Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Back to all news

Martin Fodder considers EAT guidance on dealing with amendments at trial in discrimination (and whistleblowing) claims involving alleged manipulation

Martin Fodder considers the EAT guidance on achieving a balance between the rights of claimants and respondents in discrimination (and whistleblowing) claims involving alleged manipulation or allegedly tainted information: The Commissioner of Police of the Metropolis v Denby [2017] UKEAT 0314_16_2410 (Kerr J presiding) 24 October 2017.

Denby is the first appellate level case to consider the difficulties that may arise when an employment tribunal applies the principles laid down by the Court of Appeal in Reynolds.

Tainted information/ the CLFIS v Reynolds principle

In CLFIS v Reynolds Ms Reynolds was employed by CLFIS. She was in her seventies. Her contract was terminated after rumblings of discontent with her performance within management.  Mr Gilmour was the dismissing officer. Ms Reynolds’s claim that she had been discriminated against on the ground of her age was dismissed by the ET. The ET had said that the decision-maker was Mr Gilmour and no one else, and it acquitted Mr Gilmour of being influenced by Ms Reynolds’s age. The EAT remitted the case to the ET, accepting the argument advanced on Ms Reynolds’ behalf that, even if the sole decision-maker was Mr Gilmour, his decision might have been shaped and informed by others within CLFS and the ET should have considered that possibility. The Court of Appeal, Lord Justice Underhill giving the substantive judgment, allowed CLFIS’ appeal. The ET had focused solely upon Mr Gilmour and his reasons for acting as he did, because it had not been suggested on behalf of Ms Reynolds that anyone else discriminated against her. The burden of proof provisions in the Equality Act did not place a blanket obligation on CLFS, as the employer, to prove the absence of discrimination in respect of every act of every employee that had formed part of the chain of events leading to the dismissal. Ms Reynolds had to establish a prima facie case that the dismissal had been because of her age. Whether that case was made out had to be decided by reference to the case that Ms Reynolds had advanced. Since the case she advanced only referred to Mr Gilmour, the ET had not erred in only considering Mr Gilmour’s motivation.

As explained by the Court of Appeal in CLIFIS v Reynolds, where it is contended that one employee of the respondent had been influenced by another employee into acting in a way detrimental to the claimant, an ET should treat the conduct of the person supplying the information—the ‘influencer’— as a separate act from that of the person who acted on the information supplied—the ‘influencee’. The alternative—namely a ‘composite approach’ which sought to bring together the influencee’s act with the influencer’s motivation—was unacceptable in principle because liability could only attach to an employer where an individual employee, or agent for whose act the employer was responsible, had done an act which satisfied the definition of discrimination. The individual employee who did the act complained of must him or herself have been motivated by the protected characteristic. It would be quite unjust for the person allegedly influenced—Mr Gilmour in this case—to be liable to a claimant where he personally was innocent of any discriminatory motivation.

Underhill LJ said that the solution in such cases was to ensure that the claim specifically raised the conduct of or act of the influencer and cited that influence either as the detriment or as contributing to the risk of a dismissal or detriment. This would ensure that the tribunal would then evaluate that specific claim and the extent to which the influencer caused or contributed to the risk of dismissal or detriment on a loss of a chance basis—that is to say to what extent the influencer’s influence contributed to the risk to the claimant of their dismissal or a detriment.

Underhill LJ acknowledged some practical difficulties arising from this approach. Such an allegation needed to be put clearly and notice of it given so that the respondent could call the necessary evidence. However, the fact that the apparent decision-maker had been influenced by another might not become clear to the claimant until well into the proceedings, or even at trial. Underhill LJ did, however, consider that these practical difficulties could be surmounted, if need be by an amendment being sought at the appropriate time.

Application of CLFIS to whistleblowing cases

As set out above, the Court of Appeal has now decided in Jhuti that the CLFIS v Reynolds approach does not apply to a claim of unfair dismissal contrary to s.103A ERA.  That decision is the subject of Benjamin Gray’s Comment of 17th November 2017 (below) and will be considered in detail in the forthcoming Third Supplement to Whistleblowing Law & Practice OUP 3rd Edition to which there will be a link on the Publications page of this website. Conversely as recognised by the EAT in Osipov (as to which see Benjamin’s Comment and the Second Supplement to WL&P), the approach to be applied in detriment cases does involve essentially the same exercise as in CLFIS v Reynolds of attaching liability to the co-worker (acting in the course of employment) or agent (acting with the employer’s authority) who both did the detrimental act or deliberate failure to act and did so on grounds of the protected disclosure.  That is reinforced by the observations of Underhill LJ in Jhuti as to the scope to read across from the discrimination legislation in relation to those parts of the whistleblowing legislation which use similar terminology. A similar underlying principle applies: it would unjust for an innocent influencee to be liable for whistleblowing detriment; instead the claim should lie against the influencer and, of course, his or her employer subject to acting in the course of employment and subject to the reasonable steps defences.

Applying CLFIS where the evidence of manipulation or influence only emerges during the hearing

The implications of and practical difficulties that can arise from an application of the principles laid down in CLFS were explored by the EAT in the sex discrimination case of Denby. Chief Inspector Denby was in charge of one of the five arms of the Territorial Support Group, “TSG1”, based at Paddington. His female comparator, Chief Inspector (CI) Edwards, was in charge of TSG3, in Ilford. Both reported up the same chain of command and had identical roles. Similar complaints were made against both CI Denby and CI Edwards, that overtime had been booked irregularly by officers at their respective TSGs. CI Denby’s complaint compared the more rigorous action taken against himself with that taken against CI Edwards. The ET upheld his complaint. During the hearing and against the opposition of the respondent, MPS the ET allowed CI Denby to add claims (which went on to succeed) that an officer in his chain of command had made or been party to two decisions adverse to him: CI Denby’s pleaded case had not alleged that that officer was responsible for those decisions. The ET noted that the added claims had arisen from oral evidence of the officer concerned owning up that the two decisions had been his.  It decided that it would be “wholly inequitable not to allow an amendment to reflect the case which the respondents were themselves putting forward”.

In the EAT the MPS contended that the ET had misapplied CLFIS because the amended case that succeeded had not been put (or adequately put) in the cross examination of the relevant witnesses. In considering MPS’ appeal the EAT referred to Browne v Dunn [1893] 6 R 67 HL which forms the principal source for the proposition stated in the 18th edition of Phipson on Evidence, at 12-12, that

“[i]n general a party is required to challenge in cross-examination the evidence of any witness of the opposing party if he wishes to submit to the court that the evidence should not be accepted on that point”. 

However, as the EAT noted, Phipson comments that “the rule is not an inflexible one” and that failure to put a point to the witness “may be most appropriately remedied by the court permitting the recall of that witness to have the matter put to him”. Browne had been the subject of an exegesis in North Cumbria University Hospitals NHS Trust v Dr S M Saiger, UKEAT/0276/15/LA, 17 July 2017, at paragraph 80ff) where HHJ Hand QC was “inclined to treat the rule as one of evidence and practice rather than law”.  In Williams v Solicitors Regulation Authority [2017] EWHC 1478 (not an employment case) Carr J had noted that in modern litigation the parties have more advance written material than was formerly the case and the likelihood of springing of an unfair surprise at trial is much reduced by the expansion of pleadings, written witness statements and lists of issues. 

The EAT commented that the MPS and the ET had more materials stating the nature of CI Denby’s case than the defence and the jury would have had in the era in which  Browne was decided. Finally the EAT referred to Chen v Ng (British Virgin Islands) [2017] UKPC 27 (again, not an employment case) in which Lords Neuberger and Mance said at paragraphs 54-55,:

“54. … It appears to the Board that an appellate court’s decision whether to uphold a trial judge’s decision to reject a witness’s evidence on grounds which were not put to the witness must depend on the facts of the particular case. Ultimately, it must turn on the question whether the trial, viewed overall, was fair bearing in mind that the relevant issue was decided on the basis that a witness was disbelieved on grounds which were not put to him.

55. At a relatively high level of generality, in such a case an appellate court should have in mind two conflicting principles: the need for finality and minimising costs in litigation, on the one hand, and the even more important requirement of a fair trial, on the other. Specific factors to be taken into account would include the importance of the relevant issue both absolutely and in the context of the case; the closeness of the grounds to the points which were put to the witness; the reasonableness of the grounds not having been put, including the amount of time available for cross-examination and the amount of material to be put to the witness; whether the ground had been raised or touched on in speeches to the court, witness statements or other relevant places; and, in some cases, the plausibility of the notion that the witness might have satisfactorily answered the grounds.”

Summarising the case law the EAT in Denby said that the issue  was whether the outcome of the trial was fair in the light of

  • the parties’ pleaded cases
  • the written evidence
  • the list of issues (as agreed and then amended) and
  • the conduct and course of the trial including the questioning of witnesses, and
  • the matters set out in paragraph 55 in Chen v Ng. 

In the hearing of a case such as that brought by CI Denby the context also included s 136(2) and (3) of the Equality Act (the burden of proof provisions), pursuant to which it was for the MPS to advance a positive explanation of its conduct unless it was willing to risk not doing so and therefore  leave CI Denby the option of relying on section 136(2) to the extent that the conduct remained unexplained.

The EAT gave detailed consideration to the way the case had originally been pleaded, the MPS’s response, the matters raised in CI Denby’s detailed witness statement, the witness statements served on behalf of the MPS and the cross examination of MPS’ witnesses.  The EAT stressed the context of the expectation that the MPS would be likely to proffer a full and frank explanation of CI Denby’s treatment and the ET’s conclusion that the MPS had not in fact proffered that explanation. That lack of transparency was “consistent with a desire to disguise the influence of senior decision makers …”  All in all CI Denby’s case had been adequately put.

The EAT said that the CLFIS principle needed “careful handling”. It agreed with the observation of CI Denby’s counsel that

“the CLFIS decision should not become a means of escaping liability by deliberately opaque decision making which masks the identity of the true discriminator.  Where a claimant is for good reason unable readily to identify which individual is responsible internally within the employing organisation for an act of discrimination, the claimant may, as this case demonstrates, sometimes be permitted to amend during the hearing once the correct person is, or persons are, identified from the evidence.”

A tribunal should not allow an employer to hide behind its more junior officers taking responsibility for decisions dictated to them by invisible senior officers. 


The burden of proof provisions in the Equality Act, which played an important part of the context for the EAT’s approach to MPS v Denby, are not reproduced in the relevant provisions of the ERA relating to detriment inflicted on the grounds of the making of protected disclosures.  But section 48 (2) ERA provides that on a complaint that such a detriment has been inflected it is for the employer to show the ground on which any act, or deliberate failure to act, was done. In practical terms that will involve showing which employee or employees of the employer did the act complained of.  The context in which the CLFIS “problem” will be approached by ETs in detriment claims is therefore very similar to that of discrimination claims.

More generally, and as noted above, in CLFS v Reynolds itself Underhill LJ had recognised the practical difficulties that the conclusions of the Court of Appeal might cause. Whilst employment tribunals will be reluctant to allow claimants to conduct such litigation on a scattergun basis and will expect a positive case to be advanced where possible as to which individual or individuals were party to the acts of which complained is made, they will also expect respondents to be forthcoming by way of pleading, disclosure and witness statements as to:

  • how and when the relevant decisions which are complained of were reached and
  • which individuals or individuals took those decisions/contributed to them.

If the situation remains opaque at trial, then arguments that the case was not properly put to the employer’s witnesses at a stage before the position was made clear, are likely to fall on stony ground. However those acting for claimants will be expected to formulate and articulate a case – however late in the day the material for that case emerges – so that there is clarity as to what the ET is being asked to decide.

Related Members
Shortlist Updated