This month’s decision in Brown v Colt Services UKEAT/00233/17/BA is an important decision applying the recent Konczak approach for the divisibility of psychiatric injury, and also a useful summation of the law on the recoverability of damages for loss of earnings where the Claimant has a workplace-provided income replacement insurance policy.

Divisibility of psychiatric damage

The divisibility of psychiatric injury – that is, the question of whether the court can properly apportion the blame for a claimant’s psychiatric problems between a defendant’s unlawful conduct and other non-tortious causes – continues to produce appellate decisions, this time through the employment tribunal system.

The Claimant, Mr Brown, had been employed by the Respondent from 2005 in senior IT roles. He suffered a breakdown of his mental health in March 2012, and then a more serious breakdown in April 2012, and had not returned to work since (although he remained employed by the Respondent). He succeeded at first instance in establishing that the Respondent had unlawfully discriminated against him on grounds of disability, and subjected him to harassment.

Mr Brown was clearly extremely unwell, and had not recovered even by the time of the EAT hearing in 2017 (there was an appeal on a separate ground about the defensibility of the ET’s decision regarding his likely recovery period). The ET considered evidence about the impact of non-work related factors relevant to his psychiatric health: Mr Brown had been bullied at school many years before; his mother had died in 2009; his daughter had recently been diagnosed with autism; and he had an unhealthy alcohol intake. He had, however, no prior history of developing a diagnosable psychiatric disorder. The expert evidence at trial was equivocal on these factors, with both psychiatric experts agreeing that some factors were probably “not significant”.

Nevertheless, the ET apportioned blame 70/30 between tortious and non-tortious causes respectively, relying on Thaine v London School of Economics [2010] ICR 1422, EAT, and the Respondent’s liability in damages was reduced accordingly.

The Claimant’s appeal relied on the case of BAE Systems v Konczak [2017] EWCA Civ 1188, which we have considered before. Konczak was decided after the ET had conducted the remedy hearing in Mr Brown’s case, but was of significance in relation to the apportionment of  psychiatric injuries. Prior to Konczak, there was an apparent contradiction between the approach in Thaine, which viewed apportionment generally more favourably, and that in Dickins v O2 [2008] EWCA Civ 1144, where Smith LJ was rather more sceptical about the circumstances in which an apportionment would be appropriate.

The Claimant submitted that Konczak constituted a more nuanced approach than Thaine, and should be preferred. It was also submitted that before apportionment was even considered, Konczak required the ET or court to ask: is this disorder divisible, or indivisible? It was contended that the ET’s failure to address this anterior question before going on to apportion the harm was an error justifying the appeal.

The EAT broadly agreed. Lady Wise (a senator of Scotland’s College of Justice, sitting alone in the EAT) considered that Konczak largely clarified the law from Sunderland v Hatton [2002] ICR 613 onwards. This clarification, however, did require an express consideration of whether, on the facts and evidence of the case, the Claimant’s psychiatric problems were truly divisible and therefore suitable for apportionment, or not. This was the best interpretation of Hale LJ’s statement in Sunderland v Hatton that such diseases “may well” be divisible, and of the requirement in Konczak to apportion harm “if possible”. On the facts, although there were some indications in Mr Brown’s case that there were factors pre-dating the Respondent’s discriminatory acts which might have led to his later problems, the ET’s statement that they were assessing the “contribution of the discriminatory acts to the Claimant’s overall condition” indicated that were adopting a very different approach to that approved in Konczak, with no consideration of the question of whether the harm was divisible, or not.

The case has been remitted back to the ET for a fresh remedies hearing on that specific point.

Income replacement insurance and damages for loss of earnings

Mr Brown’s case is also an important summary of the law in relation to the recoverability of damages for loss of earnings where an employee has the benefit of an insurance policy, provided through his workplace, which compensates him for some or all of his lost income when he is off work through sickness or injury.

In general, it is settled law that a claimant must give credit for income which replaces their lost earnings (e.g. through accepting other employment, though sick pay, or through the benefits system),  unless its source is either (a) benevolence, or (b) an insurance benefit which he has contributed to directly or indirectly. The rationale for the so-called “insurance exception” is that the Claimant is simply reaping the fruits of a policy to which he has contributed through his premiums (see Gaca v Pirelli General plc [2004] 1 WLR 2683).

The insurance exception does not apply where the premiums for the insurance policy have been provided by the employer, without any contribution by the employee (Haddock v Atos Origin IT Services UK Ltd [2005] ICR 277).

Mr Brown had the benefit of a similar policy through the Respondent, as part of a flexible benefits account. The minimum protection available was 50% of an employee’s salary, but the default level was 75%. When choosing from a “menu” of benefit options, an employee had the option of leaving this level of protection in place, or lowering the level of benefit in exchange for other benefits, or extra salary. Mr Brown had left his income protection insurance level at 75% when choosing his benefits package (he had made other selections in respect of other available benefits). The ET considered that, applying Haddock and Gaca, Mr Brown had contributed to this policy to the tune of 25% (i.e. the difference between the “compulsory” level of benefit and the level actually obtained through his flexible benefits options), and so required him to give credit only of 50%, not 75%, against his past and future loss of earnings.

The Respondent’s appeal on this issue was refused. The Respondent argued that it was wrong to equate personal insurance with a choice from a fixed menu of flexible benefits, all of which were supplied by the employer. The EAT did not agree: what was required by Gaca and Haddock was that there was a contribution, either direct or indirect, from the employee, and this meant a provision of some form of consideration for the benefit received. Although Mr Brown had left his protection at the default level, he had foregone higher wages (or other benefits) by doing so. By considering the authorities and a number of Canadian cases on the same point (most notably Cunningham v Wheeler [1994] 113 DLR (4th) 1) the EAT held that Mr Brown had made an indirect contribution to the insurance scheme, and that this was sufficient to justify a finding that he did not have to give credit to the extent that the insurance was paid in respect of that contribution.