1 Introduction

Corporate liability standards for punishing human rights abuse and environmental damage occurring in the context of corporate group activities are crucial to the success of transnational claims against MNEs. However, legal barriers can arise where the way in which liability is attributed among members of a corporate group under domestic laws facilitates the avoidance of appropriate accountability.¹ As a result, such barriers may prevent legitimate cases from being addressed, thus leading to corporate impunity. In France and the Netherlands, a number of plaintiffs have sought to hold parent companies liable for their direct or indirect involvement in activities harmful to humans and the environment in host countries. In most instances, cause-lawyers and CSOs litigating these cases have attempted to demonstrate the absence of an effective regime of liability applying to MNEs. They have also demonstrated the inequality arising from the benefits relating to the corporate form, such as ‘the limited liability for its members and a legal personality separate from that of its members’,² when business-related abuse occurs in the context of corporate groups.

It is important to make a few clarifications before starting the analysis. First, this chapter mainly explores how liability can be attributed to the parent company for harm resulting from its subsidiaries’ activities. It does not specifically investigate the liability of a company which may arise as a result of damage occurring in the context of its subcontractor’s activities or in joint ventures. Second, in the context of transnational civil litigation against MNEs, the law of the host State generally applies to the facts of the claim as a result of the Rome II Regulation. Consequently, it is less likely that French or Dutch corporate liability standards will apply to these claims. Nonetheless, the Rome II Regulation creates a number of exceptions which allow plaintiffs to choose the law of the home State (ie French or Dutch law) as the applicable law. As a result, the study of corporate liability standards in France and the Netherlands remains relevant. Third, France and the Netherlands have recently enacted legislation imposing mandatory HRDD obligations on companies that may be directly relevant to corporate liability standards. A detailed analysis of these new legal regimes is provided in Chapter 7 of this book.

2 Challenges to establishing the liability of MNEs

In order to understand the issues surrounding MNE’s liability for human rights violations and environmental pollution, one must understand the legal complexity of corporate groups under French and Dutch company law. The absence of a legal definition of the ‘corporate group’ as well as the existence of the separate legal personality of a company and the nature of limited liability companies pose problems in holding MNEs liable.

MNEs as corporate groups

Before delving into the subject of corporate group liability, it is important to first define what the corporate group is. Vandekerckhove broadly defines the corporate group as ‘an aggregate of legally independent corporations that are related to each other through patrimonial, contractual or personal links and that come under a common centre of control’.³ She explains that:

There is no statutory definition of the ‘corporate group’ in France,⁵ and the responsibility to define or deal with the corporate group has been left to the French courts. In general, French courts recognize the ‘interest of the group.’ However, different definitions have been adopted by the courts in different areas of law. For example, the Commercial Chamber of the Court of Cassation has held that a corporate group is characterized by economic unity and a community of indivisible interests led by one person.⁶ On the other hand, for the Criminal Chamber of the Court of Cassation, a corporate group exists when its companies share a common economic, social, or financial interest (intérêt économique, social, ou financier commun), which must be appreciated with regard to the policies developed for the whole group.⁷ While these definitions reflect the priorities of the various legal branches, they show a lack of consistency.

Although there is no statutory definition of the corporate group, the French Commercial Code describes how companies may be linked to each other through financial ownership or control.⁸ Generally, where a company owns more than 50 per cent of the capital of another company, the second company is to be regarded as a subsidiary (filiale) of the first company (Article L233-1). Furthermore, where a company owns between 10 and 50 per cent of the capital of another company, the first company shall be regarded as having a participation (ayant une participation) in the second company (Article L233-2).

In the context of listed companies and for specific purposes,⁹ any person, whether natural or legal (such as a company), is deemed to control another one in various situations, namely where: (1) it directly or indirectly holds a fraction of the capital that gives it a majority of the voting rights at that company’s general meetings; (2) it holds a majority of the voting rights in that company by virtue of an agreement; (3) it effectively determines the decisions taken at that company’s general meetings through its voting rights; or (4) it is a partner in or shareholder of that company and has the power to appoint or dismiss the majority of the members of that company’s administrative, management, or supervisory structures (Article L233-3(I)). Furthermore, a person is presumed to exert control when it directly or indirectly holds a fraction of the voting rights above 40 per cent and no other partner, member, or shareholder directly or indirectly holds a fraction larger than its own (Article L233-3(II)). Finally, two or more persons acting in concert shall be regarded as jointly controlling another when, in fact, they determine the decisions taken at the general assembly (Article L233-3(III)). Ultimately, Article L233-3 deals with different types of control: de jure or de facto control, exclusive or with other partners.¹⁰

There is also a definition of control applicable to accounting matters. Article L233-16(I) French Commercial Code provides that companies must, on an annual basis, draw up and publish consolidated accounts and a group management report for any company which they control, either solely or jointly. Article L233-16(II) provides that sole control of a company exists where: (1) a majority of its voting rights are held by another company; (2) a majority of the members of its board of directors, management board, or supervisory board are appointed by another company for two consecutive financial years;¹¹ and (3) a dominant interest is exerted over the company by virtue of a contract or the terms and conditions of its memorandum and articles of association, when the applicable law allows this. Furthermore, according to Article L233-16(III), joint control exists where the control of a company operated jointly by a limited number of partners or shareholders is shared and decisions are made on the basis of an agreement between them.

In contrast to French law, Dutch company law provides a general definition of the ‘group’ in Article 2:24b Dutch Civil Code. According to this provision, a ‘group is an economic unit in which legal persons and commercial partnerships are organizationally interconnected. Group companies are legal persons and commercial partnerships interconnected in one group.’¹² This definition applies to all types of legal persons, and foreign company types may also form part of a group.¹³ According to Dutch scholars, an important feature of a group, which is not included in this definition, is central management.¹⁴ This concept is difficult to define, as it depends on factual circumstances. However, central management will generally be present ‘when there is a joint strategy that forms the basis for the dissemination of plans and coordination of a strategy at lower levels within the group’.¹⁵ The definition found under Article 2:24b is referred to as the economic definition of the group concept in Dutch law, as it refers to actual control being exercised and economic unity within the group, rather than the power to control merely being available.

In contrast, the definition of ‘subsidiary’ refers to the power to control and therefore provides a more legal definition.¹⁶ Article 2:24a(1) Dutch Civil Code defines a subsidiary (dochtermaatschappij) as a legal person:

Based on this definition, most legal persons can qualify as a subsidiary, and foreign companies can also qualify as subsidiaries.¹⁷ It is generally accepted that, in most cases, there will be a group relationship where there is a subsidiary.¹⁸ Importantly, a group relationship does not require a capital investment. It may be established by means of a contractual relationship (eg on the basis of a contract establishing personal unions between two legal persons).¹⁹

In addition, Dutch company law provides a definition of ‘participation’ (or participating interest). Pursuant to Article 2:24c(1) Dutch Civil Code, a legal person has a participation in another legal person where it, either directly or through one or more subsidiaries for its own account, solely or jointly, has provided or has caused the provision of the capital of the second legal person with the aim of having a long-term association with the second company for the benefit of its own activities. Participation is presumed where the legal person holds one-fifth or more of the issued capital.

Finally, Dutch company law provides the existence of dependent companies, which are legal persons to which a public or private limited liability company or one of its dependent companies has provided, for its own account, either solely or jointly, at least one-half of the issued share capital.²⁰ Here, participation through share capital is required, and providing half of the issued capital is sufficient.²¹

This overview of the definition(s) of the ‘group’ in France and the Netherlands shows that French and Dutch company legislation generally lacks a definition of the group that takes into account its legal reality. That being said, French and Dutch law addresses the group through various capital or control relationships that may develop between companies. For example, they recognize the existence of subsidiaries or controlled or dependent companies. They do not, however, define the parent company. More importantly, the group is not recognized as a separate legal entity in both countries. In France, the Court of Cassation recognized that the group has no legal personality,²² which means that the group is not recognized as a unified business legal entity.²³ In the Netherlands, the single entity is the starting point of Dutch company law, and the group is not recognized as a separate legal entity.²⁴ One consequence of this lack of legal personality is that groups cannot have rights and obligations or be bound to pay damages. Therefore, the liability resulting from any obligations belonging to entities of the group shall be borne by those entities.

Separate legal personality and limited liability within MNEs

The principles of separate legal personality of the company and limited liability of shareholders are two important aspects of modern company law.²⁵ They are applied at national level by a large number of jurisdictions through either domestic legislation and/or case law.²⁶ In the context of this book, ‘separate legal personality’ means that some types of companies become autonomous legal entities once they are incorporated. As such, they exercise rights and assume certain obligations. The law ignores the artificial nature of these companies by giving them a legal personality which is separate from that of the persons who manage it (directors) or own it (shareholders when the company is limited by shares).²⁷ Furthermore, there is ‘limited liability’ when, for certain types of companies, the liability of investors, owners, or shareholders is limited to the amount of their investment, contribution, or shares in the company.²⁸ Both principles are of crucial importance for companies, especially MNEs. The principle of separate legal personality allows the company to act on its own while insulating the persons participating in the business (whether natural or legal persons) from personal liability. Furthermore, the principle of limited liability is essential to the proper operation of corporations in the market. Muchlinski states:

However, both separate legal personality and limited liability may pose problems in the context of corporate groups – including MNEs – especially where one company owns and controls another.³⁰ They generally prevent one MNE member from being held liable for the activity of another member of the same MNE, even when the former member owns and controls the latter one. For instance, the parent company may be the shareholder, or one of the shareholders, of a subsidiary and, at the same time, control or be engaged in the business activities of that subsidiary. In such a situation, the application of separate legal personality and limited liability often shield the parent company from liability for human rights abuse and environmental damage committed through its subsidiary.³¹ A parent company will not be liable for the harm caused by its subsidiary, even when the parent company owns and controls the subsidiary.³² MNEs may use complex and confusing corporate structures to distance and separate the parent company from the local operating subsidiaries, thereby protecting the MNE from legal liability.³³

Scholars have criticized the strict application of separate legal personality and limited liability where human rights abuse or environmental damage are involved.³⁴ They have also suggested that the traditional image of the company as ‘an isolated and free-standing commercial entity with a sole aim of making profit, often at any cost’ should be revised. Policy-makers should recognize that companies are integrated parts of society and that their economic persona should not be separated from their social and political persona.³⁵ Problems with the application of separate legal personality and limited liability have been particularly visible in the context of transnational claims against MNEs, where plaintiffs have repeatedly challenged the relationship between the parent company based in a home country and foreign subsidiaries under its control or ownership operating in host countries.³⁶

The technique of piercing the corporate veil may provide a solution for limiting unfair consequences for victims of MNE abuse. The expression ‘piercing the corporate veil’, or corporate veil piercing, emerged from the lexicon of company law. It refers to the situation where a corporate shareholder is held liable for the debts of the company of which it is a shareholder notwithstanding separate legal personality and limited liability.³⁷ Some commentators have argued that corporate veil piercing should be extended to cases raising human rights abuse or environmental damage by MNEs through amendments to national company laws.³⁸ However, piercing the corporate veil is a problematic solution for a number of reasons. First, it involves judicial discretion,³⁹ and scholars have argued that criteria for corporate veil piercing are not very clear-cut. It may also be very difficult to establish the factual relation required to pierce the corporate veil.⁴⁰ Furthermore, complex corporate structures, coupled with the use of separate legal personality and limited liability, have an influence on the legal strategies used by plaintiffs to hold MNEs to account. For instance, limited liability forces plaintiffs to focus on acts or omissions of parent companies rather than seeking to pierce the corporate veil. However, an emphasis on parent companies limits the potential for holding accountable corporate groups that operate under a vertically hierarchical management structure. In more complex management structures, it is even harder to match existing legal principles of negligence to the reality of control.⁴¹ In addition, there is a lot of confusion as to the exact meaning of corporate veil piercing. For instance, courts often do not distinguish between statutory rules and corporate veil piercing theories when they hold parent companies liable. As a result, parent companies are sometimes held liable based on corporate veil piercing theories where the case could have been solved by reference to existing rules of company or civil law.⁴² Finally, corporate veil piercing theories are less developed outside common law countries.

Separate legal personality and limited liability in France and the Netherlands

Both France and the Netherlands apply these principles through statutory law. First of all, pursuant to French and Dutch law, a number of business forms, including corporate entities, have their own legal personality. In France, Article 1842 French Civil Code states that partnerships (sociétés)⁴³ enjoy legal personality from the time of their registration.⁴⁴ Furthermore, Article L210-6 French Commercial Code (Code de commerce) provides that trading companies (sociétés commerciales), which include limited liability companies (sociétés à responsabilité limitée) and joint-stock companies (sociétés anonymes),⁴⁵ shall have legal personality with effect from their registration in the commercial and companies register. Therefore, once a trading company is registered, it has a legal personality separate from that of its shareholders, directors, or officers. In the Netherlands, Article 2:3 Dutch Civil Code provides that a number of companies possess legal personality, including public limited companies (naamloze vennootschappen or NV) and private limited companies (besloten vennootschappen or BV). Both types of companies are legal persons with an authorized capital divided into transferable shares.⁴⁶

The application of separate legal personality to MNEs means that a (parent) company that owns or controls another company belonging to the same group, whether a subsidiary or a controlled or dependent company, or through a participation in that company, cannot be held liable for the obligations of that company. According to French case law, companies operating as part of a corporate group remain separate legal persons.⁴⁷ Courts have held that a subsidiary must be regarded as an autonomous company, solely responsible for the consequences of its activities. Despite the close links that may exist between a parent company and its subsidiary, the latter is legally distinct from the natural and legal persons of which it is composed, irrespective of the size of the participation of the parent company in the capital of its subsidiary or the existence of common directors.⁴⁸ Similarly, under Dutch law, each company within the group has separate legal personality and is therefore responsible for its own obligations.⁴⁹

Moreover, in France and the Netherlands, for certain types of companies, the liability of investors, owners, or shareholders is limited to the amount of their investment, contribution, or shares in the company. In France, Article L223-1 French Commercial Code states that a limited liability company may be established by one or more persons who shall bear their losses only up to their contributions. Furthermore, Article L225-1 French Commercial Code provides that a joint-stock company is a company whose capital is divided into shares and which is formed among members who shall bear any losses only up to their contributions. In the Netherlands, in both public and private limited companies, shareholders shall not be personally liable for acts performed on behalf of the company and shall not be liable to contribute to company losses exceeding the amount to be paid on their shares.⁵⁰

One potential issue with limited liability entities is that they may be used to avoid liabilities. For example, an operator of a hazardous activity may carry out its activity through a limited liability entity in order to avoid having to bear the full cost of environmental damage. This problem may be exacerbated in the context of corporate groups. The use of limited liability entities in corporate group structures, where a (parent) company may be a shareholder of another company of the group, may create difficulties for voluntary or involuntary creditors seeking to recover losses or obtain compensation.⁵¹ If the company does not have sufficient assets to cover its liability, the assets of its shareholders, whether natural or legal persons, may be affected only to the extent of the value of their shares in the company.⁵² This situation may create an incentive for corporate groups to externalize risks and, therefore, avoid liability by organizing themselves in such a way that the burden of hazardous or unsafe activities is borne by companies that might turn out to be insolvent.

Nonetheless, there may be specific circumstances where courts may disregard separate legal personality and limited liability in order to hold the parent company liable for the obligations of the companies it owns or controls.

3 Bases for liability of the parent company

There are several situations in which a parent company may be held liable for the obligations of its subsidiary in the context of MNEs’ activities. This section explores the legal grounds for holding parent companies liable in different areas of law and, where relevant, describes how these grounds have applied in transnational cases against MNEs for human rights violations and environmental pollution.

Corporate veil piercing

In certain situations, courts may set aside separate legal personality and limited liability to hold the shareholder, which may be the parent company, liable for the actions or debts of the company of which it is a shareholder. This situation is called ‘corporate veil piercing.’⁵³ While corporate veil piercing may be accepted in exceptional cases under company law, it is generally applied where a subsidiary is wholly-owned by its parent company under competition law.

Company law

In French and Dutch company law, courts may exceptionally pierce the corporate veil to hold the parent company liable for the obligations of its subsidiary, most notably to protect the creditors of the subsidiary in the context of insolvency or bankruptcy proceedings. However, courts are usually reluctant to do so.

France

In France, corporate veil piercing may, in certain circumstances, be possible on the basis of statutory law and theories developed by the courts.

One statutory provision for piercing the corporate veil is Article L621-2 French Commercial Code on safeguarding proceedings (procédures de sauvegarde) under insolvency law.⁵⁴ These proceedings may be extended to one or more other persons where their assets are intermingled with those of the debtor, or where the legal entity is a sham. Article L621-2 provides for the application of two types of corporate veil piercing theories. The first one is the theory of ‘commingling of assets’ (confusion de patrimoine), which applies when it is no longer possible to distinguish between the assets of the parent company and those of its subsidiary.⁵⁵ The second one is the theory of the ‘fictitious legal person’ (fictivité de la personne morale), which provides that a legal person is deemed to be fictitious where its sole purpose is to serve the interests of the natural or legal person behind it, and that person is engaging in high-risk activities under the cover of separate legal personality and limited liability.⁵⁶ Both theories involve the notion of fraud.⁵⁷

French courts have shown a strong reluctance to pierce the corporate veil on the basis of commingling of assets in cases of relationships within a group. The Court of Cassation usually requires the existence of ‘abnormal financial relationships’.⁵⁸ For example, it refused to pierce the corporate veil in Theetten v SA Metaleurop on the grounds that cash-pooling, staff exchanges, and fund advances by the parent company did not automatically reveal abnormal financial relationships that constituted a commingling between the assets and liabilities of the parent company and those of its subsidiary.⁵⁹ In a corporate group, these acts may be justified. This case law is likely to make it even more difficult to prove a commingling of assets in a transnational case.⁶⁰

French courts may also ignore the application of separate legal personality and limited liability in the context of a corporate group on the basis of the theory of ‘interference’ (immixtion). A parent company can be held liable in respect of its subsidiary’s creditors when it has interfered in the activities and the management of its subsidiary.⁶¹ If a parent company makes decisions for its subsidiary, the latter cannot be considered an autonomous legal entity. Generally, French courts use ‘indicators’ (faisceau d’indices) to determine, on a case-by-case basis, whether there is interference. In one case, the Court of Cassation found that similarities between two companies (such as telephone numbers, email addresses, head offices, or managers), important cash flows between them, and the intervention of the parent company’s technicians in the context of a contract between the subsidiary and a third party pointed to interference.⁶² However, French courts are reluctant to recognize such interference.⁶³ Corporate groups often share a common strategy, which makes it difficult to assess the degree of the parent company’s interference in the management of its subsidiary.⁶⁴

In addition, French courts may use the theory of ‘appearance’ (apparence) to hold the parent company liable for its subsidiary’s acts. This theory enables the contractual commitment made by one group company to be binding on another in order to protect the co-contracting third party who acted in good faith. In the event of insolvency, a parent company may be liable to the creditors of its subsidiary if the creditor has a good faith reason to believe that the two companies are the same entity. Two conditions must be met to apply the theory of appearance: a sufficiently strong deceptive appearance and the good faith of the contracting third party. French courts also use indicators to determine, on a case-by-case basis, whether a parent company may have led a third party to believe that it formed a single entity with a subsidiary or that it wanted to enter into a commitment alongside a subsidiary (eg similar head offices or managers).⁶⁵

Practice shows that French courts are often inconsistent in applying these theories or mix them.⁶⁶ For example, the Court of Cassation seems to require the criteria used for interference to pierce the corporate veil on the basis of appearance.⁶⁷ This approach is, however, more demanding in terms of proof, while being less protective of third parties.⁶⁸ Furthermore, similar facts in separate cases may sometimes lead to different outcomes, as a result of the arbitrary application of these theories.⁶⁹ Scholars have criticized the French courts for being more concerned with the result of corporate veil piercing than with its legal underpinning. However, French courts are generally reluctant to hold parent companies liable for the activities of their subsidiaries, and these theories continue to be used only in exceptional circumstances.⁷⁰

The Netherlands

In the Netherlands, corporate veil piercing takes place through the application of the identification theory developed by Dutch courts. According to Vandekerckhove, identification applies mostly in cases of statutory and contract interpretation. In these cases, courts may decide to set aside the separate legal personality of the different actors involved when the application of a legal or contractual rule that does not explicitly deal with legal persons requires that abstraction be made of the separate identity of the persons concerned. Identification occurs in cases where two persons have acted where only one should have acted. As a result of identification, affiliated corporations are considered to be one legal person, and the acts and liabilities of one corporation may be attributed to another corporation. Identification depends on the factual circumstances of the case. Dutch courts have identified various circumstances or factors that may give rise to identification, such as dominance of one company over another, intensive involvement in the management of a company, creation of expectations vis-à-vis third parties, commingling of assets, or close intermingling. In general, courts will identify affiliated corporations when respecting the formal, separate existence of both would lead to consequences that would be contrary to good faith. Courts should strike a balance between the purpose and the content of the contractual or legal norm and the rule that each affiliated corporation has its own dependent legal personality. However, the Dutch Supreme Court is reluctant to apply the identification theory and requires sufficient reasons to conclude the existence of identification (eg a close commingling of assets is not enough). In the opinion of the Supreme Court, the fact that affiliated corporations are closely related legally and economically and commingle their affairs does not provide a sufficient reason to conclude the existence of identification. In most cases, identification concerns parent and subsidiary corporations.⁷¹

The single economic entity in competition law

Over the years, EU competition law, which largely influences French and Dutch competition rules, has gradually accepted the recognition of parent companies’ liability for their subsidiaries’ acts in specific circumstances.⁷² In the landmark Akzo Nobel NV case,⁷³ the CJEU held that where a parent company had a 100 per cent shareholding in a subsidiary, the parent company could be held jointly and severally liable for the payment of the fine imposed on its subsidiary.⁷⁴ The fact that the subsidiary is wholly-owned by the parent company is sufficient to presume that the parent exercises a decisive influence over the commercial policy of the subsidiary.⁷⁵ The parent company has the burden of rebutting that presumption by adducing sufficient evidence to show that its subsidiary acts independently on the market.⁷⁶ This solution is seen as an application of the ‘single economic entity’ doctrine, which sees the parent company and its subsidiary form a single economic unit and therefore form a single undertaking.⁷⁷ When the parent company does not wholly own the subsidiary, the CJEU must seek additional evidence of the absence of the subsidiary’s autonomy and of the determining influence of the parent company on its subsidiary’s behaviour on the market. Such evidence can be demonstrated by showing the parent company’s influence on fixing prices or on the subsidiary’s management and commercial strategy.⁷⁸

The reception of EU case law by domestic courts is relevant here. In France, courts have adopted an interpretation that slightly departs from the Akzo Nobel NV judgment,⁷⁹ seeking additional proof of the lack of autonomy of the wholly-owned subsidiary.⁸⁰ For example, French courts will take into account the parent company’s financial participation in the capital of the subsidiary, the nomination of the managing body, or the possibility for the subsidiary’s managing body to freely determine an autonomous industrial, financial, and commercial strategy.⁸¹ In the Netherlands, Dutch courts have established that, if a parent company exercises ‘decisive influence’ over its subsidiary’s commercial behaviour, then both form part of the same economic undertaking. As a result, the parent company and the subsidiary can be fined for infringement of competition law.⁸²

Voluntary liability

Both France and the Netherlands accept ‘voluntary piercing’, which occurs when the parent company, as the shareholder of its subsidiary, voluntarily abandons its right to limited liability and agrees to be held jointly liable for its subsidiary’s acts.⁸³ In that event, the parent itself lifts the corporate veil, mostly vis-à-vis one particular creditor or group of creditors.⁸⁴

In company law, voluntary piercing in France may result from a guarantee by the parent company for liabilities of its subsidiaries to the benefit of third parties.⁸⁵ One example of voluntary piercing allows a parent, holding, or controlling company to assume liability for the environmental obligations of its subsidiary or controlled company where the latter has defaulted. More specifically, Article L233-5-1 French Commercial Code provides that a company which holds more than 50 per cent of the capital of another company,⁸⁶ or has a participation in⁸⁷ or controls another company,⁸⁸ may choose, in the event of the failure of the subsidiary or controlled company, to bear liability for all or part of the obligations to prevent and restore environmental damage caused by the subsidiary or controlled company.⁸⁹ Article L233-5-1 does not call the principles of separate legal personality and limited liability into question. It only makes it possible for a parent company to adopt ‘virtuous behaviour’ for reasons that are consistent with the protection of its image or with ethical rules or social commitments without creating a risk for the parent company of being accused of misuse of corporate assets.⁹⁰ Its added-value is therefore limited as it does not create an obligation on the parent company. Moreover, it targets only a limited number of environmental damages.⁹¹

In the Netherlands, the parent company can declare that it assumes joint and several liability for any obligations arising from the legal acts of its subsidiary in order to allow the latter to obtain an exemption from the duty to publish its annual accounts (Article 2:403(f) Dutch Civil Code).

Fault-based liability

A parent company may be held liable for the obligations of its subsidiary on the basis of a ‘fault’ or wrongful act. This fault-based liability is based on the traditional principles of tort law and applies in different areas of law, from company law to environmental law.

Tort law

Under general tort law, persons may be liable for harm caused by their own act (ie personal or direct liability). In certain circumstances, persons may also be liable for harm caused by the act of others (ie vicarious liability). In the context of corporate groups, while a parent company could potentially be held directly liable for harm caused by its subsidiary’s activities on the basis of its own misconduct, its liability for harm caused solely by the misconduct of its subsidiary is generally excluded.

Direct liability

In France, Articles 1240 and 1241 French Civil Code govern liability for one’s own act (responsabilité du fait personnel).⁹² First of all, Article 1240 lays down the basic principle of civil liability for misconduct. It states that ‘Any act whatever of man, which causes damage to another, obliges the one by whose fault it occurred to provide compensation for it’. Moreover, Article 1241 provides for civil liability where the damage is caused by negligence. It reads as follows: ‘Everyone is liable for the damage they cause not only by their act, but also by their negligence or imprudence.’ Both articles establish a fault-based liability regime (ie resulting from unlawful conduct). Initially, Article 1240 dealt with intentional faults (delicts), while Article 1241 governed imprudent or negligent faults (quasi-delicts), but this distinction has lost its meaning in practice.⁹³

Articles 1240 and 1241 lay down common rules that apply to all areas of liability for one’s own act. Furthermore, they apply to natural and legal persons (including companies), and to all protected rights and interests. Therefore, Articles 1240 and 1241 may be invoked as a legal basis in civil actions for the damage resulting from human rights violations and environmental pollution, such as in transnational litigation against MNEs. They require the satisfaction of three elements to give rise to liability: damage, fault (faute), and causation. In order to establish a fault, the judge will assess the unlawfulness (illicéité) of the tortfeasor’s conduct. A fault may result from the infringement of a number of pre-existing obligations, most likely a written rule contained in a statute or regulation (obligation légale ou réglementaire). However, in some cases, it can result from the infringement of an unwritten duty derived from custom practised in a particular region, sector, or profession, or private norms (eg codes of conduct or guidelines). The commission of a criminal offence causing harm to another person is also considered to be a fault in the sense of Article 1240.⁹⁴ Finally, a person commits a fault if they abuse a right (abus de droit), meaning they used a right to which they are entitled with the intent to cause harm. In general, judges have a broad power of appreciation for the standard of conduct. In most situations, courts will use a standard of reference (eg bonus pater familias). The Court of Cassation has recognized that a civil fault may result from the positive act or the mere omission of the tortfeasor.⁹⁵

In theory, Articles 1240 and 1241 provide a basis for holding a parent company liable in the context of corporate group activities. However, victims must show that the parent company has committed a fault, either intentionally or negligently, that has caused the damage.⁹⁶ They must also prove causation between the parent company’s fault and the damage, which is challenging when the damage occurs as a result of corporate group activities. While judges may accept the parent company’s fault in cases involving abuse of legal personality or mismanagement of the controlled company,⁹⁷ it is unclear whether they will accept the parent company’s fault in other cases.

In the Netherlands, Article 6:162 Dutch Civil Code lays down a general rule on fault-based liability under which both natural and legal persons can be held liable for their own intentional or negligent conduct. This provision requires a wrongful act or omission, imputability, causation, and damage.⁹⁸ Under Article 6:162(1), a person who commits a tort against another that is attributable to him must repair the damage suffered by the other in consequence thereof. Furthermore, Article 6:162(3) provides that a tortfeasor is responsible for the commission of a tort if it is their fault or results from a cause for which they are accountable by law or pursuant to generally accepted principles. As a result, tortious liability is incurred not only through subjective fault, but also through objective ‘answerability’. Article 6:162(2) Dutch Civil Code specifies the types of acts which are deemed tortious. There are three main categories: (1) the violation of a right; (2) an act or omission breaching a duty imposed by law; and (3) an act or omission breaching a rule of unwritten law pertaining to proper social conduct. Some of these acts may be more relevant than others in the context of transnational litigation against MNEs. First, there is a tort where the right of a person is infringed, such as the right to life, the right to physical integrity, or the right to freedom.⁹⁹ This category is directly relevant to transnational tort claims against MNEs in which plaintiffs raise human rights abuse claims. Second, liability arises where a wrongful act or omission violates a clear legal norm, such as Dutch laws and regulations or directly applicable norms of public international law.¹⁰⁰ However, few Dutch statutory norms apply in the context of transnational tort claims against MNEs.¹⁰¹ Third, transnational claims raising the liability of a parent company for its subsidiary’s activities abroad might also be built on the breach of unwritten norms pertaining to acceptable social behaviour.¹⁰²

In the context of transnational claims against MNEs, tort may be a valid way to hold parent companies liable when human rights violations occur in the context of their subsidiaries’ activities. Scholars have notably suggested that a parent company may have a general duty of care to prevent foreseeable harm to stakeholders caused by the actions of its subsidiaries.¹⁰³ However, courts have not yet relied on Dutch law to establish the liability of a parent company in the context of transnational claims against MNEs. Having said that, in Milieudefensie v RDS,¹⁰⁴ the high-profile climate change litigation case against Shell, the District Court of the Hague ruled that RDS, Shell’s parent company, owes an obligation, under the unwritten standard of care enshrined in Article 6:162 Dutch Civil Code, to reduce the Shell group’s CO2 emissions by net 45 per cent in 2030, compared to 2019 levels, through the Shell’s group corporate policy. This decision demonstrates that Dutch tort law can be a legitimate and effective tool for holding parent companies accountable for human rights and environmental violations that occur in the context of corporate groups activities.

Exclusion of vicarious liability

Both France and the Netherlands recognize vicarious liability under tort law. This means that a person may be held liable to repair the damage caused by a third party, not because of their own wrongdoing, but because of their relationship with the tortfeasor. However, the French and Dutch vicarious liability regimes do not recognize that a parent company may be liable for the torts of its subsidiary or, more broadly, any entity under its control or business partners of its supply chain.

In France, Article 1242 French Civil Code provides that a person is liable not only for the damage they cause by their own act, but also for the damage caused by the acts of persons for whom they are responsible (responsabilité du fait d’autrui).¹⁰⁵ However, Article 1242 does not establish a general vicarious liability regime and applies only to a limited number of relationships (ie liability of parents for damage caused by their children; liability of teachers and craftsmen for damage caused by their students and apprentices; liability of masters for damage caused by their servants;¹⁰⁶ and liability of principals (or employers) for damage caused by their agents (or employees)).¹⁰⁷ The relationship between a parent company and its subsidiary is not defined as one that could give rise to vicarious liability. Consequently, this absence prevents parent companies from being held liable for the acts of their subsidiaries pursuant to Article 1242.

Nonetheless, it has been suggested that vicarious liability could be imposed on the parent company for the wrongdoing of its subsidiary. There are two main reasons. First, the Court of Cassation has accepted that more relationships could give rise to vicarious liability under Article 1242.¹⁰⁸ For example, institutions dealing with minors and sport associations have been held strictly liable for the torts of persons under their control or whose activities they control.¹⁰⁹ However, the existence of a general vicarious liability regime has not yet been recognized by the Court of Cassation. Second, it has been argued that the principles of vicarious liability applying to the principals/agents relationship could be extended to ‘any relationships capable of meeting the tests of subordination or the right to give instructions’.¹¹⁰ To date, however, French courts have been reluctant to extend vicarious liability to the relationship between a parent company and its subsidiary.

Since the beginning of the 21st century, a number of official studies, which aimed to inform the reform of French tort law, have made various proposals regarding parent company liability for its subsidiaries. For instance, in 2005 the Catala report¹¹¹ suggested that the category of persons under now Article 1242 should be extended to natural or legal persons who organize and have an interest in the activity of professionals or businesses (not being their employees). Furthermore, it suggested that a new Article 1360 extend such liability to the relationship between parent companies and subsidiaries. Interestingly, the Catala report promoted the creation of a strict liability regime. Similarly, in 2012 the working group led by Professor Terré suggested the creation of a fault-based liability regime for corporate groups (ie Article 7).¹¹² However, to date, the French Government has not followed up on these suggestions.

The Dutch Civil Code provides vicarious liability for damage caused by the acts of a number of other persons (ie children, subordinates, non-subordinates, and representatives).¹¹³ However, similar to the French Civil Code, there is no specific mention of the relationship between a parent company and its subsidiary.

The Shell case in the Netherlands

In Shell, the Dutch courts applied Nigerian law and English tort law, and not Dutch tort law, to the facts. Nonetheless, an analysis of this case remains relevant as Shell highlights the substantive legal challenges plaintiffs face when seeking to establish the liability of the parent company in the context of corporate group activities.

In Shell, the plaintiffs sued the parent company, RDS, and its Nigerian subsidiary, SPDC, for damage resulting from oil spills from pipelines and a wellhead at various locations in the Niger Delta (Oruma, Goi, and Ikot Ada Udo) between 2004 and 2007. They alleged that RDS violated its duty of care by failing to properly oversee its Nigerian subsidiary SPDC.¹¹⁴ RDS had an obligation to act in a socially responsible manner and ‘should exert its influence and control over its subsidiary [SPDC] in such a way that it is prevented as much as possible that its subsidiary [SPDC] causes damages to human beings and the environment during the oil extraction’.¹¹⁵ RDS and SPDC claimed that the oil spills were the result of sabotage.

In 2013, in one of the claims the District Court found that SPDC was liable for damage resulting from the oil spills because it failed to take appropriate preventative and remedial action against the spills.¹¹⁶ However, it dismissed all the claims against RDS.¹¹⁷ It ruled that, pursuant to Nigerian law, ‘there is no general duty of care to prevent third parties from inflicting damage on others’. This implies that parent companies like RDS have no general obligation to prevent their (sub-) subsidiaries, such as SPDC, from inflicting damage on others through their business operations.¹¹⁸ The District Court also ruled that RDS did not have a duty of care to prevent oil spills occurring in the context of SPDC’s activities based on the English Chandler precedent (see Chapter 3 of this book).¹¹⁹ It found that the proximity between a parent company and its subsidiary’s employees when both companies operate in the same country, which was the situation in Chandler, ‘cannot be unreservedly equated with the proximity between the parent company of an international group of oil companies and the people living in the vicinity of oil pipeline and oil facilities of its (sub-) subsidiaries in other countries’. In the latter situation, ‘the requirement of proximity will be fulfilled less readily’. As a result, the District Court held that:

The District Court concluded that the special circumstances that can create a duty of care on the part of the parent company according to Chandler did not occur in this case.

The District Court adopted a narrow view of the duty of care of parent companies towards third parties in the context of MNE activities. Enneking questioned whether this narrow focus on the facts of the Chandler case and the criteria set out in that case was justified. In particular, she argued that it may be possible that under different circumstances, parent companies of MNEs may owe a duty of care to third parties in host countries who are adversely affected by the activities of groups there.¹²¹

Having said that, the District Court, and later the Court of Appeal of The Hague, also signalled that they would not exclude potential parent company liability in the context of corporate groups in this type of litigation, and in Shell in particular.¹²² In an interlocutory judgment, the Court of Appeal stated:

This approach turned out to be true. In January 2021, the Court of Appeal of The Hague overturned the 2013 decisions of the District Court. It delivered three judgments in the case against RDS and SPDC.¹²⁴ In two of them, it found that SPDC was strictly liable for the damage resulting from the leakages from the pipelines on the basis of Article 11(5)(c) of the Nigerian Oil Pipeline Act (OPA). Although SPDC had argued that the leakages were the result of sabotage, it did not meet the high evidence threshold that exists under the OPA (ie sabotage must be proved beyond any reasonable doubt). However, the Court of Appeal rejected that RDS could be held liable for the damage resulting from the leakages, since the strict liability regime under the OPA only applies to the holder of the oil pipeline licence (ie SPDC) and it could not be established that SPDC had acted negligently or unreasonably.

Furthermore, in two of the judgments, the Court of Appeal found SPDC liable for its failure to provide a timely and adequate response to the leakages on the basis of the tort of negligence. In one judgment, SPDC was found liable for neglecting to install a leak detection system (LDS) on the pipeline in Oruma. Importantly, the Court of Appeal found that RDS owed a common law duty of care to the people living in the vicinity of the Oruma pipeline. Influenced by the 2019 UK Supreme Court’s ruling in Vedanta¹²⁵ (see Chapter 3 of this book), the Court of Appeal held that ‘if the parent company knows or should know that its subsidiary is unlawfully causing damage to third parties in an area in which the parent company is interfering with the subsidiary, then, as a starting point, the parent company has a duty of care towards those third parties to intervene’.¹²⁶ Furthermore, whether RDS, as the parent company, owed a duty of care depended on ‘the extent to which, and the way in which, the parent availed itself of the opportunity to take over, intervene in, control, supervise or advise the management of the relevant operations … of the subsidiary’.¹²⁷ The Court of Appeal took the view that, at least from 2010, RDS had quite intensively interfered with the decision to equip the Oruma pipeline with an LDS. It considered several aspects, such as RDS’ substantial financial interest in Nigeria, its awareness of previous oil spills in the area, the bonus of members of RDS’ executive committee being partly dependent on the number of oil spills, specific documents and witness testimonies discussing the possibility of installing an LDS, and the structure/governance of the Shell MNE.¹²⁸ In addition, RDS was aware that SPDC was not able to respond adequately to leaks resulting from the Oruma pipeline due to the lack of an LDS. Therefore, RDS had a duty of care. By ignoring this duty and not compelling SPDC to install an LDS, RDS acted in tort. In the second judgment, the Court of Appeal dismissed the claim against RDS on the ground that it had not been made aware of the oil spill.

The judgments of the Court of Appeal mark a turning point in the search for the liability of the parent company in the context of its group activities in Europe. While in the past, courts in other countries have agreed to hold parent companies liable for damage resulting from the activities of their foreign subsidiary in a limited number of cases (eg COMILOG, AREVA), all of these decisions have ended up being reversed upon appeal. In Shell, the judgments of the Court of Appeal are likely to be final,¹²⁹ which means that they are the first judicial decisions in Europe to hold a parent company accountable for harm caused by its subsidiary. More importantly, it is the first time that a parent company has been found to owe a common law duty of care to the local communities affected by the operations of its subsidiary. According to Roorda:

However, the possible impacts of the Court of Appeal’s judgments on future litigation against parent companies should be taken with a grain of salt. The Court of Appeal interpreted that the parent company’s duty of care is based on two main elements: interference and knowledge. RDS was found to have a duty of care because it had sufficiently interfered with the decision to equip the Oruma pipeline with an LDS. However, as Roorda argues, this duty of care stems from RDS’ actual intervention in SPDC’s operations ‘rather than from its central position of authority in the corporate group’ or ‘its capacity to intervene’.¹³¹ This could create an incentive for parent companies not to interfere, or to limit signs of interference, in the activities of their foreign subsidiaries in order to avoid liability claims.¹³² In that event, victims may find themselves without the possibility of seeking redress against the parent company.

Company law

In disputes relating to company law, French and Dutch courts may hold the parent company liable for the damage caused by its subsidiary on the basis of the parent company’s fault or misconduct. They may do so on the basis of general tort liability or specific statutory schemes, such as directors’ liability in bankruptcy.

In France, the courts may accept the liability of the parent company for its subsidiary’s acts on the basis of Article 1240 French Civil Code. For example, in insolvency law, courts have found that a parent company had committed a fault where it maintained its subsidiary’s operations even though insolvency was clearly inevitable or where it gave harmful instructions to its subsidiary.¹³³ It is worth mentioning that the recently adopted Act on the Duty of Vigilance creates a fault-based liability regime for the parent company where its failure to comply with its mandatory HRDD obligations results in human rights or environmental damage in the context of its activities, as well as those of its subsidiaries and subcontractors (see Chapter 7 of this book for a detailed analysis).

In the Netherlands, tort is usually the principal basis for establishing the liability of parent companies for their subsidiaries’ debts under company law.¹³⁴ In the landmark Osby case of 1981,¹³⁵ the Dutch Supreme Court found that a parent company may commit a tort vis-à-vis its subsidiary’s creditors when it has such an influence over the management of the subsidiary that, at the time of the creation of the security, the parent company knew, or should have foreseen, that new creditors would be harmed by the lack of the subsidiary’s assets but nevertheless failed to satisfy the debts of those creditors.¹³⁶ Since this case, Dutch courts have refined the idea that a parent company may have a legal duty of care towards its subsidiary’s creditors. As such, it must prevent a subsidiary from taking on a new debt if it is clear that this debt will not be satisfied.¹³⁷ Dutch courts usually ask two questions. First, did the parent company know, or should it have known, that its act or omission would harm the creditors of the subsidiary (duty of care)? Second, what was the degree of involvement of the parent company in the management of its subsidiary (control)? When the parent company intensively influences the subsidiary’s daily management, it may be considered as a quasi-director and it may incur the same liabilities in the event of a breach of duty of care.¹³⁸

Both France and the Netherlands allow their courts to hold a company, acting as a de jure or de facto director of its subsidiary, liable for the debts of the subsidiary where the company has committed a fault. Article L651-2(1) French Commercial Code provides that, in the context of proceedings for insufficient assets (responsabilité pour insuffisance d’actif), the court may, in the event of a fault in management (faute de gestion) that has contributed to insufficiency of assets, decide that the debts of the legal entity will be borne, in whole or in part, by all or some of the de jure or de facto directors who have contributed to the fault. This type of liability is interesting in the context of groups, as legal persons can be considered directors. As a result, a parent company acting as the de jure or de facto manager of its subsidiary may be required to pay the debts of its subsidiary.¹³⁹ French courts appear to have considerable flexibility in finding parent companies liable in such cases.¹⁴⁰ A controlling shareholder and/or a parent company can be held to be a de facto director if they directly manage or take part in the management of the company.¹⁴¹ However, simple negligence on the part of the directors in the management of the company is not sufficient for them to be liable.

Similarly, the Dutch Civil Code provides that, when public and private limited companies become bankrupt, their director(s) shall be held liable for the amount of liabilities that cannot be satisfied out of the liquidation of the other assets if the director(s) have manifestly performed their duties improperly and it is plausible that the improper management was an important cause of the bankruptcy.¹⁴² This rule applies to a parent company acting as a de jure or a de facto director of its subsidiary. A parent company may be considered as a de facto manager when it has had a direct influence over the subsidiary’s management and when, in reality, the subsidiary’s formal management has been set aside.¹⁴³ This rule also applies to a foreign company acting as director of a Dutch company.¹⁴⁴

Environmental law

Similarly, the French and Dutch courts may hold a parent company liable for the environmental damage caused by its subsidiary on the basis of the parent company’s own fault or misconduct. They may do so on the basis of general tort liability rules or specific statutory schemes.

France

Legal framework

Under Articles 1240 and 1241, French courts may hold a parent company liable for the environmental damage caused by its subsidiary if the parent company has committed a fault that has caused the damage. However, courts are strict about the existence of the parent company’s fault. In the Ademe v Elf Aquitaine case,¹⁴⁵ the Court of Cassation rejected a claim to extend the liability of a subsidiary to its parent company for the clean-up of a landfill site. In the court’s view, there was no evidence that the parent company had committed a fault that would have justified making it liable pursuant to Article 1240. The simple fact of controlling or having a participation in another company is not sufficient to demonstrate the parent company’s fault. This rule applies even if the subsidiary is responsible for a public service that could pose a risk to the public interest.

Article L512-17 French Environmental Code is a specific statutory scheme creating a fault-based liability regime for the parent company. Where a subsidiary¹⁴⁶ enters liquidation proceedings, a court may hold its parent company liable for the cleaning up of the subsidiary’s operation site if the parent company has committed a ‘characterized fault’ (faute caractérisée) which has contributed to the lack of assets of the subsidiary. When the parent company is itself insolvent, the tribunal can hold the ‘grand-parent’ or the ‘great-grand-parent’ company liable, provided that it has committed a characterized fault. This provision prevents parent companies from using shell companies or insolvent subsidiaries as a means of avoiding liability. However, scholars have criticized the requirement of characterized fault, arguing that it does not exist in bankruptcy proceedings.¹⁴⁷ As a result, the necessary criteria for proving that the parent company committed a characterized fault, which contributed to the subsidiary’s insufficient assets, are uncertain.¹⁴⁸ Furthermore, only a limited number of individuals, such as the liquidator or the State, can apply for such a procedure, thus limiting opportunities for other actors, such as environmental NGOs, to bring a claim before a tribunal.

France recently created a civil regime aimed specifically at repairing ‘pure environmental damage’.¹⁴⁹ Under the new Article 1246 French Civil Code, ‘Any person responsible for ecological loss is obliged to repair it’. Ecological loss is defined as the ‘non-negligible harm to the elements or functions of ecosystems or to the collective benefits derived by man from the environment’.¹⁵⁰ Therefore Article 1246 aims to repair damage to the environment itself, not to humans or their property. Importantly, Article 1246 is applicable to both natural and legal persons responsible for ecological loss. In theory, it may be possible for a parent company to be held liable for the ecological loss arising in the context of its subsidiary’s activities. However, the parent company must have committed a fault leading to such damage. Another obstacle is that only a limited number of public authorities and environmental NGOs have standing to bring a claim under Article 1246.¹⁵¹ In practice, this means that affected local communities cannot invoke Article 1246 on their own. The concept of ecological loss was first introduced in the landmark Erika case, which deserves attention for its contribution to the liability of the parent company in the context of corporate group activities.

The Erika case

The Erika case was a landmark case for a number of reasons. Not only did the courts recognize the concept of ecological loss for the first time, they also held the parent company of Total, an oil MNE, liable for environmental pollution caused by its subsidiary on breaking grounds.¹⁵² Erika was concerned with oil pollution on the high seas, which is an area governed by specific rules of international law. The International Convention on Civil Liability for Oil Pollution Damage (CLC)¹⁵³ establishes a specific system of civil liability that deals with damage resulting from maritime casualties involving oil-carrying ships. Erika was a complex case raising various legal and procedural issues. However, the following summary focuses mainly on the search for criminal and civil liability of Total SA, the French parent company.

In 1999, the 25-year-old Maltese-flagged tanker Erika sank off the coast of Brittany and spilled 31,000 tons of heavy fuel oil belonging to the Panama subsidiary of Total SA along 400 kilometres of French coastline.¹⁵⁴ Following the oil spill, several natural and legal persons, including the parent company, Total SA, were prosecuted on various grounds. Victims, including local communities, fishermen, and NGOs, also introduced ancillary civil actions in order to obtain compensation for the damage caused by oil pollution. In this case, the Paris Regional Court and Court of Appeal, and the Court of Cassation successively ruled on criminal and civil proceedings.¹⁵⁵

The French courts found Total SA guilty of the criminal offence of involuntary ship pollution for having exercised de facto control and direction in the management or operation of the ship. This offence normally applies to the person responsible for the operation of the ship, which in this instance was Total SA’s subsidiary. Nonetheless, the courts concluded that Total SA had control over the management of the tanker. First, although it was not a contracting party to the charter party,¹⁵⁶ Total SA had to enforce a number of the obligations in the contract. For example, in the event of an accident the captain of the ship had to inform Total SA immediately. Second, Total SA had retained a right to check vessel compliance under its vetting procedure. The charter party allowed Total SA to verify the care and the diligence with which the shipment was transported, as well as the ways in which the ship and the crew were managed. The courts concluded that Total SA was the real decision-maker and that the Panama subsidiary was ‘an empty shell’, as it did not have any team or building in Panama where it was registered and lacked legal and financial autonomy. Furthermore, the courts found that Total SA had made an abusive use of the charter party to separate the legal and financial risks of the tanker management and, therefore, avoid liability. An important aspect of Erika is that the courts assessed Total SA’s behaviour on the basis of its own internal rules of control. Total SA had voluntarily set up a number of procedures for its own activities, including a specific vetting procedure to control the quality of tankers. The courts concluded that, by ignoring this procedure and not vetting the Erika, Total SA had neglected its duty of care. Total SA’s voluntary commitment to control the quality of tankers became a norm upon which the company’s misconduct was assessed.

However, the Court of Appeal rejected the claim that Total SA could be held liable in tort for the damage caused by the pollution.¹⁵⁷ It held that the CLC places the liability for damage resulting from maritime casualties involving oil-carrying ships on the owner of the ship from which the polluting oil escaped or was discharged. This liability is strict and exonerates other potential parties from being held civilly liable unless these parties committed gross negligence. The Court of Appeal did not find that Total SA had committed any gross negligence, as it had not expected that pollution would occur, even though it did not respect its own vetting rules. However, the Court of Cassation overturned this point of the ruling, holding that Total SA had acted recklessly (faute de témérité) within the meaning of the CLC and that it was necessarily aware that damage would probably result from such behaviour. Ultimately, Total SA was held criminally and civilly liable for the oil pollution caused by the activities of its subsidiary on the basis of its own fault.

The Netherlands

In the Netherlands, the courts have held parent companies liable for environmental pollution caused by the activities of their subsidiaries on the basis of their own fault. It is worth mentioning two important environmental cases where general tort law was applied. First, in the Roco BV v De Staat der Nederlanden case, Rouwenhorst was the owner of premises that were heavily polluted. In order to escape liability, he transferred the business to a newly incorporated limited liability company, Roco BV (Roco), which continued to operate at another location. Rouwenhorst’s spouse and Hoekstra BV (a holding controlled by Rouwenhorst and his spouse and of which Rouwenhorst was the sole director) held the shares of Roco. The Dutch State claimed reimbursement for costs related to the environmental clean-up. At first instance, the District Court dismissed the State’s claim because Roco had been incorporated after the pollution had been caused and it had not accepted liability for the pollution upon the business transfer.¹⁵⁸ However, the Court of Appeal held Roco liable since the sole purpose of the transfer was to evade potential claims by the State, as Roco continued the business of its predecessor.¹⁵⁹ Eventually, the Supreme Court upheld that Roco was liable for the environmental clean-up.¹⁶⁰ However, it rejected the use of the theory of identification as a basis for liability since ‘the case did not concern an identification of two legal or natural persons but rather an identification of an “enterprise” with the company’.¹⁶¹ Furthermore, Roco could not be held liable on the basis of a successor liability theory, as this did not exist under Dutch law. Ultimately, the Supreme Court held Roco liable in tort for having continued the business with the clear intent of frustrating the State’s claim for damages.¹⁶²

Second, in the Bato’s Erf BV v De Staat der Nerderlanden case,¹⁶³ which concerned soil and groundwater pollution, the parent company had modified its charter and name, and had transferred its operations to a newly incorporated wholly-owned subsidiary in order to avoid liability. The Court of Appeal held that the parent company and the subsidiary had to be identified because the two companies were closely intermingled. In doing so, it took into account the following factors: the parent company had incorporated the subsidiary; it had transferred most of the assets and liabilities; the subsidiary was the true operator conducting the business of the parent company, which controlled the activities of the subsidiary; both companies had the same directors; the subsidiary was wholly-owned; and the financial statements of both companies were consolidated.¹⁶⁴ However, the Dutch Supreme Court overturned this decision, holding that the mere fact that a parent company determines the business policy of the subsidiary and directs or influences its implementation, either by having its managing directors also act as managing directors of the subsidiary, or in its capacity as managing director and/or sole shareholder, does not mean that these activities become the activities of the parent company, as a result of which the parent company would automatically be held liable for the tortious activities of the subsidiary.¹⁶⁵ This case showed that mere directorship is insufficient for liability, and that negligence or gross negligence must be established.¹⁶⁶

Labour law

In many countries, labour law plays an important role in protecting workers from harmful or unfair working conditions and ensuring that workers can exercise their labour rights. It often lays down the rules on liability to ensure that employers comply with labour law standards designed to protect workers. In France, the theory of ‘co-employment’ (or ‘co-employers’) is a relevant basis for finding parent companies liable to employees of their subsidiaries in the context of MNEs.

Co-employment

The theory of co-employment is a judicial creation that has prompted renewed interest in France over the last few years.¹⁶⁷ It challenges not only the legal principles of separate legal personality and limited liability in corporate groups, but also the contractual foundation of employment relationships.¹⁶⁸ Co-employment generally enables employees of a loss-making company to hold their employer and its parent company jointly and severally liable for financial compensation for the loss of their jobs. French courts have applied the co-employment theory to protect employees and to punish abnormal practices within corporate groups, especially when parent companies are also holding companies that benefit from an advantageous tax regime.¹⁶⁹ In general, the co-employment theory applies to situations in which a (parent) company owns some of the capital of another company.¹⁷⁰ Furthermore, the Court of Cassation has accepted that it could apply to MNEs.¹⁷¹

In general, the fact that two companies belong to the same group is not enough to justify co-employment.¹⁷² French courts accept that there is co-employment in two situations. First, there is co-employment where there is a relationship of subordination between the parent company and the subsidiary’s employee. This relationship of subordination may be demonstrated by the parent company’s interference in the management of the subsidiary’s employees.¹⁷³ Second, there is co-employment where there is a ‘commingling of interests, activities, and management’ (confusion d’intérêts, d’activités, et de direction) between the two companies.¹⁷⁴ In this second situation, French courts consider various indicators, such as the economic control of the subsidiary by the parent company or the lack of the subsidiary’s independence to define its own strategy.¹⁷⁵ In recent years, however, the Court of Cassation has adopted a strict approach to co-employment where there is a commingling of interests, activities, and management. It held that ‘a company belonging to a corporate group cannot be considered the co-employer of the employees of another company, outside the existence of a relationship of subordination, unless there is between them, beyond the necessary coordination of economic actions between companies belonging to a same group and the state of economic domination that belonging to the same group may produce, a commingling of interests, activities, and control demonstrated by the interference in the economic and social management of the latter’.¹⁷⁶ French courts seem to require strict criteria to prove such interference.¹⁷⁷ It should also be noted that the Court of Cassation has rejected the possibility that co-employment might be deduced from the sole ownership of a wholly-owned subsidiary. In a situation of co-employment, the subsidiary must not have autonomy in the management of its human resources.¹⁷⁸

The AREVA case

The AREVA case showed the limits of the co-employment theory as a basis for parent company liability. Pursuant to Article L4541 French Code of Social Security (Code de la sécurité sociale), the plaintiffs brought a compensation claim for occupational disease against AREVA for gross negligence (faute inexcusable)¹⁷⁹ in respect of Venel, an employee of its Nigerien subsidiary AREVA NC.

In 2012, the Melun Social Security Tribunal (TASS) held that AREVA, as the co-employer of Venel, was liable for gross negligence. The TASS proceeded to apply a two-stage analysis. First, it found that AREVA and AREVA NC were co-employers since they ‘pursued, in collaboration, simultaneously, indivisibly, and permanently, a common activity in a common interest, under a single authority’.¹⁸⁰ Indicators included AREVA NC’s charter, the identity of its main shareholder, and interconnections between AREVA and its subsidiary (same address, same activities, same involvement in the exploitation of the same mining site). Moreover, AREVA appeared to assume technical, economic, social, and financial liability for the potential impact on the health and safety of individuals working in its uranium mines by setting up ‘health observatories’ and signing a memorandum of understanding on occupational disease caused by ionizing radiation with Sherpa in 2009.¹⁸¹ AREVA’s voluntary commitment demonstrated that, while its subsidiary acted as the contractual employer of Venel, AREVA acted as the employer with the authority and power to control and organize working conditions, especially with regard to occupational risk management. Therefore, a subordinate relationship existed between AREVA and the employee. Second, the TASS found that AREVA had committed gross negligence by not setting up safety measures to protect workers in its mines, which caused the development of the disease.

However, in 2013, the Paris Court of Appeal overturned the TASS’ judgment, rejecting the claim that AREVA was the co-employer of Venel.¹⁸² It found that there was no subordinate relationship between AREVA and Venel, as there was no evidence that AREVA had exerted any power of direction, control, or discipline over Venel. Furthermore, it held that there was no commingling of activities, interests, and control between AREVA and its subsidiary. First, AREVA NC could not be considered AREVA’s subsidiary pursuant to Article L233-1 French Commercial Code, which requires that a company owns more than 50 per cent of the capital of another company in order for the second company to be regarded as a subsidiary; AREVA owned only 34 per cent of AREVA NC’s shares while the Nigerien State and other foreign companies owned the rest. Second, there was no evidence demonstrating that AREVA NC had lost the autonomy to manage its own activities. The fact that both companies shared a common interest, as a result of AREVA being AREVA NC’s shareholder, did not constitute a commingling of management or activities. Third, even though AREVA owned AREVA NC’s mining concession, the Court of Appeal rejected the claim that there were interconnections between both companies demonstrating dependence. Fourth, the Court of Appeal rejected the claim that AREVA’s voluntary commitment made it the co-employer of Venel. The Court of Cassation upheld this ruling in 2015.¹⁸³

The position of the French courts in AREVA is in line with the approach adopted by the Court of Cassation, which requires the demonstration of strict criteria showing the existence of co-employment. However, a lack of consistency, clarity, and certainty remains as to the exact criteria required. Furthermore, this approach restricts the situations that may qualify as co-employment. For instance, it appears that a parent company must own 50 per cent of its subsidiary’s capital within the meaning of the French Commercial Code in order for the relationship between the parent company and the subsidiary to qualify for co-employment. This criterion does not allow a parent company to be held liable in the context of a joint venture, such as in AREVA. Ultimately, such a position reduces the possibilities the co-employment theory could potentially provide to victims of labour rights abuse by MNEs in host countries.

Criminal liability

Under French and Dutch criminal law, the parent company may, in theory, be liable for a crime committed by its subsidiary if it is itself a primary or, in some situations, a secondary perpetrator of the crime or an accomplice to it. It should be noted that, to date, no transnational case has yet reached the merits stage.

Corporate criminal liability

The concept of the criminal liability of legal persons is relatively new in France and the Netherlands. Both countries were among the first European countries with a civil law tradition to adopt a comprehensive regime of corporate criminal liability.¹⁸⁴

In France, discussions over the use of criminal sanctions to regulate corporate misconduct emerged in the 1980s after the country was confronted with the increase in MNEs’ power and their ability to evade local regulatory requirements.¹⁸⁵ The new Criminal Code introduced corporate criminal liability in 1994. At that time, companies could be held criminally liable only if such liability existed under statutory law. This legal restriction had a direct impact on transnational claims against MNEs. In Total, the plaintiffs targeted Total’s executives, and not the company, as the Criminal Code did not provide for corporate criminal liability for the alleged offences.¹⁸⁶ It was not until 2004 that the criminal liability of legal persons was extended to all criminal offences.¹⁸⁷ Article 121-1 French Criminal Code now provides that legal persons are criminally liable for offences committed on their behalf by their organs or representatives. Furthermore, the criminal liability of legal persons does not exclude that of natural persons who are perpetrators or accomplices of the same acts.

In the Netherlands, in 1951 the Economic Offences Act (Wet economische delicten) recognized that legal persons, including companies, could be criminally liable for a number of economic crimes. However, in 1976 a major criminal reform introduced general corporate criminal liability. Since then, Article 51(1) Dutch Criminal Code provides, in broad terms, that criminal offences may be committed by natural and legal persons. Furthermore, Article 51(2) Dutch Criminal Code states that when a legal person commits a criminal offence, criminal proceedings may be instituted and punishments may be imposed not only on the legal person but also on persons who ordered the commission of the criminal offence or directed the unlawful acts.

Criminal liability in corporate groups

Corporate groups cannot be held criminally liable in France and the Netherlands. Pursuant to Article 121-2 French Criminal Code, only business entities that have legal personality can be held criminally liable. Since corporate groups do not enjoy legal personality, they cannot be criminally liable.¹⁸⁸ Similarly, in the Netherlands, only business entities with legal personality can be criminally liable. Nonetheless, Article 51(3) Dutch Criminal Code states that criminal offences may also be committed by certain entities without legal personality, such as unincorporated companies or partnerships. However, the corporate group is excluded from this provision. Ultimately, only group entities may be liable for criminal activities, such as those related to human rights violations and pollution, in France and the Netherlands.

In general, under French and Dutch criminal law, the parent company will be held liable for the offences it perpetrated. In France, Article 121-1 French Criminal Code states that no one is criminally liable except for their own conduct. Therefore, this principle of personal liability prevents the emergence of criminal vicarious liability in the context of corporate group activities.¹⁸⁹ This is reinforced by the fact that it may be difficult to determine which company of the group committed the offence.¹⁹⁰ At times, French courts may take into account the economic reality of the corporate group to hold the parent company criminally liable for criminal offences involving its subsidiary. However, even where this is the case, the liability of the parent company is always based on its own misconduct. For example, in Erika, the parent company was held criminally liable for involuntary pollution caused by the oil tanker chartered by its subsidiary, because while the parent company had an effective power of control over the tanker, it did not carry out controls that would have prevented an unseaworthy tanker that subsequently sank being used for the shipment.¹⁹¹

The Lafarge case is relevant here because the Court of Cassation addressed the parent company’s potential criminal liability for interfering with the management of its foreign subsidiary. In 2019, the Investigating Chamber upheld Lafarge’s indictment for endangering others through a manifestly deliberate breach of the employer’s safety obligation under the French Labour Code on the grounds that the Syrian employees who ensured the continuity of the plant’s operations had been exposed to a risk of death or injury even though they had not received adequate training in the event of an attack. Lafarge challenged this decision, in particular the fact that the employees of the foreign subsidiary could be considered as employees of the French parent company, as well as the application to this specific case of the employer’s safety obligation imposed by French law. In 2021,¹⁹² the Court of Cassation ruled that the Investigating Chamber was correct in finding evidence of a subordinate relationship between the Syrian employees and Lafarge, or of permanent interference by Lafarge in the management of the employing company, resulting in the latter’s total loss of action autonomy. On the other hand, it considered that the Investigating Chamber could not deduce the applicability of the French Labour Code and should have examined, in the light of international law, the provisions applicable to the employment relationship between the French company and the Syrian employees, and then determined whether these provisions provided for a specific safety obligation that had been breached. As a result, the Court of Cassation overturned the Investigating Chamber’s decision to uphold Lafarge’s indictment for endangering the lives of Syrian employees.

In the context of mergers and acquisitions, the Court of Cassation has consistently rejected the possibility that the acquiring company could be held liable for the criminal offences committed by the acquired company prior to acquisition based on the principle of personal liability found under Article 121-1.¹⁹³ However, in a recent landmark ruling, the Court of Cassation reversed its position on the transfer of criminal liability of a legal person in the event of a merger of one company into another.¹⁹⁴ It ruled that, under certain conditions, the acquiring company may be subject to a fine or confiscation for acts constituting an offence committed by the acquired company prior to acquisition. The Court of Cassation’s new interpretation may prevent future mergers from impeding corporate criminal liability.

The Dutch Criminal Code provides possible grounds for the criminal liability of the parent company in the context of corporate group activities. Pursuant to Article 51(2), a parent company may be liable if it has committed the criminal offence (primary liability). Importantly, Article 51(2) provides for secondary liability where an offence is committed by a legal person. It applies to persons who have ordered the commission of the offence and to persons who have actually directed the commission of the offence.¹⁹⁵ This provision does not distinguish between natural and legal persons, which means that, theoretically, a parent company may be liable if it has ordered or actually directed the commission of an offence by its subsidiary. In addition, this secondary liability makes it possible to punish a merely passive involvement in an offence committed by a legal person. The Dutch Supreme Court has ruled that ‘conditional intent’ (dolus eventualis) is sufficient for this form of secondary liability.¹⁹⁶

Importantly, a parent company may, in theory, be held liable as an accomplice to a crime in both France and the Netherlands. Article 121-7(1) French Criminal Code provides that an accomplice to a crime or misdemeanour is a person who knowingly, by aid or assistance, has facilitated the preparation or commission of a crime or misdemeanour. Furthermore, Article 121-7(2) French Criminal Code states that an accomplice is also the person who by gift, promise, threat, order, abuse of authority or power, has provoked an offence or given instructions to commit an offence. Therefore, a parent company must have committed these acts in order to be held liable as an accomplice to a crime committed by its subsidiary. Where crimes are committed by its foreign subsidiary abroad, the parent company may be held criminally liable as an accomplice under the conditions laid down in Article 113-5 French Criminal Code. However, Chapter 5 of this book has shown that these conditions are generally restrictive. A number of transnational cases against MNEs have raised the criminal liability of the parent company as an accomplice in crimes committed abroad, including crimes under international law (ie Rougier, Amesys, Qosmos, Lafarge, BNP Paribas). It should be noted, however, that the parent company has been accused of complicity in crimes committed not only by its subsidiary but also by foreign governments.

In the Lafarge case, the Court of Cassation issued a landmark decision on the definition of complicity in crimes against humanity.¹⁹⁷ In this case, the Investigating Chamber determined in 2019 that there was sufficient evidence to conclude that the armed groups committed crimes against humanity (concerted plan of abuses, widespread and systematic attack on the civilian population) and that Lafarge paid them funds despite being aware of the nature of the abuses. The Investigating Chamber concluded, however, that there was no serious or corroborating evidence of Lafarge’s complicity because the financing of the armed groups was intended to allow it to continue its activity in the middle of a war zone, rather than to be associated with the crimes committed. In September 2021, the Court of Cassation overturned the Investigating Chamber’s decision to annul Lafarge’s indictment for complicity in crimes against humanity. It considered that one can be an accomplice to crimes against humanity, even if they do not intend to be associated with the commission of these crimes. It is necessary and sufficient to have knowledge of the preparation or commission of these acts, as well as that aid or assistance facilitated them. It is therefore not necessary to belong to the criminal organization or to participate in the conception or execution of the criminal plan. In this case, the knowing payment of several million dollars to an organization with an exclusively criminal purpose qualifies as complicity, regardless of whether the person is acting in furtherance of a commercial activity.

In the Netherlands, on the basis of Article 48 Dutch Criminal Code a parent company may be held criminally liable as an accomplice to a criminal offence committed by its subsidiary if it either intentionally aides and abets in committing a crime or provides opportunity, means, or information to commit a crime. Article 48 applies to the most serious criminal offences. Article 52 Dutch Criminal Code provides that complicity is not punishable for minor offences.

Elements of corporate criminal liability

In general, corporate criminal liability will be established when the objective element (actus reus) and, in some circumstances, the subjective element (mens rea) of the criminal offence are gathered.

Actus reus

In France, Article 121-2 French Criminal Code provides that legal persons are criminally liable for the offences committed on their account by their organs or representatives. Two conditions are therefore required. First, the criminal offence must have been committed on behalf of the company.¹⁹⁸ This means that the criminal offence must have been committed for the benefit of the company and not just for the individual benefit of the organ or the representative. Importantly, it is not required that the company gained a financial benefit from the criminal offence.¹⁹⁹ Second, an organ or a representative of the company must have committed the criminal offence. An organ may be defined as the person, either an individual or a group, who has the power of direction or organization within the company, such as a director, a board of directors, or a general assembly.²⁰⁰ A company can also be liable when the criminal offence was committed by a de facto director.²⁰¹

The situation is more complex when a representative commits the criminal offence. Previously, French courts held that a representative was an individual possessing the power, either general or special, to represent the company. Nonetheless, French courts recently extended the concept of representative to individuals, whether employees or not, who intervene on behalf of the company.²⁰² They also accept that a company may be liable, even though it is not possible to identify the perpetrator²⁰³ or the company’s representatives did not commit a fault.²⁰⁴ Furthermore, French courts accept that the person who holds a delegation of authority is a representative of the legal person. As a result, a legal person can be held criminally liable for any offences a delegatee commits on its behalf. It is worth mentioning the situation where several companies appoint a joint delegatee in the context of a common project. In the event of a violation of health and safety standards committed by the joint delegatee, the Court of Cassation has consistently held that the violation only engages the criminal liability of the company which is the employer of the victim. In the context of transnational litigation against MNEs, this means that only the company that has hired the victim can be held liable. Even if another company has benefited from the delegation, such as the parent company, it cannot be held liable.

In the Netherlands, a legal person will be liable for criminal offences if the relevant behaviour can be reasonably attributed to that legal person. In Drijfmest, the Dutch Supreme Court ruled that a corporation could be held criminally liable only if there was an illegal act or omission that could be reasonably imputed to that corporation.²⁰⁵ The Supreme Court provided a guiding principle to assess this ‘reasonable attribution’: ‘the attribution of certain (illegal) conduct to the corporation may under certain circumstances be reasonable if the (illegal) conduct took place within the “scope” of the corporation.’²⁰⁶ There are four situations in which conduct will, in principle, be carried out ‘within the scope of a corporation’. First, the act or omission was allegedly committed by someone who works for the corporation, whether or not under a formal contract of employment. Second, the conduct was part of the everyday ‘normal business’ of the corporation. Third, the corporation profited from the relevant conduct. Fourth, the alleged course of conduct was at the ‘disposal’ of the corporation that ‘accepted’ the conduct. In the latter situation, the failure to take reasonable care to prevent the conduct from being carried out may establish acceptance.²⁰⁷ Furthermore, any employee can cause its employer to commit an offence as long as the facts can be construed to show that the corporation ultimately ‘committed’ the offence.

Dutch scholars suggest that the Supreme Court’s approach towards corporate criminal liability can be characterized as flexible and ‘open’, as there is no rigorous theory to turn to for guidance.²⁰⁸ This approach has several advantages, as it ‘leaves room for “tailor-made” jurisprudence, in which the courts are free to weigh relevant circumstances and factors. It acknowledges that the possible variation in cases is, in fact, endless’.²⁰⁹ As a result, the Dutch approach may leave room for relevant jurisprudential developments in the context of transnational litigation against MNEs.

Mens rea

Specific categories of criminal offences require proof of a subjective element (intent). While French criminal law requires proof of the subjective element for both crimes and misdemeanours, Dutch criminal law requires this proof only for crimes (misdrijven) and excludes it for misdemeanours and contraventions (overtredingen).²¹⁰ As a result, where intent is required, there can be no liability without the intent of committing a crime. This rule creates difficulties in the context of corporate criminal liability, as legal persons are incapable of possessing intent to commit a crime. Therefore, courts have adopted creative approaches to adapting mens rea in a company context.

Under French law, corporate criminal liability does not depend on the commission of a fault by the legal person. The Court of Cassation has clearly stated that the criminal misconduct of the organ or representative of the legal person is sufficient to engage the criminal liability of the legal person, when it is committed on behalf of the legal person, without the need to establish a separate fault on the part of the legal person.²¹¹

In the Netherlands, in order to establish corporate criminal liability, it is necessary to prove that the corporation has acted intentionally, recklessly, or with gross negligence. Dutch case law demonstrates two main approaches for proving corporate intent and negligence in the Netherlands.²¹² First, the mens rea of a natural person is attributed to the company (indirect approach).²¹³ Such imputation is dependent on the internal organization of the corporation, and the position and responsibilities of the natural person within the corporation. It is also possible to combine the intent of multiple natural persons and impute such ‘united intent’ to the corporation.²¹⁴ Second, the mens rea of the corporation is established by the existence of intent or negligence of the corporation itself (direct approach). Corporate mens rea can be derived from circumstances closely related to the company itself, such as its policies and decisions. A company may confess by means of its agents,²¹⁵ for example stating in court that management did not act to prevent fraudulent acts that it knew were taking place within the company.²¹⁶ The latter approach is particularly suited to cases of gross negligence, which can be derived ‘objectively’ from the failure of a person to act in accordance with standards of conduct. According to this approach, corporate criminal liability is established based on deficiencies within the structures, policies, and culture of the corporation itself.²¹⁷

4 Corporate social responsibility: an increasing source of liability?

The proliferation of soft law instruments and the trend towards corporate self-regulation may inadvertently contribute to the development of liability regimes.²¹⁸ Some NGOs and scholars argue that CSR instruments, whether soft law instruments or voluntary commitments of companies, are capable of creating obligations for MNEs, the breach of which may trigger corporate liability.²¹⁹ In practice, CSR instruments may be useful to courts in assessing misconduct of companies. Plaintiffs in transnational litigation against MNEs have used the CSR commitments of MNEs to support liability claims for human rights abuse and environmental damage. However, their arguments have been met with a mixed reception from the courts.

In France, labour courts have sometimes used ethical codes to demonstrate the employer’s management power or to assess the gravity of the employee’s failure to comply with a professional duty.²²⁰ In some transnational cases against MNEs, judges have creatively used CSR instruments to assess MNEs’ breach of their duty of care. In AREVA, the TASS held that by setting up health observatories and signing a memorandum of understanding on occupational diseases caused by ionizing radiation with Sherpa, the parent company AREVA appeared to assume liability for the potential impact on the health and safety of individuals working in its uranium mines. The TASS used AREVA’s voluntary CSR commitments to find that AREVA had acted as the employer of Venel with the authority and the power to control and organize his working conditions. In addition, in Erika, the parent company Total SA had voluntarily set up a specific vetting procedure to control the quality of its tankers. The French courts found that by ignoring this procedure, Total SA had neglected its duty of care. Total SA’s voluntary commitment to control the quality of tankers became a norm for assessing the faulty conduct of the company.²²¹

However, courts do not automatically accept that a company’s voluntary commitments create enforceable obligations or that a breach of a company’s voluntary CSR commitments may give rise to liability. In the same AREVA case, the Court of Appeal ultimately rejected that AREVA’s voluntary commitments led, either directly or through its subsidiaries, to a situation where AREVA automatically became the employer, or co-employer, of Venel. The argument that soft law instruments may lead to enforceable obligations, or that the breach of voluntary commitments may give rise to liability, has also revealed its limits in Alstom. In this case, the plaintiffs argued that the defendant companies had failed to fulfil their commitments to comply with the relevant rules of public international law enshrined in their code of ethics and the UN Global Compact that they had signed.²²² In particular, they insisted upon the binding nature of the norms contained in the CSR instruments the companies had committed to respect. In 2013, however, the Versailles Court of Appeal rejected their argument on the basis of the non-voluntary nature of the soft law instruments relied on. First, it held that the UN Global Compact’s application is ‘based solely on the goodwill of the corporations. It has no binding effect … The Global Compact being no more than a point of reference, non-compliance with its principles cannot be invoked to justify a claim for violation of international rights.’²²³ Second, the Court of Appeal found that the companies’ codes of ethics stated that they are of a ‘strictly voluntary’ and non-binding character. As these codes stemmed from a personal initiative and provided no sanction, they could not be considered binding, nor could they be relied upon by third parties. Ultimately, the Court of Appeal concluded:

In the Netherlands, the question whether soft law instruments could create enforceable obligations was raised in the BATCO case.²²⁵ The Amsterdam Court of Appeal annulled an English parent company’s decision to close down its Dutch subsidiary in order to concentrate production at a Belgian subsidiary. It held that the lack of appropriate consultation with the trade unions and the work council by the Dutch subsidiary amounted to mismanagement in breach of the OECD Guidelines, to which the English parent was committed. For some authors, BATCO has shown that Dutch courts may consider the OECD Guidelines when determining the duty of care of companies under Dutch tort law.²²⁶

In Shell, the plaintiffs have argued that the parent company, RDS, had a duty of care to influence and control its subsidiary, SPDC, to prevent damage to humans and the environment in Nigeria.²²⁷ It is worth noting that they have argued that this obligation is reinforced by the fact that the MNE committed to implement various CSR instruments, such as the OECD Guidelines, the UN Global Compact, and the Global Reporting Initiative, which prescribe an active duty of care for parent companies.²²⁸ For instance, pursuant to the OECD Guidelines, RDS should have set up and maintained a suitable environmental management system, including the development of emergency plans, and best practices procedures and technologies to be available in the event of oil spills.²²⁹ Furthermore, the plaintiffs alleged that RDS failed to respect precautionary measures set in the UN Global Compact²³⁰ and to report the oil spills in accordance with the Global Reporting Initiative.²³¹ However, in its 2021 decision, the Court of Appeal did not address these arguments.

5 Conclusions

This chapter has discussed the challenges to and opportunities for holding MNEs liable for damage arising in the context of group activities under French and Dutch law.

The first difficulty arises from the fact that the corporate group is not recognized as a legal entity in France and the Netherlands. As a result, groups cannot have rights and obligations or be liable to pay damages. Consequently, the liability arising from any obligations of the entities of the group shall be borne by those entities. Another difficulty arises from the fact that French and Dutch law recognizes the separate legal personality of the company and provides for limited liability of shareholders for certain companies. In the context of MNE activities, separate legal personality and limited liability generally prevent the parent company from being held liable for a subsidiary, even where the parent company owns or controls that subsidiary.

However, French and Dutch law lay down a number of bases on which to hold the parent company to account for its subsidiary’s obligations. First, courts may, in some circumstances, disregard the application of separate legal personality and limited liability by piercing the corporate veil. In company law, this practice is accepted in exceptional cases. French and Dutch courts are generally reluctant to pierce the corporate veil and require strict conditions to do so. At the same time, corporate veil piercing is generally accepted in competition law where a parent company with a 100 per cent shareholding in a subsidiary can be held jointly and severally liable for the payment of the fine imposed on its subsidiary. Furthermore, voluntary corporate veil piercing may take place where a parent company accepts liability for the obligations entered into by its subsidiary.

In addition, courts may hold a parent company liable for damage resulting from its subsidiary’s activities where the parent company has committed a fault that has caused or contributed to the damage. This fault-based liability can be found in various situations across legal fields, including tort law, environmental law, and labour law. In the context of transnational litigation against MNEs, this means that a parent company could potentially be held liable for human rights violations or environmental pollution occurring as a result of its subsidiary’s activities where its misconduct has contributed to the damage. To date, however, few cases have been brought or heard on the basis of the fault of the parent company under French or Dutch law. Moreover, courts tend to require gross negligence on the part of the parent company to find it liable for damage arising from its subsidiary’s activities. This requirement, coupled with the lack of access to relevant evidence to demonstrate the involvement of the parent company in the damage, makes it difficult for plaintiffs to hold the parent company liable on that basis.

Finally, a parent company may, in theory, be held liable for a criminal offence committed in the context of group activities if it has committed a criminal offence either as a primary or secondary perpetrator. It may also be possible to engage the criminal liability of the parent company where it has acted as an accomplice. However, the procedural and substantive conditions to engage criminal liability limit the potential opportunities provided by criminal law.

In a limited number of cases, French and Dutch courts have taken into account the voluntary CSR commitments of the parent company to assess its misconduct in liability claims. Given the current predominance of soft norms and CSR policies to regulate business activities, this approach could be an important step in ensuring corporate accountability. However, this practice remains too rare to conclude that voluntary commitments can have legally enforceable consequences.

Ultimately, the current French and Dutch legal framework on the liability of the parent company for its subsidiary’s obligations is fragmented, and it remains uncertain as to whether victims can hold the parent company accountable. A legislative intervention is necessary to establish a specific regime governing the liability of parent companies for human rights abuses and environmental pollution in the context of group activities.

The next chapter discusses the possibility of holding MNEs liable for human rights abuse and environmental pollution through the emergence of mandatory HRDD legislation across Europe.