The EU Inc Regulation (COM(2026) 321 final) codifies a Business Judgment Rule (BJR) in Article 44. This post argues that the safe harbour is formally defective. The standard of conduct and the standard of review for duty of care violations are textually identical, producing a protection gap of zero. For a corporate form designed to attract entrepreneurship, this is self-defeating.
In a forceful critique posted on this blog, Enriques, Nigro, and Tröger have argued that the EU Inc Proposal falls short of its ambition in important respects, including scope bloat, the preservation of national law patchwork, and the distance between the Proposal's digitalisation agenda and the substantive legal fragmentation it leaves intact. Much of their analysis is compelling, and I share it. One provision, however, merits separate attention, because it goes to the heart of whether the EU Inc will attract the entrepreneurial activity it was designed to serve, that is, the business judgment rule in Article 44. My analysis complements theirs.
The Protection Gap and Why It Matters
A corporate law system that seeks to attract directors willing to take efficient risk must distinguish between what the law requires of directors (the standard of conduct) and what the law will sanction when things go wrong (the standard of review). The distance between these two thresholds is the 'protection gap’. This is the economic logic of the BJR. It rests on the recognition that directors are risk-averse and that courts, reviewing decisions with knowledge of the outcome, cannot reliably distinguish bad luck from bad judgment.
Consider a director evaluating a project whose outcomes are inherently uncertain, such as a pivot into a new market or the pricing of an early-stage financing round. If the review threshold equals the conduct threshold, the director's expected liability increases with the project's variance, regardless of its expected value. This holds even where indemnification or Directors and Officers (D&O) insurance is available, because a judicial finding of breach depletes the director's reputational capital, impairing future board appointments in ways no policy can offset. The rational response is to avoid risk, which is a subtle agency cost (ie, underinvestment) and the opposite of what the EU Inc is designed to encourage.
Delaware corporate law has long understood this. Its BJR operates as both a substantive shield and a standard of review: it creates a rebuttable but strong presumption that directors acted on an informed basis, in good faith, and in the company's best interests, enabling early dismissal of claims through a motion to dismiss without reaching the merits for ordinary care breaches. Beyond the presumption, the Delaware General Corporation Law (DGCL) § 102(b)(7) permits charter-based elimination of monetary liability for care breaches, pushing the effective threshold to the floor of loyalty and bad faith. Germany achieved a functionally comparable result through § 93(1) sentence 2 AktG, codified in 2005 following the BGH's ARAG/Garmenbeck decision. The German rule protects directors who could reasonably believe they were acting on the basis of adequate information; courts have read the adequacy requirement as confining liability to cases of irresponsible conduct.
The Collapsed Separation in Article 44
Article 44(1)(c) imposes a duty of 'reasonable care, skill and diligence’. Article 44(3)(a) provides a safe harbour for directors who 'exercised the care that a reasonably prudent person would use’. Despite differences in drafting source (the first echoes the UK Companies Act 2006, section 174; the second draws on the tradition of the ordinarily prudent person in torts), these formulations designate the same substantive threshold: the care of a director of ordinary prudence. The safe harbour restates the duty.
This is particularly damaging because entrepreneurial innovation requires a tolerance for risk incompatible with the standard of ordinary prudence. The founder who launches a product into an untested market is making a forward-looking bet under radical uncertainty. A liability regime that measures her conduct against that standard penalises the very quality the Regulation purports to incentivise: the willingness to accept variance in pursuit of expected value. In the absence of a functioning liability shield, every directorial decision, however made in good faith, is exposed to substantive judicial review on the merits. Courts are invited to evaluate the substance of commercial decisions rather than confining themselves to the adequacy of the decisional process.
A Regime at War with Its Own Objectives
The Regulation introduces venture-compatible financing tools because these are what startups and scale-ups need. But the decisions those tools require, from pricing a round to certifying solvency for a distribution, are precisely the forward-looking commercial judgments for which directors need a meaningful liability shield. A regime that facilitates such transactions while denying that protection is internally inconsistent.
Article 44(4), which provides that directors’ liability shall be 'further governed by the applicable national law,' compounds the problem. In jurisdictions without a real national BJR, such as Italy and France, the EU Inc offers no improvement over existing national liability exposure and may even diminish it, since national courts that have been moving toward a degree of deference to business judgment in practice (as Italian courts have) would find their path constrained by the fact that the ordinary prudence standard is now embedded in a directly applicable EU regulation, subject to autonomous interpretation by the Court of Justice of the European Union. Indeed, the interaction with Article 44(3)(a) raises a deeper question: is the Regulation’s review standard mandatory or default? The better reading is that it is mandatory, since Article 44 is directly applicable under Article 288 of the Treaty on the Functioning of the European Union (TFEU). If so, a German court applying Article 44 to an EU Inc director would have to apply the Regulation’s standard, not § 93 AktG’s more protective construction, and the EU Inc would represent a step backward in jurisdictions where national law currently provides a wider gap. Even on the alternative reading, the 'further governed' clause opens the door to 27 nationally differentiated liability regimes, reproducing the fragmentation the Regulation was designed to overcome. On neither reading does Article 44 improve upon existing national law.
An Interpretive Path Forward
Since the Regulation is still a proposal, the most urgent remedy is legislative: amend Article 44 during the ordinary legislative procedure to introduce a functioning safe harbour on the Delaware model, combining a rebuttable presumption of regularity with a qualified fault threshold and an exculpation mechanism for care breaches. The window for intervention is narrow.
If the text remains unchanged, the requirement that every provision of EU law be given useful effect demands an interpretation that gives the safe harbour independent content, but the construction needed would be an audacious one. Recital (34) provides the critical bridge. It introduces the informational adequacy dimension that the operative text omits, directing that directors should 'ensur[e] that they have sufficient information’. Through a functional interpretation informed by the Regulation’s objective of incentivising entrepreneurship, this language supplies material for a construction in which Article 44(3)(a) incorporates a process-based qualified fault standard, confining judicial inquiry to whether the director’s decision-making process was informationally adequate rather than whether the decision itself met the standard of ordinary care.
As noted, this interpretive path is not without difficulties. Reading a recital as supplying an element that the text of article omits is doctrinally uncomfortable. The UK illustrates the difficulty. Under a statute that codifies an ordinary care standard, courts have attempted to develop a degree of de facto deference to business judgment, with mixed results. Section 174 of the Companies Act 2006 codified the pre-existing duty of care without changing it in substance, and section 1157 gives courts discretion to relieve directors who acted honestly and reasonably. Yet rates of director liability for business judgment rose after codification. The codification of the standard of care may have contributed to this increase. But whatever its role, this evidence illustrates a more general problem: a statute that fixes a liability standard without providing a clear safe harbor leaves directors dependent on judicial discretion for whatever protection they can obtain, and this creates uncertainty. There is no reason to expect that 27 national courts, without shared precedent or a common interpretive tradition, would resolve that uncertainty consistently, if at all.
The Proposal now enters the ordinary legislative procedure under Article 114 TFEU. As the European Parliament and the Council examine the text, Article 44 should be one of the priorities for amendment. The co-legislators can give the provision the operative content that the Commission's draft fails to supply.
Simone M Sepe is Professor of Law and Finance and Honourable Frank Iacobucci Chair in Capital Markets Regulation, University of Toronto, Jackman Faculty of Law and Rotman School of Management; Yale Law School Center for Private Law; European Corporate Governance Institute; American College on Governance Council.
Share:
With the support of
