|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of Otago Law Theses and Dissertations |
Last Updated: 18 September 2023
MULTINATIONAL ENTERPRISES AND THE INCENTIVES TO TAKE RISKS:
RETHINKING LIMITED LIABILITY FOR PARENT COMPANIES IN
CASES OF TORT
James Alan Tocher
A dissertation submitted in partial fulfilment of the degree of Bachelor of Laws (Honours) at the University of Otago.
October 2016
Acknowledgements
To my supervisor Associate Professor Shelley Griffiths for your guidance and insights, as well as for stimulating my interest in commercial law over the years.
To Dr Maria Hook for your helpful discussions about the conflict of laws.
To my friends and family for your confidence and support.
In particular, to Joe for helping me understand economics and providing useful criticism.
To Erwin for sharing your practical advice about the process of law over the years.
And finally to Chrystal for your many contributions to this paper, as well as for your love and encouragement.
TABLE OF CONTENTS
INTRODUCTION
On December 3, 1984, a tank housed in a pesticide plant in Bhopal, India exploded, unleashing forty tons of deadly gas. The explosion killed 3,500 people within days and has killed more than 15,000 people over the ensuing thirty years. The release can be traced back to oversights and defective equipment by U.S.-based Union Carbide Company, which immediately attempted to dissociate itself from legal liability for the tragedy.1
The Bhopal disaster is just one of many examples of “mass torts” committed by subsidiaries of multinational enterprises.2 These torts come in a number of different forms; most commonly defective products, toxic substances and environmental injuries.3 These torts can have a drastic impact on their victims, yet it can be very difficult to successfully obtain relief in the status quo, because these subsidiaries typically have limited assets and occasionally (especially in the case of historical torts) no longer even exist.4 Additionally, it is near impossible to sue the parent company because of what Peter Muchlinski refers to as the “corporate veil” and the “jurisdictional veil”.5
The corporate veil refers to the fact that parent companies are protected from their subsidiaries’ misdeeds by the principle of limited liability.6 This protection is important, because multinational enterprises tend to operate large portions of their businesses through subsidiaries.7 In 1982, the largest 1,000 American companies had an average of 48 subsidiaries, while particularly large multinational enterprises such as British Petroleum operated through 1,200–1,300 subsidiaries.8 The ability to avoid liabilities using the corporate veil has itself created an incentive for corporate groups that operate especially risky businesses to do so through subsidiaries.9 It is common practice in many industries,
1 Sarah Coleman and Jonathan Friedler “The Road to Reform in the Wake of Kiobel: Multinational Corporations and Socially Responsible Behaviour” (2014) 13 J Int’l Bus & L 191 at 191.
2 At 191–192.
3 Nina Mendelson “A Control-based Approach to Shareholder Liability for Corporate Torts” (2002) 102 Colum L Rev 1203 at 1237.
4 Peter Nygh “The Liability of Multi-national Corporations for the Torts of their Subsidiaries” (2002) 3 EBOR 51 at 55.
5 Peter Muchlinski “Limited Liability and Multinational Enterprises: A Case for Reform?” (2010) 34 Camb J Econ
915 at 920.
6 Note that almost everything said in this paper about the corporate veil will also apply to wholly domestic corporate groups. However, as this is not the key focus of this paper, reference will usually just be made to
multinational enterprises for the sake of simplicity.
7 Nygh, above n 4, at 51.
8 Phillip Blumberg “Limited Liability and Corporate Groups” (1986) 11 J Corp L 573 at 575, n 1.
9 Henry Hansmann and Reinier Kraakman “Towards Unlimited Shareholder Liability for Corporate Torts” (1991) 100 Yale LJ 1879 at 1888.
such as the tobacco and hazardous waste industries, to incorporate separate subsidiaries to own high-risk assets.10 For instance, an oil transport company on Lake Champlain operated each barge through a different subsidiary, leading to a number of spills resulting from “employees continuing to pump oil into the water after encountering a leak, employee failures to cap pipelines, and disregard of navigation agreements.”11
This problem is exacerbated further by the jurisdictional veil, which refers to the problems arising from the conflict of laws that make it difficult to obtain an enforceable judgment against the parent company if it is located in a different country from its subsidiary.12 This scenario is common given that, in 2006, there were 777,647 foreign subsidiaries (many located in the developing world) associated with 78,441 different multinational enterprises worldwide.13 When considering instances of tort liability, the importance of multinational enterprises cannot be overstated. In 2000, the 200 largest multinational enterprises accounted for 27.5 per cent of global economic activity, and 51 of the largest 100 economic entities in the world were actually companies rather than states.14
This paper seeks to re-open the question of limited liability in this particular context, in an attempt to provide a solution to the problem faced by torts victims. The proposal is to reintroduce unlimited shareholder liability for parent companies, in cases of tort. Chapter One explores the existing legal avenues for redress, such as direct tortious liability against the parent company, and the doctrine of piercing the corporate veil. It concludes that there are severe drawbacks of leaving torts victims to rely on these avenues alone. Chapter Two looks at the historical origin of limited liability in an attempt to understand why it currently applies in this context. This chapter also addresses the reasons for confining this proposal to torts cases, and within that only to parent companies, by looking at the economic rationales behind limited liability. Chapter Three focuses in more depth on the one economic rationale that is relevant in this context; the incentive to take risks. After considering the opposing incentive that tort law seeks to impose, we arrive at the conclusion that the current balance of incentives is skewed too heavily in favour of risk taking.
10 Hansmann and Kraakman, above n 9, at 1881.
11 Mendelson, above n 3, at 1235.
12 Muchlinski, above n 5, at 920.
13 Bastian Reinschmidt “The Law of Tort: A Useful Tool to Further Corporate Social Responsibility?” (2013) 34 Co Law 103 at 104.
14 At 103.
Given that revelation, Chapter Four attempts to design the best possible alternative to the current regime of limited liability. It goes into detail about how a system of unlimited liability would operate, and highlights some difficult choices that have to be made. Chapter Five considers what the implications of this specific alternative regime would be, by looking at how corporate groups are likely to respond, and any opportunities they have to avoid the effect of the reform. Chapter Six focuses on one particular way to avoid this reform, which is to use the jurisdictional veil already referred to. Chapters Five and Six identify numerous harmful outcomes of this reform. This paper does not seek to make a judgment call as to whether these harms are sufficiently large so as to outweigh the benefit of solving the original problem. That is a political decision that can only be made by lawmakers. The purpose of this paper is instead to identify that there is a problem with our current legal regime, and to outline the legal consequences of the best possible alternative, so as to inform lawmakers and enable them to make a decision.
CHAPTER ONE: EXISTING LEGAL AVENUES
1.1 Direct Liability in Tort
There are two legal tools that can be used to overcome the obstacle posed by the corporate veil. The first is to find the parent company directly liable in negligence, for its own involvement in the actions leading to the tort’s commission.15 The problem with this tool is that liability must be established on the facts of each case, and not all torts victims will be owed a duty of care.16 Direct liability has been found in situations where the parent company was actually involved in the activity that lead to the tort, such as where the parent company was the manufacturer of defective products distributed by the subsidiary.17 Another example is Dagi v Broken Hill, where the parent company directly managed a mining operation ostensibly owned by its subsidiary.18 While control over the actual operations can lead to tortious liability, Peter Nygh makes it clear that mere control over the management of the subsidiary will not be enough.19 This is because failure to prevent negligence while merely in control of management would be an omission, and it is generally accepted in tort law that omissions only lead to liability if there is a positive duty to act.20
This distinction is generally seen in the case law of various jurisdictions. In James Hardie v Hall, the New South Wales Supreme Court held that an Australian parent company was not liable for asbestos-related damage to the employees of its New Zealand subsidiary, in spite of “evidence of some control and influence” over the subsidiary.21 In United States v Bestfoods, the United States Supreme Court reiterated that “active participation in, and control over, the operations of a subsidiary could not, without more, render a parent company liable for the acts of its subsidiary”.22
15 Nygh, above n 4, at 64.
16 At 64.
17 At 75.
18 Dagi v The Broken Hill Pty Ltd (No 2) [1997] 1 VR 428 as cited in Nygh, above n 4, at 76.
19 Nygh, above n 4, at 77.
20 At 78. See generally Stovin v Wise [1996] UKHL 15; [1996] AC 923 (HL) at 926. See also Home Office v Dorset Yacht Co Ltd
[1970] UKHL 2; [1970] AC 1004 (HL) at 1063 where the House of Lords required there to be a special relationship between the tortfeasor and the torts victim in order to establish liability when the omission involved failure to control a third party. A subsidiary would be a third party here in much the same way.
21 James Hardie & Co Pty Ltd v Hall (1998) 43 NSWLR 554 as cited in Muchlinski, above n 5, at 922.
22 United States v Bestfoods [1998] USSC 65; 524 US 51 (1998) as cited in Peter Muchlinski Multinational Enterprises & the Law
(2nd ed, Oxford University Press, New York, 2007) at 312.
Cases where direct liability has been found involve exceptional circumstances. In Bowoto v Chevron Texaco Corp, a federal district court in California found a parent company liable for the actions of its Nigerian subsidiary, which acted with the Nigerian military to suppress protests, leading to deaths and injuries in violation of international human rights norms.23 The extreme nature of the wrongs likely influenced the result here, but even so, the parent company had engaged in a high level of communication with the subsidiary over its response to the protests, meaning this was effectively a case of direct involvement anyway.24 Similarly, in the Amaco Cadiz case, a federal district court in Illinois held the parent company liable for the oil spill caused by the grounding of its subsidiary’s tanker off the coast of France in 1978.25 Although the Court purported to find liability on the basis that it was an “integrated multinational corporation”, it is evident that at the time of the accident, the parent company had been the one operating the tanker, which it effectively treated as its own property.26
In Chandler v Cape, the English Court of Appeal found that a parent company owed a duty of care to the employees of its subsidiary with respect to asbestos liability, because it possessed superior knowledge on relevant aspects of health and safety, and the employees’ reliance upon it to use that knowledge for their protection was foreseeable.27 However, this duty only protects a limited class of torts claimants; employees. In Akpan v Royal Dutch Shell,28 the District Court of The Hague refused to recognise a duty of a parent company to a wider class of individuals; local residents who had been affected by oil spills; because doing so would extend liability to “a virtually unlimited group of people”.29 This would run contrary to the principle that tortfeasors cannot owe duties to the whole world.30
This demonstrates the general problem with relying on direct tortious liability; even on the rare occasion the court finds that the parent company had been sufficiently involved, the claimant might still be held back by various principles of tort law. Furthermore, it can be difficult to find evidence demonstrating direct involvement in the first place, and the
23 Bowoto v Chevron Texaco Corp 312 F Supp 2d 1229 (ND Cal 2004) as cited in Muchlinski, above n 22, at 312.
24 Muchlinski, above n 22, at 312.
25 The “Amaco Cadiz” [1984] 2 Lloyd’s Rep 304 (USDC) as cited in Muchlinski, above n 22, at 310.
26 Muchlinski, above n 22, at 311.
27 Chandler v Cape Plc [2012] EWCA Civ 525, [2012] 1 WLR 3111 at [79]–[80].
28 Akpan v Royal Dutch Shell Plc DC The Hague LJN BY9854/HA ZA 09-1580, 30 January 2013.
29 Lee McConnell “Establishing Liability for Multinational Oil Companies in Parent/Subsidiary Relationships” (2014) 16 Env L Rev 50 at 54.
30 Caparo Industries Plc v Dickman [1990] UKHL 2; [1990] 2 AC 605 (CA) at 621.
separate legal identity of the subsidiary effectively acts as a presumption that there was no such direct involvement in the absence of evidence to the contrary.31
1.2 Piercing the Corporate Veil
If a torts claimant is unable to establish a direct duty of care in tort, it can attempt to pierce the corporate veil in order to attach liability to the parent company, even when it is not actively involved.32 This doctrine is, in effect, an exception to the rule of limited liability, which is designed to curtail the worst excesses of that regime.33 However, the doctrine is largely ineffective at that task, because it has been treated strictly, since it is seen as a violation of the foundational principle of company law: separate legal identity.34 In Commonwealth jurisdictions the rule has been limited to cases where the company is a “mere façade” used to defraud others or defeat their rights.35 In Adams v Cape, the English Court of Appeal refused to accept that the creation of a subsidiary to avoid future liabilities amounted to such a defeating of rights.36
In the United States the doctrine has been given slightly wider scope, as it also applies to cases where the parent company has ignored the formalities of its subsidiary’s separate legal existence.37 This includes actions like the commingling of funds and the failure to maintain distinct accounting records.38 Furthermore, an American court will take into account the fact the a parent company has left the subsidiary with insufficient assets to satisfy a potential judgment when deciding whether to pierce the veil.39 However, Sarah Coleman and Jonathan Friedler note that the doctrine has never been applied to a publically held company, and is “functionally irrelevant” to multinational enterprises.40
31 Muchlinski, above n 5, at 919.
32 Muchlinski, above n 22, at 308.
33 Coleman and Friedler, above n 1, at 199.
34 Phillip Blumberg The Multinational Challenge to Corporation Law: The Search for a New Corporate Personality (Oxford University Press, New York, 1993) at 67.
35 Prest v Petrodel Resources Ltd [2013] UKSC 34, [2013] 2 AC 415 at [20], [28], [34] and [35] per Lord Sumption.
36 Adams v Cape Industries [1990] 1 Ch 433 (CA) at 554.
37 Nygh, above n 4, at 66.
38 Coleman and Friedler, above n 1, at 199.
39 At 199.
40 At 199.
In New Zealand, there is s 271 of the Companies Act 1993, which allows the court to order a parent company to contribute to the assets of an insolvent subsidiary that has been unable to satisfy all its debts, if the court deems it “just and equitable to do so”. However, in practice this section has hardly ever been used.41 The guidelines provided for when it is “just and equitable” broadly mirror the considerations used in the United States.42
The feature shared by both existing legal avenues is that their success varies from case to case, because they are based around contextual exceptions to the rule of limited liability. This uncertainty creates its own problem, because it is difficult for companies to predict ex ante when they will face liability, which makes management planning harder.43 That is the main reason why this paper will seek to challenge the justifiability of the underlying rule itself, rather than advocate for expanding the existing exceptions. Such a solution would be better, not only for torts victims, but also for companies themselves.
1.3 Enterprise Liability
The proposal made by this paper should not be confused with enterprise liability, which is a wider ranging jurisprudential doctrine that considers corporate groups as a single entity for a number of different legal purposes, not just tort liability.44 There is little support for this doctrine around the world, other than in India, where it was developed by the courts in Mehta v Union of India as a response to the Bhopal disaster.45 Though originally introduced to deal with ultrahazardous torts, it has since been applied in other cases.46
41James Tocher “Contribution Orders – A Case Note on Lewis Holdings v Steel and Tube Holdings” [2015] NZLawStuJl 10; (2015) 3 NZLSJ 601 at 601. The only major case is Lewis Holdings Ltd v Steel and Tube Holdings Ltd [2014] NZHC 3311.
42 Companies Act 1993, s 272(1). See generally Lewis Holdings, above n 41, at [34]–[52]. See also Tocher, above
n 41, at 609–610.
43 Muchlinski, above n 5, at 923.
44 Blumberg, above n 34, at 119–120.
45 Mehta v Union of India [1986] INSC 282; (1987) 1 SCR 819 at 844. See generally Abhi Raghunathan “The Grand Trunk Road
from Saloman to Mehta: Economic Development and Enterprise Liability in India” (2012) 100 Geo LJ 571 at 597 and 601.
46 Raghunathan, above n 45, at 595. See State of Uttar Pradesh v Renusagar (1988) 1 SCR 627 at 629-630
where the Supreme Court of India found that a subsidiary was entitled to pay the same lower duty rate for electricity as its parent company. See also Novartis Ag v Adarsh Pharmaceuticals (2004) 3 CTC 95 (MHC) at [14] where the Madras High Court found that a subsidiary enjoyed the rights of a patent owned by its parent company.
It is unlikely that a similar doctrine would develop judicially in other Commonwealth jurisdictions, because past attempts have already failed.47 In DHN Food Distributors v Tower Hamlets, Lord Denning MR attempted to create a judicial doctrine of enterprise liability.48 However, the correctness of this decision was subsequently challenged by the House of Lords in Woolfson v Strathclyde Regional Council, and then effectively confined to its facts by Adams v Cape.49 That is why this paper is written from the perspective of how lawmakers should respond, rather than the courts, which are confined to applying the traditional principle of limited liability.
47 Muchlinski, above n 5, at 924.
48 DHN Food Distributors Ltd v Tower Hamlets London Borough Council [1976] 1 WLR 852 (CA) at 860 as cited in Nygh, above n 4, at 69.
49 Woolfson v Strathclyde Regional Council 1978 SLT 159 (HL) at 161 per Lord Keith of Kinkel as cited in Nygh,
above n 4, at 69; Adams v Cape, above n 36, at 536 as cited in Nygh, above n 4, at 70.
CHAPTER TWO: JUSTIFICATIONS FOR LIMITED LIABILITY
It might be thought that limited liability is simply a corollary of the principle that a company is a separate legal entity, distinct from its shareholders. However, these two concepts are logically distinct and also have different historical origins. This chapter will argue that the basis for allowing limited liability is pragmatic rather than principled.
Limited liability is the idea that a shareholder will not be held personally responsible for the liabilities of their company. While the separate legal identity of that company is a relevant factor behind such a rule, it is not sufficient to explain it alone, because the law does sometimes hold people responsible for the liabilities of others.50 As Phillip Blumberg puts it, “[limited liability] arose in the wake of the acceptance of the entity concept, but not as a necessary consequence.”51
2.1 History
This proposition is evident in the historical development of company law in England. Unlike in the United States, where modern companies developed out of state granted incorporation charters, in England the company began life in joint stock associations,52 because incorporation charters were too expensive and rarely granted (typically only to banks).53 During the first half of the 19th century, limited liability was a “highly controversial political issue” in England.54 It was not until 1855 that general incorporation with limited liability was allowed, after some early experiments in granting limited liability to railways proved economically beneficial.55 This is important for a couple of reasons. Firstly, the fact that joint stock associations were viewed more as partnerships, than separate legal entities, indicates that separate legal identity did not play a decisive role in adopting limited liability. Secondly, the fact that corporations with separate legal identity existed prior to 1855
50 Mendelson, above n 3, at 1213. Vicarious liability is one example. For further discussion see 2.2 Principles.
51 Blumberg, above n 8, at 577.
52 “Originally conceived as a form of partnership, the joint stock association developed into an unincorporated association organized under a deed of settlement, with trustees owning the stock and holding it for the benefit of the members, who held transferable shares.” (At 581).
53 At 585.
54 At 583.
55 At 584.
indicates that limited liability is not a necessary consequence of that principle. Finally, the fact that the political decision to adopt limited liability was based on achieving economic outcomes, demonstrates the doctrine’s underlying pragmatic basis.
2.2 Principles
Accepting that there is no necessary connection between limited liability and separate legal identity, it is still worth considering whether there is nonetheless a sufficient connection between the two to justify continuing to allow limited liability on principled grounds. One way to arrive at such a conclusion is to argue that the law should not impose obligations on anyone to pay for the debts of others. Frank Easterbrook, suggests that limited liability, in this sense, is not unique to companies, but is the default rule for any transaction:56
Suppose a bank lends $100 to a partnership, and the partnership's liabilities later greatly exceed its assets. (Perhaps the partnership buries toxic waste and incurs stupendous costs of cleaning up the mess.) The bank may lose the $100, but it will not be required to contribute any additional capital. Its liability is limited to its investment, exactly as the shareholder's liability is limited in a corporation.
From this analogy to contractual creditors it can be argued that the separate identity of the company should be sufficient to limit the liability of its shareholders, and to do otherwise would be to unjustifiably impose liability on shareholders.
Others, however, such as Coleman and Friedler, see limited liability as a grant of privilege by the state, which can be taken away if it no longer satisfies the pragmatic purpose it was designed to achieve.57 They justify this position by reference to state incorporation charters which were designed to be for the public good, meaning this argument is less applicable to the historical English context. Kenneth Arrow justifies the position slightly differently, describing limited liability as a state intervention to enable risk shifting not available in the free market due to imperfections in insurance markets.58 Either way, this school of thought clearly views shareholder liability as the default position. Paddy Ireland identifies the likely reason for this; unlike a loan, or a supply of goods or services, shareholding is a form of
56 Frank Easterbrook and Daniel Fischel “Limited Liability and the Corporation” (1985) 52 U Chi L Rev 89 at 90.
57 Coleman and Friedler, above n 1, at 194.
58 Kenneth Arrow Essays in the Theory of Risk-bearing (North-Holland Publishing Company, Amsterdam, 1971) at 139–140.
ownership that comes with a degree of control over what obligations are incurred.59 Typically the law associates control with responsibility, though the degree to which shareholders actually have control in modern companies can be questioned, due to the separation of ownership and control.60 Therein lies the dilemma these two opposing schools of thought are attempting to resolve.
It does not really matter whether limited liability or unlimited liability is the ‘natural’ position, as there would be exceptions to both rules. For instance, a contractual creditor might have liability for a tort committed by their contractual partner if they are vicariously liable, or if they are found to owe a direct duty of care.61 These exceptions must be explained on pragmatic grounds, as they represent a reversal of the default. Ultimately then, whichever way you look at it, limited liability must be justified on pragmatic grounds.
2.3 Torts and Corporate Groups
Next we consider how limited liability came to apply in this particular context. From the outset, limited liability has applied just the same in cases of tort as in cases of contract.62 However, tort liability was much rarer in the 19th century, especially on a scale that could threaten the solvency of a company, and thus shareholders’ wealth.63 For that reason, it seems that the unique concerns raised by tort liability were largely ignored in the debates over the introduction of limited liability, which focused instead on contractual creditors.64
Application of the rule to subsidiaries, on the other hand, was not an issue at all, because at the time companies were not allowed to hold shares in other companies.65 In the United States, it was not until 1889 that this was first allowed (starting in New Jersey), which was more than half a century after limited liability had become accepted in that country.66 In England this was possible earlier, in 1867, by the inclusion of a power to acquire shares in
59 Paddy Ireland “Limited Liability, Shareholder Rights and the Problem of Corporate Irresponsibility” (2010) 34 Camb J Econ 837 at 848.
60 This is probably not the case for subsidiaries, a point which will become important in Chapter 4.
61 The analogy is not perfect because in those cases the creditor will have technically committed their own
independent tort, however, it does suffice to show that under a limited liability regime the state still imposes arrangements approaching unlimited liability in some cases.
62 Hansmann and Kraakman, above n 9, at 1925.
63 At 1925.
64 At 1925.
65 Blumberg, above n 8, at 605.
66 At 607.
the company’s memorandum of association.67 By the time the question finally arose whether parent companies would enjoy limited liability with respect to their subsidiaries, “it was without discussion resolved by reference to the entity concept to which limited liability had become inseparably annexed.”68
The leading American case, subsequently relied on to demonstrate that limited liability applies within corporate groups, was actually about whether the proper interpretation of a railway carriage supply contract also required using those carriages on the railways of the railway company’s subsidiary.69 The only relevance of this case to limited liability was that it established the scope of an obligation by reference to the principle that a company is a separate legal entity from its shareholders. What was supposed to be a pragmatic doctrine was expanded dogmatically, and once again, with no consideration of the unique challenges that subsidiaries pose to its justifications. Blumberg describes this as a result of the “formalistic jurisprudence of the times”; a parent company is a shareholder, and shareholders enjoy limited liability, therefore parent companies enjoy limited liability.70
2.4 Economic Rationales
This revelation gives us reason to examine how well the pragmatic impetus for adopting limited liability applies to our context. A key economic reason for limited liability is that it results in more efficient contracting, by reducing transaction costs.71 When contracting, limited liability essentially involves shifting some of the risk of business failure from the shareholders to the contractual creditor.72 The creditor is able to compensate itself for taking on board this additional risk by altering the terms of the contract to its favour in other ways, such as by increasing the price.73 If the creditor is in a better position to bear that risk, then both parties have an incentive to place that risk on the creditor. The company will be able to reduce its risk for a lower cost than attempting to take its own precautionary measures.
67 At 608.
68 At 610.
69 Pullman's Palace Car Company v Missouri Pacific Railway [1885] USSC 243; 115 US 587 (1885) at 595–596 as cited in Blumberg, above n 8, at 609.
70 Blumberg, above n 34, at 59.
71 Hansmann and Kraakman, above n 9, at 1919.
72 At 1919.
73 At 1919.
Meanwhile the creditor can earn a profit, because the amount it charges for taking on board the risk can be greater than its own cost of taking precautionary measures against that risk.
Since parties have the freedom to contract around any rule of limited liability (or unlimited liability), through a personal guarantee or indemnity, the rule merely provides a default position from which to negotiate.74 The best default position is the one that the majority of contracts would end up reflecting, because this reduces the transactions costs involved in negotiating a reversal of the rule.75 It is desirable to have lower transaction costs as these deter economic activity on the margins, by reducing the profitability of some transactions. Evidence suggests that the better majoritarian default is to have a rule of limited liability, because creditors are typically better risk bearers than shareholders.76 It is also worth adding that without limited liability all the risk of investment would lie with shareholders, so limited liability facilitates a degree of risk sharing with creditors, which is the optimal contractual arrangement to deal with systematic (economy-wide) risk that cannot be diversified away.77
None of the above applies to a tort claimant, who does not have the opportunity to negotiate compensation ex ante for the risk they bear by virtue of being an involuntary creditor.78 Thus, this reason is inept to justify extending limited liability to torts, however, it does justify retaining limited liability for voluntary creditors generally.79 This also extends to voluntary creditors of subsidiaries, as (except in cases of fraud already dealt with by veil piercing) those creditors should be fully aware that they are negotiating with the subsidiary alone, and not the parent company.
The justification for allowing limited liability in torts cases is that it enables the efficient operation of equity markets.80 Companies, especially those that are publically traded, typically involve “agency costs”, which arise from the separation of ownership and
74 At 1919.
75 Blumberg, above n 8, at 616.
76 Barak Atiram “Market Dynamics in Corporate Tort Externalization: The Hidden Assumption of Corporate Social Efficiency” (2009) 86 U Det Mercy L Rev 347 at 376.
77 Easterbrook and Fischel, above n 56, at 101.
78 Hansmann and Kraakman, above n 9, at 1920.
79 Note that this justification might also apply to other creditors who are arguably involuntary as well, which is an issue that will be addressed in Chapter Four. See 4.7 Who is a Torts Creditor?
80 Mendelson, above n 3, at 1217.
control.81 This is where shareholders are forced to monitor managers, because they cannot trust managers to act in their best interests, as mangers do not receive the full benefits of their actions.82 Without limited liability, a shareholder’s personal wealth is at risk, which increases their incentives to engage in monitoring of company operations, potential liabilities and the degree to which they would be responsible for those liabilities.83 This information could be costly to obtain and act on (especially as the shareholder might not possess the relevant skills to properly understand and value the company’s particular risks), which in some cases might make shareholding in that company not worthwhile for a person who would have otherwise invested under a regime of limited liability.84 In effect, limited liability enables a class of passive shareholders who deal with their risk by diversification rather than monitoring, which gives companies greater access to capital at a reasonable price.85 This is an important consideration, as one of the key benefits of the corporate form has been to enable large scale projects that would have otherwise been difficult to fund.86
Admittedly, a pro-rata regime of unlimited liability would not entail the costs of monitoring other shareholders, as Henry Hansmann and Reinier Kraakman point out, but the need to monitor directors would merely be reduced, not eliminated entirely.87 However, the efficiency justification does not apply at all to a parent company, which acts more like a manager than an investor.88 The parent company will either be fully informed as to the operations of its subsidiary, or will easily have access to such information.89 Furthermore, corporate groups, especially large multinational enterprises, are able to internalise capital markets by using group wealth to allocate funds to various operations within the group, thus reducing the need for access to capital.90 Thus, the efficiency of equity markets might generally justify limited liability for torts claims, but this reason will not work in the case of a subsidiary.
81 Easterbrook and Fischel, above n 56, at 94.
82 At 94.
83 Mendelson, above n 3, at 1218.
84 At 1218.
85 Easterbrook and Fischel, above n 56, at 94.
86 Muchlinski, above n 5, at 917.
87 Hansmann and Kraakman, above n 9, at 1906. A pro-rata regime is where each shareholder is only liable for a proportion of the company’s debt equivalent to the proportion of their shareholding.
88 Blumberg, above n 8, at 624.
89 Mendelson, above n 3, at 1219.
90 Blumberg, above n 34, at 139.
This leaves one central justification for extending limited liability to parent companies in cases of tort; that it incentivises beneficial risk taking. The remainder of this paper will assess the viability of that claim as a justification for retaining the current scope of limited liability.
CHAPTER THREE: A BALANCE OF INCENTIVES
Arguably the most important benefit underlying the rationales for limited liability is that it decreases the risk that a shareholder must bear and thus incentivises investment in particular activities that would not occur but for the existence of such a rule.91 These risky activities can provide social benefit, as the increased production might have flow-on effects for the rest of society.92 However, this reduction of risk does not occur in a vacuum, it is made possible by shifting that risk onto someone else; torts victims. As Coleman and Friedler note, this is where limited liability comes into conflict with one of the key economic functions of tort law; to avoid moral hazard by internalising externalities.93 These opposing objectives in the law are opposite sides of the same coin; that coin being the taking of risks by companies that impact on third parties.
3.1 Conflicting Objectives in the Law
Tort law also seeks to achieve other objectives, such as compensating victims and penalising wrongdoers, but those objectives are not intrinsically opposed to the incentive structures established by limited liability. It would be possible, and perhaps preferable, to compensate victims through a state system like the Accident Compensation Scheme in New Zealand, and to punish companies through safety standards and the criminal law. However, forcing companies to take on board the cost of their actions to third parties is something intrinsically tied to whether shareholders can ignore the company’s tort obligations.
Hansmann and Kraakman argue that if companies are able to externalise the risk of their tort harms onto torts victims then those companies will have a number of poor incentives, including to spend less on precautionary measures than would be cost effective, and to overinvest in hazardous industries.94 Precautionary measures include safety standards such as requiring guard rails, but also technological upgrades that, for instance, could reduce
91 Blumberg, above n 8, at 616.
92 Stephen Presser “Thwarting the Killing of the Corporation: Limited Liability, Democracy, and Economics” (1992) 87 Nw U L Rev 148 at 170.
93 Coleman and Friedler, above n 1, at 198.
94 Hansmann and Kraakman, above n 9, at 1882–1883.
pollution.95 If companies fail to take up these measures when they would be cost effective, this not only results in a social harm, but according to Atiram might also result in an economic harm if this perverse activity creates a knee-jerk public reaction, leading to rushed government regulation.96
The way this moral hazard manifests is more complex than companies ignoring the consequences of all risks; the actual impact will depend on the projected magnitude of tort liability. If tort liability would exceed corporate assets then any additional risk becomes irrelevant to any decision made by the company, as it would not bear that cost if it eventuated.97 It is thus worth noting that unlimited liability would not entirely eliminate moral hazard, because there would still be instances where the parent company would not have sufficient assets to cover the liability of a tort claim, meaning any additional risk beyond that would still not factor into its decision making calculus.98
The countervailing concern is that tort liability is fairly unpredictable, both in probability of occurrence and in magnitude of damages, especially once consequential damage rules are taken into account.99 This uncertainty can be a significant deterrent to investment. For that exact reason though, it is a fair assumption that this uncertainty will be dealt with by insurance in most cases.100 Purchasing insurance introduces a contractual partner into the relationship who will negotiate terms with the company that reflect the degree of risk it bears, thus bringing the company’s incentives back in line with minimising risks when it is cost effective.101 This is because insurance premiums would go up if the company refused to implement precautionary measures. The involvement of insurance allows the company to continue operating with certainty of risk, however, now at an increased cost. Thus, unlimited liability acts more as an on-going operating cost, than a rare expense that only arises if tort liability exceeds the value of the company, as suggested by Hansmann and Kraakman.102
95 Mendelson, above n 3, at 1234.
96 Atiram, above n 76, at 366. Note that regulation itself provides an alternative way for a state to create incentives for a company to take precautionary measures, if it is actually done well rather than rushed. However, see below at n 159.
97 At 365.
98 Easterbrook and Fischel, above n 56, at 104.
99 Blumberg, above n 8, at 612.
100 Hansmann and Kraakman, above n 9, at 1890.
101 Easterbrook and Fischel, above n 56, at 108.
102 Atiram, above n 76, at 380.
Importantly, this interpretation highlights that the choice whether or not to extend limited liability to parent companies in cases of tort will have an immediate impact on the profitability of particular companies, and the attractiveness of investing in them. Easterbrook points out that “the social loss from reducing investment in certain types of projects - a consequence of seriously modifying limited liability - might far exceed the gains from reducing moral hazard.”103 Especially considering that very few individuals are affected by torts, while everyone benefits, to some degree, from increased economic activity (through taxes and access to cheaper goods and services). Therefore, in order to determine which side of the coin provides the more appropriate balance, we must consider all potential consequences of limited liability.
3.2 Empirical Evidence
The best balance to strike is an empirical question.104 Unfortunately there are few empirical studies on the effect of limited liability on risk taking, because regimes of unlimited liability have been rare throughout history.105 There are only two countries that have anything approaching unlimited liability today. The first is Germany’s Konzernrecht system, which, among other things, requires a parent company to pay outstanding debts of its subsidiaries.106 The German rule is not useful for our purposes as it is so limited in scope that it has been referred to as a “dead letter” by commentators.107 It only covers minority shareholders, not creditors, and it also only applies to Aktiengesellschaft, not Gesellschaft mit beschrankte Haftung, which are the more common type of company in Germany.108
The second is India’s judicially developed doctrine of enterprise liability. This is not a particularly useful example either, since the Indian court system is notoriously inefficient, and their tort law in particular is underutilised, meaning the rule probably has little effect.109 Some commentators have suggested that because India has maintained high levels of economic growth since the Mehta decision, we can conclude that enterprise liability has not
103 Easterbrook and Fischel, above n 56, at 104.
104 Hansmann and Kraakman, above n 9, at 1880.
105 Blumberg, above n 8, at 611.
106 Blumberg, above n 34, at 161.
107 At 162.
108 At 162.
109 Raghunathan, above n 45, at 597.
significantly deterred investment.110 However, Abhi Raghunathan notes that continued investment is largely a result of the government’s policy of deregulation, and in any case there is some evidence to suggest that, while the economy on the whole is growing, the manufacturing industry in particular (which would presumably be more affected by exposure to tort liability) is lagging substantially behind similar countries like China.111
There are also three notable historical instances of limited liability. The first is Victorian England, which as mentioned earlier did not adopt limited liability until well into the industrial revolution, yet still underwent massive industrial expansion during this period.112 The problem is that we obviously do not have data from an alternative world in which England adopted limited liability earlier on, so it is extremely difficult to compare.113 The second instance was California, which had pro-rata unlimited shareholder liability between 1849 and 1931, but again still maintained a reasonable rate of growth.114 California is an interesting example as their regime continued substantially into the 20th century when the rest of the world was wholly committed to limited liability.115 Finally, for a comparative example, Massachusetts and Rhode Island retained unlimited liability until 1830 and 1847 respectively, while all neighbouring New England states had already adopted limited liability.116 These two states had the “most intense manufacturing activity” in the Union, and did not lose that position during the time they held out, perhaps indicating that unlimited liability did not impact their growth.117
There are numerous problems with relying on these examples. Firstly, the actual economic impact in each case is mere speculation by Blumberg, which is not based on economic data.118 Secondly, since these examples only assess economy-wide effects, they are a very blunt mechanism to assess the balance of risk taking by individual companies, in specific industries, at the time. Thirdly, as Blumberg himself recognises, the impact of the rule change would be predominantly influenced by the perceived importance of limited liability
110 At 600.
111 At 600.
112 Blumberg, above n 8, at 585.
113 This objection equally applies to the other examples given of India and California.
114 Blumberg, above n 8, at 597 and 599.
115 At 598.
116 Blumberg, above n 34, at 13.
117 At 13–14. Note, this could equally be reverse causation; the fact that those states had such strong manufacturing industries might have lessened the need to spur economic growth by adopting limited liability. 118 At 14.
by investors.119 In the 21st century limited liability has become so dominant that it is simply taken for granted by anyone outside of a particularly small niche of legal academia, so chances are investors would not react well to such a law change.
This perception harm is further exacerbated by the remaining observations. Fourthly, vastly reduced communication and transportation costs have massively increased the interconnectedness of the global market, meaning that capital flight is far more likely than in the 19th century. Fifthly, large modern corporate enterprises require capital on a scale not fathomable in the 19th century, meaning the necessity of limited liability might not be accurately reflected in historical examples.120 Finally, the cost of risk taking today is hardly comparable with that imposed by the sparse tort law of the 19th century. Therefore, it will be necessary to rely on theoretical economic analysis to assess the appropriate balance.
3.3 Degree of Problem in the Status Quo
The first theoretical consideration is the degree to which excessive risk taking, without adequate precaution, occurs in the status quo. Muchlinski notes that, as a matter of commercial practice, the vast majority of multinational enterprises would be unwilling to allow one of their subsidiaries to go bankrupt, and they will often guarantee their debts, or pay out creditors even if there was no guarantee.121 For example, after the 1989 Exxox Valdez oil spill off Alaska, Exxox could have had its subsidiary (which owned the ship) declare bankruptcy, but because of a “high level of public indignation” it chose instead to voluntarily pay for the clean-up costs.122 Obviously not all cases will attract such a high degree of public attention, but at the very least a parent company’s concern about its brand image will mitigate the extent to which limited liability is a problem, because it will provide an incentive to avoid embarrassing torts claims.123
Easterbrook attempts to justify the status quo, by claiming that companies still have an incentive to insure under a regime of limited liability.124 The reasoning for why companies
119 Blumberg, above n 8, at 586.
120 At 595.
121 Muchlinski, above n 22, at 334.
122 Mendelson, above n 3, at 1243.
123 Note this consideration would also apply to wholly domestic corporate groups, though arguably to a lesser extent, as such groups are less likely to be exposed to a high degree of public visibility.
124 Easterbrook and Fischel, above n 56, at 107.
would not insure is that under limited liability shareholders can easily diversify the risk of insolvency, removing the need for insurance. However, the decision to purchase insurance is usually made by managers, whose investment in the firm (their job and their reputation associated with it) cannot be so easily diversified.125 This supposedly gives managers an incentive to purchase insurance, so they can keep their job. Unfortunately, this reasoning is far less applicable to a subsidiary, where the director is typically also a manager of the parent company, meaning their job is therefore not at risk (they can also be given a newly incorporated subsidiary to run).126
This argument is also flawed for another reason; it assumes that the parent company is at risk of being sued for its torts, and that managers will be concerned about this.127 Barak Atiram suggests that the likelihood of a lawsuit is very low for a number of reasons.128 Firstly, torts victims have to find the company responsible, which is not always straightforward, especially for torts involving latent damage, and where complex group structures are involved. Secondly, in cases of group torts, they have to overcome the collective action problem (every victim has an incentive to leave the effort of initiating litigation to another victim).129 Thirdly, collection costs involved in the lawsuit, such as legal fees, have to be lower than the expected return.130 This can be problematic for group torts where the harm to any one individual is too low to justify legal action.131 Additionally, the expected return will often be artificially low as it is affected by the high degree of legal uncertainty in torts cases.132 Finally, the expected return is also affected by the existence of limited liability itself, which severely reduces the pool of assets from which to collect, even if the litigation is successful.133 For all these reasons, many torts victims will simply choose not to litigate at all, or to settle for much less than their claim is worth.134
125 At 107.
126 At 111.
127 Note that this assumption is not required in cases where the company expects to voluntarily co-operate, such as Exxox did. However, as pointed out above, this will not always be the case.
128 Atiram, above n 76, at 380.
129 At 385. Note, however, that this difficulty could be overcome by litigation structures, such as class actions, if they are available in the particular jurisdiction.
130 At 385.
131 At 384.
132 At 385.
133 It might be argued that all the preceding observations equally apply to suing the parent company under a regime of unlimited liability, and therefore Atiram’s argument does not provide a reason to introduce
unlimited liability. Note, however, that this last requirement is particular to limited liability, saving his argument from this objection.
134 At 385.
According to Atiram, this also acts as a cyclical incentive to incorporate subsidiaries and transfer risky activity to them, further exacerbating the above problem.135 As discussed earlier, under a regime of limited liability, shareholders will be ambivalent to risk exceeding in magnitude the total value of corporate assets, since they will not bear that cost if it eventuates. Therefore, smaller companies, like subsidiaries, are able to externalise greater proportions of their tort liability, because those companies are not in a position to bear much of that risk (especially compared to larger companies), meaning they are not required by their shareholders to take precautionary measures against it.136 This has the effect of lowering relative production costs for smaller companies, and therefore creates an incentive to operate business through such smaller companies, as it will be cheaper.137 This results in a race-to-the-bottom, where parents companies are forced to operate risky businesses through subsidiaries in order to keep up with their competitors who have done so.138 This further increases difficulty in identifying the responsible tortfeasor, legal uncertainty and decreases the expected return, as these subsidiaries will tend to have fewer assets. This results in even fewer torts victims litigating their cases, and therefore an even lower incentive to take precautionary measures, such as insuring.
3.4 Cheapest Cost Avoider
We have now established that there is a reasonably substantial problem in the status quo; that companies are unlikely to purchase insurance and take precautionary measures. Technically it is open to either party to insure against risks, as the potential torts victim could insure themselves as well.139 One pragmatic way to consider the rule of limited liability is on which party the law ought to place the burden of seeking insurance. The best answer to this is the party that is able to deal with the risk most efficiently (referred to as the “cheapest cost avoider”), which in the present context predominantly means the party able to insure most efficiently.140 There are a number of reasons why the parent company is the cheapest cost avoider here.141
135 At 383.
136 At 382.
137 At 382.
138 At 383.
139 Blumberg, above n 8, at 617.
140 Hansmann and Kraakman, above n 9, at 1917.
141 This is not to suggest that the parent company insure itself against the risk of unlimited liability, although that is another possibility. The comparison is to the parent company exercising its position as controlling
shareholder to require that the subsidiary purchase adequate insurance for itself.
Firstly, parent companies are more likely to be risk neutral than torts victims. A risk neutral party will be a cheaper cost avoider than a risk averse party, because risk averse parties over-react to risk and respond inefficiently, such as by purchasing more comprehensive insurance than they need.142 A risk neutral party, however, will only be deterred from causing harm and no more.143 Torts victims are most commonly natural persons, not companies.144 Companies are seen as more risk neutral than a natural person, as they make decisions in a boardroom context with a profit motive, which means those decisions are more likely to be informed.145 This indicates the burden of insurance is better placed on the parent company than the torts victim.
Secondly, the parent company has much better information about the nature of the risks involved, in order to make an efficient decision, than a potential torts victim would.146 The parent company will be aware of the operations of its subsidiary, and how dangerous those operations are, while a torts victim might not even be aware that the company exists, let alone how its operations could eventuate into a harm against them. Thirdly, insurance is cheaper to obtain for the company, as it can do so in bulk, whereas potential torts victims would each have to arrange for individual insurance, which would involve much larger transaction costs.147 Fourthly, the parent company will likely be able to diversify its risk by owning multiple subsidiaries, whereas a potential torts victim cannot really diversify the risk of being a torts victim.148
Fifthly, all of this assumes that such insurance is actually available on the market to purchase. Liability insurance is readily available to companies.149 On the other hand, insurance for a potential torts victim will vary in availability depending on the type of harm.150 For instance, health insurance is widely available in most countries, however, something more complex, like water contamination insurance, might not be readily
142 Hansmann and Kraakman, above n 9, at 1887.
143 At 1886.
144 Mendelson, above n 3, at 1226.
145 At 1224.
146 Muchlinski, above n 5, at 923.
147 Paul Halpern, Michael Trebilcock and Stuart Turnbull “An Economic Analysis of Limited Liability in Corporation Law” (1980) 30 UTLJ 117 at 146.
148 Except arguably in the case of product liability where they could purchase from a variety of different manufacturers. However, expecting consumers to take this into account when making a purchase is fanciful.
149 Mendelson, above n 3, at 1225.
150 At 1223.
accessible (at least not at an affordable price).151 Sixthly, even if insurance is technically available for potential torts victims, that does not mean they will all take it up. Many torts victims of multinational enterprises live in developing countries, where most people do not have such insurance.152 Furthermore, people do not typically insure for non-economic harms such as pain and suffering, loss of function, and death.153 This all indicates that many torts victims will not be insured, and will suffer loss if the burden of insurance is placed on them.
Paul Halpern, Michael Trebilcock and Stuart Turnbull aptly conclude that even if there are some torts victims who might be in a better position to insure than the parent company, they are the minority and it would be too difficult to distinguish them from the typical case.154 Therefore, the burden of insurance ought to be placed on the parent company rather than the potential torts victim. This conclusion is further supported by the observation that, if the company insures, that will influence its behaviour and reduce the likelihood of the tort occurring in the first place, whereas that is not true if the potential victim is the one to insure. Prevention is always better than a remedy after the fact.
151 At 1223.
152 Peter Muchlinski “The Bhopal Case: Controlling Ultrahazardous Industrial Activities Undertaken by Foreign Investors” (1987) 50 MLV 545 at 582.
153 David Leebron “Limited Liability, Torts Victims, and Creditors” (1991) 91 Colum L Rev 1565 at 1603.
154 Halpern, Trebilcock and Turnbull, above n 147, at 147.
CHAPTER FOUR: DESIGNING A REGIME OF UNLIMITED LIABILITY
Now that it is apparent that limited liability does create flawed incentives, it is necessary to consider what the best possible alternative would look like. Any reform would be best done by legislation, which could neatly lay out the various intricacies involved. Hansmann and Kraakman suggest that any such reform should only be implemented prospectively, and there should be a short transition period of around two years to allow companies to adjust their behaviour to the new regime, such as by arranging appropriate precautionary measures and buying insurance, or to sell their shares if they do not wish to bear that risk.155 Although this would leave some historical torts victims without redress, it would be unfair to impose retrospective liability on parent companies that were at the time following the law.156
4.1 Alternatives to Unlimited Liability
Although there have been other suggestions on how to counter the perverse incentives acting on companies under limited liability, the best solution is a regime of unlimited liability. Two such suggestions are minimum capitalisation requirements and compulsory liability insurance.157 These alternatives both suffer from the same problem; they are too generalised and inflexible, when in reality the requirements ought to be different for different industries in order to be efficient.158 Even if attempts were made to tailor to particular industries, these decisions would be made by government, which means they would be susceptible to political pressures. Atiram argues that price signals from unlimited liability would allow the market to respond efficiently, while government technocrats would be less informed about the risks faced by that particular industry, and (because these decisions often occur in the wake of a disaster) they would also be subject to post hoc public outrage, which might lead to setting the minimum levels too high, leading to overburdening
155 Hansmann and Kraakman, above n 9, at 1923.
156 It is also worth noting that the main reason for adopting unlimited liability, to alter the incentives of
companies to take precautionary measures, cannot possibly apply to historical torts anyway as the behaviour occurred in the past.
157 Hansmann and Kraakman, above n 9, at 1927. Note that minimum capitalisation requirements were
actually implemented in New Zealand under the capital maintenance doctrine until it was abolished by the Companies Act 1993; Andrew Beck and others Morison’s Company Law (NZ) (online looseleaf ed, LexisNexis) at [13.3].
158 Hansmann and Kraakman, above n 9, at 1927.
of those industries.159 The other downside to a minimum requirement is that companies are all forced to be risk averse, whereas under unlimited liability they are at least given the option to run the risk if they deem it efficient to do so.
4.2 Defining Subsidiary
Unlimited liability could take many forms. The first decision to be made, given this reform is only targeted at subsidiaries, is which companies will be caught by the rule. Wholly-owned subsidiaries are the easy case, because there is only one shareholder; the parent company. They are also the norm; 72 per cent of subsidiaries in American multinational enterprises are wholly-owned.160 However, partially-owned subsidiaries must also be covered by unlimited liability, otherwise companies could avoid the effect of the rule by simply adding a token shareholder with a single share.161 This raises two questions: what is a partially-owned subsidiary, and which shareholders of such a company would have unlimited liability?
Section 5 of the Companies Act 1993 already provides a reasonable answer to the first question by defining subsidiary in five different ways.162 Broadly speaking the Act defines subsidiary as a company that another company (the parent) holds more than half of the shares in, or is otherwise in a position to exercise majority control over, through specified mechanisms. This definition will be sufficient to identify when we are dealing with a subsidiary at all, and we will return to consider the intricacies of what is meant by “control” momentarily.
4.3 Which Shareholders are Liable?
In terms of which shareholders will have unlimited liability, there are broadly two schools of thought: share-based or control-based. The share-based approach would remove limited liability for all shareholders in any company classified as a subsidiary, while the control- based approach would only remove limited liability for shareholders in a position to control the company; basically just the parent company. Advocates of the share-based approach,
159 Atiram, above n 76, at 364. This criticism also applies to another alternative method raised earlier (at n 96); attempting to create regulations for the specific industries which raise concerns of excessively risky behaviour. 160 Blumberg, above n 8, at 626.
161 Hansmann and Kraakman, above n 9, at 1931.
162 See Schedule. Section 5 is also supplemented by sections 7 and 8.
like Hansmann and Kraakman, suggest adopting pro-rata unlimited liability, which means that each shareholder would only be liable for a proportion of the subsidiary’s liability that aligns with the proportion of their shareholding.163
This approach can be criticised for imposing undue liability on shareholders who are not the parent company. None of the special rationales, discussed in Chapter Two, for why the subsidiary context is different from a typical company apply to these shareholders. For them, this investment is no different from an investment in any other company, meaning the share-based approach would make limited liability an arbitrary feature, which could change based on the actions of other shareholders (if another company acquired a majority shareholding in the company, for instance). Such a rule would make investments in partially-owned subsidiaries far less attractive than other companies, which could have a detrimental effect on the ability of these subsidiaries to raise capital. Richard Posner notes that all of this is true whether the shareholder we are discussing is a natural person or another company, so we cannot solve this unfairness problem by only applying unlimited liability to companies.164
The advantage of the control-based approach is that it gets around this unfairness and is consistent with control being the factor that distinguishes a subsidiary from other companies. The challenge with this approach lies in the difficulty of defining the boundaries. There is a classic trade-off between certainty and flexibility here; the more clearly the rule is defined, the easier it will be to avoid; but the more scope that is given to judges in applying the rule, the greater the concerns about application in unintended or unpredictable cases. That is why s 5 of the Companies Act provides a nice compromise between a flat cut-off percentage of shareholding, and an entirely contextual test of control.
The philosophy behind s 5 seems to be that ‘more than half’ is the shareholding required for control. All the alternatives tests are essentially ways of having the equivalent of such a shareholding.165 It does not seek to catch a shareholding company that has effective control as a matter of fact, in spite of owning less than half of the shares (minority control). Hansmann and Kraakman raise the concern that such a blunt cut-off creates an incentive to
163 Hansmann and Kraakman, above n 9, at 1892.
164 Richard Posner Economic Analysis of Law (4th ed, Little, Brown and Company, Boston, 1992) at 407.
165 Such as by owning more than half of the shares eligible to vote at a meeting, even if that is less than half of all shares; s 5(1)(a)(ii); or by having a power to appoint a majority of the board; s 7.
own 49 per cent of shares, and ensure that the rest of the shares are widely held, such as in a publicly listed company, so that the other 51 per cent cannot effectively oppose its interests, thus giving it de facto control while retaining limited liability.166 In fact, according to the United States Securities and Exchange Commission there is a common belief that “ownership of 20% voting power in a widely held company in most instances constitutes control”.167 However, it is equally possible that a 20 per cent shareholding would not constitute control in many other instances, most obviously if there is only one other shareholder who owns 80 per cent.
To deal with this type of scenario, Nina Mendelson suggests an entirely contextual judicial test of whether the company has the “capacity to control” the subsidiary.168 Any solution, such as this, which can account for the wide vagaries of minority control, must necessarily be vague, and therefore uncertain. David Leebron provides an important reminder that the context of this reform – fixing the incentives of corporate groups with regard to risk-taking – requires a degree of certainty on behalf of those companies, so they can respond accordingly.169 As a shareholder, whether or not a company will expose you to unlimited tort liability is likely to be an important factor in any investment decision, so too much uncertainty on whether this rule applies in any particular case would likely have a chilling effect on investment. Leebron further makes the point that the marginal benefit of extending unlimited liability to minority control scenarios, as compared to just imposing wholesale unlimited liability on all companies, is small.170 Although a 50 per cent cut-off is not perfect, it avoids uncertainty, and still forces parent companies of partially-owned subsidiaries to make a choice between seeking to retain limited liability while risking that they lose control, and giving up limited liability in order to guarantee control.171 The gravity of this trade-off is apparent when considering that the parent company might on occasion need a supermajority to exercise control, such as for decisions requiring a special resolution under s 106 of the Companies Act.
166 Hansmann and Kraakman, above n 9, at 1931.
167 Mendelson, above n 3, at 1273. This is also the same proportion used in s 237 of the Financial Markets
Conduct Act 2013 to define a “relevant interest” for the purposes of that legislation. However the implications of that Act can fairly be described as less severe than the removal of limited liability.
168 Mendelson, above n 3, at 1272.
169 Leebron, above n 153, at 1619.
170 At 1622.
171 The downsides to this approach will be considered in greater depth in Chapter Five. See 5.3 Avoidance Methods.
4.4 Degree of Liability
A further choice that must be made is the degree to which the parent company will be liable. Leebron argues it should only be liable in proportion to its shareholding, as any greater liability than that would make the parent company “inefficiently risk averse”.172 On the contrary, Mendelson argues the parent company should be liable for the full amount of the loss, pointing out that its position of control gives it benefits in excess of its proportion of shareholding, and thus pro-rata liability would not provide sufficient deterrent.173 These benefits include synergies with related businesses such as the ability to bulk purchase, the ability to engage in self-dealing, tax and accounting advantages (corporate group rules), and the certainty of being able to direct the subsidiary’s business operations.174 Thus, either approach will overshoot the ideal degree of liability, but something in-between is not practically feasible, so a choice must be made. Mendelson’s approach is the better option, for reasons that will become apparent after discussing how parent companies are likely to respond to this reform, in Chapter Five.175
4.5 Attachment of Liability
Another technical matter is at which point liability will attach to the parent company. This is important if ownership of the subsidiary changes, or if it ever ceases to be a subsidiary altogether. The two options are when the tort is committed (occurrence rule) and when the judgment is delivered (judgment rule).176 The trade-off, as Hansmann and Kraakman describe it, is that the occurrence rule is fairer, but more difficult to apply (it involves complex questions about the precise time a tort occurred), while the judgment rule is simple, but creates opportunities for evasion.177 Leebron warns of the strong incentive to “sanitize” high-risk subsidiaries by periodically selling them and starting new ones if the judgment rule were to be adopted.178 Mendelson claims that the occurrence rule is better for a control-
172 Leebron, above n 153, at 1622.
173 Mendelson, above n 3, at 1283.
174 At 1253-1255.
175 See 5.1 Over-deterrence of Risk-taking.
176 Hansmann and Kraakman, above n 9, at 1896.
177 At 1896.
178 Leebron, above n 153, at 1625.
based approach, because the administrative concerns of that rule are much less when the only relevant shareholder is the parent company.179
However, this does not address another problem with the occurrence rule; that there are perverse incentives on the present owners of such a company to abuse their position of present control to shift any burden of liability to an historical parent company, such as by asset stripping.180 An obvious solution to this problem would be to impose liability for historical torts on both past and present parent companies. However, this has its own problems. For one, it would not assist if the company no longer has a majority corporate shareholder, because then no shareholder would be subject to unlimited liability. More importantly though, imposing unlimited liability for actions over which the new owners have no control, and probably very little ability to assess before purchase,181 would create a massive deterrent to corporate takeovers (which are typically seen as positive because they result in better management).182 Leebron points out that it would be open for the purchaser to negotiate an indemnity, but this simply represents another cost, and would still act as a deterrent.183 In reality, many purchasers would seek to avoid this rule altogether by simply purchasing the assets of the subsidiary, rather than shares in the subsidiary itself. A better solution, suggested by Hansmann and Kraakman, would be to impose a duty on the current owners to avoid opportunism, in much the same way that the director’s duty to act in the best interests of the company (Companies Act, s 131) extends to creditors when the company is nearing insolvency.184 Thus, the occurrence rule is better on balance.
4.6 Enterprise Liability
A final concern with the control-based approach is that it only applies to the parent company, while corporate groups, especially multinational enterprises, are often complex
179 Mendelson, above n 3, at 1279.
180 Hansmann and Kraakman, above n 9, at 1899.
181 It might be argued that financial reporting rules in most jurisdictions would require the seller to divulge this
information. However, any tort which would affect a new owner would have to be an historical tort, which in many instances the seller will not even be aware of, because it involves a latent defect, or requires knowledge which is unknown at the time (as happened in Cambridge Water Co Ltd v Eastern County Leather Plc [1993] UKHL 12; [1994] 2 AC 264 (HL)).
182 Leebron, above n 153, at 1624.
183 At 1624.
184 Hansmann and Kraakman, above n 9, at 1899. The details of this duty would have to be refined by
Parliament when enacting this reform. To avoid becoming overbearing it might be desirable to set the standard for this duty rather low, and simply have it prevent asset stripping when the director is aware of a potential torts claim against the company. What constitutes asset stripping would again have to be defined.
decentralised webs of companies.185 This has led some commentators to explore the idea of imposing liability on the entire enterprise as a whole.186 If the parent company is the only entity held liable, there is an incentive for corporate groups to create shell companies to own their subsidiaries, which have no assets of their own. However, the scale of this problem is overblown, because necessarily those parent companies will have to own at least one asset; shares in the other subsidiaries. Even if the immediate parent company does not own shares in any other companies, some parent company higher in the hierarchy must (assuming that there are other subsidiaries, which there must be for this to even be a problem). Such higher parent companies are also considered to be the parent company of the subsidiary by s 5(1)(b).
Leebron suggests that it is insufficient to rely on the parent company’s shareholdings for compensation, as this raises difficult issues of priority upon insolvency.187 However, insolvency is really a special case, where priority rules might need to be adjusted. To some degree this has already been recognised in New Zealand with s 271 of the Companies Act, which was referred to in Chapter One. In most cases a torts claim will not force the entire corporate group into insolvency, and so in order to continue business, at some point up the hierarchy a parent company will have to foot the bill by sourcing money from one of its subsidiaries.188
This avoids the need to venture into the dangerous territory of imposing liability on the entire enterprise. There are good reasons not to adopt such a regime. Unlike imposing liability on a parent company, which is simply a matter of determining the obligations attached to equity investment, enterprise liability violates the principle of separate legal identity, because the sibling subsidiaries have no investment in the company whose debts they are held liable for. Creating liability at the level of the enterprise would involve a conceptual endorsement of the enterprise as a legal unit. 189 Meanwhile the law would continue to allocate rights, such as to contract or own property, only to individual companies. As Blumberg puts it, “a legal unit without rights, particularly the rights
185 Muchlinski, above n 22, at 321.
186 At 321–326.
187 Leebron, above n 153, at 1623.
188 Another option would be to find external financing, such as a loan or issuing shares. Even more likely is that parent companies, being aware of this inevitability, simply will not bother divesting themselves of assets.
189 The law does already recognise enterprises as a legal unit in other areas, such as taxation and financial
reporting. However, this particular mismatch of rights and liabilities is not present in those contexts.
traditionally recognized as fundamental in the case of all other units, is an odd unit indeed.”190
4.7 Who is a Torts Creditor?
Now that we have defined who will be liable, we need to address who they will possibly be liable to. At first glance it may seem simple to just say this reform applies to torts victims, which brings to mind the obvious cases like property damage and personal injury.191 However, this reform explicitly excludes contractual creditors, and the line between tort and contract can often be blurry. Many torts arise from contractual relationships, or are at least closely related to them, such as product liability, liability for professional services, and misrepresentation.192 More generally parties might agree to contractual terms relating to their tort liability, such as exclusion and limitation clauses.
One approach to dealing with this would be to retain limited liability in any torts case where a contractual relationship governs the allocation of risk between the parties. This is consistent with the rationale for treating torts cases differently from contract, as it recognises that there was an opportunity for the potential torts victim to negotiate ex ante compensation for any risk of a tort occurring. However, this approach would exclude the torts victims identified above, which runs contrary to the notion that tort is a separate and distinct claim from contract, and can exist even where there is a contractual relationship.193 Product liability in particular has been developed to override contractual relationships, in some cases, in order to protect consumers in the face of negligent manufacturing and minimal bargaining power.194 To be fair, misrepresentation can probably be excluded, as the contract in that case is the very context for the tort, rather than a pre-existing relationship governing tort liability.
190 Blumberg, above n 34, at 237.
191 Note, however, that victims of a personal injury tort will not be able to sue in New Zealand if their injury
was suffered in New Zealand. See Accident Compensation Act 2001, ss 20(1)(a) and 317(1)(a). Personal injury torts are, however, still relevant to this paper as some of the torts victims covered by this reform will be located overseas.
192 Muchlinski, above n 22, at 325.
193 See generally RM Turton & Co Ltd v Kerslake and Partners [2000] NZCA 115; [2000] 3 NZLR 406 (CA) at [8]–[9].
194 Peter Muchlinski “Holding Multinationals to Account: Recent Developments in English Litigation and the Company Law Review” (2002) 23 Co Law 168 at 174.
Hansmann and Kraakman offer a refined version of this approach, which is a judicial test assessing whether that category of victim would have been able to assess the risks of dealing with the company prior to the loss occurring, and whether they would have had the opportunity to seek adequate compensation for those risks.195 Such a test is vague and leaves too much scope to courts to judicially develop this reform into the future. For instance, some commentators have already suggested that unlimited liability should extend to cover employees and trade creditors, who also have low-bargaining power.196 There are good reasons to suggest that even though these groups are not individually able to negotiate, they are still compensated for their risk through the market price, which itself is often set by more powerful groups, such as indenture trustees and unions.197 Either way, this is a policy choice that should be made by Parliament, and in the interim it is important to have certainty in this area of the law, as the removal of limited liability is a large change for many businesses.
To solve both these problems it would be best to apply unlimited liability to any action arising in tort, even where there is a contractual relationship. This has the certainty of only applying to torts cases specifically, while at the same time not leaving out some torts victims. This would make bringing a claim in tort vastly superior to bringing a claim in contract, but this is not a cause for concern, as it merely represents an implication of allowing concurrent liability in contract and tort.198
Problematically though, this approach might also catch some truly voluntary arrangements, which would disturb the principle in contractual negligence cases that the court should not disturb the parties’ allocation of risk.199 However, the very existence of this principle mitigates the problem, since it means that the court will not find a duty of care in the first place if they consider that there was a clear allocation of risk between the parties. Thus, there would be no liability at all, and the question of unlimited liability would not even arise. This means that there are no large concerns of the rule overextending beyond the principled basis upon which it was founded.
195 Hansmann and Kraakman, above n 9, at 1921.
196 Blumberg, above n 8, at 618.
197 Easterbrook and Fischel, above n 56, at 105.
198 See generally Henderson v Merrett Syndicates Ltd (No 1) [1994] UKHL 5; [1995] 2 AC 145 (HL) at 193–194.
199 Rolls-Royce New Zealand Ltd v Carter Holt Harvey [2004] NZCA 97; [2005] 1 NZLR 324 (CA) at [118].
It is also worth noting that it would be open to companies to write limited liability clauses for tort into their contracts, in an attempt to override this rule. Since the purpose of imposing unlimited liability in tort is that the parties do not have the opportunity to bargain, it seems inconsistent to allow such clauses. However, the same tension exists generally with exclusion and limitation clauses, which are often put in contracts to avoid consumer protection type liability, which is also based on a lack of opportunity to bargain. It therefore seems consistent to allow the courts to use the existing principles of contractual interpretation to deal with limited liability clauses, by limiting their effect when necessary, or placing expectations on how they are used.200
200 See generally DHL International (NZ) Ltd v Richmond Ltd [1993] 33 NZLR 10 (CA) at 17–18.
CHAPTER FIVE: HOW PARENT COMPANIES WILL RESPOND
Now that it is clear what a regime of unlimited liability would look like, it is possible to return to the concerns of Chapter Three and assess whether this new regime would result in a better balance of incentives for risk taking than the status quo. Obviously this regime will increase the cost of business, which on the margins will make some companies that were otherwise profitable unprofitable, and therefore might cause them to shutdown (or never start up in the first place). Hansmann and Kraakman embrace this outcome, saying it is the intended effect of the reform, as those businesses were too hazardous to exist in a world where they had to bear the cost of their externalities.201 However, it is not that simple, because there is not necessarily a one-to-one correlation between profit and social benefit, especially when that profit is dependent on the complexities of the insurance market. Arrow provides the counterexample of research and development, which is one of the most risky activities in terms of profit, but also one of the most beneficial to society if it is successful.202 For instance, pharmaceutical research can often lead to tort liability, but is widely agreed to be beneficial.203 Mendelson suggests these are exceptions rather than the rule, and that those particular industries could be appropriately compensated through targeted subsidies, or beneficial tax treatment.204 However, adequate compensatory measures are not guaranteed to occur, and governments are not always good at determining which activities are beneficial, so some general concern remains.
5.1 Over-deterrence of Risk Taking
The above concerns apply, even if there is no over-deterrence. If there is over-deterrence, those concerns are not only exacerbated, but they are also generalised to any industry, as the lost production occurs with insufficient corresponding benefit to potential torts victims. We identified earlier that holding the parent company liable for the full amount of the tort liability will result in some over-deterrence. However, the larger the shareholding that the parent company has, the smaller the effect of over-deterrence (because it now enjoy more of
201 Hansmann and Kraakman, above n 9, at 1888.
202 Arrow, above n 58, at 138.
203 Mendelson, above n 3, at 1297.
204 At 1298.
the benefits).205 The over-deterrence is therefore likely to be minimal, because under a control-based regime many parent companies will probably fully buy out their partially- owned subsidiaries.206 This is because any parent company that decides to maintain a controlling shareholding has explicitly made a trade-off to favour its ability to control the company over its ability to shield itself from liability, meaning it will be less willing to accommodate minority interests that might jeopardise its new high-risk position of control.207 This was the reason alluded to in Chapter Four, for why Mendelson’s approach is preferable.208
Other than this structural over-deterrence, the parent company is unlikely to be irrationally over-deterred as it is probably risk neutral. This means it is likely to purchase an adequate amount of insurance, and continue its behaviour (though adjusted accordingly).209 Ideally, having insurance means there will be no over-deterrence, because the insurance contract will put a reasonably accurate price to the risk of any activity engaged in.210 However, insurance markets are imperfect, as there might not be competitive rates, and insurance is not offered to cover every loss.211 This is because insurance companies need to be profitable as well. They operate by diversifying the risks of their insurance contracts, but some risks are economy-wide and cannot be diversified, so no insurance will be offered for those risks.212 Furthermore, offering too comprehensive insurance would create a moral hazard; the insured party might act more carelessly as a result of not bearing enough of the costs of their actions.213 Since the insurance company calculates its price based on expected behaviour, moral hazard makes that calculation more unpredictable, and therefore insurance companies tend not to offer such insurance.214 Finally, the price of insurance is often too high to be economically efficient for many companies, due to the adverse selection problem.215 This problem arises from the fact that insurance purchase is voluntary, so only companies with higher than average risk purchase the insurance (otherwise it would be
205 At 1295.
206 At 1293.
207 Leebron, above n 153, at 1621.
208 See 4.4 Degree of Liability.
209 Hansmann and Kraakman, above n 9, at 1890.
210 Mendelson, above n 3, at 1294.
211 Easterbrook and Fischel, above n 56, at 102.
212 At 103.
213 Arrow, above n 58, at 142.
214 At 142.
215 Halpern, Trebilcock and Turnbull, above n 147, at 141.
inefficient for them).216 This has the impact of pushing the price up because the insurance company’s customer pool is more risky than anticipated in its original pricing, so it is forced to adjust it.217
On the whole, there will be some loss of beneficial risk taking as a result of imperfections in this reform and the insurance market. However, because this reform only applies to subsidiaries, the top layer of corporate veil (for the parent company) continues to protect the individual shareholders of the group. Thus the group as a whole is still able to take risks as any other company could, it is just not afforded extra protection for isolating particular parts of its business.
5.2 Disaggregation of Industry
For that reason, corporate groups are much less likely to segregate their businesses into subsidiaries under a regime of unlimited liability; there is a comparative advantage to operate as an entirely separate firm.218 The first consequence of this is that companies will be less likely to purchase controlling interests in other companies, as control would expose them to unlimited liability.219 This will, on net, result in fewer large shareholders in widely- held companies, which will increase agency costs (mismatch in incentives between managers and shareholders), because large shareholders, referred to as “activist shareholders”, are particularly effective at holding managers to account.220
Additionally, companies would now be likely to use separate companies to engage in risky activities, in place of ones they own a controlling interest in, such as by outsourcing to contractors.221 This might be less desirable than vertical and horizontal integration, for a couple of reasons. Firstly, smaller companies will be less able to afford liability insurance.222 However, this must be compared to the status quo, where many of these companies do not purchase insurance anyway. Secondly, this would mean firms would be smaller on average,
216 At 141.
217 At 141.
218 Easterbrook and Fischel, above n 56, at 111.
219 Mendelson, above n 3, at 1293.
220 At 1293.
221 Atiram, above n 76, at 396.
222 Easterbrook and Fischel, above n 56, at 111.
resulting in reduced benefits from economies of scale.223 However, this loss of efficiency should not be overstated, because these companies are still profit-maximising, and to the extent that the efficiency gains from integration outweigh any costs from exposure to tort liability, they will still integrate.224 Further, there is some evidence that disaggregation is not the dominant strategy. After the Exxon Valdez oil spill some oil companies stated that the best way to avoid liability was to own the tankers directly and ensure they were managed safely.225
While outsourcing might not be a huge problem, some integration will inevitably be lost, so it is worth considering whether that is a cost worth bearing for the benefits of this reform. The value of that integration varies widely depending on the type of subsidiary involved. For artificial divisions of a company that more closely represent departments, there will be little value lost, but in the case of conglomerates there might be a genuine loss.226 Conglomerates are corporate groups consisting of distinct businesses, each often run by separate managers, prioritising the profit of each individual company, rather than the group.227 It seems odd to penalise such companies just for the identity of their shareholders, especially when conglomerates might be considered particularly effective forms of business due to their access to the internal capital market of the corporate group.228 ‘Foreign branch’ subsidiaries are another example that are often run as independent businesses.229
Formulating a clean test to distinguish between the ‘good’ subsidiaries and the ‘bad’ subsidiaries would be too difficult, so these losses are a necessary cost of this reform. It is likely, however, that many conglomerates will continue to exist, because there are other advantages to corporate groups than limited liability, such as reduced transaction costs, improved information sharing, coordination benefits, reduced contractual enforcement costs, and for multinational enterprises the ability to take advantage of international trade
223 Hansmann and Kraakman, above n 9, at 1914.
224 Leebron, above n 153, at 1615.
225 Mendelson, above n 3, at 1301.
226 Leebron, above n 153, at 1617–1618.
227 Posner, above n 164, at 408.
228 Blumberg, above n 34, at 145. Contrast Brandon Cline, Jacqueline Garner and Adam Yore “Exploitation of
the Internal Capital Market and Avoidance of Outside Monitoring” (2014) 25 Journal of Corporate Finance 234 at 234–235, who argue that conglomerates are actually less efficient than separate companies because reliance on internal capital markets enables directors to exercise investment discretion without the market providing a check that such investment is efficient.
229 Muchlinski, above n 22, at 51.
agreements and tax treaties.230 It is also worth remembering that corporate groups still retain limited liability for contract, and in most industries the risk of torts claims will probably be seen as too low to warrant changing the entire structure of the group.
5.3 Avoidance Methods
Thus far we have assessed the costs of reduced beneficial risk taking, but these must be weighed against the benefits of reduced excessive risk taking. The size of those benefits depend on the effectiveness of this reform. It is therefore worth addressing the various ways corporate groups could seek to avoid its application.231
The first method of avoidance, known as the “high-roller problem”, is where high-risk subsidiaries are sold to a natural person shareholder with few personal assets, and then financed through debt rather than equity.232 By converting its equity to debt, the parent company escapes the application of unlimited liability. However, for this to be an effective avoidance method the parent company also needs some way of maintaining control over the subsidiary’s operations. This could be done by a contractual arrangement with the high- roller, or by having the high-roller own the shares on trust for the parent company, or the high-roller could simply be a director or other employee of the parent company. The trustee scenario would be caught by s 8(b)(i), which treats a “nominee” as though they are the parent company. The employee scenario might be caught by s 5(1)(a)(i), which deals with when the parent company “controls the composition of the board of the company”. Section 7 defines this as when the parent company has a “power” to appoint the board, including when appointment requires its consent or automatically follows from an appointment at the parent company. An employment requirement to act as a director could reasonably be interpreted as a “power” in a similar manner.
230 Atiram, above n 76, at 373.
231 It is worth noting at the outset that many of these avoidance methods depend for their effectiveness on the
particular way this reform was designed in Chapter Four. Some of these methods might have been prevented by adopting a different design, however it must be remembered that doing so would come at another cost. One might conclude that the cost discussed in Chapter Four, such as greater uncertainty, is more acceptable than the cost of a particular avoidance method. These are difficult calls to make, but ultimately the conclusion is still the same, that implementing this reform requires the acceptance of at least one of these two alternative costs.
232 Hansmann and Kraakman, above n 9, at 1911.
The contract scenario is more difficult, however, because beyond the letter of the contract the other party is still an autonomous agent. However, Hansmann and Kraakman suggest there are unlikely to be many people willing to take on the role of this type of high-roller, as there is a very particular set of requirements.233 They must have few personal assets, be willing to enter personal bankruptcy, have sufficient skills that the parent company trusts them to run the business, and have no moral qualms with involving themselves in what is essentially a scheme to cheat torts victims out of compensation.234 While this might seem like a limited category of people, it would only take a few to make a business out of it, and there are already people who have expressed interest in engaging in this activity, such as Texas oilman Kyle McAlister who has said he would “pack up his wife and three children, liquidate his US holdings and move to Switzerland” in order to do this.235 Parent companies could substantially reduce the risk high-rollers face by offering “support packages” in the case of personal bankruptcy, which would probably involve them leaving the country. However, it would be easy enough to prohibit this particular type of contract. To conclude, Mendelson makes the observation that this will at least be better than the status quo, because not every subsidiary would be substituted by a high-roller.236
The next avoidance method also uses contract, but is less focused on the use of debt. As Muchlinski points out, some multinational enterprises already run subsidiary-like business structures through contractual networks, such as franchise agreements, licensing agreements and international consortia.237 More corporate groups might now reorganise into contractual networks, retaining the advantages of interconnectivity, but without unlimited liability.
A further avoidance method, arising from the 50 per cent threshold, is for two independent companies to enter into a joint venture each with exactly half of the shares, which is therefore not ‘more than half’.238 Unlike the widely-held scenario originally considered, there is little uncertainty here, because there is only one other shareholder, and by the nature of a joint venture, the two shareholders will have come to some agreement about the manner in which the company will be run. Including the ‘exactly half’ scenario within the rule would not fix this problem, because all it would take is a token third shareholder to drop
233 At 1912.
234 At 1912.
235 At 1911, n 90.
236 Mendelson, above n 3, at 1300.
237 Muchlinski, above n 22, at 52–54.
238 Leebron, above n 153, at 1622.
both shareholdings below half, and equally such joint ventures could be entered into by more than two parties, as long as none holds more than half of the shares. Leebron suggests that competition law could be left to deal with this practice, though joint ventures are legitimate business arrangements, and it is questionable whether this practice would be disallowed.239
Any of these avoidance methods could be dealt with by anti-avoidance rules, similar to those found in tax law, specifically designed to expand the scope of unlimited liability to those situations.240 However, much like in tax law, it would be difficult for these rules to discriminate between cases where a debt contract, licensing agreement or joint venture is merely being used to, in effect, re-create a limited liability corporate group, and typical cases where those business forms are being used ‘properly’. This is detrimental, because those are legitimate forms of business, which ought not to be universally discouraged, as they do not typically exhibit the same problems that corporate groups do, since there is usually far less control.241 So either we have to accept the cost of new harms created in an attempt to make this reform effective, or we have to accept that the reform will be fairly ineffective. While ineffectiveness also reduces any harmful disincentives, it is still undesirable on the whole as effort is wasted in creating and using these structures.
Finally, note that just because these avoidance methods are possible, does not mean that all companies will take them up. This sort of behaviour could negatively affect a company’s public image. Additionally, there is always some residual risk that a court will disapprove of its behaviour and find some way to hold it responsible. Investors in the parent company might be unwilling to accept such a risk, and demand compliance.
239 At 1622.
240 For a similar example, see again ss 235–238 of the Financial Markets Conduct Act 2013. Of particular interest is s 238, which attempts to outline scenarios which the rule would not apply to, in an attempt to reduce the harmful effects of uncertainty.
241 One might disagree with that statement, and argue that we should also have broader unlimited liability for
various contractual arrangements, but that would be a topic for another paper.
CHAPTER SIX: JURISDICTIONAL VEIL
The previous Chapter ignored perhaps the most obvious method for avoiding this reform; incorporate overseas. Modern companies, and in particular multinational enterprises, are highly mobile, and often tend to base themselves in the most favourable jurisdictions.242 But multinational enterprises need not even go that far, as there are still significant barriers to enforcement, even if they retain a presence in New Zealand, thanks to what Muchlinski refers to as the “jurisdictional veil”.243 There are slightly different considerations depending on whether New Zealand is the “host country” (a subsidiary of a foreign-based multinational enterprise is incorporated here) or the “home country” (the parent company of a multinational enterprise is incorporated here).244 For ease of explanation, we will assume that the tort has been committed in New Zealand by a subsidiary of a foreign multinational enterprise, and any differences for the home country scenario will be pointed out as we go.245
6.1 Bringing a Claim in New Zealand
Our New Zealand plaintiff could either bring a claim in New Zealand, or in the foreign court.246 Problems arise in both cases, but we will start with the New Zealand case first. The initial hurdle would be establishing personal jurisdiction over the parent company. This is probably already permitted under the High Court Rules, r 6.27(2)(a). In the home country scenario, jurisdiction is not contentious, as the parent company will be subject to service as of right.247 However, to be sure, it would be simple enough to include a section in this reform granting the New Zealand courts jurisdiction over any parent company of a New Zealand incorporated subsidiary that is involved in a torts claim. Such a section would also send the courts a strong message not to order a stay of proceedings (an issue we will return to momentarily).
242 Coleman and Friedler, above n 1, at 201.
243 Muchlinski “Corporations in International Litigation: Problems of Jurisdiction and the United Kingdom Asbestos Cases” (2001) 50 ICLQ 1 at 17.
244 Blumberg, above n 34, at 169.
245 Note that more complicated scenarios exist, such as where the tort is committed in a third country, however, this goes beyond the scope of this paper and will not be considered.
246 Note, again, more complicated options are also available such as bringing an action in a third jurisdiction where the multinational enterprise also has a presence. Such options will not be considered by this paper.
247 High Court Rules, r 6.12; Companies Act 1993, s 387.
The second step would be to establish that New Zealand law is applicable to the question of unlimited liability, otherwise our rule would not be applied. Nygh outlines four different choice of law rules that could apply: law of the forum (in this case, New Zealand), law governing the parent company, law governing the subsidiary, and law governing the tort.248 Only the law of the forum rule would guarantee that our law always applied in both the home country and host country scenarios. While, again, it would be simple enough to include such a choice of law section in the reform, it is worth considering the implications of this.
Such an indiscriminate application of New Zealand law would violate principles of prescriptive jurisdiction, governing when it is appropriate for states to apply their laws extraterritorially.249 The first of these is the territoriality principle, which limits a state’s law to conduct occurring within its territory, or having an effect within its territory.250 If we take the conduct here to be the tort, as Blumberg seems to do, then this principle would not be applicable to the home country scenario, but would arguably apply to the host country scenario.251 However, there is equally a claim to be made that the conduct we are concerned with is the ownership of a controlling share of another company, especially given that is the basis for liability under this reform. In theory it is debatable whether ownership is more connected with the owner or the property. However, we can look to choice of law rules to get a good indication that it is the latter, because the choice of law rule for property is the lex situs (law of the place where the property is located). As typically the tort will be committed where the subsidiary is located, these two routes both lead to the same conclusion, that the principled choice of law rule would, one way or another, be the host country’s law.252
Another possibility for justifying the imposition of the law of the forum, in the home country scenario, is the nationality principle; that a state may use its own law to govern its nationals.253 However, this will not help, as it has long been accepted that the nationality of a company is determined by its place of incorporation, not the nationality of its
248 Nygh, above n 4, at 74.
249 Blumberg, above n 34, at 171; Lord Collins of Mapesbury (ed) Dicey, Morris & Collins on the Conflict of Laws
(15th ed, Sweet & Maxwell, London, 2012) at [1-009].
250 Blumberg, above n 34, at 175.
251 At 176.
252 In the host country scenario, it might also be possible to justify a choice of law rule which selects the forum
on the basis of the protection principle, which allows states to apply their law extraterritorially when it is necessary to protect vital state interests; see generally Muchlinski, above n 22, at 127.
253 Blumberg, above n 34, at 171.
shareholder/s.254 So without violating these principles New Zealand could restrict itself to governing the host country scenario. However, it is also important to consider the practical impact of applying New Zealand law extraterritorially. The diplomatic response from other states might actually be worse in the host country scenario, because in that scenario it is their multinational enterprises being exposed to additional liability, which impacts the national economic interests of those states.
For instance, in 1973, the Argentine Supreme Court held all associated companies liable for the unpaid debts of the country’s largest meat-packing company after it collapsed (the Deltec litigation).255 A law was then passed enshrining in statute that foreign companies would be liable for the obligations of their Argentinean subsidiaries.256 This law “evoked severe international hostility” and contributed to a “severe economic downturn”, which probably played a role in the collapse of the Peron government three years later.257 This was not a typical case, as it occurred in a highly politicised context of “strident hostility towards Western capitalism”.258 However, there are other examples illustrating this point in the home country context. The United States regularly imposes its own laws on the foreign subsidiaries of American companies when it has a strategic interest in doing so (it is alone in doing this).259 There have been large diplomatic costs to this policy, evidenced by a number of examples including the ITT-Standard Electric Affair, where the United States tried to force compliance with its trade embargo against China on a British subsidiary, much to the ire of the British government.260 So this is a cost New Zealand would have to bear if it selected any choice of law rule that applied New Zealand law.
However, even if a successful judgment was acquired in New Zealand, the claimant would still have to get enforcement of that judgment against the parent company in the foreign court, except if New Zealand is the home country, where this is obviously not an issue. Enforcement will often be difficult to obtain, because states do not want to disadvantage
254 At 171.
255 Compania Swift de La Plata SA Frigorific s/convocatoria de acreedores (1973) 19 JA 579, 151 La Ley 516 as cited in Blumberg, above n 34, at 187–188. See Michael Gordon “Argentine Jurisprudence: The Parke Davis and
Deltec Cases” (1974) 6 U Miami Inter-Am L Rev 320 at 330–345 for an English translation of this case.
256 Blumberg, above n 34, at 189.
257 At 189.
258 At 188–189.
259 At 174.
260 At 182. See also Muchlinski, above n 22, at 134–135, for a discussion on the application of American antitrust law to the Swiss watchmakers cartel in the 1960s and the subsequent protest of the Swiss
government.
their own companies, so have a strong incentive not to enforce the judgment.261 Firstly, the foreign court could refuse enforcement by finding that New Zealand did not have jurisdiction over the parent company.262 One test for assessing this, used in common law countries, is whether the defendant was present in the jurisdiction.263 In Adams v Cape, the English Court of Appeal made it clear that the presence of a subsidiary was insufficient to demonstrate the presence of the parent company within the jurisdiction, and the Court therefore refused to enforce a Texas judgment.264 Alternatively, the foreign court could refuse enforcement on the grounds that applying unlimited liability is contrary to public policy.265 For instance, when a New York court refused enforcement of the Deltec litigation, the judge suggested that the action might amount to “confiscation of property”.266 Finally, many countries have passed “blocking statutes” in the past to prevent the enforcement of foreign judgments adversely affecting their commercial interests.267 Enforcement would therefore be difficult to obtain, meaning that a claim brought in a New Zealand court would ultimately be ineffective where New Zealand is the host country.
6.2 Bringing a Claim in the Foreign Court
Directly bringing the claim in the foreign court, involves the same problem of incentives. Where New Zealand is the host country, jurisdiction is uncontroversial, as the parent company would be present in the foreign country. However, this time a stay of proceedings would likely be issued, on the grounds of forum non conveniens.268 In Commonwealth jurisdictions, this involves a two-part test: the defendant must show that a particular foreign jurisdiction is the appropriate forum, then a stay of proceedings will be granted unless the plaintiff can show special circumstances where justice requires that the case be tried in the forum.269 In Lubbe v Cape, which was another asbestos claim involving Cape Industries, the House of Lords found that South Africa was the appropriate forum because the evidence
261 Reinschmidt, above n 13, at 105.
262 Simon Baughen “Multinational and the Export of Hazard” (1995) 58 MLR 54 at 55.
263 Lord Collins, above n 249, at [14-055].
264 Muchlinski, above n 5, at 921.
265 At 926.
266 Blumberg, above n 34, at 192.
267 Muchlinski, above n 22, at 161.
268 Reinschmidt, above n 13, at 105.
269 Spiliada Maritime Corporation v Cansulex Ltd [1986] UKHL 10; [1987] AC 460 (HL) at 476. Note, however, that different jurisdictions take slightly different approaches to this test. Australia, for instance, is regarded as very pro-
plaintiff; see generally Muchlinski, above n 22, at 159.
was located there.270 A stay of proceedings was actually ordered in that case, because the claimants were able to demonstrate, on the second limb, that substantial justice would not be done in South Africa, but this is hardly likely to be a problem with New Zealand, given that our law will be more favourable if anything.271 Likewise, where torts claimants have attempted to directly sue a parent company in the United States, its courts have refused to hear the cases on similar grounds.272 Where New Zealand is the home country, jurisdiction would depend on the particular long-arm jurisdiction rules of the foreign state.273 Though, if jurisdiction is established, it is unlikely that a stay of proceedings would be issued, as the foreign state would probably be the appropriate forum.
However, even if a stay of proceedings is not ordered, the court would still have to select a choice of law rule that applies New Zealand law, which it is unlikely to do. Hansmann and Kraakman assume this would not be a problem, because the issue would be governed by the lex loci delicti (law of the place where the tort is committed).274 However, this assumes that the issue of unlimited liability would be characterised as a torts issue, which is not necessarily the case, as it could equally be an issue of company law. There is at least a risk the foreign court would select a different rule. Even if this did not occur, the foreign court could still apply the public policy exception, which allows it to refuse to apply New Zealand law if it is contrary to important public interests of that state.275 Given the aforementioned incentives acting on the foreign state, the risk of one of these options being pursued is high. Finally, note that bringing a claim in a foreign court is financially out of reach for many torts victims anyway.276 On the whole, whether it is the home state or the host state, bringing a claim in the foreign court is likely to be even less viable than bringing a claim in New Zealand.
270 Lubbe v Cape Plc [2000] UKHL 41; [2000] 1 WLR 1545 (CA) at 1556.
271 At 1558–1559.
272 See In Re Union Carbide Corporation Gas Plant Disaster At Bhopal, India In December, [1987] USCA2 35; 1984 809 F 2d 195 (2d Cir 1987) at 200-202 where the Court dismissed a claim on forum non conveniens grounds. See also Kiobel v Royal Dutch Petroleum Co 133 S Ct 1659 (2013) at 1677-1678 where the Court refused to extend application of the Alien Tort Claims Act to actions of a Nigerian subsidiary in Nigeria. Contrast Wiwa v Royal Dutch Petroleum Company and Shell Transport and Trading Company Plc [2000] USCA2 359; 226 F 3d 88 (2nd Cir 2000) at 103–106 where the Court held that, in special circumstances where torture is involved, that would create a significant reason not to dismiss the claim.
273 These vary between jurisdictions. In the United States, for example, service on an American subsidiary is
accepted as valid service on a foreign parent company, while the same is not accepted in England; see generally Muchlinski, above n 22, at 141–142.
274 Hansmann and Kraakman, above n 9, at 1922.
275 Muchlinski, above n 5, at 926.
276 Reinschmidt, above n 13, at 105.
Muchlinski notes that these problems will persist unless some international consensus is reached to adopt this reform, which seems unlikely in the current legal climate.277 Blumberg agrees, suggesting that “profound changes of legal principles by less than a critical mass” will not be sustainable, referring to the United State’s short-lived unique regime of maritime liability in the mid 19th century.278 If New Zealand did attempt to go it alone, the only way to prevent this reform from being confined in its effectiveness to domestic corporate groups, would be to negotiate special enforcement treaties with other nations. Whether or not this would be possible, or desirable, is a diplomatic question that could be considered by lawmakers.
277 Muchlinski, above n 5, at 926.
278 Blumberg, above n 34, at 199–200.
CONCLUSION
This paper identified a problem in the current law; that the principle of limited liability leaves many torts claimants without a remedy and therefore creates distorted incentives for corporate groups. Chapter One found that direct tortious liability and the doctrine of piercing the veil are inadequate to deal with this problem, because they only apply on a case-by-case basis, and are difficult to invoke. Rather than relying on those traditional limitations to the principle of limited liability, this paper sought to challenge whether the principle is even justified in the first place when dealing with torts claims against parent companies.
After looking into the history and economics behind limited liability, in Chapter Two, it became evident that the only justifiable basis for retaining limited liability in this context stems from the social benefit it provides by encouraging beneficial risk taking. In Chapter Three, this benefit was weighed against the countervailing benefit from reducing overly hazardous risk taking. It became apparent that the predominant outcome of allowing limited liability is a reduced incentive to purchase liability insurance. Since insurance provides the best balance of incentives, it became clear that limited liability produces more social harm than good in this context. One important observation is that the purchase of insurance is about more than just guaranteeing compensation for torts victims, as ensuring companies have the right incentives actually helps prevent torts occurring in the first place, as precautionary measures are taken.
Identifying that it would be justifiable to remove limited liability in this context, led to the consideration of the alternative in Chapter Four. The best alternative system seems to be a regime of unlimited liability that holds any company responsible for the full amount of any tort liability incurred by other companies that it owns more than a 50 per cent shareholding in. Chapter Five found that unlimited liability could be avoided in a number of ways, including by use of debt, contract, and joint ventures. In order to obtain the full benefits of unlimited liability it might be necessary to implement anti-avoidance rules. Further, Chapter Six identified how the jurisdictional veil would essentially create another avoidance method in the case of multinational enterprises.
So even though it would be theoretically justifiable to introduce unlimited liability in the manner outlined, there are serious questions about whether doing so would be worth the effort. As Chapter Two explained, limited liability is ultimately a pragmatic doctrine designed to produce social good. This means that if lawmakers were considering removing limited liability as specified, they should weigh up the benefits of correcting the malincentives of limited liability against the harms that must be accepted in adopting this regime of unlimited liability.
Those harms broadly fit into a number of categories. Firstly, some social benefit would be lost from decreased risk taking. This is generally true of particular industries, such as pharmaceuticals, but is also true in other industries if the parent company is unable to (or chooses not to) fully buy out its partially-owned subsidiaries, and/or if an insurance package that suits the needs of the company is not available at an efficient price. Secondly, some social benefit would be lost by making it inefficient for companies to opt into what would otherwise be more efficient business structures. This could occur if parent companies chose to outsource instead of vertically integrating, or arguably if conglomerates demerged into separately owned companies. Thirdly, if the reform turns out to be ineffective, either resources would be wasted in avoidance attempts, or anti-avoidance rules could cause further social loss by deterring legitimate business structures. Fourthly, there would be some diplomatic costs if this policy were to apply to multinational enterprises. These include both diplomatic backlash from imposing our laws on foreign companies, as well as any concessions required to negotiate enforcement treaties with other states.
With so many potential harms, and doubtful enforceability, it is questionable whether reintroducing unlimited liability for parent companies in cases of tort would provide greater social benefit than the status quo, as undesirable as the status quo is. The international context makes this reform essentially unworkable in full, though it might be worth considering a limited version of this reform, which only applies to domestic corporate groups. While three categories of harm would still remain, those are all marginal and contestable harms. While I would personally incline toward the conclusion that those costs still outweigh the benefits, it is not my place to say, as the final call is ultimately a political trade-off, which can only be made by lawmakers.
SCHEDULE
5 Meaning of holding company and subsidiary
(1) For the purposes of this Act, a company is a subsidiary of another company if, but only if,—
(a) that other company—(i) controls the composition of the board of the company; or(ii) is in a position to exercise, or control the exercise of, more than one- half the maximum number of votes that can be exercised at a meeting of the company; or
(iii) holds more than one-half of the issued shares of the company, other than shares that carry no right to participate beyond a specified amount in a distribution of either profits or capital; or
(iv) is entitled to receive more than one-half of every dividend paid on shares issued by the company, other than shares that carry no right to participate beyond a specified amount in a distribution of either profits or capital; or
(b) the company is a subsidiary of a company that is that other company’s subsidiary.
(2) For the purposes of this Act, a company is another company’s holding company, if, but only if, that other company is its subsidiary.
(3) In this section and sections 7 and 8, the expression company includes a body corporate.
For the purposes of section 5, without limiting the circumstances in which the composition of a company’s board is to be taken to be controlled by anothercompany, the composition of the board is to be taken to be so controlled if the other company, by exercising a power exercisable (whether with or without the consent or concurrence of any other person) by it, can appoint or remove all the directors of the company, or such number of directors as together hold a majority of the voting rights at meetings of the board of the company, and for this purpose, the other company is to be taken as having power to make such an appointment if—
(a) a person cannot be appointed as a director of the company without the exercise by the other company of such a power in the person’s favour; or
(b) a person’s appointment as a director of the company follows necessarily from the person being a director or other officer of the other company.
In determining whether a company is a subsidiary of another company,—(a) shares held or a power exercisable by that other company in a fiduciary capacity are not to be treated as held or exercisable by it:
(b) subject to paragraphs (c) and (d), shares held or a power exercisable—
(i) by a person as a nominee for that other company, except where that other company is concerned only in a fiduciary capacity; or
(ii) by, or by a nominee for, a subsidiary of that other company, not being a subsidiary which is concerned only in a fiduciary capacity,—
are to be treated as held or exercisable by that other company:
(c) shares held or a power exercisable by a person under the provisions of debentures of the company or of a trust deed for securing an issue of debentures shall be disregarded:
(d) shares held or a power exercisable by, or by a nominee for, that other company or its subsidiary (not being held or exercisable in the manner described in paragraph (c)) are not to be treated as held or exercisable by that other company if—
(i) the ordinary business of that other company or its subsidiary, as the case may be, includes the lending of money; and
(ii) the shares are held or the power is exercisable by way of security only for the purposes of a transaction entered into in the ordinary course of that business.
BIBLIOGRAPHY
Legislation
Companies Act 1993.
Financial Markets Conduct Act 2013. High Court Rules.
Cases: New Zealand
DHL International (NZ) Ltd v Richmond Ltd [1993] 33 NZLR 10 (CA). Lewis Holdings Ltd v Steel and Tube Holdings Ltd [2014] NZHC 3311. RM Turton & Co Ltd v Kerslake and Partners [2000] NZCA 115; [2000] 3 NZLR 406 (CA).
Rolls-Royce New Zealand Ltd v Carter Holt Harvey [2004] NZCA 97; [2005] 1 NZLR 324 (CA).
Cases: Argentina
Compania Swift de La Plata SA Frigorific s/convocatoria de acreedores (1973) 19 JA 579, 151 La Ley 516.
Cases: Australia
Dagi v The Broken Hill Pty Ltd (No 2) [1997] 1 VR 428 (SC).
James Hardie & Co Pty Ltd v Hall (1998) 43 NSWLR 554 (CA).
Cases: India
Mehta v Union of India [1986] INSC 282; (1987) 1 SCR 819.
Novartis Ag v Adarsh Pharmaceuticals (2004) 3 CTC 95 (MHC).
State of Uttar Pradesh v Renusagar (1988) 1 SCR 627.
Cases: Netherlands
Akpan v Royal Dutch Shell Plc DC The Hague LJN BY9854/HA ZA 09-1580, 30 January 2013.
Cases: United Kingdom
Adams v Cape Industries [1990] 1 Ch 433 (CA).
Cambridge Water Co Ltd v Eastern County Leather Plc [1993] UKHL 12; [1994] 2 AC 264 (HL).
Caparo Industries Plc v Dickman [1990] UKHL 2; [1990] 2 AC 605 (CA).
Chandler v Cape Plc [2012] EWCA Civ 525, [2012] 1 WLR 3111.
DHN Food Distributors Ltd v Tower Hamlets London Borough Council [1976] 1 WLR 852 (CA).
Henderson v Merrett Syndicates Ltd (No 1) [1994] UKHL 5; [1995] 2 AC 145 (HL).
Home Office v Dorset Yacht Co Ltd [1970] UKHL 2; [1970] AC 1004 (HL).
Lubbe v Cape Plc [2000] UKHL 41; [2000] 1 WLR 1545 (CA).
Prest v Petrodel Resources Ltd [2013] UKSC 34, [2013] 2 AC 415. Spiliada Maritime Corporation v Cansulex Ltd [1986] UKHL 10; [1987] AC 460 (HL). Stovin v Wise [1996] UKHL 15; [1996] AC 923 (HL).
Woolfson v Strathclyde Regional Council 1978 SLT 159 (HL).
Cases: United States
Bowoto v Chevron Texaco Corp 312 F Supp 2d 1229 (ND Cal 2004).
In Re Union Carbide Corporation Gas Plant Disaster At Bhopal, India In December, 1984 [1987] USCA2 35; 809 F 2d 195 (2d Cir 1987).
Kiobel v Royal Dutch Petroleum Co 133 S Ct 1659 (2013).
Pullman's Palace Car Company v Missouri Pacific Railway [1885] USSC 243; 115 US 587 (1885).
The “Amaco Cadiz” [1984] 2 Lloyd’s Rep 304 (USDC).
United States v Bestfoods [1998] USSC 65; 524 US 51 (1998).
Wiwa v Royal Dutch Petroleum Company and Shell Transport and Trading Company Plc [2000] USCA2 359; 226 F 3d 88 (2nd Cir 2000).
Looseleaf Texts
Andrew Beck and others Morison’s Company Law (NZ) (online looseleaf ed, LexisNexis).
Books
Kenneth Arrow Essays in the Theory of Risk-bearing (North-Holland Publishing Company, Amsterdam, 1971).
Phillip Blumberg The Multinational Challenge to Corporation Law: The Search for a New Corporate Personality (Oxford University Press, New York, 1993).
Lord Collins of Mapesbury (ed) Dicey, Morris & Collins on the Conflict of Laws (15th ed, Sweet & Maxwell, London, 2012).
Peter Muchlinski Multinational Enterprises & the Law (2nd ed, Oxford University Press, New York, 2007).
Richard Posner Economic Analysis of Law (4th ed, Little, Brown and Company, Boston, 1992).
Journal Articles
Barak Atiram “Market Dynamics in Corporate Tort Externalization: The Hidden Assumption of Corporate Social Efficiency” (2009) 86 U Det Mercy L Rev 347.
Simon Baughen “Multinational and the Export of Hazard” (1995) 58 MLR 54. Phillip Blumberg “Limited Liability and Corporate Groups” (1986) 11 J Corp L 573.
Brandon Cline, Jacqueline Garner and Adam Yore “Exploitation of the Internal Capital Market and Avoidance of Outside Monitoring” (2014) 25 Journal of Corporate Finance 234.
Sarah Coleman and Jonathan Friedler “The Road to Reform in the Wake of Kiobel: Multinational Corporations and Socially Responsible Behaviour” (2014) 13 J Int’l Bus & L 191.
Frank Easterbrook and Daniel Fischel “Limited Liability and the Corporation” (1985) 52 U Chi L Rev 89.
Michael Gordon “Argentine Jurisprudence: The Parke Davis and Deltec Cases” (1974) 6 U Miami Inter-Am L Rev 320.
Joseph Grundfest “The Limited Future of Unlimited Liability: A Capital Markets Perspective” (1992) 102 Yale LJ 387.
Paul Halpern, Michael Trebilcock and Stuart Turnbull “An Economic Analysis of Limited Liability in Corporation Law” (1980) 30 UTLJ 117.
Henry Hansmann and Reinier Kraakman “Towards Unlimited Shareholder Liability for Corporate Torts” (1991) 100 Yale LJ 1879.
Paddy Ireland “Limited Liability, Shareholder Rights and the Problem of Corporate Irresponsibility” (2010) 34 Camb J Econ 837.
David Leebron “Limited Liability, Torts Victims, and Creditors” (1991) 91 Colum L Rev 1565.
Lee McConnell “Establishing Liability for Multinational Oil Companies in Parent/Subsidiary Relationships” (2014) 16 Env L Rev 50.
Robert McCorquodale “Waving Not Drowning: Kiobel Outside the United States” (2013) 107 AJIL 846.
Nina Mendelson “A Control-based Approach to Shareholder Liability for Corporate Torts” (2002) 102 Colum L Rev 1203.
Peter Muchlinski “Corporations in International Litigation: Problems of Jurisdiction and the United Kingdom Asbestos Cases” (2001) 50 ICLQ 1.
Peter Muchlinski “Holding Multinationals to Account: Recent Developments in English Litigation and the Company Law Review” (2002) 23 Co Law 168.
Peter Muchlinski “Limited Liability and Multinational Enterprises: A Case for Reform?” (2010) 34 Camb J Econ 915.
Peter Muchlinski “The Bhopal Case: Controlling Ultrahazardous Industrial Activities Undertaken by Foreign Investors” (1987) 50 MLV 545.
Peter Nygh “The Liability of Multi-national Corporations for the Torts of their Subsidiaries” (2002) 3 EBOR 51.
Oserheimen Osunbor “The Agent-only Subsidiary Company and the Control of Multinational Groups” (1989) 38 ICLQ 377.
Stephen Presser “Thwarting the Killing of the Corporation: Limited Liability, Democracy, and Economics” (1992) 87 Nw U L Rev 148.
Abhi Raghunathan “The Grand Trunk Road from Saloman to Mehta: Economic Development and Enterprise Liability in India” (2012) 100 Geo LJ 571.
Bastian Reinschmidt “The Law of Tort: A Useful Tool to Further Corporate Social Responsibility?” (2013) 34 Co Law 103.
James Tocher “Contribution Orders – A Case Note on Lewis Holdings v Steel and Tube Holdings” [2015] NZLawStuJl 10; (2015) 3 NZLSJ 601.
Lutz-Christian Wolff “Offshore Holdings for Global Investments of Multinational Enterprises: Just Evil?” (2015) 6 JBL 445.
Newspapers
Allanna Sullivan “Oil-firms, Shippers Seek to Circumvent Laws Setting No Liability Limit for Spills” The Wall Street Journal (United States, 26 July 1990) at B1.
NZLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.nzlii.org/nz/journals/UOtaLawTD/2016/32.html