Principles into Practice: Protecting Offensive Beliefs in the Workplace – by Amir Paz-Fuchs

[Republished with permission from the UK Labour Law blog]


Over the past two decades, there has been a growing interest in the impact of human rights discourse on the employment relationship and employment rights. In particular, in light of the increased opportunity, or risk, of the public exposure of an individual’s life outside of work, more attention has been drawn to the implications of employees’ private life to their place of employment.

Image by xuanklpt from Pixabay

For progressives, the interest in preserving and protecting one’s right to privacy happily overlapped, for many years, with the substantive actions that the demand for privacy protected. In other words, progressives were determined to stand by the claimants and, where relevant, criticise courts that did not do so, not only in the name of privacy, but also because we/they asserted that the claimants in question did nothing wrong to begin with. Thus, in Pay v Lancashire Probation Service, a probation officer was dismissed because of his involvement in a business venture that was connected to sado-masochistic activities. In X v Y, a gay man was cautioned and placed on the Sex Offenders Register after a policeman arrested him for engaging in sex with another man in a public toilet. In Saunders v Scottish National Camps, the employee’s homosexuality was deemed sufficient grounds for dismissal. In Crisp v Apple Retail UK Limited [2011] ET/1500258/11 an employee was dismissed after he posted a critical appraisal of the new iPhone. And in Mathewson v RB Wilson Dental Laboratory [1989] IRLR 512 EAT the employee was dismissed after being caught in possession of cannabis. In all those cases, the employers’ decisions were upheld by the courts, and these decisions were criticised for neglecting to respecting an employee’s right to privacy, which should be constructed in a way that would ‘protect individuals from employer domination’.

In all these cases, it would seem that the liberal, progressive intuition recoils not only from the conclusion that homosexuality and the private use of cannabis can be grounds for dismissal, but even that they should be the subject of severe moral condemnation.

I thought of these cases recently, following the aftermath of the football match between Manchester United and Manchester City, in which City fans were caught on camera racially abusing United players. The Football Association investigated, and a man was quickly arrested for racially aggravated public order. So far, so reasonable. But the story took a peculiar turn, at least for me, when it was reported that the individual was identified as working for a construction company – Kier Company – which was quick to suspend him, citing the company’s ‘zero-tolerance policy towards any racist and discriminatory behaviour’.

It is here where I became slightly uncomfortable. And perhaps it is worth saying that my discomfort is, in no way, meant to minimise the, quite frankly, terrible times we are currently facing in this country. I could not agree more with the charge that the way politicians, including the Prime Minister, address the issue of migration, fuels this hateful vitriol.

But, those politics notwithstanding, the pertinent question is: should such behaviour, which occurred, of course, outside the workplace, have employment ramifications? Two of the main arguments that favoured the claimants in earlier cases – privacy and freedom of speech – clearly apply here as well. If we (as progressives) are to be taken seriously about our principled claims, i.e. that human rights should apply in the workplace even when the employer does not hold the same progressive values, we should be ready to bite the bullet and apply the same principle when those acts are offensive, and are clearly such that we (as a society) wish did not exist.

This position, simple as it may sound, must confront several objections. First, it is argued that we want to allow employees who express racist views, and hate speech in general, to be dismissed because we want to eradicate those expressions in society. ‘Society’ does not tolerate those views in this day and age. However, this view falls into a couple of familiar traps. First, the historical one: LGBT employees, and those with questionable morals (like the adulterous employee who lost her job for that reason in Spiller v Wallis) suffered from their divergence from public norms at the time, in a way we (justifiably) scoff at now. Of course, we feel that racism and homophobia are different, and that they truly should be eradicated, but some humility should guide us to assess the historical analogies. Second, there is the utilitarian argument: do we really think that sanctioning, shaming and publicly ostracising such behaviour would facilitate its eradication? There is a strong argument to suggest that it would only lead individuals harbouring those inclinations to go underground, feel victimised and even, at times, start wearing their foul opinions as badges of pride. There are examples for all of the above. This, in other words, is the argument that has at its axis the free speech of employees, and finds its justifications in the rationales of free speech in general. And yet, the workplace is a unique micro-cosmos, where people are forced to interact on a daily basis, and where concepts such as (industrial) democracy and (managerial) prerogative are deployed by way of analogy to the relationship between citizens and governments. And yet, those are still analogies, and the different context should be acknowledged.

So if we are to contextualise further, we find that there is a more sophisticated counter-argument to the one just discussed. That is, instead of focusing on public morals and public norms, one can place the centre of gravity on the employer’s values, as expressed by Kier Co (‘zero tolerance policy towards any racist and discriminatory behaviour’). Surely we cannot bar an employer from advancing such a laudable policy in her own workplace? I was considering this issue when a case just published – Forstater v CGD Europe – generated plenty of heat.

In this case, a public policy researcher who had a consultancy agreement with a not-for-profit think tank claimed that the respondent refused to renew her contract because of comments she had made on Twitter, etc expressing her beliefs about trans issues. The tribunal’s decision was limited to the preliminary question of whether the claimant’s beliefs qualified for protection under the Equality Act 2010 section 10(2) as any religious or philosophical belief in accordance with the Grainger criteria (which, in turn, relied on the ECtHR case of Campbell and Cosans v UK, which concerned corporal punishment. In particular, the tribunal found that the claimant’s belief faltered on the fifth, and final, criterion, namely: a belief that ‘must be worthy of respect in a democratic society, be not incompatible with human dignity and not conflict with the fundamental rights of others’. The tribunal did not consider in this preliminary hearing any question of unfair dismissal or the application of the Human Rights Act 1998 (including the rights to privacy and freedom of expression, although the latter was mentioned in brief [74]).

Why did the tribunal find that Forstater’s views are incompatible with human dignity and the rights of others? Probably the most problematic of those beliefs, from the tribunal’s point of view, is the claimant’s (factually incorrect) assertion that there are only two sexes in nature – men and women, along with the position that, from the claimant’s point of view, a person cannot transition from one sex to another [84-85]. The tribunal emphasised, at some length, the fact that the first position is scientifically misguided, and the second is legally mistaken. While it is true that the claimant was, in fact, mistaken on both counts, it is not clear how these mistakes affect her power to hold those beliefs.

Starting with the latter – a person is perfectly entitled to dispute any position taken by the legal system at any time. In fact, many academics have made a career, and scores of activists made history, by doing so. It is true that British law recognises a person’s right to transition and gain a Gender Recognition Certificate. The claimant recognised the legal state of affairs but held that this was a ‘legal fiction’ because a man can never truly become a woman, and vice versa [84]. This, perhaps, is an immoral belief, but it is not undermined by the fact that it contests the current state of affairs. One can think, for example, of (radical, for the time) 18th century advocates for women’s rights who argued that a man’s ownership over a woman is ‘legal fiction’, whilst still acknowledging the practical consequences of that fiction.

And as for the former, clearly the factual veracity of a position is not a precondition for its protection, or else no religion would overcome such an obstacle. And whilst we’re analogising ‘religion’ and ‘belief’, shouldn’t we pause to consider how the fifth criterion, on which the claimant’s belief failed (in a manner that it is controversial in itself), should have barred any of the three major, monotheist religions from protection? In other words, it would seem that as a belief (as opposed to a manifestation of a belief – more on that below) – the claimant failed in her quest for protection because she grounded her positions in her understanding of feminism and the protection of women and girls. However, if she were to ground her position in the tenets of a major religion (e.g. Christianity or Judaism) – it would have been very clear that the religion itself would be covered by the scope of the Equality Act. This distinction between political (or ideological) beliefs and religious beliefs seems, indeed, arbitrary. The question would then turn to the manifestationof the belief. This, I would argue, should have been the focal point of the tribunal’s analysis here as well. Alas, no such discussion is apparent.

Indeed, while the tribunal offers a detailed account of letters and tweets by the claimant, there is not even one occasion in which the claimant’s beliefs are asserted to have targeted a colleague in the workplace. For example, she repeated the need to be polite to others, including referring to all people using the pronouns they prefer; and argued for a ‘broader national conversation about how to reconcile the welfare of people who seek treatment for gender dysphoria and the basic human rights of women and girls’. In other words, this is an occasion where the right to privacy and the right to freedom of speech should have been front and centre to the analysis. And yet, the latter is only mentioned, and its impact is not assessed; whereas the former isn’t even mentioned. Moreover, I would argue that the right to privacy (in this sense – the separation of work from life outside of work) is the more important of the two.

Thus, I believe that the court in Smith v Trafford Housing Trust was absolutely right to conclude that the employer wrongly demoted a housing manager who posted on his Facebook wall a post in opposition to gay marriage. Crucially, Briggs J focused on the fact that the post could not be attributed to the manager’s employer, that he used moderate language and never offended any employee in their interaction with them. Pace Paul Wragg, I don’t find this analysis to be ‘intellectually unsatisfying’ in the least. Quite the contrary. It would be quite problematic for an employer to ignore, in the name of free speech, hate speech of one employee towards another. Indeed, just as acts that may be classified as hate speech can and should be addressed by the state, and prosecuted accordingly (see the very recent ECtHR case of Beizaras and Levickas v Lithuania, with thanks to Virginia Mantouvalou for the reference), it would seem implausible for an employer to ignore similar speech if it happens within the workplace, leading other employees to feel unsafe, and justifiably so. But if such speech happens outside the workplace, it is the preserve of the government to act against those individuals.

Returning to the case at hand: if an employee, such a Forstater, suffers a detriment for actions that happened outside the workplace, the only conclusion that one can reach is that she is sanctioned for her beliefs, and not for the manifestation of those beliefs. To repeat, the claimant asserted that she ‘would of course respect anyone’s self-definition of their gender identity in any social and professional context’. And no examples were given that she behaved differently.Similarly, to close circle on the case of the Manchester derby – whilst it could be that the City fan does harbour racist beliefs, his employer gave no indication that those beliefs were manifested in the work environment.

Sanctioning someone for their beliefs is a dangerous road to follow. I have consistently taught my students that there are very few absolute rights, of which the ‘right to freedom of thought’ is one (a position accepted by the Equality and Human Rights Commission). An employer has a right, indeed a duty, to protect the work environment, to enhance inclusivity and tolerance, and make sure that all are felt welcome. When realising that some workers are (surprise, surprise) impacted by the relentless torrent of hateful narratives coming from the tabloid press and the highest political echelons, but do not act on them in a way that affects others in the workplace, employers should not respond by removing such workers from employment and into victimhood. The authoritarian control of employers, which John Gardner and Hugh Collins recently alluded to, is wide enough within the workplace, and should not extend to ‘thought control’. And as Collins and Virginia Mantouvalou wrote, apropos Redfearn v UK, ‘a democracy cannot eliminate obnoxious views by permitting the imposition of economic hardship through dismissal’. Instead, employers should have the courage of their convictions, opt for opening the discussion, through training and dialogue. That is where their social responsibility lies and so, I would argue, is where the law should be.



Amir Paz-Fuchs is Professor of Law and Social Justice at the University of Sussex, where he teaches Employment Law and is Director of Sussex Clinical Legal Education.



(Suggested citation: ‘A Paz-Fuchs, ‘Principles into Practice: Protecting Offensive Beliefs in the Workplace,’ UK Labour Law Blog, 12 February 2020, available at

The white and blue (collar) divide of worker status recognition

Amir Paz-Fuchs
Amir Paz-Fuchs

In late May 2014, the Supreme Court delivered a ruling that should be a welcome development for those seeking the expansion of employment status and to those who have protection for whistleblowers close to their hearts. And yet, it can be argued (and will be, in what follows) that the ruling sheds light on a troubling discrepancy between the approach that courts take in cases involving white collar workers, and those involving blue collar workers.

First, the facts: the case involved a qualified solicitor, Krista Bates van Winkelhof [‘the appellant’], an equity partner in a law firm, Clyde & Co LLP [‘the LLP’]. In November 2010 the appellant became aware that the managing partner in a Tanzanian law firm which was involved in a joint venture with the LLP, was paying bribes to secure work and the outcome of cases. She reported these facts to the money laundering officers of the LLP. In response, she was suspended, was subject to allegations of misconduct and, in January 2011, ultimately dismissed. The appellant claimed protection under the whistleblowing provisions of the Employment Rights Act 1996. The central issue, however, is that such protection is awarded only to a ‘worker’ within the meaning of section 230(3) of the Act. This section is composed of two ‘limbs’. The first, limb (a), covers an individual who works ‘under the contract of employment’, and is thus ‘an employee’. It was not suggested that the appellant was an employee. The question was whether the appellant falls under limb (b), which covers an individual who works under

“any other contract, whether express or implied and (if it is express) whether oral or in writing, whereby the individual undertakes to do or perform personally any work or services for another party to the contract whose status is not by virtue of the contract that of a client or customer of any profession or business undertaking carried on by the individual.”

The employment tribunals and courts delivered a see-saw of decisions.

The rulings: In the first instance, the Employment Tribunal found that the appellant was not a worker, as she worked “in business in her own right receiving a share of the profits in relation to the work carried out”. The Employment Appeal Tribunal allowed her appeal, focusing on the fact that the appellant could not offer her services to anyone else, that she was in a subordinate position, and the LLP was not her client. The Court of Appeal reversed again. While Elias LJ acknowledged that that “there is a powerful case for saying that, focusing solely on the language of section 230, the terms of the statutory definition of worker were satisfied in this case” and that that the LLP “could in no sensible way be said to be either the client or a customer of the claimant” (thus accepting the analysis of the Employment Appeal Tribunal), he found that the appellant could not be viewed as a ‘worker’, because an LLP could not be described as an employer of its members. Or, from the reverse perspective, Elias LJ states that “underlying the statutory definition of ‘worker’ is the notion that one party has to be in a subordinate relationship to the other”. In addition, the CA found that section 230(3) was impliedly modified by section 4(4) of the Limited Liability Partnerships Act 2000. This provides:

“A member of a limited liability partnership shall not be regarded for any purpose as employed by the limited liability partnership unless, if he and the other members were partners in a partnership, he would be regarded for that purpose as employed by the partnership.”

So now the question became whether the restriction embedded in section 4(4) covers both limb (a) and limb (b) of section 230(3) of the ERA. If so, as the LLP argued, a partner cannot be an employee nor a worker. If a more restrictive interpretation is accepted, then section 4(4) will apply only to employees, but not to workers. The CA favoured the wider interpretation, thus concluding that those who would enjoy worker protection under the tests developed for section 230(3), would be denied such status as a result of section 4(4).

The Supreme Court disagreed.

It is important for the argument that follows to make note of Lady Hale’s statement, opening the substantive discussion. Taking a birds-eye view of the proceedings, she states (para 16):

The immediately striking thing about this case is how much hard work has to be done in order to find that a member of an LLP is not a worker within the meaning of section 230(3)(b) of the 1996 Act. It is common ground that the appellant worked “under a contract personally to perform any work or services”. It is now common ground that she provided those services “for” the LLP. It is also now common ground that the LLP was not her “client or customer”. The Court of Appeal … accepted that there was a “powerful case” that the definition was satisfied. How then can it be said that she was not a “worker” for this purpose?

Lady Hale states, first, that the ‘natural and ordinary meaning’ of ‘employed by’ is employed by a contract of service, and thus should not apply to a wider range of contracts. Second, the law now recognised a range of employment status, from ‘employee’ through ‘worker’ and on to self-employed individuals who provide services as part of a service to others, but are still protected under health and safety laws or the Equality Act 2010, for example. Therefore, if Parliament would have wished to include workers in the meaning of section 4(4), it could have done so explicitly. Thirdly, the Limited Liability Partnership is a new corporate structure, the parameters of which are yet to be determined. It is not obviously clear whether partners in such a structure may be workers (in the wider sense) of the partnership. If Parliament wished to limit the possibility, again, it could have done so.

Lady Hale also rejected the Court of Appeal’s observation that subordination is a necessary ingredient in the employment relationship. Subordination would be observable for those employed under a contract of employment. However, for workers and those under a contract “personally to do work” (Equality Act 2010, s 83(2)), subordination is ‘not a freestanding and universal characteristic’ (para 39).

Discussion: over the past decade, Professor Mark Freedland has advanced (more recently, with Nicola Kountouris) a ‘radical approach’ to expand statutory protection by detaching employment rights from the traditional dichotomy of employee/self-employed, and provided a new organising idea of personal work relationship to replace the old categories. This


flexible, purposive approach would take into account dignity, fairness, personal development and economic performance in the regulation of the personal work relationship.

It may be that if this approach suffered a setback in the 2011 case of Jivraj v Hashwani (at least according to Chris McCrudden, but see the reply by Freedland and Kountouris), it has regained force in the current case. While still operating within the parameters of the traditional categories, the Supreme Court’s common sense approach, on clear display in section 16 of its judgment (quoted above), suggests that such categories may be transformed from strict boundaries to signposts, en route to decisions that take into account the employment context, power relations, distributional and efficiency considerations. Expanding worker status in this case, it should be made clear, has implications not only for whistleblowing protection, but also for the application of the National Minimum Wage Act (probably only of theoretical relevance in this case), Working Time Regulations 1998 and Part-time Workers (Prevention of Less Favourable Treatment) Regulations2000, to give but a few examples. It may also pave the way for the expansion of employment status in other cases.

And yet, the juxtaposition of this case against a series of cases where worker status was routinely denied to agency workers employed as cleaners, social service workers, prison maintenance workers, for example, raises the concern that courts are more willing to expand protection to ‘white collar’ workers, all the while being far more judicious and legalistic where ‘blue collar’ workers are concerned. This critique is made less casuistic, and more structural, when analysing the Court’s comments concerning the (ir)relevance of subordination as an essential element in the analysis of worker status. In fact, one may say that if subordination

Not protected
Not protected

(whether administrative or economic – see here and here) was viewed as a necessary and sufficient ingredient for the assertion of worker status, the Court’s reasoning seems to reject not only the former, but also the latter. The implications are significant: minimising the importance of subordination as a necessary element would, as it did in the present case, allow employment protection to expand to independent, professional and expert individuals employed under a wide range of contracts. However, dismissing the relevance of subordination as a sufficient element would narrow (or, at least, bar the effort to further expand) protection currently denied to agency workers, regular casuals, zero-hour workers and others. This would be regrettable, for it is those workers whose subordination is most apparent, and who are most in need of protection.

Workfare: Alive and kicking between the constitutional and the administrative

Amir Paz-Fuchs
Amir Paz-Fuchs

On 28th April, 2014, the Department for Work and Pensions (DWP) announced ‘new intensive measures to help the long-term unemployed into work’. Despite the fact that the new rules will affect 200,000 people, there was only scantmention in the national press, reflecting an ambivalence in the government’s attitude. This ambivalence can be attributed to the central plank of the DWP’s initiative – the Community Work Placement scheme, referred to, mainly by its opponents, as workfare. Workfare schemes, developed originally in the United States and Australia, require claimants to ‘work for their benefits’. Why the ambivalence? First, workfare initiatives are incredibly unpopular amongst many people. So much so, in fact, that it has managed to galvanise a campaign to Boycott Workfare which, to date, has managed to persuade over dozens of providers to refuse to take part in the scheme. Secondly, the current scheme replaces a previous effort, which ran into embarrassing (for the government) legal difficulties, mentioned below.

I will suggest that the two causes for embarrassment, the constitutional and the administrative, correlate the two main legal challenges to workfare seen thus far: the first objects to workfare by reference to human rights arguments, and the second focuses on procedural failures. Between the two, I argue, lie the substantive objections based on labour law and labour relations, which should form the central platform to challenge workfare.

The first type of objection is constitutional, or human rights based, in nature. Requiring claimants to work for benefits has been seen by opponents as tantamount to forced labour, or even slavery. This argument (referring to Article 4 of the ECHR) was rejected by the Strasbourg Court time and time again, including Talmon v the Netherlands(1997) and slaverySchuitemaker v the Netherlands(2010), and by the UK Supreme Court in R (Reilly and Wilson) v Secretary of State for Work and Pensions (2013) to which I will return below. Feeling, as I do, that forced labour and slavery are matters that are of the utmost gravity, and that should thus be dealt with the severity, I find such decisions compelling. And yet, ideologically, or even broader – politically, there is an uncomfortable feeling about demanding claimants to ‘work for’ their benefits. That is, if one rejects the vacuous rhetoric of reciprocity (constantly spewed by wealthy tax evaders and the like, who seem to identify reciprocal requirements only among the working class) and accepts that people gain ‘social rights of citizenship’, in TH Marshall’s timeless phrase; and that unemployment is, at least partially, the result of structural forces that are beyond the control of the individual, then unemployment benefits should be granted to individuals as a right. And since benefits did not, in the past, depend on participating in work schemes, adding this condition suggests that poor individuals are required (as they have very little choice in the matter) to work without pay, constituting almost the dictionary definition of forced labour. The plot thickens when we ask cui bono? As with slavery and forced labour, there will always be someone benefiting from cheap labour. A freedom of information request filed by Boycott Welfare revealed that local councils in UK benefited from over half a million hours of unpaid labour through government back to work schemes, in libraries, parks, etc. Similarly, in a series of articles in the NY Times, and in a fascinating book, Jason DeParle exposed numerous occasions of workers being dismissed from low-paid jobs, only to be replaced by (free) workfare scheme participants (sometimes – they were the same people, now coming to work for free!).

What of the administrative legal structure that governs workfare? Here, some brief background is required. The current scheme is, in fact, a revised version of a previous attempt by the Coalition government.

Caitlin Reilly
Caitlin Reilly

The previous Employment, Enterprise and Skills Scheme, put in place by the Jobseekers’ Allowance Regulations 2011, was found to be flawed by the High Court, the Court of Appeals and the Supreme Court following a successful challenge by Caitlin Reilly and Jaimie Wilson. In both cases, the High Court found that the claimants were not given written notice concerning their participation in the scheme. On appeal, it was determined that the flaws in the scheme lie not only in its implementation, but actually run deeper. The Court of Appeal concluded, and the Supreme Court agreed, that the 2011 Regulations are ultra vires section 17A of the Jobseekers Act 1995 because they fail to prescribe the description of the relevant schemes.

This finding was the cause not only of serious embarrassment, but could have been very costly, as it had potential implications for thousands of other participants sanctioned under the scheme. In a highly controversial move, the government (with support of members of the opposition) swiftly passed a retroactive measure – the Jobseekers (Back to Work Scheme) Act 2013 and the Jobseekers’ Allowance (Schemes for Assisting Persons to Obtain Employment) Regulations 2013. The intention of the 2013 Act and Regulations was to validate the 2011 Regulations retrospectively, a measure derided by some as an affront to the rule of law, and a ‘shameful retroactive stitch-up’. As to their substance – the 2013 Regulation simply incorporated the schemes that were purportedly included in the 2011 Regulations. And so, now we can find, in Regulation 3(8), that the Work Programme scheme authorises the provider (very often – a private company) to subject the claimant to work placements for the benefit of the community for a period of up to two years!

The administrative failures of the earlier scheme were easy to amend: include details of the schemes into the new regulations, inform participants very clearly of their duties, include details of the extent of the scheme, and all is well. Or is it? We find, then, that between narrow (and thus extremely successful, but short lived) challenge of administrative law, and the broad, constitutional, human rights based challenge of aligning workfare with forced labour (which never had a chance to succeed) lies a vast terrain that was not addressed: the substantive area of employment rights and employment relations.

Workfare is stealing jobs
Workfare is stealing jobs

Workfare is, after all, an extremely harsh employment measure. The government’s incentive to put claimants in the scheme can be explained, for example, by the fact that the Office of National Statistics classifies workfare participants as being in employment, thus helping push up the employment rate. Within the family of human rights violations, even within the ambit of Article 4, it is a far cry from being a serious violation (incomparable, for example, with the case of Siliadin v France). Administrative challenges, on the other hand, when established, are easy for courts to accept, and easy for governments to amend. But between the two, at the heart of the matter, lie the effect that workfare has on the employment prospects of the participants, and on the terms and conditions of other workers, who are now in impossible competition with unpaid labour. Thus, it may be worth considering that a DWP report on the pilot programme, was published (with no press release, for reasons that will be very clear) in 2012. According to Jonathan Portes, who conducted a review of the report on behalf of the NIESR, the scheme “has no impact on employment; it leads to a small and transitory reduction in benefit receipt; and worst of all, it may even lead to those on the programme moving from Jobseekers’ Allowance to Employment and Support Allowance”.

Claimants, and other low-paid workers, are subjected to serious labour rights violations as a result of workfare schemes. But if workfare is to be challenged successfully, it should be done by addressing the true nature of its harm.

Amir Paz-Fuchs is Lecturer in Employment Law at the University of Sussex, and author of Welfare to Work: Conditional Rights in Social Policy (OUP).