Painting a picture of the “close-connection” test – a critical evaluation on whether Mohamud has given a liberal brush of justice.

 -  -  5

Spread the love


When evaluating the law, it is often important to try and view it as snapshots, but like photographs, it is important to try and view it through the lens of hindsight to give context. Plunkett, for instance, maintains that:

‘despite the difficulties with the reasoning in Mohamud, it could…be said that the result provides considerable clarity in relation to how the close connection test is likely to be applied …. In particular, an extremely liberal understanding of the “close connection” test, it is now difficult to conceive of many circumstances that will fall outside it. Indeed, finding…a “frolic of their own” are now bound to be few and far between.’[1]

Plunkett’s quote holds many truths regarding Vicarious Liability (VL), summarizing the general state, as a whole, accurately. However, it falls short in that it fails to distinguish the nuances of how the close connection test is applied. Instead, the non-static nature of Vicarious Liability has instead shifted focus following Mohamed: moving from a question of when it is applied to how. The answer to this lacks clarity as demonstrated by case law.[2]  This stems from the difficulties associated with the close connection test which remain around VL and as such, VL remains problematic, albeit purposely so in an effort to create a balance between certainty and fairness.

VL is composed of two elements: The relationship between the employer and the defendant, and whether that tort is closely connected to that employment relationship.[3] This close connection test will be our primary focus.
Therefore, we will look at the basis for Vicarious Liability, observing the development driven by public policy issues and societal influences. There will also be an analysis on whether the problems underpinning the test remain today and whether frolics are a thing of the past.[4]  Policies will be seen as both societal policies and William’s four pillars underpinning tort which drives these policies.[5]

Ultimately, the non-static nature which perpetuates Vicarious Liability is relatively unique for a variety of reasons. It emphasizes fairness over certainty to a greater degree than other areas of law with an intentional design that focuses on appeasement, justice, compensation and deterrence.[6] While the close connection test can be argued to be flawed, these flaws are mostly intentional and necessary in order for VL to be fulfilled in the way the courts see fit. While the direction undertaken by Mohamud may have caused some confusion, academics believe this to be necessary to maintain the status quo of fairness and compensation. [7]

The policy reasoning dictating Vicarious Liability.

The purpose of Vicarious Liability can be grounded in public policy and tort theory, with the four pillars suggested by Williams: Appeasement, justice, compensation and deterrence.[8] It is that reasoning which Lord Phillips believed VL was justiciable, claiming to be “fair, just and reasonable” made it of a tortious nature.[9] Phillips’ rationale of “policy reasoning” provided certain criteria are fulfilled highlights that there are many factors that drive VL. Those 5 criteria: insurance availability, relationship between the tort and work, employer’s control and risk creation have over time expanded, explaining Plunkett ‘s view of a ‘liberal understanding’.[10] However, precedence dating from the 1700s show where two apparently innocent parties are involved, the burden of loss being placed on the master (employer).[11] This demonstrates the roots of reasoning being guided from social policy and to create a fair, just and reasonable outcome for the claimant who might otherwise lack sufficient compensation.

While the idea of master-servant is arguably anachronistic, Holt’s belief of social convenience to pin the tort to ensure justice justifies VL, even if seen as “rough”.[12] Holdsworth acknowledges this approach believing that VL has muddied the water with confusion that is intrinsically found when compromising certainty with fairness.[13] While Salmond acknowledges this master-servant is less relevant today, the policy rationale continues for employees. This confused approach has been partially intentional due to the case-by-case basis of Vicarious Liability and because, Glofcheski states, a strict test with little flexibility would be unable to reach fairness and justice.[14]

Development of the close connection test.

The close connection test was borne out of an attempt to create a more flexible test, as the Salmond test was found to be too inflexible with illegal acts, becoming “inadequate” for its task.[15]

The law’s difficulty in interpreting the Salmond test to cover illegal acts of sexual abuse led to an adaptation according to Lister, which asked whether the activities could be “closely connected” to the employment.[16]  This purposely broader remit allowed for this flexibility and as long as there was a meaningful connection between the tortious activities and the nature of the employment, the employer could be liable. Additionally, a lower threshold is demonstrated by the fact that factual closeness, such as time in relation to the employment hours, and location of the act leads to vagueness and open-ended decisions, something criticized by Desmond. [17]  Lister’s trend of widening application can be argued to be the stepping stone that allowed Mohamud and consequent cases like Cox to carry forward, with Cox stating that the Law of Vicarious Liability “has not stopped”. [18]

Mohamud makes it clear that there must be a real connection, often factually based, between the terms of employment and the tortious actions themselves, however, there are some obvious loose applications in order to reach the outcome.[19] While, the outcome is justiciable in that there was a connection to be found as his job’s remit, i.e. customer service and interaction could overlap with his abusive behavior, this was a liberal interpretation designed to serve the justice that tort protects. It was stressed by Toulson that Lister was still the approach and there was no need to change the wording to create further clarity. [20] Instead, this allows for flexibility, finding justice on a case-by-case basis, something that some academics have found necessary – for the test to remain effective, it must be flexible and liberal.[21] Mohamud indicates that to ensure justice, employers are held to a greater degree, liable for their employee’s actions than before while also continuing the rationale of Lister.[22] As such, the doctrine of close connection is consistent, albeit liberally applied and dictated by policy designed to seek fairness and justice, at the cost of greater certainty.

This liberal application has been acknowledged within Mohamud as non-ideal but the best practical solution to ensure a fair result for the claimant. Crowther points out that societal changes led to the broader application of the test, aided by its nature.[23] This can be seen by various cases and the evolution of the close connection test in relation to how societal and employment arrangements have changed from the master-servant analogy, to Various Claimant’s discussion on whether employment necessitated wages. [24] Instead, it was decided other factors could be relied upon to suggest an employee-like relationship. This suggests Plunkett’s point of a difficulty “conceiving” circumstances that fall outside the test is well-founded.[25] The nature of the test, the liberal approach and societal policies dictating its applications mean that the changing nature of employment and tort would always result in a liberal application of the test, something merely reaffirmed by Mohamud.[26] Bell summarizes this as rather than the law or test remaining constant, it is society which as changed hence VL’s evolution.[27]

Ultimately, Bell’s belief that the test now works in a much greater range than prior to Mohamud can be demonstrated by the slew of cases following it, such as Bellman or James-Bowen and Ors.[28]

Problems with ‘The Test’.

While Mohamud can be said to have increased the scope of usage, there does remain an issue of clarity. Phillip’s belief in Various Claimants that the test is merely accepted with little to no explanation of how to find the connection is continued by Mohamud lacking even a clear definition of what exactly close-connection is. [29]

While the vagueness has allowed the test to be used where it might have been rejected in pursuit of fairness, Mohamud’s use of public interaction means links in other cases are contentious. Bellman held the company liable for the director attacking an employee at a Christmas work party.[30] Cox relied on the actions of the prisoner and its ‘role’.[31] There appears to be an emphasis on the connection requiring some factual points such as location, both physically and on a timeline. By this logic, the location of a prisoner and the activities within a prison to justify close connection is debatable.

These examples demonstrate how tenuous the ‘connections’ can be when finding the employer liable and demonstrate how policies have shaped this test. While this broadness can be unfair, this is potentially a purposeful continuation of Lister and Mohamud and necessary for justice.[32]

The close connection test does have certain situations where public policies have dictated boundaries to application however, namely cases involving the NHS, Police or other public bodies. Razumas held that the Ministry of Justice could not be found liable despite acknowledging the overlap in assessing both duty of care and VL.[33] This same case also found that the NHS were not subject to the same standards as the general public were, with a more specialized scope of employment dictating their close connection[34].

DSD suggested that the Police had their own standards, with Lord Kerr stating “only obvious and significant shortcomings”” would lead to VL claims and investigations, a far cry from ‘liberal understanding’.[35]

Hyde believes that DSD and Michaels allowed for more police claims but highlighted the difficulty in what is exactly required to establish a close connection. [36]

This is followed by the confusion on Morrisons, who were held liable for an action undertaken outside of company working hours.[37] Overall, tenuous links amid confusion is a key issue, and although it is seen as justifiable due to policy, it remains an issue.

Further reflections indicate that while the test is incredibly broad, there also remain clear indications on when the test would fail. This is purposeful, allowing justice to be done, and something Giliker [38]acknowledges befits the nature of tort, being fairer and a more justiciable test than a firmer test with stronger boundaries might otherwise allow. Mohamud and subsequent cases show however that the liberal nature has caused issues with predictability. While this may cause confusion in future cases, this an acceptable compromise to balance fairness with certainty.

Another issue is the stance of the test, which favours claimants to an incredibly strong degree. While Cox states that employers insuring themselves to prevent liability, are “not liable because they insured themselves”, [39] Morrisons appears to contradict this, as the courts, despite acknowledging they would not reasonably have been expected to prevent the breach, remained liable due to their insurance and its relationship with the risk.[40] This insurance and risk allocation have led to contradictory stances based on policy, specifically fair compensation. Employers, if liable, are in a better position to undertake the burden of compensation compared to the employee who may not be able to afford it themselves as per Various Claimants.[41] This can be seen as an unfair onus on employers, and while it is not the primary factor, employers appear to be more liable simply because they can better deal with the risk financially. Morgan acknowledges this issue,[42] instead suggesting that foreseeability should be a factor in deciding limits. Morgan, however, fails to consider the reasoning for the doctrine – social justice. By making cases stricter, it may unfairly penalize claimants. While Morgan may view the pro-claimant stance of Vicarious Liability as a problem, this approach is necessitated to ensure relative justice and compensation, the fundamental principles of tort law.[43]

Bell has led concerns that Mohamud has opened the floodgates, indicated by the numerous cases which have cited it.[44] The liberal application of the test has led to cases such as Barclays where the bank was held liable for a medical practitioner’s sexual abuse, a continuation of various claimant’s own arguably tenuous connection.[45] Barclay’s acknowledges that changes in the nature of employment which permeates everyday life has meant the liberal approach of the test and increase in cases is inevitable. However, in the same way that Lister has been seen as a stepping stone; the same can be said of Mohamud. Despite the court’s attempts to use incremental approaches where possible, opening floodgates are inevitable in the quest to achieve fairness via flexibility. [46]

Within the test however, there remains a key exception: frolics. Frolics are actions which an employee undertakes that fall outside the scope of their employment, a definition following Mohamud was said to hold little value. Plunkett explains that this is in part from the difficulty locating them, [47] for instance, Rose, a pre-Mohamud case which overruled Conway, had an employee deliberately disobeying an employee’s command.[48] This was a tortious action as it was a furtherance of the employee’s duty. This was found not to be a frolic and a decision that was followed by Mohamud, despite its arguable link. Racial abuse is difficult to find in the remits of most employment contracts, and despite the court’s justification that the primary duty of the employee was public relations; this was what partially led to Mohamud’s decision.[49]

Changes to employment and the need for social justice reduces how easily a frolic can be found but also highlights that they still exist albeit for exceptional circumstances. As such a frolic remains a necessary ingredient, but as an exception seldom used due to these changes.

There does however appear to be contradictions demonstrated by Bellman and Warren.[50] In Bellman, it was stated that the Christmas party was merely an extension of work, the issue at hand was work-related and the director was trying to pull rank.[51] These were considered sufficient reasons to not consider it a frolic despite there being several hours between work and the event. While justified in terms of seeking fairness, this contradicts with Warren, where they were not continuing events due to a break in the chain by the police.[52] Regardless, Vicarious Liability was found via a different route, but in light of Mohamud and Morrison, frolics have obviously changed over the years giving Plunkett’s views vindication.[53] This part of the close connection test comes from the wider interpretation and leeway given, combined with the changing nature of how employment contracts work . While frolics now appear to be “few and far between”, this is again to ensure the fairness sought by tort – allowing those who have been wronged to seek justice or escape liability where most applicable. [54]


While there are clear problems with clarity regarding the close connection test following Mohamud’s liberal application, this has further opened the avenue of VL to the public. With the relaxation of the boundaries of the test, the courts can experiment using the plethora of new potential cases. It has been demonstrated that the changes in determining whether a frolic has occurred has allowed the claimant to seek both justice and compensation they might otherwise failed to find with an alternative rigid test. While this may result in unfavorable terms for employers, this deliberate impreciseness allows for social justice by maintaining the four pillars of tort and thus the floodgates should remain open. [55] 

[1] James Plunkett, ‘Taking stock of vicarious liability’ L.Q.R. 2016, 132(Oct), 556-562, 561.

[2] Paula Giliker, Vicarious Liability in Tort: A Comparative Perspective (CUP, 2010), 254; Mohamud v Wm Morrison Supermarkets plc [2016] AC 677.

[3] Lister v Hesley Hall [2002] 1 AC 215.

[4] Mohamud (n 2); Richard Glofcheski, ‘A frolic in the Law of Tort: Expanding the Scope of Employer’s Vicarious Liability’ (2004) TLR 1.

[5] Glanville Williams, ‘The Aims of the Law of Tort’ (1951) CLP 137.

[6] ibid.

[7] Mohamud (n 2).

[8] Williams (n 5).

[9] Various Claimants v Institute of Brothers of the Christian Schools [2013] 2 AC 1, [35].

[10] ibid; Plunkett (n 1).

[11] Joel v Morison [1834] EWHC KB J39.

[12]Jennifer Arlen, William Bentley Macleod, ‘Beyond Master-Servant: A Critique of Vicarious Liability’ (2005), New York University Law and Economic Working Papers; Limpus v London General Omnibus Co [1862] 1 H&N 526, 539; 158 ER 993.

[13] William Holdsworth, A History of English Law, vol 8 (1st edn, Methuen & Co, 1966), 546.

[14] Glofcheski (n 4).

[15] Lister (n 3).

[16] John William Salmond‚ The Law of Torts: A Treatise on the English Law of Liability for Civil Injuries‘ (1st edn, Stevens and Haynes, 1907); Lister (n 3).

[17] Ryan Desmond, ‘Close Connection and Akin to Employment: Perspectives on 50 Years of Radical Development in Vicarious Liability’ (2016) 56 Irish Jurist (NS) 239, 242.

[18] Mohamud (n 2); Cox v Ministry of Justice [2018] 1 All ER 1.

[19] Mohamud (n 2).

[20] ibid.

[21] Lloyd v Grace, Smith & Co [1912] AC 716, 736.

[22] Mohamud (n 2).

[23] Sarah Crowther QC, ‘Tort Guest Lecture on Robinson and Mohamud’, (University of Exeter, 20 March 2019).

[24] Joel (n 11); Various Claimants (n 9).

[25] Plunkett (n 1).

[26] Andrew Bell, ‘Double, Double Toil and Trouble: Recent Movements in Vicarious Liability’ [2018] JPIL 235, 247.

[27] ibid.

[28] ibid; Bellman v Northampton Recruitment Ltd [2018] EWCA Civ 2214; James-Bowen & Ors v Commissioner of Police of the Metropolis [2018] UKSC 40.

[29] Various Claimants (n 9).

[30] Bellman (n 28).

[31] Cox (n 18).

[32] Lister (n 3); Mohamud (n 2).

[33] Razumas v MoJ [2018] EWHC 215 (QB).

[34] ibid.

[35] Commissioner of Police of the Metropolis v DSD & Anor [2018] UKSC 11.

[36] Richard Hyde, ‘ The Role of Civil Liability in Ensuring Police Responsibility for Failures to Act after Michael and DSD’ (2016) 22(1) EJoCLI.

[37] Wm Morrison Supermarkets Plc v Various Claimants [2019] 2 WLR, [78].

[38] Giliker P, Vicarious Liability in the UK Supreme Court UK (Supreme Court Yearbook, Appellate Press LTD, 2016).

[39] Cox (n 18).

[40] Wm (n 37).

[41] Various Claimants (n 9).

[42] Phillip Morgan, ‘Certainty in Vicarious Liability: A Quest for a Chimaera’ (2016) CLJ 202, 205.

[43] Williams (n 5).

[44] Bell (n 26).

[45] Barclays Bank Plc v Various Claimants [2018] EWCA Civ 1670.

[46] Wm (n 37); Crowther (n 23).

[47] Plunkett (n 1).

[48] Rose v Plenty [1976] 1 WLR 141; Conway v George Wimpey & Co Ltd [1951] 2 KB 266.

[49] Mohamud (n 2).

[50] Bellman (n 28); Warren v Henlys Ltd [1948] 2 All E.R 935.

[51] Bellman (n 28).

[52] Warren (n 50).

[53] Mohamud (n 2); Wm ( n 37).

[54] Plunkett (n 1).

[55] Williams (n 5).

5 recommended
comments icon 0 comments
0 notes
bookmark icon

Write a comment...

Your email address will not be published. Required fields are marked *