Mental health and wellbeing at work in the UK: current legal approaches

Abstract In this paper we outline and critique legal approaches to poor mental health at work in the UK. We argue that the current legal framework is not ‘fit for purpose’. Overall, the existing framework promotes a problematic model that is ineffective because each element, individually and as part of the whole, fails to adequately engage with the nuanced realities of the relationship between undertaking paid work and suffering poor mental health. It is, we suggest, disjointed because it has evolved from a patchwork of provisions, each with different foundations, motivations, ambitions and flaws. The need for a re-focus, and what this might entail, is considered, and the capacity of a model centred on addressing workplace mental health as a manifestation of broader notions of vulnerability is explored.


Introduction
Experience of poor mental healtha category which includes, most commonly, anxiety or depression, but also encompasses severe psychosis, addiction, personality disorders, obsessive compulsive disorders, eating disorders, and bipolar disorders 1has, in the past, been a taboo subject. 2 That legal and social engagement with mental health and wellbeing remains problematic is evident, for example, in the fact that, globally, around 700,000 people die by suicide each year 3 and more than 970 million people are suffering from some form of mental health condition at any given time. 4 There are also worrying signs that mental health conditions are becoming more prevalent, with estimates suggesting a global increase of 13.5% in mental health conditions between 2007-17, so that there are now around a billion instances per year. 5 During the Covid-19 pandemic there have been further increases in the numbers experiencing poor mental health: adults showing signs of depression doubled 6 and probable mental health disorders amongst children increased. 7 The impact on some has been particularly intense: for example, mental health concerns amongst NHS staff quadrupled during the first wave of the pandemic, 8 nine out of ten working mothers experienced greater stress and anxiety, 9 and the increase in poor mental health has been particularly pronounced amongst working parents in poorer families. 10 It is also important to recognise the impact of years of underinvestment and (since 2010) austerity-related cuts on mental health services and welfare provision. 11 Austerity policies have had a devastating impact on already struggling services and users; 60% of Local Authorities, when forced to reduce costs, made severe cuts to mental health provision. 12 Waiting lists for Child and Adolescent Mental Health Services (CAMHS) have lengthened dramatically, so that only the most acute cases can access services, 13 and adult services have also felt the pressure as the NHS has been forced to reduce its investment in mental health provision. The Covid-19 pandemic has added another, deeply problematic and enduring, layer of complexity and challenge for the UK's already-overstretched mental health provision, 14 prompting the Government to launch a significant program of investment in post-pandemic mental health recovery. 15 These realities impact on workplace relationships ensuring that mental health remains 'one of the foremost challenges facing workers and employers in the contemporary labour market'. 16 Poor mental health is correlated with above-average unemployment rates, as many who are experiencing such difficulties struggle to access and retain jobs. 17 The current economic climate has had a significant impact on the labour market, with redundancies, unemployment, in-work poverty, and job insecurity all increasing markedly and impacting on the mental health of adults of working age. Recent research has found that 20% of unemployed adults in the UK are experiencing suicidal thoughts and feelings, 34% of adults in the UK in full-time work are concerned about the potential of losing their jobs, and one third of all adults are worried about their household finances. 18 There is a clear link between mental wellbeing and workplace productivity: even pre-Covid, stress, anxiety and depression were the largest cause of absenteeism in Britain, accounting for 54% of workplace absences in 2018-19. 19 Pre-Covid it was estimated that 300,000 people annually left paid  15 HM Government 'COVID-19 mental health and wellbeing recovery action plan' (March 2021) available at https://assets. publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/973936/covid-19-mental-health-andwellbeing-recovery-action-plan.pdf. 16 M Bell 'Mental health at work and the duty to make reasonable adjustments' (2015) 44 Industrial Law Journal 194 at 194. 17 Thornicroft, above n 2. 18 Mental Health Foundation 'Coronavirus: mental health in the pandemic' (2021) available at https://www.mentalhealth. org.uk/news/more-third-uk-adults-full-time-work-are-worried-about-losing-their-jobs. 19 HSE 'Health and safety at work: statistics for Great Britain' (2020) available at https://www.hse.gov.uk/statistics/overall/ hssh1920.pdf. employment because of long-term mental health conditions, 20 and the impacts of poor mental health were costing businesses £33-£42 billion every year. 21 In 2017, it was estimated that lost taxation, benefit provision, and healthcare delivery cost the government around £27 billion a year and the cost to the economy as a whole, in terms of lost output, was estimated at between £74 and £99 billion. 22 The lived realities of Covid-19 and its impact on mental health can only have aggravated the situation.
Against this backdrop, this paper offers a timely reflection on the legal and regulatory approach taken in the UK to the relationship between mental health and engagement in paid work. Sections 1, 2 and 3 will briefly explain the scope, nature, and limitations of three distinct limbs of the UK's existing legal framework. That there is no unified legal approach to mental health and wellbeing at work reflects the fact that, for most of our recent history, this issue received no real legal attention, and was seen instead as due simply to individual susceptibility. This changed only recently, as manufacturing industries began to be replaced by a more service-based economy, 23 and the three limbs of legal intervention outlined here loosely reflect three phases of this development. Legal remedies were first sought only when there was substantial evidence of potential psychiatric harm having been caused by an employer's actions or inaction. Here the laws of tort and contract (section 1) provided the obvious routes as they reflected the common law's attempts, however patchy and flawed, to better regulate relationships, including those of employers and workers. Whilst providing important generic standards for employers, the nuanced realities of working life have increasingly tested the importance of this overarching function of tort law, and have exposed its flaws as a useful mechanism, in practice, for regulating mental health wellbeing concerns. Over time, equality laws relating to individuals with disabilities developed in the UK (section 2) and this paradigm was seen as offering a new hope: a framework emerged based on the need to treat people disabled by psychological conditions equally to non-disabled people and, by implication, to compensate where that was not feasible because of the difference that the disability caused. Finally, as our understanding of the 'complex and multifaceted' 24 relationship between wellbeing, mental health and paid employment has grown, broader workplace practices and standards (section 3) have developed, influenced in part by the EU, to address issues of mental ill-health in a more preventative, and less individual-oriented, way. Our analysis of the three limbs of the UK's current legal framework reveals a tendency towards a restrictive individualism that undermines the potential to better support the health and wellbeing of workers and the need to implement a perspective that more fully 'emphasise[s] the importance of social factors influencing health and illness'. 25 A detailed discussion of tangible recommendations and practical solutions is beyond the parameters of this paper, 26 which focuses instead on providing an account of, and reflecting upon the disjointed and ineffective nature of, law's approach to mental health and wellbeing at work. In the concluding section we do, however, argue that progress is possible only if the restrictive individualisation of existing approaches to the mental health and wellbeing of workers is challenged. There has been ample engagement with how adherence to the social model, as opposed to the medical model, can benefit existing legal frameworks, especially in relation to equality/disability. 27 However, 20 Lord Stevenson and S Farmer 'Thriving at work: a review of mental health and employers' (2017) Annex C, available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/658145/thriving-at-workstevenson-farmer-review.pdf. 21 Deloitte 'Mental health and wellbeing in employment: a supporting study for the independent review (2017) available at https://www2.deloitte.com/uk/en/pages/public-sector/articles/mental-health-employers-review.html. given the broad scope of this piece and the flaws that are identified, we suggest that a more significant shift in our embedded perspective is required. We encourage, drawing on vulnerability theory, 28 greater exploration of the legal and policy consequences of viewing mental health and wellbeing as an ongoing and changing expression of the human condition.

Negligence and contract (a) Negligence claims
A duty of care has long been implied into employment relationships but it was the landmark case of Walker 29 which first introduced the notion that an employer might be liable in tort for psychiatric harm. General guidance for courts in terms of how to approach tort cases of this nature was provided some years later in Hatton v Sutherland, 30 which was refined and confirmed by the House of Lords in Barber v Somerset County Council 31 and subsequently. 32 The threshold for proving that the duty of care has been breached is high and requires any harm, physical or psychiatric, to have been 'reasonably foreseeable': 'the threshold question is whether this kind of harm to this particular employee was reasonably foreseeable'. 33 Psychiatric injury caused by such negligence is often found by courts to be 'too remote' to elicit compensation. In Walker 34 for example the employee, a manager of a social work area office with a very heavy case load of child abuse cases, experienced two nervous breakdowns but the employer was only held to have breached the duty of care in relation to the second. The High Court found that the first breakdown was not foreseeable despite the complaints he had raised about his excessive workload and the fact that he could not cope with the volume of work. The case was only successful because the court found that the Council failed, once aware of Mr Walker's situation, to then engage in a way that might have prevented the second breakdown.
Walker established that a clear awareness of the difficulties (there, a breakdown inducing acute anxiety, mental exhaustion, irritability, headaches and insomnia) and its impact on the employee is key to establishing foreseeability. 35 Regulators, policymakers, and the press had begun to voice concerns about levels of stress in British workplaces at this time, and the decision in Walker reflected an aspiration to resolve this by rebalancing the rights of employers and employees, extending existing protections to new types of harmalbeit in a form circumscribed by established notions of legal duty. 36 As such, even once a breach is established, courts are required to consider what was reasonable in the circumstance and a wide number of issues are relevant; 'the size and scope of its operation will be relevant to this, as will its resources, whether in the public or private sector, and the other demands placed upon it. Among those other demands are the interests of other employees in the workplace'. 37  addition to reasonable foreseeability, the claimant must be able to show that a particular breach of duty caused, or at least significantly contributed to, the psychiatric harm that s/he experienced. Research has indicated a reluctance amongst lawyers to support tort claims involving mental health where there was any evidence that the claimant had experienced personal difficulties such as a divorce or bereavement, 38 which would presumably have hindered any attempt at proving causation and would impact decisions to quantify damages. 39 Whilst the role of tort law as standard setter, and the broadly historical significance of this, ought not to be undervalued, over time that important contribution has been diluted by the significant flaws inherent in this prong of the legal framework. Indeed, the effectiveness of tort law at supporting those with poor mental health who are in employment or want to access paid work is clearly limited. First, it is reactive: tort law compensates financially for harms already experienced and provides no real incentive for employers to meaningfully engage with issues of mental health and wellbeing at work more generally, except to avoid liability. Potential causes of poor mental health and factors that may exacerbate existing conditions, such as heavy workloads and long hours, 40 a particularly emotionally stressful role, 41 poor management/leadership 42 or bullying 43 are only explored in retrospect, in order to apportion responsibility for the harm inflicted in the particular case. The duty places no onus on employers to foresee the potential harm that might be caused by these factors, nor does it require employers to generally mitigate them where they exist. The courts also seem unwilling to impose such a duty; in Hatton v Sutherland, where the claimant had argued that her breakdown was dueat least in part to the stress of a heavy workload in a secondary school, the Court of Appeal found that the harm was not reasonably foreseeable and noted that whilst 'all employers should have had in place systems which would overcome the reluctance of people like Mrs Hatton to reveal their difficulties and seek help… it is not for this court to impose such a duty upon all employers, or even upon all employers in a particular profession'. 44 Yet such a duty could easily be introduced, if the will was there: it might help in certain cases, for example, to require staff to be screened to determine their ability to cope at any given time or require adequate training or regular monitoring of staff workloads and wellbeing. 45 Secondly, tort is an individual-focused legal remedy that centres on the need to establish a culpable failure by an identified employer or colleague. It does not ask whether a workplace or sector is, as a whole, asking too much of its staff or failing to build resilience within its workforce. Moreover, an action in tort requires the person who has suffered the harm to enforce the law by pursuing a private legal action. Given that only 2% of people with mental health problems feel comfortable speaking about this in the workplace, 46 making legal protection reliant upon the exercise of employee voice demonstrates how this prong of the legal framework fails to accommodate the lived realities of the employment relationship. In addition, in order to overcome the hurdle of reasonable foreseeability, a claimant has to effectively make the employer aware of the true extent of their illness and connect 38 J Earnshaw and C Cooper 'Employee stress litigation; the UK experience ' (1994)  Legal Studies the impact of the relevant work system, process, or behaviour to that resultant psychiatric harm. This requires individuals to construct themselves and others as failing: the only narrative possible, if the claim is to succeed, is one where the claimants portray themselves as ostracised victims who would have been able to function adequately but for the (in)actions of their employer. The focus is placed onto the person who has suffered harm to mark themselves out as deficient and weaker than the mythical autonomous and capable worker and show that this lapse from their normal persona as a productive worker was the fault of the employer. Thirdly, tort law can disincentivise employing those whose mental health is known to be poor or who are perceived, rightly or wrongly, as being susceptible to experiencing such challenges; 47 such knowledge triggers a duty of care and requires that that relationship be treated with extra sensitivity. There is no doubt that, as Hale remarked, 'the law of tort has an important function in setting standards for employers' but, as she also commented, '… if the standard of care expected of employers is set too high, or the threshold of liability too low, there may also be unforeseen and unwelcome effects upon the employment market'. 48

(b) Contract
The duty of care, discussed above, is implied into contracts of employment; contract law can also offer a retrospective remedy when harm has occurred following breaches of other key terms of the contract, most likely the implied duty of mutual trust and confidence (hereafter 'the duty') 49 . In the leading decision of Malik v BCCI, the House of Lords set out the duty as one whereby the employer is not to 'without reasonable and proper cause, conduct itself in a manner calculated and likely to destroy or seriously damage the relationship of confidence and trust between employer and employee'. 50 Employee claims based upon a breach of this term became more common during the 1970s, following the introduction of statutory protections against unfair dismissal via the Industrial Relations Act 1971 and the subsequent confirmation in Sutcliffe v Hawker Siddeley Aviation 51 that constructive forms of dismissal fell within the scope of that Act, and a wide range of behaviours have since been recognised as capable of breaching this duty. 52 They include a failure to provide or follow a grievance 53 or disciplinary procedure, 54 suspending an employee when it is not warranted, 55 or suggesting without evidence that someone is incapable of doing their job, 56 is unsuitable for promotion, 57 or is dishonest. 58 They have also included cases of bullying 59 and overwork 60 and where an employer's fraudulent behaviour has damaged the employee's reputation because s/he is then associated with that employer/behaviour. 61 Where these behaviours have caused psychiatric illness and that connection 47 A Bogg and S Green 'Depression in the workplace: an employment law response' in C Foster and J Herring (eds) Depression: Law and Ethics (Oxford: Oxford University Press, 2017) p 253. 48 Hatton v Sutherland [2002] 2 All ER 1 at para 14. 49 For a general discussion of, and critique of academic views around, the potential scope and limitations of the duty see D Cabrelli 'The implied duty of mutual trust and confidence: an emerging overarching principle' (2005)  can be demonstrated, this offers a route to securing (sometimes relatively large) compensation payouts. 62 The duty, therefore, offers another means of challenging behaviours at work that have led to psychiatric illness.
At a broader level, the importance of the duty, and its evolution to encompass a range of scenarios, has been widely recognisednot least because it has a key role as 'a quintessentially relational norm' 63 in reconstructing employment relationships as something more than a financial exchange. 64 Its significance as a means of supporting positive mental health and wellbeing at work is linked to this overall development: the duty challenges historically embedded constructions of employment relationships as ones of servitude, and suggests an overall acceptance (now backed up with statutory rights and policy ambitions) that employers are no longer at liberty to treat employees with the disregard that was permitted in the past. Despite this, it would be misleading to overstate the duty's potential as a means of changing workplace cultures, as it does not impose a requirement to act reasonably but rather, one to avoid acting in a manner that is so intolerable that it will irrevocably harm the employment relationship. As with the duty of care, its significance manifests only once the employment relationship is damaged, usually irreparably, and compensates financially for damage/harm that is the fault of the individual employerbecause s/he acted in a way that breached the duty of trust and confidence.
In sum, while common law provisions have been applied to relationships of employment, their very nature has meant that this application is awkward. Their main focus is, as Barrett noted, to distinguish 'between stress that is caused by work and stress that manifests itself at the workplace but may be caused either by other aspects of the victim's life… or the peculiar sensitivity of the victim to normal working life', 65 determining where, within a narrow range of possible answers, blame should lie for the mental health issue in question: with the employer, the individual, or non-work related elements of the individual's life. Actions in tort or contract cannot adequately engage with the realities of mental health and wellbeing at work because their scope and nature is inherently limitedthe legal framework has developed in an ad hoc fashion and is an awkward fit for employment relationships strained by mental health issues. It was not created with these issues in mind but rather evolved to fill gaps left in the legal framework by the limitations of alternative protections, including the second and third prongs of the legal framework, which are discussed below.

Equality laws
Perhaps the most obvious source of legal protection for workers with poor mental ill-health is the legal framework governing equality and non-discrimination. Substantive protection from discrimination on grounds of disability in the workplace has existed since the Disability Discrimination Act 1995 and now stems primarily from the Equality Act 2010. On its face, the legislation provides far-reaching protection: in addition to the protections from direct and indirect discrimination, discrimination arising from disability, victimisation and harassment are covered by the Act. Indeed, given the relationship between harassment and bullying and poor mental health, it is important to note that the Act protects against harassment across all protected characteristics (although, as discussed below, the complexity of the legal regulation of workplace harassment and bullying, involving a mixture of civil and criminal sanctions, brings its own difficulties).
In addition to these protections, employers are under a duty to make individualised reasonable adjustments for workers who they know (or should know) have disabilities, in order to remove disadvantage and barriers to inclusion. In its inclusion of the duty to make reasonable adjustments, in particular, the protective framework recognises both the paucity of formal equality approaches to disability equality and the individual nature and experience of disability, thus rendering standard approaches to countering group disadvantage and exercises in comparison more complex than for other protected characteristics.
This section will outline three key concerns about the ability of the existing provisions to provide comprehensive protection for workers with mental illness. First the scope of protection offered by the legislation; secondly, difficulties in the application and implementation of the duty to make reasonable adjustments; and, finally, the onus on the individual worker to enforce the protections the Act provides. The flaws highlighted with this limb of the current legal framework echo the concerns raised throughout this paper.

(a) Scope of protection
On the one hand, the personal scope of protection afforded by the Equality Act is broader than some of the other protective statutory and contractual rights discussed here. The Equality Act covers not only those working under a contract of employment but also the much broader category of working relationships comprising those under a 'contract personally to do work'. 66 Importantly, it also protects job applicants 67 and indeed prohibits employers from asking questions of applicants about their health or disability. 68 On the other hand, the protection from disability discrimination in the workplace is only available to the subset of those with mental ill-health who meet the definition of disability. Under section 6 of the Act an individual has a disability if they have a physical or mental impairment which has a substantial and long-term adverse effect on their ability to carry out normal day-to-day activities. This has proved a high and often elusive threshold 69 and reflects a largely medical model of disability where the disability is located in the impairment. As Lawson argues, it also appears to reflect an approach which, rather than focusing on the behaviour of the employer, regards protection from discrimination (and access to reasonable adjustments) as akin to a welfare benefit and the test for disability as the 'gatekeeper'. 70 It contrasts with the much broader social and rights-based approach found in the Convention on the Rights of Persons with Disabilities (CRPD), which sees disability as arising from the interaction between impairment and attitudinal and environmental barriers. 71 While it has frequently proved a challenging test to satisfy in relation to any disability, the definition creates some particular difficulties in relation to mental ill-health. Evidence presented in 2016 to a House of Lords Select Committee enquiry into the impact of the Equality Act on disabled people suggested that those with mental health problems were less likely to be regarded as disabled and be able to access the Act's protections, including access to reasonable adjustments. 72 Given the stigma that still attaches to mental health conditions, this is unsurprising. 73 Individual workers have an incentive to conceal or to understate the severity or long-term nature of their condition in order to minimise the likelihood of discrimination. 74 Even where this is not the case, the fluctuating and episodic nature of some mental health conditions can give rise to particular challenges in proving that the elements of the test are satisfied. 75 In retaining this threshold test for access to protection, the Equality Act treats obligations relating to mental ill-health largely as a response to something that exists independently of the workplace rather than something which may be created or exacerbated or, sometimes, improved by it. While the Act acknowledges (via the obligation to make reasonable adjustments) the existence of barriers in the workplace which may make equal participation more difficult for those with an existing disability, it fails to acknowledge the role of the workplace in creating or impacting upon an existing disability. There are, for example, no obligations under the Equality Act to arrange work so that a worker with anxiety which is not yet severe enough to meet the threshold condition for disability does not become disabled by that anxiety in interactions with the workplace. It is only once the anxiety has worsened, to a degree that meets the threshold, that the obligations arise.

(b) Reasonable adjustments
Employees who meet the test for disability under the Act are entitled to protection from discrimination under it, including the right to reasonable adjustments. Employers are under a duty to make reasonable adjustments when a provision, criterion or practice in the workplace puts a disabled worker at substantial disadvantage compared to persons who are not disabled. The obligation is to take reasonable steps to avoid that disadvantage, but this duty only applies where the employer knows, or has constructive knowledge, of the worker's disability. It is an individualised approach to removing disadvantage in the workplace which ideally involves worker and employer in a collaborative effort to identify barriers and develop creative solutions to dismantle them. The duty is generally seen as a transformative legal tool which has significant potential to make workplaces more inclusive, 76 and there have been instances where courts have taken a very expansive approach to interpretation of the duty, including cases relating to mental health. Thus, for example, in Croft Vets the Employment Appeal Tribunal held that the employer's duty extended to funding psychological therapy to help the return to work of an employee who had been absent with depression and anxiety triggered by work-related stress. 77 However, Bell's recent review of case law on reasonable adjustments and mental health found inconsistencies in the approach of tribunals to interpreting the obligations of employers, and a tendency to take a narrow and literal interpretation of the duty. As an example, tribunals were generally reluctant to extend sick pay as a reasonable adjustment (where sickness absence is disability related), reflecting an 'evident concern to avoid imposing burdens on employers'. 78 A second concern relates to the gap between law and practice, even among well-intentioned employers. In evidence given to the Parliamentary enquiry into the effectiveness of the Equality Act in relation to disability discrimination, the charity Mind suggested that there is an 'obvious lack of confidence and understanding about what an adjustment could look like for someone living with a mental health problem'. 79 Bell argues that although research suggests some improvement in employers' familiarity with making reasonable adjustments in relation to mental health, important challenges remain for employers in imagining and implementing appropriate adjustmentsnot least a stigma which may make employees reluctant to engage in candid conversations in this regard. 80 There is good reason to question, therefore, what further steps can be taken to ensure the legal framework delivers on its potential to effect real inclusion in the workplace.

(c) Enforcement
A third concern about the protection afforded by the Equality Act relates to enforcement. Realising the rights to equality and non-discrimination contained in the Act depends on the willingness of individuals who have faced discrimination in the workplace to bring claims to the employment tribunal. However, the ongoing stigma attached to mental health issues, noted above, is again likely to act as one significant barrier (among others) to the effectiveness of this approach. Fear of the consequences of disclosing mental ill-health means that 'a reactive approach to mental health that relies on individuals asserting their rights is unlikely to be sufficient'. 81 Even without a fear of disclosure, there are likely to be other barriers to initiating tribunal claims for individuals struggling with their mental health, particularly given the short (three month) timeframe for bringing a claim. 82 While evidence of ill health may lead a tribunal to extend this deadline where it is 'just and equitable' to do so, 83 persuading a tribunal to do this will depend on the ability of the claimant to adduce evidence of the impact of mental health difficulties at the relevant time, something which the nature of some mental health conditions may make difficult. In Castell v Society of Motor Manufacturers and Traders Ltd 84 the Employment Tribunal was asked to admit a late claim for disability discrimination where the claimant argued that his mental health had prevented him from bringing a claim in time. While not the only factor in its refusal to extend the deadline, the Tribunal noted that there was no evidence of the claimant's mental health in the months immediately following his dismissal, until he deteriorated and was sectioned nine months later. While accepting that the claimant may not have approached his GP in the early months because 'those with mental health issues often do not want to admit, even to themselves, that they have a problem', 85 the tribunal nonetheless stated that it would have expected some independent evidence of his condition during this initial period, when the claim should have been made.
Two key, but limited, obligations under the Act require employers to take proactive, rather than simply reactive, steps to make the workplace more inclusive. First, the protection from indirect discrimination requires employers to adopt only justifiable policies and practices which may disadvantage those with a disability as a group. Because this obligation arises even where an employer does not have actual or constructive knowledge of an individual's disability, it appears to offer a way round the difficulties presented by the stigma that may prevent some individuals from claiming their rights. However, it is notoriously difficult to apply in the context of disability discrimination because the individual nature of disability means that it can be difficult to establish the group disadvantage. 86 Secondly, under section 149 of the Act, public sector employers are bound by the Public Sector Equality Duty (PSED) which requires them to have 'due regard' to the need to eliminate discrimination and advance equality of opportunity, among other things. While the duty has led to important success in challenging decisions and in organisational change 87 it remains limited in that, even for the subset of employers it binds, it does not impose an obligation to achieve a particular outcome. As the Court of Appeal noted, the PSED 'is not concerned with the lawfulness or even the adequacy of the solution that was adopted. It is only concerned with the lawfulness of the process'. 88 Shifting the burden of ensuring compliance with the Act's requirements away from individual workers will therefore require a rethink of the shape and role of these positive obligations. Achieving such a shift should result in an approach that accords more closely with the social and rights-based model of the CRPD which demands proactivity from policymakers in identifying and dismantling structural barriers to participation and inclusion. 89 The Equality Act provisions attempt a more meaningful engagement with disability at work and awkwardly engage with issues of mental health as a sideline to this broader ambition. As discussed above, however, it is limited in its usefulness. It suffers from the same core flaws experienced in the older contract and tort provisions. The onus remains too heavily weighted on the individual to enforce the laws, and it promotes the 'othering' of those with mental health impairments through its high thresholds in relation to eligibility, application and enforcement. While private law systems address questions of workplace mental health by recognising individual rights of action that stem directly from the embodied interests of workers with diagnosed mental health conditions, labour standards approach these issues from the other direction. However, as our assessment of these will demonstrate, they too are fundamentally flawed and undermine the potential of laws to adequately engage with mental health in workplaces.

Labour standards
Rather than giving rise to individually-actionable rights per se, labour standards establish publicly-enforceable requirements that indirectly protect individuals' interests and try to 'accommodate' work with the social citizenship rights of workers 'to live the life of a civilised being according to the standards prevailing in the society'. 90 This moves beyond a purely transactional understanding of the employment relationship to potentially address the needs of workers as human beings and not just as rational economic agents. 91 Contextually, the challenges of achieving this have been exacerbated by the growth of flexible and informalised forms of work (such as zero-hours contracts and the 'gig' economy of platform work) which circumvent established employee-protective norms and impact on employees in terms of lower job quality, increased precariousness of employment, and increased experience of associated mental health stressors. 92 In this section we outline the key 'facilitative' measures (working time and pay provisions) that indirectly, and unintentionally, lay the groundwork for good mental health by limiting individual workers' exposure to workplace demands via reducing the number of hours worked or level of pay provided; and 'direct' measures that impose enforceable obligations to ensure healthy working environments, including in relation to mental health.

(a) Working time provisions
While the UK's history of regulating working time goes back to the Factory Acts of the nineteenth century, 93 modern provisions in this area stem primarily from the European Working Time Directive, 94 which provides rights to workers in relation to maximum hours of work, work patterns, rest breaks, night work, and the duration of certain types of monotonous work. The Directive constitutes perhaps the most 'social policy'-oriented element of EU health and safety provision, 95 recognising the contribution to poor mental health of stress and overwork, and particularly of repetitive work and night work, which place significant psychological demands on employees as well as increasing fatigue and hence the likelihood of accidents. 96  week 97 and minimum daily, weekly, and annual rest periods, 98 something that the UK Government viewed as contrary to its policy of removing 'barriers' to labour market flexibility: 99 indeed, it challenged the Directive on the basis that issues like hours of work were national competencies not covered by Article 118a EEC, which empowers the Community to legislate on workplace health and safety issues. 100 The article was held to define health and safety broadly and so the UK was compelled to implement the Directive, albeit reluctantly and with an opt-out provision, via the Working Time Regulations 1998. 101 The 1998 Regulations exclude certain sectors of employment (such as junior doctors and the emergency services), certain types of workers (such as domestic staff and autonomous executives), and certain job responsibilities (such as ensuring security). 102 They allow for time limits to be set aside via collective agreement or when activity in an industry 'surges' (such as harvest time in agriculture) 103 and for workers to 'opt-out' of the 48-hour limit so long as this is agreed in writing. 104 Enforcement is via a mixture of criminal offences and individual recourse to employment tribunals, with the former backed by Health and Safety Executive (HSE) and Local Authority inspections, 105 a blend which reflects the social citizenship basis of the Directive. The HSE, the principal body responsible for the enforcement of the regulations, appears to have brought only one working time prosecution since the Regulations were introduced. 106

(b) Minimum wages and sick pay
Another indirect means of improving mental health outcomes is by imposing minimum rates of pay and remunerated sick pay for employees. While these measures are not health and safety-related as such, they embed conditions that limit the impact of workplace risk factors on mental health and wellbeing; for example, the mental health of working parents in the UK deteriorated significantly during the Covid-19 pandemic and this is strongly related to financial insecurity. 107 The UK introduced a mandatory enforceable minimum rate of pay via the National Minimum Wage Act 1998; the minimum wage for over-25s has since been reframed as a 'living wage', and is set at a higher level than the minimum wage for under-25s, younger workers, and apprentices. 108 This provision also extends to workers on 'zero-hours' contracts ('time workers'), including non-working periods when they are required to be present and available to work. 109 Non-compliance with the Act can lead to civil sanctions and criminal liability, 110 but ACAS, the conciliation and arbitration service, is the first-instance body handling complaints about breaches of the law. And while HMRC imposed some £17 m in penalties onto employers in around 3,000 cases during 2018-19, involving more than £25 m of underpaid wages for 24,000 workers, 111 prosecutions remain rare, with only 15 ever brought. 112 This perhaps reflects a view of the minimum wage as a matter of corrective labour market economics, rather than of relieving workers of at least some pressures associated with stress-related mental health conditions. Underpayment is most prevalent in the childcare, transport, hospitality, and retail industries; 113 measures such as these thus contribute to the protection of the welfare of workers in some of the most commonly precarious and casualised sectors of the labour market.
In addition to the above, for those unable to work as a result of ill health, the UK introduced a state-funded, employee-administered system of statutory sick pay via the Social Security and Housing Benefits Act 1982, 114 which established a universal but minimal safety net through which employees are compensated by their employer to a set level of remuneration for a time-limited period in which they were unable to work due to illness; 115 any sums or time-periods beyond those limits were left to the individual employment contract. Several problems with this are worth mentioning. The UK's sickness benefit replacement rate (20%) is one of the least generous in Europe, 116 and the scheme does not differentiate between transitory periods of sickness and the longer-term conditions (which may typify mental ill-health), and does no more than advise employers to 'consider' extending sick pay in such circumstances. 117 Enforcement relies upon individuals making a complaint to a HMRC dispute resolution team, and the scheme is of limited relevance in areas of 'gig work' and flexibilised employment, where traditional employer-employee relationships are absent. While it is possible for individuals engaged on zero-hours contracts to be construed as employees, 118 and hence entitled to sick pay and other employment benefits, it remains the case that, for many working people, there is no responsible employer to administer it.
Facilitative measures such as these are thus relevant to a conception of workplace mental ill-health as psychosocial and a product of stressors such as fatigue and burnout, the insecurities of uncertain work, and imbalances between job demand and reward. The potentially positive impacts of these legal developments have proved limited, with potential protections diluted by exemptions and opt-out clauses, difficulties in accommodating atypical employment relationships, a reliance on voluntarism for anything beyond minimum provisions, and limited enforcement. Ultimately, this has placed a significant onus on individuals to initiate legal proceedings, presenting barriers for those challenged by 109 National Minimum Wage Regulations 1999, SI 1999/584, reg 3. 110 National Minimum Wage Act 1998, s 19A (financial penalties) and s 31 (criminal offences). 111 Low Pay Commission 'Non-compliance and enforcement of the national minimum wage' (2020) p 17 available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/885382/Non-complian-ce_and_enforcement_report_-_2020_-_amended.pdf. 112 BEIS 'National living wage and national minimum wage: government evidence on compliance and enforcement 2018/ 19' (February 2020) para 8.10 available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attach-ment_data/file/866766/nlw-nmw-government-evidence-compliance-enforcement-2018-19.pdf. 113 Low Pay Commission, above n 111, p 7. 114 H Dean and P Taylor-Gooby 'Statutory sick pay and the control of sickness absence ' (1990)  value from employees) as opposed to mental health more generally. 129 Methodologically, the Management Standards intervene indirectly into issues of stress by focusing on employee reports of stress and the monitoring put in place in response; the focus is thus on the account of stress, not the underlying root causes. This creates a reliance on employee voice as a mechanism for managing workplace stress (in that staff surveys must highlight a stress problem in order for that problem to exist), but workplace power dynamics can constrain this exercise of voice and make reporting difficult. 130 Data has shown that exposure to workplace mental health stressors remains widespread across the UK: 828,000 workers reported that they suffered from work-related stress, depression or anxiety during 2019-20, and nearly 18 million working days were lost as a result, rates that have increased sharply in recent years. 131 Finally, there has never been a HSWA 1974 prosecution in a mental health-related case, and while improvement notices have been issued in a small number of stress cases, there have been none in the last five years. Commentators have speculated that a stress-related prosecution is only 'a matter of time', 132 but the enduring challenges of establishing a causal link between workplace conditions and employee mental health (due to the diffuseness, long-latency, and invisibility of the latter), and setting out what a criminally culpable failure to manage health and safety might look like (given that stress arises by degree, is hard to control, and requires systemic prevention), mean that the duties of care in this area are not just unenforced, but perhaps unenforceable via the criminal law. 133 Direct regulatory standards depart from the individualisation found in other areas of the law but there are aspects of workplace mental health regulation where that tendency remains. Employee experiences of bullying, harassment, or violence at work have historically been treated as matters of interpersonal wrongdoing, but they are also workplace risks which require effective management by employers. 134 The EU's Framework Agreement on harassment and violence at work requires employers to have policies in place to deal with such behaviour, 135 and the HSE's guidance frames reasonably foreseeable violence and harassment as an enforceable health and safety risk under the HSWA 1974 and the Management Regulations. 136 But this guidance emphasises that non-criminal (equality and discrimination law) and non-labour-standard (Protection from Harassment Act 1997) measures are the preferable routes for enforcement, and the HSE defers to ACAS, the Equality and Human Rights Commission, and (in serious cases) the police to deal with these cases. 137 As Bogg and Freedland argue, the fact that there are both civil and criminal avenues for addressing workplace harassment has a 'dragging' effect, making it harder to bring civil cases that do not appear to meet criminal thresholds of severity or certainty, while also rendering the criminal law as exceptional, reserved for only the most extreme and egregious cases. 138 Individuals are left with the burden of advancing their own protection via the civil law but, to do so, must show that the damage to their personal interests is sufficient to merit protection on their behalf via state enforcement. Concluding thoughtsthe need for an alternative focus We have seen that, across all three prongs of the legal framework, the relevant legal interventions have developed in an ad hoc manner and cumulatively offer only partial coverage, are limited in terms of scope and provision, and are compromised in terms of enforcement. The framework is not fit for purpose. At the heart of all three areas is found an undue reliance on individuals to bear the burden of protecting their own interestsby bringing claims to tribunals or arbitrators, negotiating contractual protections, exercising voice in ways that leave them vulnerable, or using the civil law to claim redress. Public enforcement occurs only rarely because the interests being 'protected' are not framed in terms of the values of social citizenship that underpin them, but instead as components of a commodified model of the employment contract: interests are viewed as transactional goods, and problemsif they ariseare to be resolved via individual rights-mobilisation. The detrimental consequences of this 'responsibilisation' are particularly pronounced in relation to mental health issues, where debilitating illness, stigma, and marginalisation combine to restrict the capacity of individuals to exercise autonomy, particularly where workers lack the job security or collective support needed to exercise their voice. The dynamic of 'responsibilisation' not only shifts the burden of regulating onto the affected worker (and so places barriers in the way of those seeking protection), it renders it a 'rights-defined' issue, positioning those who do not take such steps as complicit in their own suffering and restricting the scope of workplace mental health provision to those elements that are enforceable at law. 139 The embedded flaws we have discussed result in inadequate protections for those experiencing any kind of mental health issue becausewhile existing legal frameworks offer some means of standard setting and securing recompense for damage caused to an individual by particularly poor workplace behavioursthey also tend to play a role in 'othering' claimants by promoting the notion that the 'problem' and 'solution' to issues of mental health lie within the individual. This is an approach that has been challenged as perpetuating a medical model that is outdated, awkwardly sustaining a view that poor mental health is tragic and pathological. 140 Whilst attempts are made to engage more broadly with a more meaningful social model of disability 141 and the role that 'healthy' workplaces might play in improving the mental health of all workers, the mechanisms used remain weak and inadequately enforced.
What can be done to improve our approach? Whilst, as stated in the introduction above, the purpose of this paper is not to provide immediate practical solutions, our assessment of the current legal framework reveals a need for a different ethical perspectivea shift in approach to one that can provide a more ambitious grounding for legal interventions. The current legal framework as a whole perpetuates what Fineman has termed 'the myth of autonomy'; 142 the illusion that independence is attainable for all throughout the life course, regardless of their context, and that self-sufficiency, rationality and competencebeing a 'good' (unencumbered or 'disembodied', physically and mentally healthy, and flexible) worker for the majority of one's adult lifeis the epitome of what it means to be a valuable human being. This positions those afflicted by poor mental health as 'broken' components of the economic system who fall short of this ideal to some degree. Private law may occasionally offer compensation if the causes of this 'breakage' fall within an employer's legal responsibilities, equalities law may require some accommodation of the 'least broken' when feasible, and labour standards may demand proactive steps be taken to avoid such 'breakage', but such patchy mitigations are begrudgingly applied and, as discussed in the introduction, have had relatively little impact in changing the lived realities of those employers, sectors and workers who are challenged by poor mental health. 139 Almond and Gray, above n 130. Fineman's vulnerability theory, and in particular the notion of the vulnerable subject, offers a means of countering this damaging myth of autonomy and provides an alternative position from which to develop supportive policies. Vulnerability theory states that all human beings are constantly and universally vulnerable; it is the human condition, rather than a distinct character trait attaching to those who are perceived as weak, oppressed, marginalised or discriminated against. As such, we are all inescapably susceptible to positive and negative changesfor example in terms of physical and mental health and wellbeingthat may induce dependency of varying degrees, for various reasons, throughout our lives. Such a basic recognition makes evident the tendency of existing legal mechanisms to minimise our universal vulnerability and to stigmatise individuals who need support, to the detriment of the wellbeing of the working population as a whole and hence the economy. By replacing the conception of the 'liberal self' with one of the 'vulnerable self', we can challenge the 'collective, or social, injury that inevitably arises from a state unresponsive to the universal and constant human condition of vulnerability and dependency'. 143 Our societal investment in the myth of autonomy, and its framing of mental health as a personal problem experienced by 'broken' individuals who are no longer productive members of the labour market, perpetuates the awkward, patchy and ad hoc legal responses that we have explored in this paper. Although a powerful force for change, law can also perpetuate outdated foundations and demarcations; vulnerability theory encourages us to reconsider those foundations by acknowledging the fluidity and constant evolution of humans. It reminds us that poor mental health can be experienced by any of us, directly or indirectly, at any time during our life, and so must be effectively absorbed and adequately supported. The core and primary mechanism for doing this is the embedding of responsibility for the pursuit of social justice outcomes at the level of the state, via an increased responsiveness to vulnerability as a systemic issue, the promotion of greater resilience in individuals and institutions, and a move away from the current law's simplistic focus on individual failings and personal responsibility: the state is not neutral 'and cannot be passive, non-interventionist or restrained'. 144 Vulnerability theory forces us to look more critically at the state's contribution to building resilience in individuals and institutions, including workplaces, to our shared vulnerability. This includes vulnerability to poor mental health and wellbeing, and encourages us to question whose interests are privileged by current laws and policies and, crucially, whose are not.
As this paper has shown, employee experience of poor mental health which detrimentally impacts on the employment relationship remains an area where there are gaps in provision which neither law nor policy has adequately addressed. An increasing societal awareness of the consequences of poor mental health has not translated into effective legal protection, and there is a need to develop new labour laws capable of supporting workplaces that are economically productive, sustainable in practice, and psychologically beneficial. Doing so demands that we think beyond the strictures of existing measures and adopt a more holistic and preventive mode of state action centred around a recognition of the universal vulnerability of exposure to working conditions that can imperil mental health and wellbeing. The critique delivered in this paper exposes the fragility of a legal framework that highlights the state's ongoing lack of engagement as a builder of resilience. 145 To move forward we need to identify appropriate values and operationalise institutions and structures in a novel way. This is a long-term project but only once we adopt a more life-course-sensitive, non-judgmental, and holistic view of mental health, and recognise state responsibility in resilience-building, might we be able to usefully employ labour laws and relevant provisions to better support the lived realities of workers in the twenty-first century.