Transparency and Trilogues: Real Legislative Work for Grown-Ups?

Abstract

Trilogues represent a decisive stage in the European Union (EU) legislative process and often settle the substantive content of EU legislation. During trilogues, negotiations move fast and new solutions are actively identified by the negotiators. The (lack of) transparency of trilogues has been repeatedly criticised in recent years, yet the EU institutions have defended their “space to think”. Relying on a set of interviews with trilogue participants, this paper mirrors the institutional practices in the final stages of EU law-making against the requirements of openness in the EU Treaties, which aim to strengthen “democracy by allowing citizens to scrutinize all the information which has formed the basis of a legislative act”. The paper argues that, despite noble proclamations, the EU’s legislative practices are characterised by institutional discretion and the lack enforcement of transparency requirements. The paper describes how trilogues are conducted and how questions involving risk management and technically complex issues are assessed in this process. Greater transparency would also help to ensure that risks and alternatives are properly assessed and would thus contribute to better-quality risk regulation in the EU.

I. Introduction

Trilogues have taken over as the main forum for making legislative deals in the European Union (EU). While unknown to the Treaties, the Court has described them as “an informal tripartite meeting in which the representatives of the Parliament, the Council and the Commission take part” with the aim of reaching “a prompt agreement on a set of amendments”, which “must subsequently be approved by each of the three institutions in accordance with their respective internal procedures”. ¹ During this phase, the EU democratic process is in the hands of very few: the European Parliament rapporteur(s), the representatives of the Council Presidency and a few Commission officials or, more recently, Commissioners.

During trilogues, negotiations move fast and new solutions are actively identified by the negotiators. Their lack of transparency has been repeatedly criticised in recent years and has led to both Court litigation and investigations by the European Ombudsman. ² However, EU institutions have remained reluctant to increase the transparency of trilogues and defend their “space to think” – this despite clear Treaty provisions that place the Council and the European Parliament under an obligation to ensure the publication of the documents relating to the legislative procedures subject to Regulation 1049/2001 “in such a way as to ensure the widest possible access to documents”. ³ The Court has consistently stressed that the principle of openness

guarantees, inter alia, that the administration enjoys greater legitimacy and is more effective and more accountable to the citizen in a democratic system. By allowing divergences between various points of view to be openly debated, it also contributes to increasing those citizens’ confidence in those institutions. ⁴

In this case law, the standard of legislative transparency is set high:

Openness in that respect contributes to strengthening democracy by allowing citizens to scrutinize all the information which has formed the basis of a legislative act. The possibility for citizens to find out the considerations underpinning legislative action is a precondition for the effective exercise of their democratic rights. ⁵

Regulation 1049/2001 on public access to documents held by the three institutions only includes general provisions on legislative documents. ⁶ Many of the key issues have been addressed internally by the Parliament’s and the Council’s Rules of Procedure ⁷ – at their own discretion, which sets the level of transparency far below the level established by the Court. The benefits of transparency remain contested in the literature, and in the EU institutions the fears about its effect on efficiency continue to dominate the mindset. ⁸ Institutional practices and responses to Court findings are characterised by feet-dragging and a lack of commitment to realising the Treaty ambitions of legislative transparency in their daily activities.

In practice, much of the discussion on trilogue documents has focused on public access to the “four-column document”. This is the only jointly drafted report of discussions that tracks progress: “It is in effect the full ‘map’ of the informal, but decisive, trilogue negotiation process.” ⁹ It includes the initial Commission proposal, the Parliament’s position as adopted in Committee and the Council position. In addition, a fourth column shows the compromises suggested by any of the three institutions or considered as agreed to between the negotiators. ¹⁰ Four-column documents are drawn up in collaboration between the Secretariats of the three institutions, which also fill them in as negotiations progress. ¹¹ Only the fourth column with comments and compromise solutions changes during the negotiations. ¹² However, transparency in the final stages of EU law-making procedure is also affected by the institutional practices of all three institutions involved in the legislative process. There are various other ways in which trilogue transparency could be improved, ¹³ some of which have been discussed in the context of Ombudsman investigations, involving the publication of agendas and participants and communication about and from meetings.

A trilogue is a political negotiation and therefore inevitably political in nature. Today, a large part of the EU’s legislative activity takes place in scientifically and technically salient areas where it tackles complex problems that seem to demand expert knowledge for their resolution. The role of experts is important in EU legislative work. While experts are involved in the preparation of Commission and Council positions, the expert capacity of the European Parliament is limited and often compensated by input from interest representatives. The proper role of experts and the impact of expertise in decision-making processes are much-debated topics, ¹⁴ while much less attention has been offered to how the use of expertise is legally structured. Access to information about the conduct of trilogues and the use of expertise in the final decisive stages of EU law-making illustrate how all decision-making processes involving expertise are governed by legal rules and informal practices and understandings, which affect what we know and do not know about legislative compromises and their justifications. To the extent that legal and quasi-legal mechanisms to make processes visible and hold decision-makers and the way they use expertise to account exist, they are created by legal rules.

In this paper, I compare the existing legal rules on transparency with the institutional practices reaching beyond the tripartite meeting, as well as with how positions are prepared and communicated in the three institutions. Relying on a set of interviews with trilogue participants, the paper describes how trilogues are conducted and how questions of technical knowledge are assessed in this process. My main frame of reference is trilogues in the ordinary legislative procedure (Article 294 TFEU), but this decision-making format is also frequently used for breaking ground in special legislative procedures. I use sixty-three semi-structured and anonymised interviews conducted in the European Parliament, Commission, Council (including Member State administrations), the European Central Bank and the European Ombudsman’s Office during 2015–2020. ¹⁵ I argue that greater trilogue transparency is not just a democratic ideal but would help to ensure that risks and alternatives are properly assessed and thus contribute to better-quality legislation in the EU. My focus is on legislative files that require scientific or technical knowledge – something that sets specific challenges for the generalists that are in charge of negotiations at the trilogue stage.

II. The law and practice of trilogue transparency

The co-decision procedure (today known as the ordinary legislative procedure) was introduced by the Treaty of Maastricht largely to alleviate Europe’s democratic deficit. Article 294 TFEU establishes the main formal stages of the ordinary legislative procedure. It builds on the idea of the Council and the European Parliament co-legislating as more or less equal partners in three readings and, when necessary, in the Conciliation Committee. Yet, many of the key decisions during this procedure are produced with little scope for public oversight. ¹⁶ They are taken in a “plethora of informal and semi-formal meetings in which many of the real decisions about legislation are taken”. ¹⁷ Interinstitutional deals are, as the main rule, made in a fast-track procedure in the first reading. ¹⁸ Their use is in no way limited to technical, urgent or uncontested files. ¹⁹ Trilogues are the most important example of such informal meetings as they involve all three institutions.

The conduct of trilogues relies on institutional practice, codified in a Joint declaration on practical arrangements for the codecision procedure (2007), ²⁰ which stresses flexibility. The Interinstitutional Agreement (IIA) on Better Regulation addresses the organisation of the ordinary legislative procedure in general and the transparency of trilogues in particular. ²¹ In the IIA, the three institutions commit to ensuring “the transparency of legislative procedures, on the basis of relevant legislation and case law, including an appropriate handling of trilateral negotiations”. The IIA also includes a commitment to “improve communication to the public during the whole legislative cycle” and to “undertake to identify, by 31 December 2016, ways of further developing platforms and tools to this end, with a view to establishing a dedicated joint database on the state of play of legislative files”. So far, no such joint database exists, and legislative documents remain difficult to chase.

Trilogues are defined as the interinstitutional stage, which follows after each of the institutions has established its own position independently. However, no institution is an island, and institutional decision-making is intertwined from its early days. Many respondents point to the tightened relationship between the Parliament and the Commission, “and all the commitments given by the Commission President in order to be elected and enjoy the trust of the Parliament”. ²² The Commission prepares its legislative plan as part of its annual work programme, which is negotiated with the Parliament. ²³ This leaves the Commission very responsive to the Parliament’s wishes. ²⁴ As regards the Council, the Commission is almost always present at its meetings ²⁵ and routinely receives all Council documentation. ²⁶ During the legislative process, the institutions are in a more or less permanent dialogue at all levels, making informal contacts “something very normal and natural”. ²⁷ A Member of the European Parliament (MEP) explains that as rapporteur she spoke every day with the Commission official to make sure that her compromise proposals would be acceptable for the Commission. ²⁸ She describes how contacts are made long before the Parliament settles its position: “I speak with the Council Presidency many times a week on which states think what, where is the Council’s sense of humour, what is a definite no-go, could we find a common will somewhere.” ²⁹ Very little is formally disclosed about discussions within the Council, yet its political dynamics are well known in the Parliament. ³⁰ Position building in each of the three institutions is affected by what happens in the other two institutions.

“Formal informal trilogues” are launched when the Parliament has adopted its formal position and the Council has adopted its general approach, with a view to reaching “a prompt agreement on a set of amendments acceptable to the Parliament and the Council”. ³¹ The question of what counts as “technical” and what counts as “political” is always a contested topic. ³² In this context, however, the division into “technical” and “political” matters defines the participants, the format of negotiations and the visibility and accountability of solutions. (Informal) political trilogues involve the political level (MEPs, Commissioners and Ministers), while technical meetings are conducted among civil servants and political advisers with a view to preparing for political discussions. Technical trilogues also involve brainstorming and the informal exchange of ideas; for example, European Parliament officials are not expected to have a mandate for the solutions they propose. ³³ Most matters are settled in technical trilogues, sometimes so efficiently that they cover most of the agenda. ³⁴ The function of meetings at the political level is then condensed to confirming the deal already made or agreeing on the proposals for solving the questions that remain formally open. Therefore, the distinction between “technical” and “political” is difficult to maintain:

Sometimes you have political dialogues which are so technical that you wonder what would be a technical trilogue. But you also have very political trilogues, where it’s really a lot of politics: more political messages and all the work was shifted to the technical level and the technical trilogues lasted much longer than the political trilogues, and they would come back and say OK that’s fine go to the next topic. And I have to say this is where the deal is being made. ³⁵

Negotiations may also be delegated to the Legal Services of the Institutions, who then act as negotiators representing their institutions while also advising them with regard to legal matters. ³⁶ Delegation to the technical level builds on the assumption that important matters are decided by actors that remain politically accountable for their actions. Yet, all respondents indicate that such a presumption remains an illusion:

I’d like to think that there are some criteria to evaluate what is technical and what is political, but it is not always straightforward. Everything that is controversial is political. Things that seem technical can suddenly become political, and vice versa. There is an ongoing discussion between the different layers of decision-makers. Important elements can still be considered technical; elements can change nature. It is a trade where closed items can be re-opened, and nothing is agreed until everything is settled. ³⁷

The latter principle entails that, until the complete text has been agreed by the co-legislators, changes can be made to any part, including those provisionally closed. ³⁸

There are no joint or agreed minutes or reports of trilogue meetings. Instead, reporting takes place within each institution according to its own practices. ³⁹ In principle, any kind of document that is seen by the parties as facilitating the negotiations is admitted, ⁴⁰ and as Section III illustrates, trilogue participants gain knowledge and solutions from various directions. The attempts to make trilogue agendas publicly accessible are considered useless, as agenda items are frequently postponed and matters outside the agenda are taken up. ⁴¹ Four-column documents represent a means to control and keep track of the negotiation process. While prepared at the technical level, legislative solutions build largely on discussions between the Presidency, the Commission and the Council Secretariat when preparing these documents. ⁴² As noted above, this involves a great deal of informal “technical” brainstorming also taking place outside institutional mandates. If a deal proves impossible within the existing mandate, the Presidency may indicate to the Parliament its willingness to explore a new mandate in Coreper or solutions going beyond the mandate, subject to Coreper approval. ⁴³ On the Council’s side, there is a feeling that sometimes the rapporteurs and shadow rapporteurs operate rather autonomously from the Parliament, and doubts remain as to their capacity to gather sufficient majorities. ⁴⁴

The impact of future legislative measures is primarily assessed during the preparation of legislative proposals. Yet, changes made during the legislative process affect its outcome, highlighting the objectives of the EU Better Regulation agenda, including the aims to improve quality and public participation and to promote science-based governance. ⁴⁵ Trilogues are justified with reference to efficiency, which takes priority over transparency or good-quality legislation. ⁴⁶ The urge to close files reaches beyond those that have a clear deadline: “more time will only help more problems to emerge”. ⁴⁷ A European Parliament official explains how “exhaustion is often the real motivation for closing a file”; in addition to the “need to make results, this is what the public expects from us”. ⁴⁸ This process of EU law-making – going through the various levels of Council decision-making followed by compromises with the European Parliament – is seldom conducive to good-quality legislation, in particular since

the Parliament’s approach often seems very political, you have occasions where you think that they’re not concerned about the law at all, and all that has to be reconciled and, you have to put in a “where appropriate” to get one delegation on board and “if necessary” to get another on board, and that’s how you end up with such impenetrable drafting. ⁴⁹

In many Member States, (specialist) lawyers are in charge of drafting legislation. ⁵⁰ In the EU, this task lies primarily with generalists and politicians. In particular, the final stages of adopting EU legislation are often chaotic, and the quality of the drafting tends to rank low on the list of priorities. Ambiguity is also regularly and intentionally used as a legislative technique to enable compromises. The Parliament is known to make proposals that are either legally impossible or internally conflicting, even if such aspects are primarily sorted out at the technical level when preparing the four-column document. Towards the end of trilogues, such aspects are sometimes approved if they help close the file “even if this makes no sense and is completely unclear but just let it go so that we get a solution”. ⁵¹ Quite often, the Council and the Commission also rely on each other to defend “something sensible” or “regulation that is possible to apply and implement”. ⁵² As a respondent elegantly observes, “[n]ational parliaments don’t produce this kind of rubbish”. ⁵³ On the Parliament’s side, the Council is seen as reluctant to change anything, while the Parliament is more willing to take risks. ⁵⁴ Sometimes aggressive lobbying leads to new debates, ⁵⁵ which provokes heated debate in the Parliament during the final stages of legislative negotiations. In such debates, “technical” proposals building on “evidence” are “particularly prone to misrepresentation in the media”. ⁵⁶

When trilogues end, formal approval in the institutions is usually a mere formality. ⁵⁷ An ambassador explains:

During my three years in Coreper I, we have only had one case where a trilogue deal was re-opened, on car emissions where Germany was late and Merkel intervened. There may be other cases where the Presidency loses Member States’ support and they ultimately vote against, but usually trilogue deals are unanimously approved. ⁵⁸

An assistant working in the European Parliament confirms that in the more than thirty trilogues she has experienced, files have never been reopened once provisionally concluded. ⁵⁹ After the end of negotiations, the outcome is only subject to formal linguistic control.

The formal procedural stages established by Article 294 TFEU bear little resemblance to how EU legislation is actually approved. As regards transparency in this process more specifically, beyond the Treaty-based broad principles of legislative transparency, trilogues exist in a regulatory near-vacuum. The Parliament’s Rules of Procedure are the only institutional rules that recognise their existence. They regulate entering trilogues, enabling the Committee responsible to empower the European Parliament representatives to launch negotiations based on the Committee’s report. ⁶⁰ While transparency is emphasised, ⁶¹ this comes with serious limitations. While debates in Parliament are public, parts of Committee meetings can be held on camera. Public access is limited to “Parliament documents” that are

drawn up or received by officers of Parliament … by Parliament’s governing bodies, committees or interparliamentary delegations, or by Parliament’s Secretariat. … [D]ocuments drawn up by individual Members or political groups are Parliament documents for the purposes of access to documents only if they are tabled in accordance with the Rules of Procedure. ⁶²

These limitations also have significant implications for trilogue transparency.

The Council Rules of Procedure only address questions relating to Council meetings in ministerial composition, ⁶³ while the General Secretariat has discretion to make various other legislative documents available to the public as soon as they have been circulated. ⁶⁴ Decisions on proactive document disclosure are taken at the technical level by Secretariat staff, who usually make them available only once the act is approved. The Commission has no comprehensive public register, but merely one that contains “various types of Commission documents such as proposals, impact assessments, communications, delegated and implementing acts and other Commission decisions, agendas and minutes of meetings held by the College of Commissioners”. ⁶⁵ In practice, most legislative documents in its possession can only be accessed by submitting a request. This means that the possibility of the broader public following the conduct of legislative negotiations is highly restricted unless unauthorised disclosure takes place through leaks. Beyond that, the only form of control is the institutional control of those who are present in trilogues.

III. Who controls trilogues and their outcomes? Institutional scrutiny of expertise

1. European Parliament: outsourcing expertise

The formal key actors for trilogues in the European Parliament are the Committee responsible and the rapporteur and shadow rapporteurs appointed by it. ⁶⁶ The plenary stage is seldom problematic – the real political debates take place at the Committee stage or in trilogues. ⁶⁷ While some MEPs “specialise” in specific policy fields, ⁶⁸ as to their background they are generalists and politicians, and seldom are they in possession of any specific expertise relating to the Committees where they serve. In general:

Rapporteurs are chosen on the basis of whether they agree with the group, how sensible they are, what kind of passions they have. It is a completely normal foggy political process, and if you follow it closely you understand that from the outside this Vatican remains closed and it remains unclear how the Pope has been selected. ⁶⁹

In position building, political groups matter, ⁷⁰ “but it also matters what the Germans think”. ⁷¹ Group positions remain at a general political level. An MEP explains, “[m]ost of these positions would be like, do we wish for more or less European supervision, or do we wish to maintain something at national level, or how do you define a risk”. ⁷² Parliament’s position building depends on identifying the right political arguments – such as consumer or development interest. ⁷³ However, in many more complex files, the political question that sets the lens for viewing the file is difficult to grasp. ⁷⁴ Knowing what the political group thinks about details is often difficult: there may be questions that are a clear “no-go” for the group, while in many others no group position exists. ⁷⁵ In practice, the rapporteur – and, if sensible, together with the most important shadow rapporteurs ⁷⁶ – makes the political choices. These are then translated into details by three people: the MEP’s assistant, an official working for the political group and an official from the Committee Secretariat. ⁷⁷ Rapporteurs have tremendous power in particular as regards the technical aspects of legislative files:

[F]ew other MEPs in the same group will make the effort to learn them, and then it is just easiest to rely on the rapporteur, especially if she is known as a reliable MEP who does not push her own agenda too far. Then we just go to plenary and vote based on the lists; very few would check anything at that stage. ⁷⁸

An MEP estimates that perhaps a third or half of MEPs know what they are voting about, while others take care of their own reports and expect others to back them up. ⁷⁹ In these debates, “better regulation is purely an argument against new regulation. For example, if we do not wish for new environmental standards because we wish to support the industry, then we say that there is a need for better regulation.” ⁸⁰

Compromises are prepared by the Committee Secretariat, even when identifying them is actually a political exercise. ⁸¹ Sometimes there are Committee votes that leave the Parliament’s position ambiguous. The Secretariat will then try to improve its coherence, but generally matters are left to trilogues to be sorted out. ⁸² The role of Parliament officials is central. Trilogues are difficult for generalists since an understanding of the substance is needed for negotiating and finding alternatives. ⁸³ They are carefully managed by the Committee Secretariat, which prepares a manuscript for the MEPs, agendas and other documentation. ⁸⁴ An MEP recalls that some trilogue participants clearly understand what is going on while others do not, which emphasises the role of assistants. ⁸⁵ Completely new issues may also emerge at the trilogue stage, requiring an immediate reaction, ⁸⁶ which emphasises the role of the negotiating team further.

Much of the “trilogue transparency” in the European Parliament is directed at ensuring a flow of information within the Parliament itself, while broader transparency is more of a side effect of institutional practices aiming at ensuring all political groups have access to key developments in legislative negotiations. Rule 74 of the Parliament’s Rules of Procedure addresses the circulation of documents within the negotiating team and reporting back to the Committee or its coordinators about trilogues and possible provisional agreements. These regular briefings seldom provoke any discussion as it is not customary for Committees to intervene in ongoing trilogues. ⁸⁷ So while the Parliament gladly makes the point about legislative transparency, gaps exist:

First, meetings between rapporteurs and shadow rapporteurs are never open, and that’s where the Parliament decides what kind of compromises are approved and what is the preferred direction in negotiations. … I have never seen a situation where this discussion would have taken place in a Committee meeting. And second, there is no hearing of experts. … In a national parliament, there are experts present, and some of them are for and some are against, and you have the chance to cross-examine and also learn. … This is also helpful for understanding the many contexts in which the law needs to be applied. ⁸⁸

The need for expertise is mentioned in most interviews. ⁸⁹ Legislative work in the European Parliament requires knowledge that politicians seldom possess. An MEP explains her experiences with trying to brief national parliamentarians about her work:

For example, I tell them that now we talk about cloning animals to be used as foodstuff and the Commission has changed the definition of cloning to mean that clone is a first-generation animal and not a second-generation animal. A national parliament may say, we are not concerned with minor issues such as definitions, but only wish for no clones, clean food and agricultural products produced nationally. … European Parliament cannot operate in the manner where we just say common good, have a nice day. We need to define percentages, years, amendments, it is about real legislative work for grown-ups. Every proposal has consequences and effects, it has to be operationable. ⁹⁰

Unlike the Commission and the Council, the European Parliament employs few experts and has limited formal access to them. ⁹¹ In the context of more significant legislative files, Committees may organise a hearing of experts. But the selection of experts is a political exercise: “this is not a balanced or neutral hearing but highly political one”. ⁹² The most important legislative Committees have no time to hear experts, and locating and acquiring the expertise becomes the responsibility of individual MEPs.

Who gets consulted then? One assistant mentions “Member State representatives and a lot of operators in the field”. ⁹³ For one MEP,

[i]t is an indefinite number of self-acquired contacts including national and private associations and European and American manufacturers who tell you about emission filters and their efficiency, and which motor works in what kind of traffic. You end up learning so much substance. ⁹⁴

Another MEP recalls, “I try to listen to many kinds of interests, researchers, Commission officials, national officials … but the final adjustments come from the experts that work for your own group, who then consult the Commission”. She points out that in matters involving the environment, lobbying by the industry is massive. ⁹⁵ She continues, “I did my best to be equal and meet everyone wishing to speak to me, when acting as rapporteur or shadow. But not everyone does this.” ⁹⁶ Another MEP agrees: “Some only listen to those who support their own opinion, while I also wish to check that I am not entirely wrong in the opinion of some others. Some only look at things from their own national angle.” ⁹⁷

Among all respondents in the Parliament (MEPs, political assistants, members of Committee Secretariats), gaining expertise is considered vital. Lobbying becomes the channel for accessing information about different positions. ⁹⁸ One MEP recalls,

I was the shadow rapporteur on a file that concerned the energy efficiency of buildings. What do you think I knew about this topic? Nothing. And I listen to the lobbyists that I contact and those that reach out to me, and I listen to those that I wish to listen to. I need to listen to someone, otherwise I’d have nothing to say, and I need to find the information somehow. … And it is also a part of better regulation that you hear experts or lobbyists, however you wish to call them. ⁹⁹

Lacking necessary technical expertise is an issue not only for MEPs but also for the people who assist them and who also come from a generalist background. One member of the Parliament’s Secretariat agrees:

Technical files are difficult, it is clear that politicians cannot really know about these matters. Neither do we in the Secretariat, it is about learning by doing. I read academic things, things that are provided by lobbyists. It is often said that we should not speak to outsiders, but how are we otherwise supposed to get information about the things we should be doing? ¹⁰⁰

One assistant argues, “my job as assistant to an MEP is to meet lobbyists. They all have their own agendas, and it is my job to identify whether this agenda is good, whom does it benefit.” ¹⁰¹ She continues,

[l]obbyists are very useful for gathering information and identifying big problems that then define the small problems. There are issues we might miss entirely without lobbyists. And the resource in the [European Parliament] is very limited – in this file, it would be only me. The MEP would meet with the most important lobbyists … to make sure she can defend her choices and knows which ideological choices we have made. ¹⁰²

But lobbyists are also used to influence positions in the Member States:

We kept constantly in touch with consumer organisations that were our ally and support, because they can organise strong lobbying through their national branches. … [W]e would also provide them with information about the state of [trilogue] negotiations, and they would leak it to the German media. This is how I see comprehensive, strategic impact. ¹⁰³

Therefore, while public access to trilogue documents remains constrained, informal access by stakeholders remains strong. While MEPs generally see meeting with lobbyists as necessary, they do criticise the way colleagues take amendments from them:

I remember one lobbyist lunch at the time we were negotiating the ETS [EU Emissions Trading System] for air traffic, organised by European airline associations, who said “we will table amendments”. And we would all laugh, because of course they have no role in the process, but we would also know that they will find MEPs who will table those proposals. ¹⁰⁴

The way MEPs propose amendments received from lobbyists has been systematically studied – for example, in the context of the EU Regulation on Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), where of 1,413 amendments tabled in certain sections of the Regulation, sources within the lobbying sector could be identified for 61 percent. Amendments were tabled by MEPs from all political groups. ¹⁰⁵ One MEP points out that some colleagues probably

sincerely think that they will do a favour to their own country’s industry or European industry or economy by putting forward amendments suggested by these lobbyists. … I think that if you think it is OK to take an amendment from outside, then it should also be OK to explain publicly where it came from. ¹⁰⁶

The European Parliament Rules of Procedure stress that MEPs should only meet interest representatives that have registered in the Transparency Register, and they should publish online all of their scheduled meetings with such interest representatives. For rapporteurs, shadow rapporteurs and Committee chairs there is an obligation to publish online, for each report, all scheduled meetings with interest representatives. ¹⁰⁷ In addition, the European Parliament website explains that MEPs drafting reports or opinions can choose to attach a legislative footprint to their reports to “demonstrate the range of outside expertise and opinions the rapporteur has received”. ¹⁰⁸ This remains voluntary, and the Parliament has no rules that would even attempt to make external influence in legislative work efficiently visible with a view to enabling political accountability. Meetings conducted by political assistants or secretariat staff remain below the radar, even when their impact on legislative work during both the preparation of European Parliament positions and the final stages of identifying compromises may sometimes be bigger than that of the rapporteur. Work that is considered “technical” remains invisible and falls outside political scrutiny.

2. Ambassadors as “university-level experts”: the role of Coreper

The key institutional setting in the Council is Coreper, which approves most trilogue mandates, is kept updated about trilogue negotiations and, in practice, approves final deals. ¹⁰⁹ Nearly nothing is known about these discussions. ¹¹⁰ Communication regarding substantive Coreper discussions is non-existent beyond publishing which items are on its agenda. Coreper documents are not public, and matters are brought to public attention only once deals have been closed. ¹¹¹ Coreper serves as a “smoke screen … where you can move without losing face”. ¹¹²

Coreper discussions are conducted among ambassadors. They are generalists, and they cannot possibly know everything about every technical detail. However, at the Coreper stage, relatively few questions remain open, making it possible to study and prioritise the ones that matter for one’s country. ¹¹³ Ambassadors follow instructions from capitals, but they also have some leeway. One ambassador recalls, “as regards the sustainability criteria for biomass, I am soon a university-level expert”, but with less important files, you need to learn to let go: “you have to choose your battles, but some technical details must be studied in depth”. ¹¹⁴ Needing to cover the technical details of every single ongoing legislative file is a particularly challenging issue for the Presidency, who, of course, chairs Coreper and leads trilogue negotiations on all ongoing legislative files. The ambassador continues,

an absolute horror file is the one involving medical equipment, everything is so technical …. Another ambassador just pointed out that railway packages may have felt tough at one time, with all the technical compatibility requirements and stuff, but it is child’s play when you compare with medical equipment. ¹¹⁵

While experts are involved in assessing trilogue results in the Council, effective scrutiny is often hindered by their rapid advancement:

[T]he speed can be, “could we please do this document of 100 pages before the working party which is in two days?”. And let’s be honest, sometimes you do miss things. ¹¹⁶

Towards the end of each Presidency, negotiations move fast. Presidencies are eager to close files, which is often helped by how delegations also become more willing to make and accept more compromises than they had done some months earlier. ¹¹⁷ Such bargaining is not visible to outside audiences, even if some governments provide (usually confidential) information to their parliaments and engage them in adjusting national positions when necessary. In most legislative files, the presidency remains within its mandate and communicates whether they need more flexibility, also with a view to a better overall compromise. ¹¹⁸ One legal adviser underlines that the Member State representatives in Coreper “would be able to explain to their authorities what happened and how you ended up with something that may not have been entirely in line with what they had wanted in the beginning”. ¹¹⁹ Therefore, to the extent that any accountability relationship exists, it remains between the presidency and Coreper on the one hand and Council members and their national parliaments on the other. The possibility of the broader public following the decision-making in the Council beyond the limited number of streamed, largely orchestrated and superficial Council sessions remains non-existent. Knowledge of the state of play in Coreper is dependent on leaks to trusted Politico or Financial Times journalists and avenues of privileged access. The most decisive stages of legislating in Coreper are completely beyond public reach, and the reasons underlying any particular compromise are covered by a general culture of secrecy.

3. Politics of expertise in the European Commission

What is the Commission’s role in the legislative procedure in general and in trilogues in particular? Is it that of a technical assistant or political body? These questions are decisive for settling whether it is covered by legislative transparency, as has been recently addressed by the European Court of Justice (ECJ) in a case that concerned public access to impact assessments, which the Court defined as “key tools for ensuring that the initiatives of that institution and EU legislation are developed on the basis of transparent, comprehensive and balanced information”. ¹²⁰ The Commission had opposed their disclosure, claiming that it, as a result of public disclosure, ”would be constantly engaged in multiple dialogues with interested parties, so that it would be impossible in practice for it to have space for autonomous deliberation and to make a decision in a fully independent manner”. The Court accepted that in the preparatory phase of legislative action the Commission does not act in a legislative capacity and that the legislative procedure sensu stricto formally begins when a proposal is submitted by the Commission. ¹²¹ Despite this, it recognised that the Commission is a “key player in the legislative process”, ¹²² which also placed it under obligations relating to transparency and public participation. For the Court,

the expression by the public or the interested parties of their views on the choices made and the policy options envisaged by the Commission in the context of its initiatives … before that institution has made a decision regarding the planned initiative, is an integral part of the exercise by EU citizens of their democratic rights. ¹²³

In another case involving specifically the role of the Commission in trilogues, the Court confirmed its right to withdraw a proposal on the very day that the Parliament and the Council were preparing to formalise their trilogue agreement. ¹²⁴ Thus, the Commission’s legislative powers are operative in trilogues, and it takes them very seriously.

Any acceptance of amendments deviating from the Commission’s initial proposal must be backed by the College of Commissioners. The updates to those mandates are prepared in weekly meetings of the Commissioners’ Cabinet members. ¹²⁵ Discussions are confidential, ¹²⁶ and “[m]embers shall refrain from disclosing what is said at meetings of the Commission”. ¹²⁷ The minutes of College meetings are published online, but they simply give a general account of the discussions and list adopted decisions without revealing diverging opinions. ¹²⁸ Panning argues,

[t]he thorough preparation of trilogue positions, red lines and strategies questions the image of the Commission as a neutral honest broker. In this role, the Commission’s interest should only be to help solve differences between Parliament and Council. The intensive preparation inside the Commission, however, paints the picture of a committed broker; an institution that sees its job in facilitating negotiations and providing expertise for the co-legislator, but also tries to safeguard its own preferences and interests. ¹²⁹

Her findings are supported by my interviews. In the negotiations, the Commission generally facilitates the negotiations, provides budgetary estimates and provides the only available impact analysis. ¹³⁰ One adviser to an MEP explains,

the Commission has an expert role: they know the facts, what is sensible and might succeed and what might not. But sometimes the Commission also takes a highly political role, and does not remain in its role as facilitator; it may have firm political opinions …. They do take an active role in leading the discussion to a direction they see as sensible; visit our internal meetings and explain things, and provide compromise proposals. They may also take a clearly proactive role when they see that something would be important or some proposal is inherently bad. ¹³¹

One MEP recalls having acted “as rapporteur for files where the Commission pressured me to push for certain solutions in the trilogues”. ¹³² One assistant to an MEP recalls,

[t]he Commission representative asked me privately after the first trilogue what it is that we on the Parliament’s side need for a deal on the dossier. This is a borderline case for me on who actually exercises power. I indicated that this might be a possible compromise. ¹³³

Commission services systematically gauge and analyse the positions of key players in the Parliament and try to anticipate and follow developments. Commissioners participate in the meetings of their respective political groups or presidents in the Parliament, which often provide useful opportunities for selling new political ideas. ¹³⁴ In the context of trilogue negotiations, the Commission’s approval of a solution tends to reflect the Parliament’s position. ¹³⁵ All of this is highly influential for legislative outcomes. However, the Commission’s internal position building is covered by high levels of confidentiality. ¹³⁶

IV. External scrutiny of trilogue transparency

Transparency of trilogues or the surrounding institutional practices has never been great. For those outside the meeting room, it is difficult to know how deals have been made and whose input has been used to strike those deals. While political responsibility may be somewhat easier to conceptualise for general political choices more easily grasped at the level of generalists, it is nearly impossible for choices that are considered “technical”.

Trilogue transparency has been recently addressed by the European Ombudsman, who recognised a general difficulty in tracing and locating existing public information and recommended the establishment of a joint database. ¹³⁷ The most useful part of her enquiry is that its scope reaches to how trilogues are arranged: she urged the institutions to provide information on trilogue dates and to make lists of trilogue documents, including a list of the politically responsible representatives or civil servants in charge of negotiations. However, as regards actual trilogue documents, she limited herself to general summary agendas before or shortly after the trilogue meetings, and she was satisfied with information that does not reveal individual strategies nor compromise negotiations. ¹³⁸ Transparency during trilogues was reduced to communication – a strikingly “pragmatic” capitulation of the democratic principle of legislative transparency ¹³⁹ and an uncharacteristically a-legal approach. ¹⁴⁰ Disclosing the result after the trilogues have ended is certainly too late for enabling democratic debate.

The specific question of public access to the full contents of four-column documents in various trilogue negotiation procedures was soon thereafter addressed by the General Court, ¹⁴¹ which underlined that trilogues, even when informal, form an integral part of the legislative procedure, and that “the [protection of the] effectiveness and integrity of the legislative process cannot undermine the principles of publicity and transparency which underlie that process”. ¹⁴² To withhold access to four-column documents for the duration of trilogue negotiations would deprive citizens of their right to participate at a pivotal stage of the legislative process:

in a system based on the principle of democratic legitimacy, co-legislators must be held accountable for their actions to the public. If citizens are to be able to exercise their democratic rights they must be in a position to follow in detail the decision-making process within the institutions taking part in the legislative procedures and to have access to all relevant information. ¹⁴³

The Court found that four-column documents “could [as such] not be regarded as sensitive by reference to any criterion whatsoever”. ¹⁴⁴ However, since the matter concerned a public access request made on the basis of Regulation 1049/2001, the Court did not address the general question of proactive disclosure of the fourth column, ¹⁴⁵ which maintained the option of disclosing trilogue documents one by one, and only in case there is a public access request. Such requests routinely take several months to handle, since the institutions seldom follow the time limits specified in the Regulation. ¹⁴⁶ Strategic delays in handling the applications enable the institutions to postpone disclosure until trilogues have ended, and this makes it difficult to challenge their decisions.

The ruling has been met with mixed institutional commentary. Van Nuffel (a member of the Commission Legal Service) argues that it “makes it now very difficult for the co-legislators to generally refuse access to four column documents”, but it may make them reluctant to actually “enter compromise text in the fourth column at an early stage of the negotiations”, or at least “clearly indicate the conditional and provisional status of any partially agreed text”. ¹⁴⁷ Rebasti, the Council agent in the de Capitani case quoted above, however, argues,

trilogues remain exempted from other requirements that would normally be associated with the formal steps of the legislative process: the pro-active publication of documents, the publicity of the debates, and the need for a fully-fledged linguistic regime of the documents used for deliberations. ¹⁴⁸

Rebasti’s reading conflicts with the view of the Ombudsman in a later investigation, ¹⁴⁹ where the Council granted access to only parts of the requested documents, redacting parts that contain the Council’s negotiating strategy – its “red lines”, points where it could be flexible and its fallback options – in ongoing negotiations with Parliament. The Ombudsman agreed that releasing such details could seriously undermine the Council’s negotiating position. However, the provisional compromises found between the co-legislators, as well as the evolving positions, proposals and comments of the three institutions in relation to those parts of the legislative text on which no agreement has yet been found, should be made public upon request in line with recent case law of the General Court so as to enable the public to participate in trilogue negotiations and to influence the legislative process at this crucial stage.

As discussed above, trilogue negotiations are in many ways intertwined with institutional position building. Documents that are relevant for trilogue negotiations and their outcomes are also produced in the three institutions involved in them. The Ombudsman has recently conducted an enquiry relating to access to Council documents in the legislative process and its preparatory bodies. ¹⁵⁰ In that context, she argues for direct, proactive and generally full access to its legislative documents, since “[e]nsuring that citizens are able to follow the progress of legislation is not something to be desired; it is a legal requirement”. ¹⁵¹ Parliament documents drawn up by MEPs in their function as rapporteurs or shadow rapporteurs and (the lack of) minutes from their preparatory meetings were addressed in the context of another earlier complaint. The Ombudsman accepted the Parliament’s argument that the requested documents “do not exist”: “there is no prima facie case that the Parliament was under a legal obligation to create minutes of the meetings in question or to give information about the discussions at the meetings”. ¹⁵² This relaxed approach to the Parliament is at odds with earlier case law where the General Court has held that

it would be contrary to the requirement of transparency which underlies Regulation No 1049/2001 for institutions to rely on the fact that documents do not exist in order to avoid the application of that regulation. In order that the right of access to documents may be exercised effectively, the institutions concerned must, in so far as possible and in a non-arbitrary and predictable manner, draw up and retain documentation relating to their activities. ¹⁵³

Overall, the Ombudsman’s approach towards the Parliament’s transparency practices has been remarkably accommodating. It is unlikely that legislative negotiations within and between political groups would take place without any supporting documentation including drafts for compromise proposals. Public access now formally reaches to various “internal” documents of the Council – for well-justified reasons, which are currently not applied to the Parliament’s working documents. When such documents are drafted by MEPs or political groups, public access can be avoided by simply not tabling them “in accordance with the Rules of Procedure”. The matter should not be about internal rules of registration but about whether the documents concern the institution’s work, in which public access rules are applicable – as the Ombudsman argued recently in relation to the Commission’s text messages. ¹⁵⁴ Why would a similar logic not apply to documents produced in the Parliament?

Did these interventions by the Ombudsman and General Court produce tangible transparency effects? Following the Ombudsman trilogue enquiry, the Council has committed to disclosing certain documents proactively, including: the initial Council mandate enabling the Presidency to engage in trilogues; four-column documents but excluding the fourth column; the final position of the Council reflecting the final outcome of negotiations after examination by Coreper; and documents relating to the adoption stage of a legislative file. ¹⁵⁵ Before that, these documents can be made available based on individual requests under Regulation 1049/2001. ¹⁵⁶ While the more systematic disclosure of Council negotiating mandates may be traced back to the Ombudsman enquiry, in other respects the Council only commits to measures that are

purely cosmetic in nature. Day-to-day experience shows us that in the future, most of the Council’s preparatory documents concerning the legislative processes will not be directly accessible, but marked as documents covered by confidentiality requirements (using the word LIMITÉ). ¹⁵⁷

The Parliament has assured that its negotiating mandates as well as the names of members of negotiating teams are publicly available. ¹⁵⁸ The status of four-column documents is unclear. ¹⁵⁹ The Parliament has adopted a resolution in 2019 endorsing the Ombudsman’s findings, but it focuses on blaming the Council’s reluctance and failure to comply with it. ¹⁶⁰ Finally, the Commission has agreed that the responsibility for trilogues could be made clearer by identifying the responsible Commissioner and Directorate-General. Overall, it maintained its traditional position that legislative transparency only concerns the co-legislators. ¹⁶¹ It seems that each of the three institutions involved thinks that the fault for the lack of transparency is with the other legislative institutions, while its own secretive practices should remain intact.

V. Inclusion and exclusion in trilogues

On 14 July 2021, the European Commission adopted a massive package of proposals to make the EU’s climate, energy, land use, transport and taxation policies fit for reducing net greenhouse gas emissions by at least 55 percent by 2030 (“Fit for 55”). ¹⁶² In April 2022, I searched all institutional websites to find out whether any of the files had reached the trilogue stage and which documents might be publicly accessible. This task proved impossible, which led me to send an enquiry to the Europe Direct Contact Centre run by the European Commission, which redirected my enquiry to the Parliament’s Citizens’ Enquiries Unit. I was informed that “there is no website that indicates Commission proposals under the ordinary legislative procedure that have ongoing trilogues”. ¹⁶³ Instead, I was advised about the existence of certain search tools that can be used to identify potential files, but that “even if such a list existed, it would always be a situational snapshot, that can quickly be outdated”. When applying the suggested filters ¹⁶⁴ in the Legislative Observatory, 151 files are identified; however, each of them

would need to be checked individually, whether a mandate was voted in plenary to enter into interinstitutional negotiations or if a committee mandate was confirmed (to be found in OEIL [EU Legislative Observatory] under Key events: either mandate confirmed or vote and referral back to committee). … Furthermore, you would need to check individually, for each of the results of this exercise, whether the Council also has a mandate, an information that can be found in OEIL under Council documents. Interinstitutional negotiations can take place only when both institutions, Parliament and Council, have a negotiating mandate. ¹⁶⁵

What does this tell us about the possibility of following trilogues in some of the most important risk regulation files that the EU has introduced so far? It is a nearly impossible exercise. I finally gained the information I was after through an enquiry placed on social media, which an MEP reacted to, and by bumping into one of the European Parliament’s rapporteurs in the security control queue at Helsinki Airport. These informal channels proved useful, while information through formal channels remains impossible to find.

Trilogue transparency involves inclusions and exclusions. Based on my interviews, information flows freely within and between the three institutions involved in trilogues. In addition, many external stakeholders have privileged access to knowledge about the process. Those who remain excluded are citizens, civil society and national parliaments – key participants in the democratic process. Formal channels to key information about trilogues remain dysfunctional, and addressing this problem is of low priority for the institutions who wish to safeguard efficiency and avoid public debate about legislative outcomes before their conclusion. As a result, control of outcomes is difficult, and compromises are not put to public test before their formal conclusion. While problematic as a matter of principle, this also comes at a cost to legislative quality, as greater transparency would help to ensure that risks and alternatives are properly assessed and would thus contribute to better-quality risk regulation in the EU.

Despite recent attention to trilogue transparency, very little progress has taken place. No effort has so far been able to engage seriously with the objective of improving the visibility of the final stages of law-making in a comprehensive manner, encompassing all three institutions. Currently, the lack of a clearly defined legislative framework dictating which legislative documents should be disclosed and when enables nearly all of them to be kept confidential until negotiations have ended. A joint legislative database that would be crucial for identifying the relevant documents and seeking access to them has not advanced, even though the deadline set by the institutions themselves expired six years ago. This is not a question of a lack of technical solutions but about a lack of interest in finding such solutions. Instead, “outsiders” are left to seek documents one by one, facing strategic delays from the institutions handling them, and then appealing the denials one by one, which is both time-consuming and involves financial risk. This reluctance of the institutions to apply Court rulings beyond individual cases is frustrating. Citizens’ are far from having the right to be able to “scrutinize all the information which has formed the basis of a legislative act”.