HomepageCommercial LawPrivate LawPublic Law & Human RightsCriminal LawEU & International LawCareers


Have Irlen Syndrome, or need different contrast? Click the button below for options.

Background Colours


Enter you email address below to subscribe to free customisable article notifications.

Alternatively, click the button below for our various RSS Feeds (available journal wide, or per section).

Impossible to Bank on it: Vicarious Liability on the Move

Article Cover Image

About The Author

Ming Lu Ang (Regular Writer)

Ming is a second-year law student at the University of Birmingham. Through writing for Keep Calm Talk Law, she hopes to further her understanding in various areas of law and be kept up to date with the latest legal and commercial developments. Outside the law, she enjoys foreign language dramas and sketching.

© Håkan Dahlström

You can delegate authority, but you can never delegate responsibility for delegating a task to someone else.

Richard E. Krafve

In Various Claimants v Barclays Bank plc [2017], her Honour Justice Davies in the High Court commented that the law pertaining to vicarious liability has been on the move for a while. Indeed, as Samuel Cuthbert has reviewed in a previous Keep Calm Talk Law article, this is especially apparent after Mohamud v WM Morrison Supermarkets plc [2016], in which the employee tortfeasor’s foul-mouthed response and physical assault of the victim at the victim’s car were held to be within the ‘field of activities’ allocated to him by the employer despite his leaving of the employer’s premises to do what he did. The same can be said for Cox v Ministry of Justice [2016], where the prisoner tortfeasor was found to be an employee of the Ministry despite the non-commercial nature of the activities carried out, i.e. only in furtherance of the Ministry’s own interest due to the circumvention of incurring additional costs in employing staff by the Ministry itself.

More recently, Barclays Bank plc v Various Claimants [2018], discussed below, also follows this trajectory. However, the imposition of vicarious liability in Barclays Bank seems to be influenced particularly by the vulnerability of the claimant victims. This article argues that this is unjustified, as it risks blurring vicarious liability (VL) and non-delegable duty liability (NDDL).


In Woodland v Essex CC [2013], a case concerning whether the local authority could be found liable for negligence of an independent swimming contractor towards a child claimant in school, Lord Sumption explained that NDDL is predicated on an antecedent relationship to the victim. This positive duty to protect the victim against risks is so personal as to be non-delegable. As this imposes a higher care standard, the use of vulnerability discourse is apt. It highlights the victim’s reliance on the defendant, making it fair in deeming the latter to have assumed responsibility for the former.

However, vulnerability discourse in vicarious liability is unjustified. As made clear in Catholic Child Welfare Society and Others v Various Claimants [2012], VL is predicated on the nature of employer-employee relationships and enterprise liability. Employers, having chosen their employees, delegated duties to them, and having benefitted from the performance of such duties, should have calculated and accepted risks inherent in the nature of their businesses. Imposing liability on defendants for their employees’ acts can therefore be deemed fair. Moreover, not all customers are vulnerable, like the claimants in Mohamud and Cox. Therefore, making vulnerability discourse an integral rather than collateral aspect of VL risks blurring the two doctrines.

The Facts and Decision of Barclays Bank Plc v Various Claimants

In Barclays Bank, a group of claimants attempted to seek damages from Barclays Bank (BB) under vicarious liability for their being sexually assaulted by a doctor. They alleged that they were sexually assaulted by a doctor during medical examinations that were compulsory prior to being employed by BB. They argued that there was a relationship of employment, or that akin to employment, and that the assaults were sufficiently closely connected to that relationship. However, BB argued that the doctor was self-employed and was an independent contractor. Therefore, they could not be vicariously liable for his actions.

The High Court and the Court of Appeal both agreed that the dual stage test for vicarious liability was that expounded by Lord Philips in Catholic Child Welfare Society:

  1. Was there a true relationship of employer/employee or that akin to employment between the employer (D2) and the employee (D1)?
  2. Was D1 acting in the course of his employment when he committed the tortious act, i.e. is there a sufficiently close connection between the tortious act and the employment?

This was also reviewed not long ago in Mohamud and Cox. In determining whether the first stage has been satisfied, the Court of Appeal in Barclays Bank applied the five indicia which Lord Philips listed, and which were reaffirmed by Lord Reed in Cox:

  1. The employer is more likely to have the means to compensate the victim than the employee and can be expected to have insured against that liability;
  2. The tort will have been committed as a result of activity being taken by the employee on behalf of the employer;
  3. The employee's activity is likely to be part of the business activity of the employer;
  4. The employer, by employing the employee to carry on the activity, will have created the risk of the tort committed by the employee; and
  5. The employee will, to a greater or lesser degree, have been under the control of the employer.

The Court of Appeal in Barclays Bank then found that:

  1. The doctor was already deceased and there was unlikely to be any insurance cover in such a case;
  2. The medical examinations were carried out for the benefit of BB as they were a prerequisite for employment at the bank;
  3. The doctor was carrying out his medical examinations so that BB could recruit a sufficient workforce to carry out its business activity;
  4. BB had created a risk of the sexual assaults occurring by giving the claimants no choice over which doctor to see and dictating that the claimants, most of whom were teenagers, see the doctor at his home on their own, and by providing specific instructions on a pro forma form headed with the BB’s logo, including the executing of physical examinations such as a chest measurement;
  5. BB can still have control over the doctor’s actions despite the medical examinations occurring outside its premises as it is able to direct what the doctor does – the manner in which the doctor carries out the instructions being irrelevant.

Having established the first stage, the court found the second stage to be easily satisfied. Eventually, as it was fair, just and reasonable to impose liability, the Court of Appeal found BB to be vicariously liable for the harm suffered by the claimants.

Non-Delegable Duty Liability

In Woodland, Lord Sumption listed five indicia – underpinned by the defendant’s assumption of responsibility – necessary for NDDL to arise, including where the claimant is a patient and/or child, or is particularly vulnerable and dependent on the defendant’s protection against risk(s). Indeed, patients are generally vulnerable. Recognizing hospitals/clinics owe direct duties to provide non-negligent treatments reflects both reality and societal expectations. Especially when many services are contracted out and patients are unaware of this, such that VL is unlikely to arise, NDDL ensures compensation if negligence occurs. However, this may not accord with Montgomery v Lanarkshire Health Board [2015], which concerned the failure of the doctor to warn the claimant of the 9 – 10% risk of serious injury from shoulder dystocia should she deliver her baby vaginally. The tenor of Montgomery reflects that patients may not necessarily be vulnerable as they can and should have a right of informed consent. Nonetheless, viewing patients in such a light parts company with reality. Montgomery is exceptional. The claimant herself held a BSc degree and then worked as a hospital specialist for a pharmaceutical company. In fact, she was described by the Lord Ordinary in the Scottish Outer House as ‘a clearly highly intelligent person’. Furthermore, her mother and sister are both GPs. The background of the general populace is unlikely to be similar to hers and therefore more vulnerable. Vulnerability discourse therefore rightly stands centrally in NDDL. 

However, in Farraj v King’s Healthcare NHS Trust [2009], Dyson LJ held that non-delegable duties (NDDs) will not be owed to ‘out’-patients in screenings/genetics-testing. Here, the defendant NHS Trust had sent unsuitable tissue sample to the second defendant - an independent laboratory - for it to be cultured and tested for the possibility of the Jordanian claimant’s fetus possessing the Beta Thalassaemia Major gene. However, the court found that the NHS Trust did not owe an NDD to the Jordanian ‘out’-patient despite the second defendant’s non-communication of its doubts as to whether the sample contained fetal tissue.

Of course, even if the Woodland indicia have been observed, NDDs may not necessarily arise due to policy reasons making it unfair, unjust and unreasonable to impose any liability. However, Dyson LJ acknowledged patients’ vulnerability when placing themselves in the care/control of hospitals. Therefore, NDDs will fairly be owed to ‘in’-patients. However, ‘out’-patients are not necessarily less vulnerable. This can be discerned from Denning LJ’s view in Cassidy v Ministry of Health (1951), which concerned the negligence of surgeons that had rendered the claimant’s entire hand useless when he had only undergone operation for two fingers. Denning LJ stated that it should not matter whether the claimant knew nothing regarding the terms of how relevant staff were employed but only knew he was treated by staff whom the defendant had appointed and so the defendant must be accountable for how he was treated. Therefore, ‘out’-patients cannot be expected to hold different expectations.

As a result, Dyson LJ’s disregarding of the claimant’s (lack of) knowledge of to whom her sample was sent misses the point. As Nayer argues in ‘Outsourcing genetic and diagnostic services: a consideration of the principles for establishing a hospital’s non-delegable duty and why it matters’ (2011) 2 JPIL 61, it is not the claimant’s expectation itself that gives occasion to NDDL but how that expectation informs her vulnerability and whether it is reasonable for her to have relied on the defendant. Adopting such analysis aligns itself better with Lady Stacey’s view in S v Lothian Health Board [2009], a case which similarly concerned the sending of a mouthwash sample for screening of the Cystic Fibrosis gene to an independent contractor by the defendant. Lady Stacey was persuaded by the impression the defendant hospital gave to the claimant, that they would be the one analyzing the sample.

Vicarious Liability

Stage One: The ‘Akin-To-Employment’ Test

Stage one in establishing VL is whether primary tortfeasors are akin to being employed by the employers. Discussing vulnerability here risks expanding the notion of ‘employee’ too much, such that engaging independent contractors may now no longer be a defense. In Barclays Bank, vulnerability discourse which appears more apt for NDDs (e.g. victims’ youth and their lack of choice to the doctor) has implicitly influenced the courts in regarding the doctor as an ‘employee’ despite the claimants’ pleading of NDDL not actively being pursued.

Perhaps, Lord Reed’s dicta in Armes v Nottinghamshire CC [2016] discussing the potential overlap between NDD and VL has played an influence. This concerned whether the local authority was vicariously liable for the foster parents’ deliberate wrongful act of abusing the claimant or that it was personally liable to the claimant via NDDL. Lord Reed found no reason to impose VL where defendants have direct liability to victims for harm caused by third-parties. NDDL is thereby considered first, implying VL does not arise where NDDL does.

If this is true, no relationship between the claimants and BB (under which BB would have a positive duty to protect the claimants from sexual assault risks) would arise. The only route to find liability would be through expanding the notion of ‘employees’, since suing the already deceased doctor was pointless due to his death. Vulnerable and abused claimants ought to be sympathized with and compensated to secure some semblance of justice. However, the need to compensate them should be balanced against the impossibility of obtaining any insurance cover in such a case. Further, Lord Reed’s priority proposition is not conceptually sound. Indeed, Bell in his article, ‘Double, double toil and trouble: recent movements in vicarious liability’ (2018) 4 JPIL 235, at 245-6, highlights the usefulness of VL despite direct liability existing, i.e. where primary tortfeasors are liable in deceit and employers are liable only in negligence to claimants. It may also be easier (or only possible) to prove subordinates’ wrongs and their relationship to defendants, rather than independent wrongs of defendants. Also, despite Irwin LJ’s link of the expanded notion of ‘employees’ to modern employment structures’ changes in Barclays Bank itself (e.g. the rise of the gig economy), it ought to be questioned whether this is fair. It is now trickier to know the exact boundaries of VL’s expansion in work relationships.

Stage Two: The ‘Close-Connection’ Test  

Stage two for establishing VL is whether employees’ actions are ‘so closely connected’ with their authorized duties such that they may be fairly deemed as within their course of employment. The extent of close connection needed is explainable using the enterprise risk (e.g. the Court of Appeal’s reasoning (i)-(iii) in Barclays Bank), assumption of responsibility, and/or policy rationales. The assumption of responsibility rationale can be gleaned from Lord Hobhouse’s statement in Lister v Hesley Hall [2001], with whom Lord Steyn and Lord Millett concurred. This case concerned whether vicarious liability can be imposed on the school boarding house for the sexual abuse of boys by its warden:

The liability of the employers derives from their voluntary assumption of the relationship towards the plaintiff and the duties that arise from that relationship and their choosing to entrust the performance of those duties to their servant. Where these conditions are satisfied, the motive of the employee and the fact that he is doing something expressly forbidden and is serving only his own ends does not negative the vicarious liability for his breach of the ‘delegated’ duty.

On the other hand, the policy rationale can be gleaned from the influential Canadian case of Bazley v Curry (1999), which concerned whether the defendant could be held vicariously liable for a paedophile whom they had hired and had acted as a substitute parent to the claimant while in its residential care facility. In this case, MacLachlin J stated:

First, a court should determine whether there are precedents which unambiguously determine on which side of the line between vicarious liability and no liability the case falls. If prior cases do not clearly suggest a solution, the next step is to determine whether vicarious liability should be imposed in the light of broader policy rationales behind strict liability.

The broader policy rationales considered include the vulnerability of the victim to the wrongful exercise of power(s) by the employee and the scope of power(s) conferred on the employee in relation to the victim. However, opinions can differ regarding their significance in policy, as in the contrasting judicial opinions in Jacobi v Griffiths [1999], delivered by the same court, on the same day, and with facts similar to Bazley. This indeterminacy is unwarranted – given VL’s conceptual meaning.

Further, the assumption of responsibility and policy rationales necessitate a common factor previously largely irrelevant to VL – the customer’s position. By focusing on the customer’s position, NDDL’s language, like customer’s vulnerability, permeates VL. This risks unprincipled blurring of the two. 

For instance, this could be seen in Judge Seymour’s reasoning in the High Court in Mattis v Pollock  [2002], which concerned whether the nightclub owner was vicariously liable for the actions of its bouncer who left the nightclub, went back to his own flat, armed himself with a knife, and returned to stab the claimant who stood outside the nightclub. Although the Court of Appeal reversed the finding of the High Court that the nightclub owner was not vicariously liable for its bouncer’s action as it was an act of personal vengeance, the Court of Appeal did not explicitly address and reject Judge Seymour’s reasoning that an assumption of responsibility is necessary for establishing VL. This highlights the danger of focusing too much on the position of the customer. Whether employees were acting in the course of employment is therefore replaced by an inquiry into the scope of employers’ duty towards victims – where vulnerability discourse permeates. VL therefore shades into NDDL. This offers courts little guidance in determining the existence of NDDL or VL, and/or contentiously, to effect ‘palm tree justice’ like in Barclays.


The use of vulnerability discourse is less justified in VL as judicial enthusiasm in finding VL in novel cases has thus far not been matched by an equal clarification of its ambit. This evinces most clearly where NDDL and VL are likely to overlap. This is certainly regrettable. VL clearly plays an important role. Therefore, demarcating the two is advisable by avoiding vulnerability discourse in VL, lest one consumes the other.

For the latest articles straight to your inbox, you can subscribe for free. Alternatively, follow @KeepCalmTalkLaw on Twitter or Like us on Facebook.

Tagged: Commercial Law, Employment Law, Tort Law, Vicarious Liability

Comment / Show Comments (0)

You May Also Be Interested In...

Armes v Nottinghamshire CC: Vicarious Liability Spirals out of Control

21st Aug 2018 by James Smith

A Legal Headache: Negligence, Concussion and Rugby Union

13th Jan 2017 by Ben Cisneros

Escaping Daddy’s Dance Hall? The Impact of Brexit on Employment Law

25th Mar 2016 by Anirudh Mandagere

The Uber-Complex Law of Vicarious Liability

16th Mar 2016 by Keir Baker

Sufferers of work-related stress: doubly neglected

9th Sep 2014 by Hannah Larsen

Employee Shareholders - Would You Sell Your Rights?

15th Nov 2013 by Emily Clements

Section Pick February

Routier v HMRC: Redefining 'Charitable Purposes' For Inheritance Tax

Editors' Pick Image

View More


Keep Calm Talk Law: Moving Forward

3rd Sep 2019

Changing of the Guard: Moving Keep Calm Talk Law Forward

12th Aug 2018

An Anniversary or Two: Four Years of Keep Calm Talk Law

11th Nov 2017

Rising from the Ashes: The Return of Keep Calm Talk Law

18th Nov 2016

Two Years On, Keep Calm Talk Law’s Legacy is Expanding

11th Nov 2015


Javascript must be enabled for the Twitter plugin to function. Click below to visit us on Twitter.

Free Email Subscription

Subscribe to Keep Calm Talk Law for email updates, and/or weekly roundups. You can tailor your subscription on activation. Both fields are required.

Your occupation / Career stage is used to tailor your subscription and for readership monitoring.

Uncheck this box if you do not want to receive our monthly newsletter.

By clicking the Subscribe button, you agree to our privacy policy and terms of service. Please ensure you read these in full.

Free Subscription