Can employees refuse to attend work because of Covid-19?

A. INTRODUCTION

The Covid-19 pandemic has wrought sudden and drastic changes to the workplace. Many are now working from home, to suppress the spread of the virus and protect themselves or their families from catching it. However, working from home is a precaution which is not available to all, and which is available unequally. There has been no better example of this than last week’s drama (prior to the announcement of the third lockdown) over the reopening of schools, with the government, local authorities, unions and individual schools and teachers at loggerheads over whether pupils can safely return to schools after the Christmas break.

For those employees who are required to attend work, workplace safety will be a major concern. Many workplaces are simply not designed to operate safely in a context where a highly transmissible and potentially fatal virus is circulating in the community. Corridors will be too narrow, workstations too close together, protective equipment inadequate, and communal surfaces not cleaned sufficiently. Even after a year of Covid-19, there are many ways in which workplaces and working practices will fall short. The Government has long indicated awareness of this problem. An urgent question was put on 6 May 2020 in the House of Commons to Paul Scully, the Parliamentary Under-Secretary for Business Energy and Industrial Strategy, by the Shadow Secretary of State for Employment Rights, asking the former to make a statement on guidelines for workplace safety. The Government published guidelines on workplace safety five days later.

The question remains, for the individual employee, whether to respond to an instruction to present for work if an employer has not complied with these guidelines, or indeed, if the employer has, but the employee is still not satisfied that they will be able to work safely. The same question, indirectly, will be on the lips of employers and unions – what can they require of their employees, and what should they advise their members? In this blog post I will set out the law’s answer to this question. In short, the answer to that question is (frustratingly) that it is not at all clear how the existing legal protections for employees will apply in the pandemic. Each case is likely – to a significant degree – to turn on its own facts. Nonetheless, in what follows I will try to highlight the key questions which employees, employers, unions and advisers should consider when making decisions about requiring or refusing work in the pandemic, and which employment tribunals will consider should disputes get that far.    

B.    PROTECTION FOR EMPLOYEES WHO REFUSE UNSAFE WORK      

Employees are protected from both detriment and dismissal for certain protected health-and-safety-related conduct, including refusing to present for work in certain situations. Section 44 of the Employment Rights Act 1996 (“ERA”) protects employees from detriment on certain health and safety grounds, while section 100 ERA provides that dismissal on the same grounds will be automatically unfair.

Section 44 ERA provides:

“Health and safety cases.

(1) An employee has the right not to be subjected to any detriment by any act, or any deliberate failure to act, by his employer done on the ground that—

(a) having been designated by the employer to carry out activities in connection with preventing or reducing risks to health and safety at work, the employee carried out (or proposed to carry out) any such activities,

(b) being a representative of workers on matters of health and safety at work or member of a safety committee—

(i) in accordance with arrangements established under or by virtue of any enactment, or

(ii) by reason of being acknowledged as such by the employer,

the employee performed (or proposed to perform) any functions as such a representative or a member of such a committee,

 (b)the employee took part (or proposed to take part) in consultation with the employer pursuant to the Health and Safety (Consultation with Employees) Regulations 1996 or in an election of representatives of employee safety within the meaning of those Regulations (whether as a candidate or otherwise),

(c) being an employee at a place where—

(i) there was no such representative or safety committee, or

(ii) there was such a representative or safety committee but it was not reasonably practicable for the employee to raise the matter by those means, he brought to his employer’s attention, by reasonable means, circumstances connected with his work which he reasonably believed were harmful or potentially harmful to health or safety,

(d) in circumstances of danger which the employee reasonably believed to be serious and imminent and which he could not reasonably have been expected to avert, he left (or proposed to leave) or (while the danger persisted) refused to return to his place of work or any dangerous part of his place of work, or

(e) in circumstances of danger which the employee reasonably believed to be serious and imminent, he took (or proposed to take) appropriate steps to protect himself or other persons from the danger.

(2) For the purposes of subsection (1)(e) whether steps which an employee took (or proposed to take) were appropriate is to be judged by reference to all the circumstances including, in particular, his knowledge and the facilities and advice available to him at the time.

(3) An employee is not to be regarded as having been subjected to any detriment on the ground specified in subsection (1)(e) if the employer shows that it was (or would have been) so negligent for the employee to take the steps which he took (or proposed to take) that a reasonable employer might have treated him as the employer did.”

There are, therefore, six specific kinds of protected, health-and-safety-related conduct: (1) carrying out health and safety activities as the person designated by the employer to carry out such activities; (2) performing functions as a representative of employees on a health and safety committee; (3) conduct under the Health and Safety (Consultation with Employees) Regulations 1996; (4) bringing to the employer’s attention circumstances the employee reasonably believes to be (potentially) harmful to health and safety; (5) refusing to present for work in circumstances of danger; and (6) taking steps to protect themselves or others from such danger. While all these kinds of conduct could arise in the context of the pandemic, the most pressing question will be whether an employee who is required to return to work but does not believe it to be safe can refuse to do so. That question turns on the application of section 44(1)(d).

That is a difficult section to apply to an employee’s decision not to present for work during the Covid-19 pandemic. It is poorly suited to regulating workplace safety in such a context, so any such cases will require careful navigation. Further, sections 44 and 100 are not much litigated provisions, and there does not appear to have yet been a case decided in which a pandemic-related section 44 or section 100 claim has been determined. These are therefore difficult, largely uncharted waters.

C.    THE REQUIREMENTS OF SECTION 44(1)(d)

Section 44(1)(d) provides for one of the grounds on which an employee has the right not to be subjected to detriment. An employee who does not present for work will normally be liable to some sort of disciplinary sanction from the employer, but such consequences may not be inflicted where the employee acts as described in section 44(1)(d).

That section can be broken down as follows.

·  First, there must be circumstances of danger, which the employee could not reasonably be expected to avert.

·  Second, the employee must have a reasonable belief that this danger has two attributes: (a) imminence and (b) seriousness.

·  Third, where these circumstances and that belief pertain, section 44(1)(d) protects the following conduct: (1) leaving the place of work (or the dangerous part of it); (2) proposing to leave the place of work (or the dangerous part of it); and (3) refusing to return to the place of work (or the dangerous part of it) in circumstances where the danger persists.

D.   APPLICATION OF SECTION 44(1)(d)

(1)   “Circumstance of Danger”

It is a pre-requisite to reliance on section 44(1)(d) that there exists a circumstance of danger. It might be thought that the pandemic presents a fairly universal danger, but there are ways in which that danger might vary, which will presumably need to be taken into account by a tribunal applying the section. 

It seems unlikely that section 44 was designed to catch minimal or fanciful risks. It is likely that, faced with this question, a tribunal would consider whether the risk of the virus rose above that (or a similar) de minimis threshold. It will not therefore be enough for an employee simply to point out that Covid-19 has not been eradicated.

However, that limit aside, there are ways in which the requirement that there be a “circumstance of danger” is potentially very broad. It is apparently not defined relative to other workplaces, either within the sector in question or generally. It may not therefore be enough for an employer to show their workplace is as safe as any other workplace or as safe as it could be. Further, it is established that the relevant danger for the purposes of section 44 need not be to employees, but could be to any person: Von Goetz v St Georges Healthcare NHS Trust EAT/1395/97 §§26-27. The facts which gave rise to Ms Von Goetz’s complaint in that case are not clear from the report of the judgment, and it appears that Ms Von Goetz’s IT1 (as it then was) was inadequate. However, in the EAT, the issue was a pure point of law, namely whether the danger anticipated by section 44 and section 100 was limited to cases of danger to other employees. The EAT rejected that interpretation, and gave two examples of circumstances of danger which might affect people other than employees: a mistake spotted by an employee in a parachute factory; and bald bus tyres spotted by a depot operator.

In determining whether a circumstance of danger exists, it is likely the following factors will need to be taken into account:

·   The national and regional situation will probably be relevant. If, at the time the employee refuses to present for work, Covid-19 is highly reproductive or a large number of people are infected, that will clearly be relevant to whether there exists a circumstance of danger.  

·   The level of risk in the workplace will also be relevant. This will include whether the employer provides proper protective equipment, whether it is possible to socially distance at work, the number of people with whom the employee is likely to come into contact, whether the employee works indoors or outdoors, the opportunities for handwashing, and the type of work which the employee does.

·   Scientific and government guidance will also be relevant. The tribunal will not always be in the best position to determine what factors will give rise to a circumstance of danger on the best understanding of the contemporary science.

· Finally, it might be relevant that the person said to be exposed to danger belongs to a group most severely affected by Covid-19 (and note, again, that that person need not be the employee themselves).

Whether the pandemic presents a “circumstance of danger” or not is therefore unlikely to allow of a general answer. Each case will have to be decided on its facts. The contextual nature of the test, coupled with rapidly changing viral risk and reproduction rates, will quickly make any such determination unreliable as a guide in future cases.

The problem which will face tribunals is that “danger” must be seen in light of the fact that it is likely we will have to live with the virus for a while: at least until there can be widespread vaccination, and even that assumes that vaccine-resistant strains of the virus do not develop. The determination of what amounts to a “circumstance of danger” could be significant in determining how quickly or slowly it is possible for sectors of the economy to return to pre-pandemic operations. The virus will not be exceptional, like defective parachutes or bald bus tyres. It will be omnipresent, and the interpretation of section 44 may end up having to reflect that by adopting a construction of “circumstance of danger” which recognises, and does not encompass, the “new normal”.

(2)   Reasonable Belief

The employee seeking to rely on section 44 will need to show a reasonable belief in the imminence and seriousness of the danger which gives rise to the “circumstance of danger”. Whether an employee holds such a (subjective) belief is a question of fact, whether that belief is reasonable is an objective question to be determined by the tribunal. It is likely that this mixed objective/subjective test will be similar or the same to the approach adopted in protected disclosure claims: see Babula v Waltham Forest College [2007] EWCA Civ 174 §82. Accordingly, the task of the tribunal will be to determine first whether, as a matter of fact, the employee believed the danger to be serious and imminent; and second whether that belief was reasonable.

When the tribunal goes on to consider whether the employee’s belief is (objectively) reasonable, it will have to have regard to all the circumstances of the case, including the factors set out above in relation to the “circumstance of danger” test. In the context of the pandemic, employers are unlikely to be able to rely on the tribunal simply accepting that a reasonable person would think any particular workplace safe, and specific evidence will accordingly be required. Against that background, the following factors will be particularly relevant in determining the reasonableness of the employee’s belief:

·  The current governmental and official scientific advice;

·   The reporting of the crisis, and commentary on that advice, in the media, including possibly social media; and,

·    Any attempt which the employer has made to communicate the ways in which it has made the workplace safe. 

The employer’s belief is technically irrelevant to the application of section 44 (see Oudahar v Esporta Group UKEAT/0566/10/DA §27). It therefore will not be sufficient for the employer to demonstrate that it has satisfied itself of the safety of the workplace.

(a)   Imminence

It is necessary that the employee reasonably believe that the danger is “imminent”. The ordinary meaning of that word would suggest that there is a requirement of proximity in time (and possibly space) between risk and eventuation, and that merely hypothetical risks are not sufficient: see for example Purnell v Ashdown Medway Accommodation Trust Case No. 2300001/2018 §19-20, where the alleged danger was the possibility of injury (by means unarticulated) from confiscated weapons stored in a room in a homeless shelter.[1]  

The most difficult issue here is whether, and if so under what circumstances, an employee can argue that they reasonably believed that presenting at work imposed an imminent risk to other, unidentified people down the chain of transmission.

On the one hand, it might be said that it is in the nature of a highly reproductive virus that the danger which it presents is not just to the person who might catch it or to those to whom they might directly pass it on. The danger is also exponential, in that preventing one transmission can potentially prevent an exponentially larger number of transmissions in the future. It is simply, it might be said, a fact of the pandemic that merely being in contact with other people is imminently dangerous, and Parliament has allocated the risks associated with that by giving employees a right to stay away from work where there is such danger.    

On the other hand, there will need to be an element of reality in the way tribunals approach section 44 in the pandemic. Drastic effects – not anticipated by Parliament when it passed section 44 – might follow from a finding that section 44 gave employees a right to refuse to attend work on the sole ground that they did not want to risk spreading Covid-19 in society in the abstract. As above, it seems likely that tribunals will be driven, at least to an extent, by the need to adjust to a new normal, where the risk of Covid-19 transmission is part of everyday life. For that reason, it is likely that tribunals will use imminence as a limiting principle, and interpret it relatively strictly.

An employee seeking to rely on the protection of others will therefore need to point to something more than the mere continued existence of Covid-19. Examples might include the nature of the work being one making onward transmission particularly likely, or particular risks of transmission to particularly vulnerable people because of the employee’s home situation. Employers would do well to take such steps as they can to ensure that risk of onward transmission from the workplace is as low as possible, both because that is the right thing to do, and also because it will weaken employee arguments that the risk to other people is particularly imminent.     

(b)   Seriousness

The question will then arise whether the employee reasonably believed the risk of catching or passing on Covid-19 to be a “serious” danger. 

The requirement that the employee reasonably believes the danger to be serious has the effect of creating three categories of danger. The first kind is danger or risk which does not rise to the level of a “circumstance of danger”. The second is a “circumstance of danger”, of a sort which could not reasonably be thought to be serious. The third is danger which could reasonably be believed to be serious, and which the employee can show they reasonably believed to be serious. It is only in the third case that section 44 provides protection. It is not therefore enough for an employee merely to point out that danger exists, even danger which rises above the level of the minimal or fanciful (see above). They must show that they reasonably believed the danger to be “serious”.

The question then becomes what it means for Covid-19 risk to be “serious”. On the one hand, there is the simple point that for some groups particularly, catching Covid-19 is extremely dangerous, and could well be fatal. If that line of argument were accepted, it could be said that the starting point would always be that the danger of an unidentified vulnerable person down the chain of transmission catching Covid-19 is always serious, and thus the seriousness requirement will be easy to meet. However, on the other hand, it might be said that, in light of the above, in this context there is significant overlap between the “seriousness” and “imminence” requirements, since arguably, in relation to a transmissible disease, each affects the other. It is difficult to assess the seriousness of the risk associated with Covid-19 without considering “imminence” or risk or onward transmission.  It may be that there is only a certain extent to which these questions can be disaggregated. It therefore seems likely that the question will effectively become one of balancing the prevalence and virulence and lethality of Covid-19 in the community, the inherent transmission risk in the particular working environment, and the effectiveness of protection which the employer is able to offer.

(3)    “Reasonably Expected to Avert”

An employee may only rely on section 44(1)(d) if they can show that they could not reasonably be expected to avert the danger in question. This will be an assessment in the round, which takes into account all the circumstances of the case.

How this limiting principle is interpreted will have a significant impact on the circumstances within which an employee will be able to rely on section 44 in the context of the pandemic. It is easy to imagine steps which an employee might take to lower the risk of catching and transmitting the virus: they might, for example, cycle into work rather than take crowded public transport (see Edwards v Secretary of State for Justice UKEAT/0123/14/DM, dealing with refusal to take dangerous transport), take personal responsibility for ensuring social distancing at work, or bring in their own facemask from home. 

The starting point in applying this provision is likely to be that it is for the employer to provide a safe working environment (since it is the employer who, at the end of the day, is the one with control over the working environment, and the one balancing risk to employee health against revenue and profit), but it is for employees to make individually sensible decisions about how they conduct themselves at work. Tribunals are unlikely to look favourably on employers who do not act proactively and collaboratively with employees and take responsibility for making their workplace as safe as possible during the pandemic. Equally, employees would be well advised to respond to such entreaties in an equally collaborative fashion, and to ensure that they are doing all that could reasonably be expected of them on a personal level to reduce transmission. They otherwise will run the risk that a tribunal might find there to exist a circumstance of danger, but one which could (perhaps) be avoided or reduced if the employee wore a mask, or took similar personal precautions. What can reasonably be expected of the employee will of course depend on the facts of the particular case.    

(4)   Duration of Protection

The protection from detriment under section 44 lasts “while the danger persist[s]”. This also raises very difficult questions. It was clearly drafted with situations such as poor weather or other transient phenomena in mind (see e.g. Balfour Kilpatrick v Acheson EAT/1412/01/TC, in which construction workers refused to work because of health and safety risks presented by heavy rain). The problem with Covid-19 in this regard is that, at least until effective treatment or widespread vaccination, it will be a persistent danger, subject only to the reproduction rate and the number of people (estimated to be) currently infected.

Further, what is required is only that “danger” persists. There does not appear to be a requirement that the employee’s reasonable belief persists. This creates an additional problem for employers, because it suggests that once section 44 is triggered, the conditions for its continued application are more lenient than the conditions which trigger the protection in the first place.

If this is correct, employers will be faced with the prospect that once section 44 is engaged, the employee’s right to refuse to present at work during the pandemic will last for so long as there is a circumstance of danger which rises above the minimal or fanciful (see above). It will be important for employers, if faced with the possibility of an employee showing that section 44(1)(d) has begun to apply, to emphasise the changes which have been made in the workplace to bring down the risk of transmission in order to show (so far as possible) that the danger does not persist. Otherwise, employers could be facing extended (and fully paid) periods of employee absence. The Coronavirus Job Retention Scheme (“CJRS”) might provide some relief from this dilemma, for so long as it lasts, but it is an open question whether the political will to sustain the scheme will remain for so long as the virus circulates in the community.

(5)   Proving Reason for Detriment/Dismissal

Finally, the employee will have to show that the protected conduct was the reason for the imposition of the detriment. This will raise difficult questions about whether the reason for the imposition of an alleged detriment is the refusal to turn up for work, or the exigent circumstances created by the virus. The most obvious example is in relation to furloughing under the CJRS: is it permissible to furlough an employee for exercising the section 44(1)(d) right to stay away? It would perhaps seem odd if not, as the purpose of furloughing is to permit people to stay safely away from work, but such a decision might amount to the imposition of a detriment for refusing to present for work in a dangerous environment. Both employers and claimant employees have a difficult tightrope to walk in running their business and making their claims, respectively.

However, employers can take succour from cases such as Sinclair v Trackwork Ltd (Case no. 1801033/2019). In that case, the tribunal held that an employee with health and safety responsibilities was dismissed, not because he raised health and safety concerns, but because his colleagues objected to him doing so, and so he failed to integrate with the workforce: §§46-52. This is a generous approach to employers. Mr Sinclair was, in a real sense, dismissed because of his health and safety work, which was the very role he had been hired to perform. However, the Tribunal took a more formalistic approach, disregarding the connection between Mr Sinclair’s health and safety work and his failure to integrate. If this approach were replicated in decisions about health and safety during the pandemic, employers could be much more confident in showing that the true reason for a given action was the exigent circumstances of the virus, rather than retaliation against an employee for the exercise of the section 44 right.           

(6)   Beneficiaries of Protection

It should finally be noted that, in domestic law, the right protected by section 44 extends only to employees and not to limb (b) or agency workers: Smith v Carillion [2015] EWCA Civ 209 §17, §20. While EU law applied in the UK, the same was not true as a matter of EU law: in R (Independent Workers Union of Great Britain) v Secretary of State for Work and Pensions [2020] EWHC 3050 (Admin), the High Court held that the failure to extend the protection of section 44 to limb (b) workers was contrary to the UK’s obligations under EU law, particularly Directive 89/391/EC, and made a declaration to that effect.

The question appears to be open what the effect of this declaration will be in the absence of legislative intervention, and so the practical position for limb (b) workers is somewhat unclear. Under EU law, an unimplemented directive cannot be relied upon in litigation between private parties directly (and directives do not amount to retained EU law), and so the argument will now arise whether the duty of consistent interpretation, as retained, can achieve such a result through the interpretation of section 44. It should be noted that, if not, the lack of protection for limb (b) workers would be an especially egregious problem if transpired that, statistically, socio-economic groups more vulnerable to coronavirus are also relatively more likely to have limb (b) worker status, or be agency or casual workers.

E.    COMMENT: SECTION 44 IN CONTEXT

Section 44 sits among other protections for health and safety at work. In particular, it is an implied term of the contract of employment that the working environment be safe. Alongside this are a host of specific statutory obligations. In this context, section 44 forms an emergency provision – it is a blunt tool to protect employees where all else fails. It is not the first line of defence, which lies in sector-specific government guidance and the Health and Safety Executive’s occupational health enforcement.

This is especially so because section 44 confers no positive right to particular conduct from an employer, merely a protection from detriment. In Lewis v Dow Silicones (Case no. 1600680/2018), the ET said at §61:

“The claimant also complains that he was subject to a detriment as the issues he raised were not addressed. The Tribunal has doubts as to whether that can properly amount to a detriment. Section 44 gives the right not to be subject to a detriment by an employer on the ground that the employee has raised qualifying health and safety concerns. It is, in effect, a type of victimisation claim. It does not provide an entitlement to have those health and safety concerns investigated or personally responded to unless it could be said that the failure to do those things was to deliberately victimise the employee because the complaint was made. The Tribunal found no evidence to suggest the respondent had engaged in a practice of deliberately not responding to or answering or dealing with the claimant’s health and safety concerns as a means to cause him harm because he had raised them.” [Emphasis added.]

Further, protection for health and safety at work through section 44(1)(d) requires the employee to take a positive decision not to present at work, and possibly run the risk of detriment or dismissal, in a context where the application of the law in the pandemic is not clear and legal aid is rarely available. It is not realistic to think this risk is one which most employees can consciously afford to choose to run, especially given the socio-economic divide in the availability of home working. Section 44 therefore does not provide a firm foundation for the predictable enforcement of workplace safety in the pandemic, to the detriment of both employees, who are exposed to risks to their health and livelihoods by this uncertainty, and employers who need to know what their obligations are. This is yet another reason why employers (and employees) should act in a collaborative and understanding way: the retrospective, claim-based enforcement of workplace safety rights anticipated by section 44 is simply not an efficient or particularly effective way of protecting those rights.   

F.     CONCLUSION

Section 44 is a challenging provision to apply in the context of Covid-19. It is also one with possibly far-reaching effects. The way in which section 44 is interpreted in the context of the pandemic will determine who is responsible for bearing certain economic costs of Covid-19. A broad interpretation of the section 44 right shifts costs onto employers, a narrow one onto employees. In any event, employers would be well advised to communicate clearly and regularly with their employees about their pandemic preparedness and health and safety at work, and to engage constructively and collaboratively with them in dealing with concerns about health and safety. Employers will need to be extremely careful when making decisions about how to operate safely in the pandemic, to ensure that they do not breach the rights granted to their employees by section 44.

---

[1] It should be noted that Employment Tribunal decisions do not have binding precedential status. However, in circumstances where there is limited appellate authority, they provide useful examples of how section 44 is applied in practice, and it is in that spirit that I refer to them.