ENSURING PROTECTION

PRODUCT LIABILITY

A long journey

Implications of the revised Product Liability Directive for insurers

With an agreement having been reached to revise the 1985 Product Liability Directive (PLD) – the EU's liability regime for consumers to recover compensation for defective products – a long journey now seems to be coming to an end. For insurers, however, this is the start of a new journey, notably because they will need to review and revise their liability policies and coverage they will offer.

With this revision, the EU institutions aimed to modernise European liability rules by making them fit for the digital world. Under the revised PLD, digital products such as software and artificial intelligence (AI) systems are covered, and the rules have been adapted to tackle cyber vulnerabilities and defective updates and upgrades.

Even if the focus has shifted several times during the discussions on the revision of the Directive, the reform addresses three core issues. Firstly, with a focus on digitalisation, the EU legislator wanted to adapt the liability rules to digital products such as AI systems and software. Indeed, due to the complexity of such products and their fast development, there was a perception that the rules set up in 1985 were not appropriate anymore and did not ensure legal certainty neither for consumers, nor for manufacturers. Secondly, by focusing on remanufacturing, the EU legislator wanted to promote circularity by setting up a clear legal framework for refurbished products. Thirdly, focusing on global supply chains enabled the EU legislator to ensure that the claimant always has an opposing party, even if the actual manufacturer is difficult or impossible to find. Online platforms were therefore included in the revision as economic operators that can be held liable for defective products if no other economic operator can be found.

Key elements of the revision

Other than that, the compromise reached between the European Commission, the European Parliament and the Council of the European Union has upheld the causal link between product defect and damage, and did not drift into purely strict liability, where the defendant would be found liable by merely placing the product on the market. This is particularly important for manufacturers as it is still for the consumer to bring evidence to the courts that the product was defective and that it was this defect that caused the damage suffered by the consumer. If the compromise had drifted into purely strict liability, it would have been for the manufacturer to prove that either the product was not defective, or that the defect did not cause the damage suffered by the consumer.

Nevertheless, the reform still seeks to simplify the burden of proof, which remains with the injured party, for errors or the causation of damage in complex cases (although it is not clear what this means concretely: an implant, faulty mobile phone programming?) for the claimant. The claimant also receives assistance through a right to information (disclosure of evidence). The insurance industry has long highlighted that a reversal of the burden of proof would be challenging, because under the 1985 Directive, 60%¹ of cases are ruled in favour of the claimant. Altering the burden of proof means getting rid of the balance between the protection of consumers and the legitimate interests of producers which constitutes a precondition for producers’ strict liability.

Also, all material losses resulting from the damage are covered by the revised Directive. An extension to non-material losses – medically recognised damage to psychological health and loss or corruption of data – is also possible in principle, insofar as these are eligible for compensation under national law which increases the risk of forum shopping.

Damage to the product itself remains outside the scope of the Directive. This damage is primarily to be resolved via warranty law or the tort law of the respective member state.

“It is still for the consumer to bring evidence to the courts that the product was defective and that it was this defect that caused the damage suffered by the consumer.”

Minimum and maximum thresholds of ECU 500 for property damage and ECU 70 million for personal injury respectively have also been removed from the revision. Personal injury claims will not always expire after ten years anymore; in certain cases of latent personal injury, they will only expire after 25 years. The limitation period of three years from the date the injured person became aware or should reasonably have become aware of the damage, has been retained.

Following the publication of the new Directive in 2024, member states will have two years to transpose it into their national law. Only after these two years have elapsed, will the 1985 Directive be repealed. This makes it clear that the new Directive applies to products that were placed on the market after the old Directive was repealed. A long road will then come to a temporary end – temporary because a transitional period will start during which the two Directives will operate in parallel, as long as claims under the old Directive have not expired. Moreover, the European Commission will have to evaluate the application of the new Directive 6 years after its entry into force. This could pave the way for further adjustments to liability rules, especially considering the rapidly evolving digital sector.

The insurance industry perspective

Insurers will now have to consider what impact the new Directive will have on their insurance products, and how to adapt to this new reality. A possible consequence may be the creation of new or specialised lines, or a transfer of products, such as IT liability insurance being transferred to product liability insurance. Also, the definitions of ‘’bodily injury’’ and ‘’property damage’’ must be reviewed from a coverage point of view, as must the calculation models for liability insurance, as the extended liability must be priced in.

It seems evident that given the facilitation of evidence in complex cases, and the abolition of the deductible for property damage, issues are likely to be brought to the table that claimants would have avoided in the past due to a lack of prospects of success. When combined with the possibilities of a class action under EU law (Directive (EU) 2020/1828 on representative actions for the protection of the collective interests of consumers), this could mean a whole range of new claims scenarios appearing.

It will also be necessary to consider how the respective trigger of the liability policy fits in with the new system: is the policy already triggered if a potential claimant first has to find out, by means of a request for information, whether he has a claim at all and can present all the prerequisites required by law? An insurer will have a great interest in providing its policyholder with close support at this stage; however, it is then necessary that the policy is triggered either very early on or that insurance cover is provided for expenses incurred before the trigger occurs (e.g. assumption of the costs incurred to check what information must be given to the claimant and in what form, or assumption of the legal defence costs that the claimant is not entitled to information at all).

“There will always be cases where an insurer does not want to insure certain products due to risk considerations.”

Challenges to expanding liability cover

However, the EU legislator should not be so naive as to believe that every manufacturer can get all its products insured. Even at the time of the 1985 Directive, this was not the case, and the insurance protection gap will probably increase, given the increased risk of mass litigation due to the ease of the burden of proof and the facilitation of class actions at European level (Directive (EU) 2020/1828 on representative actions for the protection of the collective interests of consumers). In addition, there will always be cases where an insurer does not want to insure certain products due to risk considerations (e.g. exclusion of certain active pharmaceutical ingredients or certain implants); the discussion about the per-and polyfluoroalkyl substances (PFAS) is also a good example here, with some groups of substances already banned and others not (yet).

It is welcome that the legislators have not pursued the idea of compulsory product liability insurance any further. Indeed, compulsory insurance is only appropriate in a few areas, with specific characteristics, such as the motor and medical sectors. Art. 10 para. 16 of Regulation (EU) 2017/745 on medical devices may serve as an example here: “Manufacturers shall, in a manner that is proportionate to the risk class, type of device and the size of the enterprise, have measures in place to provide sufficient financial coverage in respect of their potential liability under Directive 85/374/EEC, without prejudice to more protective measures under national law.’’ However, for the majority of sectors, the right conditions for compulsory insurance to work (sufficient data, sufficient similarity in the risks being covered, sufficient insurance capacity, adequate competition and enough reinsurance capacity) are not met. When imposed without the right conditions being met, compulsory insurance may result in less loss prevention, which in turn can result in unprotected businesses or individuals because they cannot obtain the cover that they need as insurers are unwilling to offer cover.

The fact that the issue of (compulsory) insurance is not completely off the table with the EU legislator is shown by the new Article 16, which obliges the European Commission to draw up a report six years after entry into force of the revised Directive, and which, among other things, must deal with the extent to which product liability insurance cover can be obtained.

The revised Directive also foresees the set-up of a database of product liability cases and an indication of how many actually end up in court. The European Commission probably believes that cases brought to court will demonstrate how well the revised directive works. The more cases in the database, the more effective the directive? That would certainly not be correct, because it is precisely the aim of a product liability insurer not to take cases to court in the first place, but to settle them in favour of the policyholder beforehand.

1. “What do we know and what don’t we know about the consequences of the revised Product Liability Directive?”, ECIPE, November 2023