Microsoft Word - Trilogue study version with foreword and updated list of content

European Economic and Social Committee

Investigation of informal trilogue

negotiations since the Lisbon Treaty –

Added value, lack of transparency

and possible democratic deﬁcit

STUDY

Investigation of informal trilogue negotiations

since the Lisbon Treaty -

Added value, lack of transparency and

possible democratic deficit

July 2017

CONTRACT No. CES/CSS/13/2016 23284

The information and views set out in this study are those of the authors and do not necessarily reflect the

official opinion of the European Economic and Social Committee. The European Economic and Social

Committee does not guarantee the accuracy of the data included in this study.

Neither the European Economic and Social Committee nor any person acting on the European Economic

and Social Committee’s behalf may be held responsible for the use, which may be made of the

information contained therein.

Document information

STUDY FOR The European Economic and Social Committee

(EESC)

CONTRACT NUMBER CES/CSS/13/2016 23284

DATE July 2017

MAIN CONTRACTOR Dr Maja Kluger Dionigi, Think Tank EUROPA,

Copenhagen

SUBCONTRACTOR Dr Christel Koop, Department of Political

Economy, King’s College London

AUTHORS Maja Kluger Dionigi

Christel Koop

CONTACTS Maja Kluger Dionigi – mkr@thinkeuropa.dk

Christel Koop – christel.koop@kcl.ac.uk

Foreword

Trilogues have long raised questions about transparency, political accountability and democratic

participation, or rather the lack thereof, in EU law-making. The Council, Parliament and Commission

meet in closed settings to negotiate EU legislation and facilitate its adoption at an early stage of the

legislative process. Virtually all legal acts today are brokered in trilogues, even though the practice has

no legal anchoring in the European Treaties and moves a key element of EU law-making, the

interinstitutional political compromise, out of public scrutiny into a black box.

Our request for the Committee to commission the present study was sparked by widespread concern

among many of our fellow members in the European Economic and Social Committee (EESC) and by

growing public interest and debate. Our intention was to collect new empirical information on the

development and use of trilogues that would help enhance understanding of the issue within the

Committee and further the institutional, academic and public debate. This study is the first to

comprehensively investigate the evolution and patterns of trilogues from the 1990s to today. The aim is

to discuss their added value and possible democratic deficit, and derive policy recommendations for how

the concerns of civil society could be addressed.

As the study will show, trilogues have not only become normal operating procedure, but are now one

of the most important instruments for decision-making in the EU today. At the same time, they

conflict with fundamental principles of democratic law-making. If EU policy-makers and legislators

are to uphold their commitment to open and inclusive legislative conduct, solutions to this tension

need to be found in order to reconcile the legitimate demands of civil society for transparency,

participation and accountability with efficiency in law-making. This study serves as an important

source of information that will help achieve this objective.

Bernd Dittmann Erik Svensson

Member of the EESC Employers' Group Member of the EESC Employers' Group

Federation of German Industries (BDI) Almega – the Employers’ Organisation for the

Swedish Service Sector

Investigation of informal trilogue negotiations since the Lisbon Treaty

Abstract

This study focuses on the use of trilogues and early agreements in the European Union (EU). Today,

trilogues form the standard operating procedure for reaching agreements between the European

Commission, European Parliament, and the Council of the EU. The use of trilogues has long raised

concerns about public transparency and accountability. Much has already been done to improve the

way in which each institution’s negotiating team is held accountable to their respective institutions.

However, there is still scope for improving the transparency of trilogue meetings. The purpose of this

study is three-fold. First, we provide an overview of the recent developments in the use of trilogue

meetings to reach early agreements. We provide a detailed descriptive statistical overview of the use

of early agreements to conclude legislation in the period of 1999 to 2016, negotiated under the

ordinary legislative procedure. Second, we analyse EU policy initiatives taken in the areas of

transparency and accountability. Third, we suggest several avenues for improving the transparency

and accountability of trilogues. The study draws on an extensive review of both academic and non-

academic literature on the use of trilogues and early agreements. We also present some new

quantitative and qualitative data to shed light on the use of trilogues and early agreements.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Table of contents

List of tables ..................................................................................................................................................... 5

List of figures ................................................................................................................................................... 5

Executive summary ......................................................................................................................................... 6

The background of the study ............................................................................................................................. 6

Method............................................................................................................................................................... 7

Findings of the study ......................................................................................................................................... 8

Recommendations of the study ......................................................................................................................... 9

Abbreviations................................................................................................................................................. 11

Glossary.......................................................................................................................................................... 13

1. Introduction ........................................................................................................................................... 15

1.1 The focus of the study ............................................................................................................................ 16

1.2 Sources and methods ............................................................................................................................. 18

2. The rise of trilogues and early agreements.......................................................................................... 20

2.1 The ordinary legislative procedure and the emergence of trilogue meetings ......................................... 20

2.2 What differentiates informal and formal politics?.................................................................................. 26

2.3 Why have early agreements emerged? .................................................................................................. 29

3. Patterns of early agreement.................................................................................................................. 33

3.1 The volume of legislation: The use of early agreements has increased over time ................................. 33

3.2 The duration of decision-making: Early agreements now take longer to be concluded......................... 35

3.2 Early agreements and European Parliament committees........................................................................ 38

3.3 Early agreements, type of legislation and complexity............................................................................ 40

3.4 Recent findings on the use of early agreements .................................................................................... 44

4. Rules and practices in early agreement................................................................................................ 46

4.1 Rules governing access to documents and their interpretation.............................................................. 46

4.1.1 EU case-law........................................................................................................................................ 47

4.1.2 Trilogues and access to documents .................................................................................................... 49

4.2 Inter-institutional rules and practices .................................................................................................... 51

4.3 Intra-institutional rules and practices .................................................................................................... 55

4.3.1 Rules and practices in the Commission.............................................................................................. 57

4.3.2 Rules and practices in the European Parliament................................................................................. 57

4.3.3 Rules and practices in the Council ..................................................................................................... 64

Investigation of informal trilogue negotiations since the Lisbon Treaty

5. Concerns about early agreements raised in the public domain......................................................... 68

5.1 Early agreements and the ‘speed’ of negotiations .................................................................................. 68

5.2 Early agreements and deliberation ........................................................................................................ 69

5.3 Early agreements and national parliamentary scrutiny.......................................................................... 71

5.4 Early agreements and transparency ....................................................................................................... 72

6. Conclusion.............................................................................................................................................. 76

6.1 Key findings .............................................................................................................................................. 76

6.2 Suggestions for improving the transparency and accountability of early agreements............................... 79

References ...................................................................................................................................................... 81

Annex A: List of interviewees....................................................................................................................... 90

Investigation of informal trilogue negotiations since the Lisbon Treaty

List of tables

Table 1: The increase of the EP's formal powers ............................................................................................ 21

Table 2: Current rules and practices governing trilogues within the Parliament and the Council .................. 56

Table 3: Key changes to the EP’s formal rules................................................................................................ 61

List of figures

Figure 1: The ordinary legislative procedure and early agreements................................................................ 17

Figure 2: Percentage of early agreements per half year (1999-2016).............................................................. 23

Figure 3: Percentage of early agreements per stage of conclusion and parliamentary term............................ 24

Figure 4: Number of ordinary legislative files and early agreements per legislative year .............................. 34

Figure 5: Duration (days) by early agreement and parliamentary term........................................................... 35

Figure 6: Duration (days), excluding unfinished files of previous parliaments .............................................. 37

Figure 7: Early agreements by responsible EP committee and parliamentary term........................................ 38

Figure 8: Early agreement and type of legislation (1999-2016)...................................................................... 40

Figure 9: Number of recitals, by early agreement and parliamentary term..................................................... 41

Figure 10: Early agreements and legislative instruments (2009-2016)........................................................... 42

Figure 11: Mandatory consultation of the EESC and early agreement (2009-2016) ...................................... 43

Investigation of informal trilogue negotiations since the Lisbon Treaty

Executive summary

The background of the study

This study looks at the EU institutions’ use of trilogues and early agreements. Two features define early

agreements: (1) they are concluded early during the ordinary legislative procedure (either at first reading or

early second reading) and (2) the compromise is reached during informal trilogue meetings. In trilogue

meetings, representatives from the European Parliament (EP), the Council of the European Union (Council)

and the European Commission (Commission) negotiate and reach deals in closed meetings. These

agreements have to be formally approved by the ministers in the Council and the EP’s plenary before they

become binding. Today the use of trilogues has become the standard operating procedure for reaching

legislative agreements between the EP, the Council and the Commission. In the first half of the current

parliamentary term (2014-2016), 75 percent of legislation was adopted following an early agreement.

The increased use of trilogues to reach early compromises has long raised concerns about transparency and

accountability. Trilogues do not have legal reference in the 2009 Lisbon Treaty or in any earlier treaty.

However, a number of commitments and claims have been made in the Treaty about the transparency of EU

decision-making. Under the Lisbon Treaty (e.g., Article 15 TFEU), there is an obligation for all EU

institutions to conduct their work as openly as possible and to make documents relating to the legislative

procedure publicly available. The Lisbon Treaty also seeks to improve the EU’s democracy by strengthening

the EP, increasing the involvement of national parliaments, introducing elements of direct democracy into

EU decision-making (i.e., the European Citizens’ Initiative), and aiming to consult external stakeholders as

widely as possible during the decision-making process. This raises the question of what the state of informal

trilogue meetings is. In their current form, what do these meetings add to the legislative process and how can

we evaluate their transparency and democratic quality?

The transparency and legitimacy of trilogues have again become topical with the European Ombudsman’s

investigation into the use of trilogues in 2016, the Interinstitutional Agreement on Better Law-Making of

2016, and the EP’s revision of its Rules of Procedure in December 2016 – all aimed at boosting the

transparency of legislative procedures.

Against this backdrop, the EESC has commissioned this study entitled ‘Investigation of informal trilogue

negotiations since the Lisbon Treaty – Added value, lack of transparency and possible democratic deficit’.

As requested, the study covers the following three elements:

1. An overview of the recent developments in the use of trilogue meetings to reach early agreements,

particularly since the Lisbon Treaty;

2. An analysis of EU policy initiatives in the areas of transparency and criticism of democratic deficit

(that is, the involvement of the whole EP and other institutions, such as the EESC);

3. Based on the above, recommendations for how and whether transparency and accountability could

be achieved.

Investigation of informal trilogue negotiations since the Lisbon Treaty

To address these three elements, we offer a systematic overview and analysis of the patterns, challenges,

rules, and practices of early agreements. We address the three aspects in the following way:

 By providing an overview of recent developments in the use of trilogues. First, we provide a

descriptive statistical overview of all legislation finalised from 1999 until the end of 2016. We then

explain why and when early agreements are used, particularly since the 2009 Lisbon Treaty

(Chapters 2 and 3);

 By examining the policy initiatives taken to address concerns of transparency and accountability.

First, we describe and discuss the policy initiatives taken by the EU institutions to address these

concerns (Chapter 4). We then examine concerns raised by external stakeholders such as interest

groups, about the use of trilogue meetings (Chapter 5);

 By suggesting potential avenues for strengthening transparency and accountability of early

agreements (Chapter 6).

Method

The study relies on an extensive review of both scholarly and practitioners’ literature on trilogues and early

agreements. We also use new quantitative and qualitative data in the report.

To thoroughly assess the pattern of early agreements today and over time, we used quantitative data on all

finalised legislation agreed under the ordinary legislative procedure between 1999 and 2016 (Chapters 2 and

3). Data between 2009-2014 had already been collected as part of previous academic studies, and we

collected data for the first two and a half years of the current parliamentary term, covering the 2014-2016

period.

To understand the concerns of external stakeholders about trilogues and early agreements, we also conducted

three analyses of (1) the bi-annual reports from the Conference of Parliamentary Committees for Union

Affairs (COSAC) since 2009, (2) the Brussels online media’s coverage of trilogue meetings after the Lisbon

Treaty, and (3) the 50 submissions to the public consultation of the European Ombudsman’s investigation

into transparency of trilogues in 2016 (Chapter 5).

Finally, to gain a deeper understanding of the Commission’s, Council’s, and EP’s practices during trilogues,

we conducted 18 interviews with EU policy-makers in Brussels in April 2017 (please see Annex A). Many

of the practices on trilogues inside and between the EU institutions are not written down, which highlights

the importance of interviewing people inside the institutions to understand the practical conduct of trilogues.

All of the interviewees were guaranteed anonymity and confidentiality, which allowed them to speak more

freely. Information gained from the interviews is particularly used in Chapter 4.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Findings of the study

In Chapters 2 and 3, we find that:

 The use of early agreements has increased rapidly between 1999 and 2008 and has stabilised since

2008;

 Looking at the full period between 1999 and 2016, procedures with early agreements have taken less

time than procedures without early agreement. However, since 2014, procedures with early

agreements have taken significantly longer than procedures without early agreements. Most non-

early agreements concern uncomplicated legislation which can be adopted quickly and without

(major) changes;

 Since 2009, early agreements have become the default option for almost all the EP committees, even

for committees that have only gained full co-decision powers with the Lisbon Treaty;

 Files concluded by early agreement since the Lisbon Treaty came into force in December 2009 are

characterised by higher levels of complexity than files that are not subject to early agreements.

Existing academic and non-academic studies show that the popularity of early agreements reflects a number

of concurrent factors, most notably: (1) growing levels of trust and cooperation between the Council and the

EP, (2) efficiency gains, and (3) the desires of the co-legislators’ negotiating teams to be seen as successful

and reliable negotiators.

In Chapter 4, we show that the rules and practices on trilogues within the Council and the EP have become

more ‘institutionalised’ over time and have, at least to some extent, addressed the democratic deficit

associated with early agreement. In the early 2000s, the Council’s and the EP’s negotiating teams in trilogue

were largely able to negotiate in an unconstrained institutional environment. This is no longer the case: there

are elaborate intra-institutional mechanisms in place to secure the representativeness, accountability, and

transparency of the negotiating teams vis-à-vis their respective institutions. In the EP, trilogues have become

more inclusive (by including all political groups), representative of the whole Parliament (by having the

initial negotiating mandate endorsed in plenary), and accountable (by requiring the negotiating team to

regularly report back to their committee and political groups). This has made it easier for Members of the EP

(MEPs) not present in trilogues to follow what is going on and to hold negotiating teams to account. It has

also made it easier for external stakeholders (such as interest groups) and interested members of the public to

follow the progress made in trilogues by following the web-streamed committee and plenary meetings,

where updates are given by the negotiating team to other MEPs.

Chapter 4 also shows that, unlike the EP, the Council has not included its rules of procedure in its working

procedures on trilogues. Even though they are not written down, there are widely shared and entrenched

practices within the Council in terms of how the presidency (representing the Council in trilogues) obtains its

mandate and reports back to the Council after each trilogue meeting. Unlike the EP and the Commission, the

Council’s mandate is not always publicly available as only mandates given by means of a General Approach

(when the initial negotiating mandate is approved by the ministers) are public.

Investigation of informal trilogue negotiations since the Lisbon Treaty

The EESC or other EU institutions (other than the Commission, Council, and the EP) neither have access to

trilogue meetings nor enjoy a privileged prerogative in trilogues. They have the same standing as external

stakeholders.

In Chapter 5, we show that a number of concerns exist in the public domain about trilogue meetings, in

particular that:

 Early agreements are perceived to result in more delegation of powers to the Commission. This is,

however, difficult to establish empirically. A proper examination of this conjecture would require a

comparison of the delegation of powers to the Commission under conditions of early and non-early

agreements. As 75 percent of all legislation is currently concluded by early agreements, we do not

have non-early agreements with similar features to compare them against;

 Early agreements are perceived to reduce the quality of legislation due to the limited use of impact

assessment on substantive amendments introduced after the Commission has put forward its

proposal. This concern is not only confined to legislation concluded by means of an early

agreement, but to all legislation that changes considerably after the introduction of the Commission

proposal. More frequent impact assessments on substantive amendments introduced later in the

policy process may, indeed, prevent the potential negative consequences of amendments that have

not been accounted for in the Commission’s ex-ante impact assessment;

 Some national parliamentary chambers find it more challenging to conduct effective parliamentary

scrutiny. This is because they rely on their government to provide them with information about

trilogues and to give them access to LIMITE (classified) documents;

 The efficiency of early agreements is seen to come at the expense of transparency. The absence of an

official paper trail during trilogue meetings indeed makes it challenging for external stakeholders

(such as interest groups) to follow what happens during trilogue meetings and to fully grasp how

inter-institutional disagreements have been solved. This creates an uneven playing field between

stakeholders that are well-connected and resourceful and those that are not.

Recommendations of the study

A number of changes could be introduced to make trilogues more transparent to the outside world. This

would also facilitate the account-holding processes in national parliaments as well as in the plenary of the EP

and the rest of the Council. In particular, increased transparency on the initial position of the Council and

public access to the four-column document after the conclusion of negotiations would make it easier for

external stakeholders to understand the positions of all the participants in trilogue meetings and enhance the

understanding of how differing positions are reconciled.

In Chapter 6, we suggest two main avenues for improving transparency of trilogues:

 To make the Council’s initial mandate public. The Council could aim to make its initial mandate

publicly available before the first trilogue negotiations, as is the case for mandates adopted as a

General Approach at the ministerial level. Currently, when mandates are adopted solely at the level

of Coreper, they are not always made public.

Investigation of informal trilogue negotiations since the Lisbon Treaty

 To create a joint database on legislative files. As part of the institutions’ commitment in the

Interinstitutional Agreement on Better Law-Making of 2016 to create a joint database – merging

existing databases – the following information about trilogues may be included:

o Trilogue dates could be published ex-ante where they are known, together with an annotated

trilogue agenda;

o Four-column documents could be made automatically available in the joint database when

legislation is finalised, so that stakeholders do not have to apply for access to these

documents;

o A user-friendly database could be set up, allowing documents to be found without knowing

the specific document numbers (as is currently the case for the Council Register). Once the

database is up and running, it could provide an e-mail newsletter sign-up service with

updates on the trilogues and legislative process. This would foster a level playing field

among interest groups by making access to information easily available to everyone; not just

to the well-connected and resourceful ones.

Similarly to the European Ombudsman, we acknowledge the importance of trilogues for facilitating

negotiations and compromises between the EU institutions, but we also highlight the difficulties that wider

audiences face in finding out when trilogues take place, what is being discussed, and by whom.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Abbreviations

AFET European Parliament Committee on Foreign Affairs

AFCO European Parliament Committee on Constitutional Affairs

AGRI European Parliament Committee on Agriculture and Rural Development

ALDE Group of the Alliance of Liberals and Democrats for Europe

BUDG European Parliament Committee on Budgets

DEVE European Parliament Committee on Development

CJEU Court of Justice of the European Union

Commission European Commission

CONT European Parliament Committee on Budgetary Control

Coreper Committee of Permanent Representatives

COSAC Conference of Parliamentary Committees for European Affairs

Council Council of the European Union

CULT European Parliament Committee on Culture and Education

EESC European Economic and Social Committee

ECON European Parliament Committee on Economic and Monetary Affairs

EMPL European Parliament Committee on Employment and Social Affairs

ENVI European Parliament Committee on Environment, Public Health, and Food Safety

EP European Parliament

EP5 Fifth (European) parliamentary term, 1999-2004

EP6 Sixth (European) parliamentary term, 2004-2009

EP7 Seventh (European) parliamentary term, 2009-2014

EP8 Eighth/current (European) parliamentary term, 2014-2019

EPP Group of European People’s Party

ETUC European Trade Union Confederation

EU European Union

FEMM European Parliament Committee on Women’s Rights and Gender Equality

Greens/EFA Groups of the Greens/European Free Alliance

GRI Inter-Institutional Relations Group (Groupe des Relations Interinstitutionelles)

IMCO European Parliament Committee on Internal Market and Consumer Protection

INTA European Parliament Committee on International Trade

ITRE European Parliament Committee on Industry, Research, and Energy

JURI European Parliament Committee on Legal Affairs

JURI & IM European Parliament Committee on Legal Affairs and Internal Market

LIBE European Parliament Committee on Civil Liberties, Justice and Home Affairs

MEP Member of the European Parliament

NGO Non-governmental organisation

PECH European Parliament Committee on Fisheries

REFIT European Commission’s 2012 Regulatory Fitness and Performance Programme

REGI European Parliament Committee on Regional Development

SD Standard deviation

S&D Group of the Progressive Alliance of Socialists and Democrats in the European

Investigation of informal trilogue negotiations since the Lisbon Treaty

Parliament (abbreviated PES before the eighth parliamentary term)

TEU Treaty of the European Union (originally signed in Maastricht in 1992)

TFEU Treaty on the Functioning of the European Union (originally signed in Rome in

1958 as the Treaty establishing the European Economic Community)

TRAN European Parliament Committee on Transport and Tourism

Investigation of informal trilogue negotiations since the Lisbon Treaty

Glossary

B-point – An item on the agenda of the Council that requires discussion before approval by the ministers. B-

points are different from A-points, the latter being be approved by the Council without debate, although

ministers can express their opinion on the item at the time of the approval.

Comitology – A set of procedures through which the Commission exercises the implementing powers

conferred to it by the Council and the Parliament. The implementation is scrutinised by committees with

representatives from the member states.

Coreper – Committees of the permanent representatives of member states to the EU, which meets weekly

and is responsible for the preparation of the work of the Council. Depending on the policy area, the

committee takes the form of Coreper I or Coreper II. The committee discusses (proposed) EU legislation,

serves as intermediary between the member states’ capitals and Brussels-based EU institutions, supervises

the work of groups of experts, and monitors the work of working parties and committees consisting of

member state officials who prepare legislative dossiers at the technical level.

Early agreements – ordinary legislative dossiers that have the following two features:

1) They are concluded early during the ordinary legislative procedure, either at first reading or at early

second reading;

2) An inter-institutional compromise is reached on the contents during trilogue meetings.

Four-column document (also known as multi-column document) – A working document that forms the basis

of the discussions between the Parliament, the Council and the Commission in trilogues on ordinary

legislative files. Four-column documents are produced at different stages of the legislative process, and

include the positions of the three institutions as well as the compromise made on the text of the proposal, as

negotiated in trilogues.

General Approach – A political agreement reached in the Council on its first reading position at the

ministerial level. If such an agreement is reached, and inter-institutional negotiations are started in trilogues,

the agreement serves as the negotiating mandate for the presidency.

Ordinary legislative procedure (formerly called ‘the co-decision procedure’) – The standard decision-

making procedure in the EU since the Lisbon Treaty came into force in 2009. Following this procedure, the

Commission tables a legislative proposal and both the Council and the Parliament have a deciding vote on

the (amendments to the) proposal.

Political trilogue – A trilogue meeting that focuses on the political aspects of an ordinary legislative dossier

and is attended by political representatives of the Parliament, the Council and the Commission.

Rapporteur – The MEP responsible for a certain dossier on behalf of the relevant committee of the European

Parliament. The rapporteur is responsible for the preparation of the committee report on a Commission

proposal and acts as EP representative in (political) trilogue meetings.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Shadow rapporteur – An MEP responsible for following the work on a certain dossier on behalf of an EP

political group, which does not hold the rapporteurship on the dossier. Shadow rapporteurs coordinate and

negotiate with the rapporteur and are invited to attend the (political) trilogue meetings on the dossier.

Technical trilogue (also known as technical meeting) – A trilogue meeting that focuses on the technical

aspects of an ordinary legislative dossier, including the formulation of the political agreement reached in

political trilogues, and which is attended by technical-level representatives of the Parliament, the Council

and the Commission.

Trilogues – Informal and secluded meetings on ordinary legislative dossiers attended by representatives of

the Parliament, the Council and the Commission.

Investigation of informal trilogue negotiations since the Lisbon Treaty

1. Introduction

The introduction of the co-decision procedure with the Maastricht Treaty in 1993, and its gradual extension

to new areas with successive treaty revisions is largely viewed as a major step forward in attempting to

consolidate European democracy by empowering the European Parliament’s (EP) legislative powers. With

the Lisbon Treaty of 2009, co-decision was renamed as ‘the ordinary legislative procedure’, covering almost

all European Union (EU) policy areas and elevating the EP to a genuine co-legislator with the Council of the

EU (hereafter Council). However, the ordinary legislative procedure has also brought with it new

developments that have put into question the transparency and accountability of the EU’s decision-making.

First, there has been a steady rise in first reading agreements since the 1999 Amsterdam Treaty, when it

became possible to finalise legislation at first reading. Second, the use of informal trilogues – where the

European Commission, the EP, and the Council negotiate and reach deals in closed meetings – has grown

exponentially over time.

Today, the use of trilogues is a standard operating procedure for making legislative deals between the three

institutions at an early stage of the legislative process. While this has greatly expedited the ordinary

legislative procedure by reducing the number of readings and facilitating compromises early in the process,

the use of trilogues has also raised transparency and legitimacy concerns. Political dissatisfaction with early

agreements appeared in the EP in the early 2000s

and later attracted criticism in the public sphere. In the

Brussels media, the use of trilogues has been referred to as ‘infernal, undemocratic’ and ‘secret’.

This

criticism is serious as the majority of EU legislation is agreed upon in trilogues, and policymakers view them

as essential for facilitating compromises between the institutions.

The transparency and legitimacy of trilogues and early agreements have again become topical with the

European Ombudsman’s (Emily O’Reilly) recent investigation into the use of trilogues, the recent

Interinstitutional Agreement, and the EP’s revision of its Rules of Procedure in December 2016 – all aimed

at boosting transparent law-making in the EU.

The 2009 Lisbon Treaty in many ways sought to improve the EU’s democracy by strengthening the EP,

increasing the involvement of national parliaments, and introducing elements of direct democracy into EU

decision-making (i.e., the European Citizens’ Initiative). A number of articles in the Lisbon Treaty explicitly

state that the EU institutions should operate as transparently as possible and have an open, transparent, and

regular dialogue with external stakeholders.

These various developments raise two questions: (1) What is

the current state of informal trilogue meetings? (2) In their current form, what do these meetings add to the

legislative process and how can we evaluate their transparency and democratic quality?

Huber, K. and Shackleton, M. (2013) ‘Codecision: A practitioner’s view from inside the Parliament’, Journal of European Public Policy 20 (7):

1040-55.

Mahoney, H. (2013) ‘Complex EU law-making dubbed “infernal, undemocratic”’, 4 March, EUobserver, available at

http://euobserver.com/political/119181 (last accessed 14 April 2017); Fox, B. (2014) ‘Secret EU lawmaking: The triumph of the trilogues’,

EUobserver 4 April 2014, available at https://euobserver.com/investigations/123555 (last accessed: 19 April 2017).

Roederer-Rynning, C. and Greenwood, J. (2015) ‘The culture of trilogues’, Journal of European Public Policy 22 (8): 1151

See for instance Article 11 of the Treaty on the European Union and Article 15 of the Treaty on Functioning of the European Union.

Investigation of informal trilogue negotiations since the Lisbon Treaty

1.1 The focus of the study

This study examines the EU institutions’ use of trilogue meetings and discusses the transparency and

legitimacy concerns raised regarding their extensive use. The study is commissioned by the EESC under the

title ‘Investigation of informal trilogue negotiations since the Lisbon Treaty – Added value, lack of

transparency and possible democratic deficit’. As requested, the study covers the following three elements:

1. An overview of the recent developments in the use of trilogue meetings to reach early agreements,

particularly since the Lisbon Treaty;

2. An analysis of EU policy initiatives in the area of transparency and criticism of democratic deficit

(that is, the involvement of the whole European Parliament and other institutions, such as the

EESC);

3. Based on the above, recommendations for how and whether transparency and accountability could

be achieved.

To address the above points, we offer a systematic examination of the pattern, challenges, rules, and

practices of early agreements today and over time. The individual chapters aim to answer the following

questions:

 Chapter 2: How can we explain the rise of trilogues and early agreements over time?

 Chapter 3: What is the overall pattern of trilogues and early agreements when examining all

finalised legislation since 1999 (when first reading deals became an option)?

 Chapter 4: What policy initiatives have the EU institutions taken to address concerns of

transparency and accountability? What are the current institutional practices of early agreements?

 Chapter 5: What concerns do external stakeholders have about the widespread use of trilogues to

reach an early agreement, and how can we evaluate these?

 Chapter 6 (Conclusion): What are potential avenues for strengthening transparency and

accountability in early agreements?

Chapters 2 and 3 provide an overview of recent developments of early agreements (Point 1 of the EESC’s

commissioned study). Chapters 4 and 5 examine the policy initiatives in the area of transparency and the

criticism of democratic deficit (Point 2). Based on the first five chapters, Chapter 6 provides suggestions for

how and whether transparency and accountability could be achieved (Point 3). It is important to state that our

study only discusses the transparency of EU decision-making as far as it relates to the use of trilogues in

early agreement.

In this study, we focus on early agreements in particular, which are files that are (1) concluded at either first

or early second reading, and (2) based on an informal compromise between the co-legislators negotiated

during trilogue meetings.

This definition excludes procedures that are concluded early without having been

pre-agreed during trilogue meetings. Figure 1 gives an overview of the ordinary legislative procedure. Early

agreements can be reached either at the first or early second reading:

Héritier, A. and Reh, C. (2012) ‘Codecision and its discontents: Intra-organisational politics and institutional reform in the European Parliament’,

West European Politics 35 (5): 1136.

Investigation of informal trilogue negotiations since the Lisbon Treaty

1. First reading early agreements are the product of concluded negotiations before Parliament has

adopted its formal position and before the Council has adopted its common position. If the co-

legislators reach an informal compromise, Parliament includes the Council’s proposition in its own

first reading amendments; hereafter, the Council adopts the Commission’s proposal as amended by

Parliament;

2. Early second reading agreements are informal compromises agreed after the EP has adopted its first

reading position, but before the Council has reached its common position. The EP approves – rather

than amends – the Council’s common position as the Council has already incorporated Parliament’s

amendments into its common position.

Commission proposal

EP position (1st reading)

Amendments by simple majority

Council 1st reading

Common position by qualified majority

EP 2nd reading

Amendments by absolute majority

Commission position

Council 2nd reading

Council does not approve all EP

Conciliation Committee

Agreement

EP + Council 3rd reading

Approves all

amendments by qualified

Act adopted

Approves common

position by simple

Council approves all EP

amendments

Act adopted

Adoption by

absolute/qualified

Rejects common

position by absolute

majority

Act rejected

Council or EP do not

agree

Act rejected

Rejects common

position by absolute

Figure 1: The ordinary legislative procedure and early agreements

Early second reading

agreement may be reached

First reading early

agreement may be reached

Investigation of informal trilogue negotiations since the Lisbon Treaty

1.2 Sources and methods

All chapters in this study rely on an extensive review of both academic and non-academic literature on the

use of trilogues and early agreement. The study also draws on new quantitative and qualitative data.

To assess the patterns of early agreements today and over time, presented in Chapters 2 and 3, we have used

quantitative data on all finalised legislation concluded between 1999 and 2016 and negotiated under the

ordinary legislative procedure. Data between 2009 and 2014 had already been collected as part of previous

academic studies

, and we collected the data for the first two and a half years of the current parliamentary

term (2014-2016). While previous academic studies have analysed the overall pattern of the use of trilogues

before 2014, we are, to the best of our knowledge, the first to analyse the entire period from 1999 to 2016.

This allows us to gather up-to-date information about how and when trilogue meetings are used today and

over time.

To gain a deeper understanding of the co-legislators’ practices during trilogues, covered in Chapter 4, we

conducted 18 interviews with EU policy-makers. These include interviews with two high-ranking officials

from the Council secretariat, six high-ranking officials from six permanent representations in the area of

Coreper I, one Commission official, nine EP officials from seven committees and one official from the

Conciliation and Codecision Unit, and one Member of the European Parliament (MEP). For the Council, we

chose to focus our attention on Coreper I, as it covers the policy areas most often subject to the ordinary

legislative procedure and trilogue meetings. The national diplomats interviewed from the permanent

representations were chosen to have a diverse group of interviewees from countries that differ in size and

geography, with different durations of membership, and with experience in holding the EU presidency. We

did not interview stakeholders such as interest representatives. Instead, we relied on a Brussels media

analysis and the 50 submissions from stakeholders to the public consultation of the Ombudsman’s report to

obtain information on their concerns about trilogues.

In the EP, we chose to interview officials from various committees (please see Annex A for the selected

committees), including both those that have had extensive experience in working under the ordinary

legislative procedure (pre-Lisbon) and those who have acquired co-decision powers after the Lisbon Treaty

in 2009. This helps to reveal whether the amount of experience with ordinary legislative procedures matter

when it comes to internal committee practices in trilogues. The interviewees were ensured anonymity and

confidentiality, which allowed them to speak more freely. This means that information gained during the

interviews can be used as background information in this report to shed light on the practice of trilogues.

Annex A provides a list of the types of actors interviewed for this study. We are grateful to all of the

interviewed officials who kindly gave up their precious time to meet us and gave invaluable insights into the

practices of early agreements in their respective institutions.

To understand the concerns about trilogues and early agreements raised in the public domain, presented in

Chapter 5, we analysed all the Conference of Parliamentary Committees for Union Affairs’ (COSAC’s) bi-

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the European

Union,’ Comparative Political Studies 46 (9): 1112–42; Bressanelli, E., Koop, C., and Reh, C. (2016) ‘The impact of informalisation: Early

agreements and voting cohesion in the European Parliament,’ European Union Politics 17 (1): 91-113; Koop, C., Reh, C. and Bressanelli, E. (2017)

‘When politics prevails: Parties, elections and loyalty in the European Parliament,’ Paper presented at the Fifteenth Biennial Conference of the

European Union Studies Association, Miami, 4-6 May.

Investigation of informal trilogue negotiations since the Lisbon Treaty

annual reports on EU practices and procedures (a total of 19 reports) since the Lisbon Treaty entered into

force, and the 50 submissions to the public consultation of the European Ombudsman as part of her

investigation into the transparency of trilogue meetings. We also analysed the Brussels online media’s

coverage of trilogue meetings since the entering into force of the Lisbon Treaty (a total of 303 articles in

EurActiv, EUobserver, and Politico).

We would like to thank Lauge Pedersen for his assistance in the data collection and the collection of

newspaper articles on early agreements.

Investigation of informal trilogue negotiations since the Lisbon Treaty

2. The rise of trilogues and early agreements

2.1 The ordinary legislative procedure and the emergence of trilogue meetings

The introduction of the co-decision procedure with the Maastricht Treaty in 1993 has profoundly changed

the EU’s institutional balance. Whereas the EP had only had a minor role in the legislative process for

decades, it was now allowed to co-legislate with the Council, gaining the power to amend and veto proposed

legislation. The early days of the co-decision procedure were marked by a ‘clash of cultures’ between the

Council and the EP. In the words of the EP’s former president Nicole Fontaine, ‘the acquisition by

Parliament of full legislative powers was a fabulous adventure, [but] it represented a veritable earthquake for

the Council, which, from 1993 onwards, had to contend with a co-legislator and therefore fundamentally

adapt its legislative mind-set’.

The Council was unaccustomed to sharing legislative responsibilities and

remained largely sceptical about the EP’s ability to handle the co-decision process.

This was largely a

reflection of the EP’s past tendency as a consultative body to take up ‘maximalist habits’, where it would

push for its ideal scenario and take a confrontational stance.

The Co-decision I procedure – introduced with the Maastricht Treaty until it was amended by the 1999

Amsterdam Treaty – still gave the Council the upper hand in negotiations as it could reintroduce its common

position if the two legislators failed to reach agreement in the conciliation committee during the third

reading. When this happened, the EP could either choose to accept the Council’s common position as a

whole or reject it; a situation the EP felt uneasy about and quickly sought to change.

In its Rules of

Procedure, the EP inserted a rule stating that Parliament would ask the Commission to withdraw its proposal

following a failure in the conciliation committee. If the Commission declined and the Council decided to

reinstate its common position, the EP would automatically put forward a motion to reject the Council’s text

at the following plenary session, irrespective of whether the Council’s position was closer to the EP’s policy

position or to the status quo.

The EP was, thus, willing to sacrifice specific legislation to make an example

and to gain long-term institutional powers and respect. This threat came to fruition in July 1994 when

Parliament vetoed the Council’s common position on the draft directive on open network provision (ONP) to

voice telephony after the Council reinstated its common position following a breakdown in conciliation.

Afterwards, the Council never sought to reaffirm its common position again. In practice, this meant that co-

decision ended with the conciliation committee.

It also helps to explain the reform of the co-decision

procedure with the Amsterdam Treaty in 1999 (Co-decision II), which put practice into law by removing the

right of the Council to reaffirm its position after a failure in conciliation. This change characterises the

European Parliament (2013) ‘Twenty Years of Codecision Conference Report’, 5 November 2013, Brussels: European Parliament, Conciliations and

Codecision Secretariat European Parliament, p. 3.

European Parliament (1999) ‘Activity Report 1 November 1993 to 30 April 1999 (4th parliamentary term)’, PE 230.998, Brussels: European

Parliament, Delegations to the Conciliation Committee, p. 5.

Dionigi, M.K. (2017) Lobbying in the European Parliament: The Battle for Influence, London: Palgrave Macmillan; Shackleton, M. and Raunio, T.

(2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation and change’, Journal of European Public Policy 10 (2): 171-88.

Huber, K. and Shackleton, M. (2013) ‘Codecision: A practitioner’s view from inside the Parliament’, Journal of European Public Policy 20 (7):

1040-55; Héritier, A. (2007) Explaining Institutional Change in Europe, Oxford: Oxford University Press.

Farrell, H. and Héritier, A. (2004) ‘Interorganizational negotiation and intraorganizational power in shared decision making: Early agreements

under codecision and their impact on the European Parliament and Council’, Comparative Political Studies 37 (10): 1184-1212; Hix, S. (2002)

’Constitutional agenda-setting through discretion in rule interpretation: Why the European Parliament won at Amsterdam’, British Journal of

Political Science 32 (2): 259-80.

Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation and change’, Journal of European

Public Policy 10 (2): 171.

Hix, S. and Høyland, B. (2011) The Political System of the European Union, Third edition, Basingstoke: Palgrave Macmillan, p. 71.

Investigation of informal trilogue negotiations since the Lisbon Treaty

functioning of the ordinary legislative procedure today: if an agreement cannot be found in conciliation, the

draft act is rejected and the procedure can only start again with a new proposal from the Commission. Table

1 provides an overview of the key changes to the co-decision procedure over time, from its introduction with

the Maastricht Treaty until the Lisbon Treaty.

Table 1: The increase of the EP's formal powers

Treaty (Effective) Powers given to the EP

Treaty of Maastricht

(November 1993)



Co-decision procedure was introduced

 Co-decision areas: internal market, trans-European transport

networks, education, culture, and health

 Union action covered by co-decision: 15 areas

Treaty of Amsterdam

(May 1999)



Simplification of the co-decision procedure by removing the Council’s

option to re-introduce its common position after a failure in

conciliation

 Possibility to conclude agreement at first reading

 Co-decision extended to: transport, environment, development

cooperation, and employment and social affairs

 Union action covered by co-decision: 38 areas

Treaty of Nice

(February 2003)



Co-decision extended to: a few additional legal bases, primarily justice

and home affairs

 Union action covered by co-decision: 43 areas

Treaty of Lisbon

(December 2009)



Co-decision officially becomes the ‘ordinary legislative procedure’

 Significant extension of the scope of the ordinary legislative

procedure

 Co-decision extended to: freedom, security and justice, international

trade, agriculture and fisheries

 Union action covered by co-decision: 85 areas

Source: European Parliament 1999; 2004; 2009; 2014.

Following the first year (1993-1994) of the co-decision procedure’s application, the co-legislators quickly

learned that the legislative agreement required frequent interaction. The plenary meetings of the conciliation

committee at third reading were not regarded as being conducive to reaching compromises. To overcome the

difficulties of negotiating an agreement between thirty or more people in the full-scale conciliation meetings,

informal tripartite meetings between the Commission, the Council, and the EP emerged ahead of the formal

conciliation meetings. The first trilogue meetings were initiated under the German Presidency of the Council

of the EU in the second half of 1994 and the informal meetings became standard practice under the Spanish

Presidency in the second half of 1995. They were largely modelled after the budgetary conciliation

procedure.

The informality of trilogue meetings injected a degree of flexibility into the negotiations and

allowed the negotiators to speak more freely, to iron out their differences on divisive issues, and to find a

European Parliament (1999) ‘Activity Report 1 November 1993 to 30 April 1999 (4th parliamentary term)’, PE 230.998, Brussels: European

Parliament, Delegations to the Conciliation Committee, p. 6.

Investigation of informal trilogue negotiations since the Lisbon Treaty

compromise that could later be approved by the full conciliation committee.

Hence, the initial clash of

cultures between the co-legislators was progressively overcome by going informal.

Until the Amsterdam Treaty entered into force in 1999, trilogue meetings were confined to the conciliation

phase and were not yet a defining feature of earlier readings.

This was partly a result of most deals being

agreed late in the legislative process, as the Maastricht Treaty did not allow for finalisation of legislation at

first reading. It was not until the Amsterdam Treaty that first reading agreements became possible under

primary law.

The Amsterdam Treaty made it possible for the Council to accept the EP’s first reading

amendments without adopting a common position. The added value of this was a more efficient decision-

making process, enabling the co-legislators to accelerate the adoption of legislation. The change was also

made in response to the institutions’ increased workload, the progressive extension of the scope of the co-

decision procedure, and the ability of the co-legislators to work together in a constructive manner.

The institutions soon realised that they needed to change their working methods when the Amsterdam Treaty

came into force, which led to the 1999 ‘Joint declaration on practical arrangements for the new co-decision

procedure’.

In the declaration, the three institutions established that informal contact between them should

cover all stages of the co-decision procedure and that they ‘shall cooperate in good faith with a view to

reconciling their positions as far as possible so that wherever possible acts can be adopted at first reading’.

Indeed, the declaration explicitly stated that the conclusion of legislative procedures should be sought as

early as possible. The ability to conclude legislation at first reading extended the use of trilogues beyond that

of the conciliation phase and trilogues soon became the standard negotiating format at first and second

readings.

The Amsterdam Treaty significantly strengthened the parity between Parliament and Council and extended

co-decision from 15 to 38 areas of Community action.

The Treaty of Nice further extended the scope of the

co-decision procedure, but did not modify the co-decision procedure per se. The Treaty of Lisbon almost

doubled the scope of what is now called the ‘ordinary legislative procedure’ from 43 to 85 legal bases of the

EU treaties. It extended and introduced co-decision powers in the EP’s Committees on Agriculture and Rural

Development (AGRI), Regional Development (REGI), International Trade (INTA), Civil Liberties, Justice

and Home Affairs (LIBE) and Fisheries (PECH). Despite introducing co-decision powers to committees

previously unacquainted with the procedure, these committees easily adapted to working closely with the

Council and reaching compromises early and informally in the decision-making process.

Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation and change’, Journal of European

Public Policy 10 (2): 171-88.

European Parliament (2004) ‘Activity Report 1 May 1999 to 30 April 2004 (5th parliamentary term)’, PE 287.644, Brussels: European Parliament,

Delegations to the Conciliation Committee, p. 7.

Roederer-Rynning, C. and Greenwood, J. (2016) ‘The culture of trilogues’, Journal of European Public Policy 22 (8): 1149.

Article 251 TEC; now Article 294 TFEU.

Brandsma, G.J. (2015) ‘Co-decision after Lisbon: The politics of informal trilogues in European Union lawmaking,’ European Union Politics 16

(2): 300-19.

European Parliament, Council and Commission (1999) ‘Joint declaration on practical arrangements for the new codecision procedure (Article 251

of the Treaty establishing the European Community)’, OJ C 148/01, 28 May, Brussels: European Parliament, Council and Commission.

European Parliament, Council and Commission (1999) ‘Joint declaration on practical arrangements for the new codecision procedure (Article 251

of the Treaty establishing the European Community)’, OJ C 148/01, 28 May, Brussels: European Parliament, Council and Commission, p. 1.

Shackleton, M. (2000) ‘The politics of codecision’, Journal of Common Market Studies 38 (2): 325-42.

European Parliament (2004) ‘Activity Report 1 May 1999 to 30 April 2004 (5th parliamentary term)’, PE 287.644, Brussels: European Parliament,

Delegations to the Conciliation Committee p. 19.

Investigation of informal trilogue negotiations since the Lisbon Treaty

The trend witnessed under previous parliamentary terms towards adopting legislation at first reading

continued in the seventh parliamentary term (2009-2014) and the first half of the eighth parliamentary term

(2014-2016).

Furthermore, as shown in Figure 2, the percentage of ordinary legislative files concluded by

early agreement – based on an inter-institutional compromise reached in trilogue meetings – has remained

high in recent years.

The dotted vertical lines in the figure indicate the EP elections. The figure presents the

early agreements concluded per half year (for 1999-II, 2000-I, 2000-II, and so forth) as a percentage of the

total number of ordinary legislative files concluded in the same period.

Looking at the development over

time, we can see that the use of early agreements as a percentage of all files increased rapidly in the period

between 1999 and 2008, with a stabilisation having taken place since 2008. The popularity of early

agreement reflects the growing levels of cooperation and mutual trust between the co-legislators, at least

when it comes to the ordinary legislative procedure.

Figure 2: Percentage of early agreements per half year (1999-2016)

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit; European Parliament (2017) ‘Activity Report on the Ordinary Legislative

Procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’, PE 595.931, Brussels: European Parliament, Conciliation and Codecision

Unit.

The figure is based on data on 1392 co-decision files concluded between July 1999 and December 2016, excluding the 88 codification files that

have been concluded in this period under the responsibility of the JURI committee. Codification files have been excluded because they merely

combine a legislative act and its amendments into one single new act. As there is no change in contents, no use is made of informal trilogue meetings.

Thus, codification files are never concluded by early agreement. For a detailed description of the identification of early agreements, see Bressanelli,

E., Héritier, A., Koop, C. and Reh, C. (2014) ‘The informal politics of codecision: Introducing a new data set on early agreements in the European

Union’, EUI Working Paper 2014/64 RSCAS (EUDO). In our analyses, we primarily use data collected as part of the following two studies: Reh,

C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the European

Union,’ Comparative Political Studies 46 (9): 1112–42; Bressanelli, E., Koop, C., and Reh, C. (2016) ‘The impact of informalisation: Early

agreements and voting cohesion in the European Parliament,’ European Union Politics 17 (1): 91-113. Data on the second half of the seventh

parliamentary term have been collected for: Koop, C., Reh, C. and Bressanelli, E. (2017) ’When politics prevails: Parties, elections and loyalty in the

European Parliament,’ Paper presented at the Fifteenth Biennial Conference of the European Union Studies Association, Miami, 4-6 May. Data on the

first half of the eighth parliamentary term have been collected for this study. We are grateful to Lauge Petersen for his valuable research assistance

with the data collection.

Please note that while the x-axis of Figure 2 only shows labels for the first half of each year (2000-I, 2001-I, and so forth), the line in the figure

connects the data points for each half year. We only include labels for every other half year to enhance the readability of the figure.

Investigation of informal trilogue negotiations since the Lisbon Treaty

In the period between 2008 and 2016, early agreements constituted, on average, 80 percent of all legislative

files subject to the ordinary legislative procedure (excluding codification files). In the seventh parliamentary

term (2009-2014), early agreements represented 84 percent of the total number of ordinary legislative files,

while they constituted 75 percent of ordinary legislative files finalised in the first half of the eighth term

(2014-2016).

These percentages are considerably lower than those presented in the EP activity reports. For

instance, in the most recent report from 2017, it is pointed out that early agreements represented 97 percent

of the files concluded in the first half of the eighth parliamentary term.

The reason for the difference in the

percentages of early agreements between the EP activity reports and our study is that our definitions of early

agreement differ. Following the academic literature on the topic, we consider ordinary legislative files early

agreements if they are (a) concluded at first or early second reading, and (b) based on an informal

compromise between the co-legislators, negotiated during trilogue meetings (see Chapter 1). In the EP’s

activity reports, on the other hand, early agreements are defined as files concluded at first or early second

reading, regardless of whether they are based on inter-international trilogue negotiations and compromises.

Figure 3: Percentage of early agreements per stage of conclusion and parliamentary term

Excluding codification files, the number of co-decision files concluded in the seventh term is 481. In the first half of the eighth term, 119 files were

concluded. During the fifth and sixth term, 379 and 413 files were concluded, respectively. In Chapter 3, we will have a closer look at the volume of

legislation.

European Parliament (2017) ‘Activity Report on the Ordinary Legislative Procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’,

PE 595.931, Brussels: European Parliament, Conciliation and Codecision Unit, p. 9.

Investigation of informal trilogue negotiations since the Lisbon Treaty

This difference in definitions is particularly important for our understanding of what happens at the first

reading of the legislative process. As Figure 3 demonstrates, concluding ordinary legislative files at an early

stage of the legislative process is not the same as concluding files by early agreement. The figure presents

early agreements (as percentage) by parliamentary term and stage of conclusion (excluding codification).

Although early agreements represent the majority of the files concluded at first reading in the sixth, seventh

and first half of the eighth term, there are always a few files that are not based on an inter-interinstitutional

agreement reached in trilogue.

To give more insight into the nature of the small number of non-early agreements, it is necessary to take a

closer look at the files concluded in first half of the eighth parliamentary term (2014-2016). In this period, 25

percent of all concluded files are non-early agreements (excluding the JURI committee’s codification files

from the analysis). While 67 percent of the files concluded at first reading are early agreements, 33 percent

are not. These non-early agreements include various types of files that may be described as relatively

uncomplicated. Files concluded at first reading without trilogue meetings tend to be:

a) Acts in which the Commission’s proposal is adopted without amendments, often with a high level of

temporal urgency;

b) Acts which introduce only minor amendments to the Commission’s proposal, (again) often with a

high level of temporal urgency;

c) Acts which repeal legislation; for instance, in the context of the European Commission’s 2012

Regulatory Fitness and Performance Programme (REFIT);

d) Act which recast legislation (recasts).

Three (eight percent) of all second reading files concluded in the first half of the eighth parliamentary term

are non-early agreements. It is important to note that this does not mean that these files have not been

negotiated in inter-institutional trilogue meetings. In fact, the non-early agreements that are concluded at

second reading all build on a compromise negotiated in trilogue meetings. However, as the acts have not

The exact percentages in the figures are the following. In the fifth parliamentary term, 7.7 percent of all files were concluded as early agreement

(5.5 percent at first reading; 2.1 percent at early second reading). In the sixth term, 63.2 percent of all files were concluded as early agreement (56.7

percent at first reading; 6.5 at early second reading). In the seventh term, 83.8 percent were concluded as early agreement (76.9 percent at first

reading; 6.9 percent at early second reading). In the first half of the eighth term, 74 percent were concluded as early agreement (46.3 percent at first

reading; 27.6 percent at early second reading).

For instance: Regulation (EU) No 1383/2014 of the European Parliament and of the Council of 18 December 2014 amending Council Regulation

(EC) No 55/2008 introducing autonomous trade preferences for the Republic of Moldova, OJ L 372, 30 December 2014; Regulation (EU) 2015/1146

of the European Parliament and of the Council of 8 July 2015 fixing the adjustment rate provided for in Regulation (EU) No 1306/2013 for direct

payments in respect of the calendar year 2015, OJ L 191, 17 July 2015; Regulation (EU) 2016/1014 of the European Parliament and of the Council of

8 June 2016 amending Regulation (EU) No 575/2013 as regards exemptions for commodity dealers (Text with EEA relevance), OJ L 171, 29 June

2016.

For example: Directive (EU) 2015/413 of the European Parliament and of the Council of 11 March 2015 facilitating cross-border exchange of

information on road-safety-related traffic offences (Text with EEA relevance), OJ L 68, 13 March 2015; Regulation (EU) 2016/401 of the European

Parliament and of the Council of 9 March 2016 implementing the anti-circumvention mechanism provided for in the Association Agreement between

the European Union and the European Atomic Energy Community and their Member States, of the one part, and Georgia, of the other part, OJ L 77,

23 March 2016; Regulation (EU) 2016/580 of the European Parliament and of the Council of 13 April 2016 on the introduction of emergency

autonomous trade measures for the Republic of Tunisia, OJ L 102, 18 April 2016.

Examples include: Directive (EU) 2015/254 of the European Parliament and of the Council of 11 February 2015 repealing Council Directive

93/5/EEC on assistance to the Commission and cooperation by the Member States in the scientific examination of questions relating to food (Text

with EEA relevance), OJ L 43, 18 February 2015; Regulation (EU) 2016/93 of the European Parliament and of the Council of 20 January 2016

repealing certain acts from the Schengen acquis, OJ L 26, 2 February 2016. On the REFIT programme, see: European Commission (2012)

Communication from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of

the Regions on EU Regulatory Fitness, COM (2012) 746 final, 12 December 2012.

For example: Regulation (EU) 2015/755 of the European Parliament and of the Council of 29 April 2015 on common rules for imports from certain

third countries, OJ L 123, 19 May 2015; Regulation (EU) 2016/1076 of the European Parliament and of the Council of 8 June 2016 applying the

arrangements for products originating in certain states which are part of the African, Caribbean and Pacific (ACP) Group of States provided for in

agreements establishing, or leading to the establishment of, economic partnership agreements, OJ L 185, 8 July 2016. The former was originally a

codification file, but it was turned into a recast for the reason that some minor substantive amendments were made to the file.

Investigation of informal trilogue negotiations since the Lisbon Treaty

been concluded at (first or) early second reading, they do not fall under the definition of early agreement

used in this study. In the few cases of ‘late’ agreement at second reading, trilogue negotiations either started

at second reading – after the adoption of the Council’s common position – or after the first reading in

Parliament, with the compromise amendments not yet being agreed upon by the time the Council decided to

adopt its common position.

It should be noted that the percentage of second reading conclusions in the first half of the eighth

parliamentary term is somewhat higher than in the seventh parliamentary term (see Figure 3). Various

interviewees explained that this is because the previous (seventh) Parliament wanted to finish its own first

reading of pending ordinary legislative files before the EP election of 2014, even if this meant having a first

reading before an inter-institutional agreement was reached with the Council. As a consequence, the current

(eighth) Parliament took over a considerable number of ordinary legislative files which had already passed

the first reading in Parliament, but had not yet been agreed upon with the Council, thus leading to an early or

late second reading agreement. The percentage of files concluded at second reading is likely to go down once

the files concluded in the second half of the eighth parliamentary term is taken into consideration.

Interestingly, Figure 3 shows that no files have been concluded at third reading (the conciliation procedure)

in the first half of the eighth parliamentary term. As the activity report of the EP points out, this is hardly a

surprise. According to the report, the increase in early conclusion and the absence of conciliation procedures

‘are signs of a well-functioning legislative procedure, of a productive and ever-closer working relationship

between the co-legislators, and, within each of the institutions, of procedural methods that establish the

framework and the boundaries for general efficient (if not always straightforward) negotiations’.

We will

return to the (normative) evaluation of the trends in Chapter 5.

2.2 What differentiates informal and formal politics?

When solely comparing the nature of negotiations in trilogue meetings versus negotiations made outside

trilogue meetings regardless of whether legislation is finalised early or not, it is possible to identify four key

differences between informal and formal decision-making arenas. In their study of early agreements, Reh

and her colleagues distinguished four dimensions on which the two types of decision-making differ:

1) The nature and status of rules: formal rules which are codified and enforced through official

channels versus informal rules which are non-codified and non-enforceable by third parties;

2) The boundaries of participation: inclusive participation of formally restricted boundaries versus

restricted participation and unofficially drawn boundaries;

3) Transparency: public access or justified seclusion versus seclusion or lack of transparency;

For the former case, see: Directive (EU) 2015/412 of the European Parliament and of the Council of 11 March 2015 amending Directive

2001/18/EC as regards the possibility for the Member States to restrict or prohibit the cultivation of genetically modified organisms (GMOs) in their

territory (Text with EEA relevance), OJ L 68, 13 March 2015. For the latter case, see: Directive (EU) 2015/719 of the European Parliament and of the

Council of 29 April 2015 amending Council Directive 96/53/EC laying down for certain road vehicles circulating within the Community the

maximum authorised dimensions in national and international traffic and the maximum authorised weights in international traffic (Text with EEA

relevance), OJ L 115, 6 May 2015.

European Parliament (2017) ‘Activity Report on the Ordinary Legislative Procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’,

PE 595.931, Brussels: European Parliament, Conciliation and Codecision Unit, Foreword.

Investigation of informal trilogue negotiations since the Lisbon Treaty

4) The type of outcome: binding decisions versus informal agreements, which require rubberstamping

in formal arenas.

Overall, decisions taken in the formal arena are structured by a set of codified and enforceable rules and are

characterised by inclusive participation, transparency and binding decisions. In contrast, informal politics

rarely follows a set of codified rules, restricts participation to a sub-set of actors, lacks transparency, and any

decision taken in the informal arena requires formalisation in the formal arena.

First, the nature and status of rules differ between formal and informal decision arena. Whereas formal

politics is structured by codified enforceable rules laid down in the EU treaties, informal politics is usually

non-codified and often not enforceable by third parties.

The Treaty on the Functioning of the EU (TFEU)

allows for first reading agreements to be struck, but trilogues have no reference in the treaties, yet they have

become a defining feature of the functioning of the ordinary legislative procedure. Instead, informal politics

is generally ‘created, communicated and enforced outside the officially sanctioned channels’.

This does not

mean that trilogues operate in a rule-free environment. On the contrary, they have become increasingly

institutionalised over time in the EP’s own Rules of Procedure and in interinstitutional agreements between

the institutions. As we will describe in more detail in Chapter 4, the EP’s Rules of Procedure include specific

rules regarding the composition of the EP’s trilogue negotiating team, mandates given to the team, and

reporting requirements. These rules bind the EP’s own conduct and are enforceable by Parliament’s bureau.

The General Court can, under certain conditions, challenge the EP’s Rules of Procedure. It is possible for

individual MEPs to take a case against the EP to the General Court when individual members have a legal

standing (locus standi). Individual MEPs can bring a case before the General Court in a personal capacity

when they can demonstrate to the court a sufficient connection to, and harm from, an action taken by the EP.

An individual parliamentary committee would, however, not be able to the sue the EP if they find that

Parliament’s Rules of Procedure have been breached; for instance, if an EP’s negotiating team were not to

obtain a committee mandate before commencing trilogue meetings. This is because Parliament’s committees

do not constitute legal entities in their own rights and their legal existence is derived from Parliament. This

means that third parties cannot uphold Parliament’s internal rules governing trilogue meetings unless an

individual MEP can prove that he/she has been personally harmed by the breach. EU institutions’ breaches

of inter-institutional agreements, on the other hand, can be tested at the Court of Justice; for instance, if the

EP claims that the Commission or the Council have breached rules in the agreement.

Second, the boundaries of participation differ between formal and informal arenas regarding the actors

participating in trilogues and the transparency surrounding it. Only a limited number of representatives from

the Commission, the EP, and the Council are involved in trilogue meetings and the names of the people

participating are not normally made public, although the type of actors attending is determined by intra-

institutional rules and practices. Indeed, as our interviewees explained, the political trilogue meetings have a

number of recurrent participants. The rapporteur always participates in the political trilogues on behalf of the

EP and all of the shadow rapporteurs are invited to attend. Committee chair attendance varies from one

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1115-6. For the more general analytical framework of decision-making structures, see

Ostrom, E. (1986) ‘An agenda for the study of institutions,’ Public Choice 48 (1): 3-25.

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1116

Helmke, G. and Levitsky, S. (2004) ’Informal institutions and comparative politics: A research agenda’. Perspectives on Politics, 2 (4): 727

Investigation of informal trilogue negotiations since the Lisbon Treaty

committee to the other: whereas the chairs of some committees almost always attend (and chair) the trilogue

meetings, the chairs of other committees hardly every attend, with the rapporteur then chairing the meetings.

The political actors may bring their assistants and political group’s policy advisers to the meetings. Support

is also provided by the relevant committee’s secretariat, by the legal service of the EP’s Secretary-General,

and by lawyer-linguists. The Council presidency represents the Council in political trilogues, with assistance

from the Council’s General Secretariat, including the legal service. The Commission is typically represented

by the relevant head of unit in the directorate general in charge of the dossier, and by the desk officer

responsible for the proposal (see Chapter 4 for further information about the participants). Formal decision-

making arenas, on the other hand, normally allow for the inclusion of all legitimate decision-makers or

include a formally restricted and a publicly known subset of decision-makers. For instance, all MEPs (are

invited to) participate in decisions in the EP’s plenary sessions, and all ministers (are invited to) participate

in the decisions of the relevant Council of the European Union.

Third, the transparency of negotiations differs between formal and informal arenas. Formal decision-making

arenas tend to be more transparent than informal ones in terms of openness of negotiations, access to

documents produced, and justification of restrictions on public scrutiny.

In the formal arena, the decision-

making process is relatively easy to follow by those not directly involved in legislation. Documents outlining

the positions of the institutions are public, such as the EP’s and the Council’s respective positions at different

readings. If the procedure goes as far as conciliation, it is on the basis of the institutions’ publicly known

second reading positions. In the absence of inter-institutional negotiations in trilogues, the positions of the

different institutions at the different readings are known, and we also know on which provisions an

agreement and disagreement between the institutions exists. This gives the public more insight into the

different steps leading to the eventual compromise. When trilogues are used, we have limited information on

the initial position of the Council, unless its position as adopted as a General Approach (approved by the

ministers and publicly available). If the co-legislators wish to finalise legislation at first reading, the Council

does not adopt a common position, but instead has to convince Parliament to adopt amendments that the

Council can also support. Unlike formal decision fora, the institutions do not respond to each other’s

positions in turn. Early agreements sometimes lack a publicly available paper trail of the timing of

negotiations, and the position of all negotiators, both before and during the negotiations. This has led some

scholars to conclude that the actual practice of co-decision renders ‘passive the democratic potentials

characterising the formal procedural rules’.

Information about what goes on in trilogue meetings is restricted to feedback given by the trilogue members

to their respective institutions, and documentation of the decision process in trilogues (such as minutes and

positions) are not automatically made publicly available, although access may be granted upon request. The

limited transparency in the case of early agreements is not only a procedural matter, but it challenges the

distinctive role of Parliament as a debating chamber. This means that the EP has experienced the largest

adjustments to the increased informality of the ordinary legislative procedure as early agreements come close

to the confidential form of consensus formation used in the Council. Legitimacy concerns of early

agreements have mainly been directed towards the EP as the only EU institution with a direct link to the

elected through direct elections. It is also the institution that has done the most to increase the accountability

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1117.

Stie, A.E. (2013) Democratic Decision-Making in the EU: Technocracy in Disguise, London: Routledge, p. 270.

Investigation of informal trilogue negotiations since the Lisbon Treaty

and transparency of the EP’s negotiating teams in trilogues vis-à-vis the rest of the EP and the outside world,

a topic we will return to in Chapter 4.

Fourth, informal and formal arenas differ when it comes to the type of outcome produced. Only decisions

that have been approved by the EP’s full plenary and the ministers in the Council are binding regardless of

whether the compromise is struck with or without the use of trilogue meetings. Everything that is agreed

informally, during trilogue meetings, must still be rubberstamped by formal arenas to enter into force.

However, deals struck in trilogue meetings can constrain the room of manoeuvre in the formal arena by

putting pressure on decision-makers not to challenge or alter the informal agreement. For instance, MEPs are

in principle able to table amendments to the compromise reached in trilogue during committee and plenary

scrutiny. In practice, however, they are under immense pressure not to jeopardise the informal agreement.

To sum up, informal decision-making arenas are characterised by non-codified and non-enforceable rules

and by a more secluded, restricted, opaque decision-making process, generating outcomes that still require

rubberstamping in formal arenas. Inter-institutional trilogue meetings share many of the features of informal

decision-making arenas, although considerable changes have been made, leading them to become more

formal.

2.3 Why have early agreements emerged?

As the previous section has demonstrated, early agreements have become ‘the new normal’. Scholars and

practitioners alike have often posed the question of what accounts for the rise in early agreements since the

Amsterdam Treaty. There is no single explanation for the rise in early agreements, but three broad

explanations are often put forward in the academic and non-academic literature:

1. The functionalist explanation: Early agreements represent a coping strategy to harvest efficiency

gains and to reduce the transaction costs associated with the ordinary legislative procedure covering

more policy areas and actors.

2. The cultural rapprochement explanation: The Council and the EP have become socialised into

shared norms of cooperation, characterised by legislative pragmatism and anticipatory compliance,

facilitating the willingness and ability to find an agreement early.

3. The image-building explanation: The Council’s rotating presidencies wish to be seen as successful

because they manage to finalise what are, in their eyes, important files during their presidency. Most

outgoing Council presidencies produce a ‘scoreboard’ to show what they have achieved during their

Rasmussen, A. and Shackleton, M (2005) ‘The scope for action of European Parliament negotiators in the legislative process: Lessons of the past

and for the future’, Paper presented at the Conference of the European Union Studies Association in Austin, Texas; Reh, C., Héritier, A., Bressanelli,

E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the European Union,’ Comparative Political

Studies 46 (9): 1112-42.

Brandsma, G.J. (2015) ‘Co-decision after Lisbon: The politics of informal trilogues in European Union lawmaking,’ European Union Politics 16

(2): 300-19; Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in

the European Union,’ Comparative Political Studies 46 (9): 1112-42.

Burns, C. and Carter, N. (2010) ‘Is co-decision good for the environment? An analysis of the European Parliament’s green credentials,’ Political

Studies 58 (1): 123-42; Ripoll Servent, A. (2015) Institutional and Policy Change in the European Parliament: Deciding on Freedom, Security and

Justice, London: Palgrave Macmillan; Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation

and change’, Journal of European Public Policy 10 (2): 171-88.

Investigation of informal trilogue negotiations since the Lisbon Treaty

presidency. This assessment is obviously subjective in nature as no objective and externally defined

criteria exist for assessing a successful presidency. Often, however, presidencies point to the number

of legislative files that they have reached an (early) agreement on during their presidency to create

an image of an efficient and effective presidency.

Likewise, key MEPs, and particularly

rapporteurs, want to show the Council and their EP colleagues that they have managed to get to an

early agreement with the Council.

From a functionalist perspective, the growing popularity of early agreements is explained by the absence of

time limits and the lower majority requirements for passing legislation during first reading. At the first

reading, the EP only needs to muster a simple majority of votes cast to amend or adopt the Commission’s

proposal, whereas an absolute majority is needed at second reading to reject or amend the Council’s position,

and a simple majority is required to approve it without amendments. Similarly, the Council has a strong

interest in persuading the EP to propose amendments that are desired by a qualified majority in the Council,

as it is hard for the Council to amend the Commission’s proposals. At first reading, the Council decides by

qualified majority voting unless its position differs from that of the Commission, in which case unanimity is

required. At second reading, the Council votes with qualified majority requirements on Parliament

amendments for which the Commission has delivered a positive opinion, while unanimity is needed on

amendments where the Commission has delivered a negative opinion. The task of finding a Council position

has become more challenging since the 2004 enlargement of the EU, which resulted in a large increase in

member states. In this context, early input from the EP can be regarded as helping to facilitate the Council’s

internal consensus building.

Both the Council and the EP have an interest in anticipating the each other’s

position early in the decision-making process to avoid uncertainty, high voting thresholds, and time limits

later in the process.

Trilogues are examples of institutions that diminish the transaction costs involved in

decision-making by reducing the likelihood of cooperation problems and conflict.

Trilogues as an institution also have a constitutive effect on the actors involved. They create an environment

of enhanced interaction between the EU institutions, which are instrumental in building mutual trust and

shared norms. In the words of the EP, ‘[t]ime and practice have led to a cultural rapprochement of Parliament

and the Council: each institution’s distinctive internal rules, procedures and methods are better understood

by the other, as are their respective administrative and political needs and constraints’.

The wide

application of the ordinary legislative procedure has not only changed the power balance between the EU

institutions, but has also brought with it new working methods and norms. Legislative procedures are more

than just rules of procedure: they represent norms that guide behaviour. According to Ripoll Servent, the

increased use of early agreements has led the EP to ‘acquire a feeling of shared responsibility – revealed in a

softer use of language and more moderate stances in its reports’.

In trilogues, the Council and the EP have a

tendency to engage in a process of anticipatory compliance, whereby they moderate their demands to

Mühlböck, M. and Rittberger, B. (2015) The Council, the European Parliament, and the paradox of inter-institutional cooperation,’ European

Integration Online Papers 19 (4): 1-20; Rasmussen, A. (2011) ‘Early conclusion in bicameral bargaining: Evidence from the co-decision legislative

procedure of the European Union,’ European Union Politics 12 (1): 41-64.

European Parliament (2009) ‘Activity Report 1 May 2004 to 13 July 2009, (6th parliamentary term)’, PE427.162v01-00, Brussels: European

Parliament, Delegations to the Conciliation Committee, p. 14.

Brandsma, G.J. (2015) ‘Co-decision after Lisbon: The politics of informal trilogues in European Union lawmaking,’ European Union Politics 16

(2): 309.

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit, p. 11.

Ripoll Servent, A. (2012) ’Playing the co-decision game? Rules’ changes and institutional adaptation at the LIBE committee,’ Journal of European

Integration 34 (1): 60.

Investigation of informal trilogue negotiations since the Lisbon Treaty

increase the likelihood of reaching an early agreement.

In addition, Council presidencies often use the number of compromises reached by early agreement as

markers of a successful presidency. It is in the interest of Council presidencies to strike deals early as their

reputation often depends on their capacity to consolidate different opinions and get legislation through.

However, previous research has also shown that early agreement is statistically not more likely when the

legislative file is an ex ante defined agenda priority of the Council presidency.

The image building

argument may also explain why the EP has been willing to go informal despite the trilogue setting fitting

uneasily with the parliamentary style of open public debate. Ever since the EP acquired co-decision powers,

it has been determined to make the procedure work and to counter the Council’s early scepticism concerning

the EP’s abilities to legislate. In the words of the former Secretary-General of the EP (1997-2007), Julian

Priestley, ‘[the EP] had to demonstrate that it was capable of getting through the business’.

From this

perspective, the Parliament considered that its legislative powers would be more likely to be extended in the

next round of treaty negotiations if it proved to act as a responsible legislator.

Together, the three explanations elaborated on above help explain the rationale behind ‘going informal’ and

increasing this practice over time. The EP itself also explains the rise of early agreements with reference to a

higher number of uncontroversial and rather technical proposals.

However, this is not a trend born out of

empirical research, which finds that uncontroversial and technical drafts acts are not more likely to be

subject to early agreement than controversial and (re-)distributive files.

Adding to the list of possible

explanations, the EP has suggested that the Commission ‘often pushes for an early adoption because it will

be able to demonstrate efficiency and hopes that its proposal will be adopted with as few changes as

possible’.

However, scholars have not tested this claim.

What is beyond doubt is that informal tripartite meetings have become the drivers of much of the inter-

institutional work, both in terms of trilogue meetings on specific legislative proposals and the more general

working relationship between Parliament and the Council, enabling the institutions to coordinate their work.

Examples of general informal meetings are the ‘speed dating’ meetings between the EP’s Conference of

Committee Chairs and the Council’s Presidency at the beginning of each new six-month presidency. These

meetings serve to discuss legislative priorities and to uphold a fruitful working relationship between the

institutions.

The high percentage of legislation finalised at first reading would not be possible in the absence of trilogue

meetings. In the seventh parliamentary term (2009-2014) alone, over 1,500 trilogue meetings were held on

See also: Burns, C. and Carter, N. (2010) ‘Is co-decision good for the environment? An analysis of the European Parliament’s green credentials,’

Political Studies 58 (1): 123-42; Ripoll Servent, A. (2015) Institutional and Policy Change in the European Parliament: Deciding on Freedom,

Security and Justice, London: Palgrave.

Mühlböck, M. and Rittberger, B. (2015) The Council, the European Parliament, and the paradox of inter-institutional cooperation,’ European

Integration Online Papers 19 (4): 8.

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1133.

Priestley, J. (2008) Six battles that shaped Europe’s Parliament, London: John Harper Publishing, pp. 125-6.

European Parliament (2009) ‘Activity Report 1 May 2004 to 13 July 2009, (6th parliamentary term)’, PE427.162v01-00, Brussels: European

Parliament, Delegations to the Conciliation Committee pp. 11-2.

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1112-42.

European Parliament (2012) ‘Activity report on codecision and conciliation: 4 July 2009 to 31 December 2011 (7th parliamentary term)’,

DV\903361, Brussels: Delegations to the Conciliation Committee, p. 5.

Investigation of informal trilogue negotiations since the Lisbon Treaty

approximately 350 pieces of legislation.

Most files negotiated in trilogues need only one to four meetings,

while a small number of cases require more frequent interaction. For instance, Regulation (EU) No

1303/2013 on laying down common provisions for EU funds relating to regional development (the Common

Provision Regulation or CPR) had 54 trilogues.

Research on the seventh parliamentary term (2009-2014)

shows that the average number of trilogue meetings required per file varies across committees.

The

committees on Budgets (BUDG), Regional Development (REGI), Employment and Social Affairs (EMPL),

and Economic and Monetary Affairs (ECON) needed more trilogue meetings per dossier than other

committees. On average, they needed 8, 7.9, 4.8, and 4.7 trilogue meetings respectively per dossier.

Interestingly, some of the committees that were new to the ordinary legislative procedure after the Lisbon

Treaty needed fewer trilogues per dossier compared to the EP average. The committees on Agriculture and

Rural Development (AGRI), International Trade (INTA), and Fisheries (PECH) needed, on average, 1.2, 1.1,

and 0.9 meetings respectively per file. Possible explanations might be that these committees wanted to prove

that they could handle their newly acquired co-decision powers well.

Informal politics has become commonplace in the majority of draft acts and it also comes with many

advantages. Most notably, trilogues tend to make decision-making processes more efficient as fewer

readings are usually required, with the first reading being particularly ‘generous’ when it comes to voting

requirements, and improve the working relationship between the institutions. However, there is also an

awareness that the added value of increased efficiency may have come at the cost of transparency of

negotiations and accountability of the negotiation teams to their respective institutions and the outside world.

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit, p. 19.

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit, p. 19. For the Act itself, see: Regulation (EU) No 1303/2013 of the European

Parliament and of the Council of 17 December 2013 laying down common provisions on the European Regional Development Fund, the European

Social Fund, the Cohesion Fund, the European Agricultural Fund for Rural Development and the European Maritime and Fisheries Fund and laying

down general provisions on the European Regional Development Fund, the European Social Fund, the Cohesion Fund and the European Maritime

and Fisheries Fund and repealing Council Regulation (EC) No 1083/2006, OJ L 347, 20 December 2013.

Brandsma, G.J. (2015) ‘Co-decision after Lisbon: The politics of informal trilogues in European Union lawmaking,’ European Union Politics 16

(2): 311.

Cf. Earnshaw, D. and Judge, D. (1995) ‘Early days: The European Parliament, co-decision and the European Union legislative process post-

Maastricht,’ Journal of European Public Policy 2 (4): 624-49.

Investigation of informal trilogue negotiations since the Lisbon Treaty

3. Patterns of early agreement

In Chapter 2, we showed that early agreements have become a common feature of ordinary legislative

procedures, and we discussed the different explanations behind the popularity of early agreements, as put

forward in the academic and non-academic literature. By focusing on the general trend, however, we did not

address the (remaining) variation in the use of early agreements. Therefore, in this chapter, we look at the

variation using the same dataset of ordinary legislative procedures (1999-2016) as in Chapter 2. We zoom in

on the volume of ordinary legislative legislation, the duration of ordinary legislative procedures, the variation

across EP committees, and other characteristics of legislation. In the last part of the chapter, we also review

some recent academic studies on the political implications of early agreement.

3.1 The volume of legislation: The use of early agreements has increased over time

As we pointed out in Section 2.3, one of the key explanations for the use of early agreements is that it

expedites the legislative process. Starting early in the process with inter-institutional negotiations in trilogue

meetings would be more efficient than following a sequential process of position-taking and position-

exchange that allows for co-legislators to get closer to one another. Making an early agreement has the

additional advantage of lower majority requirements and absence of specified time limits at first reading.

These advantages became particularly important after the 2004 enlargement, which, in correspondence with

the increase in member states, also significantly raised the number of actors involved in the legislative

process. The advantages also became more relevant with the increase in the number of policy areas subject to

the ordinary legislative procedure, which led to a higher volume of legislation. Let us have a closer look at

the volume of legislation.

Figure 4 presents the number of ordinary legislative files concluded in each legislative year since the

legislative year 1999-2000, excluding codification files.

The figure shows the increase over time in the

number of concluded ordinary legislative files by the co-legislators. The number of ordinary legislative files

has increased from 379 in the fifth parliamentary term (1999-2004) to 413 in the sixth term (2004-2009) and

481 in the seventh term (2009-2014). There are also important differences within parliamentary terms, with

the number of concluded legislative files going up later in the term, particularly in the legislative year before

EP elections. This is unsurprising: in the year before EP elections, the EU institutions – and particularly

Parliament – seek to finish as much business as possible.

In the first half of the eighth parliamentary term (2014-2016), 119 ordinary legislative files were concluded.

This is not very different from the number of files concluded in the first half of previous parliamentary

terms, even though the new Commission has tabled fewer legislative proposals than previous commissions

as part of its endeavour to ‘be doing less, but [to be doing its work] more effectively’.

Indeed, the number

of proposals tabled by the new Commission in the first half of the eighth parliamentary term is 192, which is

We followed the EP’s activity reports in determining when legislative years started and ended. See: European Parliament (2017) ‘Activity Report

on the Ordinary Legislative Procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’, PE 595.931, Brussels: European Parliament,

Conciliation and Codecision Unit, Footnote 13. To enhance comparability, we only included full legislative years in the figure, and thus excluded the

second half of 2016. As a consequence, the figure builds on data on 1362 files (rather than on the 1392 files in the dataset).

Juncker, J.-C. (2014) ‘The Juncker Commission: The right team to deliver change’, press conference by the President-elect of the European

Commission, 10 September, available at: http://europa.eu/rapid/press-release_SPEECH-14-585_en.htm (last accessed: 30 April 2017).

Investigation of informal trilogue negotiations since the Lisbon Treaty

rather low compared to the 321 and 244 proposals tabled by the Commission in the first half of, respectively,

the seventh and the sixth term.

The reason why we do not yet see the effect of this change of the

Commission is that we look at finalised legislation in the first half of the eighth parliamentary term rather

than at new legislation proposed by the Juncker Commission since it entered office in 2014. The finalised

legislation in 2014-2015 and 2015-2016 includes a large number of files that were carried over from the

seventh term.

Figure 4: Number of ordinary legislative files and early agreements per legislative year

Figure 4 shows both the change in the volume of legislation and the change in the number of early

agreements over time (with the vertical dotted line representing EP elections). The majority of concluded

ordinary legislative files since 2006-2007 are early agreements, which means that the use of early

agreements increased in parliamentary terms in which the total amount of legislation went up. This

observation is in line with the efficiency-based explanation of early agreement, which linked early

agreements to the increase in the volume of legislation.

However, we need to be careful not to draw too many conclusions here: although we can observe that the

two developments took place during the same period, this does not mean that one is a consequence of the

other. Many other changes took place in the EU during the period in which there was an increase in use of

early agreements, and the trend towards early agreement could equally correspond with the cultural

European Parliament (2017) ‘Activity Report on the Ordinary Legislative Procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’,

PE 595.931, Brussels: European Parliament, Conciliation and Codecision Unit, p. 4.

And in some cases even from the sixth parliamentary term.

Investigation of informal trilogue negotiations since the Lisbon Treaty

rapprochement explanation. This explanation stated that early agreements came to be used more often

because the co-legislators became accustomed to working together in trilogue meetings. The implication is

that even if the volume of legislation goes down in the coming years, we should not expect a decline in the

use of early agreements.

3.2 The duration of decision-making: Early agreements now take longer to be concluded

What are the implications of the increased use of early agreement for the duration of ordinary legislative

procedures? This is a difficult question to answer – for reasons we discuss below – so we aim to give more

insight by carefully assessing our data on the duration of files. When mentioning the duration of legislative

files, we refer to the period of time between the introduction of the Commission proposal and the moment

that the agreement between the co-legislators is signed. In Figure 5, we compare the average duration of

early agreements with non-early agreements, taking into consideration the parliamentary term of the

concluded files.

The term ‘non-early agreement’ refers to legislative procedures where the co-legislators

agreed on the final legislative text without an agreement being negotiated in informal trilogue meetings at

first or early second reading. If we look at the duration of all legislation concluded between 1999 and 2016,

we see that legislative procedures without early agreements last 717 days on average. Procedures with early

agreements take 579 days on average. In order words, procedures without early agreement take significantly

longer than procedures with early agreements.

Figure 5: Duration (days) by early agreement and parliamentary term

As in previous figures, we exclude codification files.

We conducted an independent t-test to compare the duration of co-decision files under conditions of non-early agreements and early agreements.

For the full period, there is a significant difference in the average duration of non-early agreements (mean = 716.79; SD = 566.48) and early

agreements (mean = 578.63; SD = 313.18): t (894)) = 5.41, p = 0.00.

Investigation of informal trilogue negotiations since the Lisbon Treaty

However, as Figure 5 shows, the average duration hides important variation over time. Procedures without

early agreements took significantly longer in the fifth term (1999-2004) and the sixth term (2004-2009), but

the difference between the two types was no longer significant in the seventh term (2009-2014). In other

words, in the seventh term, procedures with and without early agreements took about the same amount of

time. Moreover, in the first half of the eighth parliamentary term, it was the procedures with early

agreements that took significantly longer.

This is a finding that requires more elaboration. As we pointed out in Section 2.1, early agreements are so

common nowadays that they have become the default option for ‘normal legislation’. In other words, they

are business as usual for legislation on which the three institutions do not immediately agree, and on which

some form of compromise is therefore needed. The procedures that are not concluded by early agreement

tend to be those that are relatively ‘uncomplicated’ and do not need any negotiation to be adopted.

It should

not be surprising that uncomplicated procedures do not take a lot of time to be adopted: the institutions can

agree on these procedures immediately, without the need for inter-institutional negotiations. We return to this

point when looking at the complexity of ordinary legislative files in Section 3.3.

There is an additional reason as to why early agreements in the first half of the eighth parliamentary term

(2014-2016) took such a long time to be concluded: in the first half of each parliamentary term, co-

legislators deal with a lot of ‘unfinished business’ carried over from the previous parliamentary term(s).

Indeed, 92 unfinished ordinary legislative files were carried over to the fifth parliamentary term, and the

numbers are similar in subsequent terms: 103 files in the sixth parliamentary term, 100 files in the seventh

parliamentary term, and, so far, 100 files in the eighth parliamentary term. As the incoming Parliament and

Commission need time to get up-to-date on pending legislation, ‘carried over’ files take significantly longer

to be concluded.

Moreover, as these files are primarily dealt with in the first half of the parliamentary term,

the average duration of procedures in the first half of the eighth parliament is very long.

Figure 6 shows that the average duration of legislative files is considerably shorter if we exclude unfinished

files that are carried over to the next parliamentary term; in other words, if we solely include files that are

both introduced and concluded within a single parliamentary term. This makes a considerable difference for

the duration of procedures in the first half of the eighth parliamentary term.

Ordinary legislation proposed

and adopted within one single parliamentary term takes, on average, 529 days, whereas legislation that is

carried over to the next term takes, on average, no less than 961 days.

In the future, the average duration of legislation concluded in the full eighth parliamentary term – the term

for which we now only have partial information – is likely to lie somewhere between the duration presented

The results of our independent t-tests are as follows. In EP5, there is a significant difference in the average duration of non-early agreements (mean

= 715.03; SD = 583.16) and early agreements (mean = 429.79; SD =151.33); t (124)) = 6.80, p = 0.00. In EP6, there is also a significant difference in

the average duration of non-early agreements (mean = 856.32; SD = 588.18) and early agreements (mean = 503.41; SD = 228.16); t (178)) = 7.09, p =

0.00. In EP7, there is no significant difference in the average duration of non-early agreements (mean = 598.01; SD = 399.61) and early agreements

(mean = 585.21; SD = 325.53); t (98)) = 0.27, p = 0.79. Finally, in the first half of EP8, there is again a significant difference – in the opposite

direction – in the average duration of non-early agreements (mean = 339.2; SD =365.23) and early agreements (mean = 817.92; SD = 378.27); t (52))

= –6.15, p = 0.00.

As mentioned in Section 2.1, there are a few exceptions; notably, the few files which were concluded at late second reading (based on inter-

institutional negotiations, but not concluded early).

Again, we conducted an independent t-test to compare the duration of these two categories. There is a significant difference in the average duration

of procedures which are started and finished within a single parliamentary term (mean = 529.25; SD = 267.03) and procedures which were carried

over from one term to another (mean = 961.49; SD = 663.29): t (393) = –11.94, p = 0.00.

The files that are carried over to the next term are concluded in the first half of the new term, and thus contribute heavily to the average duration in

that part of the term. When the full term is taken into consideration, the weight of the ‘unfinished business’ reduces.

Investigation of informal trilogue negotiations since the Lisbon Treaty

in Figures 5 and 6. On the one hand, the co-legislators will mostly deal with newly proposed legislation in

the second half of the eighth parliamentary term (because most legislation carried over from the seventh term

would have been concluded by then), which will reduce the high average duration in Figure 5. On the other

hand, the ‘more difficult’ new (Juncker) Commission proposals are still under negotiation because more

difficult proposals tend to take longer to be concluded. Therefore, many of the new Commission proposals

will be concluded in the second half of the eighth term, which means that the average duration in the full

eighth term is likely be higher than Figure 6 would suggest.

Figure 6: Duration (days), excluding unfinished files of previous parliaments

As early agreements have evolved from being exceptions to being business as usual, it is difficult to draw

conclusions on the difference they make to the duration of the ordinary legislative procedure. Nonetheless,

we can compare the average duration of ordinary legislative files over time, which were 693 days in the fifth

parliamentary term, 633 days in the sixth term, and 587 days in the seventh term.

These differences are not

large, and we cannot yet include the eighth term in the comparison.

Yet, given that the decline in duration

took place in the same period as the workload of the co-legislators increased, and the number of EU member

states went up, we believe it is likely that the use of early agreement has played a role in the reduction of the

time it takes to conclude ordinary legislative files.

In sum, we have seen that as early agreements are now used for almost all ordinary legislation, including

difficult legislation, they actually take a long time, on average, to be concluded. On the other hand, as non-

early agreements are today used mainly for the small number of uncomplicated legislative files that the

institutions have to deal with, they do not take very long to be concluded. Nonetheless, on average, the

ordinary legislative procedure takes less time to be concluded than in the past, and the use of early

Using ANOVA followed by a Bonferroni test, the average duration of co-decision files in these terms is not significantly different, with the

exception of the duration in the seventh term, which is significantly shorter than the duration in the fifth term.

It is difficult to predict the average duration of co-decision in this term based on the figures for the first half of that term (as explained above).

Investigation of informal trilogue negotiations since the Lisbon Treaty

agreements is likely to have contributed to this decrease. This leads us to conclude that while early

agreements can take very long to be concluded, their increased use has still expedited the ordinary legislative

process as a whole.

3.2 Early agreements and European Parliament committees

The next question we address is about the variation we see across responsible EP committees. In Figure 7,

we present the percentage of ordinary legislative files that committees concluded by means of early

agreement, presented per parliamentary term (with the total number of files per committee in parentheses).

Please note that we use the acronyms of the committees in the figures.

The order of the committees in the

figures is alphabetical, based on the committee acronyms. Please also note though that some committees had

few files.

Figure 7: Early agreements by responsible EP committee and parliamentary term

Again, we exclude codification files. This solely has implications for the JURI committee, as it is this committee that is responsible for codification

files. Had codification been included, the percentage of JURI files concluded as early agreements would have been much lower (since codification

files are never concluded using inter-institutional trilogue meetings).

AFET: Foreign Affairs; DEVE: Development; INTA: International Trade; BUDG: Budgets; CONT: Budgetary Control; ECON: Economic and

Monetary Affairs; EMPL: Employment and Social Affairs; ENVI: Environment, Public Health and Food Safety; ITRE: Industry, Research and

Energy; IMCO: Internal Market and Consumer Protection; TRAN: Transport; REGI: Regional Development; AGRI: Agriculture and Rural

Development; PECH: Fisheries; CULT: Culture and Education; JURI: Legal Affairs; JURI & IM: Legal Affairs and Internal Market; LIBE: Civil

Liberties, Justice and Home Affairs; AFCO: Constitutional Affairs; FEMM: Women’s Rights and Gender Equality. With the exception of JURI &

IM, these are the current committee names, but the names and competences of some committees have changed somewhat over time.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Figure 7 demonstrates that most EP committees did not use early agreements in the fifth parliamentary term

(1999-2004). Of the committees that were responsible for at least five ordinary legislative files, only the

Committee on Budgets (BUDG) and the Committee on Civil Liberties, Justice and Home Affairs (LIBE)

used early agreements with some frequency. This changed in the sixth term (2004-2009), when almost all

committees regularly used early agreements. The LIBE and BUDG committees further increased their use of

early agreement (81 and 100 percent, respectively), and other frequent users now included the Committee on

Economic and Monetary Affairs (ECON) and Legal Affairs (JURI) (80 and 83 percent, respectively). The

Committees on Fisheries (PECH), International Trade (INTA) and Agriculture and Rural Development

(AGRI) were not yet ‘co-decision committees’ – this would only happen when the Lisbon Treaty entered

into force in 2009 –, but they did already use early agreements for the few ordinary legislative files they dealt

with.

By the seventh parliamentary term, early agreement had become the default option for almost all committees

and this has continued to be the case in the first half of the eighth term. Of the five committees that became

full-fledged ‘co-decision committees’ in 2009 (the AGRI, INTA, LIBE and PECH committees and the

Committee on Regional Development (REGI)), it is only the INTA committee that does not often use early

agreements.

In one of our interviews with EP committee officials, it was suggested that this is because the

relevant Council still thinks of international trade as an inter-governmental policy area, which makes

negotiating in trilogue meetings more of a challenge. Indeed, the EP’s 2009-2014 Activity Report

emphasises that it was the INTA committee that faced the biggest change after 2009, because it had to deal

with, among other things, a large number of ordinary legislative files ‘mostly in areas on which it was

previously not even consulted’.

The INTA committee has also been responsible for a lot of ‘urgent’

ordinary legislative files, which were quickly adopted by the co-legislators without (major) amendments. As

we pointed out in Chapter 2, such files are more often adopted without inter-institutional negotiations

because the co-legislators can, and have to, agree on them quickly. The other ‘new’ co-decision committees

seem to have adapted quickly to the use of early agreements. As the 2009-2014 Activity Report stated, these

committees ‘rapidly became acquainted with the tried-and-tested informal interinstitutional practices for the

adoption of legislation, engaging in sustained periods of negotiation’.

This suggests that EP committees do

not need to have extensive experience with the ordinary legislative procedure to become frequent users of

early agreements.

To conclude, while the use of early agreements was limited to a few EP committees, almost all committees

now strongly rely on early agreements in the ordinary legislative process. Moreover, the use of early

agreement is not limited to committees that have a lot of experience as co-legislators: the ‘new’ co-decision

committees after the Lisbon Treaty of 2009 have quickly adapted to the use of early agreements.

Although the LIBE Committee only became a full co-decision committee in 2009, it already had considerable experience with co-decision, and it

was one of the first adopters of early agreements.

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit, p. 13.

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit, p. 13.

Investigation of informal trilogue negotiations since the Lisbon Treaty

3.3 Early agreements, type of legislation and complexity

In the last set of analyses, we look at some of the characteristics of finalised ordinary legislative files. First,

the use of early agreements may vary across types of legislation. In Figure 8, we present early agreements as

a percentage of all ordinary legislative files by type of legislation, using the categories that were identified by

Reh and her colleagues in their study on early agreements (though excluding codification).

Figure 8: Early agreement and type of legislation (1999-2016)

The figure demonstrates that early agreement has been most common for adaptations of the procedure to

Comitology (93 percent) for the full period from 1999 to 2016. Such files include adaptations to the new

regulatory procedure with scrutiny, and adaptations to the implementing powers conferred on the

Commission following the adoption of Council Decision 2006/512/EC of 17 July 2006.

These changes

required the adaptation of some procedures that had been laid down in existing legislation. These adaptations

were mainly adopted in the sixth parliamentary term – a term in which early agreement had already become

common. Hence, the percentages for this type of legislation are to be expected.

Other types of legislation for which early agreements have often been used are recasts, which introduce

minor substantive changes to legislation (77 percent), and new legislation which also includes provisions to

repeal old legislation (68 percent). Completely new legislation (without repeals) and files that amend the

contents of existing legislation, are in between when it comes to the frequency of early agreements (51 and

53 percent, respectively). Finally, early agreements are not used very often for legislation that changes dates

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1121.

Bressanelli, E., Héritier, A., Koop, C. and Reh, C. (2014) ‘The informal politics of codecision: Introducing a new data set on early agreements in

the European Union’, EUI Working Paper 2014/64 RSCAS (EUDO), p. 10.

Investigation of informal trilogue negotiations since the Lisbon Treaty

in a previously adopted legislative file (38 percent). Examples include changes of the date on which

legislation enters into force, changes of the date of the review of an act, and changes of the date by which the

Commission needs to report on the implementation of the legislation. Early agreements are least common for

files that repeal existing legislation (32 percent). The last two categories are usually concluded early (at first

reading), but no inter-institutional negotiations are used in the process.

The next characteristic we look at is the complexity of files. It has been hypothesised that early agreements

are particularly attractive for highly complex files. As these files ‘involve uncertainty about substantive

cause-and-effect relations, […] their negotiation requires special expertise as well as considerable time and

effort’.

At the same time, complex files are more under the radar, as it is less obvious to a general audience

what their distributional consequences will be.

Hence, highly complex files do not only require a lot of

special expertise – and are, thus, costly – but they also attract low levels of public attention. Together, these

features may make the use of early agreements particularly attractive. The complexity of legislative files is

difficult to capture with a quantitative measure, but we follow Reh and her colleagues in using a proxy

measure: we consider ordinary legislative files more complex if more recitals are included in the

Commission’s proposal.

Figure 9: Number of recitals, by early agreement and parliamentary term

Figure 9 shows the number of recitals in the Commission’s proposal for early agreements and non-early

agreements. For the full period (1999-2016), the average number of recitals in early agreements is 22.2,

while the average in non-early agreements is 15.5.

Yet, as the figure shows, there is some important

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1123-24.

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1124.

Indeed, for the full period, there is a significant difference in the number of recitals in non-early agreements (mean = 15.53; SD = 10.92) and early

agreements (mean = 22.18; SD = 18.36); t (1306) = –8.40, p = 0.00.

Investigation of informal trilogue negotiations since the Lisbon Treaty

variation over time. In line with the findings of Reh and her colleagues

, there is no significant difference in

the number of recitals in non-early agreements and early agreements in the fifth and sixth parliamentary term

(1999-2009).

However, this has started to change in the seventh term (2009-2014), and the change has

continued in the first half of the eighth term (2014-2016).

Therefore, since 2009, files concluded by early

agreement include significantly more recitals than files that are not concluded by early agreement. In other

words, the more complex legislation tends to be dealt with in informal trilogue meetings leading to early

agreements.

These findings should not completely surprise us. We know that, in recent years, early agreements have

become the default option. On the other hand, the absence of an early agreement is – with some exceptions –

associated with relatively uncomplicated files. These are files that the co-legislators can agree on quickly,

without having to make a lot of amendments. Our findings on the number of recitals – a proxy measure of

the complexity of the file – are fully in line with this observation. That is, non-early agreements are

nowadays associated with low levels of complexity. In regard to the first half of the eighth term, the

relatively high levels of complexity in early agreements may be related to the high volume of unfinished

business that the new Parliament had to deal with. The legislative files that were carried over to the eighth

term were files that the previous Parliament had not been able to finalise by the end of its term, which

suggests that the files were complex, and perhaps too complex to be adopted by the end of the previous term.

Figure 10: Early agreements and legislative instruments (2009-2016)

We have also collected data on the legislative instruments that have been used in the ordinary legislative

procedure in the seventh and first half of the eighth parliamentary term (2009-2016). Ordinary legislative

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining secluded decision making in the

European Union,’ Comparative Political Studies 46 (9): 1131.

In EP5, there is no significant difference in the number of recitals in non-early agreements (mean = 14.72; SD = 9.32) and early agreements (mean

= 15.38; SD = 10.11); t (32) = –0.34, p = 0.74. In EP6, there is no significant difference in the number of recitals in non-early agreements (mean =

16.42; SD = 11.42) and early agreements (mean = 16.77; SD = 12.72); t (344) = –0.28, p = 0.78.

In EP7, there is a significant difference in the number of recitals in non-early agreements (mean = 18.54; SD = 15.86) and early agreements (mean =

23.81; SD = 18.63); t (122) = –2.61, p = 0.01.

In the first half of EP8, there is a significant difference in the number of recitals in non-early agreements (mean = 12.7; SD = 8.10) and early

agreements (mean = 32.88; SD = 25.77); t (117) = –6.50, p = 0.00. For the full period (1999-2016), there is also a significant difference in the number

of recitals in non-early agreements (mean = 15.53; SD = 10.92) and early agreements (mean = 22.18; SD = 18.36); t (1306) = –8.40, p = 0.00.

Investigation of informal trilogue negotiations since the Lisbon Treaty

files take the form of regulations, directive, or decisions. However, as Figure 10 shows, there are no clear

differences in the use of early agreement between these different legal instruments in the 2009-2016 period:

early agreements are business as usual in the case of regulations, directives and decisions.

Figure 11: Mandatory consultation of the EESC and early agreement (2009-2016)

Finally, we looked at the files for which consultation of the European Economic and Social Committee

(EESC) was mandatory, both in the seventh parliamentary term and in the first half of the eighth term (2009-

2016). The EESC may also adopt opinions on its own initiative, but we only have information on the files

which – following the Treaties – require consultation of the Committee. As Figure 11 demonstrates, the

percentage of ordinary legislative files on which the EESC is consulted is slightly higher in the first half of

the eighth parliamentary term than in the seventh term. Of the files on which the EESC was consulted, 90

percent were concluded by early agreement. This is slightly higher than the overall percentage of early

agreements, when all ordinary legislative files in the same period are taken into consideration (82 percent).

This section has demonstrated that the use of early agreement varies with the characteristics of the legislative

files. Early agreements are most common for legislative files which adapt procedures to Comitology, which

introduce substantive changes to legislation, and which introduce new legislation as well as provisions to

repeal old legislation. We may think of these three types of legislation as moderately demanding. Early

agreements are slightly less common (though still often used) for the most demanding types of legislation;

namely, completely new legislation and legislation that amends the contents of previously adopted

legislation. Interestingly, early agreements are least common for the least demanding legislation: that which

makes changes to the dates in existing legislation or solely repeals existing legislation.

Looking at the complexity of ordinary legislative files, we have seen that in the early period of co-decision,

the level of complexity of early agreements – as measured by the number of recitals in the Commission

Indeed, using the Wilcoxon-Mann-Whitney test, we find that early agreements are significantly more likely on files on which the EESC has to be

consulted (z = –6.034; p = 0.00).

Investigation of informal trilogue negotiations since the Lisbon Treaty

proposal – was not different from the level of complexity of non-early agreements. But this has changed in

recent years: levels of complexity in early agreements are now significantly higher than levels of complexity

in non-early agreements. In other words, early agreements have become the default option for complex files,

while they are not used for the least complex files because the latter may not need inter-institutional

negotiations.

In contrast to the previous findings, there are no clear differences in the use of early agreement across

legislative instruments: early agreements equally likely in regulations, directives, and decisions. Finally, files

that require consultation of the EESC are somewhat more likely to be concluded as early agreements, which

may be related to the substantive features of these files; that is, files for which EESC consultation is

mandatory are likely to be more demanding and more complex files.

3.4 Recent findings on the use of early agreements

In the last section of this chapter, we take a look at some recent findings from academic studies on the use of

early agreements. A recent study by Brandsma focuses on the number of informal trilogue meetings held on

ordinary legislative files at first reading in the seventh parliamentary term (2009-2014). Brandsma addresses

the question how we can account for the variation in the number of trilogue meetings held on individual

files.

He finds that a few factors increase the number of trilogue meetings on individual files:

 Politicisation in the Council: ordinary legislative files that have more often been a B-point on the

Council’s agenda – that is, files that cannot be adopted without debate in the Council (unlike A-

points) – have a higher number of trilogue meetings;

 Contestation in the EP: ordinary legislative files with a large number of amendments tabled against

the EP committee’s report have a higher number of trilogue meetings.

In addition, more trilogue meetings are held on ordinary legislative files that include more articles, with the

latter being used by Brandsma as another proxy measure of the complexity of the file. Finally, files

negotiated by EP committees that have become new co-decision committees with the Lisbon Treaty (the

PECH, AGRI and INTA committees) have not had more informal trilogue meetings than files negotiated by

the ‘old’ ordinary legislative committees. Although this is an interesting finding, we should not be

completely surprised given that our previous analyses have suggested that the new co-decision committees

after the Lisbon Treaty adapted quickly to the use of trilogues.

The use of early agreements has also raised the question of implications for legislative behaviour in the EP.

The inter-institutional compromises negotiated in informal trilogues still need to be formalised – or ‘rubber-

stamped’ – in the plenary meeting of Parliament and the Council. Yet, although MEPs can still table

amendment to the negotiated compromise, they are under considerable pressure not to challenge the inter-

institutional deal.

A recent study by Bressanelli and his colleagues looks at the impact of early agreements

Brandsma, G.J. (2015) ‘Co-decision after Lisbon: The politics of informal trilogues in European Union lawmaking,’ European Union Politics 16

(2): 300-19.

Rasmussen, A. and Shackleton, M. (2005) ‘The scope for action of the European Parliament negotiators in the legislative process: Lessons of the

past and for the future’, paper presented at the EUSA Biannual Meeting, Austin, TX, March-April.

Investigation of informal trilogue negotiations since the Lisbon Treaty

on the cohesion of the political groups in Parliament.

The authors argue that failure to reach an early

agreement in plenary can come at a cost at several levels, namely:

 Reputational costs: Failure of a negotiated compromise in plenary harms the credibility of

Parliament (and the EP’s negotiating team) in the eyes of the Council and the electorates;

 Political costs: Failure of a negotiated compromise at first reading increases the likeliness of a

‘normal’ second reading which has stricter majority requirements and time limits. From a long-term

perspective, failure also makes reaching inter-institutional compromises in the future more

challenging, which negatively affects the ability of EP groups to deliver policies;

 Transaction costs: Negotiations in trilogue meetings are associated with considerable transaction

costs – the costs of collecting information, reaching an intra- and inter-party consensus, and

participating in the trilogue meetings. Such costs cannot be recovered in any way in case of failure

of the compromise in plenary.

These considerations should lead the political groups in Parliament to invest more heavily in discipline

among their MEPs and in finding a way to achieve consensus, which would make MEPs more inclined to toe

the party line in plenary. Bressanelli and his colleagues find partial empirical support for this argument:

levels of cohesion are indeed higher when political groups vote on inter-institutional compromises, but only

among the centrist parties in Parliament. This includes the European People’s Party (EPP), the Progressive

Alliance of Socialists and Democrats (S&D; formerly the Party of European Socialists or PES) and the

Alliance of Liberals and Democrats for Europe (ALDE), which vote more cohesively on early agreements.

In a nutshell, recent academic studies have found that the number of informal trilogue meetings is higher

when the legislation attracts more contestation in Parliament and when it is considered to be more political in

nature in the Council. In terms of legislative behaviour, levels of cohesion in Parliament are also higher

when votes are cast on early agreements, but this is only the case for the centrists groups – the EPP, the

S&D, and the ALDE.

Bressanelli, E., Koop, C., and Reh, C. (2016) ‘The impact of informalisation: Early agreements and voting cohesion in the European Parliament,’

European Union Politics 17 (1): 91-113.

Bressanelli, E., Koop, C., and Reh, C. (2016) ‘The impact of informalisation: Early agreements and voting cohesion in the European Parliament,’

European Union Politics 17 (1): 97-99.

4. Rules and practices in early agreement

All three main EU institutions (the Commission, the EP, and the Council) have had to get used to the

ordinary legislative procedure and the increased use of early agreements. However, the EP is the institution

that has undergone the most significant changes as it has altered its rules of procedure on numerous

occasions to make trilogues more inclusive, transparent, and accountable. The EP’s handling of trilogues has

become more formalised over the years. This particularly concerns the composition of the trilogue

negotiating teams, the ways in which the EP’s mandates for entering into trilogue negotiations are given, and

how the outcomes are reported back to the rest of the Parliament. Although the Council has not ‘formalised’

its practices of trilogues into its Rules of Procedure, it has developed a fairly routinised way of preparing for

trilogues and reporting back.

This chapter provides an overview of the rules and practices governing informal trilogues both within and

between the EU institutions. It is divided into three main sections. First, we present the commitments made

in terms of transparency in EU decision-making in primary and secondary legislation, and its interpretation

by the Court of Justice of the European Union (CJEU). This gives an understanding of the, at times, differing

interpretation of how to strike a balance between efficient and open decision-making. Second, we provide an

overview of the inter-institutional rules and practices of informal trilogues in the past and at present. Lastly,

we turn to the rules and practices prevalent within the institutions and how they have developed over time. It

is important to know about these changes to fully comprehend the nature of today’s rules and practices. As

informal trilogues are by definition not documented, we gained insight into the practices by relying on

existing academic literature and our interviews with 18 EU decision-makers. Please see the Chapter 1 and

Annex A for more information on the interviews.

4.1 Rules governing access to documents and their interpretation

There are a number of commitments to, and claims made about, the transparency of EU decision-making in

primary and secondary legislation, and in the rulings of the CJEU. Under the Lisbon Treaty (Article 15 (1)

TFEU), there is an obligation for all EU institutions to conduct their work as openly as possible. Articles

15(2) and 15(3) TFEU explicitly state that the EP shall meet in public, as should the Council when

considering and voting on a draft legislative act. In addition, the two institutions are under the obligation to

act publicly and make documents relating to the legislative procedure publicly available. The right of access

to documents (freedom of information) constitutes a fundamental right as emphasised by Article 42 of the

EU Charter of Fundamental Rights, which enjoys the same legal status as the Treaties (Article 6(1) TEU).

However, neither the Lisbon Treaty nor any previous treaties make reference to informal trilogue meetings,

although principles of transparent, open, and inclusive decision-making are enshrined in the Treaty.

The right of citizens to access documents – such as those used in trilogue meetings – is also reflected in

secondary legislation dating back to 2001. Regulation No. 1049/2001 on public access to documents held by

the EU institutions (the Access to Documents Regulation)

establishes that all legislative documents held by

Regulation (EC) No 1049/2001 of the European Parliament and of the Council 30 May 2001 regarding public access to European Parliament,

Council and Commission documents, OJ L 145/43, 31 May 2001.

Investigation of informal trilogue negotiations since the Lisbon Treaty

the Commission, the Council, and the EP are in principle public unless they are protected by the

Regulation’s specific exemptions under Article 4.

The EU institutions can decline access to a document where disclosure undermines the protection of (a) the

public interest in regard to public security, defence and military matters, international relations, and the

financial, monetary or economic policy of the Community or a Member State; (b) privacy and the integrity

of the individual; (c) commercial interests, including intellectual property; (d) court proceedings and legal

advice; or (e) the purpose of inspections.

Disclosure can also be refused in the case of documents for internal use by the EU institutions where a

decision has not yet been taken, giving the institutions ‘space to think’. Public access to such documents can

be denied even after a decision has been taken ‘if disclosure of the document would seriously undermine the

institution’s decision-making process, unless there is an overriding public interest in disclosure’ (Article

4(3)).

4.1.1 EU case-law

The interpretation of the legislative transparency requirements in primary and secondary law has been a bone

of contention between the CJEU, often promoting transparency of core legislative documents, and the

Council, often emphasising flexibility in the negotiations and prioritising effectiveness over transparency.

In its case-law, the CJEU has in principle required that the application of the exemptions under Article 4 in

the Access to Documents Regulation is examined on a case-by-case basis. However, case-law also shows

that an assessment of the potential harm to the interests safeguarded by the Regulation if documents are

made public must be reasonably foreseeable and not merely hypothetical.

The CJEU has made a number of landmark rulings regarding transparency in EU decision-making and

access to documents, both prior to and after the Lisbon Treaty. The Turco ruling (pre-Lisbon) stands out as

an example of the Court setting a high threshold for public access to documents by referring to all

information, which has formed the basis for a legislative act. The Turco case concerned the extent of access

to Council Legal Service opinions. As it appears from the extract of the ruling below, the CJEU establishes a

clear link between openness and public trust in EU decision-making in its ruling:

‘[Openness] enables citizens to participate more closely in the decision-making process and

guarantees that the administration enjoys greater legitimacy and is more effective and more

accountable to the citizens in a democratic system. Those considerations are clearly of particular

relevance where the Council is acting in its legislative capacity […]. Openness in that respect

contributes to strengthening democracy by allowing citizens to scrutinise all the information,

which has formed the basis of the legislative act. The possibility for citizens to find out the

Leino, P. (2014) ‘Transparency, Participation and EU Institutional Practice: An inquiry into the Limits of the “Widest Possible”’, EUI Working

Paper, Law 3/2014.

Leino, P. (2016) ‘On knowledge as power: transparency of EU law-making procedures’, EU Law Analysis Blogpost, 10 January 2016, available at

http://eulawanalysis.blogspot.dk/2016/01/on-knowledge-as-power-transparency-of.html (last accessed: 10 April 2017).

Leino, P. (2016) ‘On knowledge as power: transparency of EU law-making procedures’, EU Law Analysis Blogpost, 10 January 2016, available at

http://eulawanalysis.blogspot.dk/2016/01/on-knowledge-as-power-transparency-of.html (last accessed: 10 April 2017).

Investigation of informal trilogue negotiations since the Lisbon Treaty

considerations underpinning legislative action is a precondition for the effective exercise of their

democratic right.’

In the Turco case, the CJEU explicitly stated that the Access to Documents Regulation imposes, in principle,

an obligation to disclose the opinions of the Council’s legal service relating to a legislative process’ (para. 68

of the ruling). According to the CJEU, access to documents can only be rejected in exceptional

circumstances if the information from the legal service is ‘of particularly sensitive nature or having a

particularly wide scope that goes beyond the context of the legislative process in question’. This exception

can only be used ‘for the period during which protection is justified on the basis of the content of the

document’ (paras. 69-70). Despite this, the Council continues to restrict access to its legal service opinion

and the Council’s Rules of Procedure have not been updated to reflect the ruling.

Another relevant example is the Access Info case

from 2013, where the non-governmental organisation

(NGO) Access Info Europa, which promotes freedom of information in the EU, challenged the Council’s

rejection to grant access to a legislative document including footnotes indicating the position of individual

member states. The disagreement focused on the question of whether access to member state positions would

adversely affect the efficiency of decision-making and, if so, whether efficiency should be prioritised over

transparency. The Council maintained that its flexibility in decision-making ‘would be reduced if the identity

of the delegations were disclosed too early in the procedure, in that it would have the effect of triggering

pressure from public opinion, which would deprive the delegations themselves of the flexibility needed to

ensure the effectiveness of the Council’s decision-making process’.

The Council did not find that the public

identification of national delegations’ positions was necessary to ensure a democratic debate and prioritised

effectiveness over openness. The Council lost the case in both the General Court and, after appealing it, in

the Court of Justice. According to the Court, the requested document is not regarded as sensitive or

constituting a real risk to fundamental interests, as the member states often move away from their initial

positions in order to reach agreement.

Together, the Turco and Access Info rulings establish suggest that access to the Council’s legal service

opinions and national delegation positions in the Council should be open to scrutiny by the public, even

when a legislative procedure is ongoing. In contrast, the Council interprets the application of primary and

secondary law on publicity of its meetings as being restricted to documents that are submitted to the Council

for preparation of its deliberation or vote.

This raises the question of at which decision-making stage

legislative transparency should apply. If the publicity of documents were, indeed, limited to documents

submitted to ministers a few days before a formal ministerial Council meeting, transparency would only

apply to the closing stages of decision-making, when ministers formally adopt compromises and close a file.

The Council’s interpretation, as laid out in the aforementioned cases, would effectively prevent access to

documents from its lower decision-making levels.

Joined Cases C-39/05 P and C-52/05 P Kingdom of Sweden and Maurizio Turco v the Council, paras. 45-46

Curtin, D. and Leino, P. (2016) ‘Openness, transparency and the right of access to documents in the EU: In-depth analysis for the PETI committee’,

PE 556.973, Brussels: European Parliament, p.6.

C-280/11 P Council v Access Info Europe.

C-280/11 P Council v Access Info Europe, para. 63.

Curtin, D. and Leino, P. (2016) ‘Openness, transparency and the right of access to documents in the EU: In-depth analysis for the PETI committee’,

PE 556.973, Brussels: European Parliament.

Investigation of informal trilogue negotiations since the Lisbon Treaty

The Turco and Access Info rulings indicate that the Council and the Court have different interpretations of at

which decision-making stage publicity should apply. According to some legal scholars, this is seen as

problematic, as the responsibility for making certain legislative documents publicly available in practice

shifts from the EU legislators to the courts.

The CJEU cannot redesign the general EU transparency regime,

and can only deal with the issue on an ad hoc basis when specific cases are brought before it. The same

applies to the EU’s Ombudsman, although she also has the ability to launch own-initiative enquiries to bring

specific secretive practices to light (Article 228 (1) TFEU).

4.1.2 Trilogues and access to documents

It has been 16 years since the adoption of the Access to Documents Regulation and discussions about

reforming the Regulation have been pending since 2008.

100

While the EP has used its own-initiative reports

and resolutions to call for a revision of the Regulation on a number of occasions, the Council appears

hesitant.

101

Parliament has, in particular, criticised ‘the space to think’ exemption in Article 4 (3) of the

Access to Documents Regulation for being obsolete when it concerns situations where a decision has not yet

been taken on legislative matters.

102

In its most recent resolution on access to documents from 28 April 2016,

the EP calls on the institutions to ensure greater transparency of trilogue meetings, whilst acknowledging the

need to provide the co-legislators with adequate ‘space to think’.

103

The resolution criticises trilogues for not

giving citizens the power to scrutinise trilogue meetings and for giving interest groups unequal access to

trilogue documents as only well-connected and resourceful interest groups manage to get hold of leaked

trilogue documents (so it is argued). The resolution states that documents used in trilogues (such as agendas,

summaries of outcomes, minutes and general approaches in the Council) are part of legislative procedures

and should therefore in principle not be treated differently from other legislative documents. On this ground,

the EP suggests that lists of trilogue meetings, agendas, and summaries of agendas should be made publicly

available. The resolution does not, however, state when this should be done; that is, when negotiations are

still ongoing or are finalised.

The resolution also states that MEPs consider it regrettable that the revision of the Access to Documents

Regulation is still stalled in the Council and call on the Council to adopt a constructive approach that takes

into account the EP’s first reading position adopted on 15 December 2011 on the proposal to revise the

Regulation. For instance, the EP wants the revised regulation to reflect the new treaty provisions and relevant

EU case-law, and calls for a common approach on registers of documents and a common access point to EU

documents. In the nine years since the Commission put forward a proposal to revise the Regulation, there has

Curtin, D. and Leino, P. (2016) ‘Openness, transparency and the right of access to documents in the EU: In-depth analysis for the PETI committee’,

PE 556.973, Brussels: European Parliament.

100

See e.g. Harden, I. (2009) ‘The Revision of Regulation 1049/2001 on Public Access to Documents’, European Public Law, 15(2): 239-256;

European Parliament (2013) ‘European Parliament resolution of 12 June 2013 on the deadlock on the revision of Regulation (EC) No 1049/2001’,

2013/2637(RSP), Brussels: European Parliament; European Commission (2011) ‘Proposal for a Regulation of the European Parliament and of the

Council amending Regulation (EC) No 1049/2001 regarding public access to European Parliament, Council and Commission documents’,

COM(2011) 137 final/2.

101

Curtin, D. and Leino, P. (2016) ‘Openness, transparency and the right of access to documents in the EU: In-depth analysis for the PETI

committee’, PE 556.973, Brussels: European Parliament.

102

European Parliament (2011) ‘European Parliament resolution of 14 September 2011 on public access to documents (Rule 104(7)) for the years

2009-2010’, 2010/2294(INI), Brussels: European Parliament.

103

European Parliament (2016) ‘European Parliament resolution of 28 April 2016 on public access to documents (Rule 116(17)) for the years 2014-

2015, 2015/2287(INI), Brussels: European Parliament.

Investigation of informal trilogue negotiations since the Lisbon Treaty

been no prospect of adoption of a revised regulation due to strongly diverging views between the co-

legislators.

104

Public access to documents has received renewed attention with the ongoing De Capitani court case. On 18

September 2015, Emilio De Capitani, the previous head of the LIBE Committee Secretariat, lodged an

appeal against the EP’s decision to decline full access to documents used during trilogue meetings,

particularly the four-column documents submitted to trilogue on the proposal for a Regulation on the

European Union Agency for Law Enforcement Cooperation and Training (Europol). The fourth column

becomes publicly available when the compromise reached during trilogue meetings needs to be approved in

Parliament’s plenary. De Capitani tried to get access to the four-column document through the EP’s Bureau

on the proposal for a Europol Regulation, who declined his request. The Bureau put forward four main

reasons for declining access to all the four-column documents

105

1. Disclosing the Council’s position before the end of the negotiations would impact negatively on the

good cooperation between the co-legislators and the specific negotiation process, resulting in a loss

of mutual trust and time;

2. Disclosure might result in public pressure from national authorities and interest groups on the EP’s

negotiating team (e.g., the rapporteur and the shadow rapporteurs) due to the sensitive issues of data

protection and Europol’s management board under discussion, which would complicate the EP’s

ability to find a common line between its political groups;

3. Disclosure would make the Council Presidency more wary about sharing information and

cooperating with Parliament’s negotiating team, and would have a particularly negative impact on

the rapporteur’s standing with his/her colleagues and the Council presidency;

4. Disclosure of positions that do not necessarily reflect the institutions’ final position risks misleading

the public about the actual positions of the institutions.

Following the Bureau’s refusal to grant De Capitani access to the four-column documents used in trilogues

on the then ongoing Europol dossier, De Capitani brought the case before the General Court, where he relied

on two pleas in law. In the first plea, he argued that Parliament has misapplied Article 4(3) first subparagraph

of the ‘Access to Documents Regulation’

106

(the ‘space to think’ provision) as access to the requested

documents does not ‘specifically, effectively and in a non-hypothetical manner undermine the legislative

process’. He also argued that ‘Parliament ignores that, notably after the Lisbon Treaty, legislative

preparatory documents are subject to the principle of the widest possible access’.

107

In the second plea, De

Capitani claimed that the EP has failed to explain why it refused access to the requested document, by not

stating reasons with regard to (1) why the disclosure of the documents would seriously undermine the EP’s

decision-making process in question, and (2) why no overriding public interest exists in this case. According

104

European Parliament (2011) ‘European Parliament legislative resolution of 15 December 2011 on the proposal for a regulation of the European

Parliament and of the Council regarding public access to European Parliament, Council and Commission documents (recast) (COM(2008)0229 – C6-

0184/2008 – 2008/0090(COD))’, Brussels: European Parliament

105

De Capitani, E. (2015) ‘(EU’s) laws are like sausages, you should never watch them being made’, European Area of Freedom Security & Justice

Free Group, Blogpost, 12 July 2015, available at https://free-group.eu/2015/07/12/eus-laws-are-like-sausages-you-should-never-watch-them-being-

made/ (last accessed: 12 April 2017).

106

Article 4(2) first subparagraph of the ‘Access to Documents Regulation’ states that: ‘Access to a document, drawn up by an institution for internal

use or received by an institution, which relates to a matter where the decision has not been taken by the institution, shall be refused if disclosure of the

document would seriously undermine the institution's decision-making process, unless there is an overriding public interest in disclosure’.

107

Case T-540/15, De Capitani v European Parliament

Investigation of informal trilogue negotiations since the Lisbon Treaty

to De Capitani, the failure to state reasons for rejecting access to the requested documents contravenes both

Article 296 TFEU

108

and Article 4(2), first paragraph of the Access to Documents Regulation.

The still pending De Capitani case clearly highlights the tension between securing effectiveness and

transparency in decision-making, particularly when dossiers have not been finalised and negotiations are

ongoing. The outcome of the De Capitani case might set a precedent for how the delicate balance between

effectiveness and transparency is struck in the future, at least in Court rulings, although it is too early to say

what role it will play in future cases and institutional reforms. The Court’s decision is expected sometime in

2017. Although the Court has in several of its rulings advocated transparency of legislative documents, its

rulings have little impact as long as the EU institutions apply them selectively and do not alter their overall

transparency regimes.

109

4.2 Inter-institutional rules and practices

When early agreements first emerged following the 1999 Amsterdam Treaty, there were no clear rules within

or between the EU institutions on how to deal with trilogues early in the decision-making process. The Joint

Declaration of May 1999

110

merely talked about establishing ‘appropriate contacts’ between the EU

institutions to reach agreement in first and second readings. It was also stated that each institution should

make its own rules regarding the composition of its negotiating team and its mandate for negotiations.

The declaration was updated in 2007 and remains in force today.

111

The 2007 declaration encourages the

continued use of trilogue meetings as a means to reach compromises, and states that the EU institutions must

fully respect the principles of transparency, accountability, and efficiency. The updated joint declaration

clarifies some of the working methods and practical arrangements for pursuing trilogue meetings and states

that the early conclusion of acts should be sought when appropriate. In terms of information sharing of

documents used in forthcoming trilogue meetings, it is stated that these shall be circulated in advance to all

participants, without indicating a specific time. It also aims to announce, where practicable, when trilogues

take place in the EP and the Council in order to enhance transparency. In practice, this is usually done during

the EP’s committee meetings (which are usually open to the public), but information on trilogue meeting

dates is not announced in the EU institutions’ respective databases, such as the Council Register and the

Legislative Observatory. This suggests that there is room for improvement in meeting this commitment.

The joint declaration also spells out the procedure for reaching early agreements in terms of how the EP and

Council communicate their intentions to opt for an early agreement. In an early agreement at first reading,

the chair of Coreper (i.e., the presidency) forwards a letter to the chair of the relevant parliamentary

committee, which specifies the substance of the agreement, in the form of amendments to the Commission’s

108

Article 296 TFEU states that: ‘Where the Treaties do not specify the type of act to be adopted, the institutions shall select it on a case-by-case

basis, in compliance with the applicable procedures and with the principle of proportionality. Legal acts shall state the reasons on which they are

based and shall refer to any proposals, initiatives, recommendations, requests or opinions required by the Treaties. When considering draft

legislative acts, the European Parliament and the Council shall refrain from adopting acts not provided for by the relevant legislative procedure in the

area in question’ (italics added by the authors to emphasise the relevant parts).

109

Leino, P. (2014) ‘Transparency, Participation and EU Institutional Practice: An inquiry into the Limits of the “Widest Possible”’, EUI Working

Paper, Law 3/2014.

110

European Parliament, Council and Commission (1999) ‘Joint declaration on practical arrangements for the new codecision procedure (Article 251

of the Treaty establishing the European Community)’, OJ C 148/01, 28 May, Brussels: European Parliament, Council and Commission.

111

European Parliament, Council, and Commission (2007) ‘Joint Declaration on Practical Arrangements for the Codecision Procedure (Art. 251 of the

EC Treaty)’, OJ C 145/05, 30 June, Brussels: European Parliament, Council and Commission.

Investigation of informal trilogue negotiations since the Lisbon Treaty

proposal. The letter must reflect the Council’s readiness to accept that outcome, should it be adopted by

Parliament’s plenary. A copy of the letter is also sent to the Commission.

In an early second reading agreement, the Commission’s role is to facilitate contacts between the co-

legislators with a view to reconcile their positions. If an agreement is found at this stage, the chair of the

responsible parliamentary committee sends a letter to the Coreper chair, where he/she indicates the

committee’s recommendations to the plenary to accept the Council’s common position without amendment,

on condition of confirmation of the common position by the Council and on legal-linguistic verification. A

copy of the letter will again be sent to the Commission.

In a ‘normal’ second reading agreement (which does not constitute an early agreement), the Coreper chair

sends a letter to the chair of the EP’s responsible committee, which indicates the substance of the agreement,

taking the form of amendments to the Council’s common position. That letter should show the Council's

willingness to accept that outcome, subject to legal-linguistic verification, should it be confirmed by

Parliament’s plenary. Again, a copy of this letter is forwarded to the Commission.

The 2007 declaration has been supplemented by various commitments in interinstitutional agreements. The

current Interinstitutional Agreement on Better Law-Making, adopted on 13 April 2016, highlights the

institutions’ commitment to ensuring greater transparency in the legislative process and contains a number of

commitments to achieving this goal. It particularly includes a commitment to ‘ensure the transparency of

legislative procedures, on the basis of relevant legislation and case-law, including an appropriate handling of

trilateral negotiations.’

112

Although the term ‘appropriate’ is not further specified and leaves room for

considerations of space to think, it is clear that transparency shall at least to some extent apply to trilogues.

The section on transparency also includes commitments to ‘improve communication to the public during the

whole legislative cycle’, to look at further ways to ‘facilitate the traceability of the various steps in the

legislative process’, 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’.

113

The conception phase on the joint database is still underway and it is too early to say what information will

be included in the final joint database. According to the EP’s most recent Activity Report on the Ordinary

Legislative Procedure

114

and our interviews, the joint database will, in first instance, focus on merging the

existing databases of the EU institutions into one database; i.e., EurLex, Legislative Observatory, and the

Council Register. The joint database will then provide a ‘one-stop-shop’ for files under the ordinary

legislative procedure that covers the whole legislative process from the Commission’s proposal to

publication in the Official Journal. It is not yet possible to say if information on trilogues will be included.

112 European Parliament, Council of the European Union, and Commission (2016) ‘Interinstitutional Agreement on Better Law-Making of 13 April

2016’, OJ L 123 12.05.2016, p. 7.

113

European Parliament, Council of the European Union, and Commission (2016) ‘Interinstitutional Agreement on Better Law-Making of 13 April

2016’, OJ L 123 12.05.2016, p. 7.

114

European Parliament (2017) Activity report on the ordinary legislative procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’, PE

595.931, Brussels: European Parliament, Conciliation and Codecision Unit.

Investigation of informal trilogue negotiations since the Lisbon Treaty

The EP, however, used its recent resolution on access to documents from 28 April 2016 to call for the

inclusion of trilogue dates and agendas in the joint database.

115

Turning to the practices of conducting informal trilogues, trilogues consist of two main layers: technical

meetings and political trilogues.

116

Additionally, there are the informal bilateral contacts between the EP and

the Council that do not constitute trilogue meetings, but are preparatory meetings to nurture good working

relations, which makes the ‘actual’ trilogues run more smoothly. Incoming presidencies, for instance, start to

make contact with MEPs and EP officials several months, or sometimes even a year, in advance of taking

over the presidency.

The number of trilogue meetings and the sequence between technical meetings and political trilogues varies

across dossiers. Trilogue negotiations usually start with an exploratory meeting at a purely technical level

(although they may also start with a political trilogue), where meeting dates are agreed, points for

agreements and disagreements between the co-legislators are identified, and agendas are drawn up. The

‘real’ negotiations are not supposed to take place during the technical meeting, although it may be difficult in

practice to clearly distinguish between matters that are technical and those that are political. Political issues

can be discussed in technical meetings, but usually only insofar as there is an agreement between the

institutions. Usually the purpose of the technical meetings is for the staff in all three institutions to compare

the EP’s and the Council’s respective mandates to identify the main differences on an article-by-article basis.

The reconciliation of differences then takes place in the political trilogues. Technical meetings often take

place again after each political trilogue meeting and are used as drafting sessions that put on paper the

compromises reached in the political trilogues (i.e., in the four-column document).

MEPs do not normally participate in technical meetings, but only in political trilogues. From the EP’s side,

technical meetings typically include representatives from the responsible committee secretariat, the legal

service, the Conciliation and Codecision Unit, officials from the political groups, and the assistants of the

rapporteur and shadow rapporteurs. From the Council’s side, the chair of the relevant working party (the

presidency), officials from the Council secretariat, and the legal service attend. The Commission is

represented by its legal service, the relevant desk officer in charge of the proposal, and sometimes the head

of unit. There is a consensual norm in the EU institutions that technical meetings do not involve high-ranking

officials or politicians, although exceptions can be found.

The political trilogues are the meetings where the ‘real’ negotiations take place and compromises on divisive

issues are struck. On behalf of the EP, the rapporteur and the shadow rapporteurs attend political trilogues,

and sometimes also the chair of the relevant committee. They are assisted by various support staff/officials,

such as the committee and political group secretariats, the legal service, and the Conciliation and Codecision

Unit. The whole EP team can easily include 25 people, if not more, although it is usually only the rapporteur,

the committee chair, and sometimes the shadows who take the floor on behalf of the Parliament.

From the Council’s side, officials from the presidency at the level of the working party and the Committee of

Permanent Representatives (Coreper) attend the political trilogues. Coreper I has historically covered most

115

European Parliament (2016) ‘European Parliament resolution of 28 April 2016 on public access to documents (Rule 116(17)) for the years 2014-

2015, 2015/2287(INI), Brussels: European Parliament.

116

Roederer-Rynning, C. and Greenwood, J. (2015) ‘The culture of trilogues’, Journal of European Public Policy 22 (8): 1148-65.

Investigation of informal trilogue negotiations since the Lisbon Treaty

legislative files and thus participated in most of the trilogue meetings. Since the Lisbon Treaty, Coreper I

deals with more legislative files in the areas of economic and financial affairs and justice and home affairs,

making Coreper II a more frequent participant in trilogue meetings. The Commission’s delegation in

political trilogues consists of high-level officials, such as the deputy directors general and heads of units,

joined by supporting staff.

Both technical and political trilogues can include 50 people or more, which means that they are not as small

as one might think when looking at them from the outside. Commissioners and ministers rarely attend

political trilogues, but they sometimes attend at the end of negotiations on politically salient dossiers when a

compromise is on the horizon. Both technical and political trilogues take place on the basis of specific

mandates from each institution and are recorded in the four-column document. Key MEPs (rapporteurs and

shadow rapporteurs) can share these documents with their respective political groups after each trilogue

meeting. The documents are also available to the public once legislation has been finalised, although it

requires consent from all three institutions, as it is a common inter-institutional document.

The interviewees from the EP and the Council highlighted a number of imbalances in trilogues, which they

thought worked to their disadvantage. Many interviewees from the EP found it difficult to closely follow the

discussions in the Council prior to commencing trilogue meetings, as it is not possible for EP representatives

to attend or receive information on Council meetings at preparatory and ministerial levels. Only summits of

the European Council are open to the EP’s president. On the contrary, the Commission is a key participant

and interlocutor in the Council’s preparatory meetings. Some EP interviewees perceived this to give the

Commission and the Council an informational advantage in trilogue meetings, particularly given that the EP

committee meetings are open to everyone and are usually attended by the Commission, the Council, and

representatives from the Council presidency. This was seen as giving the Council presidency an advantage at

the beginning of trilogue negotiations, as it is able to follow the development and definition of Parliament’s

negotiating mandate and the differing parliamentary views, whereas the EP cannot benefit from such

provisions. This is a cause of concern for Parliament, which feels that there is an information asymmetry

between the institutions.

117

Several interviewees from the Council secretariat and the permanent

representations mentioned that it was also challenging for the Council to follow what is going on in

Parliament as a number of key meetings take place behind closed doors, such as coordinators’ meetings,

meetings between the rapporteur and shadow rapporteurs on specific dossiers, and political group meetings

(except the case of the Greens).

Scholars have assessed whether the EP committee meetings could be used to engender more open discussion

between the EU institutions to lift the lid of some of the discussions taking place in trilogue meetings.

118

Ever since early conclusion became possible, Council officials have followed all of the legislative

discussions in the EP closely, especially to gauge the possibility of early agreement and provide the Council

presidency with up-to-date information on all ordinary legislative files in Parliament. In practice, there are

limits to what a Council representative can say in a public committee meeting before the Council has decided

upon its negotiating mandate. If parliamentary committee meetings were used for open debate between the

co-legislators, it would require the Council to be ready to answer questions about the state of play in the

117

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014 (7th parliamentary term)’, DV\1031024,

Brussels: European Parliament, Conciliation and Codecision Unit.

118

Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation and change’, Journal of European

Public Policy 10 (2): 180.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Council. This would risk revealing possible disagreements between member states, which could potentially

weaken the Council’s hand in the negotiations with the EP. Trilogues give the Council a more comfortable

format for negotiating with a limited number of MEPs.

119

From the vantage point of the Council, the EP has certain advantages in trilogue meetings, simply because

most trilogues are held in the EP’s premises, primarily for logistical reasons. This gives the EP the

psychological advantage of ‘playing on their home ground’. Interviewees from the Council also noted that

the ‘layout’ of some of the EP’s rooms in which trilogues take place create a form of asymmetry between the

co-legislators because Parliament’s key negotiators often sit on a podium looking down on the rest. Some

Council officials believe that the physical space makes a difference. But this was found not to reflect the

otherwise strong sense of equality in trilogue negotiations. The Council hopes that with the opening of the

new Council building (the Europa building), more trilogues will take place on the Council’s premises.

Another advantage for the EP, as noted by Council interviewees, is that the workload of trilogue meetings is

more evenly spread out in the Parliament. In the Council, the presidency is in charge of representing the

Council during trilogues, whereas the workload in the Parliament is divided up between different MEPs.

Different EP negotiations teams include different MEPs, as the role of the rapporteur and shadow rapporteurs

alternates. The Council has, ever since troika presidencies were introduced with the 2009 Lisbon Treaty,

become particularly aware of securing a smooth transition from one presidency to the next. In practice, the

incoming presidency usually joins the current presidency in trilogue meetings both at a technical and

political level, mainly as an observer, to see how the negotiations are conducted, and to familiarise itself with

files it may need to take over. According to an official from the Council secretariat, however, this should not

happen too early as it may (wrongly) suggest to the EP that a file is continuing into the next presidency and

not be finalised during the current presidency.

Despite the EP and the Council both suggesting that the other institution has certain advantages in trilogues,

all interviewees supported the use of trilogues and could not imagine finalising legislation without trilogues.

Generally, the interviewees did not question whether trilogue meetings should be held or not; they were

simply regarded as ‘the way we work’.

4.3 Intra-institutional rules and practices

A long-standing concern about trilogues is that they may run the risk of short-circuiting debates inside the

institutions (such as parliamentary committees) and elevate negotiating teams to a kind of élite among their

peers with privileged access to the information given during trilogues.

120

Lack of transparency in

negotiations may indeed be problematic when decisions (1) are de facto taken in an inaccessible and

undocumented arena; (2) have repercussions for effective institutional oversight; (3) differentiate access to,

and control over, de facto decision-making and access to information. From an intra-institutional point of

view, it is important to pay attention to the representative relationship between the trilogue negotiations

teams and their respective institutions to assess the degree of leeway given to the negotiating teams vis-à-vis

119

Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation and change’, Journal of European

Public Policy 10 (2): 171-188.

120

Reh, C. (2014) ‘Is informal politics undemocratic? Trilogues, early agreements and the selection model of representation’, Journal of European

Public Policy 21 (6): 822-41.

Investigation of informal trilogue negotiations since the Lisbon Treaty

their colleagues in their respective institutions.

121

This includes asking questions such as: (1) How is the

mandate given to the negotiating team? (2) Who is represented in trilogue negotiations? (3) What reporting

requirements are in place for the negotiating team? Table 2 provides an overview of the relationship between

the EP’s and the Council’s negotiating teams and their respective institutions. The rest of the chapter

elaborates on these mechanisms by focusing on the rules and/or practices in place for trilogues in the

Commission, the EP, and the Council.

Table 2: Current rules and practices governing trilogues within the Parliament and the Council

EP Council

How is the composition of the

negotiating team for political

trilogues determined?

In the EP’s Rules of Procedure

(Rule 69f), consisting of the

committee chair, the

rapporteur, and the shadow

rapporteurs

By practice: the presidency

represents the Council in political

trilogues

What forms the basis of the

initial negotiating mandate?

The committee report, as

endorsed by plenary

The Coreper mandate or a

General Approach

Can the mandate be changed

once trilogue meetings have

started?

Yes, but it will need the

approval of the committee

responsible

Yes, but it requires a supporting

majority in Coreper

What is the negotiating team

required to report to the rest of

their institution?

The chair and/or rapporteur

must report back to the

committee responsible after

each trilogue meeting. Where

it is not possible to arrange a

committee meeting, the chair

of the negotiating team and the

rapporteur must report back to

a meeting of the committee

coordinators.

The rapporteur and the shadow

rapporteurs each report back to

their political group in the EP.

The presidency reports back to

Coreper and the relevant working

party after each trilogue meeting

Which trilogue documents are

made available to the

respective institutions?

The updated four-column

document is shared with the

shadow rapporteurs and the

rapporteur after each trilogue

All national

delegations/permanent

representation gain access to the

updated four-column document

after each trilogue meeting

121

Reh, C. (2014) ‘Is informal politics undemocratic? Trilogues, early agreements and the selection model of representation’, Journal of European

Public Policy 21 (6): 822-41.

Investigation of informal trilogue negotiations since the Lisbon Treaty

4.3.1 Rules and practices in the Commission

The formal role of the Commission in trilogue meetings is to act as a compromise facilitator between the co-

legislators and to take up the role of an honest broker. As outlined in the Interinstitutional Agreement, ‘the

Commission should carry out its role as facilitator by treating the two branches of the legislative authority

equally, in full respect of the roles assigned by the Treaties to the three institutions’.

122

The interviewees generally considered the Commission to be in a good position to suggest compromises

given its technical expertise. Many regarded the Commission as a mediator and facilitator, even when a

compromise went against its position. However, not everyone agreed that the Commission plays the role as

an honest broker, as the Commission is sometimes seen as advancing its own agenda.

According to our interviewees, the Commission does not usually oppose a compromise between the

legislators in cases where it goes against the Commission’s own position, even when it considers it to be a

bad compromise. Instead, it would issue a statement expressing its discontent and the possible court action

that might follow if the compromise is thought to be inconsistent with Union law.

The Commission usually changes its initial position (its formal proposal) throughout the trilogue

negotiations to reflect the compromises found. When the Commission negotiates with the Council and/or the

EP, it is done on the basis of a mandate given by the College of Commissioners, using the empowerment

procedure. The ‘Groupe des Relations Interinstitutionelles’, GRI, (a permanent group of cabinet members

responsible of inter-institutional relations) and Hebdo (the weekly chef-de-cabinet meetings) draw up the

Commission’s mandates or revisions of mandates. The College of Commissioners then approves the mandate

or change thereof by means of the so-called ‘GRI Fiches’ (information sheets) without discussion. GRI

Fiches provide an information update on trilogue meetings in order to be able to agree on a Joint Statement

on a new consensus. The GRI Fiches are usually not available to the public, so it can sometimes be difficult

for outsiders to see how and when the Commission’s changes its initial position.

123

4.3.2 Rules and practices in the European Parliament

The increased use of early agreements has transformed the EP’s working methods. Parliament has adapted to

the changing legislative environment and tried to address the criticism levelled against it regarding

transparency and accountability of the EP’s negotiating teams during trilogue negotiations. Much has

happened since the early days of co-decision and the use of early agreements. Co-decision had barely been

invented before the first calls for reforms to the EP’s internal rules surfaced. Research shows that revisions

of the EP’s internal rules regarding trilogues have been greatly facilitated by an overall reform agenda in the

EU to make decision-making more transparent.

124

This means that the impetus for parliamentary reforms of

trilogues is to be seen as part of a broader reform agenda.

122

European Parliament, Council of the European Union, and Commission (2016) ‘Interinstitutional Agreement on Better Law-Making of 13 April

2016’, OJ L 123 12.05.2016, p. 0001, para. 32.

123

Szaprio, M. (2013) The European Commission: A Practical Guide, London: John Harper Publishing.

124

Héritier, A. and Reh, C. (2012) ’Codecision and its Discontents: Intra-Organisational Politics and Institutional Reform in the European

Parliament’, West European Politics, 35 (5): 1134-1157.

Investigation of informal trilogue negotiations since the Lisbon Treaty

A first attempt to regulate the EP’s internal rules on trilogues was made in 2004 when the EP adopted

‘Guidelines for First and Second Reading Agreements’.

125

The guidelines recommended (but did not require)

that trilogues should not begin before the responsible committee had adopted its amendments, which would

then constitute the mandate for the EP’s representatives in trilogues. The guidelines also stated that the

coordinators were to decide who, other than the rapporteur, should be present on behalf of Parliament in

trilogues, and it required the rapporteur to report back regularly on the progress made in trilogue meetings.

Despite the EP’s good intentions, the guidelines were non-binding and full of words such as ‘should’, ‘can’

and ‘be encouraged’. They did little to address the different practices across committees and to properly hold

the rapporteur accountable to the rest of the EP, such as committees, small political groups, and rank-and-file

MEPs.

126

A few years after the introduction of the guidelines, early agreements started to be criticised in the outside

world and concerns were raised from a wider circle of parliamentary actors, partially because the use of early

agreements had increased significantly. The large room for manoeuvre given to rapporteurs and the limited

role of Parliament in trilogues raised concerns about the openness and inclusiveness of early agreements.

127

However, it was not before 2007 that Parliament picked up the baton and started looking at how its internal

rules governing trilogues could be further improved.

In 2007, the EP’s Conference of Presidents (comprising the president and the leaders of the political groups)

set up a Working Party on Parliamentary Reform explicitly tasked with examining the practice of trilogues

and suggesting improvements. The efforts of the working party and the Parliament’s Conciliation and

Codecision Unit resulted in the adoption on a Code of Conduct in 2008, setting out more detailed rules inside

the EP on trilogue meetings. In 2009, the Code was annexed to the EP’s Rules of Procedure, making it more

binding and visible. Overall, the Code introduced five changes

128

to improve the EP’s rules governing

trilogues:

1. Granting committees new powers: Committees are now the main responsible body during trilogue

negotiations and the responsible committee decides on a case-by-case basis whether an early

agreement should be sought;

2. Broadening access to trilogues: The committee (rather than the coordinators) decide on

parliamentary representation in trilogue;

3. Introducing a clear mandate: The responsible committee or the plenary’s amendments form the

basis of Parliament’s negotiating mandate in trilogues. Committees can under exceptional

circumstances start negotiations before amendments have been agreed, but it requires the committee

to provide guidance;

125

See Annex B ‘Guidelines of First and Second Reading Agreements under the Codecision Procedure’ in: European Parliament (2006) ‘Conciliation

and Codecision Activity Report July 2004 – December 2006 (6

parliamentary term. first half term), DV/651053, Brussels: Delegations to the

Conciliation Committee.

126

Huber, K. and Shackleton, M. (2013) ‘Codecision: A practitioner’s view from inside the Parliament’, Journal of European Public Policy 20 (7):

1040-55; Judge, D. and Earnshaw, D. (2011) ‘“Relais actors” and co-decision first reading agreements in the European Parliament: The case of the

advanced therapies regulation,’ Journal of European Public Policy 18 (1): 53-71.

127

Héritier, A. and Reh, C. (2012) ’Codecision and its Discontents: Intra-Organisational Politics and Institutional Reform in the European

Parliament’, West European Politics, 35 (5): 1146.

128

See Annex 2 of: European Parliament (2009) ‘Activity Report 1 May 2004 to 13 July 2009, (6th parliamentary term)’, PE427.162v01-00, Brussels:

European Parliament, Delegations to the Conciliation Committee.

Investigation of informal trilogue negotiations since the Lisbon Treaty

4. Increasing access to information: Documents used in trilogues should be made available to the

committee;

5. Introducing a ‘cooling-off’ period: A cooling-off period of at least one month between the vote on

any legislative report in committee in first reading and the vote in plenary was introduced.

Despite introducing more binding and specific rules on trilogues, the Code did little to stem the tide of

criticism of Parliament’s handling of trilogues.

129

In 2012, a new wave of reform arose as key provisions of

the Code were incorporated into the EP’s Rules of Procedure and further revisions were made, most notably

by:

1. Strengthening the role of the responsible committee(s) in defining the mandate: The opening of

trilogue negotiations is conditioned on the adoption of a legislative report in committee, constituting

the mandate for negotiations. In exceptional circumstances, negotiations can start before the

adoption of a committee report in committee, in which case the mandate may consist of a set of

amendments or a set of clearly defined objectives, priorities or orientations. Negotiations without a

committee report do need the approval of plenary for consideration with a debate and vote and the

possibility to table amendments (rule 74 in the previous rules of procedure);

130

2. Pluralising the EP’s representation in trilogues (the EP’s negotiating team): The negotiating team

shall be led by the rapporteur and presided over by the chair of the committee responsible or by a

vice-chair designated by the chair. It must comprise at least the shadow rapporteurs from each

political group. The inclusion of shadow rapporteurs in the team was, among other things, a

response to concerns about the considerable discretion that the rapporteur enjoyed in the early years.

Particularly in cases where the rapporteur came from a smaller political group, the absence of

members of other groups from the negotiations raised questions of representation;

3. Improving rules on reporting back requirements and renewal of mandates: The EP’s negotiating

team is obliged to report back to the committee responsible after each trilogue meeting to inform

members of the progress made in trilogues and to renew the mandate if necessary. If it is not

possible to organise a committee meeting in a timely fashion, the negotiating team must report back

to the committee chair, the shadow rapporteurs, and the coordinators.

The changes to the EP’s Rules of Procedure in 2012 bear witness to long-term pressure to regulate trilogues

inside the Parliament to make negotiations more transparent and inclusive. However, the pressure to

strengthen the rules further did not stop here. Four years later, in December 2016, the EP approved a new

round of revisions of its Rules of Procedure to reinforce the transparency and efficiency of trilogues. The

changes came into effect in January 2017 and made four main changes:

1. The negotiating team: The Parliament’s negotiating team must be led by the rapporteur and be

presided over by the chair of the committee responsible or by a vice-chair designated by the chair. It

must comprise at least the shadow rapporteurs from each political group that wishes to participate

(before the rules did not state that only shadow rapporteurs that wishes to participate can do so);

2. Ruling out the possibility to start negotiations without a committee report: The previous Rule 74 –

which enabled a committee under exceptional circumstances to adopt its negotiating mandate prior

129

Roederer-Rynning, C. and Greenwood, J. (2015) ‘The culture of trilogues’, Journal of European Public Policy 22 (8): 1148-65.

130

European Parliament (2014) ‘Rules of Procedure of the European Parliament (8

parliamentary term)’, Brussels: European Parliament.

Investigation of informal trilogue negotiations since the Lisbon Treaty

to the adoption of a committee report – is deleted. The overwhelming feeling in the EP is that using

the committee report as the basis for the mandate is more transparent and reflects established

parliamentary practice;

131

3. Including plenary in approving the negotiating mandate: Any negotiations between the EP and the

Council with the intention to strike a first-reading agreement will need a mandate from plenary. All

committee decisions to start trilogue negotiations (by a majority of its members) need to be

announced in plenary. If political groups or MEPs who together make up at least one-tenth of

Parliament’s members, object to commence negotiations, a confirmation vote in plenary is required

in the same part-session. If a majority in plenary do not confirm the committee’s wish to start

negotiations, the draft act and the report of the committee responsible shall be placed on the agenda

of the next plenary part-sessions, with a deadline for amendments;

4. Clarification of requirements for reporting back: As in the previous rules, the negotiating team must

report back after each trilogue meeting. Where it is not possible to arrange a committee meeting, the

chair of the negotiating team and the rapporteur must, on behalf of the negotiating team, report back

to a meeting of the committee coordinators.

The new provisions set a clear path for the adoption of the EP’s negotiating mandate and provide for a

‘check’ by plenary. All committee decisions to start negotiations at first reading need to be announced on the

first day of the following plenary session in Strasbourg or in the mini-plenary sessions in Brussels

(depending on which of the two takes place first) following the committee vote, and they can be challenged

by plenary. If the committee report is not challenged, negotiations with the Council and the Commission can

begin. If the report is challenged and the plenary adopts amendments to the Commission’s proposal in the

subsequent part-session, the plenary can either decide to refer the file back to the committee responsible and

give the green light to start trilogue negotiations on the basis of a plenary mandate, or decide to finalise its

first reading.

If the EU institutions aim for an early second reading, the revised rules state that Parliament’s negotiating

team can only enter into negotiations on the basis of a decision made by (a majority of the members of) the

responsible committee and after it has been announced in plenary. In this case, Parliament’s first reading

position constitute Parliament’s mandate.

At (‘normal’) second reading, Parliament’s first reading position still constitutes the mandate, but the

committee responsible can endorse negotiating guidelines for its negotiating team on any new elements

included in the Council’s first reading position, without having to consult plenary. This flexibility is given to

allow the EP to respond to the Council within the second reading deadlines.

132

Table 3 presents an overview

of the key changes introduced to the EP’s internal rules in the different rounds of rule reforms.

According to our EP interviewees, the recent revision of the Rules of Procedure responds to calls from both

inside and outside the EP for making Parliament’s preparation of trilogue negotiations and reporting back

more inclusive. The revised rules seek to ensure that Parliament’s mandate reflects the view of the entire

house and not just a committee majority. The rules address the concern that some committees may be biased

131

European Parliament (2017) Activity report on the ordinary legislative procedure: 4 July 2014 to 31 December 2016 (8th parliamentary term)’, PE

595.931, Brussels: European Parliament, Conciliation and Codecision Unit: 21-22.

132

European Parliament (2017) ‘Rules of Procedure, Provisional Edition, January 2017’, P8_TA-PROV (2016) 0484, Brussels: European Parliament.

Investigation of informal trilogue negotiations since the Lisbon Treaty

towards their own policy area and may not reflect the majority view in Parliament, as has indeed been found

in academic research.

133

Including the plenary is, therefore, seen as providing a more balanced and

representative mandate, which also has the advantage of strengthening the EP’s position vis-à-vis the

Council in the trilogue negotiations.

The strengthened role of plenary in confirming or changing the mandate of the committee responsible

reflects a general working practice across many committees (such as ECON, ENVI, PECH), although this

has far from been used on all files in the past. In the case of PECH, a plenary mandate was previously sought

on divisive files where there was a narrow majority in committee for a report – involving the plenary ensured

that the EP’s negotiating team had support from the full house and not just a slim committee majority.

Table 3: Key changes to the EP’s formal rules

Year of change How is the composition

of the negotiating team

for political trilogues

determined?

What forms the basis of the

initial negotiating

mandate?

What is the negotiating

team required to report to

the rest of their institution?

2007 The committee decides

on parliamentary

representation in

trilogues

The responsible

committee’s or the

plenary’s amendments form

the basis of Parliament’s

negotiating mandate in

trilogues. Committees can

under exceptional

circumstances start

negotiations before

amendments have been

agreed, but it requires the

committee to provide

guidance

Documents used in trilogues

should be made available to

the committee

2012 The negotiating team

must be led by the

rapporteur and presided

over by the chair of the

committee responsible or

by a vice-chair

designated by the chair.

It must comprise at least

the shadow rapporteurs

from each political group

The committee report

In exceptional

circumstances, negotiations

can start before the

adoption of a committee

report in committee, in

which case the mandate

may consist of a set of

amendments or a set of

clearly defined objectives,

priorities or orientations,

which needs the approval of

The EP’s negotiating team is

obliged to report back to the

committee responsible after

each trilogue meeting

If it is not possible to

organise a committee

meeting in a timely fashion,

the negotiating team must

report back to the committee

chair, the shadow

rapporteurs, and the

coordinators

133

See for instance Dionigi, M.K. (2017) Lobbying in the European Parliament: The Battle for Influence, London: Palgrave Macmillan, pp. 156-160;

McElroy, G. (2006) ‘Committee Representation in the European Parliament’, European Union Politics, 7(1): 5-29.

Investigation of informal trilogue negotiations since the Lisbon Treaty

plenary

2016 In the EP’s Rules of

Procedure (Rule 69f),

consisting of the

committee chair, the

rapporteur, and the

shadow rapporteurs

The committee report, as

endorsed by plenary

The chair and/or rapporteur

report back to the committee

responsible after each

trilogue meeting. Where it is

not possible to arrange a

committee meeting, the

Chair of the negotiating

team and the rapporteur

must report back to a

meeting of the committee

coordinators

The revisions of the EP’s rules of procedures in 2012 and 2016 indicate a conscious effort to streamline the

way in which trilogues are conducted across parliamentary committees. This has, to some extent, resulted in

a common culture of trilogues, although different committees still operate according to their own compass.

In their academic study of trilogues cultures in the EP in the seventh parliamentary term (2009-2014),

Roederer-Rynning and Greenwood find that a dominant parliamentary approach to trilogues has become

entrenched in the EP, which is characterised by two features: (1) the centrality of the chair as an integral part

of the EP negotiating team; and (2) the sharp division of labour between technical and political trilogues.

134

Within this generic ideal-type, they find three approaches taken to trilogues according to the role played by

the committee chair:

1. A gladiatorial approach: The committee chair plays a central role in political trilogue meetings and

acts as one of the key negotiators. The chair insists on being present in all political trilogue meetings

and to have a clear separation and hierarchy between technical meetings and political trilogues, with

the latter seen as constituting the ‘real’ venue for inter-institutional negotiations. The chair tries to

control the flow of information, the bilateral contacts between the EP and the Council presidency,

and demands high-level Council representation in political trilogues. The chair makes sure that

MEPs in the negotiating team toe the parliamentary mandate and do not seek to advance their own

political interests. The Committee on Economic and Monetary Affairs (ECON) committee is an

archetypical example of a gladiatorial approach; this approach was particularly developed under the

previous chairmanships of the French S&D MEP Pervenche Berès in the sixth parliamentary term

(2004-2009) and the British ALDE MEP Sharon Bowles in the seventh term (2009-2014), and seems

still to be the case in the current Parliament (2009-2014);

2. A problem-solving approach: The committee chair also plays an important role as a negotiating

chair, but this style is kept more neutral compared with the gladiatorial approach. The committee

chair does not insist on high-level Council representation. The problem-solving approach represents

a mix of committees, both those who have gained new co-decision powers before and with the

Lisbon Treaty, such as the Committees of Internal Market and Consumer Protection (IMCO),

Regional Development (REGI), International Trade (INTA), and to some extent Civil Liberties,

Justice and Home Affairs (LIBE);

134

Roederer-Rynning, C. and Greenwood, J. (2016) ‘The European Parliament as a developing legislature: coming of age in trilogues’, Journal of

European Public Policy, 24 (5): 735-754.

Investigation of informal trilogue negotiations since the Lisbon Treaty

3. An arm’s length approach: The chair plays a low-key role compared with the two other approaches

and does not frequently participate in political trilogue meetings. Committees characterised by an

arm’s length approach do not necessarily seek an early first reading agreement, and rely more often

on early or late second reading agreements. The Committee of Transport and Tourism (TRAN) is an

example of this approach (at least in the seventh parliamentary term). The TRAN committee chair in

the seventh parliamentary term (2009-2014), the British S&D MEP Brian Simpson, typically

attended political trilogues at the final stages to send the message that things were getting serious

when the chair attends.

135

Roederer-Rynning and Greenwood also find that a number of committees take a more mixed approach, at

least in the time period they study (2009-2014). The Committee on Agriculture and Rural Development

(AGRI) dealt with a comprehensive reform of the common agriculture policy in the seventh parliamentary

term, which involved over 40 trilogue meetings. Due to the high number of these meetings, the chair

delegated the chairing role to the vice-chair for several of the trilogues. In the Budgetary Control Committee

(CONT), the distinction between technical and political trilogues was less clear as the chair participated in

both technical and political trilogue meetings together with the rapporteur and shadow. The chair assumed

the role as a sort of micro-managing negotiator and was more willing to delegate responsibility to the

committee staff.

136

In our interviews with committee officials, we find a similar situation in the current (eighth) parliamentary

term (2014-2019). Our interviews show that committee chairs usually take a steering role during the trilogue

meetings to ensure that the committee mandate is followed. Chairs also often take a horizontal approach as

they have the advantage of having an overview of other trilogue meetings within their committee. This

means that chairs often deal with issues of delegated and implementing acts (‘horizontal issues’) by making

sure that they tally with what has been negotiated on other files, and that they are not traded off for more

influence on the substance of legislation. The issue of delegated and implementing acts is often dealt with as

separate items.

Not all committee chairs are present in all the political trilogue meetings on a dossier (which is the case for

the legal affairs committee); some delegate the chairing responsibility to the rapporteur. This often reflects a

high workload in committees. The ECON committee is an exception in this regard, as the chair is, in

principle, present in all political trilogues. For other committees, the chair would open the first political

trilogue meeting and be present again in the final trilogue.

The chair and the rapporteur have the responsibility to report back to their committee after each political

trilogue. This usually happens at the beginning of the subsequent committee meeting and appears on the

agenda as one of the first items, where committee members are shortly briefed on the progress made.

However, these debriefings are often very general in nature, short (about ten minutes per dossier), and are

not subject to further debate or questioning.

135

Roederer-Rynning, C. and Greenwood, J. (2016) ‘The European Parliament as a developing legislature: coming of age in trilogues’, Journal of

European Public Policy, 24 (5): 735-754.

136

Roederer-Rynning, C. and Greenwood, J. (2016) ‘The European Parliament as a developing legislature: coming of age in trilogues’, Journal of

European Public Policy, 24 (5): 735-754.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Other than the change to the EP’s Rules of Procedure, new internal practices have also recently emerged to

help the rapporteur seek advice from the Parliament’s officials throughout the policy process. On each

dossier, the rapporteur is provided with a list of parliamentary officials, called a project team, which are

available to the rapporteur for advice. The EP has done much over the years both to boost its internal

resources and to make the parliamentary conduct of trilogue meetings more transparent and inclusive to the

extent that the rules on trilogues have become institutionalised and moved from being highly informal to

becoming more formalised (cf. Chapter 2).

4.3.3 Rules and practices in the Council

Unlike the EP, the Council does not state any rules/practices governing trilogues in its rules of procedure.

None of our interviewees were able to point to any internal documents outlining the procedures to be

followed regarding how the mandate is given to the presidency and how the presidency reports back to the

other national delegations after each trilogue meeting. Instead, a number of well-established practices and

working procedures exist that seem to be followed by most, if not all, presidencies and are passed over from

one presidency to another through extensive training of the incoming presidency by the Council secretariat

and by close cooperation between current and incoming presidencies. Our interviewees from the Council

secretariat and the permanent representations thought that the lack of documented rules provided the

presidency with much-needed flexibility.

In practice, the Council’s mandate is either adopted at the level of Coreper or as a General Approach at the

ministerial level. The presidency decides on whether a General Approach is sought; practices on this issue

vary across policy areas more than across presidencies. The Council’s initial mandate is public when the

presidency decides to enter into trilogue meetings on the basis of a General Approach. By contrast, the EP’s

initial mandate is always known. According to our interviewees, a General Approach is often used as the

basis of a mandate when the Council’s schedule is ahead of the EP’s, which gives the Council enough time

to have a negotiating mandate endorsed at the ministerial level. A General Approach is usually sought in

justice and home affairs and in economic and financial affairs – policy areas that have more recently become

subject to the ordinary legislative procedure and are often politically salient. The interviewees saw the use of

general approaches from the Council as having three clear advantages. First, it was perceived to put the

presidency in a stronger bargaining position vis-à-vis the EP in trilogues, as it sends a strong signal that there

is political backing for the mandate at the highest level in the Council. Second, it allows ministers to

familiarise themselves with files earlier in the process and give them a sense of ownership and involvement,

rather than simply being presented with a compromise at the end of trilogues to rubberstamp what has been

agreed at lower levels. Third, for some countries, ministerial involvement also allows for greater

involvement of national parliaments, particularly for countries operating on a mandate-based EU scrutiny

system (such as Austria, Denmark, Latvia, Lithuania, Slovakia, Slovenia, and Sweden

137

) where the

government is required to obtain a mandate from national parliamentary committees before important

deliberations in the Council.

137

Neuhold, C., Rozenberg, O., Smith, J., Hefftler, C. (eds.) (2015) Palgrave Handbook of National Parliaments and the European Union, London:

Macmillan.

Investigation of informal trilogue negotiations since the Lisbon Treaty

A presidency opens trilogue negotiations when it has a substantive majority in the Council, either at Coreper

or ministerial level. A majority might materialise at working group/party level, but is always confirmed by

Coreper before commencing inter-institutional negotiations. Coreper (and neither working parties nor

ministers) give renewed or revised mandates to the presidency if needed between trilogue meetings.

All the interviewed national delegations agreed that there is a high level of trust between the national

delegations and the presidency, and that it is difficult for the presidency to diverge from the mandate in

practice. It was also pointed out that it is difficult for the presidency to do this because other national

delegations would quickly find out through their informal contacts with people present in the trilogue

meetings. Presidencies are bound by their mandate and are acutely aware of it, as they have a strong interest

in being seen as a successful and trustworthy presidency. The interviewees could only remember a few

exceptional cases in which a presidency had not followed the mandate or reported back to the rest of the

Council in a proper manner. But these exceptional instances had a lasting impression on the interviewees and

they clearly remembered, even though some of the examples dated back in time, which presidencies on

which cases did not follow established Council norms. The instances mentioned did not necessarily reflect

bad practice of the civil servants involved from the presidency, but indicated immense pressure from the

political level (the relevant sectoral minister from the presidency). Some interviewees found a key difference

between presidencies that are tightly ‘controlled’ by their ministers and those who have more freedom to

shape the national position – trust was seen to be greater among presidencies that were less controlled by

their political capitals. Several Council interviewees also stated that if a Council mandate went against a key

national interest of the presidency, they would often try to slow down the process and let the next presidency

take over.

According to scholarly literature, the process of giving the presidency a mandate has changed over time,

particularly since the Lisbon Treaty entered into force.

138

In the past, the presidency would circulate a draft

proposal for a mandate to all national delegations, which allowed the delegations to add footnotes to the text,

where they could indicate any national issues to be taken into consideration. All national delegations had

access to each other’s footnotes. This procedure has changed after the Lisbon Treaty. Today, the presidency,

with support from the Council’s secretariat, is in close bilateral contact with national delegations. The

presidency still sends draft proposals to the national delegations, but feedback takes place bilaterally and is

not necessarily shared with all delegations.

Once trilogues have started, the presidency reports back formally on the progress made in trilogues to the

rest of the Council both in writing, through distribution of the updated four-column document, and orally at

the following working party and Coreper meetings. A renewed mandate is sought, if necessary, at the level

of Coreper and the meeting dates of the next trilogue meetings are announced. If national delegations want

an account of the progress made in a trilogue meeting before the official reporting back, they would seek

‘live’ updates through informal contact with the presidency or other actors present in the relevant trilogue

meeting, such as MEPs. All compromises agreed on in trilogues need to be adopted by both Coreper and the

Ministers.

138

Roederer-Rynning, C. and Greenwood, J. (2015) ‘The culture of trilogues’, Journal of European Public Policy 22 (8): 1148-65.

Investigation of informal trilogue negotiations since the Lisbon Treaty

There have recently been discussions in the Council on how to increase transparency in its internal decision-

making. As it stands now, the Council’s Rules of Procedure focus on transparency at the final stage of

Council decision-making at the ministerial level when it is acting in its legislative capacity, while the formal

transparency requirements for the preparatory bodies (working groups/committees and Coreper) are less

extensive. It is specifically laid down in the Treaties that ‘the Council [Ministers] shall meet in public when

it deliberates and votes on a draft legislative act’ (Article 16(8) TFEU). Legislative deliberations, public

deliberations and public debates during Council meetings (ministers) are transmitted audio-visually (in an

overflow room) and broadcasted in all official languages of the EU institutions using video streaming. The

outcome of voting, where applicable, is indicated visually. This means that when a mandate is given at the

ministerial approach, it is publicly known.

Following the conclusion of the Interinstitutional Agreement on Better Law-Making, the General Affairs

Council of 24 June 2016 had an exchange of views on how the legislative process could be made more

transparent and understandable to the public. The basis for the exchange of views was a note prepared by the

then Dutch presidency concerning the implementation of the Interinstitutional Agreement.

139

The Dutch

presidency suggested increasing the transparency of the Council’s negotiating mandate and pursuing a more

proactive communication of the results of trilogue negotiations. Regarding the transparency of the Council’s

negotiating mandate, the note from the Dutch presidency stated that communication to the public and the

media about ordinary legislative files under negotiating is a particular challenge for the Council compared

with the EP, which constantly communicates to media and the public during all stages of the legislative

proposal and not only when files are finally adopted.

The Dutch presidency’s note suggested that the Council should improve its communication to the public

whenever ‘intermediate’ stages are successfully concluded, such as when adopting a negotiation mandate,

when the Presidency reaches a deal with the EP, or when Coreper approves the outcome of the trilogue

negotiations. The Dutch presidency’s note shows support for the Interinstitutional Agreement’s commitment

to make joint press conferences after the final trilogue meeting when a compromise is reached. Although the

commitment to make joint announcement of successful outcomes of the legislative process is enshrined in

the current and previous Interinstitutional Agreement on Better Law-Making, and is also mentioned in the

Joint Declaration on practical arrangements for the co-decision procedure

140

, joint press releases or press

conferences are uncommon and primarily occur on the politically most high profile legislative files, such as

the European Fund for Strategic Investments (EFSI). One of the main reasons is that they are logistically

challenging. The Dutch presidency encouraged the Council and the EP to send out joint press releases and

hold joint press conferences more consistently and immediately after final trilogues to provide the public

with a more balanced picture of the progress made in the negotiations.

According to our interviewees, there is no cohesive view among national delegations as to when Coreper

mandates and minutes in general should be made public. Some interviewees found that the member states’

positions on transparency are rather ‘black and white’, divided between those that are for more transparency

(here the following countries were mentioned: Denmark, Estonia, Finland, the Netherlands, Slovenia, and

139

Council of the European Union (2016) ‘Note from the Presidency to the Committee of the Permanent Representatives/Council on the

Implementation of the InterInstitutional Agreement on Better Law-Making – Improving transparency’, Brussels, June 2016, available at:

http://data.consilium.europa.eu/doc/document/ST-10120-2016-REV-1/en/pdf (accessed 22 April 2017).

140

European Parliament, Council, and Commission (2007) ‘Joint Declaration on Practical Arrangements for the Codecision Procedure (Art. 251 of

the EC Treaty)’, OJ C 145/05, 30.6, point 45.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Sweden) and those that are more hesitant or against, which makes it difficult to make changes to the exiting

practices and rules.

Some of the member states we interviewed noted that it could be useful to have some (loose) written

procedures for the Council’s trilogue meeting conduct, for example as guidelines annexed to the Council’s

Rules of Procedure. However, a point was made of not making strict rules, but allowing for flexibility. The

Council interviewees generally did not see that there was anything wrong in having a ‘diplomatic tradition’

in the Council, as this was seen to engender trust between the member states. Some member states were of

the opinion that more transparency may lead to stricter behaviour by national delegations and reduce

flexibility.

In sum, this chapter has shown that both EU law and case-law reflect tensions in trying to achieve efficiency

and transparency in decision-making, as highlighted by the pending De Capitani case. This tension is

particularly visible concerning legislation that is under negotiation. Diverging views inside the EU

institutions exist on how much information should be made publicly available during negotiations.

The chapter has also shown that although trilogues are not mentioned in the EU Treaties, they do not operate

in a rule-free environment. For instance, the EP has used its own Rules of Procedure to make detailed

transparency and accountability requirements for the EP’s negotiating team in terms of the composition of

the team, the negotiating mandate, and the reporting back requirements from the trilogue negotiating team to

the rest of Parliament. Over time, the EP has done a lot to close the democratic deficit inside Parliament, i.e.

between the EP’s negotiating teams and the whole Parliament. Despite the efforts made to improve

transparency and accountability of trilogue meetings, a range of concerns still remain in the public domain,

as demonstrated in Chapter 5.

Investigation of informal trilogue negotiations since the Lisbon Treaty

5. Concerns about early agreements raised in the public domain

Early agreements have been subject to concerns about transparency and accountability for as long as they

have existed.

141

This chapter is based on an analysis of all COSAC’s bi-annual reports on EU practices and

procedures (a total of 19 reports) since the Lisbon Treaty entered into force, and the 50 submissions to the

public consultation of the European Ombudsman’s as part of her investigation of 2016 into the transparency

of trilogue meetings. We also analysed the Brussels online media’s (i.e., EurActiv, EUobserver, and Politico)

coverage of trilogue meetings since the entering into force of the Lisbon Treaty (a total of 303 articles).

When examining the coverage of trilogues in the Brussels media, the bi-annual reports produced by the

Conference of Parliamentary Committees for Union Affairs (COSAC), and a number of statements made by

interest groups, we can identify the following four overall concerns and perceptions about informal trilogues

in the public domain:

1. Compromises in trilogues are struck too quickly, leaving limited possibilities for thorough input

from internal and external stakeholders;

2. Early agreements sometimes include new elements which have not been accounted for in the

Commission’s impact assessment;

3. Early agreements make it challenging for national parliaments to conduct effective scrutiny of their

governments in EU affairs;

4. Efficiency comes at the cost of transparency.

The first three concerns are of a more specific nature, whereas the last point – that efficiency comes at the

cost of transparency – is a recurrent and general concern related to early agreements. In the following

sections, we elaborate and discuss each of the four concerns in turn.

5.1 Early agreements and the ‘speed’ of negotiations

Both scholars and practitioners have over the years raised the concern that the speed of negotiations in early

agreements may come at the cost of deliberation.

142

While it is certainly true that the length of the decision-

making process of early agreement is reduced because of fewer required readings, early agreements can

actually take a long time to be reached and are not necessarily shorter than non-early agreements, as seen in

Chapter 3. This does not, however, rule out that there may be normative issues with early agreements and

that external stakeholders find it difficult to follow what is going on and to get their voices heard over the

course of the trilogue meetings. Yet, these concerns relate more to the nature of trilogues as being a closed

forum rather than early agreements happening so fast that there is limited time for the consideration of files.

141

See for instance: Burns, C.J. (2013) ‘Consensus and compromise become ordinary – but at what cost? A critical analysis of the impact of the

changing norms of co-decision upon European Parliament committees’, Journal of European Public Policy, 20 (7): 988-1005; Mühlböck, M. and

Rittberger, B. (2015) ‘The Council, the European Parliament, and the paradox of inter-institutional cooperation’, European Integration Online Papers

19 (4): 1-20.

142

See for instance: Héritier, A. and Reh, C. (2012) ‘Codecision and its discontents: Intra-organisational politics and institutional reform in the

European Parliament’, West European Politics, 35 (5): 1134-1157; Lord, C. (2013) ’The democratic legitimacy of codecision’, Journal of European

Public Policy 20 (7): 1056-1073; Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional innovation and

change’, Journal of European Public Policy 10 (2): 171-88.

Investigation of informal trilogue negotiations since the Lisbon Treaty

A concern related to the ‘speed’ of early agreements is the perception of some stakeholders that early

agreements sometimes push key issues into delegated and implementing acts, where procedures are seen as

being equally complex and vague.

143

It is, however, almost impossible to assess this proposition empirically.

A proper examination would involve a comparison of the delegation of powers in post-Comitology under

conditions of early agreements and non-early agreements. However, as 75 percent of all legislation today is

concluded by early agreements (at least in the current parliamentary term), with only relatively

uncomplicated files being finalised without an early agreement, it is impossible to make a valid comparison.

Moreover, even if more power had been delegated to the Commission in Comitology during the period in

which the use of early agreements increased, this would not necessarily mean that there is a causal

relationship between the two. Therefore, it is impossible to draw conclusions on whether there is a

systematic tendency to increase delegation of powers to the Commission to fill out the details in early

agreements and finalise deals between the co-legislators more quickly. It could be interesting for future

studies to assess whether more power has been delegated to the Commission over the course of trilogue

meetings compared to the initial Commission proposal, as academic literature has not yet examined this.

This type of study would require qualitative process-tracing of the evolving four-column documents and

interviews with negotiators present during the meetings on specific cases.

5.2 Early agreements and deliberation

Interest groups have also raised the concern that there is little reflection on the consequences of substantive

amendments introduced as part of a compromise during trilogue meetings.

144

For instance, according to the

European Trade Union Confederation (ETUC), trilogues ‘raise serious concerns with regard to better

regulation principles: the rapid output from the co-legislators is favoured over careful consideration of

legislative amendments’.

145

Proposals may, indeed, change significantly as a result of new amendments being

introduced after the publication of the Commission proposal. These amendments may not have been

accounted for in the Commission’s impact assessment and therefore result in a lack of evidence of how

costly or burdensome the new amendments are for affected stakeholders.

146

It is not uncommon for affected interests to express dissatisfaction with new amendments introduced later in

the legislative process if the latter have not been assessed at an earlier stage. For instance, scholarly work on

lobbying on ‘Regulation (EU) No 510/2011 on setting emission performance standards for new light

commercial vehicles’

147

shows that several car manufacturers were dissatisfied with the way in which a

143

See for instance: Mahony, H. (2015) ‘Tough battle expected on EU law-making culture’, 10 February 2015, EUobserver, available at

https://euobserver.com/institutional/127556 (last accessed: 19 April 2017).

144

Fox, B. (2014) ‘Secret EU lawmaking: the triumph of the trilogues’, EUobserver, 4 April 2014, available at

https://euobserver.com/investigations/123555; Mahoney, H. (2013) ‘Complex EU law-making dubbed “infernal, undemocratic”’, 4 March,

EUobserver, available at http://euobserver.com/political/119181 (last accessed: 19 April 2017).

145

European Trade Union Confederation (2015) ‘Proposal for an Interinstitutional Agreement on Better Regulation – ETUC position paper’, 16 July

2016, available at: https://www.etuc.org/documents/proposal-interinstitutional-agreement-better-regulation-%E2%80%93-etuc-position-

paper#.WQRwHLuLRhC (last accessed: 29 April 2017).

146

See for instance the submissions made by Danish Bankers Association/Danish Security Dealers Association, European Association of Hospital

Pharmacists, European Banking Federation, European Movement International, European Public Health Alliance, The Wealth Management

Association, and the UK House of Commons’ European Scrutiny Committee to the Ombudsman’s consultation: European Ombudsman (2016)

’European Ombudsman public consultation on the transparency of trilogues’, Strasbourg, 12 July, available at

https://www.ombudsman.europa.eu/en/cases/correspondence.faces/en/61589/html.bookmark (last accessed: 23 April 2017); Fox, B. (2014) ‘Secret

EU lawmaking: the triumph of the trilogues’, EUobserver, 4 April 2014, available at https://euobserver.com/investigations/123555 (access 19 April

2017).

147

Regulation No 510/2011 of the European Parliament and of the Council of 11 May 2011 setting emission performance standards for new light

commercial vehicles as part of the Union’s integrated approach to reduce CO 2 emissions from light-duty vehicles. OJ L 145/54, 31 May 2011.

Investigation of informal trilogue negotiations since the Lisbon Treaty

compromise was struck in trilogue, as they found it to be more driven by pressures to avoid a second reading

rather than by scientific consideration of what the best outcome would be. The crux of the discussions was

what the long-term 2020 target for reducing CO

emissions from light commercial vehicles (vans) should be.

The Council initially opted for a long-term target of 155 g CO

/km and Parliament for 140 g CO2/km. In

order to make a deal, they eventually agreed on 147 g CO

/km, meeting half way between their respective

positions.

148

This political compromise was not subjected to a revised impact assessment, but reflected a

political compromise.

It is important to note that the concern about the limited use of impact assessment later in the legislative

process when substantive amendments are made is not confined to legislation concluded by means of an

early agreement. New substantive amendments may be introduced later in the policy process regardless of

whether deals are struck early in trilogue meetings. The EU institutions are aware of the problem of adopting

new substantive amendments later in the policy process without subjecting them to impact assessments.

Better law-making is a principle recognised by the EU institutions, most recently with the Interinstitutional

Agreement on Better Law-Making of 13 April 2016. The agreement explicitly states that the EP and Council

‘will when they consider this to be appropriate and necessary for the legislative process, carry out impact

assessments in relation to their substantial amendments to the Commission’s proposal’.

149

The definition of

what constitutes a ‘substantial’ amendment is, however, left to the respective institutions to decide upon and

each institution is responsible for determining how to organise its impact assessment work, including quality

control. As the institutions find it difficult to provide a definition of ‘substantive’ that applies across the

board, the assessment is made on a case-by-case basis.

150

The Impact Assessment Handbook of the EP

151

states that it is up to the responsible parliamentary committee

to decide whether one or more of the amendments tabled during its consideration of a Commission’s

proposal are considered ‘substantive’ and whether a new (or updated) impact assessment should be

conducted. Ex-post impact assessments (i.e., after the Commission’s proposal has been tabled) can be

conducted at each stage of the legislative procedure as long as the impact assessment complies with the time

constraints specific to each reading and does not unduly delay the process. According to the EP’s Impact

Assessment Unit, the EP carried out six impact assessments of substantive amendments in its seventh

parliamentary term and, so far, two in the eighth term.

152

The Commission may also, on its own initiative or

upon invitation from one of the co-legislators, complement its own impact assessment or undertake other

analytical work it considers necessary. In practice, however, impact assessments conducted later in the

148

Dionigi, M.K. (2017) Lobbying in the European Parliament: The Battle for Influence, London: Palgrave Macmillan, pp. 67-68.

149

European Parliament, Council of the European Union, and Commission (2016) ‘Interinstitutional Agreement on Better Law-Making of 13 April

2016’, OJ L 123 12.05.2016, p. 0001, para. 15.

150

European Parliament (2012) ‘Conference of Committee Chairs, Impact Assessment Handbook, Guidelines for Committees’. 15 November 2012,

PE 499.457/CPG, Brussels: European Parliament, available at http://www.europarl.europa.eu/EPRS/impact_assesement_handbook_en.pdf (last

accessed: 27 April 2017).

151

European Parliament (2012) ‘Conference of Committee Chairs, Impact Assessment Handbook, Guidelines for Committees’. 15 November 2012,

PE 499.457/CPG, Brussels: European Parliament, available at http://www.europarl.europa.eu/EPRS/impact_assesement_handbook_en.pdf (last

accessed: 27 April 2017).

152

This information was provided to the authors by the EP’s Impact Assessment unit upon request. For the seventh term, these include amendments

on the following substantive amendments: (1) Fully paid maternity leave of 18 and 20 weeks - IMPACT ASSESSMENT (PE 425.629) - September

2010 (2) Impact Assessment of some of the European Parliament's amendments on the Commission Recasting Proposal on RoHS (Restriction on the

use of certain hazardous substances in electrical and electronic equipment) (PE 433.449) - June 2010, (3) Financing the environmentally sound

recycling and treatment of ships (PE 496.739) - February 2013, (4) Potential impact on SMEs of 18 EP amendments to two proposed Public

Procurement Directives (PE 507.505) - June 2013, (5) Registration of motor vehicles: choice of number plates in Union colours (PE 514.068) -

November 2013, and (6)EU safety tested' marking in the context of a proposal on consumer product safety (PE 528.791) - April 2014. For the eight

term: (1) Money Market Funds: Impact Assessment of Substantive EP Amendments, (PE 545.547), March 2015 and (2) Provisional legal aid for

suspects or accused persons deprived of liberty and legal aid in European arrest warrant proceedings (PE 581.410), July 2016.

Investigation of informal trilogue negotiations since the Lisbon Treaty

policy process are uncommon. This means that typically, the Commission’s impact assessment is used as a

starting point at later stages of the legislative process, and hardly any impact assessments are conducted later

in the process.

A more frequent use of impact assessments later in the process may prolong the decision-making process and

require more resources, but such measures may indeed prevent the potential negative consequences of

amendments that have not been subject to a proper assessment. Ex-post impact assessments on substantive

amendments, provided that these include consultations of affected stakeholders, would also give more equal

access for stakeholders to highlight what in their view are potential consequences associated with substantive

amendments.

5.3 Early agreements and national parliamentary scrutiny

Some national parliaments find that early agreements make it more challenging for them to conduct effective

scrutiny of their governments in EU affairs. This concern is, for instance, reflected in a number of

submissions to the Ombudsman’s consultation and in COSAC’s bi-annual reports. Two out of the 19 bi-

annual COSAC reports published since the Lisbon Treaty entered into force include a section on national

parliaments’ views on trilogues, based on a questionnaire sent out to all the European affairs committees of

the 41 national parliaments/chambers in the EU’s 28 member states.

153

The 22nd bi-annual report of 4 November 2014 asked national parliaments about their views on (1) the early

conclusion of files, (2) whether they were informed about trilogue meetings, and (3) whether the rise in first

and early second reading agreements has had an impact on their ability to conduct scrutiny of EU acts.

154

The

vast majority of responding parliaments (27 out of 36) considered it positive that legislation is finalised early

in the legislative process, although concerns about the transparency of trilogues were criticised. Trilogues

were seen to reduce the ability of national parliaments to conduct proper government scrutiny as information

about the negotiations in trilogues was often lacking.

The 25th bi-annual report of 18 May 2016 asked parliaments if they responded to the consultation on

transparency of trilogues carried out by the European Ombudsman and if the exchange of information on

trilogues among national parliaments would help parliamentary scrutiny on EU affairs.

155

Only five out of 41

national parliaments/chambers responded to the Ombudsman’s consultation (including the Dutch lower

house, the UK Houses of Commons and Lords, the Romanian Senate, and the French National Assembly).

All five chambers raised similar concerns, most notably that:

 National parliaments primarily rely on their national government to provide them with information

about the trilogue meetings and to gain access to LIMITE documents

156

, which has made the task of

parliamentary oversight difficult;

153

Of the 28 member states, 15 have a unicameral parliament, while 13 have a bicameral parliament.

154

COSAC (2014) ‘Twenty-second bi-annual report: Developments in the European Union procedures and practices relevant to parliamentary

scrutiny’, 4 November, Brussels: COSAC.

155

COSAC (2016) ‘Twenty-fifth bi-annual report: Developments in the European Union procedures and practices relevant to parliamentary scrutiny’,

18 May, Brussels: COSAC.

156

Internal Council documents which are not automatically made public.

Investigation of informal trilogue negotiations since the Lisbon Treaty

 The trilogue process is not transparent in terms of timing, duration, and substance of negotiations,

which again was seen to make the scrutiny of both legislative proposals and government activities

difficult;

 The five chambers wanted the Council’s negotiating mandate to be publicly available. They also

wanted the four-column documents to be made available as soon as possible after the completion of

negotiations, ideally after each trilogue meeting.

157

COSAC’s 25th bi-annual report showed that a majority of national parliaments/chambers (29 out of 36

responded to the particular question on information sharing) considered the exchange of information on

trilogues among parliaments a useful tool for improving parliamentary scrutiny in EU affairs. The COSAC

report, however, also showed that information sharing on trilogues between parliaments is currently rare,

indicating that there may be room for improvement.

5.4 Early agreements and transparency

A recurrent concern in the public sphere is that the efficiency of early agreements comes at the cost of

transparency. Trilogues are often depicted as something of a black box for those not directly involved in

them. The absence of an official paper trail during trilogue negotiations does make it challenging for

stakeholders to follow what is going on during trilogue meetings and to understand how EU institutions

reconcile their views. As a senior advisor at the global consulting firm FleishmanHillard phrased it in an

interview in Politico: ‘Up until the trilogue itself, it’s a fairy open process. But in the trilogue, it’s 100

percent a lottery to see if the amendments you’ve submitted to MEPs survive […] The lack of transparency

means you have to be well-connected to a fairly small group of people’.

158

Early agreements do reduce the ability of stakeholders to follow what is going on because trilogues take

place in camera and with no official minutes or attendance lists, unless interest groups are well connected

and can gain informal access to trilogue documents (such as the four-column documents) through informal

contacts with key negotiators. This is also a point the European Ombudsman has focused on in her recent

inquiry into the transparency of trilogues, adopted in July 2016. The Ombudsman’s inquiry was not based on

any complaint or concern of maladministration, but reflected a broader concern of the Ombudsman about the

conduct in trilogues and its alignment with the transparency requirements of the legislative procedure in

primary law (Articles 15(2) and (3) TFEU).

159

In her report, she called for further transparency in EU

decision-making by publishing key documents related to trilogues.

160

Increasing public disclosure of information on trilogues would – according to the Ombudsman – reassure the

broader public that there is a ‘level playing field’ in Brussels, at least when it comes to access to information.

Currently, stakeholders need to spend considerable time and effort on researching the trilogue process by

157

European Ombudsman (2016) ’European Ombudsman public consultation on the transparency of trilogues’, Strasbourg, 12 July, available at

https://www.ombudsman.europa.eu/en/cases/correspondence.faces/en/61589/html.bookmark (last accessed: 23 April 2017)

158

Cooper, H. (2016)’Where European Democracy goes to die’, Politico, 12 July 2016, available at http://www.politico.eu/article/where-european-

democracy-goes-to-die-european-parliament/ (last accessed: 19 April 2017)

159

It should be noted that the Commission, the EP, and the Council considered the Ombudsman’s inquiry to fall outside the scope of her mandate as it

concerned the organisation of the legislative process and not maladministration.

160

European Ombudsman (2016) ‘Decision of the European Ombudsman setting out proposals following her strategic inquiry OI/8/2015/JAS

concerning the transparency of Trilogues’, OI/8/2015/JAS, Strasbourg: European Ombudsman.

Investigation of informal trilogue negotiations since the Lisbon Treaty

having conversations with those involved in the trilogue meetings to find out the content of the discussions.

This risks creating an uneven playing field between stakeholders who have well-established contacts with

negotiators and the resources to seek information, and those who do not.

The Ombudsman acknowledges the importance of trilogues for facilitating negotiations and compromises

between the institutions, but also highlights the difficulties for the public in finding out when trilogues take

place, what is being discussed, and by whom. She proposed to make the following documents publicly

available prior to trilogue meetings to enable citizens to hold their representatives to account and to engage

effectively in the legislative process:

 A calendar of trilogue meetings;

 The Council’s initial position on every legislative proposal (the third column);

 General summary agendas (not the actual agenda) focusing on the main points for discussion in

trilogues. She encouraged the EU institutions to create a single easy-to-use database with the

above information;

 A list of representatives of the three institutions attending the trilogue meetings.

The Ombudsman further recommended making the multi-column/four-column documents and the final

compromise text formally adopted by the Council and the EP available after the negotiations. She proposed

to make this public after the negotiations have been concluded to make sure that public scrutiny does not

have a direct (negative) impact on the negotiations and that disclosure does not damage the negotiating

process, as a compromise early in the process might later change. She also encouraged the institutions to

make this information available through the database mentioned above.

Four-column documents are the main working tool of the institutions in trilogues and are used to facilitate

compromises early. They are joint institutional documents, indicating the position of the respective

institutions. The first column (Commission proposal) and the second column (the EP’s position as reflected

in a committee report, plenary amendments, or a plenary resolution) are publicly available, whereas the third

column (the Council’s position, usually the mandate given to the presidency) and the fourth column (the

compromise text) are not publicly available during negotiations as they contain non-adopted text elements.

The Ombudsman does acknowledge that lobbying may be problematic if it happens when public

representatives need some space to discuss proposed legislation among themselves. She also highlights that a

democratic process hinges on appropriate and timely information that is available to everyone, and not only

to the few resourceful interests. This is, however, not necessarily a view shared by officials in the EP, where

some of our interviewees indicated that access to, for instance, agendas of trilogue meetings – including the

articles of the proposal that are subject to negotiation – might make it easier for resourceful groups to employ

significant resources to lobby trilogue members ahead of trilogue meetings. Their concern was that only

active and resourceful groups would pay attention to the documents, with, for instance, consumer interests

being even more disadvantaged.

A number of concerns were shared by most interest groups that replied to the Ombudsman’s consultation

before she conducted her inquiry into the issues of transparency in trilogues. These replies are by no means

necessarily representative of how interest groups generally view trilogue meetings; they only represent the

Investigation of informal trilogue negotiations since the Lisbon Treaty

views of the interest groups that actively decided to respond to the consultation. With this caveat in mind, the

shared suggestions include:

 It is essential that relevant information is made publicly available before trilogues take place,

including agendas, a list of the type of people attending, and meeting dates. This would have to be

done early enough to allow interested stakeholders to provide the decision-makers with ‘evidence-

based arguments’, thereby increasing the quality of the negotiations and ultimately of the legislation

that will be adopted;

 The initial position of all three institutions on a legislative file (the ‘mandates’) should be made

publicly available before trilogue negotiations commence.

161

The majority of our interviewees mentioned that some interest groups get access to the evolving four-column

documents anyway through their connections with people present in the trilogues. One example to illustrate

this point is the 2016 ‘General Data Protection Regulation’

162

, where stakeholders such as the international

advocacy group European Digital Rights claimed that they relied on leaked Council documents to follow the

trilogue negotiations, including the Council’s initial position and preparatory trilogue documents.

163

According to the Brussels media, the Council and the EP held diverging views on a number of key issues

(such as fines, consent to personal data processing, and mandatory protection officers), but eventually

reached an agreement during the trilogue negotiations

164

For stakeholders without access to leaked four-

column documents during the negotiations, it can be challenging to fully understand how disagreement

between the co-legislators has been overcome, unless access to documents is sought once the negotiations

are finalised.

The potential lack of information on what goes on from the moment that each institution gains its negotiating

mandate to start trilogue negotiations and the moment when the final compromise agreement is reached

between the co-legislators means that it is difficult for outsiders to understand how initially diverging views

get reconciled. The European Ombudsman and a number of interest groups illustrated this concern by

referring to the 2015 agreement on roaming charges for mobile phone calls made abroad.

165

At the beginning

of the negotiations, the EP was pushing for mobile tariff reductions that were higher than what the Council

wanted. As no public minutes were made available after each trilogue meeting and the corresponding four-

column documents were not made publicly available either (unless they were applied for through access to

documents), it is unclear how an agreement was found.

166

161

See, for instance, the submissions made by Accessnow, Birdlife International, ClientEarth, EuroChamber, European Movement International,

European Public Health Alliance, European Banking Federation, and Transparency International, to the Ombudsman’s consultation: European

Ombudsman (2016) ’European Ombudsman public consultation on the transparency of trilogues’, Strasbourg, 12 July, available at

https://www.ombudsman.europa.eu/en/cases/correspondence.faces/en/61589/html.bookmark (last accessed: 23 April 2017)

162

Regulation (EU) 2016/679 of the European Parliament and the Council of 27 April 2016 on the protection of natural persons with regard to the

processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), OJ L

119/1, 4 May 2016.

163

European Digital Rights (2015) General Data Protection Regulation: Document pool, 25 June 2015, available at https://edri.org/gdpr-document-

pool/ (last accessed 10 April 2017).

164

See for instance, Nielsen, N. (2015) ‘National governments punch holes in EU data protection bill’, EUobserver, 3 March 2015, available at

https://euobserver.com/justice/127856 (last accessed: 10 April 2017); Nielsen, N. (2015) ‘EU ministers back weaker data protection rules’,

EUobserver, 15 June 2015, available at https://euobserver.com/justice/129122 (last accessed: 10 April 2017); Nielsen, N. (2015) ‘Data bill enters

final leg of state-level talks’, EUobserver, 24 May 2015, available at https://euobserver.com/justice/128812 (last accessed: 10 April 2017).

165

Regulation (EU) 2015/2120 of the European Parliament and of the Council of 25 November 2015 laying down measures concerning open internet

access and amending Directive 2002/22/EC on universal service and users’ rights relating to electronic communications networks and services and

Regulation (EU) No 531/2012 on roaming on public mobile communications networks within the Union, OJ L 301/01, 26 November 2011.

166

Crisp, J, (2015) ‘Secretive ‘trialogue’ talks to agree EU law face investigation’, EurActiv, 21 April 2015, available at

http://www.euractiv.com/section/public-affairs/news/secretive-trialogue-talks-to-agree-eu-law-face-investigation/ (last accessed: 29 April 2017).

Investigation of informal trilogue negotiations since the Lisbon Treaty

The initial position of the Council (that is, the negotiating mandate given to the presidency to start trilogue

negotiations with the EP) ahead of trilogue negotiations with the EP is also unclear unless the Council

negotiates on the basis of a so-called General Approach (a mandate approved by the ministers). Only

General Approaches are always made publicly available; mandates reached at Coreper-level without a

General Approach are not required to be public. The positions of the Commission (its proposal) and the EP

(the adopted committee report) are always known. Even when the initial position of the Council is publicly

available, it is difficult for non-experts to locate them in the Council Register, as this requires knowledge of

how to navigate the register and find information on the exact document code.

For non-EU experts, the lack of correspondence between the elaborate formal procedure described in the

Treaty and the actual legislative process can be rather confusing. Missing policy positions from the Council

at the beginning of trilogue negotiations and lack of access to the evolving positions of the Commission, the

EP, and the Council during trilogue negotiations make it hard to attribute (aspects of) policy outcomes and

amendments to one or the other. One way of making the initial positions clearer to the public and interest

groups is to make the Council’s initial position publicly available by, for instance, aiming for the adoption of

a General Approach on most files, provided that these are available before the beginning of trilogue

meetings, which may not always be the case today. According to Birdlife International, ‘there are often

delays between mandates being adopted and then being published […] as mandates may only be publicly

available after initial trilogue meetings have taken place’.

167

This chapter has shown that external stakeholders are concerned about the lack of transparency of early

agreements as well as the general quality of legislation where substantive amendments are introduced later in

the policy process and not subjected to ‘quality control’ through a revised or new impact assessments. The

Ombudsman’s recommendations show that there is room for improving the transparency of trilogues and for

creating a more level playing field between stakeholders. As discussed in Chapter 6, the institutions could

consider making the provisional annotated agenda and meeting dates publicly available prior to each trilogue

meeting and when they are available. The Council could consider making all of its initial mandates available

before entering trilogues, either by generally opting for a General Approach and/or by making mandates

given at the Coreper level publicly available without a General Approach. The ‘evolving’ four-column

documents could be made automatically available once negotiations are finalised and the institutions have

adopted the legislation. This would allow external stakeholders and the public to understand how the

negotiations were conducted and how potential disagreements were overcome. All of this information could

be included in the joint database on legislation, which is currently being discussed and developed as part of

one of the transparency commitments in the Interinstitutional Agreement on Better Law-Making. The issue

of the quality of legislation through more frequent use of ex-post impact assessments is also an important

issue to be discussed within and between the institutions. However, this issue does not only relate to

decisions made in trilogues, but to all proposals that are substantially amended during the policy process.

167

See submissions by Birdlife International to the Ombudsman’s consultation: European Ombudsman (2016) ’European Ombudsman public

consultation on the transparency of trilogues’, Strasbourg, 12 July, available at

https://www.ombudsman.europa.eu/en/cases/correspondence.faces/en/61589/html.bookmark (last accessed: 23 April 2017)

Investigation of informal trilogue negotiations since the Lisbon Treaty

6. Conclusion

The nature of trilogues has changed over time, from being introduced as a mechanism to avoid the

conciliation procedure to attaining the status of ‘business as usual’ under the ordinary legislative procedure.

Early agreements have increased significantly since 1999, when the Amsterdam Treaty made it possible to

finalise legislation after only one reading. Existing academic and non-academic literature shows that the

popularity of early agreement reflects a number of concurrent factors, most notably: (1) the growing levels of

cooperation and mutual trust between the co-legislators, (2) the harvesting of efficiency gains (given the

absence of time limits and the lower majority requirements for passing legislation at first reading), and (3)

the desire on part of the EP’s and Council’s negotiating teams to be seen as successful and reliable

negotiators (see Chapter 2, section 2.3).

In this study, we defined early agreements as files that are concluded at either first or early second reading,

and based on an informal compromise between the co-legislators negotiated during trilogue meetings. This

definition excludes procedures that are concluded early without having been pre-agreed during trilogue

meetings.

Our study has relied on existing scholarly and practitioners’ literature on trilogues and early agreements. We

have also gathered our own quantitative and qualitative data. To assess the patterns of early agreements and

the changes over time, we used quantitative data on ordinary legislative files concluded between 1999 and

2016, obtained mainly from the Legislative Observatory database. The data used between 1999 and 2014

were gathered and analysed in previous research, whereas the data covering the first half of the current

eighth parliamentary term (2014-2016) were collected and analysed by us. This study is thus the first

published work to include a statistical descriptive analysis of early agreements in the current parliamentary

term. The quantitative data in this report were particularly used in Chapters 2 and 3.

To gain an insight into the ways in which the EP, Council, and the Commission make early agreements, as

presented in Chapter 4, we conducted 18 interviews with EU officials about the co-legislators’ practices

during trilogues, as these practices are, by nature of their informality, not written down. To analyse the

concerns of external stakeholders in Chapter 5, we analysed all the Conference of Parliamentary Committees

for Union Affairs’ (COSAC’s) bi-annual reports published since the Lisbon Treaty entered into force in

December 2009, and the European Ombudsman’s 50 submissions to public consultation as part of her 2016

investigation into the transparency of trilogue meetings. We also analysed the Brussels online media’s

coverage of trilogue meetings since the entering into force of the Lisbon Treaty.

6.1 Key findings

Our descriptive statistical analyses in Chapters 2 and 3 show that since the seventh parliamentary term

(2009-2014), early agreements have become the default option for almost all of the EP’s committees – even

for those committees that only gained full co-decision powers with the Lisbon Treaty. The data reveal four

overall patterns in the use of early agreements:

 The overall pattern: The use of early agreements increased rapidly between 1999 and 2008, and

Investigation of informal trilogue negotiations since the Lisbon Treaty

stabilised after 2008;

 The speed of negotiations: Between 1999 and 2016, procedures with early agreements were shorter

than procedures without early agreement. However, scratching below the surface, important

variation exists over time. In the first half of the current parliamentary term (2014-2016), procedures

with early agreements have taken significantly longer than procedures without early agreements,

reflecting that the category of non-early agreements includes mostly uncomplicated legislation,

which can be adopted quickly and without (major) changes;

 Early agreements and the complexity of legislation: Our findings show that files concluded at early

agreement since the Lisbon Treaty entered into force are characterised by higher levels of

complexity than files that are not subject to early agreements, using the number of recitals in the

Commission’s proposal as a proxy measure for complexity;

 Early agreements and the type of legislation: Early agreements exist across all types of legislation,

but have been most prevalent on adaptations of Comitology, recast legislation, and new legislation

with provisions to repeal previously adopted legislation.

Despite the widespread use of early agreements, these meetings have been subject to concerns about

transparency, the possibility of systematic exclusion of some actors from the legislative process, and the

potential lack of thorough deliberation of legislation. Some of the critiques of trilogues relate to the

democratic deficit inside the institutions (i.e., the involvement of the whole EP). However, the critiques

originate from a time when there was little control over the institutions’ respective negotiating teams. In the

early use of trilogues, the EP’s negotiating teams comprised fewer actors (and excluded shadow rapporteurs),

who could largely negotiate in trilogues without being held fully accountable by the rest of Parliament. As

shown in Chapter 4, this is no longer the case as trilogues have become more inclusive (by including all

political groups), representative (by having the initial negotiating mandate endorsed in plenary), and

accountable (by requiring the negotiating team to regularly report back to their committee and political

groups). This has made it easier for parliamentarians who are not represented in trilogues to follow what is

going on and to hold negotiating teams to account. It has also enabled interest groups and the public to

follow the overall progress made in trilogues by following committee and plenary meetings. This means that

the rules and practices in the EP concerning the conduct of trilogue meetings have become increasingly

institutionalised and formalised over the years.

Unlike the EP, the Council has not included its working procedures on trilogues in its rules of procedure,

although widely shared and entrenched practices exist within the Council in terms of how the presidency

(representing the Council in trilogues) obtains a mandate and reports back. Unlike the EP and the

Commission, the Council’s mandate is not always publicly available, as only mandates given by means of a

General Approach are made public. The detailed intra-institutional rules and practices on trilogues mean that

negotiators are held to account by their respective institutions and to a much higher degree than in the early

days of trilogues. However, no rules or practices exist on including other institutions than the Commission,

the Council, and the EP in trilogue meetings. Neither the EESC nor any other consultative EU institution has

access to trilogue meetings or enjoys privileged prerogatives in trilogues (such as access to information) as a

EU institution. They have the same level of information on the trilogue meetings as any other external

stakeholder.

Investigation of informal trilogue negotiations since the Lisbon Treaty

As demonstrated in Chapter 5, a number of concerns and perceptions about trilogues exist in the public

domain, most notable that:

 Early agreements result in more delegation of powers to the Commission;

 Early agreements sometimes include new elements which have not been accounted for in the

Commission’s impact assessment (quality of legislation);

 Early agreements are seen by some national parliaments as challenging effective parliamentary

scrutiny of governments;

 Efficiency is seen as coming at the expense of transparency.

In Chapter 5, we discussed each of these concerns in turn. First, there is the perception among a range of

interest representatives that early agreements sometimes push key issues into delegated and implementing

acts, rather than including more detailed provisions in the basic act negotiated between the EP and the

Council before delegating powers to the Commission in Comitology. While this might be the case, it is very

difficult to assess this ‘working hypothesis’ empirically. A proper examination of this conjecture would

require a comparison of the delegation of powers to the Commission in post-Comitology under conditions of

early and non-early agreements. However, as 75 percent of all legislation in the first half of the current

parliamentary term (2014-2016) is concluded by early agreements and only relatively uncomplicated files

are being finalised without early agreements, we cannot make a valid comparison. A proper examination of

the position would require detailed qualitative process-tracing of the evolving four-column document and

interviews with involved negotiators. Still, this would only make it possible to draw conclusions on a case-

study basis and not at an aggregate level (see Chapter 5, section 5.1).

Second, concerning the quality of legislation, it is more the exception than the rule that updated or new

impact assessments are conducted later in the legislative process when substantive amendments are made.

However, this concern is not confined to legislation concluded by means of an early agreement. New

substantive amendments may be introduced later in the policy process regardless of whether compromises

between the institutions are struck early using trilogue meetings. Moreover, the EU institutions are aware of

the problem of introducing new substantive amendments later in the policy process that have not been

accounted for in the Commission’s impact assessment or been subject to a new impact assessment (see

Chapter 5, section 5.2). More frequent use of ex-post impact assessments later in the decision-making

process may, indeed, prevent the potential negative consequences of amendments that have not be subject to

assessment earlier in the process. Ex-post impact assessments may also make it easier for affected

stakeholders to highlight potential consequences of substantive amendments, provided that affected

stakeholders are consulted.

Third, five out of the 41 national parliament chambers (the Dutch lower house, the UK Houses of Commons

and Lords, the Romanian Senate, and the French National Assembly) who responded to the Ombudsman’s

consultation found the task of parliamentary oversight challenging as they primarily rely on their national

government to provide them with information about trilogues and to give them access to LIMITE

documents. To improve national parliamentary scrutiny, the five chambers suggested that the Council should

always make its initial negotiating mandate publicly available and for the EU institutions to make the four-

column document available after negotiations have been finalised (see Chapter 5, section 5.3).

Investigation of informal trilogue negotiations since the Lisbon Treaty

Last, the absence of an official paper trail during trilogue meetings makes it challenging for external

stakeholders (such as interest groups) to follow what is going on during trilogue meetings and to fully

understand how ostensible irreconcilable positions between the EU institutions are reconciled. While well-

connected interest groups may get access to the evolving four-column document during trilogue negotiations

through informal contacts; the less well-connected stakeholders may not. According to the European

Ombudsman, public disclosure of information of trilogues would create a level playing field between

stakeholders in Brussels, at least concerning access to information. The Ombudsman specifically proposed to

make the following documents available before trilogue meetings: a calendar of trilogue meetings, the

Council’s initial position on every legislative proposal, summary (not full) agendas of each trilogue meeting,

and a list of representatives present in trilogues. She also recommended making the four-column document

available after the negotiations had been finalised (see Chapter 5, section 5.4).

6.2 Suggestions for improving the transparency and accountability of early agreements

Similar to the European Ombudsman’s recommendations, we argue that a number of changes could be

introduced to make trilogues more transparent to the outside world. More transparency about the initial

positions of the institutions and access to the four-column document would make it easier for external

stakeholders to understand the positions of all institutions and enhance the understanding of how differing

positions are reconciled – as has also been suggested by the Ombudsman. Moreover, increased transparency

can help national parliaments hold their governments to account for their roles in the EU legislative process.

Finally, increased transparency can help Council and EP plenaries to hold their negotiating teams to account

by means of so-called ‘fire-alarm control’. That is, if external stakeholders have more access to trilogue

documents, they can help the plenaries (and to some extent national parliaments) scrutinise the process by

communicating potential concerns. Below we suggest two main avenues for improving the transparency of

trilogues:

1. To make the Council’s mandate public: The Council could aim to make its initial mandate publicly

available before the first trilogue negotiation. As this is already done when mandates are adopted as

General Approach at the ministerial level, it could be considered to expand this to the level of Coreper.

This would facilitate a better understanding of the Council’s initial position/mandate, both inside and

outside the EU institutions. The EP’s and Commission’s positions are already publicly available before

negotiations;

2. To create a joint database on legislative files: As part of the EU institutions’ commitment in the

Interinstitutional Agreement on Better Law-Making to create a joint database – merging existing

databases (EurLex, the Council Register, and the Legislative Observatory) – the EU institutions could

consider including information about trilogue meetings, particularly:

a. The institutions could make the trilogue dates accessible ex-ante, when known, and make the

annotated trilogue agenda available immediately before a trilogue meeting. Agendas might

change many times before a meeting, but at least the final agenda could be published when it is

available;

b. Four-column documents could be made automatically available in the joint database when

legislation is finalised;

Investigation of informal trilogue negotiations since the Lisbon Treaty

c. It is important that the joint database is user-friendly for the general public, so that

documentation on legislation can be found without knowing specific document numbers. Once

the database is up and running, it could offer an e-mail newsletter sign-up service, as provided

by the European Ombudsman and many government departments at the national level, to give

anyone the option to receive updates on the trilogue and legislative process. This would ensure a

level playing field for interest groups by making access to information available to all, and not

just to the most well-connected and resourceful actors.

It is important, however, to keep in mind that ‘full’ transparency does not necessarily make politics more

open, civilised or accepted by the public. One should be careful not to overestimate the positive effect of

publicity. Full transparency of trilogue meetings during the negotiating process may run the risk of impeding

decision-making efficiency by curtailing negotiators’ room for compromise. It also runs the risk that the

‘real’ decision-making is moved to less transparent settings with the net result being a further loss of

transparency (as well as inclusiveness). Therefore, making more rules about transparency needs to carefully

take into account how this will actually play out in practice, particularly if there is a risk that increased

transparency moves the negotiations elsewhere. For instance, if minutes of trilogue meetings or the four-

column document were required to be made public during negotiations, it may make the public positions of

the negotiators more rigid and less specific. Negotiations would risk presenting a ‘front-stage’ show, which

focuses on image building rather than revealing their ‘true’ preferences.

168

Increased transparency may not

lead to more accountability of decision-makers, as is sometimes assumed. As suggested in this chapter and

by the European Ombudsman, there is, however, room for improvement in the current transparency regimes

governing trilogues without risking that negotiators put on a front stage show and move the real negotiations

elsewhere.

The 2007 Joint Declaration on Practical Arrangements for the Codecision Procedure, still in force today, had

a clear aim to announce, where practicable, when trilogues take place in the EP and the Council to enhance

transparency. A similar recommendation was made by the European Ombudsman as part of her inquiry on

the transparency of trilogues. Similarly, the EU institutions have in the Interinstitutional Agreement on

Better Law-Making, adopted on 13 April 2016, made a number of commitments on transparency. The

Agreement includes a commitment to ‘ensure the transparency of legislative procedures, on the basis of

relevant legislation and case-law, including an appropriate handling of trilateral negotiations’. This broad

commitment is followed by a commitment to ‘improve communication to the public during the whole

legislative cycle’ and to look at further ways to ‘facilitate the traceability of the various steps in the

legislative process’ with a view to establishing a dedicated joint database on the state of play of legislative

files.

169

Furthermore, the De Capitani case before the CJEU is still pending and raises important questions

about the need to make trilogue meetings more transparent. Therefore, the time seems right to include crucial

information on trilogues in the forthcoming joint database to improve the transparency of trilogue meetings.

It is our hope that this report can provide a useful starting point for a discussion on how a revised

transparency regime of trilogues could be approached, maintaining the delicate balance between the need for

increased transparency and ‘space to think’ for legislators, as recognised by the European Ombudsman.

168

Goffman, E. (1959) Presentation of Self in Everyday Life, USA: Anchor Books.

169

European Parliament, Council of the European Union, and Commission (2016) ‘Interinstitutional Agreement on Better Law-Making of 13 April

2016’, OJ L 123 12.05.2016, p. 7.

Investigation of informal trilogue negotiations since the Lisbon Treaty

References

Brandsma, G.J. (2015) ‘Co-decision after Lisbon: The politics of informal trilogues in European Union

lawmaking,’ European Union Politics 16 (2): 300-19.

Bressanelli, E., Héritier, A., Koop, C. and Reh, C. (2014) ‘The informal politics of codecision: Introducing a

new data set on early agreements in the European Union’, EUI Working Paper 2014/64 RSCAS (EUDO).

Bressanelli, E., Koop, C., and Reh, C. (2016) ‘The impact of informalisation: Early agreements and voting

cohesion in the European Parliament,’ European Union Politics 17 (1): 91-113.

Burns, C.J. (2013) ‘Consensus and compromise become ordinary – but at what cost?: A critical analysis of

the impact of the changing norms of co-decision upon European Parliament committees’, Journal of

European Public Policy, 20 (7): 988-1005.

Burns, C.J. and Carter, N. (2010) ‘Is co-decision good for the environment? An analysis of the European

Parliament’s green credentials,’ Political Studies 58 (1): 123-42.

Case T-540/15, De Capitani v European Parliament

Cases C-39/05 P and C-52/05 P Kingdom of Sweden and Maurizio Turco v the Council

C-280/11 P Council v Access Info Europe.

Cooper, H. (2016) ‘Where European democracy goes to die’, Politico 12 July 2016, available at

http://www.politico.eu/article/where-european-democracy-goes-to-die-european-parliament/ (last accessed:

19 April 2017).

COSAC (2014) ‘Twenty-second bi-annual report: Developments in the European Union procedures and

practices relevant to parliamentary scrutiny’, 4 November, Brussels: COSAC.

COSAC (2016) ‘Twenty-fifth bi-annual report: Developments in the European Union procedures and

practices relevant to parliamentary scrutiny’, 18 May, Brussels: COSAC.

Council of the European Union (2016) ‘Comments on the Council’s Rules of Procedure. European Council’s

and Council’s Rules of Procedure’, Brussels: European Council and the Council.

Council of the European Union (2016) ‘Note from the Presidency to the Committee of the Permanent

Representatives/Council on the Implementation of the InterInstitutional Agreement on Better Law-Making –

Improving transparency’, Brussels, June 2016, available at

http://data.consilium.europa.eu/doc/document/ST-10120-2016-REV-1/en/pdf (last accessed: 22 April 2017).

Investigation of informal trilogue negotiations since the Lisbon Treaty

Crisp, J, (2015) ‘Secretive ‘trialogue’ talks to agree EU law face investigation’, EurActiv, 21 April 2015,

available at http://www.euractiv.com/section/public-affairs/news/secretive-trialogue-talks-to-agree-eu-law-

face-investigation/ (last accessed: 29 April 2017).

De Capitani, E. (2015) ‘(EU’s) laws are like sausages, you should never watch them being made’, European

Area of Freedom Security & Justice Free Group, Blogpost, 12 July 2015, available at https://free-

group.eu/2015/07/12/eus-laws-are-like-sausages-you-should-never-watch-them-being-made/ (accessed 12

April 2017).

Curtin, D. (2009) Executive Power in the European Union: Law, Practices, and the Living Constitution.

Oxford: Oxford University Press.

Curtin, D. and Leino, P. (2016) ‘Openness, transparency and the right of access to documents in the EU: In-

depth analysis for the PETI committee’, PE 556.973, Brussels: European Parliament.

De Ruiter, R. and Neuhold, C. (2012) ‘Why is fast track the way to go? Justifications for early agreement in

the co-decision procedure and their effects,’ European Law Journal 18 (4): 536–554.

Dionigi, M.K. (2017) Lobbying in the European Parliament: The Battle for Influence, London: Palgrave

Macmillan.

Directive (EU) 2015/254 of the European Parliament and of the Council of 11 February 2015 repealing

Council Directive 93/5/EEC on assistance to the Commission and cooperation by the Member States in the

scientific examination of questions relating to food (Text with EEA relevance), OJ L 43, 18 February 2015.

Directive (EU) 2015/413 of the European Parliament and of the Council of 11 March 2015 facilitating cross-

border exchange of information on road-safety-related traffic offences (Text with EEA relevance), OJ L 68,

13 March 2015.

Directive (EU) 2015/412 of the European Parliament and of the Council of 11 March 2015 amending

Directive 2001/18/EC as regards the possibility for the Member States to restrict or prohibit the cultivation of

genetically modified organisms (GMOs) in their territory (Text with EEA relevance), OJ L 68, 13 March

2015.

Directive (EU) 2015/719 of the European Parliament and of the Council of 29 April 2015 amending Council

Directive 96/53/EC laying down for certain road vehicles circulating within the Community the maximum

authorised dimensions in national and international traffic and the maximum authorised weights in

international traffic (Text with EEA relevance), OJ L 115, 6 May 2015.

Earnshaw, D. and Judge, D. (1995) ‘Early days: The European Parliament, co-decision and the European

Union legislative process post-Maastricht,’ Journal of European Public Policy 2 (4): 624-49.

Investigation of informal trilogue negotiations since the Lisbon Treaty

European Commission (2012) Communication from the Commission to the Council, the European

Parliament, the European Economic and Social Committee and the Committee of the Regions on EU

Regulatory Fitness, COM(2012) 746 final, 12 December 2012.

European Commission (2011) ‘Proposal for a Regulation of the European Parliament and of the Council

amending Regulation (EC) No 1049/2001 regarding public access to European Parliament, Council and

Commission documents’, COM(2011) 137 final/2.

European Digital Rights (2015) General Data Protection Regulation: Document pool, 25 June 2015,

available at https://edri.org/gdpr-document-pool/ (last accessed: 10 April 2017).

European Ombudsman (2016) ’European Ombudsman public consultation on the transparency of trilogues’,

Strasbourg, 12 July, available at

https://www.ombudsman.europa.eu/en/cases/correspondence.faces/en/61589/html.bookmark (last accessed:

23 April 2017).

European Ombudsman (2016) ‘Decision of the European Ombudsman setting out proposals following her

strategic inquiry OI/8/2015/JAS concerning the transparency of Trilogues’, OI/8/2015/JAS, Strasbourg:

European Ombudsman

European Parliament (1995) The European Parliament Delegation to the Conciliation Committee Progress

Report for the second half of 1994, Brussels: European Parliament.

European Parliament (1999) ‘Activity Report 1 November 1993 to 30 April 1999 (4

parliamentary term)’,

PE 230.998, Brussels: European Parliament, Delegations to the Conciliation Committee.

European Parliament (2004) ‘Activity Report 1 May 1999 to 30 April 2004 (5th parliamentary term)’, PE

287.644, Brussels: European Parliament, Delegations to the Conciliation Committee.

European Parliament (2006) ‘Conciliation and Codecision Activity Report July 2004 – December 2006 (6

parliamentary term. first half term), DV/651053, Brussels: Delegations to the Conciliation Committee.

European Parliament (2009) ‘Activity Report 1 May 2004 to 13 July 2009, (6th parliamentary term)’,

PE427.162v01-00, Brussels: European Parliament, Delegations to the Conciliation Committee.

European Parliament (2011) ‘European Parliament legislative resolution of 15 December 2011 on the

proposal for a regulation of the European Parliament and of the Council regarding public access to European

Parliament, Council and Commission documents (recast) (COM(2008)0229 – C6-0184/2008 –

2008/0090(COD))’, Brussels: European Parliament

European Parliament (2012) ‘Conference of Committee Chairs, Impact Assessment Handbook, Guidelines

for Committees’. 15 November 2012, PE 499.457/CPG, Brussels: European Parliament, available at

http://www.europarl.europa.eu/EPRS/impact_assesement_handbook_en.pdf (last accessed: 27 April 2017)

Investigation of informal trilogue negotiations since the Lisbon Treaty

European Parliament (2012) ‘Activity report on codecision and conciliation: 4 July 2009 to 31 December

2011 (7

parliamentary term)’, DV\903361, Brussels: Delegations to the Conciliation Committee.

European Parliament (2013) ‘European Parliament resolution of 12 June 2013 on the deadlock on the

revision of Regulation (EC) No 1049/2001’, 2013/2637(RSP), Brussels: European Parliament.

European Parliament (2013) ‘Twenty Years of Codecision Conference Report’, 5 November 2013, Brussels:

European Parliament, Conciliations and Codecision Secretariat European Parliament.

European Parliament (2014) ‘Rules of Procedure of the European Parliament (8

parliamentary term)’,

Brussels: European Parliament.

European Parliament (2014) ‘Activity report on codecision and conciliation: 4 July 2009 to 30 June 2014

(7th parliamentary term)’, DV\1031024, Brussels: European Parliament, Conciliation and Codecision Unit.

European Parliament (2016) ‘European Parliament resolution of 28 April 2016 on public access to

documents (Rule 116(17)) for the years 2014-2015, 2015/2287(INI), Brussels: European Parliament.

European Parliament (2017) Activity report on the ordinary legislative procedure: 4 July 2014 to 31

December 2016 (8th parliamentary term)’, PE 595.931, Brussels: European Parliament, Conciliation and

Codecision Unit.

European Parliament (2017) ‘Rules of Procedure, Provisional Edition, January 2017’, P8_TA-PROV (2016)

0484, Brussels: European Parliament.

European Parliament, Council of the European Union, and Commission (2016) ‘Interinstitutional Agreement

on Better Law-Making of 13 April 2016’, OJ L 123 12.05.2016, p. 0001.

European Parliament, Council and Commission (1999) ‘Joint declaration on practical arrangements for the

new codecision procedure (Article 251 of the Treaty establishing the European Community)’, OJ C 148/01,

28 May, Brussels: European Parliament, Council and Commission.

European Parliament, Council, and Commission (2007) ‘Joint Declaration on Practical Arrangements for the

Codecision Procedure (Art. 251 of the EC Treaty)’, OJ C 145/05, 30 June, Brussels: European Parliament,

Council and Commission.

European Trade Union Confederation (2015) ‘Proposal for an Interinstitutional Agreement on Better

Regulation – ETUC position paper’, 16 July 2016, available at: https://www.etuc.org/documents/proposal-

interinstitutional-agreement-better-regulation-%E2%80%93-etuc-position-paper#.WQRwHLuLRhC (last

accessed: 29 April 2017).

Farrell, H. and Héritier, A. (2003) ‘Formal and informal institutions under codecision: Continuous

constitution-building in Europe,’ Governance 16 (4): 577-600.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Farrell, H. and Héritier, A. (2004) ‘Interorganizational negotiation and intraorganizational power in shared

decision making: Early agreements under codecision and their impact on the European Parliament and

Council’, Comparative Political Studies 37 (10): 1184-1212.

Fox, B. (2014) ‘Secret EU lawmaking: The triumph of the trilogues’, EUobserver 4 April 2014, available at

https://euobserver.com/investigations/123555 (last accessed: 19 April 2017).

Goffman, E. (1959) The Presentation of Self in Everyday Life, USA: Anchor Books.

Häge, F. (2013) Bureaucrats as Law-Makers: Committee Decision-Making in the EU Council of Ministers,

London: Routledge.

Harden, I. (2009) ‘The Revision of Regulation 1049/2001 on Public Access to Documents’, European Public

Law, 15(2): 239-256.

Helmke, G. and Levitsky, S. (2004) ’Informal institutions and comparative politics: A research agenda’.

Perspectives on Politics, 2 (4): 725-740.

Héritier, A. (2007) Explaining Institutional Change in Europe, Oxford: Oxford University Press.

Héritier, A. and Reh, C. (2012) ’Codecision and its discontents: Intra-organisational politics and institutional

reform in the European Parliament’, West European Politics 35 (5): 1134-1157.

Hix, S. (2002) ’Constitutional agenda-setting through discretion in rule interpretation: Why the European

Parliament won at Amsterdam’, British Journal of Political Science 32 (2): 259-80.

Hix, S. and Høyland, B. (2011) The Political System of the European Union, Third edition, Basingstoke:

Palgrave Macmillan.

House of Commons European Scrutiny Committee (2017) ‘Transparency of decision-making in the Council

of the European Union’, HC 128, 26 May.

House of Lords European Union Committee (2009) ‘Codecision and national parliamentary scrutiny’, Hl

125, 21 July.

Huber, K. and Shackleton, M. (2013) ‘Codecision: A practitioner’s view from inside the Parliament’,

Journal of European Public Policy 20 (7): 1040-55.

Judge, D. and Earnshaw, D. (2011) ‘“Relais actors” and co-decision first reading agreements in the European

Parliament: The case of the advanced therapies regulation,’ Journal of European Public Policy 18 (1): 53-71.

Juncker, J.-C. (2014) ‘The Juncker Commission: The right team to deliver change’, press conference by the

President-elect of the European Commission, 10 September, available at: http://europa.eu/rapid/press-

release_SPEECH-14-585_en.htm (last accessed: 30 April 2017).

Investigation of informal trilogue negotiations since the Lisbon Treaty

Koop, C., Reh, C. and Bressanelli, E. (2017) ’When politics prevails: Parties, elections and loyalty in the

European Parliament,’ Paper presented at the Fifteenth Biennial Conference of the European Union Studies

Association, Miami, 4-6 May.

Leino, P. (2014) ‘Transparency, Participation and EU Institutional Practice: An inquiry into the Limits of the

“Widest Possible”’, EUI Working Paper, Law 3/2014.

Lewis, J. (2015) ‘The Council of the European Union and the European Council’, in J. Magone

(ed), Routledge Handbook of European Politics, Abingdon and New York: Routledge, pp. 219-34.

Lord, C. (2013) ’The democratic legitimacy of codecision’, Journal of European Public Policy 20 (7): 1056-

1073.

Mahony, H. (2013) ‘Complex EU law-making dubbed “infernal, undemocratic”’, 4 March, EUobserver,

available at http://euobserver.com/political/119181 (last accessed: 19 April 2017)

Mahony, H. (2015) ‘Tough battle expected on EU law-making culture’, 10 February, EUobserver, available

at https://euobserver.com/institutional/127556 (last accessed: 19 April 2017).

McElroy, G. (2006) ‘Committee Representation in the European Parliament’, European Union Politics, 7(1):

5-29.

Mühlböck, M. and Rittberger, B. (2015) The Council, the European Parliament, and the paradox of inter-

institutional cooperation,’ European Integration Online Papers 19 (4): 1-20.

Naurin, D (2007) Deliberation behind Closed Doors. Transparency and Lobbying in the European Union,

Colchester: ECPR Press.

Neuhold, C., Rozenberg, O., Smith, J., Hefftler, C. (eds.) (2015) Palgrave Handbook of National

Parliaments and the European Union, London: Macmillan.

Nielsen, N. (2015) ‘National governments punch holes in EU data protection bill’, EUobserver, 3 March

2015, available at https://euobserver.com/justice/127856 (last accessed: 10 April 2017).

Nielsen, N. (2015) ‘EU ministers back weaker data protection rules’, EUobserver, 15 June 2015, available at

https://euobserver.com/justice/129122 (last accessed:10 April 2017).

Nielsen, N. (2015) ‘Data bill enters final leg of state-level talks’, EUobserver, 24 May 2015, available at

https://euobserver.com/justice/128812 (last accessed: 10 April 2017).

Ostrom, E. (1986) ‘An agenda for the study of institutions,’ Public Choice 48 (1): 3-25.

Priestley, J. (2008) Six battles that shaped Europe’s Parliament, London: John Harper Publishing.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Rasmussen, A. (2011) ‘Early conclusion in bicameral bargaining: Evidence from the co-decision legislative

procedure of the European Union,’ European Union Politics 12 (1): 41-64.

Rasmussen, A. and Reh, C. (2013) ‘The consequences of concluding codecision early: Trilogues and intra-

institutional bargaining success,’ Journal of European Public Policy 20 (7): 1006-24.

Regulation (EC) No 1049/2001 of the European Parliament and of the Council 30 May 2001 regarding

public access to European Parliament, Council and Commission documents, OJ L 145/43, 31 May 2001.

Rasmussen, A. and Shackleton, M. (2005) ‘The scope for action of the European Parliament negotiators in

the legislative process: Lessons of the past and for the future’, paper presented at the EUSA Biannual

Meeting, Austin, TX, March-April.

Regulation No 510/2011 of the European Parliament and of the Council of 11 May 2011 setting emission

performance standards for new light commercial vehicles as part of the Union’s integrated approach to

reduce CO 2 emissions from light-duty vehicles. OJ L 145/54, 31 May 2011.

Regulation (EU) No 1303/2013 of the European Parliament and of the Council of 17 December 2013 laying

down common provisions on the European Regional Development Fund, the European Social Fund, the

Cohesion Fund, the European Agricultural Fund for Rural Development and the European Maritime and

Fisheries Fund and laying down general provisions on the European Regional Development Fund, the

European Social Fund, the Cohesion Fund and the European Maritime and Fisheries Fund and repealing

Council Regulation (EC) No 1083/2006, OJ L 347, 20 December 2013.

Regulation (EU) No 1383/2014 of the European Parliament and of the Council of 18 December 2014

amending Council Regulation (EC) No 55/2008 introducing autonomous trade preferences for the Republic

of Moldova, OJ L 372, 30 December 2014.

Regulation (EU) 2015/2120 of the European Parliament and of the Council of 25 November 2015 laying

down measures concerning open internet access and amending Directive 2002/22/EC on universal service

and users’ rights relating to electronic communications networks and services and Regulation (EU) No

531/2012 on roaming on public mobile communications networks within the Union, OJ L 301/01, 26

November 2011.

Regulation (EU) 2015/755 of the European Parliament and of the Council of 29 April 2015 on common rules

for imports from certain third countries, OJ L 123, 19 May 2015.

Regulation (EU) 2015/1146 of the European Parliament and of the Council of 8 July 2015 fixing the

adjustment rate provided for in Regulation (EU) No 1306/2013 for direct payments in respect of the calendar

year 2015, OJ L 191, 17 July 2015.

Regulation (EU) 2016/93 of the European Parliament and of the Council of 20 January 2016 repealing

certain acts from the Schengen acquis, OJ L 26, 2 February 2016.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Regulation (EU) 2016/401 of the European Parliament and of the Council of 9 March 2016 implementing the

anti-circumvention mechanism provided for in the Association Agreement between the European Union and

the European Atomic Energy Community and their Member States, of the one part, and Georgia, of the other

part, OJ L 77, 23 March 2016.

Regulation (EU) 2016/580 of the European Parliament and of the Council of 13 April 2016 on the

introduction of emergency autonomous trade measures for the Republic of Tunisia, OJ L 102, 18 April 2016.

Regulation (EU) 2016/1014 of the European Parliament and of the Council of 8 June 2016 amending

Regulation (EU) No 575/2013 as regards exemptions for commodity dealers (Text with EEA relevance), OJ

L 171, 29 June 2016.

Regulation (EU) 2016/1076 of the European Parliament and of the Council of 8 June 2016 applying the

arrangements for products originating in certain states which are part of the African, Caribbean and Pacific

(ACP) Group of States provided for in agreements establishing, or leading to the establishment of, economic

partnership agreements, OJ L 185, 8 July 2016.

Regulation (EU) 2016/679 of the European Parliament and the Council of 27 April 2016 on the protection of

natural persons with regard to the processing of personal data and on the free movement of such data, and

repealing Directive 95/46/EC (General Data Protection Regulation), OJ L 119/1, 4 May 2016.

Reh, C., Héritier, A., Bressanelli, E, and Koop, C. (2013) ‘The informal politics of legislation: Explaining

secluded decision making in the European Union,’ Comparative Political Studies 46 (9): 1112-42.

Reh, C. (2014) ‘Is informal politics undemocratic? Trilogues, early agreements and the selection model of

representation’, Journal of European Public Policy 21 (6): 822-41.

Ripoll Servent, A. (2011) ’Co-decision in the European Parliament: Comparing rationalist and constructivist

explanations of the returns directive’, Journal of Contemporary European Research 7 (1): 3-22.

Ripoll Servent, A. (2012) ’Playing the co-decision game? Rules’ changes and institutional adaptation at the

LIBE committee,’ Journal of European Integration 34 (1): 55-73.

Ripoll Servent, A. (2015) Institutional and Policy Change in the European Parliament: Deciding on

Freedom, Security and Justice, London: Palgrave Macmillan.

Roederer-Rynning, C. and Greenwood, J. (2015) ‘The culture of trilogues’, Journal of European Public

Policy 22 (8): 1148-65.

Roederer-Rynning, C. and Greenwood, J. (2016) ‘The European Parliament as a developing legislature:

coming of age in trilogues’, Journal of European Public Policy, 24 (5): 735-754.

Shackleton, M. (2000) ‘The politics of codecision’, Journal of Common Market Studies 38 (2): 325-42.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Shackleton, M. and Raunio, T. (2003) ‘Codecision since Amsterdam: A laboratory for institutional

innovation and change’, Journal of European Public Policy 10 (2): 171-88.

Stie, A.E. (2010) ‘Co-decision: The Panacea for EU Democracy?’ Arena Report 01/2010, Oslo: Arena.

Stie, A.E. (2013) Democratic Decision-Making in the EU: Technocracy in Disguise, London: Routledge.

Szaprio, M. (2013) The European Commission: A Practical Guide, London: John Harper Publishing.

Toshkov, D. and Rasmussen, A. (2012) ‘Time to decide: The effect of early agreements on legislative

duration in the EU’, European Integration Online Papers 16 (11): 1-20.

Investigation of informal trilogue negotiations since the Lisbon Treaty

Annex A: List of interviewees

1. Richard Corbett, British S&D MEP, April 2017

2. An official from the EP’s Conciliation and Codecision Unit, April 2017

3. Jeppe Tranholm-Mikkelsen, Secretary-General of the Council of the European Union, April 2017

4. An official from the Council Secretariat, April 2017

5. A high-ranking official, Permanent Representation of Denmark, April 2017

6. A high-ranking official, Permanent Representation of France, April 2017

7. A high-ranking official, Permanent Representation of Hungary, April 2017

8. A high-ranking official, Permanent Representation of Poland, April 2017

9. An official, Permanent Representation of The Netherlands, April 2017

10. Two officials, Permanent Representation of an anonymous member state, April 2017

11. Dominique Vandergheynst, Inter-Institutional Relations Group, European Commission, April 2017

12. Two officials from the secretariat of the Committee on Internal Market and Consumer Protection

(IMCO), April 2017

13. An official from the secretariat of the Committee on the Environment, Public Health and Food Safety

(ENVI), April 2017

14. An official from the secretariat of the Committee on Economic and Monetary Affairs (ECON), April

2017

15. An official from the secretariat of the Committee on International Trade (INTA), April 2017

16. An official from the secretariat of the Committee on Industry, Research and Energy (ITRE), April 2017

17. Two officials from the secretariat of the Committee on Legal Affairs (JURI), April 2017

18. An official from the secretariat of the Committee on Civil Liberties, Justice and Home Affairs (LIBE),

April 2017.

Rue Belliard/Belliardstraat 99

1040 Bruxelles/Brussel

BELGIQUE/BELGIË

Published by: “Visits and Publications” Unit

EESC-2017-80-EN

www.eesc.europa.eu

Reproduction is authorised provided the source is acknowledged.

For any use or reproduction of the cover page photo permission must be sought directly from the copyright holder.

European Economic and Social Committee

REG.NO. BE - BXL - 27

QE-01-17-783-EN-C

ISBN 978-92-830-3562-6

doi:10.2864/46932

Online

QE-01-17-783-EN-N

ISBN 978-92-830-3561-9

doi:10.2864/22305