2018/2 Court of Appeal restores burden of proof test in discrimination cases (UK)

Summary

The Court of Appeal has overruled the recent Employment Appeal Tribunal (EAT) decision in Efobi – v – Royal Mail [2017] IRLR 956 (reported in EELC 2017/41), restoring the previous position that a claimant in a discrimination case has the initial burden of proof – which ‘shifts’ to the respondent to provide an explanation of why its conduct was non-discriminatory if a prima facie case is proven.

The Court of Appeal disagreed with Mrs Justice Elisabeth Laing’s ruling in Efobi, that section136 of the Equality Act 2010 had made a substantial change to the law when it was introduced, on the basis that it could not be fair that a respondent should have to discharge the burden of proof without the claimant first showing that there is a case to be answered. Lord Singh ruled that it could not have been Parliament’s intention to remove this initial burden of proof when it enacted the Equality Act.

Legal Background

UK discrimination legislation is derived from the EU Directives on discrimination in employment, all of which include provisions dealing with the burden of proof. Prior to 2010, the various UK discrimination laws followed the burden of proof wording in the EU Directives stating that:

“…where the [claimant] proves facts from which the tribunal could, apart from this [provision] have concluded in the absence of an adequate explanation that the respondent has committed [an act of discrimination], the tribunal must uphold that complaint unless the respondent proves it did not commit that act”.

The two leading UK cases on the burden of proof, Barton – v – Investec Henderson Croswaite Securities Ltd [2003] IRLR 332 and Igen Ltd and others – v – Wong [2005] IRLR 258 were both decided under the pre-2010 legislation. Both cases emphasised that it was for the claimant to prove facts from which a tribunal could conclude that discrimination has occurred. If the claimant did not prove such facts then his or her claim would fail.

In 2010, the Equality Act was implemented, consolidating the various pre-existing pieces of UK discrimination law. Section 136 of the Equality Act provides as follows:

“…(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 [of the Equality Act] 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.”

This section does not specifically state that in the first instance the claimant must prove facts. However, the Act’s explanatory notes repeated the previous formulation of the burden of proof, as requiring the claimant to first prove facts from which an inference of discrimination could be drawn. The EAT in Efobi ruled that the tribunal shoulder consider “all evidence, from whatever source”, which marked a move away from the original position. This position has now been restored and the courts have been directed to continue to adopt the approach that it is the claimant who has the initial burden of proof.

Facts

Mr Ayodele, a black man from Nigeria, was employed as an agency worker and subsequently as an employee (from 2007) by Citylink in one of their distribution centres. In 2012, Mr Ayodele resigned and claimed constructive dismissal as a result of a repudiatory breach of his employment contract, as well as race discrimination and harassment and victimisation. His allegations included that Citylink had failed to handle his pay and requests for annual leave correctly.

Both the Employment Tribunal and the EAT dismissed Mr Ayodele’s case, as he was unable to discharge the burden of proof. He was however, able to appeal to the Court of Appeal on the basis of the finding in Efobi that there was an error in the application of the burden of proof under section136 of the Equality Act, claiming that the tribunal had incorrectly imposed the initial burden of proof on him to prove his case of discrimination.

Judgment

Mr Ayodele put forward two grounds of appeal: (1) the tribunal incorrectly considered Citylink’s explanations at the first stage of the burden of proof test; and (2) the tribunal incorrectly placed the burden of proof on Mr Ayodele, contrary to the decision in Efobi.

The Court of Appeal dismissed Mr Ayodele’s appeal in its entirety. The court held that the ruling in Efobi was wrong, meaning that the claimant still bears the initial burden of proof. The court stated that in order for a tribunal to be able to make an assessment, the claimant has to start the case, otherwise there is nothing to address or for the tribunal to assess. Furthermore, it stated that requiring the Respondent to discharge the burden of proof without first requiring the claimant to prove a prima facie case of discrimination raises an issue of fairness.

The court therefore overruled Efobi and restored the original approach towards the burden of proof in discrimination cases.

Commentary

The re-interpretation of the burden of proof test in Efobi was unexpected and had the potential to make it easier for claimants to succeed in discrimination claims and impact how all UK discrimination cases were decided in future. The judgment in this case therefore is of importance, as it restores the original position.

We are currently awaiting the outcome of an application for permission to appeal the Efobi decision in the Court of Appeal. If permission is granted, it is highly likely that the decision in Ayodele – v – Citylink will have some influence over the court and it will be interesting to see how this unfolds.

Comments from other jurisdictions

Bulgaria (Ivan P. Punev, DKGV Attorneys & Counsellors At Law): The main piece of legislation providing the general framework of anti-discrimination rules in Bulgaria is the Law on Protection Against Discrimination (the ‘LPAD’). The LPAD contains an express provision governing the burden of proof in discrimination-related cases – Article 9 of the LPAD. Pursuant to this: “In the course of proceedings concerning protection against discrimination, after the party claiming that he or she has been discriminated against has presented facts based on which it can be presumed that a case of discrimination has occurred, the respondent party must prove that it did not violate the principle of equal treatment.”

The position consistently upheld by the Bulgarian courts is similar to the findings of the UK Court of Appeal described in the summary of the case. In its case law, the Bulgarian courts have interpreted the LPAD provision to mean that the party claiming discrimination has the burden of proving that it was likely the discrimination occurred, and only after this proof has been presented does the burden shift to the respondent – which then must prove that its actions were not in fact discriminatory and did not violate the principal of equal treatment.

Croatia (Dina Vlahov Buhin, Schoenherr): The Croatian courts would have applied the relevant provisions of two Acts, these being the Labour Act and Act on Prevention of Discrimination.

According to both of these, in claims of discrimination, the ‘rule on division’ – that is, the shifting of the burden of proof from the employee to the employer must be applied. This means that the burden of proof at the beginning of the procedure lies with the employee (claimant) and he or she is required to prove only the probability of the claim. Probability should be interpreted as meaning that there appear to be more arguments in favour of the existence or absence of a fact than those speaking against its existence or absence. Therefore, an employee must prove that he or she has been placed in a disadvantaged position and that this may have happened as a result of direct or indirect discrimination. Accordingly, the employee invoking discrimination must state the basis in law for the belief that he or she has been placed in a disadvantageous position compared to other employees in a comparable situation. If reasonable suspicion of discrimination exists, the burden of proof then shifts to the employer (respondent). The rule on division is only explicitly excluded where the provisions of the Act on Prevention of Discrimination are used in the context of criminal and misdemeanour proceedings.

In general, the Croatian courts act in accordance with the division rule, but some courts seem to take different stance. For example, the County Court of Zagreb, as a second instance court, overruled one appeal in a discrimination case related to employment, stating that the fact that the first instance court failed to apply the rule on division did not represent a substantive violation of the proceedings and did not impact on the legality of the decision (County Court decision Ref no Gžr-811/11). In that particular case, instead of shifting of the burden of proof to the employer, the County Court introduced the much stricter standard (with respect to employees) which should be used only in criminal and misdemeanour proceedings. This would seem to be contrary to an explicit provision in the Act on Prevention of Discrimination and a decision based on it is likely to represent an incorrect application of the law – which itself should be reason enough for a reversal of the decision.

Abandoning the rule on shifting the burden of proof in employment-related cases may significantly complicate the already unfavourable position of employees, therefore, in my view, the courts should apply the rule as prescribed in law and shift the burden of proof to the respondent once the plaintiff has proved the probability of his or her claim.

The Netherlands (Peter Vas Nunes, BarentsKrans): It is always a pleasure to read British judgments. Continental courts would do well to take a cue from their personal and well-reasoned style.

In the Ayodele judgment reported above, the Court of Appeal overruled the Employment Appeal Tribunal’s recent judgment in Efobi, which was reported in EELC 2017/41.The author’s commentary compares the two cases, which is also what I will attempt to do below.

Before doing so, there is an aspect that attracted my attention, that is not directly related to the subject matter of this case report. It has to do with the issue of how to interpret statute. Prior to 2010, the discrimination laws in England and Wales provided: “……where the claimant proves facts from which [the court may infer a presumption of discrimination]”. The 2010 Equality Act replaced this language by: “If there are facts from which …..” (my emphases added). At first glance, this change of wording could be seen to reflect a material amendment of the law on burden of proof. However, the government’s Explanatory Notes to the Equality Act repeated the previous formulation of the burden of proof. Thus, the wording of the new statute differed from the Explanatory Notes.

In Efobi, the EAT, referencing the House of Lords’ 2002 decision in Westminster City Council v NASS, held that Explanatory Notes cannot be treated as reflecting the will of Parliament. This allowed the EAT to disregard judicial precedent based on the old law. In Ayodele, the Court of Appeal, implicitly applying the ‘mischief rule’, takes pain to identify the will of Parliament in respect of the said change of language. It does so by analyzing the initial consultation paper that started off the process leading to the adoption of the Equality Act, the response generated by that consultation paper, the White Paper, what the academic and professional community had to say on the subject prior to enactment (and, strangely, even post enactment), as well as the case law on the old statute, which “Parliament can be taken to have known”. This analysis leads to the conclusion that “Although [the EAT] was right to point out that what is said by the executive in Explanatory Notes cannot be taken to represent the intention of Parliament, it is telling that nothing in the Explanatory Notes [….] or in any other document which led up to the enactment of the Equality Act pointed to there being any perceived mischief that needed a change of substance in the law”. This purposive method of interpreting statute is familiar to Dutch lawyers. It is interesting to see how it continues to gain ground in common law jurisdictions.

As was common ground in both Efobi and Ayodele, attributing the burden of proof under Race Directive 2000/43 and, hence, in the Equality Act, involves two stages:

• Stage 1: the claimant proves facts from which it may be presumed that there has been discrimination;

• Stage 2 (in the event such a presumption is established): the respondent proves that there has been no breach of the principle of equal treatment.

 

It would appear that Mr Efobi had not asked his employer for disclosure about the race of the successful candidates for the jobs he failed to get. This failure to ask may well have contributed to him losing his case at the first level. What would have happened had he, at stage 1, stated (and the court accepted as evidence): (1) I applied for over 20 roles with my employer, but despite being qualified, I was unsuccessful; and (2) my employer has turned down, without explanation, my request to provide me with information regarding the successful applicants’ race? In that case, according to the EAT, the ET should have found the combination of these two facts to constitute sufficient prima facie evidence to reverse the burden of proof. I conclude this from the fact that the EAT held, “At the first stage of the analysis […] there is no burden on a Claimant to prove anything […] What the ET has to do is to look at the “facts” as a whole. If a Respondent chooses, without explanation, not to adduce evidence about matters which are within its own knowledge, it runs the risk that an ET will draw inferences […] which are adverse to it on the relevant areas of the case. Those inferences will then be part of the “facts”[…]”. On this basis, the ET should have considered the employer’s position (in this case: failure to disclose information regarding the other candidates) already at stage 1. This would have been to the advantage of the employee.In Ayodele, the court also considered the employer’s position at stage 1. In that case, this was to the benefit of the employer. As I understand the facts, Mr. Ayodele stated, in essence: (1) I was forced to resign because my employer treated me poorly on account of my race, and his employer responded: (2) we treat all our employees poorly, regardless of race. The Employment Tribunal, taking account of both 1 and 2, held that Mr. Ayodele’s poor treatment was unrelated to his race. The Court of Appeal approved.

How practical is it to distinguish sharply between stages 1 and 2? They are not watertight compartments. In Ayodele, the Court of Appeal quoted from a 2007 House of Lords judgment: “Although no doubt logical, there is an area of unreality about all of this. From a practical point of view it should be noted that, although [the law] involves a two-stage analysis of the evidence, the tribunal does not in practice hear the evidence and the argument in two stages. The Employment Tribunal will have heard all the evidence in the case before it embarks on the two-stage analysis.”.The Court of Appeal concurred with the ET’s rejection of Mr. Ayodele’s discrimination claim, holding: “In arriving at that finding the ET may well have taken into account facts which were adduced in evidence not only by the Appellant but also by the Respondents. However, in my view, there was nothing impermissible about that. As the authorities […] make clear, there is a vital distinction between the “facts” and any “explanation”. It is only the explanation which cannot be considered at the first stage of the analysis”. How does this apply to Efobi? Was his employer’s failure to reveal the racial backgrounds of the individuals, who had applied successfully for the jobs for which he had been turned down, not a fact? The EAT qualified it as such, not as an explanation.

In summary, the EAT’s approach in Efobi does not strike me as all that unreasonable (although its finding that, at the first stage, “there is no burden on a Claimant to prove anything” was, perhaps, worded somewhat unfortunately). Should the EAT really have reasoned: the court and the parties know that the employer has failed to give information about the successful candidates, but the court may not do anything with this knowledge until Mr Efobi has proved what we all know? I expect that a Dutch court would have taken an approach similar to that of the EAT. In fact, a Dutch court might well have treated the entire debate held in Ayodele as academic and theoretical.

As a final note, the above disregards the issue of whether the ECJ’s ruling in Meister would have benefited Mr Efobi or have worked against his claim. In the ruling, the ECJ held: “Article 8(1) of Council Directive 2000/43/EC […] must be interpreted as not entitling a worker who claims plausibly that he meets the requirements listed in a job advertisement and whose application was rejected to have access to information indicating whether the employer engaged another applicant at the end of the recruitment process. Nevertheless, it cannot be ruled out that a defendant’s refusal to grant any access to information may be one of the factors to take into account in the context of establishing facts from which it may be presumed that there has been direct or indirect discrimination. It is for the referring court to determine whether that is the case in the main proceedings, taking into account all the circumstances of the case before it.”

Subject: General discrimination
Parties: Ajayi Adodele – v – Citylink Limited
Court: Court of Appeal
Date: 24 November 2017
Case number: A2/2014/2795
Internet publication: http://www.bailii.org/ew/cases/EWCA/Civ/2017/1913.html