Verifone (UK) Ltd v K Zena

The appeal in this matter concerns the approach an Employment Tribunal (“ET”) should adopt when considering a claim of victimisation arising from an employer’s response to allegations of discrimination; the cross-appeal relates to the determination of claims of direct discrimination when an employee is singled out for redundancy and dismissal.

In giving this judgment, I refer to the parties as the claimant and respondent, as below. The decision under challenge is that of the Watford ET (Employment Judge Quill, sitting with lay members Mr Wimbor and Ms Harris; “the ET”) from 5-9 December 2022, which was sent out on 9 March 2023. By that judgment, the ET unanimously upheld the claimant’s claim of unfair dismissal but dismissed her complaints of unauthorised deductions from wages, of direct race discrimination, and two complaints of victimisation; by a majority decision, however, the ET lay members upheld two further complaints of victimisation. By its appeal, the respondent seeks to challenge the majority upholding of the two victimisation claims; the claimant cross-appeals in respect of the dismissal of her claims of race discrimination.

The respondent is a UK based company, which is part of an international group that provides end-to-end payment facilities for various clients. The claimant is a chartered management accountant (having qualified in 1999), who started working for the respondent on 29 July 2015, initially with the title of European Regional Controller. The claimant is a black woman and, as explained within her particulars of claim, was:

Throughout her employment the claimant was well thought of; she had good appraisals and, in 2017, received an award in recognition of her contribution to the respondent’s success. Although she was based in Uxbridge, from time to time the claimant’s work required her to travel to various European offices and her contract provided that she might be asked to transfer to another place of work if business needs demanded. When she started, the claimant’s remit was to create a European accounting and controllership function, and to lead the transfer of accounting to the Verifone Shared Service Centre (“SSC”) in Warsaw, Poland. In 2018, the respondent’s parent company was privately acquired, and the resultant changes gave rise to a number of redundancies, which saw the claimant’s role expand to include responsibility for the Middle East and Africa, her title changing to that of EMEA financial controller.

In or around July 2019, Mr Thomas Rebain became the claimant’s line manager. He was aware that the group was in the process of transitioning its day-to-day transactional processing and accounting and reporting functions into its two SSCs (as well as the SSC in Warsaw, there was also one in Shanghai). As part of this process, the Latin America Controller and the Asia-Pacific Controller roles (regional equivalents of the claimant’s position) had been made redundant shortly before Mr Rebain started. As he took over his responsibilities, Mr Rebain had discussions with the claimant’s former line manager (Ms Colvin), about the possibility of a restructure in the EMEA controllership team but, as the ET found, he had not wanted to take this forward at that stage. The claimant became aware of these discussions and in March and August 2020, having heard rumours about her own redundancy, questioned Mr Rebain about this but was assured there were no plans to make her redundant; that, the ET found, represented Mr Rebain’s genuinely held view at that stage. More specifically, the ET rejected Mr Rebain’s evidence regarding a meeting on 14 August 2020, when he said it was decided “to evaluate each finance professional role that was not already part of the Warsaw SSC”. The ET found that, in fact, the position only changed at the end of August/beginning of September 2020, because of Mr Rebain’s decisions relating to another employee, Mr Dawid Makowski.

On 31 August 2020, Ms Yong Chen (another Controller, at the same level as the claimant, who also reported to Mr Rebain) notified Mr Rebain that Mr Makowski, who was based in Warsaw, had handed in his resignation and was planning to leave after three months to take on another, higher paid, role with more responsibility. Mr Makowski reported to Ms Chen, and was thus at a lower level than the claimant. Asked by Mr Rebain whether Mr Makowski might be willing to stay, Ms Chen responded (on 1 September 2020):

As revealed in a document disclosed on the last working day before the ET hearing, at some point between 2 to 4 September 2020, Mr Rebain had a remote meeting with Mr Makowski, who referred to his understanding (on joining the business) that various financial functions were to be transferred to the Warsaw SSC; it was partly because this had not happened that he was resigning to take up another offer. As a result of his discussions with Ms Chen and Mr Makowski, Mr Rebain determined he would like to retain Mr Makowski, albeit there was no existing vacancy in Warsaw that was likely to persuade him to stay and the ET accepted Mr Rebain’s evidence that he made no promise of any specific post (including the claimant’s) at this stage.

The ET found, however, that it was his aim in order to retain Mr Makowski that, in early September 2020, Mr Rebain formulated a plan to transfer the claimant’s duties from Uxbridge to Warsaw, with the intention of allocating those duties to Mr Makowski, thus providing him with an increase in responsibilities (together with an increase in pay) so he would be persuaded to retract his resignation and agree to stay (ET, paragraph 239). The ET further concluded:

Mr Rebain’s plan was set out in documentation that he created, dated 27 September 2020, which showed a new structure in which the claimant’s name and job title had been replaced by a post entitled “Regional Controller and Warsaw SSC Accounting Manager”, with Mr Makowski’s name shown as the new postholder. This documentation was not disclosed to the claimant at any time prior to the ET proceedings.

Although the ET accepted that the respondent had had a longer term plan to transfer the activities undertaken by the claimant’s team to the Warsaw SSC, that plan had been put on hold and the ET was satisfied it would not have been implemented in September 2020 were it not for the threat of Mr Makowski’s resignation: it was the desire to retain Mr Makowski that led to the decision to make the claimant redundant (ET, paragraphs 241-242). In reaching that conclusion, the ET rejected the respondent’s case in this regard:

The process adopted by the respondent to implement Mr Rebain’s plan involved an HR Business Partner, Ms Blunnie, and an HR manager, Ms Wiersma (Ms Wiersma had, at least, been copied into some of the documentation; she does not appear to have attended any of the relevant meetings at this time). This process included information and consultation meetings with the claimant on 6 October 2020 (the initial notification meeting) and on 20, 23 and 28 October 2020; the ET was, however, clear:

In particular, as the ET found, Mr Rebain and Ms Blunnie had conferred about how to avoid anything the claimant might raise becoming a problem that might get in the way of the intended outcome, thus: Ms Blunnie advised avoiding telling the claimant that she could take the job in Warsaw; the documentation of 27 September 2020 was not shown to her; and Mr Rebain provided an evasive answer to her question whether anyone had been recruited to the role in Warsaw (not referring to the plan to appoint Mr Makowski).

As was common ground, at the meeting on 28 October 2020 the decision to terminate the claimant’s employment was communicated to her and the terms of her termination letter were discussed, with this letter being sent out the same day, confirming her employment would end on 13 November 2020. In the event, from 30 October 2020, the claimant was signed off work ill with stress and anxiety, her fit note covering the remaining weeks of her employment.

On 1 November 2020, the claimant submitted an appeal against her dismissal, in which she referred to the decision to make her redundant as being “personal” and “due to the lack of appreciation of the value of diversity”. Although not expressly referring to race, the claimant spoke of diversity “of thought, approach, culture and personalities” and suggested that her dismissal was due to “unconscious bias”. It was accepted that her letter of appeal constituted a protected act.

The claimant’s appeal was forwarded to Ms Wiersma. Relevant internal documents (including minutes from the consultation meetings, which had never been sent to the claimant) were sent to Ms Wiersma on 3 November 2020 but, other than an acknowledgment on 2 November 2020, no contact was made with the claimant regarding her appeal until Ms Wiersma emailed her on 19 November 2020 (after the claimant’s employment had already ended). Even then, the claimant was not sent copies of the consultation minutes until 27 November 2020, and other questions raised in her communications with Ms Wiersma were never answered.

On 30 December 2020, the claimant emailed Ms Wiersma with a formal grievance and a further document entitled “discrimination case”. The claimant’s grievance raised her concern that she had been asking to see the contractual redundancy payments made over the last five years but these had not been provided; she further asserted that her redundancy was pre-determined and personal (the restructure only involved her); and she noted there was a “startling difference” between how her case was handled and how previous redundancies were dealt with. In her “discrimination case”, the claimant gave further detail of what she contended was less favourable treatment as compared to previous redundancies and (having by then learned that Mr Makowski had taken up the role in Warsaw) she noted her replacement was to be a “young white man”.

An appeal meeting took place on 12 January 2021. Thereafter, Ms Wiersma had meetings with Ms Blunnie and Mr Rebain (and, during the course of a meeting with Ms Blunnie, was shown the documentation of 27 September 2020), before emailing the claimant on 27 January 2021, asking her to clarify certain points (which the claimant did, by email the same day).

On 29 January 2021, Ms Wiersma sent an outcome letter to the claimant, rejecting the appeal. She did not accept the redundancy was pre-determined and said the claimant had not identified alternatives during the consultation meetings and had not said she would be willing to move to Poland. It was, however, noted that a candidate had “now” been appointed. As for the discrimination complaint, Ms Wiersma responded:

As the ET found, Ms Wiersma had been junior to both of those involved in the original decision (Mr Rebain and Ms Blunnie), had herself had some involvement in that process, and had failed to liaise with the claimant until after her employment had ended; the ET was clear:

In considering the claim of unfair dismissal, the ET unanimously rejected the respondent’s case that the claimant had been dismissed by reason of redundancy. Considering the respondent’s alternative argument – that there had been some other substantial reason (“SOSR”) – the ET analysed the position, as follows:

The ET was further clear that the claimant’s dismissal was unfair and that the appeal did not cure that unfairness; as it explained:

Going on to consider the complaint of direct race discrimination, the ET was concerned with four alleged acts of less favourable treatment, of which two are relevant for my purposes: (1) that the claimant had been singled out for redundancy, and (2) that she had been dismissed. The ET was unanimous in finding that the claimant’s race was not the conscious reason for her dismissal, but then asked itself whether it might still have played any part (consciously or unconsciously) in the process. In respect of both allegations, the (actual) comparators relied on by the claimant were: “her team members and all others employed within the Respondent at her grade”, and Mr Makowski; she also relied on hypothetical comparators.

Accepting that only the claimant’s post had been affected, and that there was a difference in race between the claimant and Mr Makowski, and a difference in treatment, the ET did not find Mr Makowski could be a direct comparator for the purposes of the statute (sometimes referred to as a statutory comparator):

Constructing the relevant hypothetical comparator, the ET considered this was:

… Would it have dismissed that person (purportedly by reason of redundancy) by letter dated 28 October 2020, with termination date 13 November.”

Having regard to “all the facts” it had found, and accepting it had rejected the respondent’s case in a number of respects (ET, paragraph 265), the ET did not, however, consider the claimant had discharged the primary burden of proof placed upon her under section 136 Equality Act 2010 (“EqA”):

Addressing the claimant’s claim of victimisation, the ET noted that four detriments had been alleged. The first two were held not to be made out on the facts, and there is no challenge to that finding. The claimant had also contended, however, she had been victimised because the respondent had: (1) failed to adequately investigate her allegations of discrimination or provide a process of appeal, and (2) made superficial findings dismissing those allegations. In respect of these allegations, the ET unanimously agreed:

The ET lay members noted, however, that there had been a clear right of appeal and that the respondent “purports to have an anti-harassment policy”, which provided for allegations to be reviewed by the legal department and “for an impartial and timely investigation and follow-up”, finding that:

The lay member majority concluded:

Asking whether the claimant had met the initial burden under section 136 EqA, the majority reasoned:

Considering whether the respondent had discharged the burden then placed upon it pursuant to section 136, the ET majority was clear it had failed to do so:

In further explaining its reasoning, the ET majority referred to the fact that, in the outcome letter, Ms Wiersma had said that she had not found evidence of predetermination but had failed to mention that Ms Blunnie had shown her the documentation of 27 September 2020, which showed Mr Makowski in post in the new structure; the majority observed:

Noting that there had been no mention in the appeal outcome letter to the timing and contents of Mr Rebain’s discussions with Mr Makowski, the majority concluded:

The ET majority also referred to Ms Wiersma’s failure to engage with the points made by the claimant about how she had information that others had been given enhanced redundancy when she had not; holding:

In disagreeing with the conclusion thus reached by the majority, the Employment Judge made clear that there was no dissent from the observations made as to the inadequacy of the investigation, but was:

In the circumstances, the Employment Judge concluded that the burden of proof had not shifted to the respondent and these two further claims of victimisation fell to be dismissed.

The appeal and cross-appeal concern the claimant’s claims brought under the Equality Act 2010 (“EqA”), of direct discrimination, as defined by section 13(1), and of victimisation, as defined by section 27. In determining whether the respondent had contravened either provision, the ET was required to approach the burden of proof as provided by section 136 EqA, which (relevantly) states:

Section 136 can thus be seen to provide a two-stage test: at the first stage, the claimant has to prove facts from which the ET could infer that discrimination has taken place; it is only if such facts have been made out to the ET’s satisfaction (on the balance of probabilities) that the second stage is engaged, when the burden shifts to the respondent to prove (again, on the balance of probabilities) that the treatment in question was “in no sense whatsoever” because of the prohibited reason; see Igen Ltd v Wong [2005] EWCA Civ 142; [2005] ICR 931 CA. That said, while section 136 provides for a two-stage approach to the burden of proof, it is not necessarily an error of law for an ET to effectively assume the burden has shifted, and to look to the respondent to provide an explanation for the treatment in question; see Laing v Manchester City Council [2006] ICR 1519, EAT. In this respect, it is important to keep in mind the purpose that underpins section 136; as Elias P (as he then was) observed in Laing:

In Hewage v Grampian Health Board [2012] ICR 1054, SC, the need to avoid an overly technical approach to the application of section 136 was similarly emphasised; as Lord Hope observed:

Although it is thus important to ensure that section 136 is not used to elevate form over substance, where there is “room for doubt”, the approach it lays down provides a valuable tool for determining whether the inference of discrimination should be drawn; as HHJ James Tayler emphasised in Field v Steve Pye & Co [2022] IRLR 948 EAT:

Particular care should be taken if the reason for moving to the second stage is to avoid the effort of analysing evidence that could be relevant to whether the burden of proof should have shifted at the first stage. This could involve treating the two stages as if hermetically sealed from each other, whereas evidence is not generally like that. It also runs the risk that a claimant will feel that their claim that they have been subject to unlawful discrimination has not received the attention that it merits.

Where a claimant contends that there is evidence that should result in a shift in the burden of proof they should state concisely what that evidence is in closing submissions, …”

As for what is required to discharge the burden at the first stage, that must be something more than a difference in the relevant protected characteristic and a difference in treatment; see Madarassy v Nomura International plc [2007] ICR 867, CA, at paragraph 56. That said, the something more required at the first stage need not be a great deal; see Deman v EHRC [2010] EWCA Civ 1279 at paragraph 19.

A finding that an employer has behaved unreasonably, or treated an employee badly, will not, however, be sufficient, of itself, to cause the burden of proof to shift; that is because, as Lord Browne-Wilkinson explained in Glasgow City Council v Zafar [1998] ICR 120, at 124B:

That said, whether the putative discriminator would treat all employees in “the same unsatisfactory way” is not something that will be established by mere assertion; as Sedley LJ noted in Anya v Oxford University and anor [2001] EWCA Civ 405, at paragraph 14:

In determining whether an inference could be drawn from the facts established by the complainant, an ET must bear in mind that the relevant protected characteristic, or protected act, need not be the only reason for the decision in issue, it would be sufficient that it was a material influence; Nagarajan v London Regional Transport [2000] 1 AC 501 HL, at pp 512H – 513B. The crucial question is why the complainant received less favourable treatment (section 13 EqA) or was subjected to a detriment (section 27); see per Lord Nicholls in Nagarajan at p 511E-G:

If the reason for the conduct in issue was genuinely something other than the prohibited reason, then it matters not that it was unreasonable or unjustified; as the EAT observed in Bahl (cited with approval by the Court of Appeal in that case, at paragraph 101):

The way in which such alternative explanations (or the absence of such alternative explanations) are to be addressed in the application of section 136 EqA was considered by the Supreme Court in Efobi v Royal Mail Group Ltd [2021] UKSC 33, [2021] ICR 1263, where Lord Leggatt observed:

The inferences that an ET draws from its primary findings of fact are to be treated as questions of fact, with which an appellate tribunal should not interfere unless the conclusion reached was one that is properly to be characterised as perverse; Base Children’s Wear v Otshudi [2019] EWCA Civ 1648; [2019] IRLR 118, per Underhill LJ at paragraph 37. A challenge to an ET’s decision to draw, or not to draw, an inference of discrimination will thus need to meet a high threshold; as Lord Leggatt observed in Efobi:

In determining the reason why the impugned decision was taken, the ET is required to consider any parts of the Equality and Human Rights Commission Code of Practice for Employment (“the Code”) that appear relevant. For the claimant, emphasis is placed on the following paragraphs of the Code:

It has, however, been held to be an error of law to draw an automatic inference from the failure by a respondent to provide information or documents; as Underhill LJ emphasised in D’Silva v NATFHE [2008] IRLR 412, at paragraph 38:

In A v CC West Midlands Police UKEAT0313/14, it was observed that a particular difficulty can arise in a victimisation claim, where the detriment relied on is a failure to properly deal with the complaint that itself formed the protected act; as Langstaff P observed:

As for the approach I am to adopt to the reasoning of the ET, I note the guidance provided by Cavanagh J in Frame v Llangiwg Primary School UKEAT/0320/19, at paragraph 47; in particular, that:

Similarly, as Popplewell LJ reiterated in DPP Law v Greenberg [2021] EWCA Civ 672, the ET is not required to identify all the evidence relied on in reaching its conclusions of fact, or to express every step of its reasoning in any greater degree of detail than that necessary to be Meek compliant (Meek v City of Birmingham District Council [1987] IRLR 250 CA):

The respondent’s appeal is pursued on three grounds: (1) that the ET majority impermissibly concluded that its findings of fact shifted the burden of proof in relation to causation to the respondent; (2), that it impermissibly drew an automatic inference from the detrimental treatment itself and/or an absence of disclosure; (3) that it reached a perverse conclusion. In making its submissions in support of all three grounds, the respondent contends that the majority’s finding of detrimental treatment encompassed three acts or omissions on the part of Ms Wiersma: (i) a failure to adequately investigate the discrimination allegations; (ii) the making of superficial findings; and (iii) a failure to provide a process of appeal.

It is the respondent’s case that the majority’s conclusion that (absent any alternative explanation) the reason for the inadequate investigation/findings was the claimant’s discrimination complaint was unsustainable given the ET’s (unanimous) primary finding of fact that the reason Ms Wiersma did not adequately investigate the appeal and/or made superficial findings (not limited to the discrimination complaint) was because the outcome was pre-determined (ET, paragraph 253). Moreover, the facts relied on by the majority (ET, paragraphs 303-305) applied equally to all elements of the claimant’s appeal; none were specific to the discrimination complaint and there were no facts from which it could be concluded that the appeal would have been investigated properly or dealt with differently had it not included that complaint (see A v CC West Midland Police). In the alternative, the majority had apparently drawn an adverse inference from Ms Wiersma’s failure to disclose internal correspondence and/or from the unreasonableness of the investigation, without considering the alternative explanation provided by the ET’s (unanimous) finding that the appeal outcome was pre-determined. Further, to the extent it might be said the majority had made a positive finding that Ms Wiersma was motivated to conduct a poor investigation because of the allegation of discrimination within the appeal letter (ET, paragraph 311), that was unsupported by the evidence. As for the failure to provide a process of appeal, the ET had unanimously concluded that was “not suspicious” (ET, paragraph 297.2); as such, it was difficult to see how the majority could then conclude (absent alternative explanation) that the reason the claimant was not offered a process of appeal was because of the complaint itself.

For the claimant it is said this is an appeal on the facts: the ET majority had drawn permissible inferences akin to findings of fact (Base Childrenswear). As for the respondent’s reliance on A v CC West Midland Police, that case did not state that a failure to adequately investigate a complaint could never be victimisation; it allowed that it might, in some circumstances. This was not a case where the respondent was presenting an unreasonable employer defence, which would require cogent evidence (Anya, paragraph 14), and the majority’s decision had to be seen in context: the respondent had denied the detriments, arguing there was an adequate investigation and the findings were not superficial; the ET majority had, however, found the allegations of detriment were well-founded and no non-victimising reason had been shown.

More specifically, the ET majority had been entitled to draw an inference of victimisation given: (i) the investigation bore no resemblance to the anti-harassment policy (ET, paragraph 301); (ii) prior to the protected act, the claimant was given information about how she could appeal and the time limit, but afterwards the respondent delayed contacting her, failed to keep notes, and failed to disclose internal appeal correspondence (ET, paragraph 303); (iii) the majority did not believe there was no such correspondence and found the lack of disclosure suspicious (ET, paragraph 304), see Igen and Base Childrenswear; (iv) there was no engagement with the claimant’s points and no analysis or investigation (ET, paragraph 305); (v) the inconsistencies in Ms Wiersma’s response meant that simple, obvious steps were not taken or information was concealed (ET, paragraphs 309, 309.1 and 309.2); and (vi) Ms Wiersma failed to engage with information about potential comparators (ET, paragraph 310). The ET majority was (permissibly) not persuaded that such a poor investigation and outcome letter would have resulted from a mere appeal, finding the discrimination allegation had motivated Ms Wiersma to deny discrimination without analysis (ET, paragraph 311).

As for the respondent’s arguments: (i) finding the appeal outcome was pre-determined did not rule out victimisation; (ii) similarly, holding that the failure to provide an appeal against the discrimination decision was not suspicious did not mean the majority was debarred from making findings on victimisation or that such findings were less likely; and (iii) no automatic inference had been drawn from the absence of disclosure, rather the ET majority had carefully considered the totality of the relevant evidence.

The cross-appeal is not contingent on the outcome of the appeal but is a stand-alone challenge to the ET’s rejection of the direct discrimination claim, by which the claimant pursues six grounds of cross-appeal.

By the first ground, the claimant contends the ET failed to identify and grapple with all relevant primary facts relied on as sufficient for inferring discrimination (Field v Steve Pye; Ion v Citu); in particular it failed to engage with the matters set out in the particulars of claim and closing submissions, or with its own findings that her dismissal was pre-determined; the respondent’s actions were unreasonable; and the redundancy was a sham to retain Mr Makowski. Relatedly, by the second ground, the claimant says the ET impermissibly failed to consider whether an inference of discrimination could be drawn from those primary facts.

By the third ground, the claimant says the ET erred in failing to make a finding on a key issue, namely why Mr Makowski was stopped from leaving and then promoted to fill the claimant’s position (the reason for her dismissal; ET, paragraph 245.1). The ET found that Mr Makowski (a white male) was preferred to the claimant (a black female) without exploring why, the crucial question requiring to be answered (Nagarajan at p 511E-G). This was all the more significant, given the ET found the respondent had taken extraordinary steps - including lying - to retain Mr Makowski and dismiss the claimant whilst: (i) finding no business reasons to justify dismissing the claimant (ET, paragraph 254.1); (ii) there was no evidence to suggest Mr Makowski (more junior; less experienced) was more productive or efficient than the claimant (more senior, more experienced, with an unchallenged performance record); (iii) evidence supporting the real reason was concealed until the working day before the final hearing; (iv) the redundancy was a sham; and (v) the claimant was one of few black women working for the respondent (and the only one at her level) and had said that Mr Rebain was distant with her, whereas she felt he was more comfortable with Mr Makowski due to him his being white.

By the fourth ground, the claimant contends the ET impermissibly applied too high a threshold when applying the first stage test under section 136(2) EqA; in the alternative, by the fifth and sixth grounds, she says the ET’s decision was perverse or it failed to give adequate reasons for its conclusion.

In relation to the first ground of cross-appeal, the respondent says the ET’s decision set out its extensive primary findings of fact, demonstrating engagement with the matters relied on by the claimant (albeit it was not required to address every such matter) and considering the inferences that could properly be drawn. As for the second ground, the respondent contends the ET did not make the error identified in Field v Steve Pye, of moving directly to the ‘reason why’ question; it expressly recognised that although finding race was not the conscious motive for the claimant’s dismissal, it needed to consider whether it played any part (consciously or unconsciously) in Mr Rebain’s decision. Having properly identified the inference it was being asked to draw (Efobi, paragraph 43) – whether the respondent would have made different decisions in relation to a hypothetical comparator of a different race – the ET permissibly concluded that it could not do so.

Turning to the third ground of cross-appeal, the respondent observes there is no challenge to the ET’s finding that Mr Makowski was not a statutory comparator; it had thus asked the correct question, namely whether a hypothetical EMEA financial controller of a different race to the claimant would have been treated any differently, permissibly concluding that the burden of proof on that issue had not shifted to the respondent. As for the contention at the fourth ground, that the ET applied too high a threshold in its application of section 136(2), the respondent points to the ET’s correct self-direction on the law, such that it should be assumed to have applied the correct legal test unless its language demonstrated otherwise (DPP Law v Greenberg).

As for the perversity challenge posed by the fifth ground, the claimant could not show an overwhelming case that no reasonable ET, properly directed as to the evidence and law, could have reached the decision that it did, and, contrary to the contention in ground six, the ET’s reasons were plainly adequate.

In considering the questions raised by the appeal and cross-appeal, I have approached the ET’s reasoning in the same order as set out within its decision: addressing, first, its conclusions on the claim of direct race discrimination (the subject of the cross-appeal), before turning to the majority’s finding on victimisation (the subject of the appeal).

Addressing the claimant’s complaint of direct race discrimination, the ET was concerned with two related allegations of less favourable treatment: that the claimant had been singled out for redundancy, and that she had been dismissed. For the reasons it explained (and which have not been challenged), the ET rejected each of the possible statutory comparisons postulated by the claimant, going on to consider the complaint of direct race discrimination against the hypothetical comparator it had constructed, namely someone of a different race to the claimant who was the EMEA financial controller based in Uxbridge. Asking whether there were facts from which it could infer (absent explanation from the respondent) that different decisions would have been made in relation to that hypothetical comparator, the ET answered that question in the negative, holding that the burden under section 136(2) EqA had not shifted.

Given the ET’s primary findings of fact on the evidence in this case, I initially found this conclusion troubling. No doubt mindful of the guidance at paragraph 46 Field v Steve Pye, the claimant’s closing submissions detailed the evidence that she said demonstrated facts from which an ET could conclude that her treatment had been because of race (essentially the matters set out in support of the third ground of appeal). Many of those matters were accepted by the ET or were not really in dispute. Thus, (i) the ET had found no business reason for Mr Rebain’s decision to move the claimant’s duties to Mr Makowski; (ii) there was nothing to suggest that Mr Makowski was seen as more productive or efficient than the claimant; (iii) as a matter of record, it was only on the last working day before the hearing that the respondent disclosed the emails between Ms Chen and Mr Rebain relating to Mr Makowski’s resignation, which the ET found to be illuminative of the decision-making process; (iv) the ET had found the respondent’s reliance on redundancy as the reason for dismissing the claimant was a sham; and (v) while the ET made no finding on the claimant’s evidence about Mr Rebain being distant with her, there does not appear to have been any dispute that she was one of only two black employees at managerial level (and the only black woman). More generally, given the ET’s finding that the respondent had deliberately created a sham redundancy, having pre-determined that the claimant would be dismissed purely in order to retain Mr Makowski, it might be thought that the respondent’s actions called for explanation.

As the respondent points out, however, this is not a case where the ET (contrary to the steer provided in Field v Steve Pye) by-passed the initial stage under section 136(2); on the contrary, it actively engaged with the question whether there was evidence from which (absent explanation by the respondent) it could infer race discrimination in the claimant’s selection for redundancy and dismissal. Moreover, in so doing, the ET expressly stated that it had considered “all the facts”, and took into account that it had found the respondent’s conduct to have been unreasonable (ET, paragraphs 265-266); this is not a case where it can properly be inferred that the ET reached its decision without first grappling with the evidence in issue (paragraph 44 Field v Steve Pye). Furthermore, given that a decision whether or not to draw an inference is to be treated as a finding of fact (Base Childrenswear), regardless of the view I might have formed myself, unless some error of principle can be identified in the ET’s approach, it would not be open to me to interfere with the conclusion thus reached by the ET, unless it is properly to be said to be perverse (Efobi).

In pursuing her challenge to the ET’s decision, the claimant does not shy away from arguing that the conclusion reached at the first stage under section 136(2) was perverse. She also submits, however, that the ET applied too high a threshold - the “something more” required at the first stage need not be a great deal (Deman) – and failed to ask the crucial question, as to why the respondent strove so vigorously to retain Mr Makowski (the decision that ultimately led to the claimant’s dismissal).

Although I would agree that the ET’s decision leaves unanswered the question why Mr Rebain should have been so convinced of the need to retain Mr Makowski, the flaw in the claimant’s argument is that it assumes a direct comparison between their respective cases. That, as the ET found, would be inapt: the claimant’s circumstances were not materially the same as a more junior employee, based in Warsaw, who had given three months’ notice of his resignation on the basis that his job had not provided him with the responsibilities he had been led to expect. Understandably, there is no challenge to the ET’s rejection of Mr Makowski as a direct statutory comparator. More than that, however, the ET’s findings of fact demonstrate a clear distinction in the different stages of the decision-making process in this case, which explains why it did not answer the question the claimant identifies. On this point, it is significant that the ET accepted Mr Rebain’s evidence that, in reaching the view that he should strive to retain Mr Makowski, he made no promise about any particular post, including the claimant’s. Whatever the motivation for his decision to try to keep Mr Makowski on board (and, to the extent that this is part of the claimant’s argument, there was nothing to suggest that Mr Rebain was in any way motivated by Mr Makowski’s race), at that stage, Mr Rebain had not seen this as giving rise to a binary choice between Mr Makowski and the claimant. Rather, as the ET found, it was only when then determining how he could implement the decision he had reached in relation to Mr Makowski that Mr Rebain developed the plan to move the claimant’s duties to Warsaw. The ET thus permissibly approached the comparative exercise required by section 13 EqA as requiring it to consider whether, in circumstances in which it had been decided that steps had to be taken to retain Mr Makowski (which required the respondent to address his sense of grievance that his job had not involved the responsibilities he had anticipated), an Uxbridge-based EMEA financial controller of a different race would have been treated any differently to the claimant.

Considering that question, although the claimant had given some evidence of her perception that Mr Rebain had been distant with her, the ET made no finding that this was so, still less that this might have been indicative of some (unconscious) racial bias. As for the other matters relied on by the claimant, and taking into account the findings it had made adverse to the respondent, the ET was unable to see that any facts had been established from which it could conclude that a different decision would have been made had the claimant not been black. In so finding, the ET was not setting an unduly high threshold: it took into account its finding that the respondent had acted unreasonably but it was entitled to see that in the context of its desire to achieve its prior (non-discriminatory) objective of retaining Mr Makowski. Once it had been determined that, by moving the claimant’s duties to Warsaw, Mr Makowski could be offered the expanded role he desired, the singling out of the claimant for “redundancy” was inevitable, as was her subsequent dismissal. Similarly, given that the respondent knew the decision was pre-determined and had to be implemented if Mr Makowski’s retention was to be achieved, the failure to deal with the claimant in a transparent way was as understandable as it was unattractive. As the case-law makes clear (Zafar; Bahl) the test under section 136(2) EqA is not whether the respondent acted badly; moreover, in this case, there was a clear explanation for the respondent’s unreasonable conduct. The ET’s reasoning discloses no error of approach and, given the primary findings of fact it had made, I cannot say that the conclusion it reached was one that was not open to it applying the relevant legal principles to the facts in this case. As such, I am bound to dismiss the claimant’s cross-appeal.

Turning then to the majority decision on the victimisation claim (the subject of the respondent’s appeal), this related to the claimant’s interlinked complaints that the respondent failed to adequately investigate her allegations of discrimination, or provide a process by which she could appeal the decision on those allegations, and that it made superficial findings in dismissing the allegations. In this regard, it was conceded that the claimant’s appeal against her dismissal constituted a protected act for the purposes of section 27 EqA.

It is not uncommon for a claim to be made that the inadequacies of an employer’s investigation into a complaint of discrimination give rise to an act of victimisation. All too often, however, reliance is placed on the mere context of the employer’s failings without proper engagement with the question whether that context – the complaint of discrimination – was in fact the reason why the investigation was inadequate in the way alleged. As the EAT pointed out in A v CC West Midlands, a failure to properly investigate a complaint of discrimination need not amount to an act of victimisation: indeed, where an employer has embarked upon a process of investigating such a complaint, it might be thought unlikely that it has then failed to exercise sufficient care, enthusiasm or speed in that process because of the particular nature of the complaint. While it is thus important not to confuse context with causation, that is not to say that the particular subject matter of the complaint under investigation – raising allegations of discrimination - might not be the reason for the inadequacy in issue. Experience shows that sometimes those charged with carrying out an investigation can react in a particularly defensive way when the complaint in question includes allegations of discrimination. In such circumstances, there can be a desire not to carry out the investigative task as thoroughly, or as speedily, as would otherwise be the case; the complainant might thus suffer a detriment in the way their complaint is dealt with because it includes such allegations. Whether or not that is so will, of course depend on the ET’s findings as to what, on the facts of that case, really caused the failings in issue; as always, it is necessary to ask why the impugned conduct (the inadequate, superficial, or delayed investigation) occurred (Nagarajan).

In the present case, the ET was unanimous that the claimant had suffered the following detriments in relation to the complaints of discrimination that she raised as part of her appeal: (1) the respondent failed to adequately investigate those allegations; (2) no right of appeal was offered against Ms Wiersma’s decision in respect of the allegations; and (3) the respondent’s findings on the discrimination complaint were superficial. That said, in relation to (2), the ET was also agreed that this could be discounted (“not suspicious”): the discrimination complaint was dealt with as part of the appeal against dismissal and the ET essentially accepted that Ms Wiersma had been entitled to consider that it was neither necessary nor appropriate to offer a yet further appeal in that context. Although the ET majority referred back to the respondent’s failure to offer an appeal in respect of the finding on the discrimination complaint, I cannot see that this was then the subject of any finding of victimisation. The issue on appeal is thus whether the majority erred in finding that the burden of proof had shifted in respect of: (1) the failure to investigate, and (3) the superficial findings that were made.

In questioning whether the claimant had established facts from which the ET could draw an inference of victimisation, the majority contrasted the process for conducting investigations, laid down by the respondent’s anti-harassment policy, with that adopted by Ms Wiersma in this case, permissibly concluding that the investigation conducted in relation to the claimant’s complaint bore no resemblance to the guidance provided under the policy. It was in that context that the majority placed stress on the respondent’s failure to document the steps taken, a failure that stood in stark contrast to the process identified under the anti-harassment policy. The subsequent reliance upon the failure to keep notes and to disclose internal correspondence must thus be seen in context: the majority was not simply engaged on a “tick-box exercise” (D’Silva), it was testing the process adopted in response to the claimant’s complaint of discrimination against the standards the respondent had itself set down in its policy. The majority also took into account what it saw as a clear difference in the way the respondent communicated with the claimant before and after she submitted her appeal. Although the ET had found that the pre-termination consultation meetings were essentially a sham, and that the decision to dismiss had been pre-determined, the process had included a series of regular meetings and the termination letter had advised the claimant of her right of appeal. When she put in her appeal (the protected act) on 1 November 2020, however, there was no substantive response to her until 19 November 2020 (after her employment had come to an end), no engagement with questions raised by the claimant, and no meeting until 12 January 2021. In the circumstances referred to by the ET majority, it was entitled to find that these facts – the failure to carry out the investigation to the standards the respondent had itself laid down, and the apparent change in approach after the submission of the claimant’s appeal – were relevant matters when considering whether an adverse inference should be drawn.

More specifically, however, the ET majority found that the respondent had failed to demonstrate that the claimant’s allegation of discrimination played no part whatsoever in the inadequate investigation and superficial findings relating to that complaint (ET, paragraph 306). This, in my judgement, is the more problematic part of the majority’s reasoning. While entitled to have regard to the inconsistencies and inadequacies identified, these relate to the claimant’s appeal against dismissal as much as any complaint of discrimination, and yet there is no engagement with the ET’s earlier unanimous finding that Ms Wiersma approached her task in the way that she did because the decision to reject the claimant’s appeal was as pre-determined as the decision that she should be dismissed (ET, paragraph 253). Accepting that the drawing of an inference is to be treated as a finding of fact, I cannot see that the ET majority has taken into account the relevant unanimous finding of predetermination as part of the overall process, pre-dating the protected act. The difficulty that arises in this regard is apparent when setting the majority’s finding at paragraph 311 (“We are not persuaded that the Respondent would have carried out such a poor investigation, and written such an inadequate outcome letter, if it had been simply dealing with an appeal against dismissal. It was the allegation of discrimination which motivated Ms Wiersma to simply deny any wrongdoing by the Respondent whatsoever, without any considered analysis.”) alongside the earlier unanimous finding at paragraph 253 (“… The fact that [Ms Wiersma’s] decision was going to be to reject the appeal was pre-determined. Had she felt otherwise, she would have approached the matter entirely differently, and would be able to show us (for example) emails requesting documents from Rebain and Blunnie which she could peruse in her own time (rather than simply be shown them on a shared screen during a video meeting, which is her account of what happened) and would be able to show that she had prepared lists of questions that needed to be answered, and that she had taken a careful note of what she had been told”). Avoiding an overly technical approach to section 136(2) EqA (Laing; Hewage), and given the positive finding that the ET had made as to why Ms Wiersma undertook her investigation in the way that she did, I cannot see that the majority’s conclusion on victimisation can be upheld: it is rendered unsafe by the failure to take into account the ET’s own (unanimous) earlier finding.

Having reached the decision that the ET majority’s conclusion cannot stand, I duly allow the respondent’s appeal. The question then arises as to whether this is a matter that must now be remitted for reconsideration or re-hearing, or whether there can in fact only be one answer to the victimisation claim given the ET’s (unanimous) finding as to the reason why Ms Wiersma dealt with the claimant’s appeal in the way that she did.

In my draft judgment, circulated to the parties in advance of handing down my decision, I expressed my preliminary view that, applying the test laid down in Jafri v Lincoln College [2014] ICR 920 CA, the only possible answer to the victimisation claim, flowing from the findings made by the ET itself, must be that it fell to be dismissed. The claimant having sought to contest that view, however, I allowed the parties the opportunity to make written submissions on the question of disposal and I have duly re-visited this issue in the light of those representations.

For the claimant it is urged that there is in fact more than one possible outcome to the victimisation claim: the ET majority’s decision might be upheld on the basis that it had previously simply failed to make clear that it had had regard to the unanimous finding on predetermination, or that finding might now be taken into account but not change the earlier decision. Moreover, in assessing the possible relevance of the finding that the outcome of the appeal was pre-determined, the claimant says that the following questions of fact arise, that cannot be for the EAT to resolve: (1) at what point was the outcome of the appeal pre-determined? (2) what was Ms Wiersma’s reason for predetermination? (3) was there an obvious reason for Ms Wiersma’s conduct (there did not appear to be, which might lead the ET (majority) to hold that the burden had shifted)?

The respondent argues, however, that, given the primary findings of fact unanimously made by the ET, the only possible answer to the victimisation claim must be that it falls to be dismissed. It contends that the claimant’s submissions amount to an attempt to re-argue her case on the substantive appeal, namely that the inference that the ET majority drew was a permissible one notwithstanding the unanimous finding that Ms Wiersma approached her task in the way that she did because the decision to reject the claimant’s appeal was as pre-determined as the decision that she should be dismissed.

In determining this question, I again remind myself of the fact sensitive nature of the assessment that the ET had to carry out. Whether or not to draw an inference of discrimination is a matter for the ET at trial, and it is apparent that each of the members of this panel approached their task with great care. Moreover, in determining whether the burden had shifted under section 136(2) EqA, I have accepted that the ET majority was entitled to see it as relevant that the respondent had failed to carry out the investigation into the discrimination complaint to the standards it had itself laid down, and that there was evidence of a change in approach after the submission of the claimant’s appeal (the protected act). That said, such unreasonable conduct, without more, would not be sufficient to cause the burden to shift (Zafar), and this was not a case where the ET had found no explanation for the inadequate investigation and failure to engage with the claimant’s points (albeit that explanation was largely derived from the ET’s rejection of the respondent’s case on the fairness of the process it had undertaken).

Allowing that there might be more than one reason why an employer fails to carry out a proper investigation, I further note that the allegations of discrimination in the present case did not raise separate issues, but were part and parcel of the claimant’s appeal against dismissal. In considering the response to the dismissal appeal, the ET unanimously made a clear finding (ET paragraph 253) that the inadequacies of Ms Wiersma’s approach were because she knew she was going to uphold the dismissal and that, if her decision had not been pre-determined, she would have approached her task differently.

In suggesting that there are unresolved questions of fact in this regard, the claimant is seeking to re-open the submissions made on the substantive appeal, and fails to engage with the unanimous conclusion reached by the ET, that the outcome of the appeal was pre-determined regardless of whether it encompassed any protected act. Given that finding, I am simply unable to see how it would then be open to the ET (majority) to reach a different conclusion in relation to the allegations of discrimination. This is not a case where (as the claimant has suggested) there was no explanation for the respondent’s unreasonable conduct. On the contrary, the ET unanimously found that the unreasonable and unfair process conducted by the respondent was because it had been determined that it needed to transfer the claimant’s work to Mr Makowski in order to retain him in the organisation. As part of that unfair process, the ET found that Ms Wiersma (who was not “completely separate from the original decision”; ET paragraph 250) had conducted her role in the way that she had because her decision on the appeal was pre-determined. Acknowledging the respect that is to be shown to the decision of the first-instance tribunal, given the finding made in respect of the appeal process as a whole, I am unable to see that it would then be open to the ET (majority) to reach a different conclusion in relation to the allegations of discrimination.

For the reasons provided, I therefore: