The Fulfilment Service Provider in Product Safety and Liability: A Silver Bullet for the Enhancement of Safety and Compensation in E-Commerce, or a Leap of Faith?

Abstract

The European legislator has included the fulfilment service provider (FSP) in European product safety legislation as a new responsible economic operator (EO) in e-commerce for instances where no other EO is based in the European Union. As a matter of coherency, this adjusted personal scope of product safety law will likely be mirrored in a revised product liability directive. This article explores to what extent the aims and expectations of European legislator with respect to these legal changes (law in the books) align with the perceptions of FSPs and legal advisors in the field (law in action). This is done by means of a legal analysis of the goals and envisioned effects of these changes as well as a semi-structured interview study on how FSPs and legal advisors in the field perceive this new role, responsibility, and liability. The findings provide indications that product safety law currently in force might not (yet) lead to the desired effects. The findings also make us question the extent to which victim compensation is actually improved by the proposed adjustment of the personal scope of the product liability directive.

I. Introduction

E-commerce challenges the way the EU-legislator used to regulate product safety. Web shops and online marketplaces, like Amazon and Ali-express, offer products from outside the EU on their websites that do not always live up to EU product safety standards, therefore putting European consumer safety at risk.¹ These online marketplaces often work as mere intermediaries for sellers that may be situated outside the EU’s territory, which makes taking enforcement action difficult.

The European legislator tried to improve product safety in online markets by making the fulfilment service provider (FSP) a new responsible economic operator (EO) – in addition to the manufacturer, importer, authorized representative and distributor – under the Regulation on Market surveillance and conformity of products from 2019 (hereafter: “MSR”).² The MSR strengthens the member states’ market surveillance and forms a safety net in case sector specific legislation falls short.³ From 16 July 2021 onwards, the FSP established in the Union is the responsible EO under the MSR for product safety with respect to the products it handles, where no other EO is established in the Union.⁴

Subsequently, the MSR-model was used both as a basis for the new General product safety regulation (GPSR), applicable to consumer products as from 13 December 2024,⁵ as well as the new product liability proposal (PLDp).⁶ A final PLD-text along these lines is expected to be adopted shortly.⁷ The recently adopted European Parliament’s resolution in first reading of the PLDp (PLD EP-1) follows this “cascade.” This cascade has been criticised by legal scholars,⁸ because it lets FSP-liability prevail over liability of the online platform (or marketplace under the GSPR). Additionally, the envisaged platform liability respects the hosting exemption under Digital Services Act (DSA). It is questionable whether a corresponding change of the PLD will improve victim compensation in e-commerce sufficiently.⁹

At the same time, the scientific substantiation of the cascade suffers from important empirical-legal knowledge gaps. It is unclear what the effects have been of the already changed MSR. The empirical evidence referred to in the impacts assessment underlying these liability regimes – both MSR and GSPD, as well as the PLDp – is very limited (II).¹⁰ Also, the first evaluation of the personal scope of the MSR was scheduled for 16 July 2023, but has not been published yet.¹¹ Therefore, we do not know what the impact of the first reforms (MSR, GPSR) will be, nor can we predict the effect of the PLD-change. Presently, the European Legislator is basing the PLD-reform on coherency arguments and assumptions about the possible effects of these legal changes, yet we have no idea whether these arguments or assumptions are correct.

This study therefore addresses the following research questions:

What are the legislators aims and expectations of these legal changes and how do these align with the perceptions of FSPs and legal advisors in the field on the FSP’s new role, responsibility, and liability in e-commerce?

To answer the research question this study uses an empirical-legal methodology. First, a legal analysis is used to discuss the (proposed) legislative changes, their empirical foundations and to make an inventory of the goals and the assumptions underlying the introduction of the FSP in the legal framework (II). This way, it becomes possible to investigate whether these goals and assumptions align with the perceptions of FSPs and legal advisors in the field. If these do not align, this could indicate a problem. As Sibony and Bijleveld point out: if assumptions underlying the law do not hold or the law is not applied as planned, it is unlikely that the law would have its desired effects.¹² From this legal analysis we derive the relevant empirical sub-questions that are included in the topic list for the empirical part. Second, this legal analysis is supplemented with sociolegal insights from compliance research, which lead to additional empirical sub-questions (III).

To answer the empirical sub-research questions from paragraph II and III, a semi-structured interview study was deployed amongst several FSPs and legal advisors in the field, to determine how these legal changes have been perceived by FSP and how they see their role (IV). The design and method of analysis of the interview study is explained (IV) and the results are discussed (V).

The aim of this study is to contribute to the further evaluation of the MSR. Moreover, it provides a first insight into the possible effects and effectiveness of the corresponding changes in the GPSR and the expected new PLD, because these changes are based on similar assumptions and rationales (as further illustrated in paragraph II). The European product framework is partially based on traditional law and economics and rational choice theory, which sees product safety and product liability as two sides of the same coin. Despite their different perspectives (preventive versus remedial), both are assumed to have complementary deterrent effects.¹³ Public law in itself may also result in private enforcement: it gives substance to private law open standards (not only defectiveness in product liability, but also general duties of care). In addition, most jurisdictions allow for a claim for damages for violations of statutory provisions.¹⁴ Victims may use the violation of regulations by FSPs to support their liability claim or an injunction, hereby becoming private enforcers, possibly resulting in an optimal mix of liability and regulation.¹⁵ Therefore, the currently existing public FSP-liability could – in theory – already result in tort liability, and could incentivise FSPs to contribute to product safety.

As will be illustrated in paragraph II, the FSP’s expected contribution to product safety in the eye of the EU-legislator does not consist of a safety inspection of each parcel; it regards passing on product safety obligations to their “clients” and actively discussing product safety. The upcoming revision of the PLD could in a similar vein incentivise FSPs to take contractual measures (for example regarding insurance and recourse). From a contractual doctrinal perspective, this is a form of transnational private governance.¹⁶ An important pre-condition for this to take place is that FSPs are actually aware what their legal duties and responsibilities are and that they fully internalize these (III).¹⁷ At the same time, this study looks beyond mere deterrence and rational choice theory: it uses insights from sociolegal compliance research to help identify which findings from the interviews might stand in the way of reaching the legislative goals. Also, the findings may help to determine what other normative justifications for secondary product liability of FSPs exist in practice from a legal-theoretical perspective because it gives a first impression of who FSPs are (II and IV).

II. Legal analysis: content, goals and assumptions

1. MSR

Under the MSR, a FSP is a “natural or legal person offering, in the course of commercial activity, at least two of the following services: warehousing, packaging, addressing and dispatching (.).”¹⁸ The MSR determines that a product may only be placed on the market if there is an EO established in the Union who is responsible for the certain product safety tasks,¹⁹ such as verifying the declaration of conformity of the product and its technical documentation (a), providing these documents to authorities (b), informing authorities if a product presents a risk (c) and cooperation and taking corrective action if needed (d).²⁰ The name, registered trade name or trademark, and contact details of this responsible EO have to be indicated on the product or on its packaging, the parcel or an accompanying document.²¹ The FSP is only responsible in case no other EO is established in the Union.²² These provisions only apply in addition to nineteen sector specific product safety instruments.²³

The original proposal for the MSR 2019 did not mentioned the FSP,²⁴ hence why these changes were not substantiated with evidence in the formal impact assessment.²⁵ The FSP-provisions were included by the Council in 2019,²⁶ because allegedly market surveillance authorities found it difficult to apply the classic definitions of EOs to FSPs in practice.²⁷ They assumed that FSPs could function as a primary point of contact for information for authorities and could provide a form of secondary liability, that could enhance the effective enforcement.²⁸

Traditionally in product safety, EO-obligations and corresponding public law liabilities differed based on their assumed roles in the supply chain and capacities.²⁹ From a legal–theoretical perspective, stricter liability or duties of care are justified if the operator is capable to prevent the risk from occurring by influencing the safety of the product himself or by other means. The importer was assumed to be highly knowledgeable of the market and able to do the latter by consolidating compliance with parties up the chain, contractually passing on legal duties and/or arranging recourse which justified heavy duties and stricter liability. The distributor, who was assumed not to be able to open each individual package and to trade a big(ger) variety of products, used to have more limited duties (compare II.2 and II.3).³⁰ The MSR changed this approach somewhat by introducing a cascade of liable actors while using uniform duties for all EOs. The main justification appears that placing a product on the EU market creates a possible risk. The FSP facilitates placing on the market in the EU when no other EO is available and has financial gain of it through its services. The legislator assumes that FSPs “perform many of the same functions as importers.”³¹ At the same time, we know little about how FSPs function exactly, nor do we have a clear picture on their (contractual) capacities, risk management and position in (relation to) the supply chain, which makes it difficult to have a fully informed normative discussion on which principles should be decisive shaping their public law and private law liabilities (hence SQ 1–4 hereafter).

After the MSR’s adoption, European Commission issued non-binding³² guidelines for EOs and market surveillance authorities on the practical implementation of these provisions.³³ Herein the Commission acknowledges that FSPs do not automatically have a formal connection with the manufacturer enabling them to fulfil these tasks. Therefore, the FSPs need to actively discuss product safety and make arrangements with their “clients” to ensure that they receive – from the “client.” or directly from the manufacturer – the means to fulfil these obligations. This implies that the FSP has to contractually determine in advance who is the responsible EO and arrange the required documentation. To this extent, the guidelines contain a contract checklist for FSPs.³⁴ From a theoretical perspective, this could be seen as a form of transnational private governance (see previously I) or regulation and enforcement through contract, which could enhance safety. Capacity to contractually arrange this could, as stated before, also justify a stricter form of liability from a legal-theoretical point of view. At the same time, FSPs could contractually ensure with their clients that there is another EO in the EU to avoid liability, and leave it at that, which would imply a smaller impact or regulatory effect (see also VI).³⁵ This raises the following empirical-legal sub question:

(SQ1): Do FSPs make contractual arrangements with their counterparts to live up to the product safety obligations in the MSR and what do they look like?

Whether and with whom (contractual) arrangements can be made highly depends on the applied business model (who is the “client”?) and the FSP’s bargaining power. These business models could differ greatly, as is recognized in the guidance document.³⁶ Legal theory distinguishes between the fulfilled by merchant-model (FBM), when the FSP contracts with the seller directly, and the fulfilled by platform-model (FBP), when platform itself provides fulfilment services and a logistic service provider could be the platform’s sub-contractor.³⁷ Under the FBM-model, online platforms that explicitly intermediate in the sale as well as the fulfilment would generally not qualify as FSP, would not fall under the MSR and would enjoy the host exemption under the DSA.³⁸

This raises the following questions:

(SQ 2): A) What type of business models do we see in e-commerce related to FSPs? B) Who are FSPs contracting with? (see also III hereafter) C) Do FSPs consider themselves to be in a position to contractually ensure compliance and do they have enough bargaining power to do so?

More information on the business models is relevant to determine how the MSR obligations apply in practice and to whom, but also to inform the normative debate regarding the personal scope of the MSR, GPSR and PLD: what are the justifications for FSP liability under these various instruments and are the definitions and duties adequate in light of these justifications looking at each actors factual role and capacity? (see II.2 and II 3).

A final remark: this provision has no added value in case a product is directly imported by the consumer meaning that it is bought online and sent to the consumer by postal services, while the consumer is responsible for the custom- and VAT-declaration.³⁹

2. GPSR

On 30 June 2021 the European Commission issued a proposal for a new regulation on general product safety (GPSR) to displace the previous directive.⁴⁰ The GPSR serves as a safety net for consumer products in case no sector specific EU-rules are available or in case they fall short.⁴¹ The GPSR impact assessment report does not explicate whether FSP were part of the consultation.⁴² At the same time, 70 per cent of all stakeholders were in favour of creating an obligation to have a responsible EO in the EU.⁴³ The final text was adopted 10 May 2023⁴⁴ and follows the MSR FSP-definition⁴⁵ as well as its inclusion as responsible EO,⁴⁶ in case no other EO in the EU is available.⁴⁷

Compared to the MSR, the GPSR adds an obligation to the EO to regularly check that the product complies with the technical documentation and the requirements regarding traceability of the manufacturer and instructions of use.⁴⁸ Also, EOs – now including the FSP – are required to use any personal data they may have on consumers, to inform them in case of product safety recalls or warnings.⁴⁹ In addition, a “sales” remedy for consumers was introduced in the GPSR against the EO that performs the recall.⁵⁰ This raises the following questions:

(SQ 3): What non-contractual measures do FSPs take to comply and ensure the conformity of the products with European product safety legislation?

(SQ 4): Do FSPs already handle consumers’ personal data to take corrective action?

Through the interviews we also wanted to get an in-depth picture of what type of measures are FSPs able to take and what might be the reason why they do or do not take them. These questions are not only relevant to see whether FSPs anticipate the taking effect of the GPSR and its practical implications, but also to get a clear image of their precise role and capacity to inform the normative debate around the justifications of the personal scopes and duties under the MSR, GPSR and upcoming PLD. Although further alignment between the GPSD and the PLD has been advocated, it was already subject to debate. To what extent alignment is possible and desirable ultimately depends on the functions and goals of each framework.⁵¹

3. PLDp en PLD EP-1⁵²

Under both PLDp and PLD EP-1, the FSP is secondary liable for damage caused by a defective product if no manufacturer,⁵³ importer or authorized representative⁵⁴ established inside the Union can be identified.⁵⁵ The distributor is liable if he fails to identify who supplied him the product within one month upon receiving such a request of the claimant.⁵⁶ These conditions shall also apply mutatis mutandis to any provider of an online platform that allows consumers to conclude distance contracts with traders and that is not an EO, provided that the conditions of article 6(3) DSA are fulfilled.⁵⁷ This means that if both the FSP and the platform are based in the Union and the platform gives timely notice, the liability of the FSP prevails (under a FBM-business model, II.1).⁵⁸

It is not clear whether FSPs are able to convert the possible future increase in product liability risk in the price of their services to manufacturers/sellers,⁵⁹ or take risk-mitigating contractual measures⁶⁰ hereby possibly creating a form of private enforcement.⁶¹ If this is not the case, it seems undesirable to let strict liability of the FSP prevail over liability of the online marketplace. The latter has the economic gain of the service⁶² and could – in economic terms – be the cheapest cost avoider.⁶³ On behalf of the European Law Institute (hereafter ELI), Wendehorst, Borghetti and Koch therefore suggest to let liability of the online platform prevail over the FSP, but still make reference to article 6(3) DSA. They also suggest reformulating the definition of FSP so that it is not extended to someone providing ad hoc fulfilment services without an ongoing business relationship with the manufacturer.⁶⁴ We do not know whether FSPs are requested ad hoc regarding diverse manufacturers and sellers or not, what the role of the platform is in these transactions and to what extent the platform determines the prices, terms and conditions (hence SQ 2.A).⁶⁵

The previous evaluation of the directive did not explicitly include FSPs, nor did the extensive impact assessment underlying the proposal.⁶⁶ FSPs could not be included in the stakeholder consultation.⁶⁷ The cost-benefit analysis shows that, given the growth of e-commerce and the expected growth of fulfilment services, this could potentially affect a large group of companies.⁶⁸ At the same time, it is estimated that 50 per cent of consumer purchases are made via a platform.⁶⁹ Given the aforementioned, it is remarkable that the effects of possible platform liability are discussed in such a limited manner. The argument for giving priority to liability of the FSP appears to be the coherence with product safety law.⁷⁰ In addition, it is argued that liability of the FSP would not often occur because the appointment of an authorized representative is often mandatory in product safety law,⁷¹ but it is unclear whether it always is present in practice (this has not been further investigated, hence SQ 1). The IA-reporters expect a possible increase in insurance premiums for FSPs, but this cannot be substantiated with European data. They assume an increase prevention, but higher product prices;⁷² higher insurance premiums charged by agents and FSPs will be passed on in the price of their services. Also, contracts with producers in third countries will be adjusted to reflect new liabilities.⁷³ Whether the same effect has already occurred after the introduction of public law liability via the MSR has not been empirically investigated either. This leads to two other sub-questions:

(SQ 5): Have FSPs increased the price of their services and/or noticed an increase in insurance premiums due to an increase in liability?

(SQ 6): How do FSPs perceive the pending product liability proposal and how do they act upon it?

III. Sociolegal insights from compliance research

Sociolegal compliance research has shown that people, also in organisations, do not always behave rationally. For deterrence the perception of the chance of getting caught should be high.⁷⁴ Additionally, other drivers determine to a greater extent whether companies do or do not comply with the law, such as capacity to comply including knowledge of the law, legitimacy (like procedural justice and moral support for the measure), social norms, and – for organisations in particular – reputation and culture (risk for reputational harm is often seen as part of the deterrent factors).⁷⁵ According to Van Rooij & Brownlee a further understanding of how regulation or tort may influence behaviour, starts with recognizing that people develop a subjective view of the law and how it is enforced. Literature shows that the level of legal knowledge and understanding often is very low, both for laypersons and professionals. Moreover, perceptions of liability may not correspond to the actual liability risk.⁷⁶ “If people do not know exactly when they are going to be liable and what for, how can such liability come to shape their behaviour?” Key questions are the legal knowledge of potential tortfeasors, their views on certainty and severity of liability and their perception on their responsibility.⁷⁷ This gives rise to three – somewhat overarching or preliminary – sub questions:

(SQ 7): What is the level of knowledge of FSPs of the current and expected legal framework?

(SQ 8): What is the perception of their current and future liability risk? (relating to deterrence)

(SQ 9): What is the perception of the responsibility of FSPs for product safety? (relating to moral support for the current rules and social norms)

IV. Interview study: design and method of analysis

1. Design, procedure and sample

To answer the main and the sub-question(s) (II and III), we conducted semi-structured in-depth interviews with ten respondents from FSPs (internal staff) and external (legal) advisors. The FSP employees were legal councils or other parties involved in contracting process. We chose interviews because the aim is to capture perceptions and underlying reasons for certain attitudes/behaviours. Also there is a limited amount of previous data available in this context and it regards a very sensitive topic (confidential information regarding non-compliance for example).⁷⁸

Participant	Occupation	Company type	Company size
Participant 1	Lawyer (external)	Law firm	Medium size (50-250 employees)
Participant 2	Legal council79	FSP	Large (250 + employees)
Participant 3	Legal council	FSP	Medium size (50-250)
Participant 4	Business developer	FSP	Large (250+ employees)
Participant 5	Legal Council	FSP	Large (250+ employees)
Participant 6	Director	FSP	Small (0–55)
Participant 7	Sales and contract manager	FSP	Small (0–55)
Participant 8	Director	FSP	Small (0–55)
Participant 9	Lawyer (external)	Law firm	Small (0–55)
Participant 10	Consultant (external)	Consultancy firm	Small (0–55)

Respondents in the bigger multinationals were recruited through the researcher’s own network, snowballing and partially through the Dutch branch organisation Fenex-TLN. We discovered through the first interviews and online research, that also new, smaller logistic service providers entered the market in COVID-times. These FSPs were recruited online. We focussed on FSPs in the Netherlands⁸⁰ The external legal advisors/lawyers were approached because of their expertise and track record in product safety legislation and e-commerce legal matters.⁸¹

In the end, the interviews focussed primarily on the FBM-business model and perspective (II.2), which is absent in discussions on product safety and product liability.⁸² We decided not to recruit respondents with help of the European Commission, Dutch market surveillance authorities and/or the European stakeholder organisation for logistic service provides Clay Cat, because we wanted to minimize the risk of lobbying as far as possible and receive new, bottom-up information that has not been included in policy making yet.⁸³

The semi-structured interviews were conducted in Dutch or English; quotes from the former as presented in the Results section (V) are translated from Dutch to English. The Data Management Plan and research design was approved by the Leiden Law School Ethics Committee.

2. Qualitative analysis

The transcripts were systematically coded with software program Atlas.ti. We first inductively broke up the data from the interviews and created individual codes to label the quotes from the interviews (known as open coding). Although the interviews were semi-structured – using a topic list based on legal assumptions and existing theory – we wanted to avoid excluding possible new theoretical insights and biases as much as possible, since this group has not been studied before. All quotes were assigned one (or multiple) codes, resulting in 127 codes and 271 quotations. Next, we assessed how the codes were connected (axial coding) and what categories/groups the codes belonged to (selective coding).⁸⁴ Most of these categories, corresponded to some extent with existing theory (III).⁸⁵ For instance, we created a category for “Knowledge of the MSR,” which consisted of 16 different codes.⁸⁶ We stopped gathering new data as soon as the interviews did not yield new insights and the open coding process did not result in new codes (also known as having reached saturation).⁸⁷ The coding and the data set were checked by a student assistant.

V. Results

1. Business models and activities⁸⁸

The interviews confirmed that for FSPs e-commerce is commercially more interesting compared to offline retail, because it requires more handling (individual packages versus pallets).⁸⁹ For FSPs the added value lies in big volumes (shipping as many small packages as possible, because they do not require a lot of warehousing)⁹⁰ or additional handling (value-added logistics, VAL).⁹¹ One of the bigger FSPs, also active in medical devices, has ceased functioning as authorised representative for companies due to the lack of insurance cover available and increase in liability risk under the Medical Devices Regulation (MDR).⁹² Respondents from smaller FSPs indicated that they do not perform customs declarations or assistance in that regard; they refer to other companies for that type of services.⁹³

The FSP-respondents primarily operate under the FBM-model, which means that the logistic service contract is concluded between the seller (sometimes brand owner) and the FSP directly, without interference of the platform.⁹⁴ To be clear, also in these cases some orders may still be received by FSPs that were bought by a consumer through a platform.⁹⁵ This differs from FBP-model (see II.2): the seller then takes out fulfilment services directly with the platform and a logistic services contract is concluded between the platform and the logistic service provider.⁹⁶ Multiple respondents declare that the FBP-model is only made available by platforms to sellers that sell bigger volumes, who can be identified by platforms through their own data.⁹⁷ The smaller volumes are not interesting for big FSPs and platforms and left to new and upcoming FSPs in the market (FBM-model),⁹⁸ called “local heroes” by bigger FSPs.⁹⁹

COVID-19 has boosted e-commerce but also provided an opportunity for smaller FSPs to start their business based on the increase in demand.¹⁰⁰ Smaller FSPs saw their own e-commerce business – as sellers – taking off and noticed how bigger companies were leaving this part of the market unattended for (“we can do this better” or cheaper).¹⁰¹ Two respondents consider it to be their role to provide services to start-ups and help these new sellers (“we know where they came from”), when other bigger FSPs will not.¹⁰² A respondent from a bigger FSP indicated that their clients are bigger brands that sell throughout Europe; they want a European solution that requires arranging EDI (Electronic Data Interchange)- connectivity once.¹⁰³

It is relevant to point out that almost none of the respondents have declared to provide services to non-EU-sellers directly,¹⁰⁴ mostly because of the financial risks relating to default of payments and recoverability of invoices (see further under par. V.3).¹⁰⁵ When asked where these non-EU sellers go to sell products on EU soil, most of the respondents refer to Amazon, because it is the main way to sell throughout the whole of Europe.¹⁰⁶ Others indicate that they don’t know (exactly).¹⁰⁷ One of the respondents indicated that apart from that, it is fairly easy for non-EU-sellers to find an external (consultancy) firm that may function as an authorized representative.¹⁰⁸

2. Knowledge of the MSR, requirements and obligations for FSPs¹⁰⁹

The interviews show that the legal departments in bigger FSPs that are part of or closely related to the network of Fenex-TLN are most aware of the provisions of the MSR and their content. When asked to explain what the MSR entails regarding FSPs, they provide a fairly accurate answer.¹¹⁰ But even then, the business structure and how information is distributed within big multinationals sometimes makes it hard for legal knowledge to disseminate through the organization.¹¹¹ Smaller FSPs that mostly operate nationally are not in these traditional informational networks, do not know what the regulation entails, and were – to a certain extent – hoping that the interviewer could provide them with further information.¹¹² These smaller companies do not have legal departments and compliance is an issue for general management, often the partners that have started the business together.¹¹³

In general, the level of knowledge about the MSR in the eye of the external legal advisors is perceived to be minimal under FSPs and EOs like sellers in general.¹¹⁴ What is interesting to see is that bigger FSP get more information through the already existing informational networks, consisting of their external legal advisors, through a seminar Fenex-TLN hosted on the topic, newsletters and research of their legal staff.¹¹⁵ The smaller FSPs however often resort to social media and closed Facebook groups in which information from sellers as well as FSPs and other consultants that hope to get new clientele from this group, is shared. This information consists of practical tips and tricks on how to maximize profit in e-commerce, reputational information regarding partners in the field, but also information regarding regulations and legal obligations. We have not obtained quantitative data yet, but one respondent indicates that these groups have up to 16.000 to 17.000 Dutch users, which potentially is a large group of sellers and FSPs.¹¹⁶ Another way to get information is through their personal networks.¹¹⁷ One respondent declares that the market is conservative and that not a lot of information is shared, because of the fear of losing customers.¹¹⁸ Also, sellers do not want to share info on the product they sell in fear of other sellers moving to their products.¹¹⁹

3. Contractual and practical measures ensuring compliance¹²⁰

The main company risk management measure relating to this legislative change is to make sure a EO is in place on EU soil, other than the FSP. Staff checks whether the FSP contracts with a EU-legal entity residing on EU territory this before entering into a contract.¹²¹ This is not checked everywhere.¹²² Smaller fulfilments service providers indicate that they also make sure the seller is EU based, not for product safety and liability purposes, but just to ensure the recoverability of invoices for services.¹²³ We have not encountered respondents that have admitted to providing fulfilment services to parties residing outside of the EU.¹²⁴

With regard to practical risk management measures, most consider these measures to be solely reactive.¹²⁵ One respondent explicitly considers that the introduction of the provisions has given him some “sleepless nights,” but since the company determined that the provision was not applicable in the end – due to there always being another EO available – the company has refrained from taking further practical risk management measures, also because of the costs (such as more elaborate product registrations systems for further checks on documentation).¹²⁶

Some parties indicate that the existing contracts and terms and conditions already have general provisions requiring contracting parties or the goods to comply with applicable regulation, which might cover these new responsibilities also.¹²⁷ Fenex-TLN has also developed new indemnity clauses and contracting policy in response to the MSR to be used by its members (not publicly available) as well as a standard declaration to let clients declare who is the responsible EO and sign off on this, but these are not all implemented.¹²⁸ Only one respondent has indicated that with regard to all running contracts, such a one-sided declaration was sent and received without explicit objections by the clients.¹²⁹ The practical solution of an informal check of another EO in the Union appears the main strategy. Some bigger FSPs have standardized the check of another EO in the Union by adjusting their warehouse management system with obligatory entry fields for clients that want to use their services: services may only be obtained if a responsible EO within the EU is indicated by the client. Otherwise, the order cannot be placed, nor processed.¹³⁰ Whether it is filled in correctly is only checked reactively.¹³¹ One respondent indicates that a contractual provision is included that the FSP is allowed to notify the authorities in case something is wrong with a product.¹³²

Bigger FSPs indicate that in response to the MSR special indemnity clauses regarding recall costs have been developed and are now made policy in drafting contracts,¹³³ also because insurance cover for these liability risks and recall costs is not available, as they have been told by insurers and/or representatives of the insurance industry.¹³⁴ Respondents in smaller FSPs often do not know what (exactly) has been agreed upon in the T&Cs,¹³⁵ nor do they have a clear overview of what (exactly) is insured (V.6).¹³⁶ Some smaller FSPs have received external legal advice regarding their contracts, but have also used contracts of other FSPs (which they had access to as being also an online seller online) as a model for their own contracts.¹³⁷

Existing contracts and conditions also include provisions that exclude liability or require cooperation in case of governmental action which respondents assume will also apply in case authorities contact them based on the MSR.¹³⁸ Since the respondents assume that their EU-based clients/sellers are responsible under the MSR, they assume such information is available and will be provided.¹³⁹ The external advisors see the inclusion of duties to cooperate as one of the main responses to the MSR.¹⁴⁰

What is contractually feasible ultimately depends on the room for negotiation and the commercial interest at play. With bigger FSP, the sales team does the negotiations whereas the legal check rests with the legal department.¹⁴¹ The intensity of the legal checks of the contracts, also regarding the indemnity clauses therein, also depends on the product market. In medical devices, the FSP’s clients have their own legal advisors, and more questions are asked about the content.¹⁴² In general, respondents indicate that there appears to be room for negotiation because there is not a strong dependency on certain sellers and/or platforms. Big FSPs sometimes have to do investments to be able to provide certain logistics services (keep warehousing available, specific machinery etc.) in which case specific contractual arrangements are made.¹⁴³ Some include liability caps.¹⁴⁴ Another respondent indicates that there is some leeway regarding indemnity clauses; it remains a risk assessment.¹⁴⁵ There is still room to say ‘no’ to risks that appear too big; the final decision lies with the director.¹⁴⁶ The external legal advisors are not always involved in the day to day negotiations, so they were not able to give detailed information in that regard.¹⁴⁷

The market for local, smaller fulfilment services appears not to be dictated by sellers (at least in the eye of the respondents). As one of the respondents put it: “there are more sellers looking for logistic services than there are logistic service providers available”, which makes it possible to be selective.¹⁴⁸ At the same time, with the companies that do not have product safety top of mind, the negotiations often regard other risks such as theft (see V.4 hereafter).¹⁴⁹

We asked FSPs about the processing of personal data and whether this data might be available for corrective action. Most FSP indicated that this data is available. They do not store it specifically for this purpose, but as a safeguard, in case something else goes wrong in the process (damage to the product etc).¹⁵⁰ One respondent indicates that these are mere postal addresses, and not e-mail addresses which – in case a product is sold through a platform – should be available at the platform and/or seller.¹⁵¹

4. Perception of the liability risk and responsibility¹⁵²

In general the liability risk is perceived to be small, not only because other responsible EOs in the Union are in place and deemed responsible,¹⁵³ but also because some respondents indicate that they mostly contract with companies with a good reputation.¹⁵⁴ The type of risks that are more top of mind with the smaller FSPs are risk of theft, pests and safety inside the warehouse (risk of fire, drugs), which are the risks that are addressed and/or covered in general terms and conditions.¹⁵⁵ Two respondents indicated that because of the fact that online platforms sharpen their product safety policies (asking for more documentation, pictures of CE-marking etc.), this increases the FSP’s confidence in those sellers that they have everything in order.¹⁵⁶

FSPs perceive the risk of market surveillance or enforcement to be very small.¹⁵⁷ Some indicate that they think market surveillance authorities are severely understaffed and therefore not enforcing or will not be too strict in light of the FSP’s limited role.¹⁵⁸ Some respondents indicate that chances are very small that they will be approached as a response regarding an unsafe product because the name of the company is not on the product.¹⁵⁹ Another explained that they have changed the name of the company on the packages they send into something unrecognisable, because of the fact that they were receiving too many calls from recipients who wanted to know who sent them the package (in case of surprise presents etc.).¹⁶⁰

None of the participants have heard of a FSP ever been approached by the authorities because of an unsafe product.¹⁶¹ Other authorities do enter the premises regularly (customs, food safety, existing other certifications).¹⁶² As one of the respondents put it: “The first time a fine is imposed or liability follows, this would be world news.”¹⁶³ One of the respondents indicates that no recall have taken place in the company; “otherwise people would have been more cautious about it, but I do not think this crosses people’s minds or that this is something they are aware of.”¹⁶⁴ This might be typical for the Dutch situation. One of the legal advisors explained that in Germany, the authority under the radio equipment directive has approached German FSPs for cooperation, which was a direct result of the MSR, but the respondent is not aware of any fines being imposed.¹⁶⁵

Civil liability – because of violation for a statutory duty from the MSR and/or product liability as a result of the pending revision of the product liability directive – does not appear to be top of mind or something that is considered. Most FSP-respondents are not aware of the existence of the PLDp when asked directly.¹⁶⁶ One respondent is not aware of the PLDp, but indicates that they are aware of the possible option in the future of extending liability for bodily injury and damage to goods to the FSP, which might be a socially desirable answer.¹⁶⁷ Another respondent indicates that they also perform small adjustments to products, which could result in product liability (changing plugs of electronics).¹⁶⁸ Another respondent points out that there might be a risk in stating your company name on a product which they thought might result in liability for the products itself.¹⁶⁹

Some bigger FSPs consider compliance to be important and “not something that is (a) negotiable.” The company has a good reputation to uphold.¹⁷⁰ At the same time, they do not understand the MSR approach, since it does not cover direct import. Moreover, they find it unfair that the MSR excludes postal services from its scope, while FSPs also do not know what is in the package.¹⁷¹

The safety risks regarding products from outside the EU are seen and acknowledged by the respondents. Some of the bigger FSPs declare not to consider the supplier of such risky products their clientele or core business.¹⁷² The smaller FSPs, do encounter a lot of “Chinese junk” in their warehouses – explicitly acknowledging the doubtful safety and quality of it – but see the safety of it as the responsibility of the seller.¹⁷³ As one of the respondents put it:

[Person 1]: You know, if you know that you sell vitamin D and that it has to be within a certain temperature, then you must also ask us about this in advance. We don’t do that out of the box, so to speak. [Interviewer]: So everything is done on the instructions of the selling party? [Person 1]: Yes, because yes, that’s what I’m a little afraid of sometimes. For people on medications and those kinds of conditions, what the health risk can be. Yes, that’s a thing, so to speak, so I’m afraid of that too. [Interviewer]: But do not proactively discuss it yourself when you see it in the warehouse? [Person 1]: Yes. Look, when the customer says…. Yes, of course you know. If you see it, sometimes yes, but hey, I can’t see everything. We have 8000 products. And it comes and goes every day, so in terms of visual supervision it is really very difficult. I know that [platform] does this based on data, because it would of course also cause reputational damage if you were to do it differently, but we could also look at that;,okay, what type or type product is it’. Anyway, we’re still a bit too small for that, I think.

Only one smaller FSP recalls an anecdote that they were approached by a seller who was excluded from a platform because of an unsafe toy that apparently was lethal but was just ordered by the seller that the FSP refused to handle/store (“Nice and all, but I have a child myself. We are not going or I am not going to facilitate this”).¹⁷⁴ At the same time, they see a lack of safety as a result of the product prices that are just unrealistically low.¹⁷⁵

Interestingly, these smaller FSPs also point out the lack of knowledge of sellers about legal requirements and the lack of market surveillance in that regard.¹⁷⁶ Some indicate that more information might raise the awareness and subsequently improve product safety.¹⁷⁷

5. Perceptions regarding the price of the services and of insurance premiums¹⁷⁸

Also, we asked participants whether the change in legislation led to an increase in product prices, due to converting the increase in risk or the rise in insurance premiums. None declare that this is the case.¹⁷⁹ This is consistent with the previous answers which indicate low knowledge under the small FSPs and low perception of liability risk in general, as well as a perceived lack of (affordable) insurance options. Riskier products and riskier businesses (start-ups) are reasons to take extra financial measures (require deposits, send invoices more regularly), because of the financial risks involved.¹⁸⁰ Smaller FSPs use fixed fees per package and work with add-ons regarding specific type of handling; one FSP indicates that indexing is allowed once a year and to a maximum of 5 per cent.¹⁸¹

The participants declared that the premium of logistic policies depends on the annual turnover of the company.¹⁸² Smaller FSPs are insured against all types of risks, including liability, but do not know whether product liability is part of it.¹⁸³ Another respondent at a bigger FSP indicates not to know which risks are covered, but assumes that product liability cover is part of it.¹⁸⁴

6. Perception on the role of the platform¹⁸⁵

Although the focus of this study was the position and role of the FSP, the role of platforms remained a reoccurring theme in the interviews. One of the external advisors declares that – from a sellers perspective – there have been discussions on whether to form a branch organisation or collective to counterbalance the power bigger platforms have over small sellers.¹⁸⁶ The small FSPs and the consultant do not find the current policy used by certain platforms just or experience their clients being a “toy in the hands of [platform]”; if sellers do not deliver in time or cannot live up to the platform’s demands, they are excluded from the platform too easily, which the FSPs notice because one of the main sale channels of one of their clients falls away.¹⁸⁷ A platform may in practice also demand to lower the product price because the product is cheaper elsewhere (see IV for implications).¹⁸⁸ One FSP also finds it undesirable from a sustainability perspective, that the platform does not cooperate more with local FSPs, because more warehousing facilities are being built while others are staying vacant.¹⁸⁹ That said, these respondents are in direct competition with the platform and some are also sellers on the platform themselves, which might colour their views (see under VI). It was hard to form a clear picture on the FBP-model in practice, due to secrecy and opacity of the T&Cs of platforms under this model, the latter was also confirmed by one of the external attorneys interviewed.¹⁹⁰

VI. Discussion

The aim of this study was to make an inventory of the legislator’s assumptions regarding the goals and possible effects of the (proposed) legal changes in the MSR, the GPSR and the upcoming PLD, by means of a legal analysis, and to verify, by means of a semi-structured interview study, whether the perceptions of FSPs and legal advisors in the field align with these assumptions. The findings are discussed together with some policy implications, after which the limitations of the study are addressed and a further research agenda is given.

To start with the MSR, this study provides indications that this legislative change might not (yet) have the desired effect, due to the lack of knowledge of especially smaller FSPs, a group who is also doing business with smaller unknowledgeable sellers. Improving information on the legal framework for product safety for EOs is one of the focal points for the EU-legislator and European Commission. Article 8(1) of the MSR holds that the European Commission shall ensure that the Your Europe portal provides users with easy online access to information about the product requirements and rights, obligations and rules derived from the Union harmonization legislation.¹⁹¹ Moreover, the European Commission tries to improve national product contact points to better inform EOs.¹⁹² EOs may request information at a local office if needed, after which it will be provided free of charge.¹⁹³ This still requires an initiative from the EO itself and some sort of awareness of the lack of knowledge, which this study shows, is not always there. This study also shows that for at least the smaller FSPs interviewed, social media and their own information networks are more important sources of information regarding legislation on product safety. If most of these new companies are not related to a branch organization, it is important that the parties within these groups are also actively reached.¹⁹⁴

Another worrisome finding – that could hinder the effectiveness of this legislative change – is the lack of perceived responsibility. Even if there is knowledge, FSPs perceive the sellers to be the completely responsible for product safety. The statements by FSP’s regarding the “Chinese junk” they encounter in their warehouses show that they acknowledge the risks that online products might impose and that they condemn these. At the same time, FSPs do not actively discuss product safety with sellers (ex-ante nor ex post). This underlying goal of the new framework (raising awareness and sharing information with “clients,” see also the Guidelines, II.1) is not reached. There could be several explanations for this apparent inaction, which need to be further explored.¹⁹⁵ What is clear is that, if the internal perception nor the social norm for FSPs is indicating that product safety is also a responsibility of FSPs, chances are extremely small that FSPs will actively take on any responsibility or proactively take action regarding product safety in the supply chain, and a corresponding role in this process. On top of this, there is a perceived lack of market surveillance as well as unawareness of a possible future increase in product liability risk (due to unawareness of the legislative framework as well as low chance of receiving a claim), and therefore a lack of deterrence.¹⁹⁶ Additionally, the platforms involvement is experienced by some as an extra security, because of their sharpened product safety policies (probably due to the Product safety pledge and revised GPSR with more extensive obligations). All together this indicates that – at least for the interviewees – there are few internal motivators/drivers nor external incentives for contributing to product safety and/or compliance, other than making sure there is another responsible EO present (see III).

Product safety and liability follow a cascade model that results in residual liability for FSPs, they do not provide for joint liability between EOs. Moreover, it follows the hosting exemption under the DSA for platforms.¹⁹⁷ These findings make us question the cascade model regarding its substantial effectiveness as these FSPs do not appear to move beyond formal compliance.¹⁹⁸ Within Europe, diverse models of regulation are proposed with regards to diverse interests. For example, the recently adopted Corporate Sustainability Due Diligence Directive goes beyond mere obligations and requires in-scope lead firms to put in place due diligence mechanisms, including contractual assurances, that create a responsibility throughout the value chain.¹⁹⁹ The regulatory approaches do not always appear consistent and need further reflection and research in e-commerce regarding trade of goods.

May it be expected that the increase in liability risk of FSPs under the PLDp leads to better compensation? This study shows that details of the FSP are often not present on the packaging (or not recognisable as such) which could make a claim against a FSP practically unenforceable.²⁰⁰ Transparency throughout the supply chain remains key: both for market surveillance and for product liability. For the time being, it is not always mandatory to register your responsible EO as non-EU producer in a public register, as it is under the Directive for waste electrical and electronic equipment.²⁰¹ Transparency and traceability might improve with the introduction of the digital product passport under the recently adopted Eco-design directive and the proposed Toy safety regulation, which hopefully will be used throughout the whole EU product safety framework.²⁰² If a product passport is not in place, a product may not be released by customs authorities for free circulation. A digital register for product passports will be drawn up which could also give insights into who currently serve as responsible EOs on EU territory.

Apart from the revised PLD’s contributory role to product safety, this research provides some information as to whether any further normative justifications for product liability of FSPs exist. Although the interviewees do not have specific knowledge of the product, they do appear contractually in a position to appoint another responsible EO (or refuse service) or take other risk-management measures toward sellers. Further quantitative research could show whether this type of capacity is representative, which could justify a liability in line with the importer.²⁰³

At the same time this study has limitations. It is perception research, and people’s perceptions do not always align with their actions. Moreover, in this study most FSPs indicate that they not to provide services to non-EU-sellers directly. This could be because there is an actual seller or an authorized representative present, or this could simply be a socially desirable answer. Either way, this justifies further quantitative research to see whether this applies to all FSPs under the FBM-model, in which case the increase in liability risk for them in practice could be limited. Also, we need to study the perspectives of sellers to see whether the answers match. It also raises the question who these sellers are and whether they live up to their product safety obligations, which might be at the heart of the problem.

Regarding online platforms/marketplaces and their liability, this research provides only part of the puzzle. Under the FBM-model, where no direct contract exists between a platform and an FSP, platforms still exercise a great deal of power towards sellers, which may practically extend to and effect the business of the external FSPs. This requires further quantitative research both regarding the influence of platforms through their sellers on FSPs, as well as the relationship between the seller and the platform to determine how platforms/online marketplaces should be regulated.²⁰⁴ On the one hand, some of the answers indicate that the product safety policies of some platforms might have sharpened, possibly as a result of the Product Safety Pledge and GPSR, which is a good thing. At the same time, FSPs also report unjust practices regarding price setting (which could pre-empt application of the DSA-host exemption),²⁰⁵ content moderation and blocking from the platform, which falls partially under the DSA and other regulations.²⁰⁶

This study focussed on perception research under ten participants – a relatively small sample – but could form a relevant basis for a quantitative study such as a survey or experiment under a more representative sample of legal staff under FSPs. Such a survey should not only be distributed through the more traditional channels of the relevant branch organisations, but also through social media, being an important source of information for new and upcoming companies.