In summary:

This morning, the European Commission published its long-awaited proposal for a new Product Liability Directive, to govern consumers’ rights to recover “no fault” compensation where they suffer harm as a result of a defective product (link). The new Directive seeks to update the existing European strict liability regime, which dates from 1985, to bring new products and defendants into scope, and to make it easier for claimants to recover damages by effectively reversing the burden of proof in many cases. With the formal legislative process now about to begin, the new regime could be fully in force across Europe in 2024-25.

Our initial reaction is that the new proposal is far more claimant-friendly than many had expected. For example:

  • It expands the types of products and defendants within scope – software would be treated as a product, and software developers, authorised representatives of ex-EU businesses, and in some cases online marketplaces and fulfilment service providers, could now be sued.
  • Where a product (including software) does not comply with product safety regulations, defect would be presumed. Defect and causation will also be presumed in other cases, including those where claimants would “face excessive difficulties” because of scientific or technical complexity – the practical effect may be to reverse the burden of proof in many cases.
  • The current 10-year “longstop” period for bringing a product liability claim is to be weakened, especially in cases involving latent injury.
  • Psychological injury and loss of data would be added to the list of recoverable damage.

In greater detail:

For over 35 years, the EU’s Product Liability Directive has allowed consumers who suffer injury as a result of a defect in a product to recover compensation from the product’s producer, without having to prove negligence or fault. The consumer has only been required to show that the product in question did not meet the level of safety which the public is objectively entitled to expect (i.e. was “defective”) and that this caused them loss.

The Directive has proven remarkably adaptable to developments in science and technology, because of its product-agnostic approach and its flexible definition of “defect”. It has been applied to many types of products, from vaccines and medical devices to bicycles and cars. The Commission’s own evaluation of the existing regime, published in 2018, concluded that the regime was broadly fit for purpose.

We now have a proposal for reform. It is surprising that, rather than making minor amendments to the existing regime to bring it up to date, the Commission is proposing repealing the existing legislation and replacing it with an entirely new Directive. The stated purpose is to adapt the regime to new and emerging technologies such as AI, and to address new circulate economy business models and new global supply chains. The practical effect is likely to be to make it easier for EU claimants to bring successful product liability claims.

Despite the Commission’s view that the proposal “strikes a careful balance between the interests of industry and consumers”, at first glance, the proposals are far more pro-consumer than many had expected.

  • Representative claims permitted: The draft Directive requires Member States to ensure that “a person acting on behalf of one or more injured persons” can bring strict product liability claims. This ties in with the fact that the separate Representative Actions Directive will facilitate mass consumer claims (in some cases, on an opt-out basis) in product liability and other matters (further information here).
  • Software – and software developers – brought within scope: The draft Directive amends the definition of “product” to include software, whether or not integrated into a tangible product, and digital manufacturing files (as used for e.g. 3D-printing). The text expressly states that "The developer or producer of software, including AI system providers..., should be treated as a manufacturer", who can therefore be sued under the Directive.
  • Authorised representatives, “fulfilment service providers” and operators which modify products post-sale added to list of potential defendants: The current regime allows no fault liability claims to be brought against producers of defective products, including EU-based manufacturers and businesses which import products into the EU. The new Directive would add authorised representatives (ARs) of businesses based outside the EU to the list, and where there is no EU-importer or AR available to sue, then claims could be brought against “fulfilment service providers” which provide warehousing, packaging, addressing and/or dispatch services. Third parties which make substantial modifications to a product at any time after it has been placed on the market would also now be in-scope.
  • Secondary liability for retailers and online platforms: Where a product’s manufacturer or another EU-based economic operator can’t be identified, then in some circumstances the new Directive would allow retailers and other distributors (as now) and online platforms that allow consumers to conclude distance contracts with traders (new) to be sued.
  • Factors to consider when assessing defect: In order for a claimant to recover damages, they need to show (among other things) that the relevant product is defective. The proposal adds new criteria to the list of relevant circumstances to which courts must have regard when assessing this, including:
    • regulatory compliance: whether the product complies with relevant product safety requirements (including safety-relevant cybersecurity requirements), and whether regulators or other economic operators have intervened on safety grounds. This is likely to increase producers’ exposure in cases where e.g. there has been a product recall or an adverse regulatory decision (although this will not lead to a formal presumption of defect unless the claimant can prove that the product was non-compliant with safety regulations - see below). Equally, however, an ability to demonstrate compliance with relevant product safety regulation may make it less likely that a product is found to be defective.
    • user expectations: i.e., "the specific expectations of the end-users for whom the product is intended". This could introduce unwelcome subjectivity into the test for defect, which has always focused on the public's objective, entitled expectations of how safe a product should be. It is difficult to reconcile this with statements in the recitals to the draft Directive that "[t]he assessment of defectiveness should involve an objective analysis and not refer to the safety that any particular person is entitled to expect".
  • New obligations to produce documents and other information: As with the proposed AI Liability Directive, claimants who are able to show a prima facie case will be entitled to the disclosure of “relevant evidence” by a defendant. A court can only make an order to produce evidence that is “necessary and proportionate” to support the claim, and adequate protections must be put in place for trade secrets. Nevertheless, the introduction of document disclosure obligations of this sort will be foreign to many European legal systems. It is also unfortunate that the proposal does not create reciprocal rights for defendants (e.g. a right to request production of relevant medical records in a life sciences product liability claim).
  • New rebuttable presumptions of defect and causation: This is likely to be the most significant, and least welcome, development for industry. Although the text maintains that the claimant must still be required to prove the defectiveness of the product, the damage suffered and the causal link between the two, it also creates three important presumptions:
    • Where the defendant fails to comply with an order to produce information, or the claimant establishes that the product is non-compliant with safety regulations or that the damage was caused by an obvious product malfunction during normal use or under ordinary circumstances, the product is presumed to be defective.
    • Where defect has been established (including on the basis of the above presumption), and the damage caused is of a kind typically consistent with the defect in question, causation is also presumed.
    • Where a court believes that a claimant “faces excessive difficulties” in proving defect and/or causation, because of scientific or technical complexity, then these must be presumed to have been established, if the claimant can overcome certain hurdles.
    • These presumptions can be rebutted by the defendant, but – especially in cases of scientific uncertainty, i.e. which could involve “proving a negative” – the practical effect may well be to shift the burden of proof to industry.
  • Changes to defences available to producers: There had been some concern that the Commission would remove the “development risks” or “state of the art” defence, which producers can rely on where the state of scientific and technical knowledge at the time that the product was placed on the market did not allow a defect to be discovered. The wording has been slightly amended, but it remains. However, the language of another statutory defence has been tightened up, to make clear that a producer which e.g. updates the software of a product once it is already in the field cannot escape liability based on the fact that the product was not defective at the time of marketing.
  • 10-year longstop period weakened: There had also been concern that the 10-year longstop period, which extinguishes a producer’s liability 10 years after a product has been put into circulation, would be removed. This also remains, but the impact of the provision may be blunted, because the point where time starts to run will change in some cases (e.g. the trigger will be the date of the substantial modification of a product, where this occurs after initial marketing).
  • Longer limitation period for latent disease claims: Further, although the drafting of the relevant provision could be improved, it appears that the consumer's right to sue will be extinguished after 15, rather than 10, years in cases that involve latent personal injury or disease.
  • New types of damages recoverable: Claimants will be able to recover damages not only for death, physical injury or property damage (as under the existing regime), but also for “medically recognised harm for psychological health” and “loss or corruption of data that is not used exclusively for professional purposes.”
  • Minimum claims threshold removed: It will be possible to bring lower value claims under the regime, with the removal of the EUR 500 minimum threshold for damages.
  • Joint and several liability: It remains the case that where multiple economic operators are responsible for the same damage, they can be held jointly and severally liable.
  • Interrelationship with the new AI Liability Directive: The Commission also published a proposal for a new AI Liability Directive today, which likewise features presumptions as to liability and other claimant-friendly measures. AI-enabled products would fall within the scope both of the new AI-specific regime and the revised Product Liability Directive (with the former governing non-contractual civil liability generally, and the latter “no fault” liability in cases involving death, personal injury or property damage caused by an AI system). Applying these different regimes in practice may be challenging both for courts and litigators.

We will examine the proposals in greater detail in the coming days, focusing in particular on the implications for the life sciences and tech industries. For further background on the proposals, you can see our earlier posts here, here, here and here.

The new Directive seeks to bring new products and defendants into scope, and to make it easier for claimants to recover damages by effectively reversing the burden of proof in many cases.