8.3.1 Mobilising the Preliminary Reference Procedure

In several instances, actors have mobilised by making use of Article 267 Treaty on the Functioning of the European Union (TFEU).Footnote ²⁵ This procedure, primarily intended for resolving conflicts between national laws and policies and EU law or interpreting EU legislation, has the major downside of a dependency on national judges to pose a question to the CJEU. This has notably been explored by Passalacqua, who has researched how from the bottom up, focusing on the litigants, the decision to refer can be influenced with Euro-expertise (EU legal expertise) in the field of migration law.Footnote ²⁶ By having better resources available (i.e., knowledge), litigants are more likely to succeed in their endeavours (i.e., getting a judge to refer to the CJEU).Footnote ²⁷ Passalacqua focuses in her article on three instances of mobilisation against national laws and policies in the United Kingdom, the Netherlands, and Italy. Other such examples exist, for example, in the field of gender equality and non-discrimination law.Footnote ²⁸ The procedure has also been used to challenge acts of EU institutions due to alleged fundamental rights violations.Footnote ²⁹. One field of law where this has been particularly successful is explored here, looking at the cases, the actors behind them, the procedural hurdles they have had to overcome, and how they did so. This field of law is data protection.

The first widely known example is the case of Digital Rights Ireland, decided by the CJEU in April 2014.Footnote ³⁰ In that case, the EU Data Retention Directive was declared invalid due to non-compliance with Articles 7 and 8 of the Charter. The Directive required Internet Service Providers (ISPs) to retain internet data for a period of six months to two years.Footnote ³¹ The successful challenge was started in 2006 before an Irish court by the digital rights lobbying and advocacy NGO Digital Rights Ireland.Footnote ³² It challenged both the EU Directive and the Irish implementation thereof. Importantly, the NGO was allowed to conduct the proceedings in actio popularis, meaning that the organisation litigated for a public interest (namely the privacy rights of individuals).Footnote ³³ This possibility is not present in all EU Member States, highlighting the importance of beneficial national procedural rules. Moreover, among the members of the NGO (which has no staff or office) are practicing lawyers and legal academics.Footnote ³⁴ Digital Rights Ireland was supported by many other similar organisations, such as the Electronic Frontier Foundation and Privacy International.Footnote ³⁵ This means that the level of (Euro-)expertise within the NGO and in its relations with others was high. Despite significant delays that were suffered by the NGO in the national proceedings, eventually a preliminary reference was sent to the CJEU by the Irish court, which led to the CJEU judgment.Footnote ³⁶

The second example from the field of data protection is a fraction of a multitude of legal challenges brought by activist Maximilian Schrems. A preliminary ruling decided by the CJEU in October 2015, in the case of Schrems I, had significant consequences for the transfer of personal data from companies in the EU to the United States.Footnote ³⁷ The judgment invalidated the ‘Safe Harbor’ agreement between the EU and the United States, which allowed for the transfer of such data.Footnote ³⁸ The CJEU found that there were not enough safeguards in this agreement in light of Articles 7 and 8 of the Charter. Maximilian Schrems, an Austrian citizen (then law student) and Facebook user, challenged the transfer of his personal data from Facebook’s servers in Ireland to those in the United States (after the revelations made in 2013 by Edward Snowden).Footnote ³⁹ The Irish High Court was uncertain about the validity of the underlying Safe Harbor decision and decided to pose a question about its validity to the CJEU. The CJEU largely followed the arguments made by Schrems and declared the Decision invalid. Notably, as a third-party intervener, Digital Rights Ireland was involved in the case. Schrems himself is now a privacy lawyer and founder of an organisation (noyb) fighting for digital rights.Footnote ⁴⁰ Schrems I is one of the many legal challenges brought by the activist and his organisation.Footnote ⁴¹

These cases tell us several things. Firstly, the NGO Digital Rights Ireland benefitted from the possibility of starting an actio popularis in an Irish court, which is not possible in every EU Member State. Secondly, all litigants described in this section benefitted from legal, and specifically EU law, expertise, either through their own resources or with the help of a coalition of other individuals and organisations (sometimes formally through a third-party intervention).Footnote ⁴² This is necessary at times to provide a counterweight to the power of the national government to delay a procedure, such as in the Irish proceedings leading up to Digital Rights Ireland. It should be emphasised that these instances of mobilisation are rather exceptional: national implementation is necessary as litigation against EU acts cannot be brought before national courts. Therefore, challenging EU legislation or EU complicity in fundamental rights violations needs to be combined with a Member State ‘element’, so to say.

8.3.2 Direct Actions

The second type of mobilisation before the CJEU addressed here are direct actions, based on Article 263, 265, or 340 TFEU. Addressing fundamental rights violations by the EU through these procedures is difficult for a variety of reasons. Most obviously, the plaintiffs in the action for annulment and failure to act run into the difficulties of the Plaumann criteria of direct and individual concern.Footnote ⁴³ As emphasised in Chapter 2, the conditions under which compensation is granted in the action for damages are strict. Most likely related to these difficulties, only very few strategic procedures as direct actions before the CJEU have been initiated.Footnote ⁴⁴ Many attempts have been made to overcome these criteria especially in the field of environmental law. This line of case law, and case law more generally before the CJEU on environmental issues, is set out first in this section. The second part of this section focuses on two cases brought against the European Border and Coast Guard Agency, Frontex. One is an action for annulment, brought on behalf of two individuals by the NGO front-LEX.Footnote ⁴⁵ The other is an action for damages, brought on behalf of a Syrian family by Prakken d’Oliveira Human Rights Lawyers.Footnote ⁴⁶

The requirement of individual concern that follows from Plaumann is especially difficult to satisfy in environmental litigation: environmental measures are of a general nature and do not concern specific individuals.Footnote ⁴⁷ The first case in which this was explicitly dealt with was a case initiated by Greenpeace before the General Court (then the Court of First Instance) and later the Court of Justice.Footnote ⁴⁸ In relation to the procedural hurdle, Greenpeace defended the view that the environmental legal field is separate from other fields of law, as there is a particular type of public and shared interest at stake, unsuitable for a requirement such as individual concern.Footnote ⁴⁹ If locus standi was not granted in the case, Greenpeace argued that a legal vacuum would come into being. This case did not succeed. Shortly thereafter, the Aarhus Convention was signed by the EU in 2005.Footnote ⁵⁰ The third pillar of that Convention ensures access to justice in environmental matters, also (and specifically) for NGOs.Footnote ⁵¹ This led to a series of challenges to the strict locus standi rules in CJEU direct actions.Footnote ⁵² Still, no substantial change to the Plaumann criteria could be realised. Also attempts by individuals (together with an NGO) to overcome these criteria have proven unsuccessful.Footnote ⁵³ A different strategy was adopted by ClientEarth, who questioned the EU’s implementation of the Aarhus Convention before the Aarhus Convention Compliance Committee. This Committee found the EU to be incompliant with the Convention. This was subsequently used in litigation by the NGO Mellifera before the CJEU, but the Court rejected the Aarhus Committee’s finding alleging it is non-binding (a ‘draft version’).Footnote ⁵⁴ Environmental NGOs have not given up in this regard and continue to bring claims before the CJEU to overcome the Plaumann criteria. These cases form an example of the challenging of legal opportunity structures as part of strategic litigation: the NGOs in the cases purposefully litigate about procedural requirements to ‘loosen’ them.Footnote ⁵⁵

The action for failure to act and the action for damages brought against Frontex are pending at the time of writing, and it is uncertain whether the use of these procedures will be successful, that is, whether the applicants will have their cases declared admissible by the CJEU. Nevertheless, it is interesting to look at these cases in more detail, as they are one of the few examples in which these actions have been purposefully mobilised to address fundamental rights violations by the EU. The failure to act brought by front-LEX alleges a failure by Frontex to terminate its operation in the Aegean Sea, in the context of which the applicants suffered fundamental rights violations on their journey to Greece.Footnote ⁵⁶ Frontex argues that the complaint should be declared inadmissible as the actors that invited Frontex to act in the first place are the NGOs behind the complaint and not the applicants themselves.Footnote ⁵⁷ Moreover, the argument of a lack of locus standi (once again based on the Plaumann criteria) is put forward by Frontex as well.Footnote ⁵⁸ The second procedure, the action for damages initiated by human rights lawyers, is on behalf of a Syrian family who were allegedly victims of a ‘pushback’ by Frontex.Footnote ⁵⁹ The family applied for asylum in Greece but were deported by Frontex and Greek authorities on a plane to Turkey, where they were subsequently imprisoned.Footnote ⁶⁰ It is noteworthy that in both procedures, a more ‘strategic’ actor is acting on behalf of individuals, namely a human rights law firm and an NGO. The procedural reality of Articles 265 and 340 TFEU require this, as these actors (NGO and law firm) themselves cannot prove that they suffered damage or were directly harmed by a failure to act. Similar to the environmental field, these legal actions do not stand alone but are part of a multitude of political and legal efforts undertaken by different actors.Footnote ⁶¹ This shows the deployment of ‘integrated advocacy’, using a multitude of strategies simultaneously alongside litigation efforts.Footnote ⁶²

As can be seen from these examples, the procedural barriers in direct actions are high for strategic litigators. Direct access to the CJEU is limited, which requires legal action to overcome these barriers or legal action by making use of individual complainants. As stated in Section 8.2, EU legal opportunity structures are more ‘open’ when it comes to the availability of rights: an aspect that is used by the environmental NGOs to their advantage, by arguing for procedural flexibility due to strong substantive protection. EU legal expertise here is, once again, vital to successfully mobilise these procedures. Alongside this, relations need to be established with individuals on behalf of whom strategic litigation can be initiated. Moreover, a persistent strategy with multiple legal actions brought could increase the chances of success, but only time will tell. If the Frontex procedures prove unsuccessful, they can still provide a compelling argument for civil society in Europe that there is a lack of access to justice for fundamental rights violation by EU agencies.Footnote ⁶³ This could feed into the discussion of accession to the European Convention on Human Rights as well (see also Chapter 7).

8.3.3 Third-Party Interventions

In both the preliminary reference procedure and direct actions, there is the possibility for actors to intervene as a third party. Third-party intervention can be compared to the practice of using an amicus curiae, ‘friend of the court’, who can provide information to the judges to help them decide on the case.Footnote ⁶⁴ By providing the CJEU with arguments, the result of a court procedure can potentially be influenced. This is, therefore, a useful tool in strategic litigation. Moreover, as accessing the Court as a direct litigant is difficult, third-party intervention ensures participation in the proceedings.Footnote ⁶⁵ Nevertheless, intervening in CJEU cases is not an easy feat either, as is described below. In any case, the third party must ‘accept the case as he finds it’, meaning that no new grounds can be put forward.Footnote ⁶⁶ Still, this tool has been used by several actors before the CJEU. This section first delves into third-party intervention in the preliminary reference procedure, after which intervention in direct actions is discussed.

According to Articles 96 and 97 of the Court’s Rules of Procedure, only parties to the main proceedings can submit their observations to the CJEU in a reference for a preliminary ruling. In practice, this means that actors have to be involved at the national level already, otherwise the gates to intervene are closed.Footnote ⁶⁷ In turn, national rules on third-party intervention or submitting amicus briefs differ greatly among the twenty-seven EU Member States.Footnote ⁶⁸ This creates an uneven playing field. In the common law jurisdictions, submitting amicus briefs is relatively easy. This has resulted in some ‘strategic’ third-party interventions in important cases coming from UK and Irish preliminary references. Most notably, in the case of N.S. and M.E., there were multiple NGOs who intervened, such as UNHCR, Amnesty, and the AIRE Centre.Footnote ⁶⁹ The involvement of the NGOs in this case, a landmark case on the ‘Dublin system’ and the transfer of asylum seekers in the European Union, was the result of strategic involvement that started at the national level, otherwise intervention at CJEU level would not have been possible. The UK-based AIRE Centre has intervened in multiple CJEU cases in the field of migration and social policy.Footnote ⁷⁰ UNHCR has also intervened in a few CJEU cases, but when formal access was closed, it has turned to informally submitting its views (see Section 8.4.3).Footnote ⁷¹ As well as these examples from the migration field, interventions are also common in the aforementioned data protection cases.Footnote ⁷² Similarly, these cases originate from a common law jurisdiction and it can be assumed that the actors keep one another involved. This procedural framework places a heavy burden on NGOs who want to be involved in preliminary reference procedures at CJEU level, as they need to be aware of cases at the national level and national rules need to allow for intervention.Footnote ⁷³

In direct actions, the possibilities to intervene are a bit broader, as any person that ‘can establish an interest in the result of a case submitted to the Court’ can intervene.Footnote ⁷⁴ This interest is generally interpreted by the CJEU as meaning that the economic situation of the intervener is directly affected, but there has been a more flexible approach towards associations promoting collective interests.Footnote ⁷⁵ Several environmental cases can be found before the CJEU where parties intervened; however, a problematic aspect for the interveners is that if the applicants in the main proceedings lack standing, there is no possibility to intervene on the merits.Footnote ⁷⁶ Thus, the strict locus standi requirements in direct actions have an effect on third-party intervention as well. It is only exceptionally that third-party intervention in direct actions happens in practice.Footnote ⁷⁷

In sum, although third-party intervention could be a way for NGOs to be involved in strategic litigation, only a few successful examples can be found. A dependency on national rules and the strict requirements in direct procedures before the court form obstacles for this strategic litigation avenue. This makes this avenue resource-intensive: there needs to be awareness of cases brought at the national level for third-party intervention in preliminary rulings and intervening in direct actions is a waste of time if a case does not pass the admissibility stage. Moreover, relations among organisations seem to be of importance: keeping each other informed so that third-party interventions can be submitted.

8.4.1 Petition to the European Court of Auditors

One institution that has been called upon regarding alleged fundamental rights violations by the EU is the European Court of Auditors (ECA). The ECA is responsible for carrying out the EU’s external audit, supervising control over the EU budget.Footnote ⁷⁸ In April 2020, three NGOs submitted a complaint to the ECA about the mismanagement of EU funds through the ‘Support to Integrated Border and Migration Management in Libya’ (IBM) Programme.Footnote ⁷⁹ These NGOs are repeat-players in strategic litigation, especially when it comes to the field of migration law.Footnote ⁸⁰ In a novel way, the NGOs link fundamental rights obligations of the EU to the management of funds, alleging that the EU budget is used to violate fundamental rights (in Libya). The ECA responded in May 2020 that it would not initiate a review of the IBM programme at that time, partly due to its limited resources.Footnote ⁸¹ Still, this complaint presents an interesting case study of a route that could be used successfully. As a downside, the actors submitting such a complaint can do no more than that: submit information to the institution. There is no possibility of formal involvement in any procedure that follows.Footnote ⁸² Nonetheless, if a formal review is initiated by the ECA, this could lead to more accountability of the EU for possible fundamental rights violations and could become a useful avenue in fields other than migration as well.

8.4.2 Calling upon Other (Non-Judicial) European Institutions

The aforementioned strategies relate to the use of courts, but other EU institutions can also be called upon to challenge alleged fundamental rights violations by the EU. Under Article 227 TFEU, natural and legal persons are given the right to address a petition to the European Parliament on EU activities that affect them directly. The same three organisations that submitted the complaint to the ECA have made use of this right of petition on the same topic, the mismanagement of EU funds leading to fundamental rights violations in Libya.Footnote ⁸³ The argument made to use Article 227 TFEU is that the organisations consist of EU taxpayers, hence the mismanagement of funds affects them directly.Footnote ⁸⁴ A Committee on Petitions of the European Parliament considers and can follow up on these petitions.Footnote ⁸⁵ For strategic purposes, a downside is that the petition must be brought by the ones directly affected by the EU’s activities. Therefore, a strategic litigator cannot send the petition on behalf of a large group of individuals who suffer from EU fundamental rights violations.

A second available route that makes use of an EU non-judicial institution is the submission of information to an EU body such as the European Data Protection Board (EDPB). The NGO noyb of privacy activist Maximilian Schrems has made use of this option, by submitting a complaint to multiple Data Protection Authorities (DPA) in May 2018 about the tech company Meta running personalised ads without user consent.Footnote ⁸⁶ Four years later, in December 2022, news broke that Meta will have to pay a fine of almost 400 million euros.Footnote ⁸⁷ Although the decision on the fine was taken by the Irish DPA, it was pushed by the EDPB to do so. The EDPB coordinates the consistent application of EU privacy rules throughout the EU.Footnote ⁸⁸ Similar to the procedures mentioned before, the involvement of strategic actors stops after submitting a complaint and information. Nevertheless, if the consequent procedure has a positive outcome, the desired impact can become a reality.

A last avenue that is already elaborated on elsewhere in this volume (Chapter 5) is the submission of complaints to the EU Ombudsman. As the Ombudsman is responsible for investigating complaints about maladministration of the EU, fundamental rights issues have also been brought to the attention of the EU Ombudsman.Footnote ⁸⁹ For example, several environmental NGOs have submitted a complaint on the European Investment Bank not disclosing information about the environmental impacts of financed projects.Footnote ⁹⁰ This led to the Ombudsman issuing decisions on this topic in April 2022.Footnote ⁹¹ However, such decisions are non-binding, which poses limits to the effectiveness of this option (if decisions are positive for the strategic actors in the first place).

8.4.3 Informal Involvement

Strategic litigation is not always visible: at times, strategic involvement by actors is ‘hidden’, behind individual clients or other organisations. Such informal involvement can still be a way of influencing legal procedures, including within the EU remedies system, and therefore deserves attention here. For example, as described above, formal third-party intervention in the preliminary reference procedure before the CJEU is rather difficult. In that regard, it is interesting to note that UNHCR has resorted to informal third-party intervention submissions in the form of ‘written observations’ or ‘public statements’.Footnote ⁹² Although the influence of such informal involvement is difficult to measure, it is a way for actors to have their voice heard in Luxembourg. Another way of being involved is through the provision of expert opinions or expertise to lawyers who have cases pending before the CJEU. This is the strategy deployed by the Dutch Council for Refugees (DCR). Due to their network with Dutch migration lawyers, the DCR and their Committee Strategic Litigation is able to (at times) provide arguments to lawyers in preliminary references before the CJEU that originate from the Netherlands.Footnote ⁹³ Additionally, in their written submissions before Dutch courts, the Committee often pushes for preliminary questions to be asked.Footnote ⁹⁴ Similarly from the Dutch context, the informal involvement of a law clinic can be pointed out. At the Vrije Universiteit Amsterdam, the Migration Law Clinic has been submitting ‘expert opinions’, which support the reasoning of the lawyers in the preliminary ruling procedure.Footnote ⁹⁵ Through these lawyers’ pleadings, the submissions of the DCR and the law clinic are brought into the case. Again, these types of involvement show the relevance of coalitions and networks, such as contact with national lawyers and knowledge of pending preliminary references and the importance of Euro-expertise.

9.2.1 Defining International Arbitration

International arbitration can be defined as an extra-judicial mechanism of dispute resolution, which is considered one of the alternative dispute resolution mechanisms.Footnote ³ The term international commercial arbitration refers to a procedure where disputes between private individuals or corporations are resolved.Footnote ⁴ Moreover, it can also be between a state acting in its private capacity and a private party. The mechanism is considered to be part of the adjudication system of any legal order.Footnote ⁵ There is a group of scholars who are of the opinion that all arbitration is private, even when it concerns a dispute between an investor and a state. This idea mainly stems from the fact that the arbitrators who adjudicate on the dispute are appointed by the parties, rather than belonging to a judicial institution. However, this is not always the case. The fact that it may be a private dispute resolution system does not take away from the fact that it is a mechanism that guarantees access to justice and has final and binding arbitral awards. The author is of the opinion that private arbitration concerns disputes that are brought by private parties or between a private party and a state acting in its private capacity.

Private arbitration allows the parties to create their own arbitration process.Footnote ⁶ This is done through the principle of private autonomy, as it is an integral principle of arbitration. Hence, this also means that the elements of the arbitration process result from choices the parties make. For example, the seat of arbitration, which procedural rules apply, and the arbitrators are all choices that the parties make themselves. This means that there also has to be consent given by the parties to arbitration or it needs to be provided in an agreement. Therefore, arbitration is usually an agreed form of dispute settlement.

9.2.2 Benefits of International Arbitration

International arbitration has become the preferred mechanism for many cross-border commercial disputes due to three main benefits.Footnote ⁷ The first is the flexibility in proceedings. The parties to the arbitration agreement may choose the type of tribunal that will hear the dispute, the applicable law (lex arbitri), the seat of arbitration, and which conflict of laws to use. Hence, the parties are able to choose the evidence rules and the time limits since most of these rules are outlined in the applicable and chosen law. Therefore, the parties are able to design their own proceedings.

A second benefit of arbitration is the ability to choose one’s arbitrators.Footnote ⁸ Since disputes that arise come from different fields, they require different types of expertise to resolve the dispute at hand. The parties to a dispute are free to choose arbitrators according to their expertise.Footnote ⁹ There are arbitrators who are either professors or practitioners in different fields of law and beyond. This is particularly advantageous when the resolution of a dispute hinges on specific technical expertise that judges may not always possess.

A third benefit of arbitration is that an arbitral award is final and binding.Footnote ¹⁰ The arbitral award has res judicata effects.Footnote ¹¹ Unlike other private dispute resolution mechanisms, an arbitral award can and should be enforced in any national jurisdiction.Footnote ¹² Since arbitral awards are recognised and enforced in national jurisdictions, this means that the arbitrators will do their utmost best to resolve the case, as any arbitral tribunal would strive for the award to be enforceable (depending especially on whether the state signed the New York Convention).Footnote ¹³ Therefore, although arbitration is considered a private dispute resolution mechanism, all the regulations that concern the protection of the weaker party or fundamental rights are still applied in arbitral proceedings. Hence, while doing their jobs, arbitral tribunalstend to apply, even if not obligated to, laws and norms that any national court would apply, in order to achieve enforceability of the arbitral award.Footnote ¹⁴

9.2.3 Limitations of International Arbitration

As with any system, there are also a few limitations that can be attributed to using arbitration as a dispute resolution mechanism. The first limitation concerns the confidentiality of the dispute resolution process.Footnote ¹⁵ It poses a limit to commercial arbitration since awards are not always publicly available. It poses risks that decision-making in arbitration proceedings does not adhere to any sort of check and transparency. For EU law purposes, this is a limitation since it is not possible to check how EU law was interpreted in a dispute. Moreover, this translates into issues with legal certainty since it then becomes difficult to establish a common line taken in awards is in a certain situation.Footnote ¹⁶ Hence, it poses a limit to using arbitration.

A second limitation of arbitration, which ties into the previous one, concerns the issue of transparency and consistency.Footnote ¹⁷ Due to some arbitral awards not being made public, this raises the question of how disputes are in fact resolved by the arbitrators. This also plays into decision-making of arbitral tribunals not being public, hence not transparent. A limitation exists here in the sense that when a dispute deals with private and public interests and is not transparent, that is a problem. Since public interest can be violated when a private interest is decided, it becomes crucial for national courts who enforce the award to check this. This poses a risk that national courts when enforcing the award will want to carry out a full review.

A third limitation is that arbitration can form a limit to effective judicial protection within the EU. Given the fact that international arbitration exists outside the scope of the EU judicial mechanism, if a party to a dispute decides to take a dispute to arbitration, that party is in fact deciding to take the dispute out of the EU judicial system. From an EU perspective, practically this means that effective judicial protection as guaranteed under Article 47 Charter of Fundamental Rights of the European Union (‘the Charter’) is not achieved.Footnote ¹⁸ On the EU level, more specifically, Article 47 Charter recognises access to justice as a ‘core fundamental right’.Footnote ¹⁹ This means that arbitration undermines this fundamental principle of EU law. Secondly, if an arbitration clause in a contract or agreement exists, the dispute will go to arbitration, which means that a national court can only enforce or annul the arbitral award, hence a limited judicial review.Footnote ²⁰ Although this does not mean that it is a breach of fundamental rights per se, it does mean that the EU threshold of effective judicial protection might not be met. Therefore, from an EU law perspective, the limited review stemming from the res judicata of arbitral awards may undermine effective judicial protection.

9.3.1 The EU’s Competence and Arbitration

International arbitration is considered to belong to a branch of private international law, which predominantly relies on party autonomy and expression of will from parties to a dispute. There is no fully-fledged European Private International Law, although it is becoming more regulated within the EU sphere.Footnote ²¹ Since arbitration mainly depends on private law and domestic procedural law, it is also dependent on private international law. Hence why this framework is relevant for using arbitration.

Even though the EU has expanded its regulation into numerous policy areas, private international law has been an area of law that was originally left untouched.Footnote ²² This is also evident from the EU’s competences in matters of private international law. There is no specific field of ‘EU arbitration law’ that would be regulated in the Treaty on the Functioning of the European Union (TFEU).Footnote ²³ Generally, there is no provision in the Treaties that explicitly refers to, or deals with, arbitration.Footnote ²⁴ However, this was not always the case. A provision was inserted into the Treaty of European Economic Community (hereinafter, ‘EEC’), under Article 220 EEC.Footnote ²⁵ Here, the following is established: ‘Member States shall, in so far as necessary, engage in negotiations with each other with the view to securing to the benefit of their nationals: … the simplification of the formalities governing the reciprocal recognition and enforcement of judgments of courts or tribunals and of arbitration awards.’Footnote ²⁶

The Maastricht Treaty did not include private international law, even though there were important legal architecture changes being made. A ‘Three-Pillar Structure’ was created and placed judicial cooperation into the third pillar.Footnote ²⁷ This created important substantive changes for the Treaty. For private international law, this meant that the integration was less than in fields that found themselves in the first pillar. Through the pillar structure, Member States gained more room for cooperation.Footnote ²⁸ This was seen as a compromise whereby Member States achieved more ways for cooperation within the Pillar Structure but were not subject to the supranational integration approach and the sensitivity of the fields was still respected.Footnote ²⁹ This approach also allowed the European Council to still retain its powers. Therefore, a middle ground was found to address these more sensitive topics.

Only within the framework of Communitarisation, in the Treaty of Amsterdam, did private international law become part of the first pillar.Footnote ³⁰ Ever since then, international arbitration and EU law have come into closer contact.Footnote ³¹ Though in practice, the interaction of arbitration and the EU legal order differs to varying degrees depending on the field of EU law and the competence that the EU and Member States have in the given fields.Footnote ³²

The inclusion of private international law in the first pillar created a shift in the competence divide between the EU and Member States. What was originally a matter for Member States to decide, given that the area contains private relations, now gave the EU competence.Footnote ³³ The EU has competence to ‘take measures in the field of judicial cooperation in civil matters having cross-border implications’, as long as this relates to the proper functioning of the internal market.Footnote ³⁴ As stated under Article 3(2) Treaty on the European Union (TEU), a common area and the development of cooperation in civil matters having cross-border effects is possible with the legal basis under Articles 67 and 81 TFEU.Footnote ³⁵

Later, with the Treaty of Lisbon, the Union only gained powers to initiate development of alternative dispute settlement methods, which most of the time excludes arbitration as discussed in this chapter.Footnote ³⁶ These powers are not absolute, but this explains why there has been more reference to alternative dispute settlement as a potential remedy in secondary legislation.

Treaty amendments meant that the EU and Member States now have a shared competence in the field of private international law. For arbitration, Member States still retain most powers since most arbitration is governed domestically.Footnote ³⁷ The principles of conferral, subsidiarity, and proportionality apply.Footnote ³⁸ As also stated under Article 3(2) TEU, what can be seen within the EU is that a common area and the development of cooperation in civil matters having cross-border effects is possible with the legal basis under Article 81 TFEU.Footnote ³⁹ On the other hand, given the shared competence, EU institutions can enact legislation.Footnote ⁴⁰

9.3.2 Legal Instruments That Regulate Arbitration

On the international level, the New York Convention is an essential multilateral international treaty in the field of arbitration.Footnote ⁴¹ The Convention deals with enforcement and recognition of arbitral awards and ensures that arbitral agreements are upheld. All Member States are also signatories to the New York Convention. The EU itself is not a signatory to the New York Convention and there is no equivalent that exists on the EU level. There is the Brussels I bis Regulation, the successor of the Brussels Convention, which deals with the free movement of judgments in cross-border situations.Footnote ⁴² The Brussels I bis Regulation covers all civil and commercial matters. The term ‘civil and commercial matters’ is not defined under the regulation itself; the concept has gained an autonomous interpretation under EU law.Footnote ⁴³ The main aim of the Brussels I bis Regulation is ensuring the proper functioning of the internal market, judicial fairness, and international comity.Footnote ⁴⁴ Arbitration is, however, excluded from the scope of the regulation. This is provided in recital 12 and Article 2(d) of the regulation.

The reasoning behind the arbitration exclusion was mainly due to prior existence of international conventions, specifically the New York Convention, which dealt with arbitration when the Regulation was enforced. This is why most issues of arbitration are left to the New York Convention and national legislation to deal with. This way EU legislation avoided overlap with existing international conventions.

On the national level, each jurisdiction has their own rules that regulate arbitration. It is only on this level that one can find codes of civil procedure and rules on how arbitral tribunals and courts should interact (together with the rules that exist on the international level).

9.4.1 The CJEU Approach to International Arbitration

Constitutional limits are crucial in assessing the legal space that exists for arbitration within the EU judicial system. The main limit has developed through the CJEU’s approach to arbitration. Therefore, this section will outline the leading case law that deals with arbitration. The section is structured in a way that resembles the argumentation used by the CJEU when dealing with arbitration, namely in light of the autonomy of EU law and the uniform application of EU law.

9.4.2 Interim Conclusion

This section shows that the CJEU’s approach limits arbitration based on autonomy and ensuring effective application of EU law. Although these arguments are quite reasonable from the point of view of ensuring and protecting the EU legal order from external influence, this cautious view may not be working in the CJEU’s favour. Although this section has focused on the most integral cases while mapping out the legal room for arbitration, the discussion has also shown a broader ‘cost’ that would be incurred if international arbitration were used. This cost is twofold.

Firstly, for international arbitration to have room within the EU, a broader definition of autonomy needs to be applied. A softer definition of autonomy would allow the EU to interact with external bodies, such as arbitral tribunals, without it posing a risk to constitutional norms of the EU. The ‘cost’ would be that arbitrators apply EU law in each case. This of course poses the risk that EU law would be interpreted incorrectly and would dilute the uniform application of EU law. However, as the system currently stands, arbitrators already apply EU law when dealing with an arbitration proceeding. Hence, it may be better for the EU to reduce the risks by opening up to the possibility of using arbitration as a dispute settlement mechanism.

Secondly, from the EU perspective, having more arbitration means a balancing act needs to be struck between effective judicial protection as enshrined under Article 47 Charter and having arbitration achieve access to justice from the international perspective. From the EU perspective, utilising arbitration comes at a cost of taking the dispute outside the scope of the EU judicial system, meaning that individuals would not receive the same guarantees as envisaged under Article 47 Charter. Paradoxically, though, on the flip side, this then means that using arbitration could resolve the problem individuals have in enforcing their rights.

From the international law perspective, these safeguards exist in the international domain within the arbitration practice. This, however, is not a gamble that the CJEU is willing to take, since there is no guarantee that these safeguards would be used. Therefore, it is necessary to perform a balancing act between the principles at stake.

9.5.1 Way Forward for Potentially Using Arbitration in Fundamental Rights Violations

A distinction should be made between law enforcement and rights enforcement in the EU when arbitration could potentially be used as a supplementary tool. Since the crux of using arbitration is to induce rights enforcement, this is the main focus. The reason for this distinction is twofold. Firstly, the main argument of the CJEU where arbitration is concerned has been that it needs to be limited mainly for the purposes of ensuring effective and uniform application of EU law; this can be related to law enforcement grounds.Footnote ⁶⁸ This means that arbitration, as a consensual (commercial) dispute resolution mechanism, can be used since the parties know (or at least should) that they take the dispute out of the scope of the EU judicial system. Therefore, it comes down to a balancing act between application of EU law and party autonomy in order to guarantee rights enforcement.

It is not disputed that the EU judicial system operates on the idea that individuals derive rights from EU law, which should entail a possible remedy if such rights are infringed.Footnote ⁶⁹ This presumption also operates on the basis that a right to an effective remedy should not be harmed by national procedural requirements that place grave burdens on the individual and possibly harm a claim to be lodged.Footnote ⁷⁰ The CJEU has stated in numerous instances that an individual should not have their substantive rights hindered due to overburdensome procedural requirements of a Member State court system.Footnote ⁷¹ In precisely these situations, where strict national procedural rules hinder the possibility of access to court, for example, arbitration could be used to enforce rights. Since arbitration is predominantly made up of civil procedure, contract law, and private international law, it can be a fitting adjudication system to serve as a rights enforcement mechanism.Footnote ⁷² This way, an individual seeking to enforce their substantive rights would be able to do so with arbitration, avoiding the strict procedural hurdles of national courts or the limited standing before the CJEU.

If one were to extrapolate the reasoning of the CJEU in consumer protection cases where the Court deals with situations involving a restriction of rights of consumers to an effective remedy and an exercise of rights conferred by the Unfair Terms in Consumer Contracts Directive, much could be used as an inspiration for using arbitration.Footnote ⁷³ Given the fact that EU competences are expanding, and with it more secondary legislation is being drafted, there are more rights that are derived and need a viable remedy available. The expanding competences also result in the EU having more room to potentially violate fundamental rights, since it has more involvement within various fields. This means that a judicial system is necessary to mirror this – this is where arbitration could come in.

The caveat here is that it comes at the price of taking a dispute outside the scope of the EU judicial system and EU law would be interpreted by arbitrators. Therefore, this becomes a balancing exercise where the balance needs to be struck between effectively interpreting EU law and ensuring that individuals have a venue to seek redress for fundamental rights violations. The argument put forward in this chapter is that international arbitration should in no way replace the EU judicial system. The thesis is that arbitration could be used as a supplementary mechanism for when there is an enforcement gap in the EU. In addition to being an alternative way to remedy fundamental rights violations by the EU, arbitration can serve as a tool for rights compliance, hence stepping in when no access to an effective remedy is possible.

9.5.2 Practical Implications for Using Arbitration within the EU Judicial System

It is quite a bold statement to say that international arbitration should be used in situations where the EU violates fundamental rights of individuals, since in practice this would mean that arbitrators would apply and interpret EU law in situations where the EU has violated fundamental rights and the EU would allow for this (one would need consent for this from both sides). However, if one follows the reasoning that an individual who derives rights from EU law should also have a remedy available when those rights are violated, it would be beneficial for the EU to use arbitration as a supplementary mechanism, instead of not having any possibility to enforce their rights. Simply put, the balancing act rests with the question of whether it is better to use international arbitration and risk potentially diluting EU law than not having access to court available in situations of fundamental rights violation. This is because in practice arbitral tribunals apply EU law as developed by the CJEU because these tribunals want their arbitral awards enforced. Therefore, one could understand why the CJEU makes the argument of autonomy and uses a guarded approach in relation to arbitration, however, this approach harms the EU, especially as a global actor.

Institutionally speaking, the competence to act in this field is already established by the Treaties as discussed before, therefore no Treaty amendments would need to be made (neither are they desired). What is required, however, is a shift in approach. This is dependent on whether Member States could reach a common approach in how to treat arbitration, since most rules are still governed by domestic law. Moreover, it depends on the approach that the CJEU continues to develop towards arbitration. The question still stands whether commercial arbitration will keep being treated differently from investor-state arbitration or whether the lines will soon blur, depending on the cases that the Court receives.

The CJEU’s approach also develops constitutional principles such as proportionality, subsidiarity, and conferral, which remain the principles at the EU’s disposal to resolve questions that deal with arbitration.Footnote ⁷⁴ Therefore, it is a matter of approach and using the principles in the EU’s ‘toolbox’ in the relationship with arbitration. Practically speaking, at the time of writing, this may be hard to achieve since arbitration and the EU are not in a state of full cooperation as the two systems function independently. Taking this coexistence of systems together with the constitutionalisation of EU law, the above-mentioned discussion shows the costs that come with using arbitration. With this said, the analysis has also shown openness and some space for arbitration within the EU legal order. This is something that both Member States and the EU should keep in mind.

10.2.1 A Brief History of ODR: From Private Actors to Public Sector

To better understand the nature and capabilities of ODR, it is worth starting with a short history of ODR. ODR emerged in the mid-1990s as a response to the Internet becoming more widely available and accessible for commercial and private use. During this early period of the Web, the first online marketplaces like Amazon and eBay launched, and projects of collective content creation by internet users like Wikipedia emerged. With more and more online interactions and commercial transactions taking place around the globe, the number of conflicts originating from these interactions rose. It quickly became apparent that traditional court systems were not best suited to deal with conflicts originating from interactions on the Web. First, online transactions often had a cross-border dimension, which created a host of jurisdictional questions for national courts.Footnote ⁶ Second, the disputes were often of low value, making recourse to the traditional court system disproportionately costly.

Online intermediaries soon understood that they would have to find solutions for solving disputes between users if their business model were to succeed. In the case of online marketplaces, for example, promises of cheaper prices and greater convenience were insufficient to attract buyers and gain their loyalty. Buyers had to trust the platform. To build trust and prevent sham offers and fraudulent practices by buyers or sellers, platforms introduced rating and feedback mechanisms. These mechanisms, however, did not eliminate disputes between buyers and sellers. This is how eBay, for example, entered into a collaboration with researchers from the Massachusetts Amherst Center for Information Technology and Dispute Resolution to design an online dispute resolution system to settle disputes between buyers and sellers in 1999.Footnote ⁷ By 2010, eBay’s dispute resolution system handled sixty million claims per year.Footnote ⁸ The eBay example is useful to highlight two features of ODR. First, ODR initially developed and grew in the private sector, for business reasons. Second, the early history of ODR and its handling of extremely large numbers of complaints have given ODR a reputation of being a fast, cheap, and effective mechanism for settling disputes.

Another example of an early ODR mechanism is ICANN’s Uniform Domain Name Dispute Resolution Policy. In contrast to e-commerce platforms, ICANN performs a global internet governance function: it is a private, not-for-profit organisation that administers, among others, the Internet’s global domain name system.Footnote ⁹ ICANN’s ODR system is administered by various accredited arbitrators and deals in most part with claims from trademark owners that oppose the registration of domain names that use their trademarks without authorisation. As with e-commerce and other internet platforms, ICANN’s ODR system allows the settling of disputes in the place where they originate: on the Internet.

While ODR was initially conceived as a dispute resolution mechanism for disputes that arose online, its benefits like speed, accessibility, convenience, and expertise, as well as its trust-building function soon inspired various actors to pilot ODR mechanisms for offline disputes. The New York City Government, for example, started using the private ODR provider Cybersettle to allow individuals to settle their claims against the city relating to sidewalks, personal injury, or traffic vehicles.Footnote ¹⁰ Cybersettle was used in cases where the city’s liability was not disputed and only the amount of monetary compensation had to be determined. The Cybersettle ODR tool is an automated negotiation process, which offers a blind-bidding process where software determines a negotiated outcome based on secret offers made by the parties. It should be noted, though, that automated negotiation processes are only useful if the only outstanding issue between the parties to a dispute is the amount of monetary compensation.Footnote ¹¹

The potential of ODR for offline disputes, however, goes further. Courts and administrative agencies have found various opportunities for using ODR in the public sector for disputes that go beyond the mere settlement of the amount of compensation. States and provinces in the United States and Canada, for example, have introduced ODR systems to decide appeals against property tax claims.Footnote ¹² The ODR system for property tax appeals introduced in 2012 in British Columbia, for example, yielded an amicable solution in 75% of appeals submitted.Footnote ¹³ Overall, users of the system reported high satisfaction rates regarding the accessibility and convenience of the system.Footnote ¹⁴

ODR has also been implemented in various jurisdictions at a pre-trial stage for a variety of civil law disputes. In the Netherlands, for example, the platform Rechtwijzer started operating in 2014 as an ODR tool for family separation and divorce disputes, which allowed the parties to engage in a structured dialogue to reach a divorce settlement agreement.Footnote ¹⁵ Agreements reached through the platform can be subsequently presented before court to make the settlement binding.

One of the most ambitious implementations of ODR systems connected to the judiciary is the English Civil Online Court (Online Civil Money Claims). The original plan for the English Civil Online Court was to handle most civil claims for sums below £25,000 through a three-stage process: at the first stage, software would assist parties in filling out claims and responses. At the second stage, online or telephone facilitation by court-provided case officers would be offered to the parties to reach a settlement. Lastly, if no settlement was reached at the second stage, the claim could be taken at a third stage to a judge who would decide the case based on the documents submitted during the prior stages. If required by the circumstances of specific cases, the judge could conduct hearings online, by telephone, or face-to-face. The Online Civil Money Claims project launched a pilot in 2017 for claims brought before County Courts for money claims of up to £10,000 brought by individuals or up to £25,000 brought by represented individuals.Footnote ¹⁶ More than 100,000 claims were issued in the first eighteen months of operation of the pilot.Footnote ¹⁷ Furthermore, the pilot was expanded during the Covid-19 pandemic due to increased demand.Footnote ¹⁸

10.2.2 The Technological Component of ODR

To understand the potential of ODR, it is useful to start from its basic elements and various possible design choices. At the most basic level, ODR adds a ‘fourth party’Footnote ¹⁹ to dispute resolution. Traditional dispute resolution involves three parties: applicant, defendant, and a neutral third party (a judge, a mediator, an arbiter). ODR adds to these three parties a fourth party: technology. In ODR, technology comes in the form of software, user interfaces, databases, electronic communication, etc. In contrast to offline ADR, which, in principle, only requires a neutral individual acting as third party and can be set up ad hoc, ODR requires a more permanent infrastructure. It requires at least an electronic system or platform for ‘generating, sending, receiving, storing, exchanging or otherwise processing communications in a manner that ensures data security’.Footnote ²⁰ Technology can have an important role at the initiation, resolution, and enforcement stages of ODR procedures.

When initiating an ODR procedure, technology can enable a variety of functions. It can allow for asynchronous communication between parties and alert parties whenever a party has submitted a new communication. Technology can also streamline the completion of forms and documents, so that all relevant arguments and evidence are presented and structured in a way that allows the settlement to be conducted more smoothly. This is one of the important functions of the technological solutions in the examples of Rechtwijzer and the first stage of the UK Online Civil Money Claims. In a more sophisticated version, for example, with the help of Artificial Intelligence (AI) tools, technology can help to identify and name a legal issue for parties in the first place. In a pilot programme funded by PEW Charitable Trust, researchers from Stanford and Suffolk University, for example, are developing and testing natural language processing (NLP) algorithms to help ODR users identify and frame their legal issues.Footnote ²¹

Software tools can also help in providing options for solving conflicts at the settlement stage of the procedure. At a very basic level, ODR can simply offer video conferencing and other communication tools that enable parties to negotiate or a third party to hear parties and to propose solutions. But technology itself can also help in guiding parties to find a solution. As the example of Cybersettle above shows, automated blind bidding can help parties to find common ground to agree on monetary compensation. In the context of e-commerce, some platforms have experimented with software solutions that enable crowd-judging. On Alibaba’s Taobao platform, for example, disputes between buyers and sellers are solved by authorised users voting on the most appropriate solution for the dispute.Footnote ²² With the lightning speed at which AI solutions are developing, we can probably expect more and more automation of the solution of disputes with ever more complex questions involved, which will be available not only to ODR mechanisms but to traditional ADR and court systems, too.

Lastly, technology solutions embedded in an ODR system can ensure the enforcement of settlements. In the case of online platforms that also control payments (e.g., through collaboration with a payment service provider), the outcomes of dispute resolution between buyers and sellers, or service providers and recipients, can be enforced through charge-back functions.Footnote ²³ If the outcome of a dispute settlement between buyer and seller is that the buyer should have their money back, the chargeback mechanism will take care of implementing the remedy by reversing the payment transaction. In the case of transactions occurring via distributed ledger technologies (DLTs), like the Bitcoin blockchain, the software protocols used to record transactions could also implement automatic enforcement mechanisms for dispute settlement procedures.Footnote ²⁴ This would be possible, for example, through the operation of smart contracts that have a pre-authorisation or escrow function, which only transfer the amount of a transaction after a human or algorithmic arbiter has verified that there is no dispute or once a dispute has been resolved in favour of the recipient.

10.2.3 Designing ODR Mechanisms

The design of ODR mechanisms, including technological choices, ultimately depends on several different factors. The UNCITRAL Technical Notes on Online Dispute Resolution suggest that ODR mechanisms can comprise one or more of the following three stages: a first stage of technology-enabled negotiation between the parties, a second stage of facilitated settlement by appointing a neutral third party that mediates between the parties, and a third stage that could be, for example, binding arbitration. The difference between the stages is the level of involvement of a third party in mediating the dispute and the bindingness of the outcome. This three-stage model is followed by the English Online Civil Money Claims platform. Other ODR mechanisms comprise only the first stage, as, for example, the Brazilian consumer ODR platform Consumidor.gov.br,Footnote ²⁵ which enables consumers to file complaints against companies and companies to respond.Footnote ²⁶

Blomgren Amsler, Martinez, and Smith provide a more elaborate framework with six criteria to guide the design of dispute settlement systems,Footnote ²⁷ which also apply to ODR mechanisms.Footnote ²⁸ The six criteria are (i) goals, (ii) stakeholders, (iii) context and culture, (iv) structures and processes, (v) resources, and (vi) accountability. Before developing an ODR system, its goal(s) should be determined. At the most abstract level, the goal of an ODR system is to deliver justice in some of its normative forms, including justice in terms of outcome (substantive, distributive, utilitarian, social), process (access, voice, participation, accuracy, transparency, due process), and/or community (restorative, corrective, transitional, retributive, deterrent). In narrower terms, the goals of a specific ODR system will be tailored to a specific type of conflict that the system seeks to address. In e-commerce ODR systems, for example, such goals can be to enhance consumer trust or to encourage traders’ participation by providing speedy settlements. In pre-trial ODR systems, the goal can be to unburden the court’s docket while ensuring a fair and fast settlement of smaller claims between parties.

The second factor in designing an ODR system considers the stakeholders that create, use, and are affected by the ODR system, their background, power constellations, and respective resources. In the case of the EU ODR platform, for example, the stakeholders are consumers, companies, and national ODR bodies. Stakeholders should be consulted and actively involved in the process of designing an ODR system.

The third factor is context and culture. An ODR system has to be responsive to the context and culture in which it operates. An online marketplace’s ODR system, for example, needs to consider that it might be used regularly for handling cross-border disputes, among anonymous parties, at a high volume, and for a low value. Family ODR systems will need to consider that they are dealing with conflicts usually taking place in geographic proximity, between parties that know each other well, with a strong emotional component, and with a life-altering impact of the outcome on the parties and their affected children.

The fourth factor, process and structure, refers to the ODR’s procedural design to prevent, manage, and resolve disputes. They can involve one or more of the UNCITRAL ODR stages (negotiation, mediation, arbitration) mentioned above. The procedures should identify the ODR participants’ interests (including their fundamental rights, economic, social, and political interests) and be designed so as to ensure that these interests are recognised and furthered in the procedure.Footnote ²⁹

The fifth factor is resources. For an ODR system, these resources refer to the expenditures for running the system and individual procedures, personnel costs and training, and, importantly, maintenance of the IT infrastructure of the ODR platform. In this context, it is also of central importance who pays for these costs. Depending on who finances the system, it might be perceived as biased.Footnote ³⁰

Lastly, the sixth factor of accountability refers to periodic evaluations as to the ODR system’s functioning. Since an ODR system runs on an IT infrastructure, it will automatically generate a large volume of digital data that can be used for subsequent evaluation purposes if there are sufficient resources, cybersecurity, and privacy safeguards in place. The evaluation also entails asking whether stakeholders are actually making use of the system, whether neutral third parties (human or AI-based) involved are delivering unbiased and accurate services, and whether users of the system are satisfied with it. If the ODR system is sufficiently transparent, external parties will also be able to monitor and report on the effectiveness of the ODR system.

The six design criteria of Blomgren Amsler, Martinez, and Smith remain silent on how the remedies of dispute settlement systems or ODR should be designed, especially if procedures are not binding. Section 10.3 will present some examples of existing ODR systems to exemplify the choices made in relation to the six criteria as well as the types of remedies and redress that these various systems award.

10.3.1 ODR Mechanisms Set up by the Judiciary

As mentioned in the brief history of ODR above, one of the main examples of ODR mechanisms within the public sector is pre-trial ODR implemented by national judicial systems and administrative agencies. One of the most salient examples is the previously mentioned UK Online Civil Money Claims (OCMC) pilot, which is scheduled to run until November 2023.Footnote ³³ The pilot was launched after Lord Justice Briggs set out his vision for an online court in his 2016 Report,Footnote ³⁴ which followed the conclusion of the Civil Courts Structure Review. Given that accessing the justice system in the United Kingdom is unaffordable for many, the online court project aims at making litigation for small claims more accessible by reducing costs. Cost reductions are achieved through eliminating mandatory representation and the use of ADR before a claim is heard by a judge. The goals of the OCMC pilot are, on the one hand, to enhance access to justiceFootnote ³⁵ and, on the other hand, to help in the modernisation of the English civil court system.Footnote ³⁶

One of the notable features of the OCMC pilot is its extensive stakeholder consultation and piloting phase. Lord Justice Briggs had already conducted several consultations with the general public, judges, and other stakeholders in the process of writing his final report.Footnote ³⁷ The first phase of the OCMC pilot (August 2017 until March 2018)Footnote ³⁸ was conducted with only 1,400 selected participants before opening it to the general public. Furthermore, OCMC users are polled after submitting their claims to understand their level of satisfactionFootnote ³⁹ and to generate feedback for further improving the system.

As explained above, the OCMC platform offers in a first step an online interface to fill in the details of the claim, which can then be submitted via the platform. The fees for submitting a claim depend on the amount of the claim (between 4 to 8% of the amount claimed),Footnote ⁴⁰ and there is the possibility for applying for legal aid to cover the fees for low-income claimants. It also allows the respondent to file a response via the platform. So far, 378,000 claims have been submitted.Footnote ⁴¹ In a second stage, the platform offers mediation, which has been used in 9,560 cases to date and has led to a settlement at this stage in 50.4% of cases within 24 days on average. Settlements reached through mediation result in an agreement between the parties that is legally binding. At any point in the procedure, the claim can be referred to a judge. Overall, the OCMC has yielded positive results in the form of faster settlements, higher rates of mediation, and a faster procedure for judges to issue orders based on the digital file.Footnote ⁴²

The growth of pre-trial ODR has not only affected the United Kingdom but also other European states. In reaction to these developments, the Council of Europe issued a set of guidelines on online dispute resolution mechanisms in civil and administrative court proceedings in 2021.Footnote ⁴³ The guidelines are not binding but give guidance to states on how to design national pre-trial ODR proceedings that are in line with Article 6 (right to a fair trial) and Article 13 (right to a remedy) of the European Convention on Human Rights.Footnote ⁴⁴ The guidelines address a variety of issues around ODR procedures, including accessibility, due process, transparency, and cybersecurity. The guidelines would be a good reference for any potential ODR mechanism to offer redress to victims of fundamental rights violations by the EU.

10.3.2 The EU Consumer ODR Platform

When it comes to consumer law, the EU legislator has put a lot of emphasis on ADR and ODR as a solution to the widespread underenforcement of consumer law. The EU ODR Regulation,Footnote ⁴⁵ which came into effect in 2016, sets out the rules for setting up a European online platform that offers easy and low-cost dispute resolution through electronic means to achieve a higher level of consumer protection in the EU.

The procedure via the ODR platform is set by EU law, and the ODR platform itself is financed by the EU. The ODR platform is only a first step that allows consumers to submit a complaint and a trader to react. Subsequently, the procedure foresees that consumer and trader will agree on an accredited ADR body that will ultimately settle the dispute. Under the EU ADR Directive, accredited ADR bodies must be able to provide comparable remedies to consumers as those provided by courts.Footnote ⁴⁶ The goals of the EU ODR platform are to achieve a high level of consumer protection and to boost growth and integration of the EU’s digital single market.Footnote ⁴⁷ The ODR procedure is available to any consumer regarding complaints in respect of online purchases from a trader established in the EU.Footnote ⁴⁸ It thus follows the logic of early ODR that disputes arising online would be best resolved online.

The initial years of operation suggest that the vision for the ODR platform did not materialise as expected. Traders rarely responded to consumer complaints via the platform (80% of complaints were closed thirty days after submission due to no response from the trader)Footnote ⁴⁹ and only around 2% of claims submitted eventually reached an ADR body.Footnote ⁵⁰ This suggests that consumers and especially traders were insufficiently involved in the process of designing the platform to achieve the desired goals.Footnote ⁵¹

The Commission has published annual reports on the functioning of the ODR platforms and has run surveys among users of the platform. It eventually responded with a change to the platform’s design in July 2019, which now first offers consumers a self-test.Footnote ⁵² The self-test allows consumers various options to proceed with their complaints: either bilaterally contacting a trader (and negotiating), contacting a European Consumer Center, or contacting an ADR body through the platform. As a result, actual complaints submitted to the platform decreased significantly – the procedure offered by the platform was simply not the preferred option for consumers. In 2020, again, only 1% of complaints submitted via the platform reached an ADR body, while 20% managed to resolve their complaint in bilateral talks with traders.Footnote ⁵³ The EU Commission did react to the lack of engagement with the formal procedure offered by the ODR platform. It completely revised the design of the ODR platform and turned it more into an information portal that explains to consumers various options to proceed with their complaints. The ODR offer is now only a secondary feature of the website. This is an important example of an ODR mechanism that did not deliver on its promises. A more extensive stakeholder consultation, like that carried out for the OCMC, as well as a piloting phase could probably have helped design the EU consumer ODR platform in a more effective manner.

10.3.3 ODR Mechanisms to Redress Fundamental Rights Violations

The previous examples have shown that ODR has been used beyond a purely commercial setting: ODR has gained prominence in the judicial system and administrative agencies in various jurisdictions within and beyond the EU. Furthermore, the example of the EU ODR consumer platform is a precedent for an ODR platform that was set up by EU law. In both contexts, ODR has been used to settle small claims. In this section, I discuss two examples of ODR mechanisms that have been set up to address potential fundamental rights violations. One of these is mandated by EU law: the out-of-court dispute settlement systems to be set up in the framework of the DSA. The other has been set up as a self-regulatory measure: the Meta Oversight Board. Both deal with disputes that arise from fundamental rights violations occurring on the Internet, in particular in relation to freedom of speech.

The DSA is a regulatory framework that aims at reducing a variety of risks from illegal online content, including fundamental rights violations. It also regulates the procedures by which online platformsFootnote ⁵⁴ themselves remove illegal online content, referred to as online content moderation.Footnote ⁵⁵ The DSA foresees that online platforms should implement internal complaint mechanisms and allows for appeals to external dispute settlement bodies. The idea behind these mechanisms is that users will be better protected, and online platforms will have to take due regard of the rights and legitimate interests of users in their content moderation practices, including users’ fundamental rights. In particular, online platforms must explain how they will take due regard of the rights and legitimate interests of all parties in their content moderation activity, including users’ fundamental rights ‘such as the freedom of expression, freedom and pluralism of the media, and other fundamental rights and freedoms as enshrined in the Charter’.Footnote ⁵⁶

When a platform takes a decision to remove the content of a user, or refuses to remove content reported by a user, users must have access to an internal complaints mechanism to complain about the platform’s decision.Footnote ⁵⁷ In addition, users must have the possibility to appeal decisions made by a platform’s internal complaint handling system to a certified out-of-court dispute settlement body or online mediation instance.Footnote ⁵⁸ This can enhance access to justice, given that national court systems would be overburdened if they had to hear an appeal against every single content moderation decision and given that it might be too costly for most users to access a court in these cases.Footnote ⁵⁹ The Digital Services Coordinators of each Member State will have the power to certify dispute settlement bodies.Footnote ⁶⁰ In principle, these bodies will not have the power to impose binding solutions on the parties.Footnote ⁶¹

The certified dispute settlement bodies need to be independent from platform providers, and members should be remunerated in ways that are not linked to the outcome of the procedure.Footnote ⁶² The members should have the necessary expertise in relation to one or more areas of illegal online content or in relation to the application or enforcement of the terms of service of one or more platforms.Footnote ⁶³ This will likely also include expertise on fundamental rights, given that freedom of expression is often one interest at stake in content moderation decisions, and careful balancing against other interests is necessary. In terms of costs, the service of the out-of-court dispute settlement bodies should be free of charge for users.Footnote ⁶⁴

While the DSA out-of-court dispute settlement bodies are still in the process of being set up at the time of writing, there are some national precedents that show that such independent dispute settlement bodies can function without being perceived as biased towards platforms. In Germany, a similar ODR process was set up under the national law on illegal online content (Netzwerkdurchsetzungsgesetz – NetzDG) and is administered by an NGO, Freiwillige Selbstkontrolle Multimedia-Diensteanbieter (FSM). It is financed by a range of media and IT industry players, yet it has not been perceived as biased towards media players or online platforms.Footnote ⁶⁵

There is another example of an already functioning appeals body to review online content moderation by social media platforms: the Meta Oversight Board. In 2018 and 2019, Meta conducted a global consultation for setting up an independent appeals body, which would decide the most difficult issues and questions regarding its content moderation practices.Footnote ⁶⁶ In 2019, the charter of the Oversight Board was published,Footnote ⁶⁷ which lays down the relationship between Meta, the Oversight Board, and an independent trust that deals with the financing and staffing of the Oversight Board. Members are to be elected for a three-time renewable term of three years, and the trust decides on the compensation of the board members without the outcome of the decisions having an influence on the compensation. Furthermore, the charter of the Oversight Board lays down two procedures for appeals. On the one hand, the Board selects cases from appeals launched by users and. on the other, Meta can refer cases to the Board.Footnote ⁶⁸ When taking decisions, the Board should review the content in light of Facebook’s values and policies, as well as fundamental rights norms.Footnote ⁶⁹ Lastly, the Board’s decisions in relation to a specific piece of content are binding, while any policy recommendations are not binding on Facebook.Footnote ⁷⁰

According to the Charter of the Oversight Board, the size of the Board should be at least eleven members (likely to reach forty once fully staffed).Footnote ⁷¹ Article 1(2) of the Charter also specifies that members must have ‘a broad range of knowledge, competencies, diversity, and expertise’. They should have no conflicts of interest and must have demonstrated ‘experience at deliberating thoughtfully and as an open-minded contributor to a team’.Footnote ⁷² They should also have the skills to deliver reasoned opinions on the application of rules or policies, and they need to have familiarity with matters relating to digital content and governance, including ‘free expression, civic discourse, safety, privacy and technology’.Footnote ⁷³ When looking at professional background, two-thirds of the members are lawyers.Footnote ⁷⁴ Furthermore, two-thirds of the members have expertise in fundamental rights in general or in relation to specific fundamental rights fields (e.g., women’s rights or freedom of speech). In its decision practice, the Meta Oversight Board regularly goes beyond analysing Meta’s content moderation terms and uses international fundamental rights instruments in its interpretation.Footnote ⁷⁵

In June 2022, the Board published its first Annual Report, which gives insights into the kinds of fundamental rights violations that users seek remedy for or that pose particular challenges to content moderators.Footnote ⁷⁶ According to the report, the Board received more than 1.1 million appeals submitted by users. Most of these related to the Community Standards on Bullying and Harassment (32.4%), Violence and Incitement (28.9%), and Hate Speech (24.8%).

10.4.1 Legal Basis

To discuss the possibility of establishing an ODR mechanism to provide some form of redress requires first establishing whether there would be any legal basis in the EU Treaties for having such an ODR mechanism. The legal basis for the two legal instruments that most prominently support the establishment of ODR mechanisms, the ODR Regulation and the DSA, both have Article 114 of the Treaty on the Functioning of the European Union (TFEU) as their legal basis.Footnote ⁷⁷ Article 114 in conjunction with Article 26 TFEU gives the EU legislature the powers to enact legislation to further the function of the EU internal market. An ODR mechanism to redress the fundamental rights violations by EU bodies and agencies, however, would not have an internal market rationale. This means that Article 114 would not be an appropriate legal basis for establishing such a mechanism.

In general, there is no specific legal basis that would allow the EU to enact legislation to establish an ODR mechanism for addressing fundamental rights violations by its institutions. There are, however, some legal bases in the TFEU that could be used creatively to establish an ODR mechanism for violations of specific fundamental rights. One legal basis, for example, would be Article 19 TFEU, which allows the EU to enact legislation combatting discrimination. Another would be Article 16 TFEU on the right to data protection, which allows the EU to enact legislation regarding the processing of personal data by EU institutions. Such legislation could, in theory, also include a redress mechanism for victims of data protection violations. If we think about the type of fundamental rights violations by EU institutions that seem to be of most relevance at present, Article 78 TFEU might also be a starting point to initiate a legislative proposal. Article 78 TFEU is the legal basis for the EU’s common asylum policy and protection of third country nationals in need of protection. Under Article 78 (2) (d) TFEU, the European Parliament and Council acting by ordinary legislative procedure can adopt measures on ‘common procedures for the granting and withdrawing of uniform asylum or subsidiary protection status’. Potentially, an ODR mechanism could be implemented as a form of mechanism to address complaints by persons whose fundamental rights have been violated by the EU Agency for Asylum. As Schramm discusses in this volume,Footnote ⁷⁸ there is already a complaints mechanism in place under the European Asylum Policy Regulation implemented by the Agency’s Fundamental Rights Office.Footnote ⁷⁹ A similar mechanism exists under the Frontex Regulation.Footnote ⁸⁰ Neither of these mechanisms has been particularly visible or effective so farFootnote ⁸¹ but could potentially be enhanced through turning them into ODR mechanisms.

As an alternative, it could be possible to subsume a potential ODR mechanism under the umbrella of the European Ombudsman. Article 228 TFEU is the legal basis for the office of the Ombudsman and allows the Ombudsman to conduct inquiries into fundamental rights violations by EU institutions, bodies, and agencies upon her own motion or a complaint submitted to her. This would allow for one central ODR mechanism to complain about fundamental rights violations by any EU body, institution, or agency. The problem with Article 228 TFEU, however, is that only natural or legal persons residing in Europe can submit complaints. One would thus have to think of a network of NGOs that could submit complaints on behalf of third country citizens that have been victims of fundamental rights violations by EU bodies and agencies to the Ombudsman.

10.4.2 Design Options for an EU ODR Fundamental Rights Redress Mechanism

If the obstacles relating to the legal basis could be overcome, the process of designing an ODR mechanism subsumed under the Ombudsman’s office or other specialised EU agencies could fruitfully draw on the six design criteria for dispute settlement systems and existing ODR mechanisms.

The first step would be to establish the goals of the ODR mechanism. These seem relatively clear: to enable some form of access to redress for victims of fundamental rights violations by EU agencies and bodies, as well as fostering and demonstrating the EU’s commitment to international fundamental rights regimes.

The second step to establish such a mechanism would be to extensively engage with stakeholders. This could take the form of a stakeholder consultation, which involves victims of fundamental rights violations by the EU, as well as the organisations that give these individuals a voice, like civil society organisations. As seen in the examples above of the English OCMC and the Meta Oversight Board, several rounds of consultations with a variety of stakeholders helped to establish ODR mechanisms that are being used by stakeholders and that enjoy a necessary level of trust to function.

Third, it is essential to consider the context and culture of an EU ODR mechanism to redress fundamental rights violations. One important part of context and culture would be to set up the ODR mechanism in a truly accessible way for victims. Especially when thinking about victims outside the EU borders, access becomes a challenge due to physical, economic, and language barriers. As a starting point, the ODR platform would need to be available in more languages than the current EU languages.Footnote ⁸² Furthermore, it should have a user interface that can be easily accessed through all types of electronic devices, including mobile devices. Especially when testing prototypes of the ODR mechanism, it should be ensured that the users that are involved in tests are in a comparable situation in terms of IT literacy,Footnote ⁸³ literacy in general, language, and socio-economic background. This would avoid similar pitfalls as with the EU consumer ODR platform, which was not designed in a sufficiently user-centric manner to successfully take off.

Fourth, decisions regarding the structure and procedure of the ODR mechanism will need to be made. The procedural rules could be guided by the principles that can be found in the 2021 guidelines by the Council of Europe. In principle, the procedure would commence with the submission of a complaint – the simple registration of these complaints could already create a very important database on alleged fundamental rights violations by the EU bodies and agencies. In a second stage, a committee of members appointed by the Ombudsman or Fundamental Rights Officers could decide on the complaint and whether there has indeed been a fundamental rights violation attributable to EU agencies or bodies. This committee should be staffed by individuals that have a professional or academic background in fundamental rights law. The example of selection criteria for members sitting on the Article 21 DSA dispute settlement bodies and the Meta Oversight Board show that such criteria can be tailored to the specific context in which the ODR platform operates. Once a decision has been taken by the committee, the Ombudsman or Fundamental Rights Officers could resort to the forms of redress they can award, which are, admittedly, limited.

The complaints mechanisms administered by the Fundamental Rights Officers at the EU Asylum Agency and Frontex do not have any effective remedies available other than forwarding the message that there has been a fundamental rights violation to the executive director or national authorities.Footnote ⁸⁴ The Ombudsman seems at least to have political influence to persuade other EU institutions.Footnote ⁸⁵ She could address the agency or body that has committed the fundamental rights violation and ask for one of the internationally recognised forms of redress (cessation of the violation, damages, reparation, acceptance of responsibility, and preventing future violations) or try to influence the political process.Footnote ⁸⁶

As neither the Ombudsman nor the Fundamental Rights Officers have any enforcement powers, the entire ODR procedure would not be binding. The fact that the procedure is not binding, however, does not completely eliminate its effectiveness. In the case of the Meta Oversight Board, for example, decisions do not bind Meta regarding its overall policies but only in respect of the specific case. Nonetheless, the Oversight Board can make recommendations and has used its Annual Report to show whether Meta has made any progress on its recommendations. In this sense, even a non-binding mechanism can establish some form of accountability through transparency and reporting.

Fifth, there is the question of the resources for funding the ODR procedure. In the examples given in this chapter, all ODR platforms are funded by the fees paid by one or both parties. In the case of an EU ODR mechanism to address fundamental rights violations by EU Institutions, the mechanism would likely have to be funded by the EU. After all, it would be an accountability mechanism for its own institutions and agencies. Placing fees on the victims would not be equitable. As in several other ODR systems discussed above, the weaker party usually gets access for free to the ODR system. If the mechanism were to be administered by the Ombudsman’s Office or the Fundamental Rights Officers, they should receive an additional budget for it.Footnote ⁸⁷

Lastly, to create a sustainable and well-functioning ODR system, periodic evaluations would be necessary. Periodic evaluations could be facilitated by the publication of an annual report by the Ombudsman or Fundamental Rights Officers, similar to the reports published by the EU Commission in relation to the consumer ODR platform, the Meta Oversight Board, or the future DSA out-of-court dispute settlement bodies. This would allow for public scrutiny and for drawing lessons from the operation of the mechanism and improving it in the future.