The vicarious liability of sports governing bodies and competition organisers

Abstract

This is the first work to explore the possibility of holding sports governing bodies and competition organisers vicariously liable for the tortious behaviour of athletes that compete under their auspices. In contrast to other scholarly contributions on vicarious liability in sport, this paper examines the scope of responsibility for athletes in individual sports (as opposed to team sports). It begins by drawing upon the recent tribunal proceedings between professional cyclist Jess Varnish and British Cycling to analyse the employment status of government-funded individual athletes. In calling for a contextual and policy-sensitive approach to the definition of an ‘employee’, this paper argues that certain normative and theoretical considerations ought to be granted more or less weight depending on the particular legal issue animating the dispute. Thereafter, and with one eye on the overly intrusive regulatory provisions found in sports such as tennis and golf, this paper also demonstrates that the vicarious liability of governing bodies and competition organisers could equally be extended to cover the tortious conduct of non-funded individual athletes. In making these claims, it is demonstrated how a sport-specific application of the doctrine may help to teach us a few broader lessons about vicarious liability more generally.

Introduction

The doctrine of vicarious liability has long been recognised as a fundamental aspect of tort law. It stands for the proposition that one party will be strictly liable for the harmful conduct of another, so long as there is a close connection between the injury and the wrongdoer's relationship with the defendant. It is most commonly used to hold employers responsible for the actions of their employees, although recent case law has established that vicarious liability may also be imposed for relationships that are merely ‘akin to employment’.¹

This development was seemingly intertwined with a judicial appreciation of the various theoretical rationales for vicarious liability. As evidenced by Lord Phillips’ five-factor test in Various Claimants v Catholic Child Welfare Society,² several justifications for the doctrine can be identified. These include, for instance: loss spreading and deep pockets (which suggest that liability ought to be imposed on the party who is best able to financially bear the loss); control (which maintains that employers are usually in the best position to prevent the harm); risk (which highlights that employers ought to be held responsible for any inherent or foreseeable harm that flows from their enterprise); and fairness (which indicates that those who seek to benefit from a particular activity should also be held accountable for any losses that such activity causes). The final two justifications might be said to be two overlapping formulations of the wider concept of enterprise liability.³

The argument propounded in this paper is that an application of these theories justifies the imposition of vicarious liability on national governing bodies (NGBs) and competition organisers for the tortious behaviour of athletes in so-called individual sports (such as tennis, boxing and golf). Of course, one might suggest that such an examination of theory is erroneous following the recent case of Barclays Bank v Various Claimants. There, Lady Hale relegated the use of theory to all but the most ‘doubtful’ of cases, and she suggested – in line with the judgment in WM Morrison Supermarkets plc v Various Claimants ⁴ – that we ought to adopt a more principled, incremental approach that derives assistance from previously decided cases in this area of law.⁵ However, in contrast to some scholars who have suggested that Barclays may finally have halted the ever-growing expansion of vicarious liability,⁶ I am somewhat sceptical as to whether the Supreme Court's judgment in Barclays has had the desired effect. After all, the various possible interpretations of the term ‘doubtful’ indicate that theory could, in fact, continue to play a significant role in future vicarious liability cases. In this regard, it is notable that, following Barclays, we see barristers regularly submitting (and judges subsequently accepting) that a particular case is ‘doubtful’,⁷ and many judges also still seem keen to continue referring to Lord Phillips’ five-factor test when justifying their conclusions.⁸

Moreover, one perhaps only needs to briefly peruse the existing work on sporting vicarious liability in order to express some doubt as to whether vicarious liability is applicable to the individual sporting context. Indeed, the existing scholarly analysis on sporting vicarious liability has been predominantly, if not exclusively, directed towards the responsibility of employers in team sports.⁹ Whilst Anderson observes that vicarious liability for negligent on-field acts in this context has reached an almost ‘presumptive, uncontested status’,¹⁰ there has been an alarming lack of discussion in relation to the appropriate scope of responsibility for individual athletes. This is also reflected by the fact that case law on sporting vicarious liability – in both the UK and overseas – has been centred entirely around team sports such as football,¹¹ rugby union¹² and basketball.¹³

Now, it must be appreciated that there is perhaps good reason for this: English law has long recognised that professional athletes are employees of the clubs they represent.¹⁴ In contrast, the classic position (at least for the purposes of employment law) has generally remained that athletes in so-called individual sports are independent contractors working on their own account.¹⁵ The purpose of this paper is to challenge that assumption by examining, from a theoretical standpoint, how the overly intrusive regulatory provisions of many NGBs and competition organisers could lead to them being held vicariously liable for the athletes that compete under their auspices. This is both a descriptive and normative argument based on the first necessary element of any vicarious liability claim: the establishment of a sufficient relationship.

To be clear, there is nothing particularly novel in claiming that NGBs ought to be held responsible for on-field injury. This is illustrated by Vowles v Evans and Welsh Rugby Union, a case in which the Welsh Rugby Union was held vicariously liable for the negligence of one of its appointed referees.¹⁶ The point raised in this paper, however, is that this liability has yet to be extended to cover responsibility for the tortious behaviour of athletes. Whilst James notes that the liability of NGBs is the ‘least explored area’ of sports-related personal injury claims,¹⁷ he does highlight that the law on sports torts ‘continues to extend legal liability to new contexts and new defendants who had not previously considered themselves to be at risk from litigation’.¹⁸ As such, it may only be a matter of time before an innovative claimant looks to take advantage of the solvency of a governing body or competition organiser in their pursuit of compensation.

With this in mind, it is worth noting that this paper is separated into two (largely overlapping) parts. The first relates to vicarious liability for government-funded individual athletes. In contributing to the growing debate on the relevance of other areas of law to the determination of employment status for vicarious liability, I draw upon insights from the recent tribunal proceedings between professional cyclist Jess Varnish and British Cycling to argue for a contextual and policy-sensitive approach to employer liability. Thereafter, and in the second part of the paper, I examine how the theoretical and normative rationales for liability in the funded context can equally be transposed to justify vicarious liability for non-funded individual athletes. In this regard, I utilise professional tennis and professional golf as two instructive examples. In making this argument, I also endeavour to demonstrate how an application of the doctrine to the individual sporting context may help to teach us some broader lessons about vicarious liability more generally.

1. Vicarious liability for funded individual athletes

The reference to the ‘funded’ athlete here refers to those professional sports participants who are funded by an NGB in order to help them fulfil their athletic and medal-winning potential. With the assistance of funding from UK Sport, NGBs are able to operate a World Class Performance Programme (WCPP) in their respective sports.¹⁹ Those athletes selected for a WCPP will enter into a Performance Athlete Agreement (PAA) which imposes certain obligations on the individual participant (such as, for example, behavioural standards or restrictions relating to their image rights).²⁰ In return, NGBs provide a wide range of benefits – including the provision of world-class coaching, sports science advice and access to high-tech equipment and facilities²¹ – which are estimated to be worth up to £60,000 per athlete per annum.²² Alongside this, UK Sport also directly funds (primarily through National Lottery income) the living and personal sporting costs of WCPP-initiated athletes through the guise of an Athlete Performance Award (APA).²³ This is awarded on the basis of both means-testing and athletic potential, with athletes at the so-called ‘Podium level’ usually receiving an annual £28,000 tax-free grant.²⁴ Around 1,100 of the UK's leading athletes in a variety of individual sports benefit from such investment,²⁵ with many participants (such as those in athletics or cycling) reliant on the funding for much of their professional careers.²⁶

Unfortunately, and perhaps due to the desire for medals and national success, many NGBs have adopted a domineering approach over their athletes that might be said to imitate the role of an employer instructing an employee. Despite protestations – in both the UK²⁷ and USA²⁸ – that the agreement between an athlete and their respective sporting body only gives rise to independent contractor status, the courts have frequently stressed that they will look to the ‘reality of the situation’ rather than simply defer to the label used by the parties themselves.²⁹ This is a particularly important exercise for many sports-related employment disputes because, as Schwab highlights,³⁰ most NGBs can exert monopolistic power over athletes on a ‘take it or leave it’ basis,³¹ and it is unlikely that an athlete would be able to compete under the auspices of an NGB without adhering to their ‘legally controversial’ contractual demands.³² Consequently, it is little surprise that some funded athletes have sought to challenge their legal status as independent contractors. The most relevant case in point here involves the recent legal proceedings between professional cyclist Jess Varnish and British Cycling/UK Sport.

(a) Varnish v British Cycling Federation

The crux of this case rested on the true reason for Varnish being dropped from British Cycling's WCPP in 2016. Whilst the governing body maintained that the decision was performance-related, Varnish contended that the underlying reason related to both her criticism of certain coaches at British Cycling, and the misogynistic comments of its former technical director, Shane Sutton (who allegedly told Varnish to ‘go and have a baby’).³³ However, before she could pursue a claim for unfair dismissal or sex discrimination, she first had to prove that she was either an employee or worker of British Cycling or UK Sport under section 230 of the Employment Rights Act 1996. The Employment Tribunal (ET) at first instance concluded – much like it had done almost 20 years earlier in relation to a similar claim by former track cyclist Wendy Everson³⁴ – that Varnish was neither an employee of, nor a worker for, the governing body.³⁵ Likewise, the absence of any ‘day-to-day relationship’ between the claimant and UK Sport similarly precluded employee and worker status here.³⁶ In her appeal case that dropped the (presumably weaker) claim against UK Sport, the Employment Appeal Tribunal (EAT) reaffirmed the ET's decision on the basis that the first instance tribunal had properly and reasonably directed itself as to the relevant legal and factual principles.³⁷ The key features of this litigation were the so-called three ‘irreducible minimum’ requirements needed to establish an employment relationship: mutuality of obligation; personal performance; and control.³⁸ Given that Varnish did not challenge the initial judgment on the two latter requirements, it may reasonably be concluded that Judge Ross’ sentiments on these two factors in the ET continue to constitute good law.

On the issue of control, the ET considered this to be a ‘significant feature’, in that ‘many aspects of [Varnish's] life including what she ate how, when and where she trained were closely controlled by British Cycling’.³⁹ With reference to the PAA, Judge Ross also found that the NGB exercised control over the claimant's media image and appearances, her personal commercial work and use of social media, and when she could take time off.⁴⁰ Interestingly, the ET also found that Varnish was ‘integrated into [British Cycling's] organisation, working closely with her coach and wearing the team clothing’ at all events and training sessions.⁴¹ Although not every aspect of the purported employment relationship between the two parties was supported by the notion of control (such as the fact that Varnish could choose both her own coach and equipment if she so wished),⁴² it was clear that Judge Ross considered British Cycling's degree of control in this scenario as taking them beyond a mere regulator of the sport. Unfortunately for Varnish, however, this fact was outweighed by the absence of mutual obligations and personal service in her relationship with British Cycling.⁴³

In relation to mutual obligations, Mr Justice Choudhury in the EAT rejected the claimant's plea that the obligations under the PAA constituted work (and that the services simultaneously provided to her constituted remuneration),⁴⁴ and he reiterated the ET's findings that she was simply privy to a ‘contract where services are provided to [her], not the other way around’.⁴⁵ In other words, there was no obligation on British Cycling to provide work, nor was there a corresponding obligation on Varnish's part to ‘accept and perform the work in exchange for consideration, usually wages’.⁴⁶ This finding seemingly also influenced the view that Varnish failed to satisfy the personal performance requirement. On this issue, Judge Ross found that, whilst Varnish certainly could not substitute another rider to carry out her obligations under the PAA,⁴⁷ she was not personally performing work provided by British Cycling. Rather, she was simply ‘performing a commitment to train in accordance with the individual rider agreement’ in the hope that she would be selected for international competition.⁴⁸ Given that the determination of personal performance here hinged heavily on the so-called ‘wage/work bargain’, one might say that it was the lack of mutual obligations between the two parties that was fundamental in deciding the employment status of Varnish in this case.⁴⁹

Now, for our purposes, it is interesting to consider what might have happened in the alternative scenario if Varnish had not been dropped from Team GB in 2016, but instead had gone on to compete in future professional events whilst funded by British Cycling. Let us say that, during one such event, she negligently (or intentionally) injured a fellow competitor or spectator, perhaps in a similar manner to Miguel Angel Lopez when he punched a fan at the Giro d'Italia in 2019.⁵⁰ If the current decision in Varnish is also the benchmark for determining an employment relationship in the law of tort, then British Cycling would clearly not be held vicariously liable for this act. Nevertheless, it must be questioned whether the definition of an employee might – and indeed should – differ depending on the area of law in question. As I explore below, the generally received view now favours a context-specific approach to the definition of employment. However, there has been very little critical discussion as to what extent cases from employment law (and indeed other areas of law) might provide an instructive steer in determining the scope of vicarious liability. The following sections seek to fill in this gap in the literature, as well as hopefully shedding some light on the relevance of Varnish to the vicarious liability of NGBs.

(b) Employment law and vicarious liability: time for harmonisation?

Until fairly recently, it had long been assumed that the definition of an ‘employee’ was the same across multiple strands of the law (whether that be employment, insurance, intellectual property or tort law).⁵¹ As testament to this fact, many of the leading monographs and textbooks discussing the necessary relationship for vicarious liability refer interchangeably to labour law cases.⁵² However, recent judicial statements appear to now cast some doubt on this idea. In particular, Lady Hale in Barclays suggested that whilst it might be ‘tempting to align the law of vicarious liability with employment law’, it would ultimately be going ‘too far down the road to tidiness’ to do so in light of the differing contextual reasons in each domain.⁵³ This appreciation of context-specificity is echoed by Ward LJ's function-over-form approach in JGE v English Province of Our Lady of Charity.⁵⁴ As his Lordship explains:

If the case is one where an employee seeks a remedy against his employer, for example for unfair dismissal, then the case does require that the true relationship of employer/employee be established in order to found the claim … On the other hand, the remedy of an innocent victim against the employer of the wrongdoer has a different justification rooted, as we have seen, in public policy. The fluid concept of vicarious liability should not, therefore, be confined by the concrete demands of statutory construction arising in a wholly different context.⁵⁵

Now, it is worth highlighting that not every employment law scholar has subscribed to this viewpoint. Butlin and Allen, for instance, are perhaps the most prominent advocates of a harmonising approach to the employment relationship, and they suggest that ‘the law on vicarious liability and worker status should march hand in hand’ because they are ‘inextricably interlinked’.⁵⁶ In this, they promote a conflation of the employment and vicarious liability contexts in order to avoid a ‘damaging and undesirable’ risk of ‘regulatory dissonance’,⁵⁷ as well as to ‘ensure legal certainty for both employers and employees across different strands of the law’.⁵⁸ Pitt adopts a similar view, arguing that it would be ‘undesirable if different tests were to develop for identifying contracts of employment according to what was at stake’.⁵⁹ One of the key tenets of this view is that it helps to ensure ‘consistency between two interwoven strands of the law’.⁶⁰ However, if, as I suspect, the policies at play are oftentimes different in both the employment and vicarious liability contexts, then a lack of consistency between them becomes not just unproblematic, but also perhaps sometimes desirable. As I explore shortly, the notion of victim compensation is one policy goal that is arguably more pressing in the context of vicarious liability than it is in the context of employment law.

For such reasons, I find Butlin and Allen's view unconvincing, and this can perhaps be best illustrated by considering the mutuality of obligation requirement in both employment and tort law. More specifically, whilst the concept of mutual obligations has been described as the most ‘prominent’⁶¹ and ‘essential’⁶² feature of contemporary employment protection litigation, it is, I suggest, a largely irrelevant factor in the determination of an employer's strict liability for the torts of their employees. As Kidner opines, an analysis of the mutuality requirement is likely to skew our view of the employer-employee relationship ‘towards the demands of employment law and the policies embedded therein’.⁶³ In fact, the most recent judgment of the EAT in Varnish suggests an even more contextualised role for mutuality of obligation, with Mr Justice Choudhury highlighting that it may only be a useful criterion in employment law for cases involving intermittent working environments.⁶⁴ As such, it is suggested that the theory of control is a much more apt consideration for vicarious liability cases, and it should certainly not be viewed as possessing the same importance in this context as the concept of mutuality or other employment law-specific considerations (such as whether an individual is treated as an employee for tax purposes). This is so for two reasons.

First, it must be recognised that the concept of mutuality focusses solely on the relationship between the purported employer and employee. In contrast, and as Ward LJ in JGE referred to above, vicarious liability cases possess an added component: an innocent third-party victim. In this regard, the theory of control appears to be a far more useful factor to consider. Whilst it is true that the control is exercised by the employer over the employee, the existence or absence of such control has an important moral and normative impact on third parties. Three particular cases illustrate this point well.

The first is that of Home Office v Dorset Yacht Co Ltd,⁶⁵ a case in which the government were held liable in negligence for a failure to control the actions of several borstal boys. Whilst it is clear that the context of the proceedings was different in this case (the claimants preferring instead to base their claim on primary liability grounds),⁶⁶ Dorset Yacht is still of great importance in highlighting that ‘control imports responsibility’ when assessing the liability of a defendant for the actions of another.⁶⁷ Such comments also provide us with a useful framework for examining the relationship between employment law and vicarious liability: decisions in the former should only be instructive for the latter when the focus is on those cases, or parts of cases, which share the same underlying policies. Whilst Dorset Yacht indicates that control is one such policy that straddles both areas of law, mutuality of obligations does not. Likewise, deep pockets and loss spreading are fine examples of theories that are important for vicarious liability, but not particularly relevant for employment law purposes. The fact that many individual athletes (such as Varnish) are required to apply for funding may indicate that such athletes do not possess deep pockets, at least not vis-à-vis their NGB. However, whilst such an analysis is relevant for vicarious liability purposes, it is of far less normative importance in the employment context when deciding whether an individual was unfairly dismissed.

This argument can be sharpened by taking a second example from the facts of O'Kelly v Trusthouse Forte.⁶⁸ In this case, a hotel employed wine butlers on a regular – yet casual – basis, such that the hotel was not obliged to provide (and the staff were not obliged to accept) any work. When one of the butlers was fired, his claim for unfair dismissal was rejected by the Court of Appeal on the basis that his relationship with the hotel lacked the necessary mutuality of obligation. However, once we drop the ‘unnecessary baggage’ of additional employment law provisions such as the need for mutuality,⁶⁹ we can see how a vicarious liability claim might have succeeded on the facts of O'Kelly. As Kidner posits, in light of the ‘divergence of the needs of employment law and vicarious liability’, it would be entirely surprising if the hotel was not vicariously liable for one of their casual waiters negligently spilling wine over a patron.⁷⁰ It is likely that Peel and Goudkamp also had this case in mind when they correctly argued that there is something distinctly ‘odd’ about a scenario whereby an employer exhibits close control over a tortfeasor's manner of work, yet the employer is not held liable for the tortfeasor's acts simply because they had the option to decline him work at certain times.⁷¹ With this in mind, I can do little more than to echo McKendrick's view that the ‘test for the existence of an employment relationship should depend upon the legal question which is being asked’.⁷² In this light, we perhaps ought to treat cases such as Kafagi v JBW Group Ltd – which refer indiscriminately to purely employment law policies such as mutuality of obligation in determining the scope of vicarious liability⁷³ – with a degree of caution.

A third and final case law example illustrating the different policies at play in employment and tort law can be found in Sedley LJ's judgment in Dacas v Brook Street Bureau (UK) Ltd.⁷⁴ Whilst the employment tribunal in these proceedings had previously concluded that no employment relationship existed between the claimant and Wandsworth Council for the purposes of an unfair dismissal claim, his Lordship maintained that if the issue of employment had arisen in relation to a personal injury action – instead of the ‘more abstract question of contract law’ – a different result would have ensued.⁷⁵ In other words, had Mrs Dacas injured a third party by negligently leaving cleaning equipment in a dangerous area, it would be a ‘near-certainty’ that vicarious liability would befall the council, and that those advancing an alternative submission ‘could look forward to a bad day in court’.⁷⁶ Again, this illustrates the normative importance of third-party victims in vicarious liability cases, and it may also tentatively suggest that some judges are of the view that the right to bodily integrity is perhaps deserving of greater protection than the right not to be unfairly dismissed. Given Dagan's comments – which suggest that ‘our lives are divided into economically and socially differentiated segments, and each such ‘transaction of life’ has some features that are of sufficient normative importance… that justifies a distinct legal treatment’⁷⁷ – this may be a sensible distinction to make.

(c) Formulating a policy-oriented approach to vicarious liability for funded athletes

I have discussed in the previous section that the primacy of control in the context of vicarious liability is arguably justified in light of its normative significance in relation to third parties. To suggest that mutuality of obligations is equally as relevant here is, in the words of Posner, to allow that concept to become ‘unmoored from any plausible goal of employment’ law.⁷⁸ Nevertheless, we might also identify a second reason for our focus on control in this context: it overlaps with several other theories of vicarious liability in a way that mutuality and personal performance clearly do not.⁷⁹ This pluralistic point is an important one, as I have demonstrated elsewhere that a determination of vicarious liability based on multiple converging theories is far more justifiable than a finding based only on one theory.⁸⁰ This is predominantly because, as Giliker points out, theoretical analysis can help to ‘aid clarity’ by providing some substance to the ‘building blocks of vicarious liability’.⁸¹ Consequently, whilst the overwhelming scholarly and judicial consensus suggests that control alone ought not to be determinative as a test for employer liability,⁸² I believe that it could still be viewed as a prominent concern so long as its overlap with other relevant theories is made clear. This is largely in accordance with Morgan's view, when he suggests that control could still be a highly useful tool in helping us to distinguish between employees and independent contractors.⁸³

The most notable overlap here is that between control and the two forms of enterprise liability outlined in the introduction. We might recall that the notion of fairness – which has recently been heralded as the ‘most influential idea [of vicarious liability] in modern times’⁸⁴ – encapsulates the idea that an employer who benefits from a particular activity ought to simultaneously bear the burdens of that conduct.⁸⁵ According to Flannigan, the notion of benefit is intrinsically linked to the theory of control, in that ‘a person's ability to benefit in an equity or residual sense normally depends on whether or not that person controls the performance of the work’.⁸⁶ Morgan too makes a similar point when he outlines that ‘[a]cting on behalf of the employer, and control, also link to [the] wider notion of enterprise liability’.⁸⁷ Both control and benefit appear to overlap with the concept of integration, such that the terms ‘control’, ‘integration’ and ‘benefit’ can often be used interchangeably in most instances. Kidner makes the connection between the first two, when he highlights that the ‘degree of control exercised by the employer may well depend on the degree to which the “employee” is integrated into the activities of the enterprise’.⁸⁸ In this manner, Bell is correct to illustrate that control can be (and indeed has been in Tomlinson LJ's Court of Appeal judgment in Armes v Nottinghamshire County Council ⁸⁹) accorded ‘indirect relevance’ by assessing the integration of an activity into the defendant's enterprise.⁹⁰ Furthermore, various judges have demonstrated that integration is also closely interlinked with any benefits enjoyed by an employer. This was explicitly recognised by Lord Reed in Cox v Ministry of Justice,⁹¹ and Irwin LJ in the Court of Appeal in Barclays similarly outlined that the tortfeasor in that case was sufficiently integrated into the business activity of the defendant because his work was primarily done for the benefit of the bank.⁹²

Given that ‘control’, ‘integration’ and ‘benefit’ all appear to be cut from the same cloth, we might make the following observation: if an individual is subject to strict control by an organisation, it is likely that they are also integrated into that organisation's business activities and providing a benefit to that entity. It is no surprise, then, that an analysis of fairness appears to justify vicarious liability for many funded individual athletes. In Varnish, for instance, it was reported that the ‘ultimate goal’ for both British Cycling and its athletes was to ‘win medals for the British Team’.⁹³ Given that the benefit derived from the tortfeasor's activities need not necessarily be financial in nature,⁹⁴ one may simply point to the fact that success would ‘reflect well on the institution’ as one of the benefits received by British Cycling.⁹⁵ However, it is also evident that an NGB's failure to meet their annual medal and performance targets (as set by UK Sport) will likely lead to ‘savage cuts’ on the funding offered to that sport,⁹⁶ so it is arguably the case that British Cycling also reap a financial benefit from the high degree of control that they exercise over their athletes. Moreover, it was also noted in the EAT that the NGB were able to make use of Varnish's image ‘in connection with the promotion, publicity or explanation of the Podium Programme’,⁹⁷ again suggesting a profit-making benefit to both British Cycling and UK Sport.

In addition to fairness, we might also say that control overlaps to a significant extent with the risk-related formulation of enterprise liability. The inherent link between control and risk is illustrated once again by Flannigan when he states that the ‘ability to control is what enables the employer to take risks. When the employer has no control, he is not in a position to apply his risk set to the activity or operation’.⁹⁸ This seems applicable to the facts of Varnish, in that we might say that the existence of the PAA – which includes the package of benefits provided by the NGB and the concomitant obligations imposed on athletes – enables (and thus also increases) the risk of injury to others. The extent to which it is increased, however, perhaps depends on the type of act committed. Given that British Cycling only exercise control over (and benefit from) certain aspects of Varnish's life, it may be that vicarious liability would only be appropriate for acts intrinsically linked to that control/benefit. By way of example, we have seen that the NGB were able to exercise significant control over Varnish's use of social media, such that any tortious comments made by Varnish on Twitter could be fair game for vicarious liability. Similarly, given that British Cycling reap the (reputational and financial) benefit of medals from Varnish's performances, any negligent injury she causes during competition should also be susceptible to vicarious liability.

Contrast this with a scenario whereby Varnish assaults her coach in anger after a particularly bad performance at an event. Given that the NGB allow Varnish to choose her own coach, it is arguably far less appropriate to impose vicarious liability on British Cycling for this act. In this manner, we might say that something similar to the harm-within-risk rule from legal causation ought to operate here,⁹⁹ with the result that control, benefit and risk all need to overlap to some extent before vicarious liability is deemed justifiable. This point was recognised by Bell when he suggested that both ‘benefit and risk must be kept close together if an enterprise theory is to hold’.¹⁰⁰ With reference to the facts of Armes, he further elaborates that the benefit to the council in utilising foster parents in this case was in ‘running the child welfare system/discharging its duties, so this should then remain the focus of the risk creation point’.¹⁰¹ An application of vicarious liability to the individual sporting context appears to further reinforce the need to heed this advice.

With these points in mind, it is useful to step back and consider what lessons we might learn from a policy-oriented approach to Varnish and, more broadly, what this potentially means for the vicarious liability of many NGBs. It is arguable that, if we change the facts in Varnish to emphasise the potential negligence of the claimant (much like Sedley LJ did in Dacas), British Cycling ought to be vicariously liable for tortious harm caused by Varnish. Under the approach advocated here, control – already established as significant in the ET's judgment – would be afforded greater importance, with the mutuality and personal performance factors (which pointed in the opposite direction to control in the tribunal's decision) concomitantly being downplayed. This analysis is reinforced by the fact that control overlaps to a significant extent with enterprise liability, and both fairness and risk would seemingly also justify the imposition of vicarious liability for funded athletes in many scenarios. In addition, the existence of funding under both the PAA and APA may also illustrate that the NGB has deeper pockets than the individual athlete, and that they are also more able to adequately spread the loss of any damages award.

Now, this is not to say that Varnish was incorrectly decided, and nor does it mean that all funded athletes should now be classed as employees for the purposes of unfair dismissal. Rather, the policy-oriented approach advocated here allows us to recognise, as the Supreme Court did in Pimlico Plumbers Ltd v Smith,¹⁰² that an athlete can be an employee for one purpose, and self-employed for another. This is important, in that it helps us to avoid those doomsday arguments that are often associated with the provision of employment rights and benefits to funded athletes.¹⁰³ For instance, it has been estimated that, had Varnish been found an employee of British Cycling or UK Sport, many NGBs would have encountered ‘serious financial difficulties’, with one in five British athletes facing a funding cut.¹⁰⁴

2. Beyond funded athletes: viarious liability in other individual sports

In light of the prima facie case for the vicarious liability of funded athletes, we might take this analysis a step further and question whether a similar determination could equally apply to other professionalised individual sports where funding is absent. In particular, this requires us to examine whether the theoretical and contextual arguments applicable to funded athletes are equally as strong when applied to non-funded athletes. At first glance, the answer appears to be ‘no’. A simple deep pockets analysis, for instance, suggests that there is no need for vicarious liability at all in this scenario, as non-funded professional athletes are likely to be solvent and able to meet a significant damages award themselves. Consequently, when we see various tempestuous tennis stars engaging in potentially harm-causing activities – such as Novak Djokovic negligently hitting a line judge with a tennis ball at the 2020 US Open;¹⁰⁵ Nick Kyrgios hurling his racket into the crowd at Wimbledon in 2015;¹⁰⁶ David Nalbandian wounding a line umpire after kicking out at an advertising board in 2012;¹⁰⁷ and Juan Ignacio Chela spitting at his opponent Lleyton Hewitt in 2005¹⁰⁸ – we might justifiably conclude that there is no need to consider the potential liability of a governing body at all in these scenarios. After all, both corrective justice and deterrence seem to point towards imposing direct liability on the athletes, and the main compensation rationale for vicarious liability largely falls away when we consider the lucrative career of a professional tennis player.

However, this analysis may be criticised as under-inclusive. Imposing direct liability on the tortfeasor may only be a viable option for those athletes at the very apex of the sport who regularly compete for the biggest prizes in tennis. In contrast, a negligence claim against those athletes lower down the pecking order is far less feasible. In fact, according to a 2013 study by the International Tennis Federation (ITF), around 45% of the 13,736 professional tennis players earned nothing from the sport in that year,¹⁰⁹ and athletes placed outside of the top 200 ranking spots were unlikely to earn more than £40,000 in prize money over the course of those 12 months.¹¹⁰ As such, when we hear of stories such as those involving former world number 248 Harmony Tan – who accused a similar-ranked opponent of intentionally hitting her in the eye with a tennis ball during an ITF event¹¹¹ – we might conclude that the vicarious liability of the relevant tennis governing body is much more appropriate here. This is reinforced by the healthy financial status of many organisations that seek to regulate their respective sports. One example is that of the PGA Tour in golf. Despite its status as a non-profit organisation, the PGA Tour actually boasts an annual revenue in excess of US$1 billion.¹¹² This, it is suggested, feeds into a number of other relevant points here.

First, the PGA Tour is actually a competition organiser (rather than an NGB),¹¹³ and it may be that organisers of sports competitions could be just as susceptible to vicarious liability as governing bodies.¹¹⁴ Given that control over non-funded athletes is often exercised simultaneously by varying bodies, it might be that dual vicarious liability is sometimes appropriate in this context.¹¹⁵ This form of liability allows courts to find multiple employers strictly liable for the tort of an employee, and this seems to be an inherently useful tool for recognising that, in many scenarios, different employers may well possess ‘different levels of responsibility’ according to the degree of control they enjoy.¹¹⁶ Such dual liability could be particularly useful in sports such as boxing, where the so-called ‘alphabet soup’ of sanctioning organisations – which includes four different bodies overseeing six world champions in each of 17 different weight categories – undoubtedly complexifies the analysis.¹¹⁷

Secondly, tennis is clearly not the only sport in which it might be appropriate to hold a regulator vicariously liable, and it is for such reasons that I examine how this analysis might also be applied to the sport of golf. The potential for injury here is obvious, particularly to spectators. This is evidenced by the recent incident at the 2018 Ryder Cup, where an onlooker was blinded in one eye after being hit by Brooks Koepka's wayward shot.¹¹⁸ Whilst there was no evidence that Koepka was negligent in his drive, it would not be difficult to imagine – following the line of ‘reckless duffers’ cases such as Pearson v Lightning ¹¹⁹ and Phee v Gordon ¹²⁰ – a slightly different scenario in which negligence could be established. For instance, had it been established that Koepka was intoxicated when he took the shot – much like former professional golfer Rocco Mediate, who admitted that drinking whilst on PGA Tour courses was ‘normal’ for him¹²¹ – it may have been justifiable to hold the organisation responsible for this harm.

Third, in examining the potential scope of vicarious liability in these sports, it must be highlighted that my analysis is consistent with the Supreme Court's recent clarification of control in CCWS. There, Lord Phillips highlighted that control is now more about whether an employer can direct what an employee does (rather than simply how he does it).¹²² As such, the exercise conducted here is similar to that carried out by Morgan under his ‘dual axis’ approach to employer liability.¹²³ On this basis, he highlights various factors that go towards assessing both an employer's ‘day to day control’ and an employee's ‘discretion in role’. These include, for instance, prescription of: working hours and location; uniforms; disciplinary systems; the use of certain methods and equipment; and when breaks can be taken. An application of these factors to governing bodies and competition organisers in professional tennis and golf highlights both their control over, and benefit from, the athletes competing under their auspices, and this suggests that vicarious liability may sometimes be appropriate even for non-funded athletes. This conclusion is also consistent with Dabscheck's analysis, as he highlights how a more nuanced assessment of the restrictions imposed upon professional jockeys – such as limited control over their own attire, fitness levels, use of intellectual property rights and adherence to gambling and drug codes – suggests that they ‘should be properly regarded as employees and not independent contractors’.¹²⁴ Interestingly, this may have meant that, had negligence been established in the seminal ‘sports torts’ case of Caldwell v Maguire and Fitzgerald (which involved allegedly ‘careless riding’ by a professional jockey),¹²⁵ the British Horseracing Authority could have been held vicariously liable for this injury.

Fourth, it must be conceded that not all scholars would agree with utilising vicarious liability to hold NGBs and competition organisers responsible for harmful conduct. Some commentators (such as Gray, for instance) would likely believe that it is more appropriate to hold such organisations directly liable in negligence.¹²⁶ Given the need to prove fault for such an action, however, this may not be an adequate solution for those who are convinced by my pro-liability stance in this context. Consider, for instance, the aforementioned example involving Rocco Mediate: the PGA Tour would likely only be liable in negligence for this harm if it could be shown that they were aware of the tortfeasor's inebriated state, but failed to take action. In contrast, (strict) vicarious liability would still apply in such a scenario, irrespective of the organisations’ knowledge or any prior training they offered to the athlete.

Likewise, others may suggest that a non-delegable duty – an exceptional no-fault based form of primary liability – could be placed on sporting bodies to ensure that reasonable care is taken by participants in individual sports.¹²⁷ Now, whilst I believe that the non-delegable duty might be able to work alongside a pleading of vicarious liability (as occurred in a number of recent dental negligence cases),¹²⁸ the concept is not sufficiently developed or determinate enough, in my opinion, to completely replace a vicarious liability claim. This is evidenced by the vast uncertainty that continues to exist in relation to Lord Sumption's guidance on non-delegable duties in Woodland v Essex County Council.¹²⁹ Although his guidance was never meant to be ‘set in stone’,¹³⁰ it remains unclear as to whether any of the tests outlined by Lord Sumption would be satisfied in relation to the individual sports context.¹³¹ In this light, Giliker concludes that vicarious liability remains the ‘stronger and more predictable option’ for claimants,¹³² and it is perhaps for such reasons that the non-delegable duty was not raised at all in the recent Barclays litigation.

Finally, it may be that the discussion here could lead some NGBs to seriously consider imposing an obligation upon all athletes in their sport to take out appropriate liability insurance. Indeed, if an athlete is insured against any negligent harm he causes, the injured party is less likely to feel the need to test the vicarious liability of an NGB in a court of law.¹³³ The practicality of this solution is, however, largely dependent on two factors: the type of sport; and the extent of control exercised by the governing body. For some sports, such as tennis and golf, compulsory liability insurance for every athlete might be a useful development for the governing bodies to consider, particularly in light of the potential for injury in these sports and the stringent regulation that these athletes are subjected to. For other regulators, such as the World Darts Federation (WDF), this suggestion may be a rather pointless one. Not only is darts a sport in which injuries hardly ever occur, but the WDF also take a relatively lenient approach to the obligations of their players, and thus do not really possess the necessary degree of control over their athletes to justify vicarious liability.¹³⁴ However, even if there are NGBs that wish to mandate liability insurance for all participants, some burning questions remain: would such a development lead to the end of pro-am tournaments, where professional and amateur athletes compete together in a single event? It is one thing to require a professional athlete to take out appropriate liability insurance, but it is another thing entirely to require an amateur player to do so, particularly considering the likelihood that they may not be able to afford (or at least justify the cost of) such insurance. Likewise, we might also ponder whether insurers would be willing to provide such cover at all. Boyes makes the point that, in light of the increasing wealth of many top athletes, insurers may refuse to cover those who play high-value opponents.¹³⁵ After all (and to slightly modify a concern first raised by Boyes), who really wants to insure the person who might blind Rafael Nadal or Jordan Spieth?

(a) Tennis

Professional tennis is governed by a number of bodies, one of which is the aforementioned ITF.¹³⁶ The ITF operates as the world governing body of tennis, with responsibilities including the enforcement of the Rules of Tennis and the organisation of various tournaments (such as the four ‘Grand Slam’ tournaments, the international Davis Cup and Fed Cup, and the lower-rung ITF Men's and Women's World Tennis Tours). The Association of Tennis Professionals (ATP) – which operates several competitions under the ATP Tour and ATP Challenger Tour – acts as the governing body for men's tennis, whilst the Women's Tennis Association (WTA) acts as the global governing body for women's tennis. Although Gibson argues that the economic independence of professional tennis players means that no governing body is exercising control over them,¹³⁷ a brief perusal of the ITF, Grand Slam and ATP rulebooks suggest that this analysis is perhaps too simplistic. For example, with regard to the provisions laid down by the ITF and the Grand Slam Board (GSB), the rules outline that both bodies can exercise control over: the discipline of athletes found guilty of doping or corruption;¹³⁸ the time and location of matches;¹³⁹ the audible and visible actions of the athletes (including any ‘unsportsmanlike conduct’);¹⁴⁰ when athletes can take a break during a match;¹⁴¹ and the prompt attendance at post-match media conferences.¹⁴²

This latter requirement sparked international outrage during the 2021 French Open, when the former world number one Naomi Osaka cited mental health issues for her refusal to attend a post-match press conference.¹⁴³ She was later fined US$15,000 for the failure to honour this obligation, and one suspects that this was largely because the governing bodies in tennis stood to benefit from her interaction with the media. As the Grand Slam rulebook highlights, media appearances by the top stars provides ‘valuable exposure to the media and fans’, and this in turn helps to ‘drive engagement’ with the sport.¹⁴⁴ One can see, therefore, parallels with respect to funded athletes like Varnish, as governing bodies here are enjoying a similar level of control over, and benefit from, non-funded athletes too. This is reinforced by other stringent regulatory policies that are imposed by the ITF and GSB on professional tennis players, such as: restrictions on their participation in other events;¹⁴⁵ the degree to which they can promote their own sponsors;¹⁴⁶ and the close regulation of what attire and equipment is appropriate to wear and use during competition. In particular, participants competing at Wimbledon are famously obliged to wear white clothing,¹⁴⁷ and there are comprehensive regulations on what footwear is acceptable for different types of court.¹⁴⁸

Likewise, the ATP rulebook exhibits many similar limits on an athlete's discretion, both on and off the court. In relation to on-court restrictions, the ATP can impose financial penalties on so-called ‘commitment players’ – those ranked in the top 30 of the ATP's official rankings – who fail to participate in a certain number of competitions.¹⁴⁹ The rationale behind such a rule is, presumably, that the ATP want to accrue the financial and promotional benefits of the top stars regularly competing under their brand. Regarding the off-field control of professional tennis players, it is noteworthy that the ATP currently operate a so-called ‘STARS program’.¹⁵⁰ This initiative, which was introduced in an effort to exploit the ‘popularity of athletes through sponsorships and media’,¹⁵¹ mandates that ‘[a]ll players competing in the main draw of any ATP Tour tournament will be required, if asked, to participate in ATP sponsored activities’.¹⁵² Each player is required to dedicate up to two hours per week to the program, as well as participating in up to two activity days for promotional purposes (which may even take place ‘outside of an ATP Tour tournament week and/or location’).¹⁵³ Given that such activities sometimes place athletes in a position of power by working with children,¹⁵⁴ it could certainly be argued that, if a tennis player sexually abused a child as a result of his engagement with this program, the ATP ought to be vicariously liable for this act. This is consistent not only with case law demonstrating that the conferral of power on an employee strengthens the argument for imposing vicarious liability,¹⁵⁵ but also with the so-called harm-within-risk rule posited above. Unlike an on-court injury caused during a training session or warm-up (which the ATP exercise very little control over and reap no benefit from),¹⁵⁶ attendance at STARS events is something that the ATP explicitly mandate and profit from. In this light, we could say that the risk of off-court sexual abuse at a promotional event is within the scope of the governing body's relevant control/integration.

(b) Golf

Similar themes evident in tennis can also be identified in the context of professional golf. For instance, the aforementioned PGA Tour, which identifies itself as the ‘world's premier membership organization for touring professional golfers’,¹⁵⁷ is able to exert a significant degree of control over those participating in their events. According to its most recently published handbook covering the 2019–20 season, the competition organiser is able to regulate a players’ use of mobile devices and social media throughout the tour,¹⁵⁸ as well as impose strict limitations on any sponsorships and equipment deemed contrary to the spirit of golf.¹⁵⁹ As evidenced by PGA Tour Inc v Martin,¹⁶⁰ this even extends to the PGA Tour's attempts to control whether certain disabled competitors are permitted to use golf carts to traverse the course (which, according to some, looked ‘lousy on television’).¹⁶¹ A professional golfer's discretion in role is also inhibited by strict rules requiring them to maintain a certain pace of play whilst out on the green,¹⁶² as well as further restrictions on their signing of autographs and consumption of alcohol during competitions.¹⁶³ This latter provision suggests that the PGA Tour ought, in fact, to be held vicariously liable for any injury caused due to the negligent swing of an intoxicated golfer, as the control exercised by the organisation over this conduct is closely related to the relevant risk of harm.

Of course, this is not to say that professional golfers do not retain some form of control over their activities. They are able to choose both their own caddy and which tournaments they wish to enter, but even here the PGA Tour seem to have the last word. The organisation gives the final approval on each caddy,¹⁶⁴ and members of the tour must compete in at least 25 events over the course of the year or else risk a ‘major fine’ or suspension.¹⁶⁵ Much like with the obligation foisted upon ‘commitment players’ by the ATP, we could say that this requirement to play a certain number of games is especially important for vicarious liability purposes, as the governing bodies and competition organisers are explicitly influencing the level of risk posed by their athletes. As such, Sharpe is seemingly correct to argue that these control mechanisms ‘tend to show an employer-employee relationship between the PGA Tour and its golfers’.¹⁶⁶ Given that this comment was made for the purposes of employment and disability law, we could say that her argument is further strengthened when it is applied to the context of vicarious liability (because, as explained above, the normative importance of control is even more significant in this context).

A similar state of affairs is also identifiable in women's golf, where the Ladies Professional Golf Association (LPGA) – and its concomitant LPGA Tour – operates in a functionally similar manner to its male counterpart. One of the LPGA's more interesting policies for our purposes arose in August 2008, when the organisation attempted to enforce a new rule mandating that all participants on the tour speak English during pro-am events and when interacting with the media.¹⁶⁷ The purported reason for this rule was to ‘please corporate sponsors’ and to ‘maximize the marketability of its players’ in an attempt to improve the popularity of the tour.¹⁶⁸ Although the LPGA later backtracked on this policy (albeit with a version of the rule still remaining an option),¹⁶⁹ this example shows that, by acceding to these demands, foreign women golfers were clearly asked to carry on an activity that constituted ‘an integral part of the business activities’ of the LPGA, and were doing so for its benefit.¹⁷⁰ Lloyd persuasively argues that, ‘[b]y controlling the language that an LPGA player must speak, the LPGA is effectively exerting one of the most stifling and limited forms of employer control’.¹⁷¹ In this light, and as he further elaborates, the ‘courts must take a hard look at whether a professional golfer is really an independent contractor when the tour mandates the language that she must speak during the course of her membership on the tour’.¹⁷²

Conclusion

The adoption of a more contextual, policy-sensitive approach to an individual's employment status reveals that many NGBs and competition organisers ought to be held vicariously liable for tortious behaviour committed by athletes that compete under their auspices. This is a novel claim that has yet to be made in the scant literature on vicarious liability in sport. In relation to funded athletes in individual sports, it was maintained that the recent case of Varnish may provide an instructive roadmap for the implementation of such a policy-oriented approach. It was highlighted that certain factors prominent in employment law – such as mutuality of obligations and personal performance – ought to be downplayed when dealing with the issue of vicarious liability for the purposes of tort law. Conversely, this paper suggested that the notion of control ought to be prioritised in this context, as this theory arguably has a much stronger moral and normative impact on third parties.

As a result of this discussion, a number of important claims were made. First, it was observed that some athletes could be classed as employees for one purpose, but independent contractors for others. This is largely in accordance with recent case law in this area, and it is an eminently sensible approach insofar as it recognises the different policy considerations that underpin an unfair dismissal and vicarious liability claim. Secondly, and with reference to the facts of Varnish, it was also highlighted that control, benefit and risk may all need to overlap to a significant extent in order to produce a harmonious and logically consistent theory of enterprise liability. Given that enterprise liability has recently been touted as the strongest and most convincing rationale for vicarious liability, this is an important theoretical clarification that ought to be made in the law. Consequently, this is merely one example of how a sport-specific analysis of vicarious liability could help to teach us some broader lessons about the doctrine more generally.

Finally, this paper also established that it may be appropriate to impose vicarious liability on NGBs and competition organisers for the tortious actions of non-funded professional athletes too. This was illustrated most prominently with reference to the sports of tennis and golf. A brief perusal of the overly intrusive regulatory measures that are evident in the rulebooks of these sports demonstrates that it would certainly be feasible to impose vicarious liability on bodies such as the ATP and the PGA Tour. In this regard, and on a brief concluding note, it may be wise to end this paper with the following salutary warning: if sporting governing bodies and competition organisers wish to avoid the prospect of being held vicariously liable for the tortious behaviour of their athletes, they ought to seriously consider reducing the extent of some of their on-field and off-field regulatory policies.