Splitting hairs: A review of the ‘related to’ test under section 26 Equality Act 2010
Introduction
In Finn v British Bung Manufacturing Company [2023] EAT 165, the EAT upheld an employment tribunal’s decision that calling a male colleague a ‘bald c**nt’ was harassment related to sex. The employer’s counterargument, that baldness is not an exclusively male characteristic, failed.
Below, Imogen Brown reviews the decision in Finn v British Bung, and considers the wider circumstances in which conduct can be sufficiently ‘related to’ a protected characteristic for a section 26 claim to succeed.
Finn v British Bung: the employment tribunal’s decision
The relevant facts of the case are simple. The tribunal found that, on 24 July 2019, the Claimant’s colleague called him a ‘bald c**nt’ on the factory floor. They were satisfied the conduct was unwanted, and created “an intimidating, hostile etc” environment for the Claimant (see [232]). The question was whether it was sufficiently ‘related to’ the protected characteristic of age and/or sex: the Claimant was relying on both protected characteristics for the purposes of his harassment claim.
Before the employment tribunal, the Respondents’ counsel submitted that, because baldness is not an exclusively male characteristic, the ‘related to’ test was not met. The tribunal also considered whether baldness was ‘related to’ age in the same vein. In upholding the sex-related harassment claim, but dismissing the age-related harassment claim, the tribunal commented as follows:
[234] Plainly, some words or phrases would clearly be related to a protected characteristic. Where the link is less obvious then Tribunals may need to analyse the precise words used, together with the context, in order to establish whether there is any negative association between the two.
[235] In our judgment, there is a connection between the word “bald” on the one hand and the protected characteristic of sex on the other. [The Respondents’ Counsel] was right to submit that women as well as men may be bald. However, as all three members of the Tribunal will vouchsafe, baldness is much more prevalent in men than women. We find it to be inherently related to sex. (In contrast, we accept that baldness affects (predominantly) adult males of all ages so is inherently not a characteristic of age.)
At first blush, the tribunal’s reasoning is sound: it cannot be that an adverse comment is only ‘related to’ a protected characteristic if it refers to a feature that only those with the relevant protected characteristic possess. There are many examples that illustrate this point: in the first instance case of Brown v Young and Co’s Brewery plc ET Case No.3202099/13, for example, referring to a black colleague as looking like a ‘pimp’ was held to be harassment related to race, even though it cannot be said that all pimps are black individuals. The tribunal in British Bung referred to the ‘Hiya, big tits’ comment made to a female Claimant in the case of InSitu Cleaning Co Limited v Heads [1995] IRLR 4, EAT, which was held to be sex-related harassment. They reasoned that the fact a man with gynecomastia could also have been the victim of such a comment did not mean that the Claimant’s claim in InSitu was bound to fail.
However, it could be argued that the tribunal’s reasoning should also have applied to the issue of age, too. The tribunal were right that baldness affects “adult males of all ages”, but can it not also be said that baldness is “much more prevalent” in older individuals than younger ones? Why is baldness sufficiently prevalent in men, but not sufficiently prevalent in older people, for one claim to succeed and the other to fail? The tribunal’s reasoning did not rest on a specific statistical threshold, or any further evidence as to what level of ‘prevalence’ will be enough for the ‘related to’ test to be met. The finding was one, seemingly, of instinct alone.
The EAT’s decision
On appeal, the Respondents’ counsel argued that the tribunal was wrong to uphold the sex-related harassment claim. Even if 99 percent of bald people were male, it was argued, the claim should fail because the 1 percent of bald women (for example, women going through chemotherapy) mean that the comment was not ‘related to’ a man’s sex. The thrust of the argument was that a comment can only meet the ‘related to’ test if the nature of the comment was such that it could only apply to those with the protected characteristic in question.
Perhaps unsurprisingly, the EAT rejected this argument, finding that the tribunal had permissibly “recognis[ed] the fact that the characteristic by reference to which [the Second Respondent] had chosen to abuse the Claimant was more prevalent in people of the Claimant’s gender, more likely to be directed at such people, and, as such, inherently related to sex” (see [26]). The employment tribunal’s reasoning was held to be right.
Analysis: when will the ‘related to’ test be met?
The ‘related to’ test was seminally discussed in Bakkali v Greater Manchester Buses (South) Ltd UKEAT/176/17. At [31], the EAT held that conduct can be ‘related to’ a protected characteristic even if it is not ‘because of’ it: the ‘related to’ test requires a broader area of inquiry than that required under section 13 of the Equality Act 2010. It may involve considering the mental processes of the alleged harasser, but it will also involve the tribunal considering all the material before it, including evidence of the context in which the conduct complained of took place.
So, ‘intent’ may be relevant but not determinative; and ultimately ‘context’ is key. But when is that ‘context’ enough to found a claim of harassment as being ‘related to’ the relevant protected characteristic?
The authorities and guidance do not shed significant light on the issue. In Warby v Wunda Group plc EAT 0434/11, Mr Justice Langstaff upheld a tribunal’s decision that an employee who was accused by her manager of lying about having a miscarriage was not subjected to harassment ‘related to’ her sex within the meaning of (what is now) section 26 Equality Act 2010: “There was no necessary inherent discrimination in conduct that consisted of those words in that context” (see [24]). But on the ‘prevalence’ analysis used in British Bung, can that reasoning be correct? Whilst the EAT explained that the thrust of the comment was over the veracity of the Claimant’s miscarriage, rather than a derogatory remark related to her sex, doesn’t that distinction rest on the discriminator’s intent alone? The comment was directed towards a woman, by reference to a feature of hers (her miscarriage) which is far more, if not almost exclusively, prevalent in women. But the harassment claim was held to have permissibly failed.
An example in the EHRC’s Employment Code falls at the other end of the spectrum. At paragraph [7.10], it gives the following illustration of when the ‘related to’ test will be met:
A female worker has a relationship with her male manager. On seeing her with another male colleague, the manager suspects she is having an affair. As a result, the manager makes her working life difficult by continually criticising her work in an offensive manner. The behaviour is not because of the sex of the female worker but because of the suspected affair, which is related to her sex. This could amount to harassment related to sex.
This example resiles significantly from the high ‘related to’ threshold mooted in Warby. Indeed, it envisages that the conduct itself need not even refer to any prevalent sex-based characteristic in order for a section 26 claim to succeed. There simply needs to be a broader, sex-related context to the unwanted conduct in question. But can it even be said that the broad context is sex-related? Yes, the victim of the conduct is a woman, but the surrounding context of the behaviour is that the male manager suspects she is having an affair with someone else. His conduct does not specifically relate to the fact that she is female, but rather that she is a person in whom he has romantic interest. It could be argued that this example defines ‘context’ too broadly to be workable in practice: indeed, first instance sex-related harassment claims have failed in similar circumstances: see Zachariadis v KD Commercial Cleaning Ltd and ors ET Case No.2400380/15, where a female employee’s sex-related harassment claim, which concerned her being accused of ‘stealing [someone’s] man’ , failed on the ‘related to’ test.
Perhaps Tees Esk and Wear Valleys NHS Foundation Trust v Aslam and anor 2020 IRLR 495, EAT puts it best. At [25], it is said that there must be, in any given case, “some feature or features of the factual matrix identified by the Tribunal, which properly leads it to the conclusion that the conduct in question is related to the particular characteristic in question”. The nature of such a ‘feature’, and the categories of case to which that ‘feature’ will successfully apply, cannot be defined. The matter is one for the tribunal to consider on all the evidence before it: the cornerstone of the analysis being the ‘context’ in which the purported conduct took place.
Conclusion
The distinction the tribunal drew between age and sex in the ‘bald c**nt’ comment in British Bung is an interesting one. It characterises the tribunal’s role in assessing harassment claims as one of instinct: meticulous reasoning is left at the door. Of course, the spirit of section 26 is to proscribe the discriminatory stereotypes, assumptions, and biases which are proliferated in the way individuals recognise and interact with each other. But inevitably, those concepts are slippery, unquantifiable, and hazy. As examined above, there cannot be any bright line rules.
Practitioners may find themselves agonising over subtle distinctions regarding whether the ‘related to’ test is met, but ultimately the tribunal will step back, consider the material before it, and be guided by common sense. That makes our role as lawyers harder in predicting which way liability will fall. But as with many areas of discrimination law, the messy breadth of human experience cannot be neatly packaged in a black letter legal box. Section 26 is not the place to be splitting hairs.
