United First Partners Research v Carreras

This is an appeal from a decision of the Employment Appeal Tribunal (HH Judge Eady QC sitting alone) dated 7 April 2016. Judge Eady allowed the Claimant’s appeal from the decision of an employment tribunal sitting at London Central, chaired by Employment Judge Henderson, dismissing his claims of (constructive) unfair dismissal and disability discrimination. To avoid confusion, I will refer to the parties to this appeal as they were in the original proceedings, though the Respondent is the appellant before us.

The Claimant was represented before us by Mr John Mehrzad and the Respondent by Mr Sami Rahman, who both also appeared in the ET and the EAT.

I need only set out the facts in bare outline.

The Respondent is an independent brokerage and research firm. The Claimant started work with them as an analyst in October 2011.  Initially he worked very long hours, typically from about 8 or 9 in the morning to between 9 and 11 in the evening: the evening hours were in order to cover the US markets. But on 22 July 2012 he was involved in a cycling accident in which he was severely affected, both physically and emotionally. Although he returned to work within a few weeks – far too early, as the witnesses in the ET thought – he continued to experience serious symptoms of dizziness, fatigue and headaches, and he had difficulties concentrating and focusing. As a result he was not able on his return to work the same hours as before.

The Respondent was aware of the Claimant’s symptoms, though it never saw or sought any medical report about his injuries, and it was content for him to say how long he felt able to work. In the first six months after his return he worked a maximum of eight hours a day.  After that he began to work longer hours, starting at 8 in the morning and going on till 6.30 or 7 in the evening. 

From about October 2013 the Claimant began to be asked to work later in the evenings and, when he agreed, an expectation began to develop that he would do so. By the date of his resignation on 14 February 2014 there was, as the ET found, an assumption he would be working one or two evenings a week, the Respondent asking him which nights he would be working late rather than whether he was prepared to work at all. The Claimant found working these very long hours difficult, but he made no formal complaint until 14 February.

Two other problems upset the Claimant. The first was the late payment of bonus instalments, apparently as a result of the Respondent’s cash-flow difficulties. The second involved the supply of inaccurate earnings figures to the solicitors who were pursuing the personal injury claim arising out of his accident.

On 14 February 2014, the Claimant sent an e-mail formally objecting to working late in the evenings.  Later that day there was a heated exchange between him and one of the owners of the business, Mr Mardel, arising out of what Mr Mardel interpreted as a critical comment made by the Clamant about one of his colleagues: although the Claimant’s e-mail was not the subject of the exchange, Mr Mardel was (as the ET found) aware of it.  At para. 3.30 of its Reasons the ET summarised Mr Mardel’s conduct at the meeting as follows:

“- He raised his voice to the Claimant.

- He deliberately reprimanded the Claimant in front of other employees saying that the Claimant continually criticised his colleagues.  This was done to make an example of the Claimant.  The Tribunal did not hear any evidence from the Claimant that he felt humiliated although he did tell Ms Barlow [the Respondent’s Head of Compliance] he found Mr Mardel’s behaviour to be abusive, unacceptable and intimidating.

- Mr Mardel told the Claimant he could leave if he did not like it.

- Mr Mardel demanded an apology from the Claimant for his behaviour.

- Mr Mardel did not seek to resolve the issue with the Claimant following the incident.”

The Claimant left the office, returning after some two hours when he went to see Ms Barlow to say that, as noted above, he thought Mr Mardel’s behaviour was abusive and unacceptable and that he was resigning.  She asked him to confirm that in writing, which he did by an e-mail which said simply “I hereby resign”.  In his evidence to the ET the Claimant said that he resigned because he was unhappy with Mr Mardel’s conduct, but he also said that he expected Mr Mardel would ask him to return and that, if he had, he would probably have stayed. 

Following receipt of the Claimant’s e-mail Ms Barlow wrote to him reminding him of his post-termination obligations.  In response, on 18 February 2014 he sent a lengthy e-mail setting out in detail his reasons for resigning. He gave as his reasons not only Mr Mardel’s conduct on 14 February but also (as summarised at para. 2.14 of the ET’s Reasons):

“- failure to pay his bonuses at the correct times

- forcing him to work the evening shifts/later hours

- providing inaccurate information to his solicitors as regards his working hours (which adversely impacted on his personal injury claim)

- refusing to correct such inaccurate information.”

The Claimant brought proceedings in the ET for disability discrimination, specifically by failure to make reasonable adjustments contrary to sections 20-21 and 39 (5) of the Equality Act 2010, and for unfair dismissal. The unfair dismissal claim was advanced on the alternative bases (a) that Mr Mardel’s conduct on 14 February 2014 had amounted to an actual dismissal and (b) that the Claimant had been entitled to resign as a result of a fundamental breach of contract by the Respondent and had accordingly been constructively dismissed.

Nothing turns for the purpose of this appeal on the specific provisions of the Employment Rights Act 1996 relating to unfair dismissal nor most of the applicable provisions of the 2010 Act. But I should set out sub-sections (1)-(4) of section 20, which read:

“(1)

Where this Act imposes a duty to make reasonable adjustments on a person, this section, sections 21 and 22 and the applicable Schedule apply; and for those purposes, a person on whom the duty is imposed is referred to as A.

(2)

The duty comprises the following three requirements.

(3)

The first requirement is a requirement, where a provision, criterion or practice of A's puts a disabled person at a substantial disadvantage in relation to a relevant matter in comparison with persons who are not disabled, to take such steps as it is reasonable to have to take to avoid the disadvantage.

(4)

The second requirement is a requirement, where a physical feature puts a disabled person at a substantial disadvantage in relation to a relevant matter in comparison with persons who are not disabled, to take such steps as it is reasonable to have to take to avoid the disadvantage.”

I will, in the usual way, use the shorthand “PCP” to refer to a “provision, criterion or practice” within the meaning of sub-section (3).

The claim was heard in the ET over three days in January 2015. There had been an unsuccessful attempt at a case management hearing to produce an agreed list of issues. Such a list was, however, agreed at the start of the hearing, as follows:

“Disability discrimination

(1)

Was the claimant disabled at the material time (namely October 2013 to 14 February 2014)?

(2)

Was there a failure to make reasonable adjustments? (section 20 EA)

(3)

Was the respondent aware of the disability?

Unfair dismissal

(4)

Was the claimant dismissed? (section 95 (1) (a) Employment Rights Act 1996 (ERA)

(5)

If the claimant was not dismissed was he constructively dismissed? (section 95 (1) (c) ERA)

(6)

What was/were the repudiatory breach(es) alleged?

(7)

Were those breaches the reason for the claimant’s resignation?

(8)

Was there any delay in the claimant’s responses to the alleged breaches?”

(I have supplied the numbering for ease of reference.) There appears also to have been a discussion with counsel at which what was involved in those issues was to some extent clarified or amplified.

I should mention one feature of the hearing which is the basis of one of the grounds of appeal. The Tribunal did not hear submissions from the parties at the conclusion of the evidence because time had over-run. Instead, it set a timetable for the provision of written submissions and for responses to those submissions, to be received in advance of its deliberations in chambers. At para. 2.28 of the Tribunal’s Reasons it identifies the written submissions to which it had regard, referring to both parties’ first-round submissions but not to the Claimant’s reply submissions. It is common ground that the Claimant had indeed sent such submissions. It is not clear whether they had not reached the Tribunal as a result of some administrative error or whether it did in fact have them and accidentally omitted explicitly to refer to them.

By a Judgment with Reasons sent to the parties on 8 May 2015 both claims were dismissed. I shall have to come back to aspects of the Tribunal’s reasoning in more detail in due course, but in essence:

The Claimant appealed to the EAT against both elements in the ET’s decision: as regards the unfair dismissal claim he accepted that there was no actual dismissal, but he challenged the finding on constructive dismissal. The Respondent cross-appealed on the basis that the ET should have found that the Claimant had waived any breach on its part.

By a judgment sent to the parties on 25 May 2016 Judge Eady allowed the Claimant’s appeal and dismissed the Respondent’s cross-appeal. In brief:

I should mention one particular point about Judge Eady’s judgment at this stage. Having found that the ET was wrong in its approach to the PCP (point (1) above), she said, at para. 37, that the reason why it had erred might be that it had reached its decision without sight of the Claimant’s reply to the Respondent’s closing submissions: see para. 14 above. But she made it clear that that was irrelevant in view of the conclusion that she had reached in any event.

The Respondent’s grounds of appeal are diffusely pleaded, but Elias LJ, who gave permission, helpfully identified the three grounds on which he did so. The first concerned how the EAT dealt with the ET’s failure to refer to the Claimant’s reply submissions. The second related to its decision that the ET had wrongly found that the pleaded PCP had not been established – “the PCP issue”. The third concerned the question whether the ET was entitled to find that the Claimant had not resigned in response to a fundamental breach – “the constructive dismissal issue”.

By a Respondent’s Notice the Claimant seeks to uphold the EAT’s decision on the constructive dismissal issue on the further ground that Mr Mardel’s conduct on 14 February 2014 “amounted to a repudiatory breach of the Respondent’s [the Claimant’s] contract of employment (which was accepted by the Respondent [Claimant]) in its own right” – in which case it would be unnecessary to consider whether he had resigned in response to the cumulative fundamental breach found by the Tribunal.

I will take in turn the PCP issue and the constructive dismissal issue. As part of the former I will consider the first ground, which relates to how the EAT dealt with the ET’s failure to consider the Claimant’s reply submissions.

THE PCP ISSUE

Para. 38 of the Particulars of Claim lodged with the ET1 summarises the Claimant’s various claims. Head (c) summarises the disability discrimination claim as follows:

“The respondent failed to make reasonable adjustments for the claimant in respect to his disability in calculating his remuneration and requiring [my emphasis] the claimant to work unsuitable hours.”

(The reference to the calculation of the Claimant’s remuneration reflects an aspect of the claim which was not pursued.) The pleaded basis for that part of the claim appears at para. 17, which reads as follows:

“However, as early as 9 December 2012 Mr Mardel requested Claimant to start working to 9pm. The Claimant highlighted he still felt fatigue and dizziness in the evenings. Subsequently, Mr Mardel asked the Claimant to work into the evenings more and more often. By the latter part of 2013 Mr Mardel sent almost weekly emails to Claimant requesting the Claimant work to 9pm one or two days per week. Despite continuing to feel tired and dizzy in the evenings. Eventually and under pressure, Claimant ceded to Mr Mardel’s requests and started to work later into the evenings significantly more often despite feeling unwell.”

It should be noted that that passage does not use the language of “requiring”. The picture that it paints is of repeated “requests” which create a “pressure” to agree.

In his written closing submissions to the tribunal Mr Mehrzad said, at para. 10.1:

“The PCP relied upon was for C to work evening shifts, in particular from October 2013 until 14th February 2014. By 14th February 2014 it was not a request for C to work for the late evening shift but was, in reality, merely a choice of which days C worked that shift. … In other words there was a ‘requirement’ that C worked evenings, such that there was a practice in place.”

That picks up the pleaded language of “requirement” but it makes it clear what is meant by it, namely the development over a period of an expectation/assumption that the Claimant would work in the evenings. That is essentially the same case as was pleaded at para. 17 of the Particulars of Claim.

The agreed list of issues set out at para. 13 above does not break down the particular issues arising under the reasonable adjustments claim (issue (2)). But the ET in its Reasons says:

“2.3

The provision, criterion or practice (PCP) as set out in Section 20(3) EA relied upon, is the requirement for the Claimant to work evening shifts. The Claimant disputed the existence of ‘shifts’ as such. He preferred to refer to them as later working hours. The Claimant said that he was placed at a substantial disadvantage compared to a non-disabled person (such as Ms Wong who had originally covered those shifts).

2.4

The disadvantage was that the Claimant would be particularly tired during those later shifts/hours. The Claimant said he was ‘forced’ to work the evening shifts and referred to paragraph 38(b) of his Particulars of Claim.

2.5

The Respondent denies that the Claimant was forced to work the later hours: it says that there were requests for him to do so and he complied with those requests. The Respondent had allowed the Claimant to work the earlier shifts immediately after his return following the accident and it had believed the claimant was recovering from the accident. The Respondent says the Claimant preferred the early shifts for personal, family reasons and not because of his fatigue.”

It will be noted that para. 2.4 refers to para. 38 (b) of the Particulars of Claim. That is in fact the sub-paragraph which summarises the constructive dismissal claim (though of course the factual basis of the two claims overlaps): that identifies an aspect of the repudiatory breach relied on as “attempts to force unreasonable working hours upon the Claimant”.

I understand that that passage reflects the discussion at the start of the hearing referred to at para. 13 above; but it refers to the pleadings and there is no suggestion of a departure from the pleaded case.

The ET’s primary factual findings about the circumstances in which the Claimant was asked to, and did, work later hours, are at paras. 2.47-2.56 of its Reasons. They can be summarised as follows:

The dispositive part of the ET’s reasoning is at paras. 3.8-3.16. These can be summarised as follows:

Judge Eady addressed this issue at paras. 31-36 of her judgment, as follows:

“31.

The identification of the PCP was an important aspect of the ET’s task; the starting point for its determination of a claim of disability discrimination by way of a failure to make reasonable adjustments (see Environment Agency v Rowan [2008] IRLR 20 EAT, paragraph 27).  In approaching the statutory definition in this regard, the protective nature of the legislation means a liberal rather than an overly technical or narrow approach is to be adopted (Langstaff J, paragraph 18 of Harvey); that is consistent with the Code, which states (paragraph 6.10) that the phrase ‘provision, criterion or practice’ is to be widely construed. 

32.

It is important to be clear, however, as to how the PCP is to be described in any particular case (and I note the observations of Lewison LJ and Underhill LJ on this issue in Paulley).  And there has to be a causative link between the PCP and the disadvantage; it is this that will inform the determination of what adjustments a Respondent was obliged to make. 

33.

In the present case, the ET sought to clarify the issues with the parties at the outset of the hearing.  It would have been aware that the Claimant’s case - as characterised in his ET1 - complained that, in requiring him to work unsuitable hours, the Respondent had failed to make reasonable adjustments.  A ‘requirement’ might imply something rather narrower than a PCP; after all, the adoption of the language of ‘provision, criterion or practice’ rather than ‘requirement’ or ‘condition’ - for the purposes of defining indirect discrimination - is generally viewed as heralding a broader and more flexible approach.  It is, of course, not suggested that the Claimant was here using ‘requirement’ in any statutory sense; he was simply identifying the PCP.  Whatever he was relying on, he was saying that it was a provision, a criterion or (and this might be the more natural way of seeing it) a practice.  In explaining his case more fully in the opening discussion with the ET, the Claimant described being required or forced to work the later hours.  That was to be contrasted with the Respondent’s case, which denied that the Claimant was forced to undertake the later hours: he was initially allowed to work those hours, and subsequently there were requests for him to do so.  That was the issue between the parties.

34.

Having considered the evidence, the ET did not accept the Claimant’s description of having been forced - as in coerced - to work the later hours, as such (paragraph 2.56).  Equally, however, it did not accept he was simply requested to do so.  What it found was that things progressed from totally open requests (the Respondent’s case) to an assumption that the Claimant would be working one or two later nights; asking when - not if - he would be working the later hours (see the ET’s finding at paragraph 3.11 that the Claimant was originally requested and then expected to work one or two evenings per week on the later shift). 

35.

The ET recognised that there might be real world factors that:

‘2.56. … led the Claimant to decide that it was in his interests financially/in terms of career progression for him to work late. …’

Whilst thus politic, the ET did not consider the Claimant could thus be described as forced to work those later hours, but I do not see that means that - on the ET’s own findings - there was no requirement for him to do so.  I can see that ‘requirement’ may be taken to imply some element of compulsion, but I do not read the term as limited to that and I can equally see that an expectation or assumption placed upon an employee might well suffice.  As the ET recognised by its reference to commercial or political factors (which I take to be referable to the workplace rather than more generally), employees can feel obliged to work in a particular way even if disadvantageous to their health.  In that context, characterising an employer’s expectation as a requirement - when it was recognised the Claimant was thereby placed at a disadvantage due to his disability - would seem an entirely straightforward construction of the Claimant’s case.

36.

Given that the Claimant was characterising the requirement as a form of practice by the Respondent, I agree the ET’s approach was overly technical and led it to treat the Claimant’s case as having been put more narrowly than it in fact was.”

Mr Rahman argues that the EAT was wrong to overturn the decision of the ET. Where an ET, as is good practice and as happened here, obtains the parties’ agreement to a clear definition of the issues that it has to decide, it is not only legitimate but necessary to decide those issues by reference to the agreed language; and that is particularly so in relation to the definition of the PCP, which is a central element for the analysis of a claim under section 20. The ET correctly adopted that approach. It found as a fact that the Claimant had not been “forced” or “coerced” into working late but had made a free choice to do so – see para. 2.56, quoted at para. 26 (4) above – and that finding, which had not been (and could not be) challenged, negatived the allegation that he was subject to any “requirement”.

I of course accept that it had been agreed at the start of the hearing – though in truth this only reflected the pleaded case – that the PCP relied on was a “requirement” that the Claimant work in the evenings. I also accept Mr Rahman’s submission about the importance of a tribunal not departing from the terms of the agreed issues – though the cautionary observations of Mummery LJ in Parekh v London Borough of Brent [2012] EWCA Civ 1630 (see para. 31) should not be overlooked. But the real question is what ought to have been understood by the term “requirement” in this context.

As is clear from para. 2.56 of the Reasons, the Tribunal interpreted the pleaded reference to a “requirement” as meaning that the Claimant had been “coerced” or “forced” to work, in the sense that he had no real choice whether to do so. It adopted that interpretation consciously, on the understanding that that was the nature of the Claimant’s case, and I am very alive to the dangers of an appellate tribunal differing from a tribunal at first instance about how the parties’ cases were put. But in the end I have come to the conclusion that the EAT was right to hold that the ET’s approach was too narrow. In my view the term “requirement” does not necessarily carry a connotation of “coercion” in the sense understood by the Tribunal. On the contrary, it may, depending on the context, represent no more than a strong form of “request”. In order to understand how it should have been understood in para. 38 of the Particulars of Claim (and the submissions based on it), it is necessary to read the fuller pleading at para. 17, which I have set out at para. 22 above. The allegation there is not that the Claimant was explicitly ordered to work in the evenings, or subjected to other explicit pressures which had the effect of depriving him of any real choice; rather it is that it was made clear by a pattern of repeated requests that he was expected to do so, and that that created a pressure on him to agree. Mr Rahman in his oral submissions accepted that such a state of affairs could in principle constitute a PCP – more particularly, a “practice” – within the meaning of section 20 (3) of the 2010 Act, but he said that that did not amount to a “requirement” and was not the case which the Respondent had to meet. For the reasons which I have given, I do not agree: the equiparation of “requirement” with “coercion” is a gloss on what was pleaded.

In reaching that conclusion, I have not lost sight of the fact that in his e-mail of 18 February 2014 the Claimant referred to being “forced” to work in the evenings, and that that language was continued into the pleading of the constructive dismissal claim. But again it depends what is meant by “forced”. Although literally this is indeed much the same as “coerced”, it is a word which is commonly used hyperbolically, and in fact the ET itself at para. 2.54 of the Reasons uses it in inverted commas and says that the Claimant said in his evidence “that this took the form of being put under pressure” (see para. 26 (3) above).

Mr Rahman also said in his oral submissions that at the hearing the case advanced was that the Claimant been subjected to “bullying, browbeating and threats of redundancy” in order to get him to work evenings. But he could not refer us to any passage in the Reasons or elsewhere that substantiated the contention that the case was put that way, which of course differs from the way it was pleaded. It is, I suppose, possible that, as sometimes happens, the Claimant in his oral evidence put his case unnecessarily high, but that would not justify disregarding the actual case as more moderately pleaded, in particular at para. 17 of the Particulars of Claim.

Mr Rahman placed some weight on the Tribunal’s reference in para. 2.56 of the Reasons to “commercial or political factors” that may have led the Claimant to agree to work in the evenings notwithstanding his unwillingness to do so. But those findings are immaterial to the question whether the pattern of requests and the expectation that he would work evenings constituted a PCP. The fact that because of such considerations he chose not to refuse may be relevant to one or more of the other elements in section 20, and/or to quantum; but the issue before us is limited to the question whether there was a PCP at all.

If that is the correct understanding of the Claimant’s pleaded case the EAT was right to find that the ET erred by approaching this issue on a misunderstanding of what he had to prove.

Having reached that point, Judge Eady also held that on its findings of primary fact the ET was bound to accept the Claimant’s case. That conclusion was not challenged in the grounds of appeal or the skeleton argument, but in his oral submissions Mr Rahman did contend that the right course for the EAT was to remit to the ET the issue of whether the pleaded PCP (as correctly understood) had been proved. I do not accept that submission. I agree with the EAT that the ET”s findings of primary fact clearly establish the pleaded PCP.

I would add by way of coda that if there had been oral closing submissions the misunderstanding of the nature of the pleaded case might have been avoided: this is just the kind of point that can be clarified by the dialectical process. It has regularly been said that reliance on written closing submissions alone is undesirable: see most recently Pimlico Plumbers Ltd v Smith [2017] EWCA Civ 51, [2017] ICR 657, at paras. 119 and 147.

As noted above, Judge Eady at para. 37 of her judgment observed that the ET might not have fallen into the error which it did about the nature of the Claimant’s case if it had had regard to his reply submissions. The Respondent contends that no reliance should have been placed on its omission to do so because it was simply not known how it had come about or therefore whether any real injustice had occurred. But this is a non-point. It is perfectly clear that Judge Eady’s observation formed no part of her substantive reasoning.

THE CONSTRUCTIVE DISMISSAL ISSUE

Section 95 (1) (c) of the Employment Rights Act 1996 provides that an employee is to be treated as having been dismissed for the purpose of the Act when “[he] terminates the contract under which he is employed (with or without notice) in circumstances in which he is entitled to terminate it without notice by reason of the employer’s conduct”. It is trite law that that refers to the contractual entitlement of an employee to resign in response to a repudiatory, or fundamental, breach of contract on the part of the employer. Typically such a breach will be of the so-called “trust and confidence duty” – that is, the duty on an employer “not, without reasonable and proper cause, [to] conduct itself in a manner likely to destroy or seriously damage the relationship of confidence and trust between employer and employee” (see per Lord Nicholls in Malik v Bank of Credit and Commerce International S.A. at [1997] UKHL 23, [1998] AC 20, at p. 34 A-B),

It was also common ground before us that where an employee has mixed reasons for resigning his resignation will constitute a constructive dismissal provided that the repudiatory breach relied on was at least a substantial part of those reasons: there is a good deal of authority to that effect, but we need only refer to the judgment of this Court in Meikle v Nottinghamshire County Council [2004] EWCA Civ 859, [2005] ICR 1.

At paras. 3.25-30 of its Reasons the ET reviewed the reasons for resigning which the Claimant gave in his e-mail of 18 February 2014 (see para. 10 above), and which he pleaded in support of his claim of constructive dismissal. Para. 30 contained the summary of its findings about Mr Mardel’s behaviour which I have already quoted (see para. 8 above). It then continued:

“3.31

On the basis of these findings of fact the tribunal have to determine whether Mr Mardel’s behaviour was a breach of contract. The Tribunal was referred by Claimant’s counsel to the case of Hilton International Hotels UK Ltd v Protopapa [1990] IRLR 306. The Tribunal has considered this authority but notes that there was no evidence before it that the Claimant had found Mr Mardel’s behaviour humiliating or degrading. The Tribunal also notes its finding of fact based on evidence that the Respondent’s workplace was one where colleagues did disagree and express their views forcefully and openly.

3.32

The Tribunal notes that Mr Mardel’s behaviour was not best practice; it was not appropriate behaviour for a senior manager; it was not good management – especially raising the matter in front of the Claimant’s colleagues. However, the Tribunal finds that Mr Mardel’s behaviour was not a fundamental breach of contract.

3.33

The tribunal has found that each of the alleged breaches while some might be technical breaches of contract were not individually fundamental breaches. However, bearing in mind the case of Waltham Forest London Borough Council v Omilaju [2005] ICR 481 the Tribunal finds that the sequence of acts as set out above had the cumulative effect of amounting to a fundamental breach of the implied term of trust and confidence.”

In short, the Tribunal found that, to the extent that the matters complained constituted breaches of contract, none of them by itself was a repudiatory breach; but, taken together, they amounted to such a breach. The Respondent does not challenge that finding.

As regards whether the Claimant resigned in response to that breach, the Tribunal said:

“3.34

The Respondent says that the Claimant had other reasons for resigning namely that his wife had been seeking employment in the US and he would have left in any event.

3.35

The Tribunal heard evidence from the Claimant that his wife had been looking for work in the US at the end of 2013 and had job interviews in January 2014 and had started a job in mid-April 2014. The Claimant said that he had wanted to stay in the UK for the conduct of his personal injury case in November 2014. The Claimant did leave to live with his wife in the US in March 2014.

3.36

Bearing in mind that evidence and also bearing in mind the Claimant’s own evidence (as set out below) the Tribunal finds that the Claimant did not resign in response to the cumulative effect of the various events set out above. The Claimant said that he was resigning because he was unhappy with Mr Mardel’s conduct on 14^th February 2014. However, he also said in his evidence that he expected Mr Mardel to ask him to return. This is based on the evidence in the Claimant’s own witness statement and his answer in cross-examination that if Mr Mardel had approached him he probably would have stayed with the Respondent organisation.

3.37

The Tribunal finds the Claimant did not believe that it was impossible for him to continue to work with the Respondent. The tenor of the Claimant’s witness statement at paragraphs 121-123 indicate that the Claimant had resigned, but fully expected the Respondent to attempt to change his mind or to apologise for Mr Mardel’s behaviour. This is borne out by the brevity of the Claimant’s written resignation on 14^th February, namely the words ‘I hereby resign’. It was only when that resignation was accepted and it was also pointed out to him that he needed to observe post-termination restrictions that the Claimant wrote the more detailed e-mail on 18^th February which set out the various alleged breaches.

3.38

On this basis the Tribunal finds that the Claimant did not resign in response to those breaches and, therefore, the Claimant’s constructive dismissal claim does not succeed.”

I have problems with that reasoning. The crucial finding is at the beginning of para. 3.36 – that is, that “the Claimant did not resign in response to the cumulative effect of the various events set out above”. The reasons given for that finding – i.e. the two things which the Tribunal says it “bears in mind” – are (a) the evidence about the Claimant’s wife’s planned move to America and (b) “the Claimant’s evidence … set out below”, namely that he expected the Respondent to ask him to stay and that he would probably have done so if that had happened. But neither reason seems to me capable of justifying the conclusion. I take them in turn.

As to (a), the fact that there were other reasons why resigning was attractive to the Claimant is not inconsistent with his having done so in response to Mr Mardel’s conduct on 14 February: as noted at para. 40 above, it is well-established that it is sufficient if the repudiatory breach in question is part of the reason for the employee’s resignation. The position would be different if Mr Mardel’s conduct was a mere pretext because the Claimant was going to resign anyway, at or about the same time, because his wife was moving to America; but that would be a very strong finding, which I cannot read the Tribunal as having intended to make. And even if that was the intended finding it would be hard to reconcile with the sequence of events found by the Tribunal, where the Claimant’s resignation seems clearly to be at least in part a response to Mr Mardel’s conduct. The fact that the Claimant might in the next few months have been moving to America – on his evidence, once his personal injury claim was resolved – would be very relevant to the quantum of his claim, but that is a separate matter.

As to (b), the essential question is what caused the Claimant’s resignation at the moment that he proffered it. If, as I have said above, Mr Mardel’s conduct two hours previously must have been at least part of that reason, I do not think that it makes a difference that the Claimant might have changed his mind if asked to do so or even that he expected that that might happen. The fact is that he had resigned and that Mr Mardel’s conduct was (at least in part) the cause.

I was for some time attracted by a different analysis of the Tribunal’s reasoning – namely that, while it accepted that the Claimant had resigned in response to Mr Mardel’s conduct on 14 February (which it had held not to be repudiatory in itself), the previous history of breaches (which, together with that conduct, it had held to be repudiatory) played no part in his motivation. Read on its own, the first sentence of para. 3.36 would certainly bear that interpretation. But it is hard to reconcile with the reasoning of the passage as a whole. Neither of the two reasons which the Tribunal gives for its decision – what I have designated (a) and (b) – would be relevant. Such a conclusion could only be justified by reasoning to the effect that the Claimant’s reaction to Mr Mardel’s conduct on 14 February was, so to speak, entirely self-contained and that the previous breaches did not operate (even subconsciously) on his mind; but neither the fact that he had reasons to join his wife in America nor the fact that he might have accepted an offer to come back would be relevant to such reasoning. It would also, while perhaps logically possible, be rather odd for the Tribunal to find, on the one hand, that the Respondent’s treatment of the Claimant, culminating in Mr Mardel’s conduct on 14 February, constituted a repudiatory breach – that is, that it had destroyed or seriously damaged the relationship of trust and confidence between employer and employee – but, on the other, that the Claimant was not motivated at all by the other elements in that series or the serious damage to the relationship occasioned by the conduct as a whole. In the end, therefore, I have come to the conclusion that this is not a viable analysis of the Tribunal’s reasoning.

Judge Eady’s reasons for overturning the ET’s finding appear at paras. 40-43 of her judgment. At para. 40 she referred to the fact that in the light of Meikle it was only necessary to establish that the breach relied was “a reason for the resignation [rather] rather than … the reason for that resignation”. She continued:

“41.

Here, it seems to me that the ET was led into error by the description of the point in the list of issues: whether the Respondent’s breaches of contract were the reason for the Claimant’s resignation.  In looking at its conclusions, it is apparent that the ET focused on whether this was the only reason and - finding there were others that also operated on the Claimant’s mind - considered that it was fatal if not. 

42.

That error alone would be sufficient to render the ET’s conclusions on this question unsafe.  Even if I am wrong about that, however, I also think the ET erred more generally in how it approached this question in this case.  Appreciating Mr Rahman’s observation (for the Respondent) that an ET’s decision has to be viewed as a whole and one must be careful not to read errors of law merely from a failure to set out every point, I consider the starting point here required the ET to be clear as to when the Claimant resigned.  That was not difficult as the parties were agreed that was on 14 February.  That was not a resignation by means of words spoken in the heat of the moment, but a considered view, taken after the Claimant had left the office for some two hours and had returned to inform Ms Barlow that he considered Mr Mardel’s conduct that morning to have been abusive and unacceptable and was resigning.  When required to put that in writing, the Claimant did so, sending his simple one-line email, ‘I hereby resign’.  That being so, the next task was to determine what had led the Claimant to that point and whether that included the Respondent’s repudiatory breach of contract?  It was not necessarily irrelevant to look at what the Claimant did next - that might have been evidentially relevant in terms of any inference the ET was to draw - but it would need to be considered in that context.  It is hard, for example, to see the Claimant’s subsequent move to join his wife in the US would mean that the Respondent’s conduct was not a reason for his resignation on 14 February: prior to that morning, the evidence was that he was planning to stay in the UK at least until his personal injury claim was determined in November 2014; something happened that day to change his mind.  Equally, it is difficult to see why his case should be undermined by the fact that he did not immediately explain his reasons more fully; that would not prevent the Respondent’s breaches of contract forming part of his reasoning on 14 February.  As for whether he expected to be asked back and whether he would then have returned, that (i) did not prevent his resignation on 14 February being just that, a binding resignation; and (ii) would not prevent that being, at least in part, due to the Respondent’s breach of contract. 

43.

I therefore agree with the Claimant on this aspect of the appeal: the ET’s conclusion is unsafe.  Further, on the ET’s findings, I consider there was only one permissible conclusion: that the Claimant resigned in response to the Respondent’s repudiatory breach.”

It will be seen that the criticisms of the ET’s reasoning made by Judge Eady at para. 42 broadly echo those that I have made at paras. 44-46 above, though they are not perhaps quite identical. It is convenient to say at this stage that I am more in sympathy with those criticisms than with that at para. 41: the reference in the agreed issues to “the reason” is a venial inaccuracy in a summary of this kind, and I can see no sign that the Tribunal was misled by it.

The pleaded ground of appeal is opaque, but in giving permission Elias LJ re-formulated it as follows:

“… [T]he tribunal found that the employee did not leave in response to the accumulation of incidents (para 3.36) (which it found did together constitute a repudiatory breach (para 3.33)) but in response to the incident on the 14 April (para 3.36). However, that incident was not, taken on its own, a repudiatory breach (see para. 3.32). If that finding is sustainable, it is plainly arguable that this was at least in part the basis on which the ET found there was no constructive dismissal as a matter of law.”

As appears from para. 47 above, I was initially minded to analyse the paragraphs in question in the same way – that is, as distinguishing between Mr Mardel’s conduct on 14 February taken in isolation (which did not constitute a repudiatory breach) and that conduct viewed as the culmination of a series of acts or omissions (which did constitute such a breach); but, for the reasons there given, I have come to the conclusion that that does not represent the ET’s reasoning.

Mr Rahman’s principal submission was that the ET had made a factual finding at para. 3.36 that the Claimant did not resign in response to the repudiatory breach and that that was a factual decision which must be respected. He submitted that the question essentially depended on the Tribunal’s view of Claimant’s credibility, and he referred us to some passages in its narrative findings where it had explicitly not accepted his evidence (albeit on other points). He submitted that that was the end of the matter. I do not accept that submission. The ET’s finding is not based on an explicit finding of primary fact about what motivated the Claimant to resign when he did, still less on an explicit rejection, on the basis of his credibility, of evidence given by him about his motivation. Instead it is based on the two reasons (“(a)” and “(b)”) which it gave and which I have found to be flawed, on essentially the same basis as the EAT did.

Mr Rahman’s fallback submission was that if the ET’s reasoning was flawed the case should have been remitted and the EAT was wrong to hold, at para. 43, that there was only one permissible conclusion. I do not agree. I have already said that it is impossible, on the sequence of events as found by the Tribunal, to escape the conclusion that the immediate cause of the Claimant’s resignation was Mr Mardel’s conduct on 14 February. That conduct was the latest element in the cumulative repudiatory breach that it found. As I have noted above, it might theoretically be possible for it to have found that his response was only to Mr Mardel’s conduct taken in isolation, and not to the wider series of events which it found to constitute a breach. However, as I have also noted, that would be a surprising finding, and there is no warrant for it in the Tribunal’s reasoning in this case; nor is there any indication in the Respondent’s written closing submissions to the ET that it was invited to make that distinction.

I would accordingly dismiss the appeal on this claim also. In those circumstances I need not consider the arguments raised by the Claimant’s Respondent’s Notice, which were to the effect that the ET should have found that Mr Mardel’s conduct on 14 February constituted a repudiatory breach even if viewed in isolation.

I would accordingly dismiss the appeal on both issues. I would very much hope that the parties can now give serious consideration to compromising this long-standing claim, where it must be on the cards that the costs will exceed the amount of any eventual award.

I agree that for the reasons given by Underhill LJ the appeal on the PCP issue should be dismissed. Like him I would also dismiss the appeal on the constructive dismissal issue, for the reasons which follow, which are not essentially different from his.

Constructive unfair dismissal occurs, in the words of section 95(1)(c) of the Employment Rights Act 1996, when “the employee terminates the contract under which he is employed (with or without notice) in circumstances in which he is entitled to terminate it without notice by reason of the employer’s conduct”. That conduct must amount to a repudiatory breach of contract by the employer; and, as Waltham Forest LBC v Omilaju reminds us, a repudiatory breach of a contract of employment may consist of a sequence of incidents culminating in a “last straw”. This accords with common sense. An angry public outburst against an employee by a manager, for example, may not be repudiatory as an isolated incident, but could well be the last straw if it follows other significant incidents of breach of contract by the employer. There are difficult cases in which the “last straw” relied on is relatively trivial, but this is clearly not one of them. It is unnecessary for the departing employee to give reasons for his resignation, so long as he is in fact leaving because of the employer’s repudiatory conduct (see Weathersfield Ltd v Sargent [1999] ICR 425 at 431F). The crucial question is whether the repudiatory breach played a part in the dismissal (per Elias P in the unreported EAT case of Abbycars (West Horndon) Ltd v Ford, 23 May 2008, cited with approval by Langstaff P in Wright v North Ayrshire Council [2014] ICR 77, EAT).

The tribunal found (at para. 3.32) that Mr Mardel’s conduct on 14^th February 2014 was not itself a fundamental breach of contract and (at para. 3.33) that each of the previous breaches of contract relied on by the Claimant was not fundamental either; but that, bearing in mind Waltham Forest LBC v Omilaju, “the sequence of acts as set out above had the cumulative effect of amounting to a fundamental breach of the implied term of trust and confidence”. Thus far their reasoning is entirely clear; and, as Underhill LJ has noted, the employer does not challenge the finding of a cumulative repudiatory breach.

Following the incident involving Mr Mardel on 14^th February 2014 the Claimant told Ms Barlow that he found Mr Mardel shouting at him in front of the entire office to be abusive and unacceptable behaviour and told her that he was resigning. She asked him to confirm any resignation in writing. He sent an email to Ms Barlow, Mr Mardel and Mr Hadjej saying simply “I hereby resign”.

The employer’s case before the tribunal is that this was simply a voluntary resignation prompted by the Claimant’s wish to move to the USA together with his wife, and that his reliance on Mr Mardel’s conduct was merely a pretext. The tribunal rejected that argument and, like Underhill LJ, I would have found it hard to accept. The tribunal found as a fact that the sequence of events culminating in the Mardel incident on 14^th February 2014 amounted cumulatively to a fundamental breach of contract by the employers. That leads inevitably to the conclusion that Mr Mardel’s conduct was at least the immediate cause of the Claimant’s resignation. On that basis it is irrelevant that the Claimant had a motivation for leaving his job sooner rather than later. The attraction of a move to the USA in the near future could be highly relevant to the loss of earnings element of any compensatory award, but that is a different point.

The tribunal found that the Claimant fully expected the Respondent to attempt to change his mind or to apologise for Mr Mardel’s behaviour. No doubt he did. But I do not see why this matters. The email simply said “I hereby resign”. This was clear and unambiguous and the employers immediately treated it as such, as they were entitled to do. The question is what it meant, construed objectively as contractual documents should be.

I regard it as impossible to construe the crucial email as meaning “I have no quarrel with anything that you, my employers, have done to me prior to this afternoon. I am taking Mr Mardel’s conduct today out of context and treating it as repudiatory in its own right”, which is what the Tribunal’s finding appears to me to say. It does not acquire that meaning by reference to the Claimant’s earlier conversation with Ms Barlow: that merely indicated the obvious fact that the Mardel incident was the immediate cause of the resignation. Nor can it be read as meaning “I am minded to resign unless Mr Mardel apologises for his behaviour”. A more natural construction would be “I complained this morning, as I have done before, about being made to work long hours despite my disability. Mr Mardel’s conduct this afternoon is the last straw. I am therefore resigning.”

But perhaps all these alternatives are over-analytical. The Claimant simply wrote “I hereby resign”. In doing so he clearly and unambiguously terminated the contract in circumstances in which (on the tribunal’s finding of cumulative repudiatory breach) he was entitled to terminate it without notice by reason of the employer’s conduct. His claim of constructive dismissal should have succeeded accordingly.

I would therefore dismiss the employers’ appeal on both issues.

I agree with both Underhill and Bean LJJ in relation to the PCP issue. I also agree with them in relation to the constructive dismissal issue. In that regard, it seems to me that the Claimant’s email of 14^th February 2014 was unequivocal and unambiguous. He terminated his contract of employment in circumstances in which he was entitled to do so by reason of the employer’s conduct, the tribunal having found as a fact that the sequence of events culminating in the Mardel incident amounted cumulatively to a fundamental breach of contract by the employers. In those circumstances, neither the fact that the Claimant expected the Respondent to try to change his mind nor that he was likely to move to the USA with his wife is relevant, although as Bean LJ noted, the latter may be highly relevant to the loss of earnings element of any compensatory award.

I too would therefore dismiss the employers’ appeal on both issues.