Abraham Goldstein v Marie-Pierre Herve

This appeal raises questions about the applicability of the detriment and dismissal protections provided by sections 44 and 100 Employment Rights Act 1996 (“ERA”) in the context of health and safety concerns, and perceived circumstances of danger, arising during the coronavirus pandemic in 2020.

In giving this judgment, I refer to the parties as the claimant and respondent, as below. This is my determination of the respondent’s appeal against a decision of the London Central ET (Employment Judge Joffe, sitting with lay members, Ms Venner and Ms Brayson, on 3-7 October 2022; “the ET”), sent out on 14 November 2022. By that judgment, the ET upheld claims against the respondent of protected disclosure and health and safety detriment, of automatic unfair dismissal for health and safety reasons, and of unfair dismissal under section 98 ERA. The respondent has appealed against the ET’s judgment, limited to its findings under the health and safety protections provided by sections 44 and 100 ERA. The claimant had also pursued claims before the ET against another respondent, Mr Sareen. A complaint of section 98 unfair dismissal was upheld against Mr Sareen, but he has not sought to challenge that decision and, by Registrar’s order, has been dismissed from these proceedings.

Representation before the ET was as it has been on this appeal.

The respondent and Mr Sareen had been co-founders of a financial services company called I V Capital Limited and, in May 2009, the claimant had started working as a personal assistant for the owners of the company; at that stage her contract was with I V Capital Ltd. Other than a break in 2012, when she left for another opportunity but then returned to her former role, the claimant continued in her work as a personal assistant for the respondent and Mr Sareen until the events in issue in these proceedings. Although I V Capital Limited ceased business in October 2014, and the claimant signed an employment contract on 1 December 2015, which named Mr Sareen as her employer, the ET found that was just for administrative convenience; the reality was that she had a separate employment relationship, and separate contract, with both the respondent and Mr Sareen, attending each of their homes to carry out her work which, in the respondent’s case, also included tasks for Mrs Goldstein.

In respect of the claimant’s work for the respondent and Mrs Goldstein, this was carried out at their family home in London, albeit in a separate office area. As at 2020, considering the claimant’s working arrangements at the time of the pandemic, the ET described her workplace at the Goldsteins’ home as follows:

During the first national lockdown, in March 2020, the claimant started to work remotely.At that time, the claimant was in her late forties, the Goldsteins some decades older. The claimant lived in another part of London with her partner, who fell into a high risk group at that stage of the pandemic (the claimant’s partner is of Bangladeshi descent and suffers from asthma); as the ET found, the claimant was conscious of the risks the coronavirus posed to her partner at that time.

Coming out of the first complete lockdown, in or about late May 2020, the claimant began to attend the respondent’s house again on an ad hoc basis to carry out some of her tasks. At this stage, the respondent and his family were living in their home in France. Although there was a dispute between the parties as to the proportion of the claimant’s duties that required her physical presence at the respondent’s house, the ET considered it was only around 15% of her work that could not be carried out remotely.

In late September 2020, when the Goldsteins had returned from France, there was an exchange of emails between the parties; by email of 28 September, the respondent made clear he wished to speak to the claimant about her returning to carry out her work for some of the time at his home. The claimant responded on 29 September, asking to continue to work from home, attending the office at the respondent’s house only on a once a month basis. In making this request, the claimant pointed out that she had been successfully working from her home since March, and this was a safe and proven arrangement; she also referred to the government guidance, which stated that those who could work from home should do so, and explained that she would have to travel to the respondent’s home by the London Underground and wanted to ensure both their households were kept safe.

On 1 October 2020, the claimant attended the respondent’s home and there was an in-person meeting between her, the respondent and Mrs Goldstein, which took place at the claimant’s desk. The respondent and Mrs Goldstein were not wearing face coverings and it was accepted that they were not diligent about wearing masks when working in proximity to the claimant, as they saw this as their home. The position of the respondent at this meeting was that he could not accept an arrangement where the claimant only came in once a month; he proposed instead that she come in every other week, which the claimant agreed, albeit with what she considered was a consensus understanding that she could vary her hours so as to avoid travelling on congested trains.

On 6 October 2020, the respondent emailed the claimant again, as there seemed to be a misunderstanding about which days she would be coming in. The claimant responded, clarifying her understanding of the agreement, and saying she would be coming in on Thursdays, outside congestion hours, arriving at around 11 am and leaving at 4 pm. In fact, it was her evidence to the ET that she would be given work late in the day and would then have to stay late at the respondent’s house, which caused her concern as members of his extended family would be present and would not be wearing face coverings.

At the beginning of November 2020, covid cases had started to increase and another lockdown was announced, coming into force from midnight 5 November 2020. On 4 November 2020, the claimant emailed the respondent regarding this change, including a link to the government guidance and citing the following prohibition:

The claimant duly attended the Goldsteins’ house on 5 November 2020. On the same day, the respondent sent her an email, referring to government guidelines dealing with working in other people’s homes (and sending a link to these), saying that these clearly envisaged such work continuing. Explaining that he and his wife considered that the quality of support they had been getting was lower when the claimant was working remotely, the respondent proposed that she should continue to alternate between on-site and remote working.

The ET found the respondent’s reliance on the government guidance he had cited was inapt. As it recorded, that guidance – referring to a non-exhaustive list of those who worked in other people’s homes, such as childcare providers, those who carry out repair services, and delivery drivers – was:

In his email of 5 November, the respondent had also referred to undertaking a risk assessment, as provided under the guidance he had forwarded, albeit the ET found that he never consulted with the claimant as part of such an assessment, which meant that she had no formal way to raise her concerns about the Goldsteins not wearing face masks.

On 10 November 2020, the claimant emailed the respondent to say she was unwell and would not be in that day. The following day, she sent in a sick certificate, albeit that version did not include a stamp for the claimant’s GP’s practice or GMC number; those omissions were corrected with an updated version sent in later.

On 12 November 2020, the claimant sent a resignation letter to the respondent, explaining her reasons for resigning as follows:

The respondent replied on 12 November 2020, disputing the claimant’s characterisation of her working experience and seeking to clarify the contractual arrangements between them; it was his contention that the claimant’s employment contract was between her and Mr Sareen. On 13 November 2020, the respondent emailed again, referring to a conversation he had had with Mr Sareen, who had said the claimant had agreed she would resign under their contract and it would then be open to her to enter into a new contract with Mr Sareen. The respondent stated:

Although allowing that the respondent’s email reflected his conversation with Mr Sareen, the ET accepted the claimant’s evidence that she had never in fact agreed to resign from the latter’s employment: she had a good relationship with Mr Sareen and would not have wanted to fall out with him.

The claimant responded on 18 November 2020, making clear she was resigning only from her employment with the respondent, and asking for clarification as to his intentions in respect of her notice and holiday pay entitlement.

On 19 November 2020, the respondent emailed the claimant in the following terms:

At this time, the respondent did not pay the wages, notice and holiday pay due to the claimant; only paying those sums shortly before the ET hearing. The respondent told the ET he had felt the claimant had wronged him by walking off, and that the things she said in her resignation email were hostile towards him.

In first considering the claims of protected disclosure detriment, brought under section 47B ERA, the ET accepted that the claimant’s resignation email of 12 November 2020 amounted to a protected disclosure and that she was subjected to detriments as a result of the respondent’s demand, by email of 13 November 2020, that she should resign from her post with Mr Sareen, and from the content of his email of 19 November 2020, which had accused her of unprofessional conduct, malingering with respect to her sickness absence, and of causing potential damage and loss to his business. Separately, the ET found that the respondent had not paid sums due to the claimant by way of wages, notice pay, and holiday entitlement because of the protected disclosures she had made in her resignation email. There is no appeal against these findings.

The ET then turned to the claims of health and safety detriment pursuant to section 44 ERA, first considering that made under section 44(c). The claimant had relied on her emails of 29 September, and of 4 and 12 November 2020; the ET agreed that, in her emails of 29 September and 4 November 2020, she:

The ET accepted the claimant had thereby brought to the respondent’s attention, by reasonable means, circumstances connected with her work which she reasonably believed were harmful or potentially harmful to health and safety, reasoning:

The ET then considered the claim made under section 44(1)(d) ERA, concluding that, by her email of 4 November 2020, the claimant had refused to return to her work after 5 November 2020, and finding that that refusal was undertaken in circumstances of danger which she reasonably believed to be serious and imminent, and which she could not reasonably have been expected to avert, explaining its decision as follows:

Relatedly, the ET also found that the claimant’s refusal to return to her place of work was an appropriate step for the purposes of section 44(1)(e) ERA:

Asking itself whether the detriments complained of by the claimant were as a result of these actions, the ET did not accept that was the case with the respondent’s pressurising her to come into work: that was because he wanted her to carry out work in his home for his convenience, not because she raised health and safety concerns. The ET did, however, uphold the claims of detriment in respect of the respondent’s criticisms of the claimant’s work in his email of 5 November 2020 (referring to the quality of her work being “lower than usual”), his demand that she also resign from her post with Mr Sareen, the criticisms made of her professionalism in the email of 19 November 2020, and his failure to pay wages, notice pay and holiday entitlement.

The ET then turned to the complaint of constructive dismissal, asking itself whether the respondent’s conduct had amounted to a fundamental and repudiatory breach of the implied term of trust and confidence. Although the ET found that his actions were not calculated to destroy trust and confidence, it considered:

It further found that that conduct had been likely to destroy trust and confidence:

The ET was also satisfied that the respondent did not have reasonable and proper cause for his conduct:

As the ET recorded, there was no dispute that the claimant had resigned in response to (the) respondent’s conduct and she had not delayed or waived the breach of her contract. Moreover, for the purposes of section 98 ERA, no potentially fair reason had been postulated, and the ET was unable to see that one was made out on the evidence; the respondent does not challenge the decision that the claimant was unfairly dismissed for section 98 purposes.

Turning then to the question whether the dismissal was also automatically unfair under section 100 ERA, the ET asked itself whether the reason or principal reason for the repudiatory breach of contract had been the fact that the claimant had brought the health and safety concerns to the respondent’s attention or whether it was her refusal to return to work; it concluded it was the latter:

In the circumstances, the ET upheld the complaint of automatic unfair dismissal.

Sections 44 and 100 ERA were introduced to give effect to Council Directive 89/391/EEC of 12 June 1989 on the introduction of measures to encourage improvements in the safety and health of workers at work (“the Health and Safety Directive”). Given that the focus of the Health and Safety Directive is thus on workers’ safety, that purpose makes clear the Parliamentary intent which must be kept in mind when construing the domestic provisions in issue.

Section 44 falls within Part V ERA, which provides protection against suffering detriment in employment in certain contexts. Specifically, section 44 provides such protection in what are referred to as “health and safety cases”: in the language of the provisions in force at the relevant time, an employee has the right not to be subjected to any detriment by any act, or deliberate failure to act, by his employer on the ground that (relevantly to the present proceedings):

Part X of the ERA makes statutory provision for rights relating to unfair dismissal. Chapter 1 provides for the right not to be unfairly dismissed, either in the circumstances provided by section 98 (sometimes referred to as “ordinary” unfair dismissal), or in other circumstances where, if the reason for the dismissal is a specified reason, it will automatically be regarded as unfair (“automatic” unfair dismissal). Section 100 ERA deals with what are described as “health and safety cases” and effectively replicates the provisions of section 44 (as set out above), providing that a dismissal by reason of one of the circumstances thus described will be regarded as an unfair dismissal.

Before turning to the specific circumstances described within sections 44 and 100 ERA, it is first necessary to recall that there can be no claim of unfair dismissal (ordinary, or automatic) unless it is first established that there is a dismissal. By section 95, it is provided that an employee will be dismissed by their employer if the employment contract is terminated by the employer (with or without notice), or on the coming to the end of a limited-term contract without renewal, or where:

The question whether an employee has thus been constructively dismissed, for the purposes of section 95(1)(c), is to be determined applying a contractual test: it is not sufficient that the employer has acted unreasonably, the conduct in question must amount to a breach of contract, which is sufficiently important to justify the employee resigning (or is the last in a series of incidents that would do so), and the employee must leave in response to that breach and must not delay too long, so as to amount to a waiver of the breach in question (see Western Excavating (ECC) Ltd v Sharp [1978] ICR 221 CA).

The terms of a contract of employment include an implied mutual obligation not, without reasonable and proper cause, to act in a manner calculated or likely to destroy or seriously damage the relationship of trust and confidence between employer and employee (see Malik v BCCI [1998] AC 20 HL; Baldwin v Brighton & Hove CC [2007] ICR 680). Given the trust and confidence that is a necessary characteristic of any employment relationship, a breach of this implied term will necessarily go to the root of the contract and amount to a repudiatory breach (see Morrow v Safeway Stores plc [2002] IRLR 9 EAT).

For the respondent, it is observed that a genuine, if mistaken, belief as to the interpretation of a contractual provision cannot give rise to a repudiatory breach (see Financial Techniques (Planning Services) Ltd v Hughes [1981] IRLR 32 CA); by analogy, it is contended that, similarly, it would be wrong to find that an employer did not have reasonable and proper cause for its actions if it had relied upon what was genuinely believed to be the meaning and relevance of government guidance. For the claimant, however, it is countered that purported reliance on government guidance is not the same as a belief as to the construction of a disputed contractual provision: in assessing whether there has been a breach of the implied term relating to trust and confidence, what is significant is the impact of the employer’s conduct on the employee (rather than what the employer intended), which is to be assessed objectively (per Lord Steyn, Malik pp 622F-G and 623B-D). In any event, it is, common ground that the question whether there has been a breach of the implied term will be a question of fact for the ET (see London Borough of Lambeth v Agoreyo [2019] EWCA Civ 322, [2019] ICR 1572 per Singh LJ at paragraph 61).

Returning to sections 44 and 100 ERA, the first of the protections relied on by the claimant was that under subsection 44(1)(c), relating to the employee who (unable to raise concerns through a health and safety representative or safety committee) has brought to the employer’s attention, by reasonable means, circumstances connected with their work which they reasonably believed were harmful or potentially harmful to health or safety. As the EAT observed at paragraph 28 ABC News Intercontinental Inc v Gizbert UKEAT/0160/06 (unreported, 21 August 2006), the protection is thus concerned with the raising of an issue of health and safety connected with the employee’s work; it does not, of itself, necessarily ensure his (or others) health and safety (although, I would add, it is plainly intended to assist in the achievement of that wider purpose). Moreover, while the circumstances identified by the employee must be “connected with” their work, there is no requirement that those circumstances must be limited to those within the workplace itself.

The protection provided by section 44(1)(c) may thus be contrasted to the second relied on by the claimant, under subsections 44(1)(d) and 100(1)(d) ERA. In either case, the protection under subsection (1)(d) relates to the employee who, in circumstances of danger which they reasonably believed to be serious and imminent and which they could not reasonably have been expected to avert, left (or proposed to leave) or (whilst the danger persisted) refused to return to their place of work or any dangerous part of that place of work. As the Court of Appeal made clear inRodgers v Leeds Laser Cutting Limited [2022] EWCA Civ 1659, [2023] ICR 356, the circumstances relevant under subsection (1)(d) must thus arise at the employee’s workplace:

As in the present proceedings, the case of Rodgers related to a perception of danger arising from the coronavirus pandemic. In that case, however, the ET had found that, while Mr Rodgers considered there were circumstances of serious and imminent danger from the virus “all around”, he did not hold that belief about his place of work; in the alternative, the ET was clear it would not have been reasonable for Mr Rodgers to hold such a belief as it was not hard to socially distance in that workplace and measures were in place to reduce the risk of transmission of the virus. Accepting that, for subsection (1)(d) purposes, the perceived danger need not be exclusive to the workplace:

As for the third protection relied on by the claimant in the present case, subsections 44(1)(e) and 100(1)(e) ERA also require that the employee had a reasonable belief that the circumstances of danger were serious and imminent, but not that those circumstances exist at the workplace, or be connected with the employee’s work. In Von Goetz v St George’s Healthcare NHS Trust UKEAT/1395/97 (unreported, 18 October 2001), the EAT rejected an argument that the ambit of either subsections (1)(c) or (e) should be limited to only the possibility of harm at the employee’s workplace, or to their fellow (or any) employees.

More generally, in relation to all three subsections (whether under section 44 or section 100), it is to be noted that the employee concerned need only have a reasonable belief in the (potentially) harmful circumstances, or in the existence of the danger and in its seriousness and imminence; that their belief might be mistaken will not necessarily mean it was not reasonably held (and see Rodgers at paragraph 17).

The first ground of appeal relates to what is said to have been the ET’s conclusion that the claimant’s complaints concerning her travel to work on more than one occasion a month fell within the protections afforded by sections 44 and 100 ERA. In this regard, the respondent contends: (1) the ET thereby misinterpreted the legislation; (2) gave inadequate reasons; and/or (3) reached a perverse decision.

Addressing the first objection, it is the respondent’s case that the claimant’s concerns over what she considered to be the risk she encountered during her commute to work: (a) amounted to concerns about covid-19 generally, which could not fall within the ambit of subsections (1)(c)(d) or (e) (per Rodgers); (b) did not relate to something that was “connected with” her work - as opposed to being merely incidental to it - and was thus outwith subsection (1)(c): even on the most permissive legislative interpretation, the claimant’s commute was not “connected with” her work (per Rodgers), and to find otherwise would require an employer to conduct assessments of every employee’s commute to ensure compliance with its obligations; (c) did not amount to circumstances of danger that the claimant believed to be serious and imminent, so as to fall within the protection of subsection (1)(d), given she was prepared to attend the workplace, and: (i) the danger in issue had to be of Covid-19 in the workplace, as opposed to there being a general risk, (ii) the claimant did not leave her place of work because of this danger but remained at work on 5 November 2020 and was then signed off sick until her resignation, (iii) at no point did she leave or propose to leave - the issue was about the amount of time she would spend at the workplace, (iv) the focus had to be on there being a serious and imminent danger at the workplace; and (v) a potential danger was not an “imminent” one (ABC, paragraph 32); (d) did not amount to circumstances of danger for the purposes of subsection (1)(e), given that: (i) in this regard the concern is to protect the safety of the employee from the serious and imminent danger, but (ii) the claimant was absent from work unwell when she resigned, having proposed attending the workplace once a month.

Secondly, it is contended that the ET gave inadequate reasons for its determination that: (a) the claimant’s actions came within the framework of sections 44 and 100 ERA other than “in a broader sense” and by adopting a “purposive interpretation” (ET, paragraph 152); (b) the claimant had a reasonable belief in a serious and imminent danger when agreeing to attend the workplace once a month (her proposal), but not twice (as the respondent had proposed); (c) the respondent’s request for the claimant to attend twice a month, as and when required, was objectionable.

Thirdly, it is the respondent’s case that the ET reached a perverse decision as to the application of sections 44 and 100 ERA, basing its conclusion on the understanding or premise that the claimant was refusing to attend the workplace at all (and, thus, that the respondent was requiring her attendance in the face of her absolute refusal (ET paragraphs 172 and 174)), when it was the claimant’s evidence that she had proposed to attend her place of work once a month.

By the second ground of appeal, it is contended that the ET erred in concluding that the respondent’s reliance on the relevant government guidance was “unreasonably wrong” (ET, paragraph 174). The respondent submits that the ET failed to provide an adequate explanation of its reasoning: beyond an assessment of the volume of the claimant’s work requiring attendance on site, the ET did not explain why reliance on the guidance was “unreasonably” wrong when an employer might have a different (even mistaken) belief in something and still be acting reasonably (Financial Techniques). Relatedly, it is contended the ET further erred in law in finding the respondent did not have reasonable and proper cause for his actions, such as to give rise to a repudiatory breach of the implied duty of trust and confidence.

By the third ground of appeal, it had been contended that the ET erred in its application of the burden of proof under section 48(2) ERA. In oral argument, however, Mr Salter accepted that this was not a point he could pursue, given that there was no challenge to the ET’s findings of public disclosure detriment.

It is the claimant’s case that the premises that underpin the first ground of appeal are incorrect. The ET’s decision did not depend on a finding that her complaints concerning her travel to work fell within the relevant statutory provisions; its conclusions related to her refusal to attend work after 5 November 2020 for the duration of the lockdown (the concerns raised on 29 September 2020, regarding the commute to work, arose in a different context, before the rapid rise in Covid cases and the reimposition of lockdown). More than that, reading the ET’s judgment as a whole (as the EAT was required to do; DPP Law Ltd v Greenburg [2021] IRLR 1016 CA), it was apparent that the conclusions on the claims under sections 44 and 100 were made not only in relation to the claimant’s travel to work but also in relation to her workplace once she was there (ET, paragraphs 149-150), and the findings in respect of the latter had not been challenged on appeal. Although the ET had expressed the view that the claimant’s commute would also be connected with her work (ET, paragraph 152), that was in response to a specific argument taken by the respondent. More specifically, the respondent was wrong to contend that the claimant had been prepared to attend the workplace during lockdown; the ET had made a clear finding that she was not (ET, paragraph 155).

Addressing the specific objections raised under the first ground of appeal, for the claimant it is contended that the case of Rodgers is distinguishable: in the present case, the claimant had a reasonable belief in serious and imminent danger in the workplace, not just during her commute. In any event, the claimant notes that there was no discussion of the extent to which subsections (1)(c) and (e) apply to dangers arising other than at the workplace in Rodgers; those provisions were not limited to harm, or the possibility of harm, at the workplace (see Von Goetz at paragraph 28). More generally, the claimant submits that the ET provided adequate reasons for its decision, which could not properly be characterised as perverse.

As for the second ground of appeal, it is the claimant’s case that, reading the ET’s reasoning as a whole, it was clear that adequate explanation had been provided for the finding that the respondent’s reliance on the government guidance for working in homes was “unreasonably wrong”. In any event, Financial Techniques did not assist the respondent: that case concerned the mistaken interpretation of express contractual terms, not government guidance, and it did not concern “reasonableness” but repudiation of contract. More generally, the question whether there has been a breach of the implied term as to trust and confidence was to be determined objectively (see Malik), and that was a matter for assessment by the ET as the fact finding tribunal; absent perversity, that assessment was not susceptible to challenge. In the alternative, the claimant submits that, in any event, the ET did not find the respondent acted without reasonable and proper cause solely on the basis of his reliance on the government guidance; its conclusion in the final sentence of paragraph 174 had to be read in conjunction with its earlier findings (see ET, paragraphs 171-173).

Although the protections in issue on this appeal all fall under the heading “health and safety cases”, each of the provisions addresses a different eventuality. Consistent with the underlying purpose, these protections are clearly intended to cover a wide range of potential circumstances, with the aim of protecting employees’ health and safety.

Whether considered under section 44 or section 100 ERA, subsection (1)(c) addresses the case of the employee who raises health and safety concerns with their employer when they have not been able to do that through a relevant representative or committee. It is the raising of the safety issue that is protected, not (in contrast to subsections (1)(d) and (e)) the safety of the employee (or others) (see ABC, paragraph 28), although, of course, it is entirely consistent with the aim of protecting employees’ safety that it should be possible to raise such concerns without suffering detriment or dismissal. The protection is, however, limited to the raising of circumstances “connected with” the employee’s work. It is the respondent’s case that that expression cannot be read to include travel to and from work; he relies on Rodgers to support his argument that a commute to/from work cannot be “connected with” that work, and says that to hold otherwise would place too onerous an obligation upon an employer. I cannot, however, see why “connected to” should be read in such a limited way. Travel to and from work may well be “connected to” that work, and the examples postulated by the ET at paragraph 152 of its decision (see the citation at paragraph 24, above) demonstrate why a purposive reading of the provision might well lead to such a conclusion; certainly, the judgment in Rodgers, concerned with the construction of subsection (1)(d), provides no support for such a limitation to the ambit of subsection (1)(c). Moreover, the respondent’s argument in this regard seems to be founded upon a mischaracterisation of the obligation that would then arise: the employer would not be required to conduct assessments of every employee’s commute to/from work; the protection is simply against being subjected to a detriment, or being dismissed, for raising the concern.

The protection at subsection (1)(d) (of either section 44 or 100) is concerned with the avoidance of circumstances within the workplace, which the employee reasonably believed gave rise to a serious and imminent danger. Where the conditions envisaged by subsection (1)(d) are present, an employee is entitled to leave their place of work, or to refuse to return to it, without suffering a detriment or being dismissed as a result. As the protection expressly relates to the leaving of the employee’s place of work, or the refusal to return to that place, it follows that the perceived danger must relate to the workplace itself, although it need not be exclusive to that workplace (Rodgers paragraphs 19 and 59). In contrast, subsection (1)(e) is not limited to dangers at the workplace, but protects the employee who takes appropriate steps to avoid any danger, reasonably perceived to be serious and imminent, to himself or another.

In considering a complaint of detriment or automatic unfair dismissal arising from these provisions it is, therefore, necessary to keep in mind the particular circumstances relied on.

In its consideration of the detriment claims pursued by the claimant under section 44 ERA, the ET was astute to separate out each of the protections relied on. In finding that the claimant’s emails of 29 September and 4 November 2020 fell within subsection (1)(c), the ET expressed itself carefully: she was bringing to her employer’s attention her belief as to the potential harm to health and safety that would arise from her attending the workplace by public transport (ET, paragraph 149). That was, however, not a finding of a circumstance that was merely incidental to the claimant’s work; the claimant’s concern was that by travelling on public transport to and from the two homes, and spending time in the respondent’s house, she would be increasing the risk of transmission of the virus, and thus the potential harm to health and safety. As the ET made clear: being at the respondent’s home was connected with the claimant’s work (ET, paragraph 150). In the circumstances, it was open to the ET to find that the claimant’s concern fell within the ambit of section 44(1)(c).

Then going on to consider the claims under subsection (1)(d) (relevant to the ET’s decisions on both detriment under section 44 and automatic unfair dismissal under section 100 ERA), the ET made plain its finding that, by her email of 4 November 2020, the claimant had made clear her refusal to return to her place of work after 5 November 2020. In suggesting that this did not amount to a relevant refusal for the purposes of subsection (1)(d), the respondent: (i) elides the claimant’s clear refusal to return to the workplace, communicated by her email of 4 November 2020, with her earlier proposal to attend once a month; and (ii) seeks to go behind a clear finding of fact by the ET. The claimant’s non-attendance at her place of work after 5 November 2020 was not simply as a result of her subsequent sick leave; as the ET found, her email of 4 November was a refusal to return to work after midnight on 5 November. As such, this fell firmly within subsection (1)(d).

More generally, the respondent objects that the claimant’s concerns relating to the coronavirus at this time could not give rise to a reasonable belief in circumstances potentially harmful to health and safety that were connected to the claimant’s work under subsection (1)(c), or circumstances of serious and imminent danger relevant to either subsections (1)(d) or (e). It is the respondent’s case that this is a submission supported by the Court of Appeal’s decision in Rodgers.

Again, I am unable to see that the decision in Rodgers assists the respondent’s argument. In that case, the ET expressly found that, while Mr Rodgers had a general concern regarding the coronavirus, he did not believe it gave rise to circumstances of serious and imminent danger in his place of work (and could not reasonably have believed that it did). That was not the ET’s finding in the present proceedings; on the contrary, the ET was clear: the claimant had (reasonable) concerns that related to the specific circumstances of her employment, which included the lack of social distancing and mask wearing by the respondent, his wife, and members of their extended family (ET, paragraph 157). Moreover, the fact that the claimant might also have held a more general concern as to the risk posed by the coronavirus, would not mean that she did not have the requisite belief relevant to her work (subsection (1)(c)) or her workplace (subsection (1)(d)); the potential harm to health and safety and/or the circumstances of danger need not be present only in relation to the employee’s work or at their place of work (Rodgers, paragraphs 19 and 59).

Finally, in support of his contention that the ET erred in its approach to the statute, the respondent argues that the perception of a potential danger arising from the coronavirus could not meet the requirement that it was “serious and imminent”; to bolster this argument, he seeks to rely on the finding in ABC, that the claimant in that case faced no danger, let alone a danger that was “imminent”. That, however, is an unwarranted attempt to read across from the facts of one case (a journalist who was relying on a danger arising from a war zone posting that would never have been asked of him) to the entirely unrelated circumstances of another (an employee who, at a time of rising infections during a global pandemic, was being required to attend a workplace where safeguards could not be guaranteed, at a time when there was no vaccine and her partner fell into a high-risk group). As Ms Mankau observed in oral argument, for the purposes of subsections (1)(d) and (e), it is the danger that the employee needs to believe to be serious and imminent, not necessarily the particular harm to which it might give rise (the dangers posed by the presence of asbestos particles would be an obvious example). Ultimately, however, the answer to this part of the appeal is that the ET permissibly found as a fact that the claimant had reasonably believed there were circumstances of serious and imminent danger such that she should refuse to return to her place of work, which was an appropriate step both for her protection and for that of her partner (ET, paragraphs 156-158).

Turning then to the contention that the ET failed to provide adequate explanation for its conclusions, and/or that it reached decisions that are properly to be characterised as perverse; it is the respondent’s case that the ET’s reasoning was wrongly based on the premise that the claimant was refusing to return to work when it was her evidence that she had proposed to attend her workplace once a month. That submission, however, again elides the claimant’s position as at the end of September/start of October 2020 with that which she communicated on 4 November 2020. While it is correct that the claimant had, on 29 September 2020, responded to the respondent’s request that she return to the workplace on a more regular basis by proposing that she attend once a month, her position changed on 4 November, when (as the ET found) she plainly stated that she was refusing to return to the respondent’s house once the second lockdown commenced, after midnight 5 November 2020. The ET’s reasoning is clear in this regard: in considering the claims pursued under subsection (1)(d), it was the email of 4 November 2020 that demonstrated the claimant’s refusal (ET, paragraph 155); that was a finding that was open to the ET on the evidence and its reasoning was adequately explained.

For all the reasons provided, I therefore dismiss the first ground of appeal.

Separately, by the second ground of appeal, the respondent contends that the ET was wrong to find that he had been unreasonable in his reliance on government guidance and that, as a consequence, he did not have reasonable and proper cause for his actions, such as to give rise to a breach of the implied duty of trust and confidence. In my judgement, however, that argument is misconceived on a number of levels.

First, the ET permissibly found as a fact both that it was not necessary for the claimant to go into work after 5 November 2020 and that the respondent could not reasonably have believed otherwise (ET, paragraphs 171 and 124). Given its unchallenged finding that the guidance relied on by the respondent posed the question whether working in the homes of others was necessary (ET, paragraph 54), the ET was entitled to conclude that reliance on that guidance in these circumstances was unreasonable. Second, in holding that the respondent’s conduct had been likely to destroy trust and confidence, the ET had not merely relied on matters that might have arisen from his erroneous reliance on the guidance, but had also referred to its findings that the respondent had pressed the claimant to come into work to avoid what was “only a moderate inconvenience”, that “there was a lack of empathy and flexibility” on his part, and that he had treated a long-serving employee in a “disrespectful way” (ET, paragraphs 172-173). Third, similarly, in answering the question whether the respondent had reasonable and proper cause for his actions, the ET again not only referred to the fact that his interpretation of the guidance was wrong, but also looked at the respondent’s wider conduct: his interpretation of the guidance had thus been “distorted by his desire that the claimant should attend work and save him from some inconvenience”, when it was “not reasonable to insist the claimant attend work during the November 2020 lockdown” (ET, paragraph 174). Fourth, ultimately the ET was clear as to why the respondent had acted in breach of the obligation to maintain trust and confidence: it was the claimant’s refusal, communicated by her email of 4 November 2020, to return to her place of work in circumstances of danger which she reasonably believed to be serious and imminent, such as to fall within the ambit of section 100(1)(d) ERA (ET, paragraph 178). Again, the ET’s finding goes further than simply stating that the respondent was relying on a mistaken interpretation of the government guidance: he was “provoked” by the claimant’s email, and he responded by “insisting that she should attend in terms which … breached the implied term of trust and confidence” (ET, paragraph 178). I am satisfied that there is nothing in the second ground of appeal.

Ultimately this appeal is an attempt to re-argue the case presented below. The respondent’s arguments fail to engage with the careful and nuanced reasoning of the ET, and wrongly elide the claimant’s position at the start of the November 2020 lockdown with her earlier attempt to reach an accommodation with the respondent, at the end of September 2020. The ET’s decision clearly distinguishes between those different circumstances, reaching permissible – and time-specific - conclusions under each of the statutory provisions in issue. For all the reasons provided, each of the grounds of appeal is dismissed.