there is no alternative allegation of, for instance, a PCP of refusing requests for individual exemption. If there were such an alleged PCP it would require separate consideration of whether it was an act of indirect age discrimination and/or justified. The PCP alleged is the decision of 18 September 2019.

e. Since the PCP was applied (and time started to run) on 18 September 2019 the claimant’s claim, whether dated to July or October 2020 is substantially out of time, and the only question that remains is whether it is just and equitable to extend time.

[223]The “just and equitable” formulation gives the tribunal “the widest possible discretion” in extending time (Abertawe Bro Morgannwg University LHB v Morgan

[2018] EWCA Civ 640

). However (Morgan): “factors which are almost always relevant to consider when exercising any discretion whether to extend time are:(a) the length of, and reasons for, the delay and

(b) whether the delay has prejudiced the respondent (for example, by preventing or inhibiting it from investigating the claim while matters were fresh)”

[224]On the question of just and equitable extension of time, we find the following:a. The claimant was, through no fault of his own, unaware of the PCP until early April 2020.

b. The claimant took appropriate action to follow up on his April 2020 discovery, with correspondence following in May 2020 from his solicitors. From that point onwards the respondents were aware of the possibility of a claim, although not its eventual substance or what parties it was to be brought against.

c. The respondents appear to have responded properly to the claimant’s correspondence. This is not a case in which further delay can be substantially accounted for by delay on the part of the respondents. Page 50 of 52

d. There is no rule of law that time limits in a discrimination claim only run from when the claimant becomes aware of the discrimination, but where a claimant is (through no fault of their own) unaware of the discrimination (or potential discrimination) that would appear to be a powerful factor in considering a just and equitable extension of time.

e. The claim as originally brought was a claim against the first respondent only, alleging that it has applied a PCP to him. The October amendment added the second respondent, slightly amended the PCP and for the first time made allegations that the first respondent was liable as an accessory to discrimination by the second respondent. As acknowledged by the employment judge at the time, the October amendment was a substantial amendment.

f. The respondents have suffered no significant prejudice through the claim being brought out of time. In their closing submissions, they say that “the cogency of the evidence has obviously been impacted”. In saying this, they refer to difficulties the claimant had in recollecting a meeting. However, this is not a case in which the respondent says that relevant documents have been destroyed or witnesses are no longer available to them. Inevitably with delay the cogency of the evidence will have been impacted to some degree, but that is inherent in employment tribunal or any other litigation, and we do not consider that in this case the cogency of the evidence has been so seriously affected as to significantly prejudice the respondents. They also say that they will be prejudiced by the loss of a statutory limitation defence – but that will always be the case when the tribunal is considering an extension of time.

g. The claimant will suffer significant prejudice if time is not extended – he will have lost the right to bring the entirety of his claim.

[225]Having taken all of this into account we conclude that it is just and equitable to extend time in order to encompass the claimant’s claim.

[226]The delay from September 2019 to April 2020 is accounted for by the claimant being unaware of the PCP. From April 2020 he has properly progressed his claim, and the respondents were aware of the possibility of a claim from May 2020.

[227]There is some force in the respondents’ arguments that the claimant has not accounted for the delay between the original form of his claim in July 2020 and the final form of his claim in October 2020, but the primary question that that must relate to is whether the delay in settling the final form of the claim caused any substantial prejudice to the respondents. In our judgment, it did not. The amendment was substantial, but largely related to the technicalities of the relationship between the first and second respondents, and how each of them may have been responsible for the decision of the Remco. The underlying evidence that would be required to explain and assess the Remco’s decision Page 51 of 52 did not change. Aside from legal argument on the relationship between the first and second respondent, the evidence required to address the original and amended claim were largely the same.

[228]The claimant’s claim was not brought within the primary time limit, but it is just and equitable to extend time in order that the tribunal can consider the claim, and that is what we do in this case. E. CONCLUSION

[229]We can well understand why the claimant, after a long and distinguished career within the first respondent’s group, has sought to bring this claim in respect of the 2017 LTIPs. We have found that there was no age discrimination in the decision that ultimately denied him an entitlement that others achieved under that scheme. There may yet be more to be argued on the question of whether there has been any breach of contract (as to which we express no view) but the claimant’s claim of indirect age discrimination is dismissed. The respondents have justified the indirect age discrimination that arose from the PCP as the PCP was a proportionate means of meeting their legitimate aim of retaining senior employees.

[230]Given our decision, the provisional remedy hearing listed for 24 & 25 March 2022 is not required and will be removed from the tribunal’s lists.