The modern corporation has been made possible by the principles of limited liability and corporate personality (‘entity principles’), which have driven investment of private capital in enterprise on a scale previously unimaginable. Yet the granting of such privileges to corporations is also a ‘remarkable gift’2 whose entrenchment and extension has been more an accident of history than an evidence-based policy. These longstanding principles of corporate governance have been the subject of increasing criticism, with a range of commentators calling for a re-evaluation.3
This paper presents the case for an incremental step away from limited liability in a particular circumstance, namely the ‘risky business’ scenario. In such cases, a company incorporates a wholly owned subsidiary for the purposes of undertaking especially dangerous or risky activities and shielding itself from liability in the event that this risk is realised. The subsidiary is undercapitalised in relation to the gravity of the risk, resulting in the non- or under-compensation of tort victims. This brings the principles of corporate law into direct conflict with the principles of tort law, shifting social costs away from the business incurring them.4
A brief history of limited liability and corporate groups is first given, outlining their relationship with each other, and with the law of torts. The application of limited liability to corporate groups in the context of tortious liability is criticised, and it is argued that entity principles should be abandoned in such cases. The latter part of the paper identifies some potential impediments to the implementation of enterprise liability and assesses the different theoretical foundations. Finally, some factors are proposed upon which a test for enterprise liability could be based. It is argued that the application of enterprise liability in risky business situations is both necessary and feasible.
A Limited liability
The timing of the appearance of limited liability in its current form is difficult to pinpoint.5 Some examples of de facto limited liability can be found in ancient Rome and medieval Italy and the Mediterranean,6 though the modern emergence came from Seventeenth and Eighteenth century England.
By the second half of the 17th century, people were citing limited liability as a motive for incorporation7 and it was increasingly accepted that shareholders were not directly liable.8 The UK’s Attorney-General stated in 1784 that the ‘individuals who may compose the corporation would not be liable in their private characters’.9 The Joint Stock Companies Registration Act 1844 provided for unlimited liability, but this proved unworkable and was replaced shortly afterwards.10 Widespread acceptance of limited liability came in the context of railway companies, where shares were widely dispersed, management and shareholdings were separate, and tortious liability of individual shareholders could not be avoided. Acceptance of limited liability for the railways foreshadowed the widespread acceptance of limited liability more generally.11
In 1855, the UK Parliament enacted the Limited Liability Act, definitively accepting the principle.12 The dominance of limited liability was affirmed by the seminal case of Salomon (1897).13 In this case, the Court confirmed that all a party need do to attract the protection of limited liability is comply with the statutory requirements for company registration, thereby rejecting a purposive approach or any limitation on the strict wording of the Act.
In 1892, Germany embellished the UK’s effort and developed the GmbH,14 which subsequently became the primary model for the spread of limited liability and corporate personality across Europe. In France, for example, the French limited liability company15 had become more popular than the more traditional stock corporation by the late 1940s, comprising approximately one-third of all French companies.16
Limited liability has been called the ‘greatest single invention of modern times’, with one commentator stating, ‘even steam and electricity are far less important.’17 While others are somewhat less exuberant in their praise,18 the importance of limited liability generally is beyond dispute. However, the principle has been increasingly criticised in recent decades, particularly in the wake of a number of notorious risky business cases,19 and following the global financial crisis.20
B Corporate groups
Whereas limited liability was a deliberate decision taken due to commercial pressures,21 corporate groups emerged almost by accident. The Companies Act 1862 (UK)22 provided for incorporation, giving companies the powers specified in their memorandum of association, including, if specified, the power to acquire shares in another company. The statute did not specifically contemplate this possibility and the Courts, taking a literal approach to the statute held that a company could purchase shares in another company so long as it was authorised by its memorandum.23 From these haphazard beginnings, the use and prominence of corporate groups has grown exponentially; they now ‘conduct the great bulk of the economic activity of the industrialized world’.24
III Limited liability, corporate groups and tort law
The major consequence of giving corporations the power to purchase shares was the creation of two layers of limited liability: the original layer, protecting the ultimate investors in a company from claims against that company, and a new layer, protecting a parent company from liability from a claim against its subsidiary. The development of limited liability in the context of corporate groups was apparently accepted ‘without consideration of whether such acceptance was sound’.25
Likewise, the issues associated with applying limited liability in the context of tort creditors were not considered at the time limited liability was gaining credence.26 Indeed, it was not thought that the doctrine would entail any more than simply protecting investors in the company from the unsatisfied claims of the creditors.27 While this is problematic in relation to natural persons who are shielded from tortious liability, the scale of the risk that a corporate group is able to take in relation to potential tort victims makes this a pressing policy issue. In the leading UK case, Adams v Cape,28 the Court of Appeal vigorously reaffirmed the strict application of limited liability and rejected the many arguments made for imposing liability on a subsidiary company.
Thus the approach taken to corporate groups generally has been to squeeze them into the pre-existing body of company law, while at the same time, consideration has not been given to tort victims in the application of limited liability. Combined, it is clear that there is a need for a thorough re-examination of the application of limited liability in relation to corporate groups.
A Limited liability should not apply to wholly-owned subsidiaries
The main reason for removing limited liability in the context of corporate groups is that the justifications for limited liability generally, that is, as intended to apply to natural persons, do not hold in the group context. There is a lack of a clear and defensible policy justification for applying the principle, and the law is therefore out of step with the realities of modern corporations. Blumberg calls the law ‘anachronistic and dysfunctional’.29
The justifications for limited liability generally hinge on the characterisation of the shareholders as passive investors who have no interest in the business other than that their money is invested in it. However, in the context of corporate groups, and especially in the risky business scenario, the parent corporation is not an absentee owner, but rather the driving force behind the subsidiary.
The numerous advantages of, and justifications for, limited liability are well established, and many commentators have helpfully and comprehensively identified them.30 It is worth considering each of the major justifications as they relate to corporate groups:
Encouraging diverse sources of investment.33 This is irrelevant as the only shareholder/dominant shareholder is a parent company with the ability to diversify its portfolio and spread its risks in a way that ordinary investors cannot.34
Ensuring capital market efficiency (by enhancing the transferability and uniform pricing of shares).37 In the case of the wholly owned subsidiary, no adverse affect on the public market can occur as no such market exists.38
Encouraging risk taking. Risks must be balanced by responsibilities, as well as other policy goals. In the context of corporate groups, a holding company generally wishes to take advantage of the possibility of externalising risk using a subsidiary. This can encourage risk that is excessive because ‘owners who engage in excessively risky activities are protected from liability’.39
B Limited liability should be curtailed in relation to torts
Limited liability unduly prejudices tort victims. The key difference between a tort creditor and a voluntary creditor in the present context is that the latter can evaluate the risk associated with the business activity in advance, and consequently can charge a commensurate rate of interest40 or seek other forms of security.41 By contrast, a potential tort creditor cannot so readily account for the risk that his claim will go unsatisfied.42 The charging of interest commensurate with risk incentivises the company to reduce their risk in relation to those activities,43 whereas no such incentive exists with regard to potential tort creditors. The inability of potential tort creditors to foresee the harm means it is difficult, if not impossible, for them to contract out of limited liability,44 and in any case it is unlikely that potential tort victims will have the bargaining power to do this.
The core aims of tort law include the compensation of victims, the imposition of the costs of an activity on the activity itself and discouraging negligent or intentionally harmful activities,45 each of which is patently flouted by the application of limited liability. Firstly, a parent company is to some extent able to dictate the level of compensation paid to tort victims by undercapitalising its subsidiary. Secondly, society as a whole bears the cost of any risk, allowing companies to create externalities without paying for them. Thirdly, the need to discourage negligent or harmful activity is at odds with the risk-taking encouraged by limited liability.46
C ‘Piercing the veil’ and tort law alternatives
Veil piercing is a tool that attempts to look behind or through the existing corporate form in particular cases, generally where when the subsidiary is constituted for fraudulent purposes or to avoid an existing obligation. Veil piercing has rarely been a stable concept, and has been called ‘the most litigated issue in corporate law’.47
In Germany corporate law developed a number of theories of veil piercing based on ‘domination’ of a subsidiary by a parent, while today, shareholders can be liable in the case of a destructive interference in the corporation. These approaches are not for the benefit of tort victims, but for intra-group relations (see below). In the UK, where numerous cases have discussed the concept, the approach has been to pierce the veil based on the subsidiary being a fraud or ‘sham’, something that is difficult to prove in practice. Veil piercing remains uncommon in practice.
While veil piercing focuses on the illegitimate use of the corporate form, enterprise liability actually utilises the parent-subsidiary relationship to allocate liability.48 In so doing, enterprise analysis goes beyond the stated corporate form entirely. Rather than creating sporadic and incomplete exceptions to the corporate form, enterprise analysis tackles the problem at the critical juncture.49
Perhaps due to the ineffectiveness of veil piercing, a number of cases have attempted to utilise the ordinary rules of tort law to claim that the parent itself is primarily liable. In Lubbe v Cape (UK 2000),50 for example, the claimants sought to make use of this potential, but issues arose as to the difficulty of proving both a duty of care and causation. The Court noted that to ‘investigate, prepare and resolve these issues...would plainly involve a careful, detailed and cumbersome factual inquiry and, at least potentially, a very large body of expert evidence’.51
Subsequently in Chandler v Cape (UK 2012),52 the Court of Appeal for England and Wales upheld a High Court decision that a parent company owed a direct duty of care towards an employee of one of its subsidiaries to ensure a safe system of work. Arden LJ held that if the parent had interfered in the operations of the subsidiary, responsibility for health and safety issues would attach to the parent.53 He emphasised that piercing the veil was not necessary, as direct liability would exist if the parent exercised control; ordinary principles of tort law regarding third parties apply.
The decision in Chandler is significant as it is the first time that a claimant injured by a subsidiary company has established the parent company owed a duty of care. Arden LJ argued that the case did not involve piercing the corporate veil, though the outcome has the same effect, namely, imposing liability upon a parent company despite the fact that it is a separate legal entity to the subsidiary.
IV Enterprise liability
The present law fails to adequately address two key issues: corporate groups and tort creditors.54 While either of these issues in isolation may suggest the need for a re-evaluation, the need is particularly pronounced where the two issues intersect. In such a situation the ‘normative and economic realities necessitate a different regime of legal inquiry’.55
The regime of legal inquiry advocated here, and by a number of commentators, is enterprise analysis. Enterprise analysis can broadly be defined as treating a parent and subsidiary as one company. In the same vain, ‘enterprise liability’ refers to the holding of both a subsidiary and its parent liable for torts on the basis that they are, in fact, part of the same enterprise.
A number of jurisdictions have experimented with enterprise analysis, and this provides a contextual background for the discussion that follows.
Germany has been termed the ‘standard-setter’57 for enterprise analysis and was the first country to comprehensively approach the problem presented by corporate groups.58 German law defines two categories of group. The first, rarely used,59 is the contractual corporate group, whereby a voluntary ‘control agreement’ is made between the parent and subsidiary,60 pursuant to which the parent exercises ‘far-reaching management powers over the subsidiary’.61 The second is the de facto group, characterised by majority ownership of one company by another62 and a centralised and homogenous management structure63 in which both companies are operated as one enterprise.64
It is important to notet that the German legislature sought to correct a different problem to the one under discussion here. Their aim was to correct a perceived conflict between the interests of a subsidiary and its parent, who would ‘presumably seek to maximize its own shareholders’ welfare at the potential expense of the subsidiary’s minority or passive shareholders and creditors’.65 As a result, the focus is on compensation of a subsidiary by a parent for loss caused to the subsidiary, which does little to assist a tort victim trying to recover from a parent company.66
The law applies only to joint stock corporations, generally utilised by large public corporations,67 whereas the most commonly used form of company in Germany is the limited liability company (LLC).68 The judiciary actively extended the law to analogous situations not originally covered by the strict wording of the Act.69 The Courts extended enterprise liability to LLCs70 and other situations71 to the point where it was widely understood that enterprise analysis was more generally applicable.72 However, the court subsequently changed direction in this regard in the Bremer Vulkan case,73 severely curtailing the development of the doctrine.74 This ‘complete abandonment’ of the application of group liability principles75 has been lamented as hailing a return to a ‘casuistic, largely unprincipled and fragmental approach.’76
B German law on stock corporations: progeny
Despite the retreat of the German law, it has been ‘more influential than any other...in spurring legal reform of the parent-subsidiary relationship’,77 and a number of countries have adopted laws influenced or inspired by Germany’s example.78 Brazil, for example, has adopted a mild form of enterprise analysis79 that is broadly similar to the German law.80 Unfortunately, it also follows the German law in not giving creditors a direct cause of action, based on the belief that ‘experience shows that the creditor as a rule obtains protection via contractual agreement’.81 Portugal enacted similar provisions,82 but they only apply where a company chooses to legally formalise its parent-subsidiary relationships.83
Italy created a direct cause of action against a parent company,84 but the cause of action applies only if the holding company causes damage through mismanagement of the subsidiary.85 Overall, the Italian approach has been lamented as offering only piecemeal reforms, rather than an overarching system,86 and is of doubtful significance in practice.87
The EU has discussed enterprise liability. These discussions have not progressed as far as substantive law, though there is potential for the EU to be a ‘trendsetter’.88 The initial hope that the European Company Statute would include provisions on parent liability in the case of European Companies89 was quelled by the exclusion of such provisions in the eventual statute.90 Thus the remaining proposal of relevance here is the Ninth Directive on company groups.91 Under the directive the parent would have been responsible for the liabilities of subsidiaries on the basis of unified management and control.92 Unfortunately, the provisions have been lamented as being weakened in order to stem political opposition93 and progression of the directive has, for the time being, halted.94
There was a brief flirtation with enterprise analysis in English law, starting with Lord Denning’s ‘single economic unit’ argument in the DHN case.95 The issue in DHN was whether a group of three companies could be treated as one enterprise for the purposes of receiving compensation for the compulsory purchase of the group’s property. Noting that groups of companies are treated as one for many other purposes, Denning held that they should be treated as one concern for the purposes of compulsory purchase compensation.96
Although originally pertaining to compulsory purchase compensation claims, the single economic unit argument started to seep into the law at a more general level. In Lewis Trusts97 garments were being made in stages by different subsidiaries and May LJ suggested obiter that it was unnecessary to distinguish between them for the purposes of copyright infringement.98 During the debating of the Companies Act 1980, the Labour Party spokesman proposed that a parent company could be made liable for the debts of a defaulting subsidiary.99
Despite these nascent movements toward a more general application of the single economic unit argument, DHN was confined to its facts in Woolfson,100 and the separate entity approach was affirmed as a fundamental principle of company law once again in Adams v Cape.101
In the US, Michigan has distinguished between a company as an investor, who exercises ‘mere oversight of a subsidiary’s business in a manner appropriate and consistent with the investment relationship’, and a company that exercises ‘actual participation and control over a subsidiary’s functions and decision-making’.103 Texas courts had enunciated a doctrine of ‘single business enterprise’;104 however, this was subsequently rejected by the Texas Supreme Court.105 A landmark Louisiana case considered that ‘[i]f one corporation is wholly under the control of another, the fact that it is a separate entity does not relieve the latter from liability’106 and enumerated eighteen separate factors for consideration.107 Interestingly these factors included undercapitalistion and ‘excessive fragmentation of a single enterprise into separate corporations’, as well the more obvious factors such as the relationship between the parent and the subsidiary and commonality of offices, directors and employees. Though the doctrine has garnered some judicial support,108 it has also attracted academic criticism109 and the case has not resulted in a widespread movement to embrace the doctrine.110
By far the most extensive approach to enterprise liability has been taken by India. Firstly, in the wake of the Bhopal disaster, the Government assumed parens patriae responsibility for the resulting cases in the New York Courts,111 arguing that a corporate group is not a set of distinct entities, but in fact one entity, the ‘monolithic’ corporation,112 and that the group is in a better position to assume the risk of its activities than tort victims.113 A year after the Bhopal disaster, the Supreme Court of India, in the Oleum Gas Leak case,114 held that an enterprise has a duty to ensure that no harm results on account of the nature of its activity115 and that if such harm does result, the enterprise as a whole should be liable.116
The court noted that the tort victim is not best placed to bear the risk and social costs of the business117 and that the ‘enterprise alone has the resource[s] to discover and guard against hazards or dangers’.118 The court posited that the permission to carry on a risky business must come with the responsibility for the externalities generated.119
G United Nations Guiding Principles on Business and Human Rights
The United Nations Guiding Principles on Business and Human Rights (UNGPs) are a global standard for addressing the risk of adverse human rights impacts linked to business activity. The UNGPs were unanimously endorsed by the UN Human Rights Council in June 2011, becoming the first corporate human rights responsibility initiative to be officially adopted within the UN framework. The UNGPs have received wide support from a range of actors.
The UNGPs comprise three pillars: (i) state duty to protect human rights; (ii) corporate responsibility to respect human rights; and (iii) access to remedy for victims of business-related human rights violations.
The UNGPs are the culmination of several decades of UN efforts to create global human rights standards for business and represent a final attempt to overcome the divisive debate that had previously characterised the issue. The UNGPs are general in nature and therefore do not directly address concepts such as tortious liability or enterprise principles. However, they do mention ‘the need for access to effective remedies, including through appropriate judicial or non-judicial mechanisms’ and ‘enhancing access to effective remedies available to those whose human rights are affected by corporate activities’.
The fact that such unanimous agreement on key principles is now possible suggests a trend in favour of increased corporate responsibility and accountability with regards to human rights. In this context, it is clear that enterprise principles ought to enter into national and international discussions on how best to provide remedy to victims of risky business activities.
V Toward a test for enterprise liability
Creating a test for applying enterprise liability would ‘take a roomful of experts, legislators, and businesspeople’.120 That said, the discussion of enterprise liability above highlights a number of key issues that any such test would have to address.
A Part one: tort victims
A test for enterprise liability should allow all tort victims to be compensated. This is in line with the aims of tort and ensures that the parent company is held responsible for the risks taken.
Dearborn suggests that ‘torts’ in this context should be confined to ‘mass torts, human rights disasters, and environmental harms’.121 She notes that confining enterprise liability to mass torts will appease the business community, who will inevitably be concerned that ‘enterprise liability would cause the end of investment capitalism’.122 This argument is discussed more fully below, but suffice to say that, although the business community is likely to agitate against any measures to curtail corporate freedoms, it seems unlikely that anyone would truly think that the effects of a limited form of enterprise liability for tort victims would be so far-reaching.
Dearborn suggests that this limit on the definition of tort victims ensures that enterprise liability is ‘merely a tool to check the most egregious and socially harmful of corporate behaviours’.123 However, it is not clear why it is only these behaviours that should be checked. This would cause a lacuna between victims of the parent’s torts and victims of the subsidiary’s torts: while the former will be compensated for any tort, the latter would only be compensated where the tort happened to be particularly awful. Furthermore, whereas allowing all tort claims makes for simple application, the confinement of torts to mass torts, human rights disasters, and environmental harms adds a layer of difficulty in determining which tort victims are able to recover, and, in particular, determining what extent of harm must occur for an event to qualify as a ‘human rights disaster’.
Dearborn also makes the point that, in terms of controlling corporate behaviours, it is mass torts and the most serious torts that ‘stand to harm the corporation from a public relations and economic standpoint’.124 While this is true, allowing the parent to be held liable for all torts will further increase the corporate control effects of enterprise liability by increasing the number of torts the company can be liable for. In any case, it is arguable that the control of corporate behaviour should not be the only factor taken into account. In risky business cases, the aims of tort law should take precedence and the focus should be on the injustice caused by the inability of a subsidiary to adequately compensate tort victims.
At least one commentator has suggested that corporate groups should be able to ‘opt out’ of enterprise liability.125 This is clearly antithetical to the arguments made in this paper: an opt out provision would allow a parent company to easily preserve the limited liability that has been seen to be so problematic.
B Part two: defining ‘enterprise’
Part two of the test would focus on when exactly a subsidiary and its parent will be considered part of the same enterprise. The different bases for this are much discussed in the literature, and a number of examples can be identified in the current implementations of enterprise liability discussed above. The test for enterprise liability must protect ‘real’ investors, namely those for whom the original justifications for limited liability apply, while maintaining liability of a company that is, in reality, part of the same enterprise.
1 Control-based enterprise liability
Basing a test for enterprise liability on the control exerted by the parent over the subsidiary is superficially appealing because it appears to offer simplicity. The major downside of this approach is that it may incentivise the deliberate decentralisation of management in order to make it appear that the subsidiary acts independently, thus allowing for easy circumvention. Decentralisation may even result in an increased chance of risky behaviours as oversight is decreased. It may therefore be desirable to encourage strong central management in the hope of increasing oversight and preventing torts before they occur.126
An alternative is to establish a presumption of control based on a share holding percentage.127 This is, again, attractive due to its simplicity, but, unfortunately, along with this simplicity comes the risk that the reality of the situation will not be reflected accurately by the legal standard. Hypothetically there may be an optimal percentage which reflects the business reality in a majority of cases, but in reality such a level will be almost impossible to find, and would likely causes the law to be inflexible and unable to adapt to the specific case at hand.
2 Economic or ‘true’ enterprise liability128
Economic enterprise liability looks in detail at the structure of the group in question, beyond mere control. An economic test assumes that the picture of a parent and subsidiary company acting together as a unified economic unit is ‘generally an accurate one’.129
Whereas a control-based approach incentivises the decentralisation of management and therefore increases the chance that torts will occur, an economic approach incentivises the parent to invest in preventing torts. This approach would also mitigate the rigidity of a control-based approach. A number of factors could be considered, and this flexibility would be invaluable to legislators and the courts in drawing up and applying the test in a way that adequately allocates liability.
The following three factors would be central to an economic test for enterprise liability, though drawing up an exhaustive list would likely be impossible: (i) whether the subsidiary was created, or acquired, for the purposes of furthering the economic aims of the parent; (ii) whether the externalisation of risk plays a role in the utilisation of the subsidiary; and (iii) market and public identification of the group as a unified business.130
There are numerous real-life examples of cases where these factors could easily have been applied in practice. In the Adams v Cape case it would have been clear to the court that the subsidiary was incorporated to further the economic aims of the parent. In the Trafigura case internal emails came to light that would have informed a court that the company was purposefully externalising risk.131 In the Unocal case,132 villagers living close to a pipeline under construction were gifted Unocal-branded items during construction in an attempt to find favour with them.133 In this way, Unocal was able to hold itself out to be the company behind the construction, yet also distance itself from the risk it was creating.
The use of a factored test has been criticised. Presser contends that such an approach replaces serious purposive analysis with lists of factors,134 while Hamilton and Macey suggest that problems would arise due to poor weighting of factors and the encouragement of a mechanical approach.135 On the contrary, a factored test offers guidance and a framework in which courts can assess the economic unity of the enterprise. The factors are not a substitute for purposive analysis, but a guide to approaching such an analysis. There is little reason to think that the courts, guided by suggested factors and cognisant of the overarching question at hand, will approach the test in an unduly mechanical fashion. Indeed, courts are well accustomed to using such frameworks to interpret broad concepts and guide their decisions.
VI Arguments against enterprise liability
As with any proposal for reform, there will be detractors and arguments wielded against the implementation of enterprise liability at the intersection of corporate groups and torts. This part of the paper identifies some of the most likely arguments.
Negative economic impacts. It seems unlikely that the limited scope for liability argued for here would cause significant damage to the wider economy. Only excessively risky ventures will be deterred and any resulting decline in investment is likely to be very small. Germany provides an example of ‘an industrialized country that has adopted a milder form of enterprise principles without disastrous results for domestic or international investment capitalism’,136 while India, with its stronger approach to enterprise analysis, provides an even more pertinent example.137
Avoidance. Companies will undoubtedly attempt to circumvent such a rule, however, this is not a reasonable argument for not attempting to draft a test that identifies such attempts. Simply enacting such a law will likely act as a deterrent,138 and the courts can, and frequently do, take an active role in ensuring the purpose and spirit of the law is upheld.
Enterprise liability is insufficient. It may be argued that the application of enterprise liability advocated here does not go far enough, and that parent companies should be liable for all debts of the subsidiary.139 However, externalisation of risk is only one of many reasons that a company may wish to incorporate a subsidiary:140 if unlimited liability were to be imposed as the new paradigm in company law, the myriad of legitimate uses of subsidiaries would be frustrated. The approach taken in this paper is cautious, balancing the legitimate uses of the corporate form with the need to protect tort victims from the ill-considered application of limited liability to corporate groups.
Uncertainty. Enterprise liability has been attacked on the ground that ‘measuring the extent of an “economic unit” introduces an intolerable level of uncertainty into the question of liability’.141 This argument is commonly made against many legal reforms, yet it is unconvincing. A diligent legislature could remove some uncertainty with the construction of a sufficiently certain test, while the courts regularly apply complex legal standards with little difficulty.
Uncertainty of goals of enterprise liability. Sommer contends that, while enterprise analysis suffers from a ‘haziness of goals’ because ‘it is difficult to see why jurisdiction or liability should attach to an active investment and not a passive one’,142 Addo suggests ‘enterprise law needs a clear and distinct rationale to rival the facilitation of entrepreneurial spirit [of entity principles]’.143 The need to adequately protect tort victims in the risky business scenario provides both a justification for distinguishing between passive and active investors and a driving rationale for implementing enterprise liability.
VII The reason for inaction
Given the strength of the arguments for enterprise liability, particularly in risky business situations, why is it that courts ‘remain so willing to provide limited liability to parent corporations in tort cases’?144 It may be that the risks are seen as remote, such that even the imposition of liability would lead to no additional preventative actions being taken.145 Yet this explanation is unsatisfactory because: (i) many cases involve the use of a subsidiary company precisely to avoid the consequences of risks that they are aware of, as opposed to remote and unforeseen risks; and (ii) even if imposition of liability would not stop such torts occurring, it would at least allow the courts to provide a remedy for tort victims.
A much more likely explanation for reluctance on the part of the courts is a predisposition of deference to the legislature. Cognisant of this, legislatures could have implemented the necessary reform, but have, as yet, failed to do so. Hansmann and Krakmaan suggest that this inaction is because markets and politics do not well represent the interests of tort victims:146 tort creditors cannot contract around limited liability, while their inherently disparate nature means that potential tort creditors ‘do not constitute an easily organized political interest group’.147
This seems to have been borne out in practice in India, where it was a disaster killing an estimated 5,000 people148 that provided the impetus for reform:149 ‘The Bhopal disaster shook off the lethargy of everyone and triggered off a new wave of consciousness.’150
This paper started by observing that the current application of limited liability in the context of modern corporations is flawed, failing to take a considered approach to corporate groups and prejudicing tort victims. The former arose because the modern phenomenon of corporate groups has been awkwardly squeezed into the concepts of the corporate personality and limited liability, apparently without acknowledging that their foundational justifications are irrelevant to such groups. The latter arose because uncritical acceptance of limited liability meant that the interests of tort victims were not considered.
Despite these beginnings, the strict separation of companies has remained a fairly consistent principle of company law. This paper has argued that the time for reform in this area is long overdue and that enterprise liability at the intersection of corporate groups and torts is an incremental step in this direction. This narrow focus tempers the potentially sweeping and revolutionary nature of enterprise analysis, dealing specifically with risky business situations, and reflecting both the economic realities of modern enterprise and the aims of tort law.
Such a reform should not be delayed until a large-scale human tragedy takes place, but should be progressively implemented with a view to replacing the dysfunctional and anachronistic law at the intersection of corporate groups and torts with an appropriate and just set of rules.
I wish to thank Sandra Frisby at the University of Nottingham for her excellent corporate law course that inspired this paper, and Michelle Worthington at the Australian National University, whose interesting work on corporate governance caused me to revisit this subject.
See eg DW Leebron, Limited Liability, Tort Victims and Creditors (1991) 91 Columbia Law Review (Colum L Rev) 1565, 1566; M Dearborn, Enterprise Liability: Reviewing and Revitalizing Liability for Corporate Groups (2009) 97 California Law Review (Cal L Rev) 195; P Muchlinski, Limited Liability and Multinational Enterprises: A Case for Reform? (2010) 34 Cambridge Journal of Economics 915–928.
Blumberg (1986) 11 J Corp L 573, 583. Nevertheless, nearly 1000 unlimited companies were registered under the Act until it was replaced by the Limited Liability Act 1855. BC Hunt, The Development of the Business Corporation in England 1800–1867 (1936).
Eg JA Grundfest, The Limited Future of Limited Liability: A Capital Markets Perspective (1992) 102 Yale LJ 387, 420 (‘it has its theoretical flaws. It is not a thing of perfect beauty, but at least it works’).
Making some attempt to find authority for the practice in the statute, the Courts were partially persuaded by the fact that corporations were mentioned in the definition of ‘persons’, who could become members of a company. In re Barned’s Banking Company (1867) 3 Law Reports, Chancery Division (LR Ch) 105, 112 f; In re Asiatic Banking Corporation (1869) 4 LR Ch 252, 257.
CD Stone, The Place of Enterprise Liability in the Control of Corporate Conduct (1980) 1 Yale LJ 1, 70 (‘Only later, when corporate liability for serious wrongdoing had grown from the exception to the rule, could the principle of limited liability have taken on, imperceptibly, a meaning not originally signified’) (citations omitted).
See eg Blumberg (1986) 11 J Corp L 573, 611–616; A Muscat, The Liability of the Holding Company for the Debts of its Insolvent Subsidiary (1996) 162–175; FH Easterbrook/DR Fischel, The Economic Structure of Corporate Law (1991).
That is, the costs associated with collecting payment from numerous dispersed shareholders. See RB Thompson, Unpacking Limited Liability: Direct and Vicarious Liability of Corporate Participants for Torts of the Enterprise (1994) 47 Vanderbilt Law Review (Vand L Rev) 1, 20.
Leebron (1991) 91 Colum L Rev 1565, 1612. Indeed, even where two or more different companies own another company, it is still likely that collection costs will not be so great as to negate any benefit gained.
It is worth noting that the same argument has been applied in relation to financial institutions who are ultimate investors. See P Halpern/M Trebilcock/S Turnbull, An Economic Analysis of Limited Liability in Corporation Law (1980) 30 University of Toronto LJ 117, 298. Blumberg notes that ‘[t]his factor would be as applicable to corporate groups as to financial institutions’, idem (1986) 11 J Corp L 573, 624.
K Hofstetter, Parent Responsibility for Subsidiary Corporations: Evaluating European Trends (1990) 39 International & Comparative Law Quarterly (ICLQ) 576, 578; SB Presser, The Bogalusa Explosion, ‘Single Business Enterprise’, ‘Alter Ego’, and Other Errors: Academics, Economics, Democracy, and Shareholder Limited Liability: Back Towards a Unitary ‘Abuse’ Theory of Piercing the Corporate Veil (2006) Northwestern University Law Review (NWULR) 405, 425.
These are not the only two problems, but are certainly two key problems with the present state of the law. Other commentators have recognised this. See eg Thompson (1991) 76 Cornell L Rev 1036; Dearborn (2009) 97 Cal L Rev 195, 209; Blumberg (1986) 11 J Corp L 573, 576.
For a comprehensive account of German law on corporate groups and its evolution, see R Reich-Graefe, Changing Paradigms: The Liability of Corporate Groups in Germany (2005) 37 Conn L Rev 785. For an account of its history, see JE Antunes, Liability of Corporate Groups: Autonomy and Control in Parent-Subsidiary Relationships in US, German, and EEC Law: An International and Comparative Perspective (1994) 342–347.
German Law on Stock Corporations (Aktiengesetz) 1965, § 18. Available at <http://www.gesetze-im-internet.de/bundesrecht/aktg/gesamt.pdf>, accessed 16February 2010. Translated in H Schneider/M Heidenhain, The German Stock Corporation Act: Bilingual Edition with an Introduction to the Law (2000).
Dearborn (2009) 97 Cal L Rev 195, 216 (citation omitted). See also B Singhof, ‘Equity Holders’ Liability for Limited Liability Companies’ Unrecoverable Debts – Reflections on Piercing the Corporate Veil under German Law (1999) 22 Loyola of Los Angeles International and Comparative L Rev 166, 169.
See Autokran case (1985) Bundesgerichtshof (German Federal Court, BGH) 95 BGHZ 330; C Alting‚ Piercing the Veil in American and German Law – Liability of Individuals and Entities: A Comparative View (1995) 2 Tulsa Journal of Comparative and International Law 187, 234.
Bremer Vulkan (2001) BGH II ZR 178/99. For a detailed discussion of this case in English, see P Zumbansen, Liability within Corporate Groups (Bremer Vulkan): Federal Court of Justice Attempts the Overhaul (2002) 3 German LJ.
Dearborn (2009) 97 Cal L Rev 195, 220. See also TW Wälde, Parent-Subsidiary Relations in the Integrated Corporate System: A Comparison of American and German Law (1974) 9 Journal of International Law & Economics 454, 492 f.
The German model has also influenced jurisdictions other than those discussed here, such as Slovenia and Croatia; unfortunately these developments cannot be elucidated further due to language constraints and the lack of available information. M Andenas/F Wooldridge, European Comparative Company Law (2009) 6.
See M Ventoruzzo, Experiments in Comparative Law: The Recent Italian Reform and the Dubious Virtues of a Market for Rules in the Absence of Effective Regulatory Competition (2004) <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=556601> accessed 27 February 2010, 40.
The Ninth Directive has never been published in the Community Gazette (Orne (tr), B Pasa/GA Benacchio, The Harmonization of Civil and Commercial Law in Europe (2005) 369). A mid-1980s French draft is available in CDVA (ed), Modes de rapprochement structurel des entreprises. Tendences actuelles en droit des affaires (Brussels, Commission droit et vie des affaires, 1986). Some provisions of the Directive relevant to the current discussion are reprinted in English in K Böhlhoff/J Budde, Company Groups – The EEC Proposal for a Ninth Directive in Light of the Legal Situation in the Federal Republic of Germany (1984) 6 J of Comparative Business and Capital Market Law 163, 181–192.
Though given that he suggested that this liability could be excluded by notification of creditors, it seems that this proposal did not include involuntary and tort creditors. The proposal was rejected. See D Prentice, Groups of Companies: The English Experience, in: K Hopt (ed), Groups of Companies in European Law: Legal and Economic Analyses on Multinational Enterprises (1982) 99, 111
Note that US Courts have been more willing to look into the structure of corporate groups when it appears to be permitted by federal regulation. See CA Schipani, The Changing Face of Parent and Subsidiary Corporations: Enterprise Theory and Federal Regulation (2005) 37 Conn L Rev 691. See also Dearborn (2009) 97 Cal L Rev 195.
CPC International v Aerojet General Corporation, 777 Federal Supplement (F Supp) 549, 573 (Michigan 1991). See also Lansford-Coaldale Joint Water Authority v Tonolli Corporation, 4 Federal Reporter, Third Series (F 3d) 1209, 1222 (3d Cir 1993). This distinction tracks the argument that the case for limited liability is at its strongest in relation to investors, but at its weakest where a parent company operates the subsidiary.
Paramount Petroleum Co v Taylor Rental Center (1986) 712 South Wester Reporter, Second Series (SW 2d) 534, 536 (Texas CA). See also PI Blumberg et al, Blumberg on Corporate Groups (2nd edn 2005) secs 12.04 and 66.04 [A].
See eg Thibodeaux v Ferrellgas, Inc, 741 So 2d 34, 35, 42 f (La Ct App 1999) (Louisiana CA) and Grayson v R B Ammon & Assocs, 778 So 2d 1 (La Ct App 2000) (Louisiana CA) and Pine Tree Association v Doctors’ Association, Inc, 654 So 2d 735, 736, 738 (La Ct App 1995) (Louisiana CA), cf Town of Haynesville, Inc v Entergy Corporation, 956 So 2d 192 (La App 2d Cir) 964 So 2d 334 (La 2007) (Louisiana CA).
Bhopal Gas Leak Disaster (Processing of Claims) Act 1985. See also Brief of the Plaintiffs, Union of India v Union Carbide Corporation reprinted in U Baxi/T Paul, Mass Disasters and Multinational Liability: the Bhopal case (1986). Note that ‘Union of India’ in a judicial context refers to the government of India (art 300 of the Constitution of India).
MC Mehta v Union of India, All India Reporter (AIR) 1987 SC 1086 (India Supreme Court) (available in part at <http://www.elaw.org/node/1322>, accessed 17 February 2010; also reprinted in full in Baxi/Paul (fn 110)).
Ibid. While the Court framed the case in terms of the rule in Rylands v Fletcher ( UKHL 1) and strict liability, the language used is that of enterprise liability and it is generally considered that this is the underlying reasoning. See Baxi/Paul (fn 110) and U Ramanathan, Business and Human Rights: the India Paper <http://www.ielrc.org/content/w0102.pdf> accessed 17 February 2010, 13. Cf PT Muchlinski, Multinational Enterprises and the Law (2007) 318.
J Kluver, Entity vs. Enterprise Liability: Issues for Australia (2005) 37 Connecticut L Rev 765, 781. While this suggestion is made in the context of discussing the difficulties of meeting the demands of creditors upon insolvent liquidation, there is no suggestion in the article that tort creditors would not be considered creditors for this purpose, or that an opt out would not extend to tort creditors.
Dearborn (2009) 97 Cal L Rev 195, 226. The term ‘true enterprise liability’ is more accurate as some factors, such as public identification, are not strictly economic, however, the former term is more commonly used in the literature.
D Aronofsky, Piercing the Transnational Corporate Veil: Trends, Developments, and the Need for Widespread Adoption of Enterprise Analysis (1985) 10 North Carolina J of International Law and Commercial Regulation 31, 42.
That case involved contracting, rather than use of a subsidiary, but the salient point remains. See D Leigh, How UK oil company Trafigura tried to cover up African pollution disaster, The Guardian (London, 16 September 2009) available at <http://www.guardian.co.uk/world/2009/sep/16/trafigura-oil-ivory-coast> accessed 27 March 2010. The emails in question can be read here: <http://www.guardian.co.uk/world/2009/sep/16/trafigura-email-files-read> accessed 27 March 2010.
Though the general difficulty of attempting to discern the effects of one provision on an entire economy is exacerbated by India’s generally rapid economic growth and liberalisation of other areas of the law.
In the Thor litigation, one set of claims was settled out of court after the Court merely accepted that they had jurisdiction over the case (Ngcobo v Thor Chemicals Holdings  Times Law Reports (TLR) 579), while Unocal settled claims against them in the face of potential litigation under the Alien Tort Claims Act (MD Kielsgard, Unocal and the Demise of Corporate Neutrality (2009) 36 California Western Intl LJ 185, 189).
Simplification of management and international operations, for example, are potential reasons for creating subsidiaries. WO Douglas/CM Shanks, Insulation from Liability through Subsidiary Corporations (1929) 39 Yale LJ 193.