Journal article

Barriers to Implementing Transparency in EU Institutions

Pages 47 to 61

Cite this article


  • Novak, S.
(2024). Barriers to Implementing Transparency in Eu Institutions. Action publique. Recherche et pratiques, Special Issue #2(1), 47-61. https://doi.org/10.3917/aprp.018.0028.

  • Novak, Stéphanie.
« Barriers to Implementing Transparency in EU Institutions ». Action publique. Recherche et pratiques, 2024/1 Special Issue #2, 2024. p.47-61. CAIRN.INFO, shs.cairn.info/journal-action-publique-recherche-et-pratiques-2024-1-page-47?lang=en.

  • NOVAK, Stéphanie,
2024. Barriers to Implementing Transparency in EU Institutions. Action publique. Recherche et pratiques, 2024/1 Special Issue #2, p.47-61. DOI : 10.3917/aprp.018.0028. URL : https://shs.cairn.info/journal-action-publique-recherche-et-pratiques-2024-1-page-47?lang=en.

https://doi.org/10.3917/aprp.018.0028


Notes

  • [1]
    Hereinafter the Council.
  • [2]
    Hereinafter the Court of Justice.
  • [3]
    On the unintended consequences of transparency, see Erkkilä, 2012.
  • [4]
    On this matter see Flinn and Jones, 2009.
  • [5]
    For more on the treatment of sensitive documents, refer to Article 9 of Regulation (EC) No 1049/2001.
  • [6]
    From 2006 to 2022, we conducted over 100 interviews with Council members. These interviews were held to learn about the institution’s decision-making practices and the impacts of transparency policies on decision making. See for example Novak, 2011.
  • [7]
    Interview on the impacts of the Treaty of Lisbon on the decision-making of the Council, the Council Secretariat, Brussels, November 2012.
  • [8]
    See for example De Capitani v Parliament, 2018.
  • [9]
    See for example ClientEarth v Commission, 2018, § 13.
  • [10]
    As of June 2023, the date of publication of this paper.
  • [11]
    The Qatargate scandal erupted in December 2022, involving current and former European Parliament members who allegedly received money from Morocco and Qatar in an attempt to influence parliamentary decisions.
  • [12]
    This case was brought by the non-governmental organisation Access Info Europe against the Council. In 2008, as part of its work to raise the level of institutional transparency, Access Info Europe applied to the Council for access to a note sent by its Secretariat General to the Working Party on Information set up by the Council, concerning the proposal for a regulation of the European Parliament and of the Council regarding public access to European Parliament, Council and Commission documents. That document contains the proposals for amendments and for re-drafting entered by a number of Member States at the meeting of the working party on 25 November 2008 (Access Info Europe v Council, § 6). The Council sent this document to the organisation, with the identity of the Member States mentioned in the note redacted so that there was no way of telling what position had been defended by which Member State. The Council justified this partial access to the document by arguing that disclosing the identity of the Member States mentioned would seriously undermine the decision-making process. In 2011, Access Info Europe challenged this decision before the General Court which ruled in favour of the pro-transparency organisation.

1 Since the 1990s, EU institutions have developed policies that are designed to make their work and decision-making processes more accessible to the public. These policies are introduced against a historical backdrop in which transparency has gradually become a “global” norm (Peters, 2013). In this respect, several countries have introduced access to information acts over the past 60 or so years. EU transparency policies developed in the last few decades are the product of events and factors such as Denmark’s rejection of the Maastricht Treaty in 1992, MEPs being tasked with acquiring more information on the Council of the European Union’s [1] work, a string of corruption scandals, and decisions made by the Court of Justice of the European Union [2] and the European Ombudsman that are often in support of increasingly open institutions. Against this backdrop, EU institutions present access to information policies as solutions to a European “democratic deficit” and transparency as a tool to ensure greater accountability among decision-makers, increase citizens’ trust in their representatives and bolster the legitimacy of the institutions.

2 However, there are a myriad of ways to coordinate access to information, and reforms on the matter have not necessarily resulted in the intended outcomes. This paper will provide an overview of the main regulatory changes and their impacts, and also will attempt to identify the barriers facing EU transparency policies. In the first section, we analyse the notion of transparency. Next, we explore developments on the transparency front in EU institutions and their consequences – first and foremost taking into account Regulation (EC) No 1049/2001 regarding public access to European Parliament, Council and Commission documents – followed by the barriers to transparency in legislative institutions. In the final section, we examine the decisions made by the Court of Justice and the European Ombudsman in relation to access to documents and discussions on the limits and barriers to transparency.

Transparency : an in-vogue quality…

3 Transparency is a coveted quality to the extent that one has the impression that it is not just a means to an end – that of ensuring democratic governance – but an end in itself. It is often associated with the development of new technologies, but the principle of openness underpinning transparency considerably predates this. Enlightenment philosophers – namely Kant, Rousseau and Bentham – played a key role in the emergence of this principle, upheld to prevent abuse of power (Meijer, 2015). Very often regarded as a cure for the various ills of modern politics – corruption, low citizen participation in public forums, undue influence exerted by interest groups and scant information about decision-making processes, which makes it difficult to determine which positions the stakeholders have taken (Hood and Heald, 2006) – transparency is commonly seen as a key component of the democratic accountability of representatives. This notion of democratic accountability (Przeworski, Stokes and Manin, 1999) relates to the need for citizens to be informed of the positions backed by their representatives during political debates. This information in turn will allow them to make an informed choice during elections. Transparency can therefore be defined as “the availability of information about an actor that allows other actors to monitor the workings or performance of the first actor” (Meijer, 2013, p. 430).

…serving various purposes and with different tools used to implement it

4 Although transparency relates to public access to information, it may have various distinctive purposes and be implemented using a variety of tools. According to Grimmelikhuijsen et al., (2013, p. 576), the object of transparency can be decision-making processes (e.g. parliamentary debates), the content of the policies adopted (what problems are the policies intended to solve and how are the policies implemented?) and the outcomes or effects of these policies. Mansbridge (2009) makes a distinction between transparency in process, relating to information on the decision-making process, and transparency in rationale, which relates to representatives explaining to citizens the content of the policies they have adopted. She draws a link between transparency in process and democratic accountability, considered an opportunity for citizens to punish, with their ballot, elected officials for having backed positions or acts that they disagree with. Transparency in rationale, which focuses on the explanation given by representatives for their political decisions and not on the monitoring of their actions by citizens, is linked to narrative democratic accountability. Given that the relationship between transparency and trust is often brought to the forefront for discussion, it should be noted that these two forms of transparency have a differing relationship with trust. Transparency in rationale and narrative democratic accountability assume a high level of trust among citizens in their representatives, while transparency in process and democratic accountability – based on the possibility of punishing elected officials – are considered as standing in opposition to the tools employed to build trust. Furthermore, the definition of transparency used by the institutions is broader than the disclosure of information on decisions, including also information about the career and income of politicians (elected members of the European Parliament and European Commission) and EU civil servants, so as to mitigate the risk of corruption or conflict of interest. It also covers interest groups that must detail their objectives, resources and staff when they are listed on the EU transparency register (2021 Transparency Register).

5 There are various tools used to grant the public access to information. If the object of transparency is the decision-making process, citizens can attend debates in person (e.g., during parliamentary sessions) or remotely (e.g. in the case of the Council, within which ministers of EU Member States negotiate EU legislation). However, access to the debates may only be granted once they are over, in the form of minutes. While attending debates encourages citizen participation, reading minutes allows citizens to acquire information but not to influence the procedure underway. While the institutions do not grant access to their debates, instead favouring transparency in their policies (this is for example the case for the European Central Bank [ECB], Curtin, 2017) and/or the outcomes thereof, a variety of new tools, such as press releases and annual reports, are used to disclose information to the public.

6 To help with the analysis of the tools coordinating public access to information, a distinction should also be made between passive and active transparency. Passive transparency is when citizens may request a document – the procedures for exercising this right are generally set out in freedom of information laws. On the other hand, active transparency is when the institutions directly grant access to documents, a process facilitated by new information technologies (de Terwangne, 2004). Generally speaking, the European Union has shifted from passive to active transparency.

7 The concept of transparency can also be distinguished from that of publicity (Naurin, 2006). Transparency means that the information is available but that citizens may not necessarily be able to comprehend it (e.g. if an institution were to publish all its documentation online without providing the tools needed to understand its nature, the manner in which it is produced and drafted, and also how to use the search function in the digital archives). Publicity has the implication that the information is structured in such a way that citizens can use it. Transparency is required to ensure that permission to disclose information is granted carefully, but it is this principle of openness that drives democratic accountability.

8 Although transparency is a democratic imperative, a number of empirical studies over the past 20 years have highlighted the costs of transparency and its potential for unintended or even counterproductive consequences. [3] Some of this work shows that transparency’s impacts vary depending on the cultural and/or institutional context. For example, comparative studies posit that transparency does not necessarily build citizens’ trust in their government, and that it can even have the opposite effect (Grimmelikhuijsen et al. [2013] analyse this topic through a comparison of the Netherlands with South Korea). Some research has also examined transparency’s impact on negotiations; negotiators generally claim that secrecy is needed for such proceedings (Stasavage, 2006; Novak and Hillebrandt, 2020). This subject has fuelled debates on what limits to place on transparency. Another critical viewpoint is that the potential to publicly disclose decision-making processes is overestimated : within EU institutions, it is said to be a commonplace practice for negotiators, when subject to disclosure obligations, to push back the “real” debates to non-public spaces. This consequence raises issues in that it implies that there is only a veneer of transparency. We continue our discussion on this topic and the friction between negotiation and transparency in later sections.

Regulation (EC) No 1049/2001 regarding public access to European Parliament, Council and Commission documents and the limits of transparency

The increasing importance of the principle of transparency

9 Since the Maastricht Treaty (1992) and the Laeken Declaration (European Council, 2001), the principle of transparency – also known as the principle of openness – has gradually become more and more important within EU institutions. The right of access to documents of the Union’s institutions is referred to in Article 15 of the Treaty on the Functioning of the European Union. The arrangements for exercising this right are set out in Regulation (EC) No 1049/2001, which marked a major step for EU transparency policy. This regulation concerns access to European Parliament, Council and Commission documents, and other EU institutions have produced their own transparency rules. Regulation (EC) No 1049/2001 was adopted following negotiations between (i) stakeholders in favour of an ambitious regulation – notably the European Parliament, which had an interest in acquiring more information about the Council’s work, along with a coalition formed of Finland, the Netherlands, Sweden and Denmark, countries where transparency had been a long-standing political principle – and (ii) Member States, particularly Germany, France and the United Kingdom, more inclined to want to safeguard the confidentiality of certain documents and which feared that this new regulation would require disclosing documents pertaining to national security issues, a concern heightened by their closer ties with NATO (Bjurulf and Elgström, 2004). The regulation, adopted in 2001, is a compromise factoring in these different positions, as is often the case for the European Union. A few years later, the Commission began work on revising the regulation to address the Parliament’s requests. The adoption of the Treaty of Lisbon and its Article 15 regarding the right of access to information especially made it necessary to adapt the regulation so that it could cover all institutions (Maes, 2010). In 2008, the European Commission adopted a proposed revision of the regulation. However, negotiations have come to a standstill as a result of the diverging opinions of the Parliament, Council and Commission. The Parliament criticises the Commission’s proposal for using a definition of “document” that is far too narrow and for not encompassing enough institutions. Some Member States would prefer that certain documents (e.g. those related to state aid and competition policy) not be subject to the revised regulation (European Parliament, 2013 and 2023).

Legislative documents, accessible by default

10 Pursuant to Regulation (EC) No 1049/2001, legislative documents must be accessible by default. These are defined as “documents drawn up or received in the course of procedures for the adoption of acts which are legally binding in or for the Member States” (Article 12(2)). The regulation outlines the circumstances in which the institutions may or may not grant access to a given document. These exceptions cover documents where disclosure would undermine the protection of (i) the public interest as regards public security, defence and military matters, international relations, the financial, monetary or economic policy of the Community or a Member State, and (ii) privacy and the integrity of the individual. Furthermore, the institutions shall refuse access to a document where disclosure would undermine the protection of commercial interests, court proceedings and legal advice, and the purpose of inspections, investigations and audits (Article 4(2)). Article 4(3) stipulates that public access to a document shall be refused if disclosure of the document would seriously undermine the institution’s decision-making process. This exception applies to ongoing decision-making processes, but may also be invoked after a process has ended. This reasoning is to some extent detailed in the recitals of the regulation : “Wider access should be granted to documents in cases where the institutions are acting in their legislative capacity, including under delegated powers, while at the same time preserving the effectiveness of the institutions’ decision-making process. Such documents should be made directly accessible to the greatest possible extent”.

11 The institutions have 15 days to respond to any applications for access received. In the event of a refusal of an application, the applicant has the right to make a confirmatory application. If the institution once again refuses the application, the applicant has the right to refer to the Court of Justice or the European Ombudsman. All refusal letters from the institutions are publicly available. An analysis of replies to confirmatory applications has shown that in addition to the specific interests detailed in Article 4(2), the institutions commonly invoke the protection of the decision-making process pursuant to Article 4(3) (Novak and Hillebrandt, 2020). More specifically, the arguments given are as follows : the necessity of negotiation ; the act of publishing a document would risk bringing negotiations to a standstill as representatives would be unable to explain to citizens that they had to change their position during the process; the risk that trust breaks down between the decision-makers to the extent that open exchanges are no longer possible; the risk that stakeholders no longer want to leave a paper trail of their work, which would lead to a purge of the archives; [4] and the risk of external pressure if the negotiation process is still under way. An analysis of the methods employed by the Court of Justice and the European Ombudsman to address these arguments when considering refusals to access documents is presented below. For the moment, it should be noted that most arguments used under Article 4(3) relate not just to the decision-making process in general but more specifically to negotiations. Arguments relate on the one hand to the possibility that a decision-maker’s position can change during the process, which is a fundamental characteristic of negotiations and compromise that requires stakeholders to make concessions, and on the other to the difference between a closed-door meeting, in which stakeholders can hold honest discussions, and a public session which does not allow for real discussions. This reference to negotiations combined with a requirement for confidentiality proves problematic in a legislative context, i.e. when it is a matter of adopting rules and regulations that will be applicable to all Member States. Although this tug of war between the necessity of negotiation and public access to information is factored into Regulation (EC) No 1049/2021 and into the decisions of the Court of Justice and the European Ombudsman, this very real problem still needs to be resolved, and Article 4(3) serves to cover cases in which transparency could jeopardise the negotiation process. Furthermore, the issue is not just that the institutions refuse to grant access to legislative documents on the pretext of negotiations, but that the reasons given in this case are often opaque – citing the “ability to negotiate” (European Ombudsman, 2016) is vague and understood only by insiders for the most part. This in particular raises the issue of legitimate limits on transparency.

12 One would think that refusals to disclose documents pursuant to Article 4(3) relate primarily to the Council : its work is effectively carried out for the most part by permanent representatives and deputy permanent representatives of Member States (Council, 2009, Article 19), who are generally career diplomats. However, negotiations are carried out beyond the intergovernmental sphere; as is detailed below, negotiations are characteristic of relations between the institutions and the decisions made within them. The Commission and the Parliament also refuse to disclose certain documents citing reasons relating to the ability to negotiate.

Barriers to transparency in the legislative procedure

13 Two main aspects of the legislative procedure have been criticised for a lack of transparency : a) deliberations within the Council and b) the trilogues, tripartite closed-door meetings during which representatives of the Commission, Council and Parliament work out compromise legislation.

The Council : procedure transparency and disclosure

14 The Council is known to be the least transparent institution of the institutional “trio” formed with the Commission and the Parliament. However, this institution is gradually opening up. To some extent, the Council’s transparency is centred on the procedure (Mansbridge, 2009), but it will be evident that its impacts on democratic accountability are limited and that it is often used for political spin.

15 Since 1994, Council votes on legislative acts have been made public. Some of the plenary sessions – the initial debate on the legislative proposals and the vote to adopt legislative acts, as well as a number of intermediary debates, based on the decisions of the six-month presidency – can be streamed on the Council website (Council, 2009, Article 9). Furthermore, the agenda of the Permanent Representatives Committee (Coreper) and the Council, in addition to the Council minutes, are published directly on the institution’s website.

16 While legislative work must be made public by default, access to the procedure is still restricted. Firstly, the minutes of Coreper are not available despite the fact that in most cases this is where compromises are reached : pursuant to the Rules of Procedure of the Council, “Coreper shall endeavour to reach agreement at its level to be submitted to the Council for adoption” (Council, 2009, Article 19). Permanent representatives receive instructions from their ministers, but have a degree of leeway for negotiating agreements. This has led to the German wordplay of Vertreter-Verräter (meaning “representative” and “traitor” respectively) being coined within Coreper. Only ministers have the right to vote, and they publicly approve legislation once the permanent representatives have negotiated agreements behind closed doors. The minutes of working groups are also not published. These groups comprise experts who are nominated by Member States and the first to receive the Commission’s legislative proposals. They are tasked with discussing any technical difficulties before submitting amended proposals to Coreper. In 2020, Emilio De Capitani submitted a request for access to certain documents exchanged within the “Company Law” working group and was refused access by the Council (De Capitani v Council 2023). The EU official then embarked on an endeavour to increase the amount of public information concerning the decision-making process. A few years prior, he had requested access to documents relating to trilogues (De Capitani v Parliament, 2018, see below). In the case of the Council, it should also be noted that some documents are often filed in the register once the Council Secretariat has redacted certain information (e.g. the identity of Member States defending certain positions). For other documents, their existence is noted in the register but they cannot be viewed. [5] What is more, making the ministers’ votes on adopted legislative acts public would only be the tip of the iceberg. Indeed, votes against legislation are not published; to our knowledge, such votes rarely are cast because the Presidency of the Council would not ask delegations to vote if it believes that a qualified majority (or a unanimous vote, in cases where qualified majority voting does not apply) may not be reached. An agreement by qualified majority is also generally reached behind closed doors during a Coreper meeting, while the official public vote is held a few weeks later. When representatives know that a qualified majority in favour of an act has been reached, they may choose to not reveal their feelings of scepticism or discontent with a vote against or abstention. Based on our interviews, [6] it seems that they often feel that there is no point to the voting, and that it is even counterproductive, as in their view it draws attention from the media who would paint it as a failure, even if the Council is known to decide “by consensus”. On average, roughly 80% of legislative acts are adopted with no votes against or abstentions, as the delegations generally follow the consensus. Voting behaviour varies from one Member State to another. Some Member States do all they can to avoid casting votes against or abstentions. For certain Member States, such as Denmark and Sweden, the vote cast by ministers reflects the respective national parliament’s position : if a parliamentary majority is reached to reject a legislative proposal, the minister concerned cannot participate in the compromise. This loyalty to respective parliaments helps to make the procedure more transparent. Furthermore, votes against and abstentions declared publicly represent tools for communicating with certain interest groups : a delegation may vote in such a way to send a message to these groups, demonstrating that it has attempted to protect national interests, albeit in vain (Novak, 2011).

17 In the case of public votes in the Council, Curtin’s (2017) observation about the ECB is applicable : it is the institution which has control over transparency as a communication tool – since it decides how it will be implemented – so that transparency does not control the institution. According to Curtin, while the ECB has transformed into a powerful supranational institution as a result of the economic and financial crisis of the 2010s, it suffers from a deficit in democratic accountability. As it is not subject to Regulation (EC) No 1049/2001, the ECB has adopted its own transparency practices and reports on its work through press releases. However, this kind of transparency in rationale adopted by the ECB is more a case of disclosing than sharing information so as to uphold its responsibilities to citizens and national parliaments.

18 Cases of voting in the Council and the ECB show how transparency policies, intended to ensure the accountability of representatives, can actually be fashioned by decision-makers to into a tool for political spin. This risk emerges if the institutions decide themselves when information can be disclosed – whether it be votes, debates or documents – and if there is no third-party agent ensuring the proper implementation of transparency. Council debates should also be mentioned in this respect. Excluding debates that are required to be public, i.e. initial debates on legislative proposals, the six-month presidency is free to choose when to make sessions open. It may use the public dimension of debates to push through legislation and deter a delegation from expressing any objection without providing a strong argument, or alternatively use other strategies to do so. If there is no third-party agent involved, the public nature of debates could also encourage decision-makers to move the decision making to behind the scenes or during lunchtime – in an interview, [7] a member of the Council Secretariat compared the Council’s public debates to Potemkin villages. In certain respects, the European Ombudsman and the Court of Justice ensure the placement of third-party agents, albeit downstream of the procedure for the Court and with no obligation to do so for the European Ombudsman. However, the loyalty binding ministers to their respective parliaments in certain Member States would suggest that national parliaments could to some extent assume this role of third-party agent.

19 When stakeholders see that representatives are avoiding public debate in the decision-making process, they often argue that negotiations in public do not work. As noted previously, the friction between negotiation and transparency is not only raised by Council members. The Parliament [8] and the Commission [9] have also refused access to legislative documents on the grounds that their publication would jeopardise ongoing negotiations. This last point has prompted us to analyse the issue of closed-door trilogues.

Trilogues and four-column documents

20 The Treaty of Lisbon increased the powers of the European Parliament by extending the ordinary legislative procedure, which is currently applied in most cases. Consequently, the Council and the Parliament have the same influence and must together approve the Commission’s proposals so that they are adopted. This shift is partly the result of the “deficit in democracy”. As the Parliament is an institution whose work is more publicly available than that of the Council, it would be expected that this institutional reform would increase public access to legislative work. In her decision on the Council’s transparency, the European Ombudsman also observed that the Parliament provides more information on its work and that the Council should base its approach on this (European Ombudsman, 2018, Preliminary comments, § 9). However, under the ordinary legislative procedure, closed-door negotiations between the Commission, the Council and the Parliament have become commonplace. EU institutions work out compromises that will then be formally adopted by the Council and the Parliament (Curtin and Leino, 2017; Brandsma, 2019). In other words, alongside the progress made in transparency regulations, non-public negotiation forums have surged in number.

21 During trilogues, the stakeholders use “four-column” documents. The Commission’s legislative proposal is given in the first column, the positions of the Parliament and the Council are set out in the second and third columns, while the fourth column is reserved for the inter-institutional compromise proposal. When the legislative procedure is under way, the institutions do not proactively publish these documents. In 2015, the European Ombudsman opened an inquiry into the transparency of trilogues. One of its findings was that these meetings, that are a vital stage of the procedure, are not transparent enough. She therefore recommended that a user-friendly database be created in which the four-column documents would be classified (European Ombudsman, 2016, § 65). These documents are required to understand the political accountability of the institutions, since they indicate their positions on the articles of legislative texts, the amendments they have requested, and to what extent the final compromise text differs from their requests. According to the information at our disposal, there are plans to set up this database as per the European Ombudsman’s recommendation in 2016, but there is still no sign of it [10] (see also Leino-Sandberg, 2022, p. 4).

The key issue: when should access to documents be granted?

22 A lack of transparency in the legislative procedure is widely acknowledged by experts (Curtin and Leino, 2017). It is the subject of criticism from civil society, particularly from non-governmental organisations and citizens who have taken it upon themselves to challenge the refusal of access to documents, as well as by the European Ombudsman and, indirectly, by certain judgments of the General Court and the Court of Justice. A topic featuring in this debate is the moment when legislative documents should be published, which, in the words of the European Ombudsman, is the “key issue” (2016, § 53) : when the procedure is under way – a factor which would encourage public participation in debates – or once the procedure is closed – which would allow citizens to acquire the information but not take part. While the European Ombudsman, after her 2015 inquiry into trilogues, recommended publishing four-column documents ex post so as to maintain the “capacity to negotiate” (European Ombudsman, 2018, § 30), others call for access to documents during the course of negotiations. For example, in his appeal before the General Court, Emilio De Capitani stated that he had only received the four-column documents he requested from the Parliament once the procedure had finished, even though he had expected to receive them when the procedure was still under way (De Capitani v Parliament, 2018). Similarly, he challenged the fact that the Council had only granted him access to the legislative documents exchanged within the “Company Law” working group once an agreement was reached by the Council and the Parliament. As a result, he was not able to provide the public with the relevant information and prompt a debate (De Capitani v Council, 2023, § 16). In March 2022, 40 civil-society organisations and trade unions challenged the insufficient information provided about the trilogues on the Digital Markets Act (Transparency International EU, 2022 ; see also Statewatch, 2023) and particularly the fact that a document published without the fourth-column information (i.e. the compromise reached by the institutions) was presented as the most up-to-date version when that was not the case. These organisations alleged that it was all the more necessary to inform citizens about the trilogues under way as Big Tech had been exerting intense pressure on negotiators. Curtin and Leino posit that four-column documents should be actively published “in real time” since they relate to the legislative procedure and the most powerful interest groups have access to information on trilogues (2017, p. 1710). In terms of access to information, this situation puts citizens at a disadvantage and creates what Curtin and Leino ironically refer to as “highly selective transparency” (2017, p. 1693; Leino-Sandberg, 2022, p. 12).

The transparency register

23 As mentioned before, access to information concerns distinct elements of political life and is coordinated using various tools. Another important development in transparency policy was the creation in 2011 of the Transparency Register operated jointly by the Parliament and the Commission in which interest groups must be listed if they wish to carry out certain institution-related activities, such as being involved in a Parliament public hearing. The register contains information on, for example, the interest groups represented at EU level, their structure, their financial resources and their staff. It is designed to promote more transparent dialogue between the institutions and civil society groups, and is based on a very broad definition of lobbying since a wide variety of organisations can be listed, from multinationals and consulting firms to non-governmental organisations. When an interest group is listed in the register, it undertakes to observe certain rules of conduct in their interactions with EU institutions. Since 2021, the register has been jointly operated by the Commission, the Parliament and the Council, and organisations’ inclusion in the register has been required to carry out interest representation activities, reflecting one of the European Ombudsman’s recommendations (2016, § 28 ; 2021 Transparency Register). However, even when registration was optional, a considerable number of organisations had joined the register. It appears that the register’s success can be attributed to a reputational concern for organisations: the register may be considered by interest representatives as a means of legitimising their work. Furthermore, given the large number of organisations listed in the register, not appearing in it would leave organisations in a difficult position (Nastase and Muurmans, 2020). Nevertheless, as the Qatargate [11] scandal has shown, ineffective monitoring of external pressure on the institutions is an issue which is far from being resolved. It could be argued that it is all the more difficult to tackle this problem as the European Commission promotes a kind of consultative democracy in which lobbying holds a central position, which has subsequently been sanctioned by the Transparency Register (Robert, 2018). What is more, the actual impacts of the register on European democracy have yet to rear their head. For example, it seems that citizens use this public information to a very limited extent, and also that the European Commission believes that this register is a tool for experts and not the wider public (Robert, 2018).

Court of Justice and European Ombudsman responses to justifications for limited transparency

The impact of their decisions on transparency policy implementation

24 Transparency policies have a limited reach in terms of democratic accountability, which is partly due to there being no third-party agents to oversee their upstream implementation. When the Court of Justice hears cases in which an institution has not granted access to documents and rules in favour of greater transparency (see for example Hautala v Council, 2001; Sweden and Turco v Council, 2008; Access Info Europe v Council, 2011; Council v Access Info Europe, 2013; De Capitani v Parliament, 2018; De Capitani v Council, 2023), its decisions will influence how institutions will act in the future. The judgments of the General Court in the De Capitani v Parliament (2018) and De Capitani v Council (2023) cases are key to increasing public access to information in two stages of the legislative procedure characterised by closed-door sessions, namely trilogues and working group negotiations within the Council. The work of the European Ombudsman has an influence on at least two levels : she can inquire on single cases, as she did recently concerning the Council’s refusal to grant access to legislative documents related to the Digital Markets Act (2022); and she can also issue recommendations following broader inquiries, as was the case for the trilogue transparency case (2016) and the Council transparency case (2018). However, the institutions do not necessarily follow these recommendations – or at least not all of them – and if they do, not always in a prompt manner (e.g. the online database for four-column documents). In 2018 for example, the European Ombudsman published recommendations on the Council’s transparency and the institution failed to reply within the legal time limit of three months, which meant that she closed the case citing maladministration (2018, p. 1).

Countering arguments supporting limited transparency

25 The positions adopted by the two transparency “watchdogs”, the Court of Justice and the European Ombudsman (Hillebrandt and Leino, 2021), help to keep public access to information on the institutions’ agenda and reveal the limits of institutional transparency in spite of Regulation (EC) No 1049/2001. In addition, their decisions analyse certain arguments used by the institutions to maintain a negotiation “space” (European Ombudsman, 2016), whether these arguments are used in letters of refusal to grant access to documents, during negotiations on transparency rules and in inter-institutional relations. The Court of Justice and the European Ombudsman have for example challenged the argument that publishing documents when the decision-making process is under way would make negotiations more rigid (see for example Council, 2014) as representatives would be unable to explain to the public why they changed their positions. In response to this reasoning, the Court of Justice and the European Ombudsman stated that citizens are absolutely able to understand changes in position (Access Info Europe v Council 2011, § 69 ; De Capitani v Parliament, 2018, § 102; De Capitani v Council 2023, § 79 ; European Ombudsman, 2016). This counterargument serves to challenge a seemingly common-sense theory that is regularly cited by the institutions. However, it should be noted that this theory is shaky, in that determining whether to grant access to information on the basis of conjecture regarding the (in)capacity of citizens means using an undetermined variable as a criterion. What should be taken into account is less the capacity of citizens and the impact this would have on decision-makers’ behaviour – which varies depending on the context – and more the public’s right to be informed. Furthermore, if the (in)capacity of citizens is to be cited, all that it would take is a published document sparking outrage and bringing negotiations to a standstill for advocates of confidentiality to claim once again that negotiation and transparency are incompatible (Novak, to be published).

26 The institutions sometimes claim that a procedure that is too public would open the door for external pressure. However, non-governmental organisations and experts note that those participating in trilogues are pressured by lobbies, a factor that helped the European Ombudsman form her decision on the transparency of trilogues. She flipped this argument around by noting that it is valid for groups to try to influence the procedure; in her opinion, the true problem is that in a non-transparent environment only the groups with the most contacts and resources have the means to exert any influence. This is why opening up the procedure to all is necessary (2016, § 26).

Negotiation and transparency : can you have both?

27 Generally speaking, the “capacity to negotiate” (European Ombudsman, 2016, § 43 and 44) is a barrier to transparency, often cited by stakeholders to promote its restriction. In her decision on trilogues, the European Ombudsman sought to balance effective decision-making processes with access to information. She attempted to maintain the “capacity to negotiate” by recommending that the initial positions of the institutions are published before the trilogues begin, and that the four-column documents are published once the procedure is finished. This solution would allow the institutions to negotiate and would avoid the issue of a concession being offered and then withdrawn during a decision (2016, § 54). This issue could be poorly received by the public, and lead to negotiators no longer making concessions. The European Ombudsman’s line of argument does not seem to be fully compatible with the idea that citizens understand that negotiators have to switch positions in order to reach a compromise (2016, § 45). It is also not a solution to the issue raised by the European Ombudsman herself, namely the privileged access of certain groups to information, which would in her words justify a democratic opening of the process. It is also because of these apparent contradictions that basing access-to-information decisions on citizens’ alleged abilities (or lack thereof) is problematic. When legitimate limits to transparency are discussed, an argument which appears to be fundamental is that negotiating behind closed doors serves to strengthen the influence of certain interest groups while civil society is kept in the dark. This is the case not only because it can constitute real cases of opaque lobbying, but also because the perception of this risk generally lowers the level of trust in the institutions.

28 With that said, these considerations do not mean negotiation and transparency can easily coexist. As mentioned above, this real issue is less related to the ability of citizens to understand – which could be debated ad infinitum – and more to decision-making practices. An argument often used by the institutions when they refuse access to a document is that public meetings would prevent decision makers from having open exchanges. This was contested by the General Court in the Access Info Europe v Council [12] case, which noted that the Council did not convincingly demonstrate how publishing a document could actually damage trust between participants and discourage them to speak freely (2011, § 73). In fact, empirical studies on the Council show that negotiations make it difficult for the various parties to learn about the interests of other parties (Novak, 2011). This is particularly the case within the Council, where understanding what are the interests of each stakeholder is an ongoing problem for the various delegations and the presidency conducting the negotiations. Any negotiation situation implies a difference in preferences and interests, and in what will be said to other negotiators – the position defended. For example, it is common practice to ask for more than is really needed because then, knowing that concessions will be made, potential losses are minimised. Negotiation also entails the formulation of strategies which, by definition, are not revealed to the other negotiators. The European Ombudsman noted this, observing that making trilogues more transparent should not lead to the institutions revealing their strategies which constitute the way each institution intends to negotiate with the other two (2016, § 57). A lot can be said about the links between negotiation and a lack of transparency, but what is important to note in this paper is that discussions behind closed doors are not necessarily transparent. This lack of internal transparency within the institutions is also a barrier to public access to information : how are citizens expected to receive information about decision-making processes when stakeholders do not openly share information even among themselves?

29 It is clear that while the Court of Justice and the European Ombudsman uphold the right of access to information, there is no consensus on the limits that could be validly imposed on it. While the exceptions mentioned in Article 4(2) of Regulation (EC) No 1049/2001 are relatively specific, the protection of the decision-making process mentioned in Article 4(3) is vague and does not effectively address the friction between the capacity to negotiate and public access to information, since the decision to publish documents must be made on a case-by-case basis. Given the importance of negotiations in international and supranational bodies, and their increasing influence on national legislation, transparent negotiations and the democratic accountability of the stakeholders involved are a major political challenge.

References

  • Primary sources

    • Judgments of the General Court and the Court of Justice of the European Union
    • Council v Hautala, case C-353/99 P, 6 December 2001.
    • Sweden and Turco v Council, joined cases C-39/05 P and C-52/05 P, 1 July 2008.
    • Access Info Europe v Council, case T-233/09, 22 March 2011.
    • Council v Access Info Europe, case C-280/11P, 17 October 2013.
    • De Capitani v Parliament, case T-540/15, 22 March 2018.
    • ClientEarth v Commission, case C-57/16 P, 4 September 2018.
    • De Capitani v Council, case T-163/21, 25 January 2023.
    • Other documents
    • Conseil (2009), Décision du Conseil du 1er décembre 2009 portant adoption de son règlement intérieur (2009/937/UE).
    • Conseil (2014), Document 8622/14, Subject: Drafting of documents relating to legislative activities. Brussels, 11 April 2014.
    • Conseil européen (2001), Conseil européen Laeken. Conclusions de la présidence. Annexe 1. Déclaration de Laeken sur l’avenir de l’Union européenne. 14 et 15 décembre 2001.
    • Médiatrice européenne (2016), Décision dans l’affaire OI/8/2015/JAS – Propositions formulées par la Médiatrice européenne à l’issue de son enquête stratégique OI/8/2015/JAS sur la transparence des trilogues, 12/07/2016.
    • Médiatrice européenne (2018), Recommendation of the European Ombudsman in case OI/2/2017/TE on the Transparency of the Council legislative process made in accordance with Article 3(6) of the Statute of the European Ombudsman.
    • Médiatrice européenne (2022), Decision on the Council of the European Union’s refusal to give full public access to documents related to negotiations on the draft ‘Digital Markets Act’ (case 1499/2021/SF). 27/06/2022.
    • Parlement européen (2013), Résolution du Parlement européen du 12 juin 2013 sur la révision du règlement (CE) no 1049/2001 dans l’impasse (2013/2637[RSP]).
    • Parlement européen (2023), Revision of the access to documents Regulation In “Civil Liberties, Justice and Home Affairs - LIBE”. Rapporteure : Evin Incir. 20/04/2023. https://www.europarl.europa.eu/legislative-train/theme-union-of-democratic-change/file-revision-of-the-access-to-documents-regulation
    • Registre de transparence (2021), Accord interinstitutionnel du 20 mai 2021 entre le Parlement européen, le Conseil de l’Union européenne et la Commission européenne sur un registre de transparence obligatoire. Journal officiel de l’Union européenne no L 207/1, 11 juin 2021.
    • Règlement (CE) no 1049/2001 du Parlement européen et du Conseil du 30 mai 2001, relatif à l’accès du public aux documents du Parlement européen, du Conseil et de la Commission, Journal officiel de l’Union européenne no L 145, 31 mai 2001, p. 0043.
    • Statewatch (2023), EU : Eurodac and Joint Investigation Teams : four-column documents ready for secret trilogues, 27 February 2023, https://www.statewatch.org/news/2023/february/eu-eurodac-and-joint-investigation-teams-four-column-documents-ready-for-secret-trilogues/#:∼:text=The%20four%2Dcolumn%20documents%20are,once%20negotiations%20have%20got%20underway.
    • Traité sur le Fonctionnement de l’Union européenne (version consolidée), Journal officiel de l’Union européenne no C 326/47, 26 octobre 2012.
    • Transparency International EU (2022), Open letter : transparency for DSA and DMA trilogues, 1 March 2022, https://transparency.eu/open-letter-dsa-and-dma-trilogues/
  • Secondary sources

    • Bjurulf B., Elgström O. (2004), “Negotiating transparency : The role of institutions”, Journal of Common Market Studies 42, no 2, pp. 249–269.
    • Brandsma G. J. (2019), “Transparency of EU informal trilogues through public feedback in the European Parliament : Promise unfulfilled”, Journal of European Public Policy, vol. 26, no 10, pp. 1464-1483.
    • Curtin D. (2017), « ‘Accountable independence’ of the European Central Bank : seeing the logics of transparency », European Law Journal, 2017, vol. 23, nos 1-2, pp. 28–44.
    • Curtin D., Leino P. (2017), “In search of transparency for EU law-making : Trilogues on the cusp of dawn”, Common market law review, vol. 54, no 6, pp. 1673-1712.
    • Erkkilä T. (2012), Government transparency. Impacts and unintended consequences, New York, NY: Palgrave Macmillan.
    • Flinn A., Jones H. (2009), Freedom of Information : Open Access, Empty Archives ?, Routledge.
    • Grimmelikhuijsen S., Porumbescu G., Hong et Im T. (2013), “The effect of transparency on trust in government : A cross-national comparative experiment”, Public administration review, vol. 73, no 4, pp. 575–586.
    • Hillebrandt M., Leino-Sandberg P. (2021), “Administrative and judicial oversight of trilogues”, Journal of European Public Policy, vol. 28, no. 1, pp. 53–71.
    • Hood C., Heald D. (2006), Transparency: The key to better governance ?, Vol. 135, Oxford University Press for The British Academy.
    • Leino-Sandberg P. (2022), “Transparency and Trilogues : Real Legislative Work for Grown-Ups ?”, European Journal of Risk Regulation, pp. 1–21.
    • Maes M. (2010), « Le droit d’accès aux documents des institutions européennes », Cahiers de la documentation, 4, pp. 25–37.
    • Mansbridge J. (2009), “A ‘selection model’ of political representation”, Journal of Political Philosophy, vol. 17, no. 4, pp. 369–398.
    • Meijer A. (2013), “Understanding the complex dynamics of transparency”, Public administration review, vol. 73, no 3, pp. 429–439.
    • Meijer A. (2015), “Government transparency in historical perspective : From the ancient regime to open data in the Netherlands”, International Journal of Public Administration, vol. 38, no 3, pp. 189–199.
    • Năstase A., Muurmans C. (2020), “Regulating lobbying activities in the European Union : A voluntary club perspective”, Regulation & Governance, vol. 14, no 2, pp. 238–255.
    • Naurin D. (2006), “Transparency, publicity, accountability. The missing links”, Swiss Political Science Review, vol. 12, no 3, pp. 90–98.
    • Novak S. (2011), La prise de décision au Conseil de l’Union européenne : pratiques du vote et consensus, Paris, Dalloz.
    • Novak S., Hillebrandt M. (2020), “Analysing the trade-off between transparency and efficiency in the Council of the European Union”, Journal of European Public Policy, vol. 27, no 1, pp. 141–159.
    • Novak S., « De la “capacité de négocier” : un concept opaque peut-il justifier une transparence limitée du processus législatif européen ? », Revue des affaires européennes, à paraître.
    • Peters A. (2013), “Towards Transparency as a Global Norm”, in Bianchi A., Peters A. (dir.), Transparency in International Law, pp. 534–607, Cambridge, Cambridge University Press.
    • Przeworski A., Stokes S.C., Manin B. (dir.) (1999), Democracy, accountability, and representation, Cambridge, Cambridge University Press.
    • Robert C. (2018), « Les dispositifs de transparence entre instruments de gouvernement et « machines à scandales » : Fabrique et mobilisations des formes de connaissance sur le lobbying européen », Politique européenne, no 61, pp. 174–210.
    • Stasavage D. (2006), “Does transparency make a difference ? The example of the European Council of Ministers”, in Hood C., Heald D. (dir.), Transparency: The key to better governance ?, Vol. 135, Oxford University Press for The British Academy, pp. 164–179.
    • de Terwangne C. (2004), « Accès à l’information et organisations internationales : le cas de l’Union européenne », Éthique publique, vol. 6, no 2.