Business & economics Social commentary

Unconscious bias: now you see it, now you don’t?

Roger Kline is Research Fellow at Middlesex University Business School. Roger would like to acknowledge the helpful assistance in drafting this blog from Peter Daly, Principal Employment Lawyer at Slater Gordon lawyers.

“I am a black woman who has been harassed at work by a white man. My behaviour and the standard of my work has been praised by every witness, by my line manager and in every appraisal. It is accepted that staff surveys and workforce data on bullying, discipline and recruitment in this organisation strongly suggest there is race discrimination here. You have heard what other BME staff have said about their treatment and mine.

“I would therefore like to ask the panel to consider again how they can possibly conclude race played no part in my treatment. It is agreed I was harassed. Whilst there have been no explicit racist comments you should surely consider whether I have been subjected to race discrimination even if the person harassing me denies it or is even unaware he is a racist?”

 This member of staff then pointed out (this is quoted verbatim from a written appeal) that her employer had recently put all its managers through an “unconscious bias” training programme (at significant expense) but now appeared to deny that such unconscious bias might be a factor in her treatment. Whilst the effectiveness of unconscious bias training is in dispute no one disputes its powerful existence.

 The member of staff who shared this with me was a white colleague. He wanted to know how, in such circumstances, the employer could decide racism was not a factor in her treatment when in his view it clearly was.

 This is not a unique case. A fairly recent high profile example came in the recent Employment Tribunal which awarded Richard Hastings £1million due to his treatment by an NHS Trust. In that Tribunal the Court unanimously agreed that “unconscious” race discrimination played a role, a case I wrote about at the time. Just a few days ago the Court of Appeal used a similar approach.

The statutory framework

The law is reasonably clear and is certainly well known (or should be) to NHS Employment lawyers and senior HR staff as follows:

S.26 Equality 2010 (1) (a) and (b) consider what to do if person (A) harasses another person (B) in a manner which constitutes “unwanted conduct related to a relevant protected characteristic, which has the purpose or effect of violating an individual’s dignity or creating an intimidating, hostile, degrading, humiliating or offensive environment for that individual” (my emphasis).

Section 26 (4) then states thatIn deciding whether conduct has the effect referred to in subsection (1)(b), each of the following must be taken into account:

  • (a) The perception of person B above
  • (b) The other circumstances of the case
  • (c) Whether it is reasonable for the conduct to have had that effect.

In the case considered at the start of this blog, it was agreed by the panel that bullying and harassment had indeed taken place. They correctly noted that:

  • There was no explicit racial comment verbally or in writing
  • There was no other direct evidence that the individual manager was motivated by racism in this case
  • Though it was agreed there was an act of harassment (several in fact) which were committed by a white manager to a BME member of staff, but that in itself (rightly) is not proof that the harassment was racially discriminatory. 

However, what they did not then go on to consider, which a reasonable investigation always ought to do, is whether the explanation given for the harassment fully explained what happened.

In doing so they should have considered :

  • the evidence of the accused person
  • the evidence of other witnesses
  • evidence from within from the wider work environment which might be helpful. That would include workforce data and staff survey data which might (or might not) demonstrate a pattern of institutional discrimination.

A summary of what an investigation into such matters might include is found in my previous article about conducting an inquiry into workplace racism. If the explanation provided for the action does not provide a credible explanation for the actions complained of, then the investigator (or panel) should then consider what other cause there might be, including whether the harassment causing the detriment might be considered racially discriminatory.

Case law

In considering whether the act of harassment might be racially discriminatory the courts have made clear that direct discrimination and discriminatory harassment are unlawful, whether or not the motive or intention which led to the act in question was consciously discriminatory.

Thus, whilst demonstrating that direct discrimination or discriminatory harassment has taken place may well involve an analysis of the reasons for the discriminatory treatment complained of, the courts have made it very clear that it is not necessary to show that the person(s) alleged to have discriminated did so consciously since “unconscious” discrimination is also prohibited, as two House of Lords cases made clear.

Lord Browne-Wilkinson noted that claims under discrimination legislation present special problems of proof as those who discriminate;

‘. . . do not in general advertise their prejudices: indeed they may not even be aware of them’. (Glasgow City Council v Zatar 1998 ICR 120, HL)

In another significant case, the House of Lords similarly stated;

Many people are unable, or unwilling, to admit even to themselves that actions of theirs may be racially motivated” (Nagarajan v London Regional Transport and others [1999] IRLR 572 (HL)).

These decisions of the highest court in the land are reflected in lower court decisions. Thus one EAT ruled that a Tribunal will not assume that a person’s actions are free of subconscious bias even if the person is an honest and reliable witness, and one who genuinely believed they were acting for non-discriminatory reasons. (Gellser and another v Yeshurun Hebrew Congregation UKEAT/2016/0190.)

A similar understanding of “unconscious bias” played a crucial part in the Employment Tribunal decision (not appealed) referred to earlier (Mr R Hastings v Kings College Hospital NHS Foundation Trust: 2300394/2016.

Employers must also bear in mind that discrimination (including race discrimination) need not be the main reason for an act or omission to have been discriminatory. Case law has determined it simply needs to have a “significant influence”:

“… the discriminatory reason for the conduct need not be the sole or even the principal reason of the discrimination; it is enough that it is a contributing cause in the sense of a ‘significant influence’.  (Law Society v Bahl [2003] IRLR 640, at 83).

If it is established that there is an instance of negative conduct which could be assigned to race discrimination, and the employer cannot provide a reasonable and adequate explanation that this was not due to discrimination, then the court or tribunal in accordance with s.136 Equality Act 2010 can draw an inference that the negative conduct was caused by discrimination (see also Fox v Rangecroft [2006] EWCA Civ 1112; and Barton V Investec Henderson Crosthwaite Securities Limit [2003] I.C.R. 1205).

The burden of proof

The burden of proof may therefore shift in discrimination cases. Article 8 of the EU Race Equality Directive (No 2000/43) confirms that where a prima facie case of discrimination exists, it is for the respondent to prove that there has been no discrimination. The same approach is legislated at s.136 Equality Act 2010:

(2) If there are facts from which the court could decide, in the absence of any other explanation, that a person (A) contravened the provision concerned, the court must hold that the contravention occurred.

(3) But subsection (2) does not apply if A shows that A did not contravene the provision.”

So, there are two stages than any HR professional (or indeed trade union official) should consider:

  • First, has the person making the complaint of discrimination sufficiently established the facts from which it may be presumed on the balance of probabilities that there has been discrimination?
  • Second – if established – the burden of proof shifts to the respondent based on the balance of probabilities (Meister v Speech Design Carrier Systems GmbH [2012] Case C-415/10 CJEU).

In reaching such a decision the employer should consider whether there may have been any other non-discriminatory reason that satisfactorily and fully explained the detriment complained of. But if they are unable to do so they must bear in mind that:

  • The decision making need not be consciously racist
  • Race discrimination need not be the sole or even the main factor.

In the case summarised at the beginning, the panel agreed there had been harassment and no convincing explanation had been provided for it. The panel (and indeed HR) should therefore have considered, whether in light of (a) the local NHS staff survey and workforce data, (b) the evidence from other BME staff about their own poor treatment, and (c) the fact that the person who suffered negative treatment from a white manager was from a BME background, whether there was a prima facie case that one motivating factor in the treatment of this member was her race.

They did not do so and thus placed the Trust at risk of financial and reputation cost at a Tribunal. More importantly, perhaps, they betrayed the member of staff who plucked up courage to raise the original concern of harassment and in doing so the Trust may have deterred other staff from raising similar concerns.

Many employers now put their managers through unconscious bias training. It is difficult to understand how they can do this and not consider whether such unconscious bias might be a factor in treatment their staff receive.