by Grigoris Bacharis, Lecturer in Private Law, Edinburgh Law School
I. Introduction
In my previous post, I explored the emergence of enforcement or regulatory delicts and, in particular, how the doctrine of the undertaking transitioned from public enforcement to competition damages claims. I argued that this development represents a striking departure from private law’s commitment to the concept of legal personality. By attributing liability not to a legal person but to an “economic unit,” the doctrine unsettles, among other things, the principle that responsibility must track wrongdoing and personhood.
As a tentative explanation for why this extension only applies in competition law, I proposed that the law tolerates such a radical departure from the principle of legal personality because there are certain delicts, such as competition damages, that have a hybrid character, part private, part public.
But should the law of delict and tort accommodate this shift? I argue that while the undertaking doctrine might be defensible on enforcement grounds, its coherence with private law’s normative architecture is deeply contested.
II. The ‘Undertaking’ Principle Helps Enforcement
The rationale for departing from ordinary principles of corporate group attribution lies in the enforcement aims of competition law. Claimants act as private enforcers of public norms, and the undertaking doctrine effectively advances these aims. Traditional private law tools have in fact proven inadequate to enforce public norms.[1] Veil-piercing is confined to cases of fraud or evasion of obligations and is therefore far too narrow. Negligence liability is a possibility but is fact-intensive, requiring claimants to prove operational control, which is complicated in decentralised groups. Joint tortfeasance, as seen in Fish & Fish, has potential but remains a doctrinally uncertain concept.[2]
Against this backdrop, the undertaking doctrine offers clear advantages, supported by economic analysis.[3] Liability rules should ensure that firms internalise the costs of their activities. By making the group liable, EU law forces parent companies to account for the risks created by their subsidiaries, reducing incentives to externalise harm. As Koenig has emphasised, being able to extend liability across the group expands the compensation pool by allowing claimants to target entities with assets, particularly parent companies.[4] It thereby avoids the ‘judgment-proof defendant’ problem, where victims would otherwise be left suing subsidiaries with no resources. Group liability also prevents firms from escaping liability through restructuring,[5] and it discourages regulatory partitioning, whereby corporate law separates risks and assets. It further lowers enforcement costs by shifting monitoring obligations from regulators to parent companies, which often have superior knowledge of risks. The move also has jurisdictional advantages, as claims can be brought against the parent company. EU competition law even allows for downward liability, enabling claimants to sue local subsidiaries rather than foreign parents, reducing translation and procedural costs and increasing familiarity with local courts. Finally, in a context of systematic underdeterrence, strict liability appears more defensible especially since many infringements are intentional cartels rather than borderline conduct.[6]
Objections remain, however. Strict liability may risk over-deterrence, discouraging legitimate business practices.[7] Parents might have little incentive to monitor subsidiaries if they are liable regardless of their efforts. Yet there is little indication that Europe’s undertaking doctrine has deterred investment. By contrast, there is strong evidence of under-deterrence in cartel cases: competition harm remains uncompensated, and fines alone are insufficient.[8] In this context, stronger liability appears justified.
III. Corrective Justice and (Redefining) Legal Personality
However, enforcement success cannot, by itself, justify liability in delict. Delict law embodies distinctive interpersonal norms of responsibility, treating defendants as accountable agents for the harms they cause. Stevens’ core thesis in Torts and Rights is that tort law responds to infringements of pre-existing private rights, not to policy goals.[9] At the heart of most theories of delictual liability lies the attribution of wrongdoing to an accountable agent (which is what I define here as ‘corrective justice’). A wrongdoer violates another’s right; Weinrib claims that liability restores the normative equilibrium.[10] This bilateral structure presupposes that the defendant is a person capable of acting and bearing responsibility.[11]
The undertaking doctrine, therefore, marks a suspension of private law’s core grammar. Legal persons are accepted as bearers of responsibility precisely because they are treated as analogues of human agency.[12] Groups, however, may not be autonomous in this sense and lack legal personality. To impose liability on them appears to abandon the link between wrongdoing and responsibility.
Whether this abandonment is legitimate depends on how much weight one places on corporate personhood. For some, it is not a dispensable fiction but a ‘degree of freedom’ through which law organises agency and responsibility.[13] Corporations are group agents who can deliberate, adopt safety policies, and act on reasons[14]; their liability thus expresses juridical blameworthiness, not mere cost internalisation. Legal personhood functions as a ‘visibility cloak’ stabilising otherwise fluid entities and enabling the law to track their obligations.[15] Even so, as Miller and Gold have noted, corporations are already tenuous agents for the purposes of private law, being collectives of shareholders and employees.[16] Extending liability to broader groups risks stretching the fiction beyond recognition. Nonetheless, there are opposing considerations.
First, social and intuitive judgments rarely attach liability with fine attention to the registry of separate legal persons. When the public condemns ‘BP’ for an oil spill or ‘Volkswagen’ for emissions fraud, they attribute wrongdoing to the corporate group as a whole.[17] This suggests that, at the level of moral practice, responsibility is often already assigned to groups.
More fundamentally, the undertaking doctrine can be seen not as abandoning personhood, but as redefining it. Much as vicarious liability treats the employer as ‘acting through the employee,’ the EU treats the corporate group as a functional legal person. This construction extends legal personality to match economic reality. On this view, the undertaking doctrine may be read not as a denial of personhood but as its realignment: liability follows the level at which organisational autonomy materialises. If legal responsibility tracks the capacity to generate autonomous effects, then the relevant bearer of liability is not the formal legal person but the functional unit, the undertaking, which acts autonomously of its participants. By treating the group as one economic unit, the CJEU addresses what Hardman calls the ‘misalignment of legal personality.’[18]
Group responsibility can also be defended on grounds of outcome responsibility. Even through a corrective justice lens, liability need not always be fault-based; strict liability may serve corrective justice by ensuring victims are compensated by those who created the risk. Strict liability, therefore, may still serve corrective-justice aims by ensuring that the costs of harmful outcomes fall on those who made the activity possible. This also brings to mind enterprise liability. The group that collectively organises a risky enterprise assumes responsibility for the outcomes of that enterprise, much as an individual assumes responsibility for the consequences of their own actions. In competition law, subsidiaries and parents alike derive benefits from being part of a corporate group, and the flipside is that they may bear responsibility for the group’s wrongs.[19]
A further justification comes from the nature of corporate groups themselves. As Wells notes, they are motivated by profit in a way that differentiates them from human agents.[20] If human autonomy merits special protection, corporate groups merit less.
Finally, corrective justice also has a practical, access-to-justice-related dimension. In competition cases, victims often face insurmountable obstacles in proving claims against well-resourced defendants, in achieving, ‘substantial justice’. Extending liability to groups is the only way to realise any kind of corrective justice in practice.[21]
IV. Collective Attribution and Corrective Justice
The undertaking doctrine’s broader significance emerges when seen against established and evolving forms of collective responsibility in private law. Collective forms of attribution that sit somewhat uneasily with corrective justice are not unprecedented. Vicarious liability already imposes responsibility on employers for the wrongs of employees. Policy justifications, such as deterrence, enterprise risk, and deep pockets, have long sustained vicarious liability despite its tension with corrective justice. Still, vicarious liability differs from the undertaking doctrine, though both involve attributing liability beyond the direct wrongdoer.[22] The challenge is that in competition law, attribution is quasi-automatic. Nevertheless, the undertaking doctrine can be seen as simply completing this trajectory, eliminating the need for personhood or a fiction of control as the basis of responsibility.
Apart from vicarious liability, a similar trend towards wider notions of corporate attribution can also be observed elsewhere. As the competition-law notion of the ‘undertaking’ attributes liability across a group based on economic unity rather than legal form, the UK’s reform of the identification principle in the Economic Crime and Corporate Transparency Act 2023 extends corporate culpability beyond the ‘directing mind’ to senior managers performing key control functions. Both reflect the same instrumental trend: a turn from formal legal personhood to functional accountability.[23] Comparative law also offers useful analogies. US legislation, such as CERCLA and the Oil Pollution Act of 1990, imposes liability on groups. German law analogises certain corporate groups to partnerships, recognising a sui generis capacity that allows them to bear obligations as functional units.
Yet even on this functional reading, difficulties remain. First, there is a problem of doctrinal confusion. Mixing public and private rationales risks muddling causation, standing, and attribution doctrines. Second, there is a problem of collective guilt. Extending liability to an ‘undertaking’ risks collapsing individual and collective responsibility, raising concerns about both collective guilt and doctrinal coherence. Rights of defence under the EU Fundamental Rights Charter and the ECHR presuppose personal culpability, sitting uneasily with quasi-criminal attributions to groups.
Still, even these objections can be partly contained. The Court of Justice has required a nexus between the entity sued and the infringement, as in Sumal, and the presumption of control in Akzo, grounds liability in decisive influence rather than mere association. Although difficult to rebut, this presumption rests on a structural relationship, and the ECJ has found it to be compatible with fair trial rights. The danger of doctrinal confusion can also be reduced by treating such cases as ‘enforcement delicts,’ exceptional instances where private law momentarily accommodates public aims without abandoning its person-centred conception of responsibility altogether. The upshot is that although the undertaking doctrine conflicts with corrective justice as understood in mainstream delict law, it nonetheless reflects a broader legal evolution: a shift from formal personhood to functional accountability within regulatory contexts.
V. Broader Implications: Beyond Competition Law
The most pressing question is whether the undertaking doctrine will remain confined to competition law or spread further. There are signs of expansion.[24] The Corporate Sustainability Due Diligence Directive anticipates holding parent companies liable across their entire value chains. Environmental and human rights litigation is increasingly targeting corporate groups, especially in transnational contexts (and also in Scotland). Many civilian jurisdictions are disregarding the corporate veil in the supply chain context.[25] Data protection, digital markets, and AI regulation may generate similar pressures. The logic is consistent: where enforcement gaps exist, courts and legislators may find it attractive to impose group liability.[26]
Yet the undertaking’s specific logic does not easily travel. Competition law in Europe is exceptional as it has a parallel public law concept of the ‘undertaking’ and a history of weak private enforcement. Without these structural features, it is more difficult to justify extending group liability. Comparative experience also points to limits. The United States fills enforcement gaps in other ways, through treble damages, class actions, and criminal prosecutions. This suggests that group liability is not inevitable, but one regulatory choice among several. In many other regulatory fields, liability doctrines remain tied to delict law’s doctrines of autonomy, bilateral justice, and fault. Regulators might also be more active, resulting in a less acute enforcement gap. Additionally, courts outside the realm of competition law may be less willing to accept what resembles collective guilt, particularly where private and public enforcement are closely linked. The broader lesson is that the undertaking doctrine might make sense only where enforcement pressures are so acute that private law must be instrumentalised to achieve compliance.
As Micheler warns, indiscriminate veil-peeking undermines predictability and weakens the coherence of company law.[27] It is preferable to examine the various regulatory delictual actions one by one to identify when and how private law’s structural commitments are overridden in the pursuit of regulatory aims. Private law can either accommodate enforcement aims or adapt its doctrines to preserve personhood while reflecting economic reality. Yet any such changes are likely to remain sectoral, confined to areas closest to regulation, rather than transforming delict law as a whole.
[1] CA Witting, ‘Rethinking Tort Law in the Corporate Group Situation’ in Zvonimir Slakoper and Ivan Tot (eds), Digital Technologies and the Law of Obligations (Routledge 2021) 54.
[2] Witting (n 1) 62-3.
[3] See, C Koenig, ‘An Economic Analysis of the Single Economic Entity Doctrine in EU Competition Law’ (2017) 13 Journal of Competition Law & Economics 281.
[4] Ibid 299 ff.
[5] As happened in the infamous German ‘sausage cartel’ case.
[6] See, e.g., P Akey and I Appel, ‘The Limits of Limited Liability: Evidence from Industrial Pollution’ (2024) Journal of Finance 76(1) 5.
[7] On merits and demerits of strict liability see Steven Shavell, Foundations of Economic Analysis of Law (HUP 2004) 182–89.
[8] JM Connor and RH Lande, ‘The Prevalence and Injuriousness of Cartels Worldwide’ in Peter Whelan (ed), Research Handbook on Cartels (Edward Elgar 2023) 22.
[9] R Stevens, Torts and Rights (OUP 2007).
[10] EJ Weinrib, ‘Correlativity, Personality, and the Emerging Consensus on Corrective Justice’ (2001) 2 Theoretical Inquiries in Law 107.
[11] Ibid.
[12] B Ewing, ‘The Structure of Tort Law, Revisited: The Problem of Corporate Responsibility’ (2015) 8(1–2) Journal of Tort Law 1.
[13] A Raz, ‘Taking Personhood Seriously’ (2024) 2023(2) Columbia Business Law Review 729; Cf R Leow, Corporate Attribution in Private Law (Hart 2022) esp ch 1–2.
[14] Ewing (n 13) 27; cf PA French, ‘The Corporation as a Moral Person’ (1979) 16(3) American Philosophical Quarterly 207.
[15] VAJ Kurki, A Theory of Legal Personhood (OUP 2019) 167.
[16] PB Miller and AS Gold, ‘The Corporation as a Category in Private Law’ in Hanoch Dagan and Benjamin C Zipursky (eds), Research Handbook on Private Law Theories (Elgar 2020) 190.
[17] See eg, N Zhou and H Wang, ‘Foreign Subsidiary CSR as a Buffer Against Parent Firm Reputation Risk’ (2020) 51 Journal of International Business Studies 1256, 1275.
[18] J Hardman, ‘Fixing the Misalignment of the Concession of Corporate Legal Personality’ (2023) 43 Legal Studies 443.
[19] MJ Dearborn, ‘Enterprise Liability: Reviewing and Revitalizing Liability for Corporate Groups’ (2009) 97 California Law Review 195, 198.
[20] CP Wells, ‘Corrective Justice and Corporate Tort Liability’ (1995) 69 Southern California Law Review 1769.
[21] However, this has also been described as a form of regulatory overreach.
[22] C Koenig, ‘The Rise of Corporate Cross-Entity Liability: Which Doctrine for What Purpose?’ (2024) 35(3–4) European Business Law Review 343.
[23] J Horder, ‘Corporate criminal liability under the Economic Crime and Corporate Transparency Act 2023’ (2025) 45 Legal Studies 133.
[24] See generally Koenig (n 23).
[25] See the French loi de vigilance, the German Supply Chain Due Diligence Act (Lieferkettensorgfaltspflichtengesetz, LkSG)..
[26] Indeed, where the public-law prerogative is strong, statutory legislation has dispensed with legal personality; see, the Trading with the Enemy Act 1939; Russia (Sanctions) (EU Exit) Regulations 2019/855, reg 6; Modern Slavery Act 2015, s 54; and Building Safety Act 2022, s 130.
[27] E Micheler, ‘Separate legal personality – an explanation and a defence’ (2024) 24 (1) Journal of Corporate Law Studies 301, 328.
