Overseeing Trilogues: What Is the Impact of Administrative and Judicial Watchdogs?, by Maarten Hillebrandt

An analysis of the interventions of the European Ombudsman and of the Court of Justice of the EU in the area of trilogue (in)transparency suggests that institutional structures and dependencies may have a discernible impact on oversight and enforcement styles.

The trilogue method: beating heart of the EU legislative process

Over the past decades, a growing share of EU legislative procedures is marked by an informal stage in which negotiating teams of the European Parliament and Council, facilitated by the Commission, negotiate a closed-door deal that is subsequently ‘rubber-stamped’ by the respective legislative institutions. Known as the ‘trilogue’, this process has been applauded by its institutional proponents for increasing legislative efficiency, understood in terms of legislation time and the ability to reach compromises that are satisfactory for all three institutions.

Thus, the trilogue procedure has gained a prominent role in EU legislation. Indeed, according to data presented in a recent dissertation by Alexander Hoppe (Utrecht University), a whopping 99% of the EU’s laws adopted between 2014 and 2019 were ironed out in trilogues.

At the same time, the extreme preponderance of this phenomenon has also been met with considerable unease, from within both civil society and academic circles. As critics point out, the secrecy surrounding trilogue negotiations makes it largely impossible for the wider public – and, at best, much more difficult for national parliaments – to hold the EU’s Member States to account for their role in the negotiated outcome. Fault lines of agreement and disagreement with elements in the proposal are kept outside of the public view. Moreover, the precious little transparency that marks earlier stages of the legislative process – when MEPs, on the one side, and Member States, on the other, agree among themselves on a common position –  is largely undone by last-minute changes to the legislative proposal. These are often far-reaching but are, without exception, arrived at in secrecy.

It is therefore unsurprising that legal scholarship has strongly criticised the lack of transparency that stems from this informal negotiating method. Curtin and Leino-Sandberg, for example, have described trilogue secrecy as lacking a solid normative justification, arguing that ‘[t]he time may well be ripe for efficiency to beat a chequered retreat’.

Mapping the state of the art on trilogues

Political scientists also take a keen interest in the trilogue method, certainly for normative reasons, as well as on empirical grounds. In this context, political scientists Gijs Jan Brandsma, Justin Greenwood, Ariadna Ripoll Servent, and Christilla Roederer-Rynning convened authors for a special issue to take stock of the state of the art in trilogue research. The special issue, entitled ‘Inside the “Black Box” of EU Legislative Trilogues’, appeared earlier this year in the Journal of European Public Policy.

In a contribution to this special issue co-authored with my Helsinki Law School colleague Päivi Leino-Sandberg, we consider what role EU administrative and judicial watchdogs have played in clarifying the EU legislator’s legal transparency obligations during the trilogue stage of the legislative process. The article, entitled ‘Administrative and Judicial Oversight of Trilogues’, departs from the observation that EU Treaty Law (notably Article 1 TEU and Article 15 TFEU) and the access to documents act (Regulation 1049/2001) impose certain obligations on the European legislator, which certain watchdogs – namely, the European Ombudsman, the European Commission, and the Court of Justice of the European Union – are charged to enforce. The article thus tracks the interventions that these watchdogs have made with regard to clarifying transparency obligations during the trilogue stage and evaluating their impact along three dimensions: (i) the interpretation of key legal concepts; (ii) the interpretation of transparency procedures; and (iii) legal remedies against breaches of institutional transparency obligations. Finally, we make a number of tentative remarks and observations with regard to differences in enforcement style of the watchdogs in question.

Trilogue problems and the available routes to addressing them

Overall, the legislative institutions act in contravention to the constitutional principle of legislative transparency when they negotiate in trilogues under the current set-up. They routinely fail to immediately and fully disclose documents drawn up in the context of trilogue negotiations – even when Article 12(2) of Regulation 1049/2001 contains a clear duty to proactively disclose all legislative documents in a document register. It is true that the access regulation does leave institutions the possibility to refuse request-based access in order to protect the decision-making procedure (under Article 4(3), the so-called ‘space to think’ clause). However, the requisite standard of justification established in the case law – which entails that refusals should be exceptional, specifically and precisely motivated, based on non-hypothetical risks, and not overridden by a public interest in knowing the requested information – clearly does not square with the nearly-categorical tendency to delay the disclosure of trilogue documents until the sensitive elements of the proposal have been settled, or even until the act has been passed.

Based on a survey of interventions of the above-mentioned oversight bodies, it became clear that the Commission, in its capacity of guardian of the Treaties, has played no role whatsoever in overseeing the transparency of trilogues to the requisite legal standard. On the contrary, as a facilitator of the legislative process, it played an active role in these trilogues themselves, and defended the practice by intervening on the European Parliament’s side in the De Capitani case. In practice, it is very difficult – if not impossible – to induce the Commission into action in this matter. It is further questionable whether the Commission possesses adequate instruments for doing so.

The Ombudsman and Court of Justice, on the other hand, have made various interventions over the course of roughly two decades. These interventions came in the form of (i) individual remedies – when applicants bring, respectively, a complaint or a case; and (ii) systemic remedies – which occur when the Ombudsman conducts an own-initiative inquiry, or when a court judgment contains findings with a wider, precedent-setting applicability.

Two watchdogs and how they guard the premises

Given these partially overlapping instruments, the two trilogue transparency watchdogs have assumed remarkably different oversight and enforcement styles.

The Ombudsman began in the early 2000s by settling various definitional and procedural questions. These questions related to: the Council’s duty to legislate openly; the position of the Presidency; and permissible exceptions to the direct disclosure of legislative documents through the online register. More recently, the Ombudsman used the own-initiative instrument to highlight the trilogue transparency problematics, first directly (inquiry concluded in 2016), and thereafter in relation to the Council’s role in the legislative process (inquiry concluded in 2018). Recourse to the inquiry instrument provided her with a platform for high-profile interventions, allowing her to place the wider legislative transparency question squarely on the institutional agenda.

Upon a close reading, however, it seems hard to avoid the impression that the general tone and conclusions of the two inquiries differ markedly. In the first inquiry, the Ombudsman pleads for a general confidential ‘space to think’ and negotiates between the two legislative institutions – the Parliament and the Council. At the same time, she considers a scenario of non-disclosure throughout trilogue negotiations reasonably conceivable. This position, even when taken before the appearance of the Court’s far more stringent De Capitani judgment (see below), appears strikingly detached from the letter and logic of Regulation 1049/2001, which requires a strict and case-by-case assessment of the applicability of non-disclosure grounds rather than a categorical acceptance of ‘space to think’ arguments. This permissiveness starkly contrasts with the Council transparency inquiry issued two years later. Here, the Ombudsman promulgates the direct and (in principle) full disclosure of Council legislative documents as ‘a legal requirement’ from which the Council may only deviate under strictly circumscribed conditions.

The Court of Justice – constrained by its passive role of adjudicating only on questions when they are placed before it – has, by contrast, taken a far more piecemeal approach. Its judgment in the above-mentioned De Capitani case has been most important in this regard. In De Capitani, a former EP staff member challenged the Parliament’s refusal to grant access to the so-called ‘four-column documents’, which track progress made in trilogue negotiations. The Court agreed with the applicant’s conceptual position that trilogues, even when informal, must be considered an integral part of the legislative process, and its documents should be treated as legislative documents. Building on the previous access to documents case law of Sweden and Turco v Council (2008) and Council v Access Info Europe (2013), the Court was then able to highlight the elevated democratic importance of providing access to the requested documents.

One matter, however, that the Court evaded in De Capitani was the question of proactive (rather than request-based) disclosure obligations. It did so by simply observing that the case at hand concerned a dispute over an access request. This may be considered a regrettable omission, all the more because applicants must base their case on a grievance affecting them directly, and therefore lack standing to enforce the institution’s duty to proactively disclose information.

Different methods, different vulnerabilities

Overall, the analysis shows that both the administrative and the judicial route to oversight have brought substantial gains to the conceptual, procedural, and remedial clarification of trilogue transparency obligations.

Conceptually, trilogues have been classified as part of the legislative process, and trilogue documents as fully recognised legislative documents. Meanwhile, both watchdogs have clearly established that the Commission is formally a participant in the legislative process. Differently than with tango, it takes three to trilogue.

Procedurally, the Court has brought trilogue documents within the purview of the stringent review test pertaining to legislative transparency. Having previously found that legislative documents cannot generally be considered sensitive, in De Capitani it reiterated that access requests for specific trilogue documents must be assessed on their individual merits, applying potential justifications for non-disclosure very strictly. At the same time, neither of the watchdogs has thus far made definitive statements regarding the duty to proactively disclose trilogue documents.

Remedially, the Ombudsman has, against misgivings expressed by the Council, established a fast-track procedure for access to documents complaints in order to facilitate their swift handling. The Court, by contrast, tends to take far longer to adjudicate on transparency-related grievances. At the same time, it has tended to take a protective view of applicants’ rights, so long as it concerns individual access requests rather than general proactive disclosure duties.

The different oversight styles of the EU’s transparency watchdogs may give way to some tentative observations concerning their effectiveness and structural dependencies. The Commission might have acted in defence of the Treaties, had it not worn the double hat of participant (thereby becoming a stakeholder) in the legislative procedure. Consequently, its latter role likely diminishes the effectiveness of the former role. The Ombudsman, in turn, has tended to take an increasingly expansive role in defining the concept of maladministration, leading to a visible and sometimes high-profile defence of citizens’ access rights. At the same time, this comes with accusations – particularly from the Council – of mandate-stretching and violation of procedural rights. An elected watchdog office might here invite behaviour that some construe as partisanship. Finally, in the Court of Justice, the large number of judges and the attention devoted to consistent legal interpretation over time have likely had a moderating effect on the nature of judicial oversight. Yet, the passive nature of the Court can still be seen to create indeterminate and fragmented enforcement of transparency obligations on specific points.

In the end, it would seem that there is no silver bullet for the enforcement of trilogue transparency. Watchdogs must balance between engagement and oversight drift, activism and partisanship, and restraint and fragmentation. For this reason, external oversight is probably best served by calibrating between different actors, methods, and styles.

Posted by Maarten Hillebrandt, University of Helsinki

Maarten Hillebrandt is postdoctoral researcher at the Eric Castrén Institute, University of Helsinki, where he participates in research projects on EU transparency and separation of powers. The latter takes place in cooperation with the Universities of Amsterdam and Gothenburg. Maarten is the editor of the Open Government in the EU blog, which tracks news, research, and debates related to transparency and participation in European decision making since 2010. He is also a member of the Standing Committee of Experts on International Immigration, Refugee and Criminal Law, a think tank of legal scholars and practitioners based in the Netherlands. Maarten has published extensively on transparency questions in relation to the Council of the EU.

Suggested citation : M. Hillebrandt, “Overseeing trilogues: What is the impact of administrative and judicial watchdogs?”, published on REALaw.blog and available at https://realaw.blog/?p=598