The Art of Lobbying the EU: More Machiavelli in Brussels (revised edition) 9789048517701

Recalling the work of Niccolò Machiavelli, leading researcher Rinus van Schendelen distills the practices of successful

206 80 985KB

English Pages 388 [387] Year 2013

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents Summary
Detailed Contents
Special Preface to the Fourth Edition: Ten Years’ Anniversary of this Book
List of Figures
List of Best Websites
List of Abbreviations
Chapter 1. The Europeanization of Public Affairs
Chapter 2. The Playing-field: EU Common Decision-making
Chapter 3. Pushing the Buttons of ‚Brussels‘
Chapter 4. Getting Grip on an EU Arena
Chapter 5. Managing the Home Front
Chapter 6. Managing the EU Fieldwork
Chapter 7. The Limits of EU Public Affairs Management
Chapter 8. Public Affairs, Lobbying and EU Democracy
References
Index
Recommend Papers

The Art of Lobbying the EU: More Machiavelli in Brussels (revised edition)
 9789048517701

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

The Art of Lobbying the EU

The Art of Lobbying the EU More Machiavelli in Brussels Fourth, fully updated and revised edition

Rinus van Schendelen

Amsterdam University Press

2002 2004 2005 2005 2010 2010 2012 2013

First Edition, titled ‘Machiavelli in Brussels’ (AUP) Czech Translation (Barrister & Principal, Brno) Polish Translation (GWP, Gdansk) Second, Fully Updated Edition, same title (AUP) Third, fully updated and revised edition, new title ‘More Machiavelli in Brussels’ (AUP) Bulgarian Translation (Megachrom, Sofia) German Translation (Lexxion, Berlin) of Germanised and revised third edition Fourth, fully updated and revised edition (AUP), reversed title and subtitle

Cover design: Maedium, Utrecht Lay-out: Het Steen Typografie, Maarssen Library Cataloging-in-Publication Data: 1. Lobbying and Public Affairs Management, 2. European Union, 3. Interest groups. ISBN e-ISBN e-ISBN NUR

978 90 8964 468 8 978 90 4851 770 1 (pdf) 978 90 4851 771 8 (e-Pub) 654

© Rinus van Schendelen / Amsterdam University Press, Amsterdam 2013 All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book.

To my wife, Hedy, my best supporter in life

Contents Summary Special Preface to the Fourth Edition: Ten Years’ Anniversary of this Book

15

List of Figures

25

List of Best Websites

26

List of Abbreviations

28

1 The Europeanization of Public Affairs

33

2 The Playing-Field: EU Common Decision-Making

71

3 Pushing the Buttons of ‘Brussels’

119

4 Getting Grip on an EU Arena

165

5 Managing the Home Front

205

6 Managing the EU Fieldwork

245

7 The Limits of EU Public Affairs Management

285

8 Public Affairs, Lobbying and EU Democracy

325

References

363

Index

383

Detailed Contents Special Preface to the Fourth Edition Ten Years’ Anniversary of this Book List of Figures List of Best Websites List of Abbreviations

15

1. The Europeanization of Public Affairs

33

I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

How are we Living Together in Europe? Solving the Irritating Differences by the EU Method Not Member States, but Member Countries The Domestic Patterns of Public-Private Relationships The Europeanization of the Member Countries Vectors of Europeanization The Fuller Story of Europeanization Influence Challenges from Europeanization Old Influence Techniques and Lobbying 1. The notion of lobbying Lobbying the EU as Public Affairs Management Extra: The Growth of Public Affairs Management in Europe Public Affairs Management Practises at the EU Level

2. The Playing-field: EU Common Decision-making I. II. III.

Deus ex Machina The Skeleton of the EU Machinery: Power 1. The Skeleton’s Main Components 2. The Ever-Changing Skeleton The Flesh-and-Blood of the EU Machinery: Influence 1. The Polycentric and Multifunctional Commission 2. The Commission’s Assistant Layer of Experts 3. Intermezzo: The Commission’s PA and Lobbying by the ‘Brussels Method’ 4. The Parliament’s Influence Role 5. ESC and COR: More Influence than Power 6. The Council: More Power than Influence

25 26 28

33 35 38 40 43 45 49 52 55 57 60 64 67 71 71 73 77 83 87 88 92 95 98 101 103 9

Detailed Contents

IV. V. VI.

7. Some Other Decision-Makers The EU Playing-Field in Helicopter View Extra: Wandering Scholars The EU Playing-Field and Public Affairs Management

3. Pushing the Buttons of ‘Brussels’ I. II.

III. IV.

V.

VI.

Managing the Crucial Variables The Manageable EU Machinery: General Approaches 1. Actors to Approach 2. Factors to Use 3. Vectors to Create The Meta-Game of Triple P Structural Trends of Managing EU Public Affairs 1. From ‘National Co-ordination’ to Self-reliance 2. From Individual to Collective EU Action 3. From Ready-made to Tailor-made Action Extra: Country Profiles of EU PA and Lobbying 1. The Five Old Larger Countries 2. The Ten Old Smaller Countries 3. The Twelve New Central and Eastern European Countries Professional EU Public Affairs Management

4. Getting Grip on an EU Arena I. II. III. IV. 1. 2. 3. 4. V. VI. VII. VIII. IX. 1. 10

Whom to Lobby in an Arena and on What? The Predecessors of Arena Analysis Arena Analysis: Going Window-Out Describing an Arena Stakeholders Issues Time Boundaries Gathering Information for an Arena Description Optimal Rather Than Minimal or Maximal Homework The Quick Arena Scan From Information to Intelligence Pro-Active EU Arena Management: Window-In Short-term EU Arena Management

108 110 113 116 119 119 121 121 124 128 132 136 136 142 147 148 149 154 158 161 165 165 166 169 171 171 173 175 176 177 182 184 187 190 190

Detailed Contents

2. X. XI.

Long-term EU Arena Management Extra: The GMO Food Arena Arena Analysis: Necessary Means, No Sufficient Goal

5. Managing the Home Front I. II. 1. 2. 3. 4. III. IV.

Who Is Lobbying, Why, for What and with What Result? Organizational Assets for Public Affairs Management Sufficient Cohesion Useful Knowledge Optimal Mix of Resources and Skills Good Image Strategy: Why PA Management? Strategy Development: Long-List, Shared-List and ShortList V. Setting the Targets and the Agendas VI. With What Result? VII. The Multinational Organization’s PA in the EU 1. Dossier Level Tasks for the Central PA Desk 2. The Issues of Trust and Consultants VIII. Extra: The Multinational PA as Benchmark IX. From the Home Front to the EU Fieldwork 6. Managing the EU Fieldwork I. II. III. IV. V. VI. 1. 2. 3. VII. 1. VIII. 1. IX. 1.

Lobbying: The Final Link The Surplus of Unorthodox Actions Coping with the Surplus Lobbying Whom? Lobbying on What? How to Style One’s Lobbying Better? Supplying or Demanding Formal or Informal Direct or Indirect The Fine-Tuning of ‘Sound’ Tone Makes the Melody Lobbying Where? Cyber-lobbying Lobbying When? Early, Late and Continual Activities

195 197 203 205 205 206 207 210 211 215 216 221 225 228 231 234 236 238 242 245 245 246 252 255 257 259 259 261 262 266 270 271 273 274 276 11

Detailed Contents

X. XI.

Extra: The Ideal Profile of the EU Public Affairs Expert From Potentials to Limits of EU Public Affairs Management

7. The Limits of EU Public Affairs Management I. II. 1. 2. 3. III. 1. 2. 3. IV. 1. 2. 3. V. 1. 2. 3. VI. 1. 2. 3. VII. 1. 2. 3. VIII.

IX.

12

From Tantalus to SCARE The Sender’s Mental Limits Emotions Dogmas Myths The Sender’s Organizational Limits Dissent Scarcity Lack of Leadership The Limits of Channel Management Power Quality Configuration The Limits of Arena Management Pluralism Complexity Dynamics The Limits on the Receivers’ Side Inattention Satiation Neutralization The Limiting Environment Reputation Codes of Conduct Outside Groups Extra: The Limited National Government 1. Transparency 2. Pressures from Elected Politicians 3. The Dogma of General Interest 4. Solving the Limiting Tragedies 5. Can Capitals Become More Successful? The Limits of PA Management in the EU: Problem or Blessing?

278 282 285 285 286 287 288 290 292 292 293 295 296 298 299 300 301 302 303 304 305 305 306 307 309 309 311 313 315 316 317 318 319 322 324

Detailed Contents

8. Public Affairs, Lobbying and EU Democracy I. II. 1. 2. 3. 4. III. 1. 2. 3. 4. IV. 1. 2. 3. V. VI.

Democracy as a Criterion Notions of Democracy in Europe Input Notions Throughput Notions Output Notions Feedback Notions The Impacts of EU Interest Groups on EU Democracy Input Impacts Throughput Impacts Output Impacts Feedback Impacts Improving EU Democracy by EU PAM and Lobbying Strengthening the Positive Impacts Reducing the Negative Impacts Reinforcing the System-bound Correction Mechanisms Extra: Improving EU Democracy by the Study of EU PAM Final Reflection before the Action

325 325 327 328 331 333 335 337 338 340 343 345 348 349 350 351 354 358

References

363

Index

383

13

Special Preface to the Fourth Edition Ten Years’ Anniversary of this Book A synthesized navigator’s guide This is not a typical book about interest groups trying to influence public authorities like civil servants and politicians in general and those in the European Union (EU) in particular. More books and articles address the general question (or part of it) and even discuss its application to the EU. In each, one or more authors present their own views and/or findings and sometimes they compare theirs with those of others. The synthesis of all the different little bits of assumed knowledge they leave to the reader, however. By its ambition to create this synthesis, this book stands out as a navigator that can help the interested reader to gain a better understanding of the many pitfalls of attempting to influence public authorities and other stakeholders in general and those at the EU level in particular – in short, to increase their chances of success. The author, privileged by over forty years of both academic study and applied practise in this field, has simply collected here ‘everything that anybody might know if having had the same privileges’.

The backgrounds of the book On substance the book has two objectives. The first and general one is to synthesize how interest groups can influence public authorities and other stakeholders on what they plan to do (or not to do). As such, the book is about, in popular parlance, the art of lobbying in any arena. Our notion of lobbying refers to the mediaeval practise of entering the lobia or corridor of a prince or principal and has nothing to do with the negative meaning it has acquired in some corners of the mass media. Today, most interest groups and not least those from governments use it as a synonym for ‘making a difference as desired’ or getting influence. If done intelligently, lobbying is part of a wider body of knowledge called public affairs management (PAM, shortly PA). Its main ingredients are: the ambition to win, the study of the arena and the prudency of action. Embodying these three qualities that enhance lobbying as an art is Niccolò Machiavelli, the advisor to the ruler of Florence in the early 16th century and demonized by English moralists in the 18th century as devilish ‘old Nick’. The book’s second objective is specific, as general knowledge can only prove its value when applied to a particular setting. Here we apply it to the 15

Special Preface to the Fourth Edition

decision processes of the European Union, which annually result in about 2,500 new laws and acts that bind the member states and overrule their domestic ones if incompatible with these. Hence, its main seat at Brussels can now be called the common capital of all who belong to the EU. Not surprisingly, all sorts of interest groups want to win (or at least not lose) their interests in the EU and many now move from the domestic to the EU arenas to achieve that goal. Even national governments being unable to rely on only their few formal powers, have to (learn to) fight for their interests and to behave like interest groups. The two objectives match in our ambition to show that the EU can be influenced successfully. This ambition has, of course, nothing to do with being in favour or against europeanization. Knowledge takes no sides and can serve both camps. Influencing successfully is largely a matter of PAM expertise that belongs to the few with talents and skills. Nature distributes talents, and we cannot change that, but by bringing together here the insights of researchers and practitioners in this field of PAM in the EU, we offer many more people the chance to improve their skills. Hopefully they will take this chance, as otherwise they leave the room for influencing the EU to others and have to accept the outcomes of their greater skills in PAM and lobbying. We owe the most to three sources of knowledge. One is political science, the mother of all studies of influence and focused on how influence processes work or do not work. The second is management science, with its focus on how these processes can be managed more successfully. The third source is that of the best practitioners of PAM in the EU as they, eager to improve their performances, collect useful information and experiences that may contribute to better knowledge for all in the long run.

The changing insights on the art of lobbying the EU from 2002 to 2012 The first three editions of this book since 2002 have gathered a wide readership, which is also indicated by the many reprints and the four translations. This fourth edition has again been updated and revised for two permanent reasons. One is that the three sources of knowledge produce a steady flow of useful new insights on how decision processes and particularly those of the EU really work at the moment and can be influenced better by the expertise of PAM. The second reason is that the EU machinery of decision-making is ever-changing. Any textbook on either PAM or the EU needs revision after a few years, the more so if it is about both PAM and the EU and most of all if it wants to serve as an up-to-date navigator’s guide. 16

Special Preface to the Fourth Edition

On the occasion of this new edition, it is appetizing to sketch in broad lines for the years 2002 to 2012 some of the most striking changes of the EU playing-field, the leading practises of influencing it and the main insights from the three sources of knowledge, which – as they stand in 2012 – are all described in detail in this book. The logic behind this comparison is that influence success always depends on the specific situation of the playing-field, the best possible practises of PAM and the most useful knowledge. Trying to influence the EU in 2012 by using the ‘state of the art’ tools and techniques of 2002 is almost guaranteed to be a failure. So what are major changes?

1 Changes of the EU playing-field – More EU competences. The EU is basically a construction made by member states that by the way of treaty formation transfer or pool specific national competences to its machinery of more or less common and binding decision-making. In 2002 the EU operated under the 1999 Treaty of Amsterdam. The 2003 Treaty of Nice that had to settle its leftovers was in the making, the (rejected) 2005 Constitution for Europe was still far away and nobody could imagine the 2009 Treaty of Lisbon, let alone the 2012 debates about a Political Union Treaty on enabling the EU to cope better with the various post-2007 crises. In 2002 the legislative programme to make the internal market more open was ten years young, the euro had just started and the number of member states, largely located in western Europe, was fifteen. The EU’s relatively strongest competences were in the policy fields on markets, such as competition, open market, agriculture, industry and environment, on which the European Commission acted as the engine of proposals that after adoption bind the member states. Sensitive policy areas, such as on Justice, Home Affairs, Security and Foreign Affairs, fell under a special regime of only the Council of Ministers that largely decided by unanimity. On only a limited number of market areas the European Parliament (EP) had powers of codecision, together with and equal to the Council. The Court of Justice had relatively limited competences too. By 2012, within ten years, the EU’s powers on market areas are expanded and intensified and on the sensitive areas fully (Justice, Home Affairs) or to some extent (Security, Foreign Affairs) brought under the ordinary regime that starts with the Commission. The EP has got more codecision powers, the Court more jurisdiction and the Council’s unanimity voting has become the exception rather than the rule, by which now usually every member state can be overruled. Also due to the post2007 banking crisis and its spill-over to issues of financial and euro sta17

Special Preface to the Fourth Edition

bility, government budgets and effective governance, the EU Council of ‘heads of state’ has got a legal position. It is commissioned to develop by unanimity (non-binding) proposals for common solutions to these crises. It does so by stretching the current treaty and by drafting new treaties that bring formerly national competences gradually under a more common EU regime and that, after ratification by the member states, empower the Commission to draft legislative proposals. By acting as a temporary service-hatch for the rapid europeanization of national competences and policy domains, it makes that ever more interest groups move their focus and actions from the national to the EU capital, being Brussels. Here they usually find the Commission, EP and Court to be more relevant than the Council. – From EU-15 to EU-27. In 2002 the EU had 15 member states, all with their own representatives (Council, EP) or nominees (Commission, Court). In 2012, after the 2005 and 2007 enlargements, it has 27 member states (and more in wait). These enlargements have had modest impacts on the Commission that mainly had to create vacancies for people from the new states and to anticipate the more varied agendas of 27 instead of 15 countries. The EP expanded from 626 to 754 members, coming from hundreds of national parties, and it organized itself even tighter in order to transform its intense diversity into effective majority positions. The Court got twelve judges more and staff from the new states, all with different legal traditions and styles. On the Council the enlargements got the greatest impact. With 27 national ministers around the table, any discussion of only four minutes per minister would easily take hours. Except for the EU Council, the old style of broad consensus formation was increasingly replaced by counting (with the help of a calculator) the Council members’ voting points that deliver a sufficient majority. To escape blunt counting, more cases (‘dossiers’) of decision-making are now prepared at its lower levels of national civil servants that deliver ready-made majority positions for rubber-stamping and, at the level of ministers, by ever more informal meetings and flexible coalitions (‘block voting’). Interest groups now have to cope with an EU-27 that contains many more diverse interests and stakeholders and particularly changed the Council. – Domestic politics. In 2002 and in hindsight, domestic politics and European politics were two relatively different arenas but less than in the EU’s older past, when the EU had still limited competences, the national governments usually had fairly stable positions and most citizens (except those in the UK) showed benevolent indifference for EU integration that they entrusted to their politicians and interest groups. By 2012 much has changed. Sensitive policy domains like immigration, 18

Special Preface to the Fourth Edition

safety and privacy are now in a fast process of europeanization. The post-2007 crises have leveraged EU agendas and measures on the euro, national budget, financial institutions and more. Also due to domestic causes, most national governments are in less stable positions. Due to the more expanded and intensified europeanization of policies, many more citizens in all countries now have an attitude of ambivalent criticism. The 2005 Constitution Treaty was rejected by referendums in France and the Netherlands, two EU founders. Increasingly, EU integration has become a hot subject of domestic debate, being, paradoxically, an indicator of growing integration of citizens in the EU. This domestication of EU issues brings many a national government into a cleavage between its European and its domestic arena and weakens its position in both. Interest groups now have to be more alert on domestic arenas too, which can give a tailwind or a headwind to their interests in EU, and must invest in the friendly europeanization of their domestic interests. – Daily functioning. The treaty changes, enlargements and impacts from domestic politics have changed the daily functioning of the EU. Between 2002 and 2012 the EU got much more relevance, complexity, dynamics and openness. In 2012 hardly a domestic market sector or policy domain is left free from EU interventions that may be liked or disliked and thus are relevant. In 2002 the hottest EU debates were about the plan for the Constitution Treaty, dates for the enlargements, border control, food safety (‘Frankenstein’ foods) and, as usual, the annual budget. These dossiers are much different from the 2012 crises ones that are low in number but high in relevance and managed by the new and complex troika of EU Council, ECB and IMF, with the Commission as secretariat of the EU Council and as the implementing institution thereafter. The crises add to the dynamics of the EU Council, not least because the heads of state act under pressure of their domestic politics. At any summit any decision may look insufficient, but the many summits since 2011 make a big leap. For managing the crises, the EU Council stretched the EU’s competences under the Lisbon Treaty, drafted two treaties for giving the EU more competences and discusses an overarching treaty for a Political Union. Usually being ‘the oyster of Brussels’, it even caused willy-nilly more openness, as every head of state accounts for its doings in its national parliament and mass media in order to address the concerns of the citizens. Because no two countries show exactly the same account of what is happening in Brussels, they all together reveal a great deal of inside information. Interest groups standing at a distance try to influence the EU Council via the domestic politics of one and increasingly several countries. 19

Special Preface to the Fourth Edition

By far the largest majority of EU dossiers, however, is not crisis-like and, as usual, only more or less controversial. Many are, however, processed differently in 2012 than in 2002. The Commission, still a very small bureaucracy but most intelligent by its insourcing of interest groups from the member countries, now often applies impact assessments to legislative proposals and online consultations for gathering earlier responses from organizations and citizens. For both its implementing and delegated acts, which outnumber EU laws, it has simplified the procedures. By the new EU referendum citizens can now put an issue on its agenda. The EP has got expanded powers of codecision and more diversity of members. It tightened its internal workings and behaves now as an institutional pressure group, often in coalition with the Commission and against a hostile Council that now functions more at its lower layers, in informal meetings and by flexible coalitions inside. Thanks to the crises dossiers, many non-crisis ones now run rather smoothly through the main institutions, as the former ones take much energy (time, staff, attention) from the higher levels, including those in the national capitals. Lower levels inside and other interest groups enjoy more room for influencing the EU.

2 Changes of Influence Practises – Quantity and quality of interest groups. In Berlin, the capital of the largest member state Germany, the number of interest groups that compete on whatever dossier may easily amount to about thirty, but in the EU capital of Brussels it is easily ten-fold higher, so making the EU an extremely competitive playing-field. A reliable number of interest groups in Brussels can, however, not be given. For the year 2002 the mass media claimed that there were 15,000 lobbyists and a private directory listed 2,600 entities or groups. The EP and Commission started a common register of (only) entities in 2008. Its 2012 total is above 5,000 entities that are mainly based in the Brussels area and this does not include the entities from governments (free from registration) and the countless groups that go to Brussels in an ad hoc manner. The numbers indicate that PA and lobbying are widely accepted by both interest groups and institutions in the EU. The Council (the oyster) has never published any figures. The PA quality of the entities and their staff as measured by their prepared ambition, studiousness and prudence and their final performances of success, reputation and licence to operate, has always been highly variable. Like in all free sports, only a small minority of them belongs to the ‘premier quality league’ that has an ever-changing composition. Due to the strong competition among the 20

Special Preface to the Fourth Edition

many more interest groups and the increased social control by competitors and mass media, this small minority has in 2012 the reputation to surpass their 2002 colleagues on the PA quality indicators. Many other interest groups take them as an ‘example to follow’. – European Federations. In 2002 many interest groups could enter the buildings of the EU on their own. With a few exceptions, this practise is over and in 2012 it has been replaced by the need for collective organization and action. One impulse came from the Commission that now wants to receive more aggregated interests and another from the interest groups that smell so many relevant EU dossiers that they see no alternative for sharing their interests with other stakeholders having similar interests. Their standard solution is to join or to make a European Federation (EuroFed), permanent or ad hoc. In 2002 more than 700 permanent EuroFeds were counted and in 2012 more than thousand. In addition there are now countless ad hoc EuroFed-like networks. They all make the individual interests of their members more collective, so acquiring for their platform a better entrance at the Commission and EP. Their cross-national composition allows them to fine-tune (by passport and tongue) their actions inside these multi-national institutions. The ad hoc networks often have also a cross-sectoral composition, for example of companies, NGOs and regional governments, which increases their attractiveness to the EP and the Commission due to their broader aggregation of interests. Only the national ministries (‘capitals’) of the member states hardly have a EuroFed, but some are now starting to follow this example. – Multi-level influence practises. In 2002 the many interest groups acting on their own mainly took the direct route to Brussels. By 2012 they discovered that the shortest route is usually not the most effective one. At home they now take the national route for collecting befriended stakeholders that are connected to other countries and EuroFed(s). On the transnational route via other countries and the EU they then collect stakeholders there and establish a common ad hoc network or join a suitable EuroFed that can mobilize support in the various countries and provide an influential position at the EU institutions. Through this common platform they can also take the international route to bodies like the WTO or OECD and foreign countries like the United States or Japan, as these bodies or countries can be influential on their EU dossiers too. – Shared-listing for 2E = MI2. In 2002 ‘the short-list’ was the catchword for the list of EU dossiers to which an interest group attached the highest priority for its influence activities in the EU. By 2012 the professional interest groups (by quality) found that their number of priority dossiers is really a long-list that is impossible to manage or at only very high cost 21

Special Preface to the Fourth Edition

in time, manpower and budget. The new catchword is now the ‘sharedlist’, which still contains priority dossiers but can be influenced through platforms of stakeholders in the EU that share the priority and pool their resources and PA actions to achieve them. The ‘short-list’ has got the new meaning of priority dossiers without such a platform, which thus have to be influenced single-handedly. Shared-listing sharply decreases the costs of securing both the critical mass (M) of supportive stakeholders and the information needed for intelligence (I2) and sharply increases both the efficiency and effectiveness (2E) of PA. By the logic of 2E = MI2, the new trend among the professional groups is to make the shared-list the longest possible and the short-list the shortest possible. To achieve their different priorities they operate through various platforms, including EuroFeds and networks. – Professionalization. The changes in both the EU playing-field and the influence practises require ever more professionalization. In 2002 very few interest groups were aware of how difficult it is to come to intelligent decisions regarding, for example, MI2 acquisition, strategy or routes, influence tactics and techniques, use of internet and social media or making the home organization fit for PA the EU. In 2012, the groups that belong to the ‘premier league’ see such difficulties and respond to them seriously in order to reach better decisions about which influence actions to take. They have reduced the ‘mediaeval lobbying’ to only a small part of PA. By their (ascribed) successes they encourage other groups to follow their methods and so they spread their professionalism wider. One example is the increased respect for the limits of PA and its role in EU democracy. Another is the now more widely accepted notion that the home organization, where every PA starts and results in profits or losses, needs PA expertise with a PA desk that functions as a flexible intervention force. The theme of professionalization that embeds any lobbying in PA methodology and so enhances it to the level of an art stands central in this book.

3 Changes of Knowledge Sources – EU studies. In 2002 the state of EU studies was widely criticized for its over-supply of different and vague grand theories and big concepts, often having a normative bias, taking the EU as an intergovernmental playing-field and hardly based on facts. Only a few, mainly younger scholars took a mid-level approach with an empirical search for facts. In 2012 the grand theories and big concepts have almost faded away and the mid-level approach has produced a large and growing amount of empirical research with detailed insights into parts of EU decision-mak22

Special Preface to the Fourth Edition

ing. Departing from mid-level theories and concepts, this research is largely deductive by its testing of hypotheses on collections of mainly official documents, records and activities related to some policy sector or cases. Field research, such as by participant observation or interviewing that may give inductive rather than deductive insights, is a scarcity. Beaten roads are researched in rich detail but discoveries of perhaps more relevant side-paths are hardly made. By more inductive research the added value of EU studies to the study and practise of PA might greatly increase. – PA studies. In 2002 only few researchers with both political and management science backgrounds were active in the study of PA and even fewer were focused on the EU. Academic courses on PA were scarce and, if they existed, were usually modest in size or were elective; those on PA in the EU were even scarcer. By 2012 this has slightly improved. The major events have been the 2002 launch of the international Journal of Public Affairs and the Handbook of Public Affairs [Harris and Fleisher, 2005], both the first in the field. Their articles come from both academics and practitioners, but only few are related to the EU. In a few countries the courses are growing by number but they are still modest in size or status and seldom provide a focus on the EU. If the latter, the flag of PA often covers a lot of cargo that is useful to know but not the basics of PA in the EU, such as mass communications, ‘EU institutions, procedures and policies’ and old-styled lobbying (without PA). This limited focus on the EU is different from PA studies for national capitals like London, Berlin and Washington, D.C. By focusing more on the EU, PA studies might add greater value to the study and practise of the EU. – PA practitioners in the EU. In 2002 only few practitioners in a few countries and in Brussels held closed meetings with each other to exchange experiences, findings and questions and/or to discuss their trials (and errors) and common interests. Besides, they attended open conferences with speakers on PA in the EU. They hardly consumed the academic products in the fields of both EU and PA studies, let alone made use of them. In 2012 these circles still exist but the open conferences have largely been replaced by closed and commercialized trainings, often of dubious quality and attended by mainly young people seeking jobs in interest groups. A few polling firms now measure and publish the opinions of EU officials and PA practitioners on issues of PA in the EU. – Win-win opportunity. In both 2002 and 2012 the three worlds of EU studies, PA studies and PA practitioners in the EU are still highly separated. Only few people bridge the gaps between the three. The practitioners are hungry for easily applicable results, checklists and prescripts that hardly seduce the academics. The latter seldom make connections 23

Special Preface to the Fourth Edition

between the EU and PA and have, within their own discipline, the strongest appetite for ‘sound’ knowledge according to their theories, concepts and research, and hardly for the raw material of fresh information from the EU playing-field that the practitioners have. They miss the win-win opportunity that could greatly enhance all three sources of knowledge. By synthesizing their segmented pieces of knowledge this book aims to stimulate the win-win for all.

Final remarks The appetizer above shows that there have been major changes in the playing-field, the influence practises and the knowledge sources over the past few years. It explains the need for another new edition of this book, subtitled as ‘more’ Machiavelli in Brussels. A synthesized navigator’s guide that is not up-to-date would mislead the reader or user and thus be useless. The improved or new insights into the changes of both the playing-field and the influence practises we owe to the three knowledge sources. They all have their own strengths and weaknesses. Their common weakness is that they are hardly tapping or inspiring the other sources. My own roles in all three domains explain my personal ambition to synthesize the three sources. As such, the book stands alone in the field. My dream remains that many skilled people in these three worlds will bridge the gaps between them. The win-win at stake is that the academics in EU and PA studies benefit from both each other and the practitioners and as well the latter from the former by exchanging ever more useful information in return for ever more useful research and teaching on PA in the EU. Like in previous editions, we only refer to literature and documents in English, this being the common language in both the practise and the study of the EU. The authors, however, come increasingly from many countries, which indicates a widened awareness that the EU can really be influenced and that its decisions result from ‘what all of us cleverly make of it’. More than others, the best practitioners of PAM at the EU level have promoted this awareness. They always inspire me by their openness to inform me and their eagerness to improve their knowledge in return. To them I dedicate this new edition. Rotterdam, January 2013

Note: Professions are mentioned in the masculine form, but should be read as referring to the feminine form too. 24

List of Figures 1.1 1.2 2.1 2.2 2.3 3.1 3.2 3.3 4.1 4.2 4.3 4.4 5.1 5.2 5.3 6.1 6.2 6.3 7.1 7.2 7.3 8.1 8.2

Vectors of Europeanization Public Affairs (PA) Management Main Decision-making Flows between EU and Member Countries Adopted Binding Decisions by Procedure and Type, 1970-2011 Official Key Dates of Main EU Policy Formation The Meta-Game of Triple P Interest Groups and Routes to EU Professional Public Affairs Management/ Lobbying Getting an Issue on (or off) the EU Agenda Major Sources and Techniques of Arena Monitoring Dossier Homework The Better Tactics of Arena Management The Challenges of Public Affairs Management Downsizing the Long-List The Multinational Model Major Dilemmas in Managing the Fieldwork EU Examples of Up-Framing Examples of Direct and Indirect EU Lobbying Ten Popular EU Myths Variety of Channels for Public Affairs Management Code of Conduct 2012 Popular Notions of Democracy Self-Correcting EU Lobby-Democracy

46 61 76 81 84 132 140 162 169 180 185 192 219 223 233 247 251 263 291 297 312 331 354

25

List of Best Websites On European Union WWW.EUROPA.EU The most important EU site. WWW.ECPRD.ORG and WWW.IPEX.EU Sites of parliaments on the EU. WWW.VOTEWATCH.EU Independent site devoted to EP vote statistics. WWW.THEPARLIAMENT.COM News bulletin on the European Parliament. WWW.AMICURIA.EU and WWW.CASELEX.COM Sites on the Courts’ cases. WWW.EURACTIV.COM and WWW.EUOBSERVER.COM Information sites. WWW.AGENCEEUROPE.COM Daily news bulletin. WWW.PUBLICATIONS.EU Site on EU publications. WWW.EUROPEANVOICE.COM Weekly newspaper on the EU. WWW.NEUROPE.EU Weekly newspaper on particularly EU countries. WWW.EUREPORTER.CO.UK Weekly newspaper on European countries. WWW.EUFEEDS.EU and WWW.PRESSEUROP.EU Dailies on the EU. WWW.EUROPEANBUSINESSREVIEW.EU Business review on the EU. WWW.EUROPEANAGENDA.EU Daily EU news on people and events. WWW.THEBRUSSELSCONNECTION.BE Site on Brussels’ EU quarter.

On Public Affairs Management WWW.DODS.EU and WWW.STAKEHOLDER.EU The two leading directories. WWW.EUROBRUSSELS.COM The European Affairs jobs site. WWW.SEAP.BE and WWW.EPACA.ORG Brussels-based PA networks. WWW.DEGEPOL.DE and WWW.BDP-NET.DE Two German PA-related sites. WWW.THEECPA.EU and WWW.PUBAFFAIRS.ORG UK sites on PA. WWW.AFCL.NET The site of French PA consultancies. WWW.ALPAC.AT and WWW.BVPA.NL Austrian and Dutch PA sites. WWW.AESCOP.COM and WWW.ILCHIOSTRO.ORG Spanish and Italian sites. WWW.PAC.ORG and WWW.ALLDC.ORG Two US sites on PA. WWW.PUBLICAFFAIRSWORLD.COM PA services site.

On Academic Studies WWW.SGEU-ECPR.ORG The ECPR Political Science Research Group on EU. WWW.UACES.ORG UK-based association of scholars on the EU. WWW.EUSTUDIES.ORG US-based association of scholars on the EU (EUSA).

26

List of Best Websites

WWW.ELPRG.EU The European Legislative Politics Research Group. WWW.CEEOL.COM Central and Eastern European Online Library. WWW.PALGRAVE-JOURNALS.COM/EPS EU studies and teaching online.

27

List of Abbreviations ACEA ACP AEA AEJ AEM AMCHAM BEUC BINGO BONGO CASTER CATS CCMC CEA CEEC CEFIC CEMR CIAA CoA COM CONECCS COR COREPER DG EACEM EACF EBA EBF ECB ECF ECJ ECOBP ECOFIN ECPA ECSC EDA EEB

28

European Automobile Manufacturers Association African, Caribbean, Pacific Association of European Airlines Association of European Journalists Association of Mountain Areas American Chamber of Commerce European Consumers Organization Business Interested NGO Business Organized NGO Conference and Association of Steel Territories Committee Article Thirty-Six Committee of Common Market Automobile Constructors European Committee of Insurers (2012: InsuranceEurope) Central and Eastern European Countries European Chemical Industry Council Council of European Municipalities and Regions Confederation of EU Food and Drink Industries (2011: FoodDrinkEurope [FDE]) Court of Auditors European Commission Commission’s Consultation of Civil Society (EU site) Committee of the Regions Committee of Permanent Representatives (-tions) Directorate-General European Association of Consumer Electronics Manufacturers (now EICTA) European Airlines Consumer Forum European Banking Authority European Banking Federation European Central Bank European Construction Forum EU Court of Justice European Campaign on Biotechnology Patents Economic and Financial Council European Centre for Public Affairs European Coal and Steel Community European Defence Agency European Environmental Bureau

List of Abbreviations

EFC EFPIA EFSA EIB EICTA EIRA ELTAC EMF/FEM EMU END EP EPEE EPHA ERT ESC ETI ETUC EU EuroFed FAC FDE FELPA FIM GAC GINGO GONGO GMO GS HDTV IGC MNC MNGO MNO MEP MP NATO NGO OC OCM OECD OLP PA(M)

Economic and Financial Committee European Associations of Pharmaceutical Industry European Food Safety Authority European Investment Bank European Information & Communication Technology Industry European Industrial Regions Associations European Largest Textile and Apparel Companies (now EuraTex) European Metal Workers European Monetary Union Seconded National Expert (Expert national détaché) European Parliament European Partnership for Energy and Environment European Public Health Alliance European Round Table of Industrialists Economic and Social Committee European Transparency Initiative European Trade Union Confederation European Union European Federation Foreign Affairs (and Security) Council European Food and Drink Industries (formerly CIAA) European Federation of Lobbying and Public Affairs International Motorcyclist Federation General Affairs Council Government Interested NGO Government Organized NGO Genetically Modified Organism General Secretariat (Council) High-Definition Television Intergovernmental Conference Multinational Company Multinational NGO Multinational Organization Member of the European Parliament Member of (National) Parliament North Atlantic Treaty Organization Non-governmental Organization Online Consultation Open Coordination Method Organization of Economic Co-operation and Development Ordinary Legislative Procedure Public Affairs (Management)

29

List of Abbreviations

PGEU PR PSC PURPLE QMV SCoFCAH SEA SEAP SEM SG SGAE SLP SME TEU TFEU UEAPME UNICE VLEVA WEAG WHO WTO

Pharmaceutical Group of the EU Permanent Representative (Representation) Political and Security Committee Peri-Urban Regions Platform Europe Qualified Majority Voting Standing Committee on Food Chain and Animal Health Single European Act Society of European Affairs Professionals Single European Market Secretariat-General (Commission) French EU co-ordination centre (formerly SGCI) Special Legislative Procedure Small and Medium-sized Enterprise Treaty on the European Union (Lisbon, 2009, part I) Treaty on the Functioning of the EU (Lisbon, 2009, part II) European Association of SMEs Union of Employers’ Organizations (now BusinessEurope) Flemish Public-Private PA desk in Brussels Western European Armament Group (NATO) World Health Organization World Trade Organization

Commission DGs/Services (2012, except few internal offices) AGRI BEPA BUDG CLIMA COMM COMP CONNECT DEVCO DGT DIGIT EAC ECFIN ECHO EDPO ELARG EMPL

30

Agriculture and Rural Development Bureau of European Policy Advisors Budget Climate action Communication Competition Communications Networks, Content and Technology (formerly INFSO) Development & Cooperation (Third World) Translation Service Informatics Education and Culture Economic and Financial Affairs Humanitarian Aid European Data Protection Officer Enlargement Employment, Social Affairs and Inclusion

List of Abbreviations

ENER ENTR ENVI EPSO ESTAT FPIS HOME HRM IAS JRC JUST MARE MARKT MOVE OLAF OP REGIO RTD SANCO SCIC SG SJ TAXUD TRADE

Energy Enterprise and Industry Environment European Personnel Selection Office Statistical Office Foreign Policy Instruments Service (formerly RELEX) Home Affairs Human Resources and Security (Personnel) Internal Audit Service Joint Research Centre Justice Maritime Affairs and Fisheries Internal Market and Services Mobility and Transport European Anti-Fraud Office Publications Office Regional Policy Research and Innovation Health and Consumer Affairs Interpretation Service Secretariat General Legal Service Taxation and Customs Union External Trade

Member States AT BE BG CY CZ DK DE EE EL ES FI FR HU IE

Austria Belgium Bulgaria Cyprus Republic Czech Republic Denmark Germany Estonia Greece Spain Finland France Hungary Ireland

31

List of Abbreviations

IT LV LT LU MT NL PL PT RO SE SI SK UK

32

Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Sweden Slovenia Slovakia United Kingdom

Chapter 1 The Europeanization of Public Affairs If wanting to launch a campaign in foreign terrain successfully, one should first study the terrain’s history, geography, people and anything else that may determine success. This basic belief of Machiavelli in his Discorsi (book 3-8) is applied to the EU here. If wanting to influence EU, what should one know beforehand?

I How are we Living Together in Europe? ‘Europe’ is almost a synonym for variety or diversity. In the mid-2010s, there were almost 500 million people living in the European Union (EU), which is more than the 420 million people of the United States and Japan combined. The negotiations about entry into the Union with nine states more, from the remnants of the former Yugoslavia to Iceland and Turkey, may add more than 100 million people to the EU. All these countries have a great deal of internal diversity by ideology, ethnicity, religion, education, language, income, culture and much more [Calder and Ceva, 2011]. Their states are frequently subdivided into more or less autonomous regional, provincial and local governments that reflect different territorial and power ambitions. Their histories often have a long record of both civil wars and foreign ones with neighbours and (former) colonies. The domestic political ideologies range from extreme left to right and their religious belief-systems from Catholicism, Protestantism (in many variants), Islam and many more exotic ones, plus a substantial proportion of people who are atheists or agnostics. In their daily lives, Europeans, living in free and pluralist societies, encounter a kaleidoscope of civil organizations such as trade unions, companies, political parties, churches and other interest groups. Their economies show much diversity such as by sector, productivity, employment and regimes that range from more free-market in the north to more state-directed ones in the south (the Mediterranean) and the east (Central Europe). For communication the people of the EU rely on twenty-three different ‘official national languages’ (acknowledged as such by the EU) and hundreds of regional ones. In France alone, one-third of the population uses one of its eight recognized dialects. In terms of socio-economic statistics, the variety in Europe is even more impressive [COM, Europe in figures]. The population size of the member state of Luxembourg is less than half a per cent of that of Germany, but 33

The Europeanization of Public Affairs

Chapter 1

has the highest GDP per capita, two and a half times that of Portugal and five times that of Latvia. In general, there exist sharp contrasts in wealth between the relatively rich north-western and the poor eastern part of the EU, on a regional level with a ratio as high as ten to one, with the southern part (including Ireland) roughly at average. Rates of labour activity and unemployment vary widely across member states and regions. Employment rates for economic sectors show a mountainous landscape, with Greece and Poland peaking in agriculture (about 20% of the labour force), Germany in industry (about 40%) and the Netherlands in services (about 70%). Social security in Europe is a volatile variable, mainly due to the differences in domestic government programmes for income maintenance, health, housing and education. This description of social variety could be extended to a range of other variables, from urbanization rates to leisure patterns and from consumer behaviour to environmental interests. Perhaps the most important but the least documented, is cultural variety [Halman and others, 2011; www.europeanvaluesstudy.eu]. In France, Belgium and Spain the dominant norms and values of social behaviour are relatively hierarchical or top-down, even inside small and decentralized units like family systems and sectoral networks. The new Central European member states, coming from a communist past, largely reflect the same pattern but show growing attachments to bottom-up networks. In Scandinavian countries and the Netherlands, national culture is more bottom-up, based on self-reliant networks. Portugal, southern Italy and Greece have a culture based mainly on family values, with elderly leadership and clearly defined loyalties. In the United Kingdom, culture is more centred on the value of free private life, as well as on the market, but limited by the rule of law. In Germany and Austria, orderly and predictable behaviour is a dominant value. These brief generalizations of national cultures deserve, of course, a more detailed examination, but for the moment they suffice to underline the cultural variety found across the EU. The cultural differences help to explain why the peoples of Europe have so little faith in each other. Except for the Italians, the Swedes and the Belgians, they put most trust in their own countrymen, less in the northern peoples and least in the southern ones, as shown in an official survey held once and never again [COM, Eurobarometer, 1996, B 46-7]. In the much enlarged EU-27, trust in other nations is assumed to be even lower, so resulting in low cohesion benefits and high integration costs [Delhey, 2007]. Not least, there is a great deal of political variety [Magone, 2011]. Some countries once had a great empire inside Europe, for example Austria, Denmark, Lithuania, Poland and Spain. Others, such as France, Britain, Portugal, the Netherlands and Belgium, built empires with colonies outside Europe. Many countries, such as Bulgaria, Hungary and Greece, are 34

Chapter 1

The Europeanization of Public Affairs

very old states, while others, such as Belgium, Germany and Italy, are around 150 years young. Slovenia exists since 1991. Many countries have undergone long-term occupation by some neighbour (Ireland and the Central European states) or internal fragmentation (Italy, Germany). All possess a democratically elected government now, but with strong variations of state capacity and civil support [Sissenich, 2010]. France and Britain have a centralist state tradition, as still reflected by the domestic position of their government leader. Others, like Germany, Spain, Belgium and Austria, have decentralized governments, which is also on the rise in Italy and particularly in many Central European states, where this existed before 1939, the start of German and subsequently Soviet occupation. Due to their socio-economic and cultural differences, all governments give expression to very different political interests and preferences. All this external variety is, of course, not unique to Europe. It also exists in other large geographical areas, such as China, Brazil and the United States [Warleigh and Van Langenhove, 2010]. But their variety is already brought together into the larger whole of their conglomerate state and only has to be kept together now. The people of Europe do not have a common identity, let alone belong to a single society [Favell and Guiraudon, 2011; Wallace and Stromsnes, 2008]. They have many different identities instead, which are now in a process of integration and create an instrumental (‘what is in it for us?’) rather than intrinsic sense of common identity [Cram, 2012]. In the EU, their interest groups compete with each other in much higher quantity and diversity than at any national level. Europe’s primary management challenge is to bring the different parts together into a larger whole; keeping it together is its second challenge. The differences that cause cross-border irritations must be solved first. Examples out of countless many are the different taxation rules, subsidy practises or civil rights among the countries. Of course, many differences are not considered irritating at all. Most Europeans appreciate the varieties of music, cuisine or architecture as a non-irritating enrichment. The integration problem lies in the differences that do cause irritation. These can be anything, and may even include the way music is produced, the hygiene of the cuisine or the safety of the architecture. The management problem here is: how can the irritating differences be solved?

II Solving the Irritating Differences by the EU Method Take, for example, the gas hose. By far the best type of gas hose, according to the French, is their ‘Dormant’: it has stainless steel helical tubing, is moulded from a continuous spiral, with flare-type seals at the end and has 35

The Europeanization of Public Affairs

Chapter 1

no covering. But the British believe they have a better-than-best gas hose: one with galvanized metal annular tubing, formed into concentric circles, and with a rubber covering. The real best one is the Italian, as the Italians believe: stainless steel annular tubing, extendable and free of covering. The German, Swedish, Spanish among others have their own types of gas hose and feel it is the best. The gas hose is just one example of variety in Europe and numerous others exist, ranging from products and services to values and ideas. So far the many varieties of gas hose across the EU do not cause any irritation. It only represents a difference and this might continue to be the case for another fifty years. Some producers may even appreciate the difference, as this provides effective protection against imports from abroad. However, other interest groups may feel irritated and start to politicize the difference. Some producers may want to export their gas hose, to create economies of scale and to become a market leader in Europe. Consumer groups may demand a cheaper and standardized gas hose, health groups a safer one, and ‘green’ groups one that is more eco-friendly. Governments may want to save on inefficiencies. An irritation is born and may grow into a conflict. What are the potential solutions? The following five traditional methods exist in Europe: 1 Resignation. The costs of resignation may be considered to be less than those of any other solution, thereby justifying living with an irritation for longer, just as those travelling across borders today must be resigned to the irritation of dealing with different electricity plugs. Some may even count the blessing of a difference or accept it as a minor element on the balance sheet. Resignation may work for a while, as in the case of educational diplomas that have different attributes and values in different countries. One day, however, an enduring irritation may become an obsession for which resignation no longer suffices. 2 Leniency. The irritation resulting from a difference may be dealt with by leniency, as a variant of tolerance. For example, the small country of Luxembourg irritates its neighbours by its practises in banking, being its major economic sector. For a long time, the neighbouring countries have accepted them leniently. Likewise, France and the Netherlands have been rather lenient regarding each other’s different (and irritating) drug policy practises. A next step may be to equalize the differences as well. Concessions and the time involved are the costs of leniency. 3 The battlefield. Many European countries once used this method for settling irritating differences. Famous cases are the Roman, Frankish, Napoleonic and Third Reich empires. Victory may solve the irritation and equalize the difference but costs a lot of resources, which is only 36

Chapter 1

The Europeanization of Public Affairs

rational if the victor keeps the upper hand and can write off its investments over a long period. Post-1945 wisdom in Europe takes this method as largely inefficient and ineffective. 4 Imitation. Governments and companies may decide to follow what is called the best practise method. They adapt their position to that of the other(s), thus removing the difference and consequently the irritation as well. Many a national welfare scheme in Europe is based on such imitation, following the British ‘Beveridge’ or the German ‘Bismarck’ model [Flora, 1988]. Industrial companies easily adopt newly developed technology, due to pressures coming from an open and competitive environment. The costs of adaptation are real, but are also rewarding and thus accepted. 5 Negotiation. A government or a company may enter into negotiations with those responsible for the irritating difference. Public treaties or private agreements can bring about the solution. There exist thousands of good examples, ranging from the Rhine Treaty to the standardization of automotive parts. The method is based on an ad hoc approach, taking each issue separately. This is inefficient if there are many issues at stake at the same time and ineffective if there is no sanctioning mechanism. The five traditional solutions have always been applied in the management of conflicts in Europe. Together they form the old menu of integration practises with their benefits and costs as sketched above. The construction of what is now called the European Union (EU) aims to provide a sixth and better method to solve irritating differences. It works through a machinery of common public decision-making, called Community or EU method, by which the Council of Ministers and representative platforms like the European Parliament decide upon legislative proposals of the Commission and under appeal at the Court. This method aims at a pragmatic accommodation of cross-border conflicts and not at any kind of assimilation, given the bad experiences with this via the old battlefield method. The new EU method results in EU laws, based on treaties and having priority over domestic law [Keleman, 2012; Lindberg, 1970] and not in an EU that is one cultural community of shared values. The EU is, instead, a collection of many national communities, called ‘member states’, and ever more regional and local ones, each having its own cultural characteristics [Marks, 2012], all held together by that legal community of shared laws based on Treaties. The EU or community method is thus a normative project under construction, which implies that its real workings may still be incomplete and imperfect, according to many. The sixth method is neither unique nor exclusive. Most current states 37

The Europeanization of Public Affairs

Chapter 1

have a past of heavy internal diversity with even civil war and came into being by initially one or more traditional methods and finally the new method of state construction. Neither is it exclusive, as most of the older solutions are still applied. The EU is an open forum allowing cross-border resignation and leniency. It is a non-violent ‘battlefield’ where fighting takes place only verbally and on paper and warfare is replaced by ‘lawfare’. By an Open Coordination Method (OCM) it stimulates organizations to follow best practises that may exist somewhere. In fields not covered by treaties it is full of ad hoc negotiations, as shown in the post-2009 euro crises. Yet, for internal conflict management in the EU the old methods are widely considered to be not so efficient and effective and as inferior to the EU method; for external conflict management (EU foreign policy) the old methods are still dominant [Hughes, 2010]. Gradually the EU method is replacing the old methods [Mabbett and Schelkle, 2009], which from historical perspective is a political miracle. But it is not undisputed, as Europeans disagree over every value and thus also over the EU method, particularly at home, where many fear a loss of national sovereignty, which indicates deepened EU awareness at mass level [McLaren, 2006].

III Not Member States, but Member Countries From the legal perspective, the national governments of the EU member states make every new treaty that binds their country after ratification by the electorate (through a referendum) or the national parliament. It is, however, a misperception to see the EU as involving only national state governments. The EU is composed of member states, to be sure, but what are they? Take the case of France [Culpepper and others, 2008; Grossman, 2007; Knapp and Wright, 2006]. France has the reputation of being the most ‘statist’ and most centralized state of the Union, as illustrated by the saying ‘Paris gouverne’. The state intervenes strongly in French society, usually by way of specific and hard regulations. For almost any sector, trade or activity there is a detailed book of rigid codes. Through a plan indicative, longterm policies are developed and made. State banking and subsidies give the central government financial power over society. What is the state? It is largely the bureaucracy that issues the regulations and controls public finance. The leaders of this technocratic apparatus are not so much a closed group as a network, mixed up with captains of industry and sociocultural elites, with whom they have enjoyed a common education (the grandes écoles) and often a change of position (pantouflage). Major companies and associations can be present or represented inside the min38

Chapter 1

The Europeanization of Public Affairs

istries by their membership of administrative comités. Who is in charge of such a state system? The legal view is incomplete. State and society are partially overlapping and different. Or watch Germany [Timmins and others, 2007; Padgett and others, 2003]. This member state is, even in legal terms, a federalized collection of sixteen Länder or regional states, each having its own government, parliament, bureaucracy and court. The umbrella state is about 150 years young, and contains all kinds of variation. Brandenburg is not the same as Bavaria. Like Europe at large, the German states have their more or less irritating differences with each other. At both the regional and the federal level, the governmental culture is mildly interventionist, with a preference for general and encouraging rather than specific and rigid policies. The regional governments play an interventionist role through, for example, their Landesbanken. All governments make, as part of their corporatist-like Konkordanzpolitik, long-lasting agreements with well-organized interest groups, such as from business, labour, consumers and environmentalists. Trade organizations, companies and civil groups have comparable forms of Mitbestimmung (co-determination). Social order is valued highly, hence the extensive provision of social security. Again: who is influencing whom? There is no major federal government policy without at least the basic consent of the leading decentralized governments and private groups. Or consider, finally, the case of Britain [Heffernan and others, 2011; Bache and Jordan, 2006]. Here, the central government shares its limited rule with regional authorities and agencies. Their interventions in society usually have a general scope (like in Germany) and rigid form (like in France). But all governments together have only a limited domain, leaving much to the private domain of pluralist market forces. Developments of privatization and deregulation during the Conservative years (1979-1997) enlarged and vitalized this private domain. The central government dislikes the continental practises of consultation and keeps organized business and labour at arm’s length. The Cabinet, acting as the first committee of the House of Commons, pushes policy-making without compromising social consultations or co-determinations. It prefers the rule of law above the rule of compromise. These basics of governance have not changed very much since the 1980s. Once again: who is influencing whom? The UK is like an open park with different public and private playing-fields. These three country snapshots make clear that member states, widely seen as the crucial parts of the EU construction, are not identical to national governments. All countries have a government that is more or less both fragmented by different layers (regionalization) and limited by the private domains. The fragmentation in the public domain is relatively low (but increasing) in France, Denmark, the Netherlands and the Central Euro39

The Europeanization of Public Affairs

Chapter 1

pean countries, but high in countries like Germany, Belgium and Spain [Flora and others, 2005]. The private organizations, which everywhere far outnumber the public ones, are usually extensively fragmented by size, sector and region. The limitations show different patterns. In the UK the public and private domains are rather separated, but in France, Italy and Portugal they are bridged by government-business courtships, while in Germany, Austria and Sweden they are often bound by public-private agreements. In the Central European countries, the governments widely intervene in private society, but see their domain shrinking. Overall, the general pattern is a somehow organized pluralism, which gives all public and private groups some domain and scope for influence at home. The member states can best be characterized as mixed public-private systems. The government is only a part of any country’s governance system, as the latter is run by a mixture of public and private organizations, such as ministries and companies. In between stand many hybrids or nongovernmental organizations (NGOs), like trade unions, institutes and consumer groups, which are often somehow linked to either the public or the private domain [Hudock, 1999]. Between the public and the private organizations, mutual interventions exist. Government organizations intervene in private organizations that act for profit or not and private organizations do so in the organizations of government. For this reason it would be better to rename the concept of ‘member state’ as ‘member country’. Before we examine the consequences of this for the processes of European integration, we need to look more closely at the various patterns of relationships between the public and the private organizations as they exist in the countries.

IV The Domestic Patterns of Public-Private Relationships For the sake of analytical clarity, we will keep society split into two parts, the public and the private, thus disregarding for the moment the many hybrids in between. On the relationships between the two parts, we pose three important questions. Firstly, do they each have their own separate world or are they interdependent? Secondly, is the public sector dominating the private one or is the opposite the case? Thirdly, is the quality of their relationship antagonistic or friendly? The central variables behind the three questions are, in short, respectively: domain, direction and affection. If we limit ourselves, again for the sake of clarity, to dichotomized answers (the extremes of the three kinds of relationship), then we get six logically possible patterns. However, as the answer to the first question 40

Chapter 1

The Europeanization of Public Affairs

shall be ‘interdependency’ and not ‘two separate worlds’, we shall end up with five both logical and relevant patterns of relationship, typified as follows [Van Schendelen, 1990]. 1 Interdependency, no ‘two worlds’. It is true that in many countries there are people who believe that the public sector of government and the private one of civil society must have autonomy, independent from each other, as two separate worlds. The normative beliefs are termed as ‘privatization’ and ‘deregulation’ for the public sector and as ‘citizenship’ and ‘self-reliance’ for the private one. Are the two sectors really independent from each other? There is no empirical evidence to support this view. For its income, support, information, legitimacy and much more, any public organization is, directly or indirectly, highly dependent on the private sector, and vice versa. Thus, public organizations want to intervene and do so in the private ones, just as the latter do in the former. This interdependency is not constant but variable, both across countries and time. For example, the central government of France usually intervenes more often and invasively in private organizations than its Irish counterpart. In Germany and Sweden, the public authorities usually undergo more interventions from private groups than their counterparts in Portugal or Poland. All countries are mixed public-private systems, thus giving validity to the subsequent patterns. 2 Public interventionism. This second pattern is one extreme of the core variable of direction. It refers to the interventions of government in civil society through both its hard legislation, ultimately maintained by police and court systems, and soft interventions, such as subsidies, procurements, statements, privileges and more. The soft ones have the good face of granting a desired item or the bad one of refusing it, and in both cases they can be as effective as hard law, for which reason they may be called ‘soft law’. The outcome is that every factor of a company’s functioning, from production and processing to sales, is subject to some public intervention, except the factor of weather, which falls beyond anybody’s control. Many governments target their interventions most easily at the private areas with low mobility (big plants, farms) and/or high concentration (chemicals, car industry). The factors and reasons behind the choosing of these targets are manifold. It may be for solving a social issue of infrastructure, security, education or welfare, or be caused by government instability, financial crisis or technology. In a pluralist society an intervention is always contested, even inside the government. Both politicians and bureaucrats have different ideologies, policy values and interests that they push inside their domain. In their internal competition, many search for external sup41

The Europeanization of Public Affairs

Chapter 1

port from private groups with similar preferences and even empower the latter to support them and to oppose their public rivals. The EU’s Commission is famous on this (see chapter 2). 3 Private interventionism is the third pattern and mirrors the former. In order to protect or promote their interests, many civil organizations like NGOs and companies intervene in governments. A formal indicator like regulation in the former pattern is largely absent for private interventions. Only in the few ‘corporatist’ countries (like Austria, Germany, the Netherlands and Sweden) may the leading groups of employers and employees have a strong say over the government’s socio-economic agenda. The standard measure of private interventionism is its influence on public officials, indicated by the latter’s decision to act differently from what they would otherwise have done and more in the interests of the private groups that exerted the influence. Many private groups push, paradoxically, public interventions that benefit them. Their behaviour can be seen as a case of participation in government. As such it is again a variable, dependent on many factors that can be downsized to four main categories [Milbrath and Goel, 1977]: – capacities to act, such as resources (size, budget, personnel), external support (coalitions) and skills – desires to deal with threats and opportunities coming from or through government – compulsions from competitions with other groups or from its calculation that the costs of influencing are lower than the benefits and/or the costs of non-action – invitations from government to participate as interest group or stakeholder. 4 Public-private antagonism is our fourth pattern and represents one extreme of our last core variable, being affection. Indicators here are distrust, evasion, insult, conflict and fight. In most continental countries of Europe, antagonism is often more latent than manifest, due to a domestic culture of preferred harmony. Yet, antagonism is a normal phenomenon everywhere. It may be caused by misperceptions and misunderstandings, but is then real in its consequences. It may also be based on real conflict of interests, injury or misconduct, or be used as bargaining chip in a negotiation game [Coser, 1956]. Public and private organizations can be antagonistic not only to each other but also to organizations in their own domain, thereby increasing the pluralism of the country. For example, on the issue of a public railway a coalition of the Ministry of Environment, regional authorities and green groups is often antagonistic against a counter coalition of the Ministry of Transport, truckers’ groups and local citizens’ groups. The dichotomy 42

Chapter 1

The Europeanization of Public Affairs

between the public and the private domain seldom or never results in full antagonism between the two in reality, as the latter shows (almost) only mixed coalitions of public and private organizations. 5 Public-private partnership is the final pattern and it mirrors the former one. Mutual trust, co-operation, respect and harmony are its indicators. Public and private organizations then feel happy with each other. This may, of course, be the result of misperceptions and misunderstandings, but become real by its consequences. More commonly it is based on shared preferences, mutual dependencies and common interests. A partnership can also be a negotiating chip, even if it is only played as a courtesy. This happens in continental countries especially, where a partnership easily garners social approval. Partnerships frequently remain, however, quite limited by time and place as happens between a local authority with land but no money and an investment company that has money but no land to develop. By its creation of outsiders and losers, a partnership can fragment both the public and the private sector even more, so increasing the country’s pluralism.

V The Europeanization of the Member Countries The implicit assumption so far is that national public and private organizations remain inside their national borders. In reality they do not, especially not in the European area with its mainly medium-sized and highly interdependent countries. For its domestic success a ministry, company or whatever interest group can be more dependent on another EU country than on its own. The EU countries are for 55% to 80% of their exports (65% on average) dependent on intra-EU trade. The 2011 Greek Debt Crisis created strong chain effects on banks, companies, households and governments across borders. Hence, every country, organization or group has to cope with other European entities and not least with the EU. All patterns of relationship mentioned above can become relevant at this European level too. True enough, a shopkeeper or a citizen may consider this level as belonging to another world far away, but in objective terms nobody can escape the interventions from other countries and the EU. Annually the EU produces about 2,500 new decisions that overrule domestic laws and acts and create chain effects at home. From their side, ministries, companies and NGOs are often active pressure groups in other countries and at the EU level. With some parts of the EU, they may have an antagonistic relationship and with other parts a partnership. The same holds true for the EU units. A Directorate-General (DG) of the Commission intervening in a member country may develop both antagonistic and 43

The Europeanization of Public Affairs

Chapter 1

friendly relationships there. Beyond the European level, there also exists the global level with its economic, financial and political forces and its international organizations like the WTO (World Trade Organization) and the WHO (World Health Organization). It strongly intervenes in the EU, down to its local level and, vice versa, it is also influenced by the EU and its parts. All patterns of relationship, in short, can recur here. We have already observed that, at the European level, the EU is relevant as a sixth approach to the settlement of irritating differences among the member states. We also observed that a member state is not a homogeneous entity, let alone a monolith as suggested by the notion of ‘member state’, and is better described as a member country. These two observations can be linked now. An irritating difference between two countries is never a difference between two nationally cohesive coalitions of all public and private organizations. It is always (without exception) an issue inside each country as well, both between and among its public and private organizations. In the 2008 fight over the protection of Volkswagen against an open car market, the company was supported by its regional and federal governments plus the trade unions, versus an alliance of the Commission, Porsche, the German employers association BDI and foreign car makers. Inside the EU they all can air their preferences and disagreements over any issue, so contributing to its europeanization at home. By raising them to the EU, they make this level mixed public-private too. This term or idea of europeanization has become increasingly popular in both public discussion and in literature, but its definition is frequently implicit or loose and not yet well established. Some take it as a synonym for European integration or, in inelegant wording, ‘EU-ization’ [Conway and Patel, 2010; Kohler-Koch, 2003; Fligstein and Stone Sweet, 2001], others as a regional case of globalization [Wallace and others, 2010]. Particularly dominant is the view that EU dynamics have become part of domestic policies and politics [Börzel and Risse, 2012; Ladrech, 2010; Graziano and Vink, 2007; Schimmelfennig and Sedelmeier, 2005; Featherstone and Radaelli, 2003], which is so to say the downstream view, which goes from the EU to the national systems [Green Cowles and others, 2001] and creates impacts on, for example, domestic social policies [Bache, 2008], public law [Jans and others, 2007] and urban government [Marshall, 2005]. The reverse view of an upstream of interventions from the national systems to the EU is seldom given equal attention but stressed here, as we define europeanization as the increase of cross-border public and private issue-formation in Europe. The core element is the transport of an irritation or an issue across state borders. The increase can take place not only by volume (more new issues), but also by content (more intensely contested issues). The definition leaves open four questions about the process: (1) the source 44

Chapter 1

The Europeanization of Public Affairs

of europeanization, (2) its direction from the start, (3) the dependent variable that is being europeanized and (4) the final outcomes. The source can be one or more of the following four: a European or a domestic pressure group, either public or private. Examples are the EU, a European federation (EuroFed), a national ministry or a trade association. On the same issue all different sources can be active at the same time, thereby reflecting the pluralism of Europe. Europeanization can occur in two opposite directions. One is the downstream from the European level, being another country or the EU, to the domestic one and the other is the upstream that starts somewhere in a country and flows to another country or to the EU. The dependent variable at either the European or the national level can vary from people’s attitudes and behaviour to policies, institutions, regime values and constitutional arrangements in the public or private domain. The outcomes of issue-formation are left open here as well. They can be a proposal or law made by the EU or a private agreement among companies from various countries, or the issue may remain as it stands, like the plug socket issue, or simply disappear, like the early 1990s HDTV issue.

VI. Vectors of Europeanization Elaborating upon our general concept of europeanization, we will maintain the two central dichotomies regarding the source: the national (or domestic) versus the European level and the public versus the private sector. Of course, every dichotomy has some grey area. For example, is an issue coming from the multinational company Siemens, headquartered in Germany, a case that belongs to the national or to the European level? The realistic answer is: formally national, but practically European. If Siemens creates the issue in partnership with the German Ministry of Environment, does the case then belong to the private or the public sector? Although dichotomies are always less shaded than reality, we can use them here for the purpose of clarifying the mechanism of europeanization. At least eight different vectors or carriers of europeanization can be distinguished: the four analytical sources multiplied by the two directions. They are presented in figure 1.1 and explained below. The first four cover the downstream from the European level to the domestic one that results in domestic (public or private) adaptations to European causes and the other four the upstream from the domestic level that influences the European one. Viewed from the EU, the first four stream down from its output side and the second four stream up to its input side. Jumping forward to the rest of this book: if an interest group is not part of the influence vectors, 45

Chapter 1

The Europeanization of Public Affairs

Vectors of Europeanization Downstream Adaptation Model (E Public

N)

Private

National European

National European

Upstream Influence Model (E

N)

National European

National European

– – – – – Indirect or derived europeanization (vectors 9-12)

Figure 1.1

46

Chapter 1

The Europeanization of Public Affairs

it has to adapt itself to the outcomes, or ‘those who are not around the table, may be on the menu’. The downstream vectors are the following: 1 From the European public sector to the national public sector. This refers to the popular idea of ‘Brussels governs’. It has given rise to a large body of literature on domestic government adaptation to EU laws that overrule domestic laws directly or, through national implementation, indirectly. A dramatic example is the post-2010 EU supervision of the national state budgets, starting with Ireland, Portugal and Greece. Due to the treaties, this vector is most active in fields like market harmonization and liberalization. This vector also applies to treaties made outside the EU Framework, such as the 1985 Schengen agreement on frontier control and the 2012 draft treaty on the European Stability Mechanism ESM. Soft EU interventions like guidelines and subsidies fall, if followed by national governments, under this vector as well. 2 From the European public sector to the national private sector. Many decisions by the EU directly bind private organizations, whether profit-oriented or not. General examples are the EU decisions regarding an open and competitive market. Specific examples are the EU regulations regarding food safety or emissions harmful to the environment. The non-EU Rhine Treaty, signed by national governments along the River Rhine, binds both the companies that transport goods and those on the banks (and the governments alongside too). 3 From the European private sector to the national private sector. Many private sectors have their EuroFeds. For the food and drink industries there is FDE (formerly CIAA), for the trade unions (ETUC), for motorcyclists (FIM) and so on. One of their functions is to influence the EU, which falls, however, outside our definition of europeanization, because no border is crossed here. Their most important latent function is to form internal agreements that the members are supposed to follow. Without any public involvement, any agreement can europeanize the members back home. Many norms and standards in industry and services are produced this way. 4 From the European private sector to the national public sector. The EuroFeds and institutions referred to above may seek to influence national public organizations as well. For example, the European Round Table of Industrialists (ERT) addressed its 1984 campaign for an open EU common market directly to the national governments, which decided accordingly in the 1986 Single European Act (SEA). Regarding the distribution of landing rights for air transport, the national carriers, united in the Association of European Airlines (AEA), orchestrated in the early 1990s informal influences on their national ministries at home. 47

The Europeanization of Public Affairs

Chapter 1

European NGOs like Amnesty and Greenpeace often launch calls to national capitals for friendlier domestic policies. Here we leave the vectors that stream down from the European to the national level and cause a national adaptation to the European level. The next four vectors regard, conversely, the upstream influences from the national to the European level. 5 From the national public sector to the European public sector. An obvious target of national governments is the EU Council of Ministers. National ministers meet together in order to suggest, approve or reject proposals for new legislation coming from the Commission and possibly amended by the European Parliament. A semi-formal example is provided by the thousands of national civil servants that partially come from decentralized governments, sit in the Commission’s expert groups and use this vector for squeezing through the interests of their government at home. 6 From the national public sector to the European private sector. In 1998 the French government wanted to create for Aerospatiale (a French aerospace manufacturer) a European consortium with British Aerospace and German DaimlerChrysler Aerospace (DASA). The French government partially privatized Aerospatiale to tempt the two companies into an agreement, and so it happened. Local governments can appeal to European companies to make a common cause, like local authorities in steel regions suffering from declining employment did in 1995. Jointly with major European steel companies and trade unions, they established Caster, a public-private umbrella for steel interests (after its 2002 merger with Reti, now Eira). 7 From the national private sector to the European private sector. Every direct or indirect membership of a private EuroFed holds this vector. The national groups of pharmacists discuss some of their issues within the EU Pharma Group PGEU, where they try to monitor and to influence each other. This sectoral approach is followed by thousands of other national private organizations, profit-oriented or not, from telephone companies to consumer groups. By making a cross-sectoral deal at home, they can even orchestrate their influence on more than one European sector. This way, the French railway company TGV strongly influences the time schedules of other transport modalities in the countries where it operates. 8 From the national private sector to the European public sector. This vector covers the many popular cases of EU lobbying by private companies and citizen groups, the many private sector people sitting in expert groups of the Commission and most complaints over unfair market 48

Chapter 1

The Europeanization of Public Affairs

behaviour that the Commission receives from private groups with grievances. Along this vector too, players and clubs in professional football have effectively challenged the rules of their European umbrella (UEFA) by appealing to the EU’s open-market rules [Garcia and Meier, 2012]. Multinational firms often use their national subsidiaries for orchestrating a pan-European symphony of lobby sound directed at the EU, as Philips did in 1986 to gain EU protection for its ‘infant industry’ of compact discs (failed), and in 2007 for getting, reversely, liberalized import tariffs for its low-energy light bulbs made in China (success in 2008). For getting a market for their anti-tobacco products, the pharmaceutical companies pushed the 2001 Tobacco Products Directive, at the cost of the tobacco industry [Duina and Kurzer, 2004]. Market competition usually drives this vector.

VII The Fuller Story of Europeanization The eight different vectors help us to understand better the reality of europeanization. They represent eight hypotheses or stories of europeanization, thereby stimulating the mental mapping for both research (description and explanation) and practise (evaluation and optimization). For example, interest groups wanting to make their organization EU-proof are usually focused on the EU’s rear flank, from which they receive the laws that require adaptation. Thanks to the vectors they now can broaden or even shift their focus to the EU’s front-side and try to get EU laws adapted to their domestic needs instead. The vectors form, however, a schematic model that can never tell the full story of any case of europeanization. Six realistic elements must be added: (1) public-private transactions not across borders, (2) the grey areas of the two dichotomies, (3) the connections between vectors, (4) the wider global level, (5) creeping europeanization and (6) rolling europeanization. Their specifications are as follows. 1 No crossing of borders. In our definition of europeanization all issueformation at only the national or the European level falls outside any vector. These events not crossing borders can yet be relevant for crossborder issue-formation. In figure 1.1 they are labelled as indirect or derived vectors and numbered 9 to 12. In 2003, the Commission called the EuroFeds of employers and trade unions to arrange more pension mobility (vector 9). Most EuroFeds try to influence the EU institutions and stakeholders there (vector 10). In many countries, ministries at home ask private groups for information and support at the European level (vector 11). Very common is that a company or NGO observing a 49

The Europeanization of Public Affairs

Chapter 1

threat or opportunity coming from elsewhere in Europe, asks domestic public officials to support its position at the European level (vector 12). Even events happening inside the public or the private domain only can become relevant for European issue-formation. Domestic companies and NGOs frequently support or fight each other over action at the European level, as do domestic governments. All these cases not crossing a border are side stories of europeanization: relevant, but not the main story. 2 Grey areas. The two dichotomies are not cleanly divided in real life. Many a public or private organization operates at both the national and the European level. The clearest example is the multinational company (MNC) acting in various countries and inside EuroFeds and EU expert groups. EU Agencies work closely with national agencies. Most national ministries and many decentralized governments and non-governmental organizations (NGOs) occupy positions at both levels too. They may suffer from a degree of ‘schizophrenia’ between europeanization and domestication [Kassim, 2003], but are able to transport their issues across borders. In 1987, UK Prime Minister Margaret Thatcher politicized this grey area by demanding clearer separation between them and, re-launching the mediaeval notion of subsidiarity, for the supremacy of the domestic level. The second dichotomy, between public and private organizations, has its own grey area too. Well-known are state-run companies and company-run agencies. Many hybrid NGOs exist in the grey areas as well [Van Thiel, 2000; Beloe and others, 2003]. Some are government-organized (and sponsored) NGOs, to be called GONGOs, and function as a hybrid agency or institution in fields like health, human rights, development and the environment. For enabling NGOs in the EU, the EU Commission funds many NGOs to enable them to function [Mahoney and Beckstrand, 2011], at a cost (2010 estimate) of more than €1 milliard a year. The antipode is the business-organized (and sponsored) NGO or BONGO, such as any trade organization and the NGOs in fields like health (patients), transport (travellers) and consumption (consumers). Some governments and companies create fake NGOs in order to get the friendlier face of ‘civil society’ and thus another vector at hand, in short a better ticket to influence the EU. The parent of a fake NGO takes the risk that its creation becomes unmasked or becomes a monster that eats the parent. NGOs can become even more hydra-headed. If wanting to sit on the government’s seat, they become a government-interested NGO or GINGO. By selling special products and services, as did the simoniacal clericals in the past, they become a business-interested NGO or BINGO. They usually do this in order to acquire power and/or money. In 2005 the hybrid relations between the 50

Chapter 1

The Europeanization of Public Affairs

public and private domains became an issue in the EU. 3 Connected vectors. The cross-border formation of an issue is seldom carried out by only one vector. Most cases of europeanization are caused by a number of vectors connected in parallel or series. An example is the Decker/Kohll Court case. In the mid-1990s two citizens (a Luxembourger and German) needed medical services during holidays in another EU country. Back home they applied for reimbursement to their health insurance companies, which refused to grant it. After appeal, the Court decided in 1998 that medical services are not exempted from open-market law, thus the costs should be reimbursed. The angry companies approached, as side story of europeanization, their domestic ministry and EuroFeds, such as the CEA. National ministries discussed the issue in the EU Health Council of April 1998. The 1999 Amsterdam Treaty’s chapter on public health stimulated the Commission to prepare proposals for liberalizing the public health insurance market. The companies soon started their lobby at the EU directly. In this single case the first two adaptation vectors and three of the four influence vectors were in operation serially or even in parallel and all together they form only one episode. The case also shows that adaptation and influence vectors are analytically different but connected in practise. This connection is often even planned. As EU binding decisions overrule domestic ones, it is rational to influence the national capital via Brussels [Fontana, 2011]. 4 The global level. Non-European countries with booming markets, like Brazil, Russia, India and China (‘BRIC’), are forces of europeanization, just as Europe is reversely influencing them, directly or indirectly via organizations like the WTO. For example, trade issues on bananas, beef and biotechnology were initially incited and settled in the EU, but in 1999 the three ‘B’s were put on the WTO agenda by the US negotiator Charlene Barshefsky, thereby becoming global issues. By looking for support from national governments and private companies in Europe, she exploited divided positions and useful vectors there. Some governments and private interest groups from the EU indeed supported the US against the EU, in order to change the EU policies that disadvantaged them. The reverse happens as well. Then, US governments and/or private interest groups, having lost in Washington, support the EU side in the WTO. 5 Creeping europeanization. Many stories of europeanization have important prologues. In policy fields on which the EU hardly has formal powers, such as education, taxation and social inclusion [Armstrong, 2010; Linsenmann and others, 2007; Borrás, 2004], the Commission often uses tools of soft law like attractive subsidies and 51

The Europeanization of Public Affairs

Chapter 1

benchmarks from an OCM. They can directly influence the member countries and, after some time, they can develop into hard law, as part of the EU’s downstream. Many interest groups that want to get their interests included in EU agendas, exploit such creeps in the upstream and often see these followed by the desired impacts and hard law. 6 Rolling europeanization. Many stories of europeanization also have important epilogues. A downstream example is the EU’s general competition law that has had strong impacts on many different sectors, such as on health, education and housing. The Habitat Directive (EC 1992/ 43) on Preserving Nature has affected water management, town planning, road construction, archaeological excavation and more. Similar epilogues exist in the upstream too. Outside the EU framework, cross-border police co-operation started in 1990 as an initiative of five of the original member states (except Italy) as the Schengen Agreement. Pressured by interest groups from trade, NGOs and governments, the agreement was included in the 1993 Maastricht Treaty (UK and Ireland opting out) and much expanded and intensified since. Vector analysis is indispensable for an improved understanding of any case of europeanization. To scholars it may bring better explanations of outcomes, how difficult it may be to assess causality [Exadaktylos and Radaelli, 2012]. Practitioners wanting to score in Europe, may gain better insights on optimizing the connection of influence vectors from it. Although the EU clearly is not the full source of europeanization, it has become the single most important source of it. The EU deals with all sorts of irritating differences between domestic public and private organizations from the different member countries. It has political, legal and other capacities to channel issues and to bring many of them to at least temporary accepted outcomes. In academic terms, the EU is the major dependent and independent variable of europeanization. In practical terms, the major and common capital of the member countries is now located in Brussels.

VIII Influence Challenges from Europeanization It is old wisdom that every organization is never independent from and always more or less dependent on its challenging environment [Thompson, 1967]. This outside world is perceived as having two faces. One looks promising and full of opportunities, the other threatening and difficult. The ever-changing complexities of the outside world make adaptive and/or influencing behaviour continuously necessary. That world is full of potential friends and enemies, in short stakeholders. Therefore, every organiza52

Chapter 1

The Europeanization of Public Affairs

tion has to satisfy two selfish interests. At its input side, it has to acquire from outside the necessary means to operate, for example budget, support and information, and at its output side it has to deliver what is demanded by stakeholders, for example special products, services, support and information. In this functional sense, every organization is an interest group. This is equally true for a ministry, the EU, a company and a group of citizens. Many government officials indeed act as belonging to an interest group [Baumgartner, 2007]. Only with regard to its internal operations can an organization act more or less independently, but this autonomy is not sufficient for survival, is always limited and is dependent on many properties, such as type (public or private), resources (affluent or poor), reputation (good or bad) and, particularly, awareness (alert or sleepy). The more an organization is conscious about its position in the larger world, the better it can exert its autonomy and do more than merely adapt to the good or the bad luck coming from outside. If it is alert, it gets a fair chance to influence the environment by pushing the opportunities and by blocking the threats. Paradoxically, an unfriendly environment full of threats frequently sharpens this awareness better than a friendly one that tends to dull the senses and to increase the risks of being caught by surprise and of losing leadership, income or function. No organization is safe in this respect. Since the rise of the EU, the vectors of europeanization enrich this old wisdom with the new experience of a much wider cross-border environment. The downstream vectors create challenges of adaptation. They can bring the good fortune of opportunities, for example for exporting products or changing policies at home. They can also bring new threats and increased costs of adaptation that no longer, as in the past, can easily be filtered or blocked by the national state. The upstream vectors create their own challenges of influence. They can offer opportunities, such as a more level playing-field or an EU decision as desired, but also new threats, for example a comparative advantage to a competitor at home or abroad. The fight for survival, in short, is shifting from the domestic to the European level and requires a cross-border or European approach. This influence approach at the European and particularly the EU level stands central in this book. Anticipating chapters 4 and 5, the major prerequisite is an open mind for all vectors of europeanization, which obliges to three early forms of preparatory homework: – Facts and trends occurring in the EU area need continuous monitoring. – Observations need to be assessed as threats and opportunities that always imply value judgements, derived from one’s own core values and objectives. 53

The Europeanization of Public Affairs

Chapter 1

– These assessments must be sharply defined as interests, being chosen positions or targets regarding those threats and opportunities that one wants to get solved. In pluralistic societies, the perceived facts, chosen values and preferred solutions are always contested by many other stakeholders that have taken different positions or interests. This makes interests practically synonymous to issues. This is true for both public and private interests, as a public interest is not an interest of a higher or lower order than a private one but merely an interest of a public entity such as a ministry, agency or party, whereas a private interest is one held by a private organization. The more alert or conscious an organization is in defining its interests, the more it is capable of survival in the challenging European environment. This alertness does not imply, however, that a public or a private organization always has to influence that environment. As will be shown later, there are various situations in which adaptive behaviour is more rational than an effort to influence a growing threat or opportunity. This rationality can only be based, again, on good preparatory homework. Often, however, adaptive behaviour is not the consequence of conscious homework but of precisely the opposite, namely the neglect of it. Then, due to its nonchalance, the organization can only adapt to, in our case, the europeanization of its environment. Sometimes it may enjoy some good luck from a free opportunity or faded threat, but usually it suffers more from the bad luck of a missed opportunity or neglected threat. If an organization comes to the decision that officials or stakeholders have to be influenced on an issue, it changes from interest group into, as this is called, a pressure group. Acting in the European area, it can choose from the full menu of vectors of europeanization the optimal mix of vectors, in series or parallel, in order to build up a potentially winning position. The quality of a choice is, as usual in life, largely dependent on expertise that to a high degree is dependent on preparatory homework and thus is frequently absent. For example, if wanting to save an opportunity or to solve a threat, an easy-going pressure group wants to go to the source directly and to influence the stakeholders or officials that have caused the challenge. It behaves like Pavlov’s dog. An intelligent pressure group, informed about the many vectors, shall search after an optimal mix of vectors by connecting the initial challenge to a different level and/or domain. In response to the 1992 EU policy programme on the Single European Market (SEM), with all its threats and opportunities for competition, many MNCs reacted not by pressuring the EU, but by moving to exploit their markets [Calori and Lawrence, 1991; Mayes, 1992; Urban and Vendemini, 1992]. In the early 1990s, the EuroFed of the Largest Textile 54

Chapter 1

The Europeanization of Public Affairs

and Apparel Companies (ELTAC) took the reverse route. Wanting (high) price protection against cheap imports from non-EU low-wage (child labour) countries, it got this market issue solved in the EU by acting as a sort of NGO against child labour.

IX Old Influence Techniques and Lobbying Trying to influence somebody else is as old as human life and everybody can choose from a long menu of traditional techniques that people use [Greene, 1998]. The following examples run from sweet to spicy: – Seduction. Famous cases in literature are Eve’s apple in Paradise (Old Testament), the Ars Amandi (Art of Love) and, for repair, the Remedia Amoris (The Cure for Love), both by Ovid, and countless more [Greene, 2001]. Examples of seduction are donations, flattery and politeness, all attuned to the taste of the other. Competitors may overbid and seduce better. In the EU, stakeholders and officials usually have a sack of fresh carrots within reach. – Reframing. It is not the text but the tone that makes the music. By reframing an issue so that it sounds nicer to others’ ears, one may collect more support in the game [Daviter, 2011]. A subsidy for overproduced food is difficult to get, but for the upgraded purpose of humanitarian aid it is much easier. Opponents then have to play the downgrade. – Persuasion. Promoting a strong belief in ‘what should be’ may convince others to do what is asked. The ‘missionary enterprise’ of the Jesuits once converted many people by ‘marketing and selling’ its belief system through persuasion plus seduction by way of promising social benefits [Wright, 2004]. Many pressure groups act as persuaders [John, 2002; McGrath, 2007], for example the ‘green’ NGOs in Europe. Opponents have to fight or neglect that belief system and/or to start their own form of persuasion. – Image building. If one impresses others with a self-image that they hold for relevant and true, one may gain influence on them. The image must reflect core social values, such as generosity (‘Robin Hood’), carefulness (‘Guardian Angel’) or power (‘Me Tarzan’), as some labour unions, green groups and governments show. Opponents may want to unmask this as robbery, self-interest and illusion. – Argumentation. Scholarly reasoning about ‘what is’, based on evidence and logic, is helpful but seldom sufficient as only one assertion has the full correlation value of 1.0, namely that between life and death. All other assertions are contestable or even falsifiable. It is commercial rather than scholarly argumentation that works fine. The former is 55

The Europeanization of Public Affairs











56

Chapter 1

salesman’s talk that conceals what is contestable, tells what is correct and, if contradicted, flatters that ‘you are right’ and thereby gathers the deal. The scholarly argument can work in a few situations: when important stakeholders are hesitant (they might be won), when the issue is in an early phase (many are uncritical), when the issue attracts publicity (the media may want better argumentation) and when it needs an upgrade (up to a more general interest). Advocacy. This has many variants. The informal one is, for example, a mass media campaign, which is frequently used by NGOs and trade associations [Holzhacker, 2012; Johnson, 2008; Warleigh and Fairbrass, 2002]. Semi-formal is the lodging of a complaint, for example by companies feeling hurt by unfair market practises and forwarding their complaints to the EU competition authorities. Most formal is litigation in court. But there is always room for counter-advocacy. Negotiation. Making a deal with an important stakeholder or official is an efficient way to get an interest at least partially met. It is the game of mutual give and take and finding a compromise or, in commercial terms, matching supply and demand. It requires from those involved much prudence and respect and from outsiders to remain aside [Gutmann and Thompson, 2010]. In the EU, though, many outsiders want to interfere and at least to avoid any deal at their cost. Authority. By some authority derived from tradition, charisma or position, one can cheaply get others to do what one wants [Friedrich, 1963]. Then, one gets a following and thus becomes a leader. Particularly in the EU, no source of authority is permanent and absolute and neither is any following. Encapsulation. By making a stakeholder or official more dependent on scarce items like budget, position or reputation, one can more easily get what one wants. BONGOs and GONGOs are dependent on the financial lifeline from their parent and thus want to please that parent. By creative use of networks and procedures one may encapsulate befriended interest groups in privileged positions and opposing ones in subordinate positions, thereby configuring a playing-field so that it is more convenient to one’s interests. The EU is full of encapsulation fights that, however, are not easy to bring to lasting success (chapter 3). Coercion. The ultimate example is the launch of a war, the ‘ars belligerendi’ as masterminded by Sun Tze in old imperial China [Griffith, 1963] and often applied in Europe. Today, officials can coerce interest groups and citizens by issuing legislation that is ultimately maintained by the police, the court and the jail systems. The EU is ultimately based on coercive legislation. NGOs may set up a blockade or hate campaign, as Greenpeace did against Shell in the 1995 Brent Spar affair. A compa-

Chapter 1

The Europeanization of Public Affairs

ny can threaten to move production to elsewhere. Coercion regards, in short, the use of sticks. Often it returns the boomerang of protest and isolation. All these (and more) traditional techniques of influencing a challenging outside world are still used in practise and frequently in combination, in either serial or parallel settings. Left outside the menu have been the illegal or controversial variants, such as bribery, blackmail, corruption or misbehaviour. Due to the high amount of social control in the EU, they are usually the fastest track to achieve failure instead of success. As old as the menu is the question: what choice is best? Every choice has possible advantages and disadvantages, for example in terms of costs to the user, effects on the people to be influenced and reactions from people watching. Choosing at random or by arbitrary trial and error is not a rational option, as this easily causes blunders and results in boomerangs. The more rational alternative is acting prudently, incrementally (step-by-step), avoiding points of no return, in short finding out by methodical trial and error or ‘the science of muddling through’ [Lindblom, 1959]. This is done by preparatory homework that to the best of one’s knowledge compares the possible costs and benefits of (among more) these various techniques of the menu, including their possible effects on one’s own position, the targeted groupings and the wider audience, and finally choosing the optimal course of action. Frequently both solutions are applied simultaneously and the metaphor of the both prudent and calculating chess-player is illustrative here [Simon, 1979]. Only amateurish players just go and see what happens. More rational ones gather information first, study it in order to improve their intelligence and, if they decide to play the game, go prudently. We return to this in chapters 4 to 6.

1 The notion of lobbying The notion of lobbying fits here [Thomas, 2004]. The word lobby comes from the Latin lobium or the mediaeval lobia, which refers to a hall or vestibule in an important building. People wanting to get something done by the important resident had to wait in the hall, where other visitors were usually waiting and servants went around. By talking to them, they tried to tap information about the resident himself, the rules of the house and the other visitors (called stakeholders today) and, better informed now, to probe support for their own interest. If permitted to visit and speak to the resident, they could push forward their request more attuned to his or her whims. In modern times, the political sense of lobbying gained popularity 57

The Europeanization of Public Affairs

Chapter 1

thanks to the US Congress in the late 19th century [Milbrath, 1963; US/GAO, 1999]. Waiting in the hall or lobby of the Congress building, representatives of pressure groups tried to gather information from Congressmen and their assistants and ultimately to urge the Congressmen to vote ‘yea’ or ‘nay’ [Matthews and Stimson, 1975]. Over the years, this ‘lobbying’ developed from merely corridor behaviour to collateral activities such as offering information and support, organizing mass attention and publicity and giving political and financial support. This latter activity sometimes developed into bribery that has given the word of lobbying a bad name among many people. Such ‘dirty lobbying’ still happens inside both the US and the EU, but mainly at the lower state or country level (lacking strict lobby regulations), around political parties and politicians (needing support for re-election in their district) and when social control is weak (low levels of lobby competition and mass media alertness). Weak and small-sized interest groups are more likely to take the risks of dirty lobbying than well-established, big ones [Bennedsen and others, 2011]. This dirty lobbying is scarce at EU level, where lobby regulations exist, party-politicians have a more indirect life-line with electors and social control is strong. In contrast to the English word of lobby, the old French word for the same place and activities, namely the antichambre, has kept its neutral meaning. The concept of lobbying we shall use here in its neutral or technical meaning and, bypassing many other definitions and labels [McGrath, 2005A, chapter 2], define as the build-up of unorthodox efforts to obtain information and support regarding a game of interest in order to get a desired outcome from a power-holder. So defined, lobbying belongs to the menu of traditional influence techniques and is perhaps the second oldest profession in history, if not the first (dependent on how Eve got Adam to eat the apple). The definition is neutral or technical as it takes no position on ‘dirty’ versus ‘clean’ lobby practises. As said, the dirty ones (illegal, controversial) are at EU level usually the fastest track to scandal, failure and expulsion and hence avoided by any person or group that wants to remain effective and by consequence remarkably scarce (in chapters 6 and 7 we shall return to its contested meanings). A power-holder is a person, group or institution on which one is seriously dependent. A lobby group we define as a pressure group that in unorthodox ways tries to gather information and support. Our keyword, unorthodoxy, falls outside orthodox (or institutionalized) patterns of behaviour, such as the right to pose a question or request during a formal meeting. As the effects of such activities are often uncertain or even risky, they can be better preceded by unorthodox approaches. Unorthodoxy has, however, no fixed contents, as it (like orthodoxy) may 58

Chapter 1

The Europeanization of Public Affairs

vary with time and place. What is orthodox (or not) in the one country or period may be the opposite in another one. The dichotomy also inevitably has a grey area, thus resulting in degrees of (un-)orthodoxy. A few EU examples can clarify this. The Commissioner who offers the Parliament’s rapporteur a fine lunch in order ‘to listen to’ or ‘to settle something’ is clearly lobbying, but their official answer to a request from the rapporteur has nothing to do with lobbying. Attending a meeting is not an activity of lobbying, but pressuring for its convocation or calling it off may be. Using a break during a hearing for dealing informally with other stakeholders is certainly a case of lobbying. The letter or lunch with some socializing talk may have nothing to do with lobbying, but every lobby-related letter or lunch can best start with such talk first, in order to create a pleasant atmosphere for conducting business. The main logic behind collecting early information and support in unorthodox ways comes from the need to discover the risks and opportunities of an arena before one enters it, in order to avoid blunders and boomerangs and to increase the chances of success. The early lobbying should have four characteristics: – Indirect. Information about stakeholders and their issues is preferably gathered in indirect ways first, for example through neighbours, friends and servants. – Informal. All this is done highly informally and preferably without leaving footprints, as otherwise some point of no return may be passed. – Silent. The lobbying is done silently, because information is gathered through ears and eyes and never through the mouth. Sound might mobilize opponents. – Charming. The supply side is pushed forward, because others are usually less interested in somebody’s demand side and more in ‘what is in it for them’. This lobbying takes place at all times and in all places, when and where somebody wants to get something from a power-holder and to minimize the risks of a refusal. This avoidance of risks requires much preparatory homework, as exemplified in several well-known sources. One is Ovid’s aforementioned manual on ‘how to win a beloved boy or girl without being refused’ (Ars Amandi), which comes down to approaching the ‘target’ indirectly, informally, silently and always with charm. He also observes that young people around 20 years old are usually perfect lobbyists in private affairs, almost by instinct; at a later age many have lost this instinct, do the exact opposite of what they should do and must rediscover the basics of lobbying in private affairs. Another source is classical diplomacy (De Callières, 1716). A third one and our icon is Niccolò Machiavelli, who 59

The Europeanization of Public Affairs

Chapter 1

always gathered information and support in unorthodox ways in order to better understand how to win and not to lose a contest. His catchwords are ambition, study and prudence.

X Lobbying the EU as Public Affairs Management Machiavelli’s catchwords are very true for the EU playing-field. On any EU dossier, numerous players coming from many countries want to promote or protect their interests but, divided as they are, do so in different directions. This competition may limit the efficiency and effectiveness of anybody’s efforts to influence. The EU with its many institutions, procedures and characters poses its own complexities. The traditional influence techniques have proven their limits since long, as a few examples may clarify. Coercion, if based on established EU law and exerted by litigation in Court, frequently takes a lot of time with short-term effectiveness, as the opponents can cause a retaliatory boomerang. Encapsulation requires stable power and many resources that are scarce. Advocacy leads to little more than counter-advocacy, as every barrister has an opponent, and thus leads easily to stalemate too. Argumentation (salesman’s talk) can be useful in a few situations but is seldom sufficient, as most competitors can produce their equally attractive position papers with plausible inferences and references. Like in ancient Greek drama, all characters may then be right in their position and, unwilling to compromise, dead by the final act. It is even more useless regarding strong allies and grim opponents, as the former are already convinced and the latter unwilling to become convinced. In the search after ever more efficient and effective influence techniques, the notion of public affairs management (PAM) has emerged [PARG, 1981; Harris and Fleisher, 2005]. The public affairs (derived from the Latin res publica) of an interest group are the threats and opportunities it perceives in its outside world, in short the external agenda that it wants to manage. This agenda contains the interests that need protection or promotion by influencing this outside world. The decision to do so belongs to strategy and thus is an essential part of general management. In the field of PAM, the term ‘influence’ is often replaced by ‘management’, derived from the Latin manu agere, which means ‘directing by hand’ or ‘having a hand in the game’, which is a synonym for influence. Ironically, the term ‘manipulation’, coming from the same language (manu pulare), has almost the same meaning but a much more unpleasant tone. So far, the new catchword PAM might be seen as an upgrade of the old idea of lobbying or as a new label on a bottle of old wine. It contains some fine 60

Chapter 1

The Europeanization of Public Affairs

old wine indeed, particularly the common focus on ‘trying to influence or to make a difference as desired’. In this sense there is no difference between a group that lobbies and one that applies PAM. Yet the bottle holds a lot of new wine too. The greatest difference between lobbying and modern PA management is the increased use of systematic methodology that must result in more rationally organized ‘ambition, study and prudence’. This is briefly stylized in figure 1.2 and detailed in chapters 3 to 6. As an appetizer we now sketch the five main steps of PAM. 1 Scanning the environment. It all starts with the extroverted (‘window-out’) radar on arising issues and events in the arenas of officials (who can decide ‘yes’ or ‘no’) and stakeholders (who may deliver support or opposition), in short with the lobby for information or, better phrased, intelligence that comes from the brains and is fed by the senses that gather information. The resulting list of identified threats and opportunities or ‘nightmares and daydreams’ coming from outside is easily a long-list above a hundred relevant interests, being only a few per cent of the EU’s new binding decisions issued annually. The long-list serves as a database for intelligent selection, as it is easily too long for managing it efficiently.

Public Affairs (PA) Management (2E = MI2) In Three Arenas: I Self

= Home front arena Daydreams, Nightmares

II Stakeholders

= Competitors’ arena Interests, Positions

III Authorities

= Officials’ arena Agendas, Policy frames

1 Window-out Intelligence Lobby: Information + Analysis = I2 2 Listing the Interests - Long-list = All Relevant - Shared-list = PA Shared With Others = M - Short-list = Do PA Yourself

3/4 Change Playing-Field by Triple P 3/4 (Remote) Control on Targets & Process 5 Window-in Support Lobby: Scoring = 2E Network> Aggregate Interests> Win by Coalition

Figure 1.2 61

The Europeanization of Public Affairs

Chapter 1

2 Shared-list and short-list. For reasons of effectiveness (influence success) and efficiency (cost-saving) the long-list is downsized into a shared-list and a short-list of interests. The shared-list is managed together with stakeholders having similar interests and the short-list by the interest group itself. Both lists are seen as relevant, but processed differently: with supporters (shared-list) or not (short-list). For influence reasons, a broader supportive mass is sooner or later necessary. For cost reasons this sharing of interests is done at the earliest possible moment, which is a matter of ‘intelligent laziness’: intelligent people like to be ‘lazy’ (efficient) by setting those with similar interest to work on common interests (but lazy people are not necessarily intelligent). Thus, the sharedlist can best be the longest possible and the short-list the shortest possible. 3 Triple P. In the early phase of an EU dossier fight every interest group can dream of an EU decision process that has befriended Persons on strong Positions in benign Procedures and thus is more comfortable to its interests. This early game of Triple P (chapter 3) is preferably done simultaneously with the following. 4 Control. As a match usually takes much time and, like in chess, is uncertain in its outcomes, it needs permanent control of its progress, including the assessment of what has been done before and what is best to do next. This control is either remotely (on shared-list) or directly (on short-list). 5 Getting the edge. Every interest group that wants to become a winner needs basic support from important stakeholders and officials. This requires a further aggregation of interests and usually indoor negotiations (‘window-in’). Then, it may come close to realizing its daydream and/or averting its nightmare. The logic of early shared-listing is evidenced here, as early partners can mobilize their own supporters and share costs. All these steps are a matter of preparatory homework and management that starts in the home organization, which is usually an arena of itself (chapter 5). Here, the PAM official or desk supports its board on getting its outside world internalized by strategy and its interests externalized among stakeholders and officials [Pedler, 2002, 4]. Both require an intelligent and prudent use of influencing tactics and techniques (chapter 6). Public affairs management can be defined now as the systematic methodology to get a desired outcome from a power-holder. Its target remains the successful influencing of a power-holder outside, but its design is much more than a new label for old wine. True enough, the basic idea of PAM, namely that one has to be prepared studiously and organized prudently for a fight outside, 62

Chapter 1

The Europeanization of Public Affairs

is far from new. In the past the Chinese chief commander Sun Tze recommended this to his Warlord, as Niccolò Machiavelli did to Lorenzo dei Medici, the ruler of Florence, and Ignatius Loyola to the Roman Catholic Church. The effectiveness and efficiency (2E) of any effort to influence the outside world they all considered as mainly dependent on both sufficient critical ‘mass’ (M) or support from other interest groups and informationbased intelligence (I2), in short on the formula 2E = MI2, to paraphrase Albert Einstein’s one. Under PAM, however, the old wine is produced along a more robust and sophisticated methodology, derived from both the logic and the experiences of winning games in a complex world such as the EU. The promotion of 2E is a matter of disciplined creativity and thus requires expertise. This short appetizer of public affairs management and lobbying needs six shading comments from empirical observation. – ‘Lobby’ has many meanings. The word is widely used for both the bad practises of bending the rules by illegal and illegitimate behaviour and the good ones of changing them in legitimate and legal manners. Traditionally, it refers to only (mediaeval) corridor behaviour, which today amounts to only an estimated 15% of the work time of a PA manager [Spencer and McGrath, 2008, 18]. Some use the word as title for totally different contents, such as institutions, actors, policies and issues [Coen and Richardson, 2009]. Others take the word as pars pro toto for all PAM or even as synonym for any effort to influence; we too shall do this sometimes, because ‘lobbying’ reads easier than the slang of ‘PAM’ or ‘PA’. – Not the same as real influence. PAM and lobbying should never be equated with real influence. It is only an effort to create this and without guaranteed success. In academic language, the effort is neither sufficient nor necessary for success, but only helpful. The chance of success increases with the quality of the effort. Using only traditional techniques, including mediaeval lobbying, usually gives results that are poor or caused by good luck. Embedding these techniques in and enriching them by the methodology of PAM tends to give outcomes that are much better, although bad luck can always happen. If PAM creates success without any lobbying, for example when one has got the power-holder to anticipate one’s interest and to comply to it [Friedrich, 1963], it even becomes an art. – Interactive shuttling. The definition may suggest a one-way flow of actions, but this is only the initial pattern. In reality, usually a two-way flow and often multiple flows of communication and interaction happen between senders and receivers. The window-out lobby for information and the window-in one for support then turn over into a mutual 63

The Europeanization of Public Affairs

Chapter 1

exchange of information and support. Interest groups, stakeholders and EU officials each play – even on the same dossier – the dual roles of sender and receiver of messages. – Only few all-rounders. Figure 1.2 contains five main steps and, as will be shown in the chapters to follow, each of them contains numerous smaller steps to be chosen rationally from menus of options, based on logic and expertise. All steps require a lot of preparatory homework and field activities. Ideally, a PA expert is a master of all. The reality is much different. By far most PA experts are, for whatever reason and to put it mildly, specialized in only a few steps and skills, for example only representing the interest group, lobbying for information, making a longlist, running the PA desk, studying issues or briefing its management. Only a very small minority can be said to be all-rounders. In a large PAM team the limited but different expertise of every member might make the team all-round. – Only few full-timers. Most people that practise PA management partially or fully do this on a part-time basis, as side-activity of another job such as general manager, civil servant, secretary of an association or policy analyst. Such a broader role-set can positively contribute to their PA performance, unless their part-time PA work makes them less all-round or even less professional, as often happens. – Only few PA professionals. Most people that say to influence the EU by PAM do this not by systematic methodology but by trial and error or, as amateurs, at random. Quality of PA performance is our landmark of professionalism, which is still in development [McGrath, 2005-B] and not identical to acting all-round and/or full-time, let alone to having a job in PAM. The (like in any sports) small ‘premier league’ of best professionals is marked by, for example, its double use of the order of figure 1.2 by simultaneously thinking backwards from the fifth to the first step and acting forwards from the first to the fifth step. The EU Commission has a reputation for such backwards planning of its precooking of the EU decision processes. To this we return in the next chapter.

XI Extra: The Growth of Public Affairs Management in Europe The foundations of public affairs management have a long history, older than its name, as exemplified by Machiavelli’s three virtues: the ambition to win, study of the arena and prudence during the match. The modern basics come from two main sources. The first is academic and has two origins: business management and political science. This should not be sur64

Chapter 1

The Europeanization of Public Affairs

prising, as the relationships between business and politics (or, more broadly, the private and the public sector) form the crux of any modern society characterized by mixed public-private governance. After World War II, scholars (primarily from the United States) developed the field of PAM [Dean and Schwindt, 1981; PARG, 1981]. Those in the study of business management took the perspective of business and particularly that of big companies and trade organizations. They took them in their dual roles of sender and receiver of influence efforts, as business both influences and is influenced by its environment. The focus of their research was on how these organizations (can and/or should) cope with the threats and opportunities coming from that environment. The political scientists took the perspective of government and focused mainly on its receiver’s role. Due to their democratic bias that a government should not act as a pressure group against society, many political scientists implicitly denied (and still deny) the government’s role as an influential pressure group. With regard to its receiver’s role they conducted many studies of private interest groups, pressure groups and lobby groups [Laurencell, 1979]. They questioned how government officials, elected or not, (can and/or should) cope with pressures and lobbies from the private sector. This focus on profitoriented organizations they extended to NGOs later on. These two areas of academic study spilled over into Europe from the US via the UK. The second main source is practice, again starting in the US, after World War II. Due to a lack of formal training in PAM, many people in the private and public domains simply had, and still have, to find out how best to cope with threats and opportunities coming from their challenging environment. Through daily experiences they collected empirical insights and wisdom, often resulting in ‘do’s and don’ts’. In the private sector, many joined peer group organizations, such as the Public Affairs Council (1954), so sharing their personal experiences with others. In the public domain, many officials politicized the supposed threats from business lobbying. In an effort to cope with these, their governments often took the traditional technique of coercion and issued regulations, registration forms and codes of conduct regarding lobbying [US/Congress, 1977; US/GAO, 1999]. Again, the UK became the first country in Europe to follow the US example in practising PAM. The first stimulus came from the 1964 Labour government, which pursued a range of antagonistic interventions in business, including the nationalization of private companies. The second stimulus came, paradoxically, from the 1979 Conservative government of Margaret Thatcher. She wanted to create a clearer wall (‘two worlds’) between the public and the private sector, and to abolish the organizations representing labour and management [Grant, 1989]. Large UK companies felt compelled to follow the US example of PAM. 65

The Europeanization of Public Affairs

Chapter 1

Thanks to internal and external developments, the small family of people interested and engaged in PAM in Europe experienced rapid growth and spread since the mid-1980s. On the internal side, the often separate worlds of academia and business became linked. In the UK the European Centre for Public Affairs (ECPA), founded in 1986 and originated at the University of Oxford, became a major meeting place for both groups. From its start it launched conferences, workshops and studies in the field of PA, and soon began to take in the wider EU level. Another major British initiative was the launch of the Journal of Public Affairs (JPA) in 2001, the first in its field. Another internal development was the dissemination of PA experiences among small groups of academics and practitioners in north-western Europe. In academia, more people study and research the practises of lobbying and PAM in Europe [Coen, 2007; Griffin, 2005; journal JPA]. Following the example of the ECPA, practitioners in many countries began to meet in peer groups in order to exchange information and share experiences. In the Netherlands, four different circles of PA practitioners (in big companies, public interest groups, trade associations and at the junior level) have been formed since 1985, all called and numbered after one of the country’s many kings named ‘William’. Similar groups in Europe exist now in Germany (like the Adlerkreis), France (like AFCL) and other countries (see List of Best Websites). Although PAM is evolving, it is still seeking to clarify its identity in academia and in practise [McGrath and others, 2010; McGrath, 2009]. On the external side, the changing methods of government in many countries in north-western Europe have become a major factor of growing awareness of the potential of PAM. Although happening a little later than in the UK, the pattern of change has come close to the British one. Many companies in Scandinavia and the Netherlands became unhappy with the government’s politicization of the economy in the 1970s and were subsequently surprised by swings towards liberalization, deregulation, privatization and public budget cuts in the 1980s [Van Schendelen and Jackson, 1987]. They could hardly control both changes and perceived the national governments as acting arrogantly, unpredictably and unfriendly. Strong antagonisms started to rise. Also in the 1970s, new NGOs were born, particularly in fields such as health and environment, which had to fight for a position at home. Hardly invited but all the more driven by their desire, compulsion and capacity, these companies and NGOs searched for new ways to influence the domestic system, and many discovered PAM. In general, the combination of a closed government and irritated pressure groups gives strong impetus for the adoption of PAM. Another change came with the EU’s launch of a Single European Market (SEM) in the late 1980s. Thanks to the previous changes at home, the private organizations 66

Chapter 1

The Europeanization of Public Affairs

in the northern countries could easily apply their PAM to the then largely unfamiliar EU.

XII Public Affairs Management Practises at the EU Level The EU is a most appropriate system for the application of PAM and the common capital Brussels is an ever more relevant place for acting [Van Schendelen, 2008]. Whether interest groups appreciate this relevance or not, is irrelevant for the appreciation of PAM. The antagonists even more than the protagonists of the EU should appreciate the expertise of PAM, as swimming against the stream requires more ambition, study and prudency than going with it. Anyhow, in the EU many interest groups feel capable, desiring, compelled or invited to participate and to influence it somehow. Unhappy about being ‘on the menu’ and having to accept any outcomes, they want to be ‘around the table’. Due to the variety of Europe, their interests will inevitably clash with those of others and cause new issues. Willy-nilly the pressure groups push europeanization forward by their issue-building. Bringing their interests into the EU, they apparently consider the EU method to be more effective and efficient than any traditional integration method. They also exhibit a great deal of diversity of influence behaviour in the EU. Some tend to choose from the menu of traditional influence techniques the ones they use to apply at home, sometimes even illegal or illegitimate ones. Others rely mostly on the mediaeval lobby and still have to discover the PAM methods for improving their chance of success. Under pressure of many factors, some convergence of influence practises is evolving, however. The EU disciplines much lobbying, for example by its procedures and rules on transparency. Cat-burgling groups are rapidly expelled by other stakeholders and EU officials and caught by journalists. A closed door or public scandal is an excellent lesson. The multinational and multicultural EU shall yet continue to allow some diversity of influence practises. Nobody knows exactly the quantity of interest groups and lobbyists that try to influence the EU system [Berkhout and Lowery, 2011 and 2010; Wonka and others, 2010]. A few sources give a range of rough figures. In 2011 the Commission and the Parliament launched a joint European Transparency Register (ETI) of interest groups. In June 2012 more than 5,000 entities had registered. Other directories, such as the European Public Affairs Directory (EPAD, 2012) and the new Stakeholder.EU (2012) come to different numbers of entities, not based on self-registration but on their own research. The three sources largely contain only Brussels-based entities. The breakdown for type of entity also varies per source, due to their 67

The Europeanization of Public Affairs

Chapter 1

different method and to the short-term volatility (around 15%) of entities that leave or enter the Brussels scene. The official ETI gives (at the end of 2012) the following breakdown of registered entities: business groups 47%, NGOs 28%, consultancies 11% and the remainder, such as thinktanks, government groups, churches and academic platforms 14%. The two other sources also enlist the permanent representations of the member-state governments, around 150 missions of foreign governments (such as the United States, Liberia and Turkey) and offices of domestic government entities, not all of which are requested to register. The EU clearly is an open area and Brussels an open city. Quoting Eartha Kitt, ‘They all do it.’ In the recent past, many interest groups felt particularly attracted by the EU policy fields of internal market, environment, health, agriculture, social affairs and transport [Fligstein and McNichol, 1998] and there is no indication that this has changed much. Runners-up in the mid-2010s are regional governments and domestic agencies in areas such as transport safety, food health and land registry. The numbers and figures above only roughly indicate the presence of lobbying people. According to survey data, most companies and EuroFeds have full-time staff of about six people and consultancies about ten people [EurActiv, 2009-A-B-C], but these surveys over-represent the established and Brussels-based entities. In the 1990s we assumed publicly for the (then about 2,500) Brussels-based entities an average staff of about six people and concluded that there was a total of about 15,000 lobbyists in Brussels, a figure that is still sacrosanct in the EU but not solid. All these entities and lobbyists are, however, only part of the fuller story. There are certainly many more than about 5,000 pressure groups in the EU. A study of more than 2,000 large companies reveals that only 7% have a Brussels office [Bernhagen and Mitchell, 2009]. According to the ETI about 60% of the Brussels-based entities come from Germany, France, the UK (each around 15%), Belgium and the Netherlands (each around 7%), and 40% from all other countries [Wonka and others, 2010]. Coming from an old and wealthy member state near to Brussels is apparently a stimulus for having a presence. By far most pressure groups rely on their national or European umbrellas or specific platforms (shared-list) and only occasionally send somebody to Brussels. To give one number: roughly 100,000 experts representing their public or private interest group attend meetings of the about 2,000 Commission’s expert groups. The total sum of all entities and people acting only part-time at EU level is really impossible counting, not to mention the EU officials and MEPs that often act as part-time lobbyists for themselves or others. What about the PAM quality of all these interest groups and lobbyists? On this question only indicative research is available. A survey among 91 68

Chapter 1

The Europeanization of Public Affairs

regional and local offices shows that most are primarily ‘listening posts’, particularly for EU subsidies, and only hope to shape EU decisions but are rarely decisive actors [Marks and others, 2002]. A survey among 149 thinktanks shows that their large majority only produces studies and training materials for domestic clients, must continuously fight for budgetary survival and have little firepower at EU level [Boucher, 2004]. According to a questionnaire among 140 PA practitioners, most of them say that they do not have a formal PA function and do not measure their PA performance, which indicates a low status of their PAM [ComRes, 2008]. My experience as a frequent participant observer at the sender’s side of different pressure groups in the EU is that, indeed, most PA managers are still far away from figure 1.2. They are busy with drafting introverted position papers that disregard the many stakeholders and officials in the EU, use their introverted long-list largely as a short-list that inevitably exhausts them without much return on their efforts, and choose from the menu of influence techniques those they use at home, to mention but a few observations. They may be ambitious, but fall short on the virtues of study and prudence, in short on PAM and are not part of the ‘premier league’. Many even have only modest ambition and feel satisfied with simply profiling their organization in the EU area (many regions and cities), writing ‘fact sheets’ (a lot of consultancies) or representing their organization as experts in a EuroFed (many companies and NGOs) or expert group (most ministries). Both the quantity and quality of PAM practises at EU level are gradually growing. From the twelve new member states many more interest groups self-reliantly enter the EU and discover PAM. Private groups from north-western Europe that initiated PAM in the EU have become leaders inside the Brussels-based peer groups, such as the Society of European Affairs Professionals (SEAP), the European Public Affairs Consultancies’ Association (EPACA), the European Federation of Lobbying and Public Affairs (FELPA) and the European Centre for Public Affairs (ECPA). Brussels-based national and regional groups on PA, such as the Dutch European Affairs Platform (EAP), exert a peer function too. Brussels-based PA consultancies, often having a north-west European background, sell their skills and are pushed to higher quality by their reputation among clients. Similarly, market competition pushes the quality of the weekly newspapers European Voice and New Europe, monthlies like Public Affairs News and EU Reporter and the quarterly Europe’s World. French-speaking groups have followed Britain’s lead by establishing in Brussels their own training centres, consultancies, peer groups and information services, such as the daily Agence Europe, the website Euractiv.com, monitors like European Information Services and Europolitics and the (now British) European Public Affairs Directory. The new directory Stakeholder.EU is a German 69

The Europeanization of Public Affairs

Chapter 1

initiative that indicates growing PA awareness there. Regular surveys among PA people about their PAM are held by EurActiv and the British ComRes. In short, the infrastructure of PA at EU level is growing by quantity and quality. The clients are mainly companies and NGOs from initially the north-western and gradually more north-eastern European countries. Once groups from new countries start at EU level, they are often followed by critical (and initially moralistic) journalists at home, who willy-nilly disseminate the message of PA at EU level back home, where after more factual reports by journalists and academics on what PA practitioners do in the EU bring the acceptance of PA into balance. Translations of this very book are another indicator of rising attention paid to the subject. There even seems to be in development a sort of premier league of EU pressure groups trying to manage their public affairs more professionally. Usually they have learnt their EU lessons through games lost in the past. Instead of giving up, they braced their nerves and body. Many have tasted the honey of fewer threats and more opportunities, less losses and more wins. Now they keep open their organizational windows for going out and getting in. These advanced professionals are, once again, not necessarily the same as those having a permanent office in Brussels or a full-time job as EU lobbyist, acting commercially or otherwise. Groups and people without a facility in Brussels and operating from elsewhere or even parttime can be excellent players. Yet, the reputation of professionalism is often assigned to full-timers with a facility in Brussels. They feel at ease there and learn from each other. EU public affairs management is, of course, no panacea. If applied professionally, it provides only better chances to win and not to lose an EU game. The professional PA manager operates by a set of methodological prescripts. Our approach, however, is not normative but advisory. Every interest group is free to manage its PA amateurishly, to follow its old preference for a traditional influence technique, to solve its challenges by non-EU activities, to feel satisfied with merely getting adapted to EU outcomes, or to rely on the old non-EU methods of solving irritations in Europe. However, if it subsequently fails to secure its interests, it should not reproach the EU but only itself. If it wants to improve its chances of success, it should practise the insights from PA management professionally. In this sense, our approach is advisory. The next step is to introduce the EU playing-field.

70

Chapter 2 The Playing-field: EU Common Decision-making In order to gain influence in a Principality, the skill of gaining familiarity with its court, dignitaries and officials is crucial. This old wisdom of Machiavelli (Prince, chapter 7) is also true for the EU, where nobody can score without intimate knowledge and information about how decisions are made. Like in the old courts, its internal ‘flesh-and-blood’ can be much different from its formal ‘skeleton’. How does the EU work?

I Deus Ex Machina The basic principle of EU integration is that every EU decision that is based on competences and procedures agreed upon by treaty text, legally binds the member states and overrules the domestic laws and acts that are not in accordance with it. The principle works, however, in a more complex way because, due to the various treaties and their differentiations, more than one legal EU order exists [Avbelj, 2011], each with a division of competences between EU and member states. Potential conflicts between them are solved by the Court and ultimately by modifying a treaty. The principle is most crucial for EU integration, as the about 2,500 new decisions produced annually would otherwise remain just on paper. As main place of production, Brussels can hence be called the capital of the member states. The interest groups wanting to influence the outputs of the EU machinery must possess useful (valid, reliable and applicable) knowledge of how they come into being [Lindblom and Cohen, 1979]. Otherwise an interest group cannot effectively influence the authorities and officials inside. This chapter is a crash course on the workings of the EU machinery. What sort of machinery is the EU? Three basic characteristics we emphasize. – Not an ordinary system. Uninformed newcomers from inside or outside EU often take it as a system like their national capital or an international organization. That is a mistake. Functionally the EU is different from both, so it could be called a system sui generis, without parallel [Almond and Powell, 1966], a sort of ‘non-state statehood’ [Neyer and Wiener, 2011]. Its main function is to make common decisions that bind the member states somehow. Therefore, it articulates and aggregates the interests of all sorts of interest groups, including governments. 71

The Playing-field: EU Common Decision-making

Chapter 2

Its administrative function is often shared with the member states. By recruiting its people from all member states, it functions like a multinational organization. It largely lacks the functions of mass socialization and communication as exist in national systems. The EU is not (yet) a union of widely shared values, and its few EU-wide mass media are chiefly English such as the BBC, Financial Times and CNN, being rather atypical for European views. – Artificial and natural. The EU machinery was originally constructed by the six member states (France, Germany, Italy, Netherlands, Belgium and Luxembourg) that came together in the 1958 Treaty of Rome and its forerunner, the 1952 European Coal and Steel Community (ECSC); in hindsight, these agreements were the pilot for the new method of integrating Europe better. (Throughout this book we refer to treaties by their year of coming into force.) Many new treaties have followed thereafter [Laursen, 2012; Luitwieler, 2009; Beach and Christiansen, 2007; McDonagh, 1998]. As a construct the EU has the age of an average subSaharan African state and is still in development. Two sorts of dynamics brought the construct to natural life. One is the internal dynamic created by spill-over effects from used competences towards new agendas and decisions [Niemann, 2006; McCormick, 1999, 15-17]. For example, the 1987 SEA Treaty for a single market created internal dynamics forcing police co-operation, codified by the 1993 Maastricht Treaty. The other dynamic is the external one coming from interest groups which articulate and aggregate their demands to the EU. The story of ‘EU-ization’, once created by the vector of six national governments, is now written by all vectors of europeanization. – Skeleton versus Flesh-and-Blood. Treaties define the formal structure of the EU, such as its institutions and authorities, their competences and their procedures, in short its ‘skeleton’. Its real functioning by ‘flesh-andblood’ is often not so much in contradiction but different, as everything not forbidden by treaty is allowed. For example, in the Council of Ministers the ministers formally decide upon legislative proposals, but the substantial decision is usually made at the lower level of working groups of national civil servants. Such differences between ‘formalité’ and ‘réalité’ are normal for every living system, such as any house reflects both its architect and its inhabitants. Uninformed newcomers often mistake the outside façade for the inner workings, and the formal powers for the real influence. Our focus here is not on the broad concept and practise of European integration, but limited to the basic characteristics of the EU machinery as it operates in practise. From it emerges like a god, but in this case a goddess 72

Chapter 2

The Playing-field: EU Common Decision-making

called ‘Europa’, the miracle of European integration. This is the playingfield for all PA management of EU affairs, the theme of this book. Our objective is also not to describe fully and in detail the machinery’s many aspects, issue areas, changes, causes and implications, on which exist various text books [e.g. Hix and Hoyland, 2011; Nugent, 2010; Dinan, 2010] and dictionaries [McGowan and Phinnemore, 2006; Bainbridge, 2002; Dinan, 2000], if this objective could even be possible at all for the ever changing and prismatic EU [Bellier, 1997]. For the PA management of the EU, it is seldom necessary to know all the ins and outs of, for example, voting points at Council level, as their relevance depends on the specific situation and thus is a matter of specific homework. In anatomizing the machinery, we shall thus pay less attention to its formal structure or skeleton than to its working practises or flesh-andblood. True enough, the skeleton of formally established procedures, powers, positions and people may set serious limits on the flesh-and-blood practises, but this usually only happens when the machinery operates under serious strains and tensions, just as the physical skeleton can set limits on a tired body. Such crisis moments are exceptional rather than regular. Usually the EU institutions are energetic rather than tired, test their skeleton to any lengths and establish different flesh-and-blood practises. Interest groups, acting from outside the EU skeleton and thus being dependent on its flesh-and-blood, usually have a long-term interest in keeping the system relaxed and not strained or tense. Basic knowledge of the skeleton may help in setting limits they consider useful, but that of flesh-and-blood practises is truly crucial for influencing the EU.

II The Skeleton of the EU Machinery: Power The architecture of the community method relies on four common beliefs of the original six member-state governments in the 1950s and laid down in the 1958 Treaty of Rome, a treaty for economic co-operation. – Primary law above all. Any common decision must ultimately be based on treaty text, called primary law, which is concluded by the memberstate governments and ratified by domestic procedure. Any subsequent legislative proposal based on treaty text, called secondary law, the governments control by their Council of Ministers, where they pooled part of their national sovereignty [Keohane and Hoffmann, 1991]. Initially they decided largely by unanimity, which implies veto power for every member state (called intergovernmentalism). Today the Council decides increasingly by qualified majority voting (QMV), an about 70% majority of voting points allocated to each government according to roughly 73

The Playing-field: EU Common Decision-making

Chapter 2

the size of its population. Also increasingly it does so with the European Parliament (EP) as co-legislator (codecision procedure), which implies that every member state can be overruled (called supranationalism). The volume of this legislation by the EP and/or the Council is annually around 15% of the aforementioned total of 2,500 binding decisions. – Autonomous Commission. It still bears the imprint of its French design. The drafting of common decisions must be managed exclusively by a secretariat, now called the Commission, independent of national politics. Its Board of Commissioners (the College) and staff of civil servants is recruited from the member states. It has decision-making powers of its own, assigned to it by treaty text (particularly on competition) or secondary law. They empower it to make binding delegated and implementing acts, the latter largely produced through special committees (comitology) with representatives of the member states. The total of delegated and implementing acts is annually around 85% of those 2,500 binding decisions. – Consultations. Both main political parties and major interest groups from the member states must be involved. Its design bears a German imprint and resulted in an Assembly of national parliamentarians, now being the EP and directly elected on a national basis since 1979, the Economic and Social Committee (ESC) that represents major national socio-economic interest groups and, since the 1993 Maastricht Treaty, the Committee of the Regions (COR) that represents regional and local governments. The three institutions are asked for their opinion on planned decisions, as specified in treaty text (consultation procedure). Only the EP has, since the 1987 Single European Act (SEA), gradually got the much stronger position of co-legislator next to the Council (codecision procedure). The three representative platforms are autonomous in their internal workings. – Hierarchy of norms. The common decisions must legally bind the member states, as otherwise there is no new method of integrating Europe. For this purpose, the member states also established the EU Court of Justice (ECJ) which in early rulings confirmed the priority of EU law over national law. The treaties are the ultimate norm, followed by the (secondary) legislation by the EP and/or the Council and lastly both the delegated and the implementing acts. The binding decisions can have one of three forms: regulations that fully bind directly, decrees that bind specific groups like farmers directly, or directives that bind indirectly through national acts that implement them before a deadline. All other texts, such as policy papers, guidelines and recommendations, are not legally binding.

74

Chapter 2

The Playing-field: EU Common Decision-making

Two remarks on nomenclature are useful here. One regards the EU secondary route of legislation by the EP and/or the Council, which should not be compared with the British secondary procedure that, by the absence of written primary law, comes after its parliamentary procedure and is in fact delegated legislation (like the Commission’s delegated acts), a confusion created by many a British textbook. Secondly, there are the new wordings and concepts as introduced by the 2009 Treaty of Lisbon. Legislative proposals approved by ‘the legislator’ EP and/or Council (secondary law) are called legislative acts now and are produced by either ‘ordinary legislative procedure’ (OLP, formerly ‘codecision’) or ‘special legislative procedure’ (SLP, formerly ‘consultation’). All formerly called ‘delegated laws’ or ‘administrative laws’ made by the Commission, are now called non-legislative acts and are split into the two categories of delegated and implementing acts, the former being the autonomy of the Commission and the latter to be produced by comitology. All acts bind the member states legally. The new labels bring the meaning of ‘legislation’ closer to national grammar, where it usually stands for laws produced by government and parliament and where administrative acts that come next are often called ‘rules’. In the future the easiest phrasing for the two might be ‘laws and acts’, the first produced by legislative and the second by non-legislative procedure. In this book we shall use both the old and new wordings, because the old ones are still part of the present and the new ones are largely only new labels. The main substantial change is the splitting of the old concept of ‘delegated legislation’ into delegated and implementing acts. Since the 1958 Treaty of Rome, the member-state governments, whenever agreeing upon a new treaty [Christiansen and Reh, 2009], have redesigned the skeleton by modification or codification of existing practises; the former really changes existing practises and the latter formalizes the changes that happened already. An example of the former is the 1987 SEA allocation of (preliminary) codecision power to the EP; and an example of the latter is the now formalized position of the European Council. Most changes have not been radical but incremental or step-by-step, such as the 1999 Treaty of Amsterdam that cleared leftovers from the 1993 Treaty of Maastricht. Other treaties took bigger steps. One example is the 1987 SEA, which launched the policy programme on the Single Open Market combined with (limited) codecision. Another is the 1993 Maastricht Treaty that established the Committee of the Regions (COR) as a platform of regional and local politicians, and brought sensitive policy domains, such as foreign-and-defence politics and justice-and-home affairs, under modest europeanization by establishing special regimes under the Council (called Pillars II and III, making all other domains Pillar I). The third 75

Chapter 2

The Playing-field: EU Common Decision-making

Main Decision-making Flows between EU and Member Countries Binding Decisions by: Treaty (Primary Law)

‘Legislation’ (Secondary) ‘draft phase’

by OLP (only EP) by SLP (all three) OLP = codecision

EU Level

Natl. Level

Commission College/ Cabinet DGs/ Interservice expert committees advisory agencies

recruitment consultations lobbying

EP/ESC/COR EP committees and intergroups ESC sections COR sections

recruitment hearings consultations lobbying

SLP = consultation

By OLP or SLP

(about 15%)

Council Coreper working groups General Secretariat

Legislation

‘Non-legislation’ Delegated acts Implementing acts

‘inspection’

(about 85%)

Jurisdiction

Figure 2.1

76

PR delegations recruitment lobbying

govt. co-ordination NP Private groups

Commission DGs comitology executive agencies

‘implementation’ (only directives) law enforcement

Non-Legislation

Court

Natl. Courts

Chapter 2

The Playing-field: EU Common Decision-making

example is the 2009 Treaty of Lisbon [Craig, 2010]. Cases of its modification are the ending of the Pillars by bringing all domains under one flexible framework of inter-institutional co-operation, the creation of both an EU Council President (the first being Herman van Rompuy) and a ‘double hatted’ Council High Representative for Foreign Affairs and Security who is also the Commissioner for External Relations (the first being Catherine Ashton), and the introduced ‘subsidiarity procedure’ that entitles at least nine national parliaments to halt a legislative proposal. Codification cases are the intensified cooperation on justice, police and home affairs, the Court’s jurisdiction over special matters of justice and home affairs and the Charter of Fundamental Rights. In retrospect, the old skeleton has undergone continually changing forms. For the main institutions we shall now, under reference to figure 2.1 (left side and middle), summarize the current features of their formal structures, powers and procedures as developed over time.

1 The Skeleton’s Main Components The Council The Council, hosted in the Justus Lipsius building near Place Schuman, experienced the following major changes since its beginnings [Werts, 2008; Hayes-Renshaw and Wallace, 2006; Westlake and Galloway, 2004]. 1 Enlargement. It developed from a group of six national governments with altogether a population of 195 million people into a body of 27 ones with a population of almost 500 million people, the world’s third largest population after China and India. More countries are in the pipeline of further enlargement. 2 Policy expansion. Starting with Coal and Steel in 1952 and expanding the range of its economic policies in 1958, the Council now covers almost the full range of policies that can be found at national level. The intensity of its involvement varies according to the competences assigned to the EU by treaty text (TFEU, article 3-6). 3 Multi-layered. It developed from a General Council of Foreign Affairs ministers to three layers now. On top stands the EU Council of heads of government or state, which is chaired by its President. Hereunder function the Foreign Affairs and Security Council with its double-hatted chair and the General Affairs Council (GAC), which is chaired by a country rotating every six months. Under the GAC function special Councils, such as on Transport and Competitiveness. They all are formally configurations of one and the same Council, as one God with many faces. If a special Council fails to come to a common decision, the 77

The Playing-field: EU Common Decision-making

Chapter 2

GAC and ultimately the EU Council can make the final decision on the issue. 4 Support staff. For its administrative preparation and management the Council relies on its General Secretariat (GS), which has about 3,500 staff (2012), working in different policy areas. The political preparation is primarily done by Coreper, the treaty-based committee of all Permanent Representatives (PRs), the ‘embassies’ of the member states at the EU. Their aggregate staffs have grown to about 2,000 people (2012). Coreper has two chambers (Coreper I and II), the first for ‘regular’ dossiers and the second for politically sensitive ones, and it gets itself prepared through about 300 working groups (200 standing and the others special or ad-hoc), largely composed of national civil servants plus PR staff. For special domains there are, next to Coreper, high-level committees, such as PSC on foreign security, EFC on economy and finance and CSA on agriculture. The new President of the EU Council has a personal staff of about 25 people (2012). 5 Decision powers. The EU Council sets general political priorities primarily for its layers, but has no legislative powers that might bind others (TEU article 15). The latter are to the ordinary Council, once the sole legislator, but now ever more in codecision with the EP. Its decision mode changed from unanimity to, increasingly, QMV, and in specified cases even a simple majority. This change from ‘intergovernmental’ to ‘supranational’ implies that every single member state can be overruled. In its final phase of 15% of binding decision-making, the Council thus has become more dependent on both QMV inside and codecision with the EP. In sensitive fields like treaty formation, defence and taxation it still has autonomous powers. The Commission The Commission, headquartered in the Berlaymont building opposite the Council, shows the following changed features [Spence, 2006; Nugent 2001]. 1 College. Every member state has one Commissioner. Their common College has grown from six to twenty-seven Commissioners now, each with a small personal staff, called a Cabinet and since 1999 selected from different countries. The President is in charge of the Commission’s internal and external operations, such as the division of portfolios (policy areas), the College’s weekly agenda and inter-institutional affairs, like the EU budget and the annual working programmes. The College decides by simple majority (at least 14 in favour). Special positions have the Commissioner on Competition (due to the treaty) and (since the end of 2011) the one on Economic and monetary Affairs. 2 Departments and Services. The real policy work is done by Directorates78

Chapter 2

The Playing-field: EU Common Decision-making

General (DGs), now also called Departments, and Services. Their number has grown from only a few to about 25 DGs plus 10 Services. The division of portfolios of the Commissioners often does not run parallel to that of policy areas of the DGs and Services, which results in a matrix organization that can produce both internal conflict and better policy integration. Conflict is managed by internal procedure, called interservice. Some DGs and Services are more powerful than others, such as DG COMP that has treaty powers, the Secretariat-General (SG) that serves the College and the Legal Service that can block any proposal. Inside a DG the director-general runs several Directorates and every director several policy units. The administrative culture still has the French footprint of formal hierarchy. 3 Staff. The number of statutory staff (‘fonctionnaires’) has grown from about 300 to about 25,000 and, if including external staff working for Agencies and seconded national experts (called ENDs), to almost 35,000 people (2012). They consist of managers at the highest level (5%), administrators (AD, 33%), assistants (AST, 36%), temporary contracted or seconded people (20%) and technical people (6%). From 1999 on, recruitment is shifting from national representation (fourchette) to qualifications, or from ‘spoils’ to ‘merits’ [Bauer, 2008]. Internally responsible for policy are Managers and ADs, totalling around 12,500 people; the many ASTs, one-third of all personnel, are desk managers rather than secretarial staff. Compared to national bureaucracies, the Commission’s statutory staffs are extremely small. The Dutch central government is five times larger and the US federal service more than forty times. On average a DG or Service has around 350 policy staff and a policy unit of less than 20. 4 Agencies. Agencies [Rittberger and Wonka, 2011] are specialized, decentralized bodies that monitor, execute or regulate specific policies, such as food (European Food Safety Authority EFSA), police (European Police Office Europol) and banking (European Banking Authority EBA). Their design has a British imprint. They grew from zero to about 40 (2012). Most are former directorates or even policy units of the Commission that got some independent status. They have their own policy tasks, staff (together about 6,000, year 2011) and a budget that is usually co-financed by the Commission and the member states. They are located anywhere in the EU. 5 Budget. The total annual EU budget, based on legislative decisions, has grown to about €150 milliard (or US billion, year 2012) and comes largely (65%) from GNI-related contributions from member states, and for the rest from value-added taxation, customs and minor sources. To compare: the Belgian federal budget is almost equal to the EU’s total 79

Chapter 2

The Playing-field: EU Common Decision-making

and the 2008 US federal bailout of the mortgage banks Freddie Mac and Fannie Mae was three times bigger. Almost 94% of EU expenditures are for external policy objectives, largely through structural funds like those for Cohesion, Agriculture, Regions and Social Affairs. The remainder of about 8% is for administration and half of this for the Commission, this being another indicator of its minute format. 6 Power positions. The Commission has (almost) ‘the exclusive privilege’ to draft legislation, which prohibits the EP and/or the Council to do so formally. It also has become the largest producer of non-legislative acts; its main production site is the Alfred Borchette building. Figure 2.2 presents the numbers of all laws and acts based on the community method (EU law), excluding the decisions that have another legal basis (e.g. Schengen, ECB and ECSC), for five-year intervals since 1970 and the years 2006-2011. The years 2010 and 2011 are transition years that not (yet) allow a breakdown of the non-legislative acts for delegated and implementing ones. It shows that often around 85% of the about 2,500 annually produced binding EU acts are non-legislative [Brandsma, 2010]. This share of ‘administrative legislation’ is normal in almost all countries and often higher than the EU’s 85% [Braun and Gilardi,

Adopted Binding Decisions by Procedure and Type, 1970-2011 YEAR LEGISLATIVE ACTS BY SLP*

1970 1975 1980 1985 1990 1995 2000 2005 2006 2007 2008 2009 2010 2011

NON-LEGISLATIVE ACTS BY OLP

TOTALS

(Delegated + Implementing)

A

B

C

%

A

B

C

%

A

B

76 126 145 77 127 129 192 240 239 219 220 252 243 319

24 49 64 66 70 38 44 10 26 5 14 17 8 6

256 417 443 477 405 263 194 123 165 121 141 117 84 122

40 39 29 32 28 12 12 13 15 15 17 18 17 21

0 0 0 0 0 6 16 11 20 16 28 23 40 30

0 0 0 0 0 10 34 28 37 18 55 77 15 21

0 0 0 0 0 0 16 19 44 21 48 60 35 40

0 0 0 0 0 0.5 2 2 3 2 6 7.5 4.5 4

166 247 655 388 392 384 487 465 499 447 460 403 348 346

2 8 22 23 24 34 39 54 77 53 57 70 68 69

C

%

364 60 685 61 931 71 912 68 1110 72 2810 88 2585 86 2022 86 1808 82 1413 83 1139 77 1111 74 1137 78.5 1195 75

% 888 1532 2260 1943 2128 3683 3607 2972 2915 2313 2162 2130 1978 2148

Legend: SLP = consultation procedure, OLP = codecision procedure A = decisions, decrees/ B = directives/ C = regulations * including decisions on which the Council acts alone / ** rounded total %. Source: Data compiled by Brandsma (2010) and forthcoming from Eur-Lex register of documents.

Figure 2.2

80

100 100 100 100 100 100 ** 100 100 ** 100 100 100 100 ** 100 100

Chapter 2

The Playing-field: EU Common Decision-making

2005]. The Commission gets few powers for this from treaties (mainly on competition) and most from legislative acts. The figure also shows that, contrary to popular beliefs, on average only 6% of all acts has the form of an indirectly binding directive (the columns B) and that 94% (columns A and C) binds the member states directly. An important note here relates to the reliability and completeness of the data of figure 2.2, as those published by the Commission fall short on these criteria and allow different compilations [Häge, 2011; König and others, 2006]; the figure’s data have been checked by the referred author on the basis of the original documents. 7 Non-legislative acts. Under the Lisbon Treaty (TFEU, articles 290 and 291) the non-legislative acts are subdivided into two categories [Peers and Costa, 2012]. One is that of delegated acts (COM 2009/ 673), defined as acts that supplement or amend ‘non-essential’ elements of a legislative act and which, in easy phrasing, detail ‘what member states must do’. These acts are the Commission’s competence and are usually prepared by its own expert groups (paragraph III, 2 below). Under a 2011 ‘modus vivendi’ (Council 8753/1/2011), the EP and the Council are informed before and, in cases of urgency, can make objections that block the act. The other are implementing acts (EU Regulation 2011/182) that ensure the uniformity of the implementation of legislation in the member states by prescribing ‘how member states must do this’. The Commission makes the implementing acts via about 350 especially empowered committees, called comitology, largely composed of representatives of the member states. Under the 2011 regulation, these committees have two variants. Most are advisory committees that deliver their opinion on a draft act by simple majority, to which opinion the Commission must give ‘its utmost attention’ but is not bound. The others are examination committees (formerly the management and regulatory committees) that deliver their opinion by QMV and, in case they oppose, forward the contested act to the Appeal Committee (EU 2011/182) that advises by QMV and is chaired by the Commission. The Commission informs the EP and the Council about the draft acts so that they can make objections (right of scrutiny) if the draft exceeds the implementing powers provided for in the secondary law and, if the Commission maintains the draft, they might review that secondary act. The Parliament The Parliament, located in the Altiero Spinelli building at a distance from the Council and the Commission, has undergone many changes to its formal features too [Corbett and others, 2011; Priestly, 2008; Judge and Earnshaw, 2008; Hix and others, 2007; Hix and others, 2003]. 81

The Playing-field: EU Common Decision-making

Chapter 2

1 Direct mandate. Since 1979 the Members of the Parliament (MEPs) are directly elected on a national party-political basis, which affords them a direct mandate and a representative licence to operate as it wants by majority. 2 Size. The number of MEPs increased from 58 in 1958 to 736 in 2009 and to 754 from the end of 2011 to the 2014 EP elections, after which its size is to be capped at 751 (750 plus its President). It is the world’s largest parliament after the Chinese one. Due to the treaty, its plenary sessions take place in the French city of Strasbourg, which makes it a unique travelling parliament. Its total staff (personel, groups and committees) amounts to almost 6,000 people (2012). 3 Political groups. The more than 200 different national parties holding a seat in the EP pose challenges of effective management. Their partypolitical integration is promoted by internal rules that give privileges (time, staff and budget) to groups having at least 25 members and seven nationalities. Almost all MEPs belong to one of the currently seven political groups. The two largest groups are the European People’s Party EPP (Christian Democrats) and the Progressive Alliance of Socialists and Democrats S&D, together holding potentially the absolute majority. 4 Work process. The EP works mainly through about 20 standing and a few temporary committees, each getting their competences and composition by plenary decision. The groups inside a committee assign dossiers for plenary decision-making to one or a few members (called rapporteur and co-rapporteur), around whom the other groups nominate their ‘shadow rapporteur’. 5 Budget powers. The EP’s budgetary powers have always been strong in cross-national perspective, as until recently it had the final say over both the total EU budget and the specific ‘non-obligatory’ (mainly nonagricultural) expenditures. The Lisbon Treaty has extended the latter competence over all specific expenditures. 6 Legislative powers. As shown in figure 2.2, the EP’s powers over secondary legislation have developed from merely consultation (advisory) until the mid-1990s to also codecision (amendment and veto), the latter under the Lisbon Treaty covering roughly one-third of the legislative acts. Thus, consultation is still its main procedure. The full codecision procedure holds two readings plus, after a ‘conciliation’ with key people of the Council and Commission (trialogue or triangle), a final third reading, and may require from the EP an absolute majority. On dossiers like enlargement and international agreements the EP must give its consent, i.e. may reject it. Thus the EP has developed from assembly to parliament, with a strong hint of legislature now. 82

Chapter 2

The Playing-field: EU Common Decision-making

Other institutions The most important other parts of the skeleton of the EU machinery are the following. 1 EU Court of Justice (ECJ), The ECJ, located in Luxembourg, has as its primary function to settle conflicts in specified treaty fields and to control (upon request) whether EU binding decisions are in accordance with treaty texts, so making its jurisprudence a higher norm than legislation [Kelemen and Schmidt, 2012; Nowak, 2007]. It has to base its judgements on treaty texts, so inevitably earning the reputation of being ‘pro EU integration’. It suffers from an overload of cases brought to it; the total staff is about 2,000 people (2010). To address the overload better, the Lisbon Treaty expanded the ECJ to three layers: the Court of Justice for major issues, the General Court for other issues and Specialized Courts for ‘first instance’ cases [Mendes, 2011; Nugent, 2010; Arnull, 2006]. The first two layers have one judge from each member state. The authority of the Court is remarkable. Without an effective sanction system (police and jail) of its own, it gets its judgements largely accepted by civil entities, member states and EU bodies. 2 Economic and Social Committee (ESC). The ESC, located in the Jacques Delors building near the EP, is the Grand Committee that represents employers’ organizations, trade unions and various other interest groups. It can be said to be the body that has undergone the least change. Due to enlargements, it has (since 2009) 344 members. In its policy domains it has still only advisory (consultation) power. Its staff amounts to 700 people (2010). 3 Committee of the Regions (COR). The other Grand Committee, the COR, which since its establishment by the 1993 Treaty of Maastricht meets in the same building as the ESC, is also an advisory body of 344 representatives, mainly directly elected politicians from decentralized governments. It has about 500 people on its staff (2010). 4 Other institutions. A few other institutions and bodies exist, such as the European Central Bank (ECB), the European Court of Auditors (ECA), the European Investment Bank (EIB), the European Ombudsman and the European Data Protection Supervisor (EDPS). Because they act at formal distance from or, such as the ECB and Agencies, only in a specific policy area of EU, they do not need further description here.

2 The Ever-Changing Skeleton The community method of the EU is, as said (chapter 1-II), still a normative project under construction. By making new treaties the member states have pooled inside the EU previously national competences. At times of a 83

Chapter 2

The Playing-field: EU Common Decision-making

crisis – such as from the early 1950s coal and steel crisis of the 1950s, to the ‘euro-sclerosis’ in the mid-1980s, to the financial crisis at the end of the 2000s – they usually feel compelled to take a next step of further integration by pooling more national competences at the EU level. Without a crisis they hardly move forward, but may allow the Commission, unless the Court disagrees, to stretch the treaty text, which can be codified in a future treaty. Any step forwards is always the best possible last compromise that is usually welcomed by most of its institutions, governments, civil stakeholders and observers, and rightly so from a historical perspective. Yet, the current state of integration is still incomplete and imperfect, as shown by the field of the open market that, in spite of much progress (on the export of goods, post and telecom), is still full of trade barriers (like on financials, energy, patents, services and railway). New steps forwards are always in the pipelines of the Commission and the EU Council, but are often blocked by Council member states. The EU policy expansion and intensification illustrate the progress made. Policy domains expanded from only involving coal and steel to dealing with nearly the same broad policy landscape of the member states. Figure 2.3

Official Key Dates of Main EU Policy Formation Coal & Steel Open Market Competition Euratom Agriculture Transport Regional Policy Social Affairs Developmental Aid Environmental Policy Consumer Policy Monetary Affairs SM Enterprises Fishing Industrial Policy Energy Education Public Health Security/Foreign Policy Justice/Home Affairs

1952-1957 + -

Member States

1952: ‘Six’

Institutional Reform

1952: ECSC 1958: ‘Rome’ 1967: ‘Merger’

Figure 2.3

84

1958-1972 + + + + + (1962 EAGGF) + + (1957 EIB) + (1957 ESF) + (1963 Yaounde) -

1973-1985 + + + + + + + (1974 ERDF) + + (1975 Lomé) + (1973 Action) + (1975) + (1979 EMS-ERM) - (1983) - (1983) -

1986-1993 + + (‘1992’) + + + + + + + (1990 EBRD) + + + + + + (1986 R&D) + + -

1994-. + + + + + (Mac Sharry) + + (Cohesion Fund) + (Social Protocol) + + + + (ECB) + + + + + + + +

1973: + UK, DK, IE 1981: + EL

1986: + ES, PT

1995: + AU, FI, SE 2004: + 10 CEEC 2007: + RO, BU

1987: SEA 1999: ‘Amsterdam’ 1993: ‘Maastricht’ 2003: ‘Nice’ 2009: ‘Lisbon’

Chapter 2

The Playing-field: EU Common Decision-making

shows the expansion (+ sign) of EU policy domains by any new treaty; almost every plus-sign stands, however, in reality for much more intensified EU policy-making, so to say a double or triple plus. For example policymaking on competition, open markets and agriculture now largely falls under the EU, while that on health, housing and civil liberties is still mainly under national competence, and still other domains have a shared regime (TFEU, articles 3-6). For most policy domains the trend, however, is from national to first intergovernmental (Council) competences and then to supranational (Commission, Court) ones. Recent examples are financial and economic governance, consumer protection, cross-border crime and education. The 1993 Maastricht Treaty can be seen as the watershed between a mainly market-orientated EU to one also broadly covering national domains of government, such as public health, security, justice and home affairs, initially under more or less intergovernmental regime and, under the Lisbon Treaty, increasingly supranational. The wider and more intense EU policy-making is only partially the outcome of new treaties. Much more happens that is not forbidden by treaty texts, which frequently provide room for interpretation, say a menu of modes of governance, such as the following [Héritier and Rhodes, 2011; Stacey, 2010]. 1 Hard or soft. An intended EU policy intervention can get either the ‘hard’ form of a legally binding measure or be ‘soft’, for example by subsidies, benchmarks, the Open Co-ordination Method (OCM) or guidelines [Kurpas, 2008]. 2 General or specific. A policy can be designed so that it generally (‘horizontally’) applies to many sectors, trades and cases, as many measures regarding the open market, competition and environment do. The alternative is the specific (‘vertical’) design for a single sector, trade or case, as often happens for food and industry. 3 Positive or negative method. The policy can be to intervene by legally binding decisions, technically called ‘positive integration’, as often happens on agriculture and environment. The alternative is to intervene by removal of barriers to free interaction, called ‘negative integration’ or liberalization, as for open markets and education. 4 Subsidiarity or not. The choice here is between a full decision-process at the EU level or leaving substantial parts of it to domestic layers (‘subsidiarity’), as is the case with the European Agricultural Fund for Regional Development EARFD (COM 2005/ 1698) [Franchino, 2007]. Even the choice for a directive instead of regulation contains an element of subsidiarity. 5 Common or varied. The definition of ‘common’ policy is open to choic85

The Playing-field: EU Common Decision-making

Chapter 2

es. It may apply fully to all member states at the same time (the EU-27), to only a group of them (‘enhanced co-operation’) like the EU-17 for the euro or with different intensity (‘opt-outs’) or deadlines (‘exemptions’) for some and altogether resulting in ‘variable geometry’ or differentiated integration, laid down in protocols (‘protocology’) [Holzinger and Schimmelfennig, 2012; Dyson and Sepos, 2010]. 6 Standing or experimental solutions. If standing solutions of governance are seen as inappropriate, new hybrid ones may be made within the treaty framework, such as the 2002 ‘Lamfalussy procedure’ (broadened comitology) for security and banking measures, the 2004 European Competition Network ECN (DG COMP and national authorities) on competition and the 2010 Troika (Commission, ECB and IMF) for supervising Greece’s financial decisions [Sabel and Zeitlin, 2010]. 7 Treaty change or new treaty. If any desired choice above is not allowed by the current EU legal order, it needs either treaty change or, if not all member states are in favour, a new treaty. Examples of changed treaties are the ones called after a city of the Council-chairing country (like Amsterdam, Nice, Lisbon). Examples of new ones are the 1990 Schengen treaty on free border-crossing among initially six states and the 2012 ESM treaty on stabilizing the euro among 17 states. Usually the EU decision-makers quarrel over the menu and create by compromise a mixture of choices that, since the 1990s, show a shift from soft to hard interventions, from specific to general measures, from positive to negative integration, from EU regime to some variant of subsidiarity, from fully common to differentiated integration and, for the euro zone, a new legal order. The volume of hard legislation is already about that of an average member state [Neyer, 2004; Alesina and others, 2002]. The choices that the treaties allow thus provide room for creeping europeanization, which marks the transition from skeleton to flesh-and-blood. Even if there is some ‘institutional paralysis’, as happened after the failure of the 2005 Constitution Treaty that had to make the EU ready for the new enlargements, the process of europeanization (and the enlargements) could go on [Wallace, 2007]. The paradox is explained by the flesh-andblood practises that can be unintended or intended. Usually unintended is the spill-over from an EU policy domain to a national one [McCormick, 1999], such as the effects of the 1992 Open Market on criminal operations that soon required cross-border police co-operation. Intended creeps are, among more, the creative use of treaty articles and the OCM, applied to matters that fall under national competence but need EU intervention, awaiting treaty modification or codification. Exemplifying is the EU’s response to the 2007 financial crises that imperilled private finance and 86

Chapter 2

The Playing-field: EU Common Decision-making

public budget, for which the treaties (Lisbon included) provided insufficient powers. After many intense debates the member states, meeting informally and in various institutions (Basel Committee, ECB and EU Council), re-pooled their sovereignty and decided upon better EU financial and economic governance. The Commission created in 2010 four platforms for supervising private finance (banks, security markets, pension funds and systemic risk), launched its first ever proposals (‘Six-Pack’) to scrutinize national public budgets and proposed a lot of specific legislation. It is the flesh-and-blood that matters now.

III The Flesh-and-Blood of the EU Machinery: Influence An interest group reading the preceding summary may conclude that it has to target the Council in particular, as this body formally controls both the treaties and the legislative acts that follow. It would do better to read on and consider the work floors of the EU decision process, this being our chief concern from here on. The metaphor of ‘work floors’ represents the places where the work of decision-making is really done. Figure 2.1 (middle and right side) shows the main parts of the machinery with their work floors. Although decision-making is, strictly, only an early phase of the full decision process, we use it as pars pro toto for all phases, from agenda building to inspection (to which we will return in chapter 4). The skeleton may be a determinant of the EU work floors, but the flesh-and-blood makes them real. The same bodies as in the skeleton approach we now regard for their functioning and outcomes in daily practise. Two differences of approach deserve attention first. In the previous section, our key concept was formal power, as distilled from the treaties. From now on, the central concept is influence. We define it as a relationship between two or more actors: A influences B, if B’s behaviour changes (either in accordance with the wishes of A or in any other direction) due to the behaviour of A [Dahl, 1991]. Like any relationship, influence is not a constant but variable factor, dependent on time, arena, intelligence, supply side and many other factors. By consequence, only repeated and controlled observations may permit generalizations. Like any causal relationship, influence cannot be proven perfectly, but only in terms of plausibility. Formal powers can, of course, be a determinant of work floors or, in our terminology, provide potential influence. A hierarchically superior person or body may indeed determine the behaviour of a lower-placed one. But reality is frequently different, as a lowerplaced person or body can be the influential force, due to key factors like expertise, support and gate-keeping. Lobbyists are the best-known exam87

The Playing-field: EU Common Decision-making

Chapter 2

ple of persons having no formal powers and yet often exerting much influence. For explaining influence, formal powers have scant value. Secondly, we reverse our starting point. In the formal approach, the treaties and thus the Council are the engine of EU decision-making, while the other institutions like the Commission and the EP have more or less derived positions. In the work-floor approach, we focus on the EU decision process as it evolves from start to finish. The best metaphor for it is the initially blank A4-format piece of paper. At the end of the process this may be published full of text as a binding decision in the Official Journal of the EU. Filling in the paper usually starts inside or around the Commission. Insofar as the EP and/or Council have any say over a proposed decision, they usually give this as a last say at the end of only the legislative (secondary) pipeline, when the paper is already full of text. In the majority of cases (around 85%) of non-legislative decisions, they have no other final say than to complain or to demand a review of the law that empowered the Commission. The difference between the two approaches is illustrated by the saying among PA practitioners in the EU, whose daily job is to influence the machinery, that ‘the most important people are not those who sign the decision, but those who write the text’.

1 The Polycentric and Multifunctional Commission Except in two respects, the Commission is a rather normal bureaucracy, such as most member countries have [Kurpas and others, 2008; Spence, 2006; Nugent, 2010 and 2001]. One exception is its multinational and thus multicultural composition with a touch of cosmopolitanism [Suvarierol, 2011]. In 2010, all member countries have to live with a Commission composed of about 95% or more staff coming from other countries, except Belgium (about 20% countrymen), France and Italy (each about 10%). At all levels the officials reflect the social and cultural variety of Europe and show that this can result in a productive apparatus, favoured by factors like common location, basic esprit de corps and a sense of pragmatism [Egeberg, 2012; Hooghe, 2012]. The second exception regards the relatively under-resourced nature of the Commission in terms of budget and staff. Due to its scarce resources, the Commission has a strong appetite for information and support from outside, to which we shall return below. The Commission, looking formally so hierarchical and united, operates in reality highly polycentrically. Viewed vertically, it has different layers. The President’s style affects the College, as exemplified by the contrast between Jacques Delors, who turned it into the engine of ‘Open Market’ legislation in the early 1990s, and José Barroso, who bypassed it to please the Council in the mid-2000s. Nominated by their national government, 88

Chapter 2

The Playing-field: EU Common Decision-making

the Commissioners usually have a party-political background and are negotiators rather than technocrats. In the College they run a portfolio, not a DG, and often have to cope with various DGs that cover part of their portfolio, which brings to many DGs the reverse challenge. As in any normal bureaucracy, the substance of policy work is done at the middle level of a DG. Most important here are the chef de dossier, usually an Administrator (AD), and the Assistants (ASTs) and the Head of Unit. They organize the draft papers for the Senior Managers, who normally make comments in the margin (and hence might be called ‘marginal’). A legislative draft is usually pre-framed by guidelines from the College and/or DG, part of an annual work programme, based on a management plan (‘roadmap’) and subject to an impact assessment that includes a consultation of stakeholders. When ready, the drafts go into the interservice, the euphemism for bureau-political involvement of other DGs. The SG and Legal Service are always involved, DG Budget and DG HRM if budget or staff is at stake, and Cabinets and Commissioners thereafter. The President can delete any proposal from the College’s agenda, as José Barroso did in 2007 regarding ‘controversial’ proposals such as on Cross-border Health-care (COM 2008/414) that could have disturbed the Irish, the EP elections and his chance of re-appointment; his gate-keeping resulted in a low EP workload in 2008. The College really votes rarely, as it prefers broad consensus, if necessary by package deal. On cases of non-legislative acts, it usually mandates (by habilitation) the responsible Commissioner(s). The Commission’s component parts have, in short, much room to manoeuvre. Viewed horizontally, every DG has its own set of policy tasks and structures to realize them. Usually this segmentation by policy sector does not fall in line with that of the substance of policy fields. For example, the europeanization of working hours for cross-bordering truckers falls under DG MOVE, but is also an issue for DG EMPL, DG MARKT, DG ENTR and DG EEAS. The result is frequently bureau-political infighting among the DGs and even inside a DG, as happened in 2003 inside DG MARKT when (French) opponents to the planned directive for Services on the Internal Market (COM 2004/2) launched their deviant paper on Public Services (COM 2003/270), which was soon withdrawn. Sometimes a DG succeeds in claiming a new policy area, as DG ENVI did in 1998 on Biotechnology, to the regret of DG ENTRE, DG AGRI and DG RTD; in 2010 it lost it to DG SANCO. In 2012 the DGs CLIMA and ENER fought openly over the sideeffects of biofuel and the DGs MOVE, CONNECT and ENTR over vehicle emergency-call systems. The Commission, in short, has a rather flat workfloor. By taking a diagonal view, one sees structures and cultures that crosscut horizontal layers and vertical units. Examples of such structures are the 89

The Playing-field: EU Common Decision-making

Chapter 2

joint groups of Commissioners and high-level Director-Generals, meetings of different Cabinets and DGs, comprehensive impact assessments and policy reviews prepared by the SG. The most important cultural crosscut among officials is double loyalty, which may quadruple. Nationality has always been one [Page, 1997], as still applies to the recruitment of Commissioners and Cabinets, but is weakening at the lower level since 1999. Secondly, there is much regional loyalty, as officials from federal countries often feel loyalty to their region rather than state. A third variant is sectoral loyalty, as shown by the ‘freaks’ on environment at DG ENVI, health at DG SANCO and farming at DG AGRI. Both regional and sectoral interest groups try to parachute their experts into relevant positions inside, to which the END positions provide opportunity. Finally, there are informal loyalties, based on personal characteristics such as education, age, friendship, seniority, housing, hobbies and character [Suvarierol, 2009; Christiansen and Piattoni, 2003]. The higher the number of loyalties and the more they crosscut each other, the more they all produce scrutinized and impartial networks. This seems to be the trend indeed [Geuijen and others, 2008; Trondal, 2008; Suvarierol, 2007]. The polycentric Commission does much more than drafting decisions and executing them after adoption. It operates, in fact, in a multifunctional way. Politically it is the main aggregator of issues raised by public and private interest groups, and also of the expertise and support that are needed for drafting proposals so that they can gather broad support among EU officials and stakeholders. Administratively it supervises the enforcement of binding decisions, for example by delegated acts (its competence), implementing acts (comitology) or setting up networks of EU inspectors that inspect domestic ones [Keessen, 2009]. Its economic function is to allocate discretionary (i.e. within its formal limits of budget and policy programmes) subsidies, tenders and procurements to interest groups. Managerial functions regard, among much more, its internal apparatus and the information flow to the EP and the Council, such as its annual working programme (which covers, however, only two-thirds of legislative plans, thus about 10% of all laws and acts in its pipeline). It also is a plaintiff and adjudicator in certain policy fields, particularly on competition issues. The Commission is, in short, somehow involved in all ‘trias politics’. Legislation, however, can be considered the Commission’s main function, for four reasons. 1 Legislative acts. It holds the (near) monopoly on drafting proposals. With this edge on agenda-setting and gate-keeping it has hidden veto power [Burns, 2004]. Usually the legislator approves the largest part of any proposal (starting after ‘whereas’), as random comparison between 90

Chapter 2

The Playing-field: EU Common Decision-making

a draft and the final version shows [Bellier, 1997, 110; Kassim, 2001, 15]. A main explanation of this success is the Commission’s informal anticipation on policy demands from inside the EP and/or the Council. Exceptions to this pattern are so scarce that most insiders know them by name, such as the directives on Reach (COM 2003/ 644) and Services (COM 2006/ 123 and forerunners 2004/1, 2). 2 Delegated acts. On these the Commission sits firmly in the driver’s seat. Under the 2011 ‘modus vivendi’ the EP or the Council can, on urgent matters, make objections and block an act, but their impacts are modest. Its forerunner, the Regulatory Procedure with Scrutiny (RPS) was seldom really blocking. In the year 2009 (COM 2010/ 354) only 131 acts (out of 1,808) fell under RPS and only one case (Energy Labelling Directive for Television) was finally blocked by the EP. 3 Implementing acts. On these the observations are not much different and show that the ‘flesh-and-blood’ of comitology is different from its ‘skeleton’ [Blom-Hansen, 2011]. On the advisory committees the Commission has always been decisive, as it only has to pay ‘utmost attention’ to their advice. The former management and regulatory committees (now ‘examination committees’) seldom overrule the Commission. In the year 2009 they rejected 87 proposals (out of 1,808), more than usual. One explanation is that the Commission anticipates well what committee members want; another that these members have a low level of confidence in what may happen after rejection, when the EP and/or the Council might formally outvote the Commission. In 2009 the Council really blocked 4 of those 87 cases and the strongly divided EP (no majority in favour or against) none. Under the 2011 new procedure, the EP and the Council cannot intervene if a committee disagrees with the Commission’s proposal, as such conflict is handled by the Appeal Committee, chaired by the Commission, but the EP or the Council may appeal at the Court, as the EP did successfully on a Commission decision on border control (‘Schengen’) in 2012 (Case C-355/10). 4 Discretionary decisions. Within the limits of a legislative or non-legislative act the Commission makes countless discretionary decisions, for example for awarding a subsidy or procurement. Until 2011 the now supplementary acts fell under this discretion. An example is the allocation of slots to airports (COM 2007/ 704 and 2008/ 227). The new procedure for delegated acts has codified this discretion. Quantitatively the Commission clearly is the main EU legislator in practise. It has a strong footprint on the 15% legislative acts and produces almost all 85% non-legislative acts directly (by delegation) or indirectly (by comitology). Quantity is, of course, different from quality. Some suggest 91

The Playing-field: EU Common Decision-making

Chapter 2

that the non-legislative acts concern mainly non-political, routine or technical details and are ‘low’ rather than ‘high’ politics [Nugent, 2003, 241]. In its procedure for delegated acts (COM 2009/ 673) the Commission itself defines this area as ‘non-essential elements’. The mentioned suggestion is true by tautology: ‘high politics’ is what high politicians like MEPs and Council ministers do (approving legislative proposals) and ‘low politics’ is what lower-ranking administrators do (making non-legislative acts). However, there exists no objective criterion to assess what is essential (or not), as this is always subjective to everyone inside or outside the EU [Warntjen, 2011]. Here are a few illustrative examples. The Council and the EP regarded the 1989 Council Framework Directive on Safety, Health and Hygiene in the Workplace (EC 89/391) as ‘high politics’, but it mainly empowered the Commission to issue through its advisory committee under comitology almost forty detailed directives that the interest groups of employers and unions considered to be more relevant [Daemen and Van Schendelen, 1998]. The EP and the Council also regarded the 2002 European Food Safety Regulation (EC 2002/ 178) ‘high politics’, but it hardly contained any substantial decision on food safety as it was about allocating powers to the Commission and its new Food Safety Authority (EFSA). The ‘highest political’ issue in the Council was the location of the EFSA, in Italy or Finland. The substantial decisions could be made afterwards under the flag of ‘low politics’, at a distance from the EP and the Council and to the great pleasure of many interest groups. The devil is usually in the details, and so is the angel.

2 The Commission’s Assistant Layer of Experts How can the Commission produce its big volume of outputs via such a small-sized policy staff, less than 20 policy staff per policy unit? The Commission totters permanently on the brink of both volume and content overload. To survive, it applies various solutions. On the one hand, it reduces the overload by standard methods like setting priorities, combining issues and phasing policy actions [Deutsch, 1963; Easton, 1965]. On the other hand, it expands its capacities in two ways. One is by outsourcing parts of policy work to national authorities (implementing directives, inspection and enforcement) and consultancies (research and advice). The second way is by in-sourcing people from the outside by four techniques: (1) hiring temporary personnel, (2) setting ENDs to work inside, (3) holding public consultations and (4) inviting experts from interest groups around its tables. The first two fall under personnel policy. The third technique attracts more interest groups beyond the circles of the established ones and the fourth takes place through many expert groups and comitol92

Chapter 2

The Playing-field: EU Common Decision-making

ogy committees [Larsson, 2003; Van Schendelen, 2003 and 1998]. By the last two techniques the Commission creates an administrative culture of wide consultation (COM 2002/ 744) or ‘deliberative democracy’ [Tanasescu, 2009] and receives from the many interest groups highly diverse expertise and signs of support at very low cost. For example, under the fourth technique the close to 120,000 experts (in expert groups and comitology), who meet on average around three times a year, are reimbursed for only their marginal costs (hotel, food and ticket), which add up to an estimated average of €50,000 per group or committee annually, less than one consultant’s report. If all experts worked for the Commission on average ten days a year, they would represent a staff equivalent of about 5,000 full-timers; this workload is a low estimate [Heard-Lauréote, 2010]. The last two techniques we shall detail now. Public consultations are an old technique to get broader input, both institutionally (ESC and COR), semi-permanently (OCMs) and ad hoc (hearings, open days, meetings and the like) [Tanasescu, 2009]. New is the online consultation (OC) that started in the early 2000s, boomed since the 2006 E-Participation Initiative and now comprises about 130 consultations annually, each with a short deadline, open to interested persons, groups or organizations and attracting usually hundreds of responses [Quittkat, 2011]. Via a questionnaire attached to a communication on a new matter, the Commission applies them in its earliest policy phase. The OC on ‘An Integrated Maritime Policy’ (COM 2007/ 575) coined this early method as blue paper. On its website (SG and Your Voice in Europe) the Commission publishes the respondents and its planned next steps (‘roadmap’). The OC receives mixed appreciation. Some established interest groups regret its open-endedness, their own free delivery of information without a corresponding chance of influence, and the influx of new challengers. Some MEPs and member states show mixed feelings about the many OCs in their ‘backyard at home’ [Bouwen, 2007]. Commission officials in charge of OCs feel overburdened with the volume of all responses (that they increasingly outsource to firms) and/or with their contents. The 2010 OC on the review of the Tobacco Products Directive (EC/ 2001/ 37) resulted in 85,500 responses, with 96% of citizens strongly supporting a milder regime. In its 2011 report on this OC the responsible DG SANCO refused to accept this citizens’ response (largely from Italy and Poland), and concluded to the opposite that its strict regime enjoyed popular support. Some officials also regret that stakeholders cannot be involved online in negotiations, as this would require meetings. To them the OC is not an alternative but a supplement that precedes the work of the expert groups. To the interest groups the blue paper gives an early warning about ‘what is coming up in the pipeline’. 93

The Playing-field: EU Common Decision-making

Chapter 2

Expert groups, also bearing names like ‘task force’ and ‘interface’, are formed by the Commission for giving it expertise and advice on its planned legislative proposals (first box of figure 2.1) and delegated acts (last box of figure 2.1) [Gornitzka and Sverdrup, 2008; Larsson and Murk, 2007]. The groups symbolize the contrast between the EU’s skeleton and its flesh-andblood as they, lacking formal powers for taking binding decisions, assist the Commission in filling blank A4-format paper with proposals that usually become binding later on. Each group is directed by a ‘chef de dossier’ having a mandate within a few crucial limits (pickets) set by the upper levels of the Commission, such as policy frame, legal basis, budget and time line. An expert group tends to work in three consecutive (‘Cartesian’) steps. It starts with putting a broadly shared definition of ‘the problem at stake’ on a green paper, followed by selecting solutions to that problem in a white paper and finally considering the pre-draft text of the proposal. The real number of expert groups is unclear. On its 2012 register the Commission’s SG mentions more than 800 (narrowly defined) expert groups. In addition there are more than 500 ‘subgroups’ and the (not yet registered) ‘special groups’ with a different status or role, the ‘social dialogue’ groups, the joint entities and the ad hoc groups. Broadly defined for their function of ‘giving expertise and advice’, inside estimates come close to 2,000 expert groups. This number would imply that a DG or Service holds on average about 55 expert groups, which is less than three per policy unit. They are composed of people, who represent a major stakeholders’ platform, organization, domestic authority or personal expertise. To newly arising and weakly organized interest groups the Commission can grant a temporary subsidy for getting empowerment. In 2010 the total number of experts is estimated at about 100,000 people, hence around fifty members (or alternates) per group. According to older research the experts come from (roughly fiftyfifty) interest groups from government (central and decentralized) and civil society (business and NGOs) [Van Schendelen, 1998]. Recent research shows roughly the same picture [Gornitzka and Sverdrup, 2011]. Political parties are not represented. Comitology committees, which exist since the late 1950s and became codified through the 1987 SEA in 1989, have to approve the Commission proposals for implementing acts (lowest box of figure 2.1) [Blom-Hansen, 2011; Vos, 2009; EP, 2009-B; Christiansen and Larsson, 2007; Bergström, 2005; Larsson, 2003; Andenas and Türk, 2000; Vos, 1999]. Their number is officially about 300 but, including autonomous subcommittees, is closer to around 350. Membership is officially one per member state but, including alternates, almost double, thus all together close to 20,000 part-time experts. Usually the members officially ‘represent their member state’, but under this formula it is up to the national capital to decide who may sit 94

Chapter 2

The Playing-field: EU Common Decision-making

behind its national flag. Thanks to their domestic PA and lobbying, often decentralized governments and private groups get their expert inside. By legislative act the membership can also directly be assigned to non-governmental organizations, as applies to the Advisory Committee on Safety, Health and Hygiene in the Workplace that must have members from ministries, employers and trade unions equally. Only representatives of political parties are fully absent in these committees.

3 Intermezzo: The Commission’s PA and Lobbying by the ‘Brussels Method’ The way the Commission uses stakeholders in expert groups for drafting legislation and delegated acts or for deciding upon the latter, to be called here the Brussels Method, is as old as it is unique. The first parliaments in Europe were essentially budget committees (‘no taxation without representation’) with experts representing citizens [Bisson, 1973]. They are a case of functional democracy that allows ‘outstanding people’, now called representative experts, to be consulted or to co-legislate. By its quantity and quality this Brussels Method is without parallel in the member states and hence unique. Now we summarize how it is engineered if its order and form are perfectly applied. 1 On a new controversial matter the head of unit of the responsible DG appoints a lower official to manage, as ‘chef de dossier’, the process of consultation. 2 If found useful, an online ‘open call for interests’ is launched (blue paper) in order to attract new and even more diverse stakeholders. 3 The access to expert groups is equally open to all stakeholders that believe to have an interest and thus it is not selective (cartel-like) to only a few. 4 If many stakeholders from one sector or region show up, they must nominate one as their common representative, which keeps the size of the expert group manageable. 5 The chef watches over the key pickets set by the Commission for promoting its own agenda, such as the policy frame, legal frame, available budget and planned deadline. 6 The members meet in sessions, thus in a parallel setting that allows a discursive and competitive exchange of views on facts and values. 7 Collateral or derived matters of interest to the members the chef sets in series for a future agenda. 8 The chef invites the members to arrive at a more-or-less common definition of the problem (green paper), solution for it (white paper) and agreement with the pre-draft. 95

The Playing-field: EU Common Decision-making

Chapter 2

If perfectly engineered, this Brussels Method creates great benefits to both the Commission and the stakeholders. An expert group, then, is a place for the mutual exchange of support and information, so creating a two-way PA process. With minimal investment of staff the Commission gets maximal quality of highly diverse field expertise and signs of support and the stakeholders get a fair chance to put some footprint on the pre-draft. The open access automatically creates a diversity of interests, which saves the Commission from an interest cartel and gives the different stakeholders a title to be represented. By blue paper and small subsidy the Commission can mobilize new interest groups. By demanding from similar stakeholders to enter common networks and to nominate a common representative, the Commission gets a better-integrated Europe; and those stakeholders, sharing their interests, get more influence at lower costs. By setting pickets for the deliberations, the Commission manages the expectations of the members, who from their side pragmatically focus on their room to manoeuvre. By setting stakeholders in parallel and interests in series, it creates a compromising culture (‘everybody wins and loses something’) that returns to it a feasible and legitimate outcome and to the members a piece of the cake ‘either now or then’. The Cartesian procedure from green to white paper and draft text disciplines and speeds up the process, which is welcomed by all. Internal competition and external representation make the process fair and transparent. The greatest paradox is that, by its Brussels Method and supplementary consultations, the Commission’s ‘chef de dossier’ usually holds the driver’s seat in the process [Lintner and Vaccari, 2007]. In short, the Commission is leading by numerous anonymous chefs and so a case of ‘anonymous leadership’, which is something hardly described in literature. The Brussels Method is sometimes applied imperfectly. Some chefs misuse or abuse its application. For example, in the mid-2000s chefs (‘freaks’) inside DG ENVI and DG SANCO restricted access on certain dossiers (e.g. Reach and Tobacco) to befriended stakeholders only, which resulted in significant legislative problems of acceptance and implementation later on. Experts can fail to behave as professionals as well, such as when they deliver false information, do not keep promises or refuse to compromise. In such cases the head of unit or the other experts usually intervene soon enough and try to correct it; otherwise, they will lose the benefit of the method. In the domain of comitology committees the Brussels Method is usually applied imperfectly, as their access tends to be under the control of the national governments. Often the Commission corrects this shortcoming by also involving its expert groups on texts for the comitology procedure, as this is not forbidden and thus allowed. The sociology of groups and comitology committees makes the legal dis96

Chapter 2

The Playing-field: EU Common Decision-making

tinction between them diffuse [Van Schendelen, 1998]. Both have the potential to be highly influential (or not). By defining ‘problems’, suggesting ‘solutions’ and framing a policy line, expert groups often have a strong impact on the Commission’s proposals. Advisory committees can do so by, for example, delaying their position and examination committees by threatening with a blocking QMV. Expert groups are often involved in drafts for comitology, while committees may be asked about early policy ideas. Some experts sit in both variants. Many hardly know the legal status of their group or committee. Their interest is not formal power but influence. In short, powerful examination committees can lack influence, and powerless expert groups or advisory committees can be influential. Their influence depends on many variable factors. Individual experts gain influence if they are seen also as representative, which image they like to cultivate but is always contestable. Experts from governments and NGOs that tend to behave as trustees rather than delegates, are challenged on this [Warleigh, 2003]. Thanks to the groups and committees the Commission itself looks to be a representative expert. The understaffed Commission can be assessed as a most intelligent interest group (see figure 1.2). It organizes its intelligence lobby not by going window-out itself, but by openly inviting different stakeholders to expert groups. In the less open comitology committees it makes the best of it. The many controversies are housed in heavy deliberative arenas that give better information and indicators of support (MI2). By its public consultations it gets even more diverse information that enhances its intelligence (I2). It stimulates stakeholders to enter or establish platforms, thereby creating aggregated supply lines of information and critical mass (M) [Coen, 2007]. In short, it organizes its own 2E well. It also promotes the bundling (or engrenage) of domestic bodies into an EU network [Leonard, 1999] that may develop into an EU Agency with its own deliberative arenas. For every upcoming arena it sets in advance by Triple P its pickets of law, policy, time and budget. By its prudent ‘divide and rule’ in Machiavelli’s style, it usually comes to sit in the driver’s seat. The ‘chefs de dossier’ act like public affairs managers and altogether like a big PA desk. With a small investment on reimbursements and management, thus with high ‘intelligent laziness’, they can monitor countless different interest groups, collect diverse information and support, test the feasibility of their ideas, and influence the participants coming windowin to it. Almost for free they get a high-quality assistant workforce that helps to depoliticize issues in a Cartesian (rational, scientific and technical) way, to combine argued consensus and bargained compromise [Prins, 2008] and to get their proposals and decisions largely accepted by the EP and the Council, whose members always see domestic stakeholders on the 97

The Playing-field: EU Common Decision-making

Chapter 2

Commission’s side. The Commission is a talented manager of interest groups [Bouwen, 2009], as the latter like to accept its invitations that provide low-cost channels to influence each other, the Commission and, ultimately, the Official Journal. All this is unique for the EU and different from most national practises that tend to allow entrance to only a few stakeholders that are structured not in parallel but sequentially. In its intelligent approach, the Commission is cleverer than most member states [Lord, 1998, 29].

4 The Parliament’s Influence Role The around seven political groups in the EP are rather loose networks of more than 200 parties from the member countries, thereby contributing to European integration [McElroy and Benoit, 2011; Judge and Earnshaw, 2008; Steunenberg and Thomassen, 2002]. They are steering rather than working platforms. The two largest groups, the (Christian) EPP and the (Socialist) S&D, command an absolute majority in theory and often in practise. Between 1979 and 2004, cohesive voting by these groups increased from on average 82% to 91%, with the other groups at lower rates and altogether with national voting on the decline [Hix and others, 2007], also after the 2004 enlargement [Hix and Noury, 2009]. Bloc voting is even higher on power issues (‘housekeeping’), such as on the distribution of chairs and rapporteurships and on the negotiations with the Commission and the Council. The high group cohesion looks miraculous, as the MEPs are not easily socialized due to their national mandate, their high turnover of around 15% during term (five years) and 40% at election time and the cross-pressures from both other national parties inside and domestic interest groups outside, which circumstances could make them trustees with free mandate rather than delegates. The miracle has some explanations. Lacking the right to form the College, the EP is not divided by the Trojan Horse of a ‘government coalition’ versus an opposition. All groups can freely enter broad coalitions. They also have a common sense of urgency to get the big EP well-organized and influential. The steering power of the groups is anchored in the EP’s Rules of Procedure and the coordinators that give the cues for voting [Whitaker, 2011, 22] and sanction their MEPs by granting or withholding important positions. The competition among the groups is overarched by the need to create a majority for winning in plenary. By their party-political background most MEPs are trained power-players and negotiators. The about 20 EP standing committees plus a few temporary ones are the main work floor, each being a kind of micro-parliament for a specific policy field [Corbett and others, 2011; Whitaker, 2011]. Their composition 98

Chapter 2

The Playing-field: EU Common Decision-making

roughly reflects the groups’ proportional share in the EP [Yordanova, 2009]. An MEP can be a member of more than one committee. The size of committees, on average about 50 people (wide range around) [McElroy, 2006], requires further downsizing. This is done by assigning a rapporteur for every dossier on the committee’s agenda and, if the dossier regards more committees or is heavy by either volume or contents, one or more co-rapporteurs [Keading, 2006]. Rapporteurships are scarce in practise as only about 250 reports are produced annually, mostly by German MEPs (around 20%), followed by French (13%), Italian (12%) and British (10%) MEPs [Hix and Hoyland, 2011, 58], with strong variation for seniority [Hurka and Keading, 2012]. The rapporteur is watched by ‘shadow rapporteurs’ and leaders from other groups [Jensen and Winzen, 2011]. The job of the rapporteur is to make the draft resolution acceptable to their committee and finally the plenary. The draft is made by a handful of rapporteurs and the ‘invisible’ personal assistants and committee staff [Winzen, 2011]. Usually they operate informally and proactively and under intense lobbying by interest groups, Commission officials and the PRs. Their resolutions usually get high support at committee level (between 80% and 99%) [Settembri and Neuhold, 2009] and rubber-stamping by the plenary. Part of the work floor are also the around fifty inter-groups, which have no formal status at all. Under an agreement among the political groups they can get facilities if registered, which requires having at least three political groups, disclosing financial interests and publishing their membership; about 20 inter-groups are registered. Inside the EP, they all act as an interest group of MEPs, who share a common interest, such as on disability, Roma rights or family protection. Stakeholders from outside can join inter-group meetings. The common objective is to push the interest onto the EU agenda through a formally non-binding resolution, again prepared through rapporteurship (so giving more MEPs that role), to be accepted by the plenary EP. Like any other ‘own initiative’ this cannot bind the Commission or Council as there is no Commission proposal. Yet, it frequently influences them. Often the Commission feels sympathy for such new issue-creation and stimulates it informally. After a period (usually about four years), the issue then appears on its official agenda, which makes inter-groups effective U-turns for agenda-building. A famous example of success is the inter-group on Animal Welfare [Corbett and others, 2011]. The case of inter-groups illustrates that the EP’s influence can be greater than is suggested by its formal powers that are limited to the about 15% legislative proposals and particularly the one-third under codecision. The latter mainly come from a few DGs and fall under a few EP committees. 99

The Playing-field: EU Common Decision-making

Chapter 2

According to the EP’s own statistics for 2004-2009 almost 500 codecision cases had been tabled for plenary vote. Of these, 63% were concluded in the first reading (similar to consultation), 30% in the second reading and, after the conciliation behind closed doors, 7% in the third reading [EP 2009-A; Rasmussen, 2008; Williamson, 2006]. Seldom has the EP vetoed a proposal under codecision; it did so on Voice Telephone, Biotechnology and Port Services. The potency of the second and third readings casts its shadow before, as the Council agrees with the outcomes of most first readings. The EP has spent most plenary time in those years, however, not on codecision dossiers (11%) but on communications from Commission and Council (23%) and its own initiatives (22%). Cases of its influence are also its 2012 rejection of the Anti-Counterfeiting Trade Agreement ACTA, the first under the new consent procedure, and its 2011 promotion of Finance Watch for countervailing, as a ‘Greenpeace on finance’, the European Banking Federation EBF, with a take-off subsidy of one million euro (so creating an EU GONGO). If formal powers equal influence, the EP would have been non-influential before the 1987 SEA that introduced a forerunner of codecision and would still largely be so now. Almost always, however, the EP’s influence exceeds its powers [Van Schendelen, 1984]. By its parliamentary questions and non-binding resolutions, it is a major creator of EU issues and agendas [Proksch and Slapin, 2010]. Under consultation, the EP can exert real influence by using resources such as delaying its report without which the Council cannot decide, linking it to one under codecision, exploiting any dissension inside the Council, dealing informally with the Commission and the Council and mobilizing support from interest groups and mass media [Kardasheva, 2009]. One strong factor of the EP’s influence are these interest groups that show intense activism on exchanging useful information and support with relevant MEPs, groups and committees [Yordanova, 2009] and on influencing them through also their national MEPs as ‘assistant lobbyists’ on the inside [Marshall, 2010] and/or by going through their (cross-national) EuroFed. They feed them with ready-made parliamentary questions, amendments and ideas for non-binding resolutions. If they oppose a legislative proposal, they use the about 20% of selfdeclared EU-sceptic MEPs (slightly higher than among voters) [Leconte, 2010]. A second strong factor is the EP’s capability to act as a pressure group, thanks to this interest group activism, its tight internal organization, the absence of a Trojan horse inside (a standing ‘government’ majority) and the presence of an often hostile Council. By informal trialogue (or triangle) its leaders often meet with key people of the Commission and, sometimes, the Council in order to secure a gentleman’s agreement like the 2010 Inter-institutional Framework Agreement or the 2011 modus 100

Chapter 2

The Playing-field: EU Common Decision-making

vivendi on comitology. Since 2004 (and increasingly 2009), the EP increasingly enters coalitions with the Commission against the Council on legislative (including budgetary) matters.

5 ESC and COR: More Influence than Power In the ESC, three groups of socio-economic interests meet: employers’ organizations, trade unions and various groups representing consumers, farmers, green movement and SMEs etc. They are consulted on secondary Commission proposals that, according to treaty texts, are relevant to them. They can also issue recommendations on their own initiative. As a Grand Committee, the ESC’s position is like an advisory committee under comitology: its advice has to be given the utmost attention, but not necessarily any response. Its logic of influence follows accordingly. It can exert influence if its expertise and representativeness are acknowledged and if it comes to sufficient consensus. Annually it produces about 200 reports, the majority upon request and the rest on its own initiative, about policies in fields like the environment, agriculture, social affairs and transport. The real work floor exists at the level of sections and subcommittees, led by a chairman, assisted by the bureau and, for every dossier, a rapporteur. Although its own annual reports praise its influence on EU decisionmaking, observers are more reserved [Van der Voort, 1997]. They see the ESC as rarely having an observable direct impact on Commission proposals and assess its influence as largely indirect and moderate [Westlake, 2009]: by reframing a policy climate, agenda or issue in the secondary process. It produces much output, but has difficulty in agreeing upon concrete matters, as its reports often sum up the diverse views or suggest a consensus at an abstract level. Its main handicaps are its internal divisions, challenges to its expertise and representativeness and competition from the EP. The employers’ and workers’ groups are strongly divided among each other and internally, such as those between members from rich and poor countries. Often they use their ESC position primarily for watching opponents and for blocking their preferences. Thus they exert negative rather than positive influence, find this better than an uncomfortable consensus and direct their positive messages at the Commission, EP and Council outside the ESC or under the flag of the ‘social dialogue’. This Social Dialogue has its own legal status since the 1993 Maastricht Treaty and is formally distinct from the ESC. It is only open to employers’ organizations and trade unions. As ‘social partners’ [Dufresne and others, 2006; Smismans, 2004] they meet in about 20 committees and informal groups. If they come to an agreement, they can formally break the Commission’s (‘almost’) exclusive privilege to propose legislation and, after 101

The Playing-field: EU Common Decision-making

Chapter 2

adoption by the EP and the Council, they may implement the decision themselves, which makes the social dialogue a remarkable case of corporatist-like self-regulation under EU public law [Gorges, 1996]. Since 1993, the social partners produced a few laws this way, such as the 1996 Parental Leave, 1997 Part-time Work and 1999 Fixed-term Contracts Laws and, including implementation, those on Tele-working (2002), Work-related Stress (2004) and Equal Opportunities for Men and Women (2005). Additionally, they largely produce solemn texts phrased in abstract terms without specific proposals, such as on disabled people, vocational training, and employment. Since the early 2000s the Commission helps them informally to reach agreements for proposals, so informally regaining its privilege of drafting texts. The other Grand Committee, COR, is in many respects comparable to the ESC. Acting for the interests of the more than 270 official regions of the EU, it too has an advisory position only, same size, similar work-floor operations and a high level of activity. Five differences are its relative youth (it was established in 1994, after the Maastricht Treaty), regional focus and interests, composition of mainly directly elected regional and local politicians, internal dominance by political groups like in the EP and its reputation for having influence, as shown by the impact it had on treaty formation [Bindi, 1998] and legislation [Neshkova, 2010]. It must be consulted on ordinary legislative proposals or affecting decentralized government, such as regional policy, cross-border co-operation, budget and territorial cohesion. Its own-initiative agenda contains varied topics, from environment to employment and public procurement. That the COR has been rather successful in reaching internal consensus and external impact is not self-evident. So far, the first four differences with the ESC work as beneficial resources. The COR is ambitious, gets tail-wind from growing regionalism in the EU [Knodt and others, 2011; Scully and Jones, 2010; Telò, 2001; Loughlin, 2001], holds close links with party fellows in the Council and the EP, and cherishes its image as the EU’s second parliament, a kind of Bundesrat next to the Bundestag. Of course, these influence resources may not last. One internal issue is that members from federalized countries like Germany, Belgium and Austria gain, due to domestic politics, positions inside the Council and lose interest in a strong COR that might challenge their position there. Since the Lisbon Treaty, the COR wants to develop closer co-operation with the EP. On its policy domain, the COR is entitled to go to the Court that has strengthened the position of decentralized governments (Constanzo ruling 1988/103).

102

Chapter 2

The Playing-field: EU Common Decision-making

6 The Council: More Power than Influence Formally, the Council is the body that acts as the gate-keeper for binding decisions, by both allocating powers to institutions through treaty formation and deciding upon proposals based on these powers. The flesh-andblood story is much different: treaty formation has become more difficult and the final say over binding decisions has become more limited, as only a small part (15%) of all proposed binding decisions is settled there, increasingly with the EP as the co-legislator. The current Council is a polycentric and multifunctional institution, as the following observations show [Naurin and Wallace, 2008; Hayes-Renshaw and Wallace, 2006; Westlake and Galloway, 2004]. The Council, consisting of leading national party politicians, is a multilayered construct. The EU Council comes close to an informal platform for discussion among ‘heads of government’ on internal or external crises, complex issues and new policy ideas, but without power to legislate. It is the most intergovernmental part of the whole EU. Its influence is very dependent on consensus among France and Germany, which requires frequent and intense negotiations that often fail [Tallberg, 2007; Beach and Mazzucelli, 2007; Warntjen, 2007; Elgström, 2003]. The EU President acts highly informal, also with his colleagues in the Commission and the EP. The Foreign Affairs and Security Council (FAC) holds a special position by its double-hatted High Representative and its lack of formal powers. The General Affairs Council (GAC) manages general affairs like the EU budget, supports the EU Council and handles the leftovers from the Specialized Councils. The latter are the major terminus of the legislative process and meet usually around five times a year each. Among them there is a sort of caste system, with the Economic and Financial Council (Ecofin) starring and the Youth Council closing the league. A member state rotating every six months chairs the GAC and its underlying layers and co-operates with the previous and next chair (troika), during which time it can subtly influence agendas, meetings (about 1,500 per half-year) and minutes. The rotating chair has no say over the EU Council, the FAC and the Euro-group of Finance ministers of the 17 Euro countries. If a Specialized Council meets an internal stalemate, it sends the dossier to Coreper or to the GAC to resolve it, but usually it prefers the former to the latter, as it has no control over what GAC does. The General Secretariat (GS) runs the household, administers activities and contributes to the Council’s policy domains, but on the latter it is often overruled by political bodies like the Presidency, Coreper and special bodies like the PSC [Christiansen and Vanhoonacker, 2008]. The real Council work is done at the much lower levels of Coreper and 103

The Playing-field: EU Common Decision-making

Chapter 2

working groups. The almost 300 working groups (about 200 standing ones and the others special or ad hoc) explore, according to the required procedure (unanimity, QMV or other), the feasibility of legislative proposals [Häge, 2012; Fouilleux and others, 2007]. They mark them as positive (Roman numeral I) or negative (II) and send them to Coreper, which gives them a final check, forwards them as A-point (‘settled’) or B-point (‘unsettled’) to a Specialized Council, which usually follows the pre-marking. The A-points are in the large majority, mostly concern legislation and are quickly rubberstamped at the start of the next Council meeting. The remaining B-points get Council discussion, remain mostly unsettled and go to Coreper again for resolution as A-points or to the GA Council. Only a very small minority of B-points are both formally and actually decided by the Council [Häge, 2012; Hayes-Renshaw and Wallace, 2006, 53; Van Schendelen, 1996]. The Council rarely blocks non-legislative acts and even then not decisively. Officially, the working groups consist of domestic civil servants from the capitals and the PRs but, dependent on domestic PA and lobbying, people from private groups and decentralized governments can sit behind national flags; Austria and Denmark exemplify the former and Germany, where the Bundesrat (Länder House) distributes the seats, the latter. The Councils are formalizing rather than deciding bodies. The general reason for this is that most ministers, like in their Cabinet at home, want to rely on preparatory layers as they, falling short on technical expertise, would run high risks of blunders if acting otherwise. Three specific reasons exist besides. Firstly, a Council meeting with 27 ‘flags’ around the table hardly allows debate and decision, as one round of four-minute statements by every minister would require about two hours and rigid formalization. Secondly, the composition of any Council is often unstable as, due to dynamic domestic politics, about one-third of its ministers change every year, thereby making long-term negotiations difficult. Thirdly, the main culture in the Council is the search for broader consensus that requires preparation rather than voting. The first QMV ever was taken on the 1992 Banana Trade dossier and was five years after its introduction [Pedler, 1994]. Council voting records show that the 343 adoptions (rejections remain secret) between mid-2009 and mid-2012 required unanimity in 34 and QMV in 309 cases. Of the latter, 35% were adopted under abstention or opposition from some countries (the UK the most; France always on the winning side) and 65% were still adopted unanimously (although often with pro forma reservations) [Vote Watch, 2012; also Hosli and others, 2011; Hagemann and De Clerck, 2007; Heisenberg, 2005]. One particular reason why ministers usually prefer quasi-unanimous agreements is that a formal vote by QMV or veto would reveal the losers. A minister may 104

Chapter 2

The Playing-field: EU Common Decision-making

want such a loss only for domestic purposes or for nuisance value in the Council, as the Dutch did 12 times in 1998 for the purpose of ‘getting our money back’. The broad consensus is built in decentralized and informal settings. The working groups, Coreper and informal Council meetings are used for precooking coalitions [Cross, 2011] that can evolve into standing platforms such as the Baltic States (Nordic agenda), the ‘Sparing Group’ (six to nine states wanting to freeze the EU budget) and the Aachen Group (public health). The coalitions develop from north-south to also old-new member states now [Mattila, 2009]. The Council is clearly a strongly fragmented arena [Naurin and Wallace, 2008; Meerts and Cede, 2004], full of internal rivalries among all layers and players, to be solved by argued consensus (working groups) or bargained compromise (Coreper and Council) [Dür and others, 2010; Prins, 2008; Lewis, 2000]. Major battle-fields are the following. 1 Sensitive areas. On foreign policy and defence, including international institutions like the WTO and the IMF, the Council often fails to reach another common position than a solemn declaration [Jørgensen and others, 2011]. On justice and home affairs it often failed, during the Pillar structure (1993-2009), to agree upon its agenda, let alone its common position, and welcomed informal proposals from the Commission, a practise largely codified in the Lisbon Treaty. The ‘German-French axis’ is usually one of conflict rather than co-operation but, if the latter, often effective. 2 Competence fights. The Specialized Councils engage in much in-fighting on their many overlapping policy domains. A field like GMO (genetically modified organisms) touches on so many issues (agriculture, health, research, industry, development, employment, environment, external relations and more) that it could be handled by half the number of Councils. 3 Inter-institutional fights. The Council’s rivalry with the EP and the Commission might bring cohesion to the Council in theory, but it often produces the opposite in practise as, in the divided Council, some governments take side with the Commission and/or the EP (which increasingly make their coalition before) [Panke, 2012; Naurin and Rasmussen, 2011]. The asymmetric work division among the institutions, on which each is autonomous, causes special fights. In 2008, the Agricultural Council handled a proposal from DG SANCO on the prohibition of cancer-causing substances in pesticides and declined it, thus upsetting the EP and the Health Council. 4 Domestic fights. Many domestic conflicts continue at the Council level, often due to also an asymmetric work division between the Council and the government at home. An example is that case of pesticides, an 105

The Playing-field: EU Common Decision-making

Chapter 2

issue that in many countries falls under the Health Ministry but was decided by the Agriculture Council. Germany often shows up in a Council with several ministers, also from the Länder level, all distrusting each other. The Competitiveness Council has acquired the nickname ‘Mickey Mouse Council’ as 50 to 75 different ministers from the 27 member states fly in and out to discuss different items on the agenda. As a result of all these divisions in the Council, there are no permanent winners or losers among the member states [Arregui and Thomson, 2009]. Although the post-2007 crises on financial, monetary, budgetary and economic issues are not representative of the usual EU decision processes that are the main concern of this book, they are illustrative for the functioning of the Council at the highest level. Normally, as stated earlier, in crisis times the skeleton of the EU takes priority over its flesh-and-blood. Following the experience of dealing with a number of measures under the Treaty of Nice (such as COM 2009/ 207 for hedge funds and the like) and Lisbon (such as COM 2011/ 203 on capital requirements) [Snyder, 2012], it was concluded that the current Lisbon treaty was insufficient to tackle the crises fully. In autumn 2010 calls for a modified or new treaty started to occur. The only place for this is the EU Council of heads of state, where in early 2011 its President, Van Rompuy, started a long series of summits, each about the ‘next urgent crisis’, from the defaulting Greece to emergency funding and surveillance of national budgets. Major first results are the 2011 Six-Pack of financial measures added to the 1997 Stability and Growth Pact on the monetary union (applying to all 27 member states) and the 2012 draft treaties on Stability, Coordination and Governance (TSCG) on economic measures for 25 countries (the UK and CZ abstaining) and the European Stability Mechanism (ESM) on emergency relieffunding for the 17 euro-countries. The summits show many of the characteristics above. The issue area is highly sensitive, as major national competences on public finance have to be europeanized. The EU Council must decide by unanimity, which often results in the common decision not yet to decide, until after tough negotiations a compromise is found. Around the table sit debtors and creditors (‘South versus North’), which is not an easy talk. Every head of state has to cope with opponents there and a critical parliament and public opinion at home, which makes achieving a quick consensus impossible. The outcomes for 27, 25 or 17 states show that intergovernmental negotiations easily promote differentiated integration [Puetter, 2012]. Other institutions (like the Commission and the EP) challenge the EU Council, as they want to be involved. Many heads of state, falling short of technical expert106

Chapter 2

The Playing-field: EU Common Decision-making

ise, need help from the ministers of the ECFIN Council (27), the Euro Group (17) and the technocrats of the Euro Working Group in order to get the decisions feasible and consistent. In June 2012 they did so, however, not before but after their summit, thereby creating much turmoil. One result of the crisis decision-making by the EU Council is that the Six-Pack and TSCG are brought under the supranational regime of the Lisbon treaty; the ESM falls under the Euro Group with, for emergency measures, a role for the Commission and the ECB (not the EP) and thus is a mixed intergovernmental and supranational arrangement. Another result is that the crises dossiers require so much attention from the high politicians and mass media that, under their lee, many non-crisis dossiers can proceed easier, to the pleasure of interest groups. The Council member states, as formal founders once acting as managing proprietors of the current EU, have changed from managing directors into shareholders. They are still crucial for treaty-making but see, as in the current time of crisis, the newly ‘pooled sovereignty’ move on to the other institutions, from an intergovernmental to a supranational regime. They still have final say over the secondary proposals, but this is close to the final say of an Upper House or Senate that has the last but seldom the decisive word. In 2001, German Chancellor Gerhard Schröder proposed to codify this change by making the Council an EU Bundesrat and the codecision a bicameral procedure [Rasmussen, 2011]. Like shareholders of a company, the 27 Council members decide upon statutory proposals made by an executive board (Commission) and, after settling their remaining Bpoints, they largely agree. In addition, the Council members intensified two functions that are typical of shareholders. Firstly, at the end of Council meetings they raise new points and, through (non-binding) declarations and memoranda, so try to contribute to EU agenda-building. Secondly, in the Council they play home politics to satisfy the national parliament or to bypass and overrule another ministry or major stakeholder at home. In the mid-1990s the ministers of Environment got the latter reputation, as they approved decisions that did not amuse many other ministries and stakeholders at home. More change may follow. The governments in Council might turn from shareholders to stakeholders. Most are already pressed by their national parliament and public opinion to become proactive on the Commission, and feel pressed by domestic interest groups that bypass or oppose them via the Commission and the EP. More ministries want to influence the Commission early, either through their staff on its work floors or through befriended interest groups. Many PRs, focused on Council negotiations, now try to influence the Commission and their national MEPs as well. In 2008 the Environment Council acted as a stakeholder by making an early 107

The Playing-field: EU Common Decision-making

Chapter 2

informal deal on CO2 Emissions Trade with the EP’s committee on Environment versus the Commission, the Competitiveness Council and member states like Germany. The governments or ministries that act as stakeholders early in the secondary pipeline can reap in the Council the fruits they sowed. These centrifugal forces are still countered by centripetal ones. Coreper, the GS, the GAC and the EU Council try to keep dossiers coherent and governments on board. By co-ordinating their ministries better, national governments try to decrease the export of domestic fights into EU. Yet, centrifugal forces seem to be on the winning side. The national ministries that already proactively try to influence the Commission and the EP gain followers. Their effectiveness depends on their capacity to apply public affairs professionally (to which we will return in the next chapters). To give one indicator, as long as their civil servants on the work floors of the Commission act as self-instructed trustees, they will still have a long way to go [Schneider and Baltz, 2005; Trondal, 2004].

7 Some Other Decision-Makers The Court of Justice is often a major EU decision-maker [Kelemen and Schmidt, 2012; Sweet and Brunell, 2012; Mendes, 2011; Nugent, 2010; Cichowski, 2007; Arnull, 2006]. Its decisions on cases brought to trial formally bind the parties involved. By their precedent-setting they are full part of EU law. The Specialized Courts handle hundreds of cases every year, mainly of three types: appeals from private parties against Commission decisions, citizens’ complaints in various policy fields and disputes involving EU personnel. Under the Lisbon Treaty the new General Court gets its workload both from the Specialized Courts as cases on appeal, and directly from private parties (individuals and organizations), national governments (including national courts) and EU institutions. The German government often leads the pack in Court cases. National courts, asking for an authoritative interpretation of EU law, bring in about half of all cases. The Commission uses the Court procedure frequently as a touchstone for its enforcement of competition law, over which it has its own authority. The EP and the Council go to the Court when they believe that their treatybased powers and/or positions are not sufficiently respected. Interest groups try to influence the Court by litigation and by lobbying its preparatory staff, particularly the referendaries. The three layers of the Court frequently act as an engine of europeanization for two reasons. One are its decisions which, having to follow the treaty texts, tend to be in favour of European integration. For example, its 1978 Cassis de Dijon ruling ordered the free circulation of goods, which was laid down in the 1958 Treaty of Rome but remained a dead letter. The 108

Chapter 2

The Playing-field: EU Common Decision-making

other reason is that the Court often functions as an alternative and indirect route to secondary legislation. The Commission often takes this route when it meets resistance in the Council to proposals based on treaty texts, once approved by the member states. A Court ruling then can empower the Commission to uphold its proposals ‘as ordered by the Court’, which the Council then tends to accept [Nowak, 2007]. An example is the directive on Environmental Protection by Criminal Law (COM 2007/ 51) that the Council first opposed and after a Court ruling approved in 2008. The Court is highly effective in getting its decisions accepted by contesting parties, which comply either voluntarily or under force by the domestic legal systems that act as its annexe. Even member-state governments, often reluctant to accept Court decisions in the past, usually comply now [Nicolaides and Suren, 2007]. Under the 1993 Maastricht Treaty, the Court gained the authority to impose financial penalties. In July 2000, the Greek government, disregarding the EU environmental laws, was the first to be fined € 20,000 for every day it continued to do so. The Court of Auditors (ECA) can act as a decision-maker as well. In fact, it is another fine example of the paradox that weak formal powers, now in the field of budgetary inspection, can run parallel to real influence in the long run. Its usually critical reports on EU revenue and expenditure can formally be left unread or be taken as read by the Commission, the EP and the Council. In reality, the ECA’s criticisms and recommendations often fuel the EP’s agenda and the mass media with regard to the financial administration of the Commission, and contribute to improvements. The 1999 fall of the Santer Commission started with the critical ECA report on the 1997 budget that the EP refused to give discharge in 1998. Finally, there are the special institutions, bodies and agencies for specific policy sectors, all having some degree of independence. Here are a few examples. The ECB, in charge of common monetary policy-making, is the most institutionalized case. The EU Ombudsman can inspect policy decisions in the case of a citizen’s complaint. The EU Data Protection Supervisor does so on privacy issues. Agencies that monitor, inspect or regulate policies are more or less decentralized extensions of the Commission and many of them have their own expert groups [Ongaro and others, 2012; Rittberger and Wonka, 2011; Kelemen and Tarrant, 2011]. Whatever their formal status is, they all can exert influence on other EU authorities by using their discretionary and/or delegated powers or by mobilizing mass media. They also prove that the distinction between decision-making on the one side and implementation and inspection on the other is often really artificial. Implementation is continued decision-making at the more operational level and inspection, which almost always reveals a gap between legal norms and social facts, usually feeds back into new EU 109

The Playing-field: EU Common Decision-making

Chapter 2

issues, agendas and decisions. Interest groups should be aware (most are not yet) of these bodies and agencies, which can be powerful in their field and influential on the Commission (for legislation and non-legislation), the EP and/or Council (legislation) and the Court (jurisprudence).

IV The EU Playing-Field in Helicopter View A popular question is which institution rules the EU? In the skeleton view that takes rule as formal power, the Council looks to be the dominant one, followed by the Court, the EP and lastly the Commission, positioned as service apparatus. This is true for new treaty formation and for dossiers that, like on the post-2007 crises, lack sufficient treaty basis, as only the Council can make a breakthrough [Van Middelaar, 2012]. Then it acts as temporary service-hatch for the europeanization of formerly national competences, such as on Pillars II and III from 1993 to 2009 and the TSCG and ESM in 2012. The outcome is usually that it empowers the other institutions, so making the skeleton view true for the short run but optical illusion thereafter. In the flesh-and-blood view that takes rule as influence, the answer is often that, reversely, the rank-order of influence goes from the Commission and the Court to the EP and the Council. In both approaches there is, however, not really a single answer to the question. Since its 1958 start, the skeleton holds many internal checks and balances such as veto positions of member states in the Council, the Commission’s own competences and the Court’s adjudication. New treaties strengthened the EP’s codecision, the Commission’s delegated powers, the Court’s jurisprudence and the Council’s loss of powers in relation to other institutions. The many divisions and conflicts inside and among the institutions [Brack and Costa, 2012] often make the skeleton fragile. For the EU’s flesh-and-blood it is anyhow impossible to give one single answer to the question, as influence always varies according to both endogenous factors like policy domain and procedure and exogenous ones like the qualities of both those who try to exert influence and those who undergo these efforts. The skeleton gets supple flesh-and-blood by, for example, informal trialogues or triangles inside and by interest groups outside. Ultimately, every EU decision gives its own answer to the question. By studying many cases with the use of methods from political science [Dahl, 1991], an overall answer may be reached for a period, but with many exceptions. In diachronic perspective, the answer is that the Commission has evolved into the main decisionmaker. In the future this may change, as influence can vary over time. Another popular question regards the balance between the EU and member states. In the skeleton approach, the member states’ power varies 110

Chapter 2

The Playing-field: EU Common Decision-making

with the position of the Council and their voting position inside. If they have ever had sovereignty [Krasner, 1999], they have lost (most of) it to the EU, as measured by indicators such as ‘independence, equality and veto power’ [Morgenthau and Thompson, 1985, 331-2] or ‘supremacy of the last word’ [Jackson, 2007, 10]. In the Council they all need support from other governments and institutions in order to get their wishes granted, have competences that by treaty vary with policy domain, are treated unequally by voting points and thresholds, can easier be overruled by QMV than by veto and see their last word be subjected to much appeal. On this second question the skeleton view thus gives not one answer but many [Laursen, 2011]. In the flesh-and-blood view, a member state’s influence is never equivalent to its national government’s position in the Council, but always the outcome of all domestic efforts to influence EU decisions. A ministry, regional government, company or NGO can be highly influential on a dossier on which other groups from its country fail to have an impact, while the success balance is reversed on another dossier. The question can better be rephrased to ask about ‘the balance between the EU and member countries’. The answer then depends on the PA qualities of the component parts of a member country, the reception at the side of the EU and the many specific circumstances in between (see the next chapters). It is never one single answer, but always extremely varied. Paradoxically, member countries can gain more influence in the EU if their states pool their sovereignty at EU level. Although there are no simple answers to the two questions, it is possible to typify the EU playing-field in four catchwords that enlighten its characteristics and the potentials for PAM and lobbying. They are: relevance, complexity, dynamics and openness to influence. 1 Relevance. The EU’s relevance is firstly based on its functioning as onestop method of integration that largely replaces the five old methods and produces a volume of weighty laws and acts that bind the member states. For the volume we refer to figure 2.3 and for the weight to the creeping europeanization before and particularly the rolling one thereafter (chapter I-VII). Measuring the volume is already difficult [Töller, 2010; Brandsma, 2010; König and others, 2006] and, if only the about 6% directives are taken (as some do), fully invalid [Christensen, 2011]. The numerous creeping and rolling effects are nearly impossible to measure. The difficulty of finding, reversely, domestic policies that are free from any EU footprint, indicates the EU’s relevance too. Secondly, relevance is also based on contents but this is a matter of taste and thus subjective. Interest groups (and citizens) value the EU decisions on, for example, health, trade or finance very different, ranging from irrelevant events to relevant ‘heaven’ or ‘hell’, dependent on their perception 111

The Playing-field: EU Common Decision-making

Chapter 2

of a dossier as strengthening or weakening their position in their policy area or commercial market. Prudent interest groups, thus, have all reasons to organize their radar on the EU proactively. 2 Complexity. Considering its atypical design and workings, the EU skeleton and even more its flesh-and-blood may look complex, like a labyrinth in which one easily loses one’s way. The EU machinery is strange indeed. On its input side it is highly open and informal by its endless absorption of different interests and interest groups and on its output side it is strict and formal by its enforcement of EU laws; in between it transforms the different inputs pragmatically into common decisions that competitive stakeholders support and officials approve. The term ‘complexity’ is, however, a value judgement too, made by those falling short of knowledge, just as any foreign system looks complex to a tourist. Apparently, many people are ‘tourists’ in the EU. Informed stakeholders hold a different judgement: thanks to complexity, any desired outcome might be acquired in different ways (multicausality) and the same way might lead to various outcomes (multifinality). Ambitious interest groups, thus, should get familiarity with the EU. 3 Dynamics. A popular game among EU watchers is to name the five most important recent EU changes. Unable to agree, they end up with a long list that includes shifts of formal powers, intensified policies, regionalization, enlargements and much more. The EU machinery looks like a trampoline on wheels [Stacey, 2010]. Due to the pace of EU decision-making time is often a crucial political factor [Goetz and others, 2009; Jordan and others, 1999]. According to statistics of the Commission (SG, 2007), the codecision-making by the EP and the Council is 12 months for the first reading (completing 71% of dossiers), 22 if including the second one (23% of cases) and 29 if also including conciliation (6% of cases); on the latter the Council takes the most time, on average 18 months. During 1984-2003 the Council’s controversial Bpoints caused the most delay, and the growth of QMV and codecision a rapid acceleration [König, 2008]. Many national capitals use for their transposition of EU directives more time than codecision does [Haverland and Romeijn, 2007]. Non-legislative acts, as run by the Commission, go much faster. Perhaps the best indicator of EU dynamics is that every EU handbook needs revision after a few years. The dynamics are no surprise to clever stakeholders. They watch events, search for trends and assess the dynamics as opportunities for taking beneficial chances. Interest groups should always be studious. 4 Open to influence. The reputation of being the most open (accessible and transparent) institution belongs to the EP, followed by the Com112

Chapter 2

The Playing-field: EU Common Decision-making

mission and, at a distance, the Council, the oyster of Brussels (although it is more open than most national capitals) [ComRes, 2011]. Outside their rapporteurship, most MEPs prefer to receive, however, stakeholders from home. The Commission welcomes any group of stakeholders with cross-national and preferably cross-sectoral interests [Dür and Mateo, 2012]. The Council is slightly open at its back doors of PRs and national capitals (ministries, parliament and civil servants). Interest groups want, however, institutions not only to be open but also relevant by scope and domain of influence. Therefore, they preferably contact the Commission that offers both [Eising, 2009, 147]. Anyhow, interest groups that are absent around the institutions’ tables, risk to be on the menu. The intelligent ones want to be present somehow. Of course, many factors may limit the efforts to influence the institutions and people inside successfully. They range from those at the receivers’ side of PA and lobbying, the competitors around and the home organization of every interest group. In order to widen and use these limits, the interest groups must gather PAM intelligence and expertise. That is the substance of the chapters to follow.

V Extra: Wandering Scholars If knowledge about the EU playing-field is that important (as quoted from Machiavelli and sketched above), one might expect many useful (valid, reliable and applicable) insights from academic experts. Often, however, they disagree on their different insights and, like wandering scholars in the Middle Ages [Waddell, 1952], stick to their own best view. Many of them voiced their widespread dissatisfaction in the recent past, so creating a consensus in disguise. To quote a few EU scholars: ‘we, prisoners of our concepts’ [Wallace, 1990, 19], ‘the crisis of legal studies’ [Shaw, 1995], ‘imprecision of theory’ [Schmitter, 1996, 137], ‘many either … or approaches’ [Wallace, 1997], ‘tribalism of specialisms’ [Jørgensen, 1997], ‘the near-chaotic state of theorizing’ [Chryssochoou, 2008], ‘the continuing fragmentation’ [Paterson and others, 2009] and ‘islands of theorizing’ [Holzinger and Schimmelfennig, 2012]. One cause of differences is the prismatic and everchanging character of the EU that ‘wrong-foots extant theories’ [Hooghe and Marks, 2008]. The following three causes belong to the academic world. Firstly there is the difference between the two disciplines or ‘bodies of knowledge’ that, in the past, have made the largest contribution to the study of the EU: international law and international relations. The lawyers 113

The Playing-field: EU Common Decision-making

Chapter 2

describe, explain and assess the EU mainly as ruled by primary law or treaties. They focus on formal institutions, procedures and powers, in short the EU skeleton. Scholars in international relations describe the EU as a regional case of multilateral foreign policy-making by state governments [Howell, 2000], focus on the negotiations among them and regard the EU as being led by the Council. That the two disciplines bring different stories about the EU tells more about them than about the EU. In the mid-1970s, political science came in as third discipline [Hurwitz, 1980; Egan and others, 2010], followed by sociology around the 2000s [Favell and Guiraudon, 2011; Saurugger and Mérand, 2010]. The political scientists started with the frames of both international relations and nation-state but, discovering that the EU is neither an international arena nor a nationstate but sui generis, increasingly developed their own concepts and theories [Neyer and Wiener, 2011; Wiener and Diez, 2009; Pollack, 2005; Chryssochoou, 2008; Rosamond, 2000]. Many shifted their focus (dependent variable) from integration to decision-making, broadly taken as the process of converting inputs from the member countries, including interest groups and citizens, into laws, acts and policies. From here we focus on political science. Secondly, the political scientists follow different approaches that can be divided into three categories: conceptualizing, grand theorizing, and midlevel approaches [Nugent, 2010, part V]. To the first category belong old concepts from domestic politics, as if EU politics is like politics at home, and discussed either analytical (‘what is’) or normative (‘what should be’), such as sovereignty and federation [Laursen, 2011]. New ones are multilevel governance [Kohler-Koch, 2003], consociation [Taylor, 1996] and europeanization [Graziano and Vink, 2007]. In the second category figure grand theories that each assume one coherent set of driving factors and again range from analytical to normative, such as (neo)functionalism, institutionalism, supranationalism and constructivism [Eilstrup-Sangiovanni, 2006]. The third category of mid-level approaches can overlap with the previous two, but holds a more modest focus on specific settings of EU decision-making. An old example is the threefold reinterpretation of the same few decisions [Rosenthal, 1975] and more recent ones are those on policy networks [Fligstein and McNichol, 1998], transactional exchange [Sweet and others, 2001] and influencing the EU [Coen, 2007]. The diversity of the many approaches indicates their still exploring state of thinking about the EU. Thirdly, differences of methodology and methods cause a lot of differences. Many political scientists have an appetite for constructing new theory, from which they deduct reasoned expectations (hypotheses) to be tested later or by others. If tested, the validity, being an essential ingredient of 114

Chapter 2

The Playing-field: EU Common Decision-making

usefulness, almost always appears to be fragile. It is no surprise, then, that one theory follows another, any theory is often amended after testing, and they altogether illustrate a field of wandering scholarship. Empirical (factfinding) research is on the rise, but mainly as a by-product of deductive theory and by study of EU documents, such as regarding comitology [Brandsma, 2010], legislative output [König, 2008] and lobbying [Coen, 2007]. Still a scarcity is open-minded field research, such as by participant observation or interviewing people, in order to develop new hypotheses and ideas inductively or to test them deductively. Old examples regard the MEPs’ views on lobbyists [Kohler-Koch, 1998] and the workings of committees [Van Schendelen, 1998]. The state of research with its different conclusions about the EU is still exploring. In spite of all their differences, the political scientists display some common traits and trends by generation. The old scholars shared some beliefs: in deductive over inductive reasoning, integration rather than decision-making, skeleton over flesh-and-blood, theory construction above empirical research, general rather than specific statements and evaluation above explanation. The young ones in southern European countries like France, Spain and Italy are still in that old and often essayistic mood, but their counterparts in northern Europe tend to adopt opposite beliefs and to take the fragmented state of EU studies for granted or as a rich mosaic [Wiener and Diez, 2009; Peterson, 2001; Rosamond, 2000]. As three new trends, the northern youngsters push mid-level study, empirical research and sometimes, to get both inductive insights and deductive tests, field methods. They follow the American scholars who, long before the birth of the EU, had gathered rich empirical knowledge about their ‘multi-country’ United States by mid-level empirical field research. At one time Americans took the lead in studying the EU, such as Ernst Haas [1958], Leon Lindberg [1963], James Caporaso [1974], Glenda Rosenthal [1975] and Leon Hurwitz [1980]. Some tried to synthesize old grand theories and concepts [Moravcsik, 2000]. Promising too are the upcoming comparisons between the EU and other cases of multi-country integration earlier or elsewhere [Marks, 2012; Warleigh and Van Langenhove, 2010]. Platforms of scholars include the (US) EU Studies Association (EUSA) and the (European) University Association for Contemporary European Studies (UACES). In EU the new trends are reflected in the Journal of Common Market Studies (JCMS), Journal of European Public Policy (JEPP), Journal of European Integration (JEI) and European Union Politics (EUP). Our position goes even farther than that of the youngsters. The fragmented state of current knowledge we too take as normal for a system that is still under construction. We strongly support the trend towards midlevel empirical field research. In two respects we differ. Firstly, we want 115

The Playing-field: EU Common Decision-making

Chapter 2

the resulting knowledge to be useful, i.e. to be scientifically sound (valid and reliable) and applicable for influencing EU decisions [Heisenberg, 2008]. Secondly, in order to gain a better understanding of the EU we also want to rely on information from professional PA practitioners. Every day these people need to find ways and means to navigate through the EU machinery, thereby collecting potentially useful experiences and discoveries. Their empirical information is not yet tested knowledge but it could be used to develop hypotheses inductively and, as one piece of information can refute a theory, to test and improve academic views deductively. So far, political science hardly delivers this sort of information, as its investigation of the EU mainly focuses on the institutions, partially on its downstream to member states and hardly on the upstream from the member countries.

VI The EU Playing-Field and Public Affairs Management Returning to our four catchwords that typified the EU playing-field by helicopter view, we can now draw general conclusions that are relevant for interest groups seeking to influence the EU. The catchwords of relevance and openness to influence imply that the EU has a supply side that stimulates interest groups, once they get it on their radar, to develop their demand side. This is, in a free world, not different from the birds and bees that, according to Jewel Akens’ song, are attracted by the flowers and the trees in the garden. The natural scarcity of bread and honey also creates in the EU a strong competition among a mixed coalition of officials from inside and stakeholders from outside. The Official Journal that continuously publishes new common decisions, proves that, over the distribution of a scarce value, sufficient consensus can often be reached. This outcome looks like a political miracle and is seldom explicable by the dominance of a single or a few officials and stakeholders, just as one bird seldom takes the most from an open garden. Even if one category of interest groups, such as technical experts, businesses or NGOs, looks dominant in a field [Burley and others, 2010], there is usually strong competition inside. The miracle comes from the EU’s functioning as a political market where, different from Akens’ metaphor, two specific demand-and-supply sides come to consensus by compromise, such as price formation on the commercial market [Hauser, 2011; Gutmann and Thompson, 2010]. The EU officials want to receive useful information and support from interest groups and are willing to provide chances of influence. Interest groups want these chances and are willing to supply information and support. Negotiations among the mixed coalitions follow and often result in the 116

Chapter 2

The Playing-field: EU Common Decision-making

miracle of compromises that are formalized by the Commission, the EP and the Council (laws) or the Commission (acts), altogether a method of common decision-making in which step-by-step (incrementally) broad and thus stable consensus is rationally sought [Lindblom, 1959]. Technocrats and activists often consider a compromise to fall short of a ‘rational’ decision, as they lose part of their core value; they forget that the probable alternative would be no decision at all. Their position is like the early economists’ debate on ‘the right price’ (pretium justum) that Adam Smith [1776] brought to a close by pointing out that, under the precondition of open competition (market pluralism), the match between demand and supply sides results in the most rational price, not by value but by method. If the EU playing-field is open to all (political pluralism), it produces by its compromises the best possible outcomes of settling irritating differences in Europe. The EU’s seeming complexity and dynamics, our other catchwords, are helpful rather than disadvantageous for creating the best possible compromises. They augment the opportunities to influence. Thanks to complexity, the labyrinth can offer a way to a desired destination, and thanks to dynamics the trampoline to a better moment. The flesh-and-blood of the EU allows every interest group to be optimistic that, one way or another and sooner or later, on any issue their demand and supply may match somehow with those of officials and stakeholders, and result in a compromise, called decision. Not surprisingly, most PA experts familiar with the EU machinery are typically optimists about possible outcomes, like scientists searching for findings [Popper, 1945]. Proactive professionals try to create even more complexity and dynamics. So they set at a distance the amateurish players who, in contrast to them, often complain about losing their ways and missing their moments. This contrast unveils the first necessary ingredient for influencing the EU playing-field so that there is a good chance of success: useful intelligence drawn from collected information [Johnson, 2008]. The best access points for both tapping information and probing support that can feed one’s own intelligence (our I2) are seldom the EU front-doors and the more often their back-doors, such as the following examples show: – Early radar. Monitor broadly like a radar (not an antenna) and proactively early speeches, debates and non-binding texts for early indicators of rising issues, as final outcomes start here. Identify inside any institution new nominations, reshuffles, procedures and anything that may affect the flesh-and-blood. Watch for rising conflicts, as they can be sources of free information about the inside. Study the institutions, units and key people and their dependencies from all angles, ranging from close nearby to those in the member countries, as they may be 117

The Playing-field: EU Common Decision-making

Chapter 2

influenced from there. The Council may be the oyster of Brussels, but every member state leaks much information at home. – Work floors. Find for any legislative draft that looks relevant, the (potentially) crucial people and groups inside the Commission (like dossier chefs, expert groups), the EP, COR and ESC (like rapporteurs, committees) and the Council (members of working groups and PRs) and for non-legislative drafts the expert groups and comitology committees in charge. Anticipate OCMs, OCs, impact assessments, blue, green and white papers and the roadmaps. Discover which officials and members may be on one’s side or not. They always vary from potential friends to opponents. – Various. Watch the Commission’s interservice procedure, the agencies that can play a role at either the input or output side of decision-making, any PR acting as voice of its government, the Court that can act as legislator, the EU Council that discusses crisis dossiers with potentially a high fall-out, and the informal meetings among all of them. Network with other stakeholders, as some of them always have the edge on information. Read national newspapers that often report on events that have an impact on the EU or were not disclosed by the EU. Not least, watch attempts to change a specific playing-field by Triple P (see the next chapter). These examples are a short summary of main tapping points of information inside the arena of the EU [Hardacre, 2011] and every point requires a lot of window-out PA and lobbying. The professional interest group is ambitious in broadening its radar to encompass the arena of stakeholders interested in the same issues or dossiers, and not least to that of its home organization. On all three arenas it studiously gathers updated information, transposes this to intelligence and uses it for the window-in support lobby that is needed to move the decision in the desired direction. All the activities that start with information are a matter of PAM expertise to promote an effective and efficient influence, the 2E of our formula 2E = MI2. It is no surprise that the PAM agenda outlined in figure 1.2 needs the rest of this book to make sense. In the applied political science of influencing the EU, with a desired outcome as the dependent variable, PA expertise is the independent summary variable that differentiates the strongest between winning and losing. Even the best expertise has its limits, however. The ambitious and studious interest group must hence also act prudently and become an allaround PA expert like Machiavelli. How this exercise of prudence creates an EU variant of Darwin’s Law, under which the most expert interest groups have the best chance to survive, and whether this still meets the criteria of democracy, are questions for the last two chapters. 118

Chapter 3 Pushing the Buttons of ‘Brussels’ In a republic with a great deal of equality, without idle noblemen, many people are trusted to participate as responsible citizens and, in Machiavelli’s view (Discorsi I-55), such a system is more stable than any alternative. How much room does the EU offer to interest groups to ‘push the buttons of Brussels’?

I Managing the Crucial Variables The ultimate goal of public affairs management (PAM) is to achieve a complete victory in a supposedly interesting game. Such an achievement requires winning in three arenas: the EU, which has to allow the desired outcome; the stakeholders, which have to deliver support; and the organization at home, which must provide backing. In real life, the chance of a threefold full score is low. Competition in the EU is usually extremely strong and hard. EU officials act under many cross-pressures, stakeholders have difficulty with trusting each other and the home organization is usually divided. Every pressure group, therefore, has reason to be satisfied if it has won a game only partially and/or has maintained its position in the fighting arena and/or has kept the home organization on its side. These three – compromise, respect and backing – might be seen as second-class prizes to win, but are in reality frequently valued as the highest attainable and thus satisfying achievements. They give a prolonged licence to operate. Of course, a pressure group may lose as well by not being party to the compromise, getting disrespect from others and/or losing backing at home. But as long as it lobbies ambitiously, studiously and prudently like Machiavelli, it seldom runs this risk. In short, the full score is normally only a daydream to all lobby groups, while the full loss is the nightmare that usually only befalls the amateurish groups. Both dreams have, however, their positive functions for the professional too. Without the daydream of becoming the complete winner and without the nightmare of ending up as the complete loser, no sportsperson can ever become a formidable player. The definition of a desired outcome from the EU is, of course, up to every lobby group and can be anything. The main field of desires usually regards the decisions that bind those whom they concern and supersede domestic laws. An interest group lobbying for such decisions may desire, 119

Pushing the Buttons of ‘Brussels’

Chapter 3

for example, a change of a secondary or non-legislative act, a new Court decision or the granting of an exemption to a rule. Equally targeted may be a decision derived from superior laws or acts, such as the allocation of subsidy or levy, the granting or withholding of a procurement and the launch or postponement of an inspection. It may even be a detail in a text: a deadline, sanction, annex, word, definition, number, comma or anything else considered relevant by a player. Whether other players consider this irrelevant or silly does not matter. Due to the lack of objective criteria to distinguish between relevance and irrelevance, every value is always a relative and subjective affair. Also ‘soft’ decisions may be desired, such as a change of an agenda, seats in committees, crucial information or financial favours. From other stakeholders an interest group can desire such items as respect, information, support, network, burden-sharing and even favours unrelated to the EU, such as a contract or joint-venture, and from groups on the home front more backing and resourcing. An exhaustive catalogue of desires cannot be made. The desired outcome is not necessarily derived from a daydream and positively targeted at getting something new from the EU. It may, feared as a nightmare, also be negatively targeted as never to become reality. An intelligent lobby group always has desires in stock for both daydreams and nightmares. Such a non-outcome can be the rejection of a legislative proposal, the delay of a decision, the fading-out of an issue, the blocking of a process and anything else that produces a desired non-event. The more competitive an arena is, the more the influence behaviour is directed at preventing undesired outcomes rather than promoting desired ones. Often it is easier to bring a fear to death than a dream to life, as every single player usually has more nuisance value than pushing power. No lobby group, however, can only play the negative game as it would become an outcast then and hence it has to push for some positive outcomes as well, which may be a nuisance in a nice disguise, such as the thorough review of a policy area, a broadened stakeholder’s consultation or a more extended impact assessment. By implication, lobbying the EU has nothing to do with being in favour of or against Europe, EU integration or policy europeanization; its only concern is the promotion of one’s interest at the EU level. If this interest is in line with a current trend, one can take the opportunity of lobbying at low costs, but if it goes counter to the trend, one has to fight the threat at high costs, as swimming upstream always takes more energy than navigating downstream. By building an agenda early the most intelligent lobby groups try to create a new trend or movement that is comfortable to their interests. The positively or negatively desired outcome is, in other words, the dependent variable of PAM in the EU. This ambition can be found among 120

Chapter 3

Pushing the Buttons of ‘Brussels’

all sorts of pressure groups, from civil society and governments and even among EU officials. The big question for everyone to answer is ‘how can we achieve this ambition successfully?’ The number of success factors is countless – most are variable and many are manageable, and these last ones are, according to Aristotle in his Politics, the most relevant in daily life. In this chapter we shall apply this basic insight to the arenas of EU decision-makers and competing stakeholders (and in chapter 5 we will return to the home front). The conclusion shall be that the single strongest independent variable relating to lobbying is the expertise of managing the variables that determine both arenas. In order to influence ‘Brussels’ successfully, one has to consider the many buttons that can be pushed, both serially and in parallel.

II The Manageable EU Machinery: General Approaches Thanks to the characteristics of the EU playing-field, there is no shortage of buttons to push. In fact, there is an oversupply that requires an intelligent selection, as otherwise one may push wrong buttons, lose momentum, shake up competitors, hit a dead end, make a short circuit, irritate officials or otherwise damage one’s interests. Before one can make the best choice, one has to become familiar with the menu of existing or potential buttons. In the following we distinguish between three general categories: the actors to approach, the factors to use and the vectors to create. The actors are the people who contribute to the making of a decision, the factors are the determinants of their decision behaviour and the vectors are the newly created factors that may influence that behaviour. As with every menu, ours too is limited and not exhaustive for all items under the three categories. Besides, in the ever-changing machinery new actors, factors and vectors always arise. Our objective here is to make the reader systematically more aware of and familiar with them. In academic speak, they are the intervening variables of influencing the EU machinery, which can often easily be applied thanks to their multicausality (many causes) and multifinality (many consequences). This is a matter of knowledge and creative PA management.

1 Actors to Approach Which actors are crucial in the making of an EU decision? A short answer was given at the end of the second chapter. The long one depends on the situation of the dossier, arena, procedure, setting, time and many other circumstances. In the flesh-and-blood of EU decision-making, some patterns 121

Pushing the Buttons of ‘Brussels’

Chapter 3

and configurations are discernible. The metaphor of the A4-format piece of paper, starting blank and ending up full of text in the Official Journal, remains helpful. Many papers, of course, never get a final and authoritative signature, perhaps as the outcome of successful blocking actions by others. Every binding decision, however, starts as a first draft on paper and is the outcome of successful pushing actions. As stated in the former chapter, those who draft the decision are frequently more important than those who sign it. Our question therefore is: which actors are, not formally but actually, usually the major contributors to the draft that ends as a decision, and thus can best be approached? In almost every case of EU decision-making, people inside the Commission play a crucial role. Usually they drive the flow of secondary legislation, frame the substance of delegated and implementing acts, prepare the annual work-programme, enjoy the habilitation freedoms and even contribute sometimes to the formation of both primary legislation and Court jurisdiction. In every DG, the key people are usually the chef de dossier and the levels of managers and ASTs, the experts being in-sourced in expert groups, committees or otherwise, the staff of other DGs and Services able to intervene by the interservice procedure, and the people at the level of Cabinets and the College. For every dossier of relevance the ambitious lobby group must, preferably in the earliest phase, identify the names of these officials and their (possibly) relevant backgrounds (such as policy position, education and career) and idiosyncrasies (private values and hobbies). All this clearly requires a lot of PA for information. On an issue of food production, a lobby group must study not only DG AGRI but also, for example, DGs as SANCO, MARKT, ENVI and TRADE and one or more of their agencies. Clear enough too, this work can best be done via a platform of befriended stakeholders, in order to save costs (efficiency) and to personalize every contact (effectiveness). Often difficult is the gathering of information regarding comitology committees, these being the oysters of the Commission. All information must be transformed into intelligence. Then, one can infer that the Commission usually has internal divisions and thus many potential friends and enemies inside. After their identification, the interest group can start to mobilize their friends, to disarm their enemies, to seduce the waverers and to approach them. The key people of the EP, COR and ESC are first of all the rapporteur, the co-rapporteur(s) and their shadows; secondly, the supervising chairmen of committees and the co-ordinators of the political groups (in the EP and the COR) or interest groups (in the ESC); and thirdly, operational staff members and assistants doing lots of substantial work. For each dossier, the total number of key people in the EP is usually less than twenty. In reality they make the plenary resolution. Under the consultation proce122

Chapter 3

Pushing the Buttons of ‘Brussels’

dure (SLP) their impact may be limited to the timing and the framing of the issues or agenda, which can be decisive for the final result. Insofar as the MEPs have a codecision position (OLP), they can formally intervene in legislation and, more in theory than practise, in non-legislation (delegated and implementing acts). Here too, the ambitious lobby group has to identify at an early stage the names of its friends, waverers and enemies inside and their footing in domestic public or private organizations through which they can be influenced. All this work may seem easy once a proposal is on the table, but then the die is largely cast. During the Commission’s drafting phase the interest group should already try to identify the players who come next. At Council level the most relevant actors are those who determine its position. They are, first of all, the national civil servants sitting in the working groups and special committees like SCA and PSC. Supervised by Coreper and every PR they make in their domain the bulk of Council decisions (A-points) rubberstamped by the ministers. They are appointed nationally, even if coming from a decentralized or private organization, and can be influenced there. The PRs, all based in Brussels, have dual relevance, as they can influence both their nationals in working groups and their capital. Relevant too are the staffs of the Council’s GS that handles the dossiers ‘administratively’, a euphemism for all sorts of manoeuvring. Particularly in times of a weak rotating Presidency, these staffs take over some of its roles, such as scheduling the meetings and thus the arenas, pushing issues up or down, arranging logistics, drafting conclusions and proposing solutions. Council ministers can be approached at home through domestic parties, parliament, media and more. Thus, even the Council, the oyster of ‘Brussels’, can be opened by intelligent PAM. Thanks to the polycentric and fragmented character of the EU, many more officials can usefully be approached. At the Court, the advocates-general and the referendaries usually have a strong impact on the Court’s decisions, which in their turn influence the Commission. The researchers at the Court of Auditors offer a similar back-door access. The inspectors of inspection agencies are useful for the identification of a difference between norms and practises of implementation that always exists, and can be used for getting an issue back on the agenda. The jurists of each institution’s legal service are influential advisors to their high officials and can, by changing their mind, produce different advice. The translators convert legally binding decisions made in working language into all official languages, but no two of them have fully identical semantics and a small change of translation can make a big difference. High-ranking officials frequently play important ‘meta-roles’ in decision-making, for example by dividing work, assigning positions and allocating resources. By influenc123

Pushing the Buttons of ‘Brussels’

Chapter 3

ing them, one may change an arena and get, for example, another DG, expert group, EP committee, rapporteur or Council working group to be formally in charge. Even a small change can make a big difference for the outcome, as a minor change of course can divert a ship miles away from its original destination. Other important actors to approach are the stakeholders not formally belonging to the EU, such as decentralized governments, companies and citizen’s groups that have special links with parts of the EU machinery and with other stakeholders. Many also hold positions at the crossings of different networks and thus can be approached from different sides and be used as carriers of interests. Since every country is divided over every dossier, one can always mobilize domestic support and use these stakeholders for influencing key people in the institutions. A potential gold mine is the ordinary domestic civil servant sitting in both an expert group for drafting a decision, a Council working group for deciding upon secondary cases and in comitology for implementing cases, so covering most dossiers in a policy domain. People with the same domestic identity often meet informally, as the Irish do in their Brussels pub Kitty O’Shea and the Wild Geese. Some also play relevant roles outside the EU, for example in non-EU countries or at the global level of OECD, WHO or WTO, three platforms that often influence the EU and thus can be used as back door into the EU. There is, indeed, no shortage of actors to approach for collecting information and support. Plenty of them are available and, as no single actor is decisive, many of them are needed. This is a lot of work but, from PAM perspective, this fragmentation of power is better than a concentration would be, as it is easier to win many friends than a single ruler. By sharedlisting the intelligent interest group can downsize the amount of work to modest volume, as its partners selfishly help to get ‘the snowball rolling’ to a wider network that crosscuts institutional, national, sectoral and other borders. In PAM one has thus, in theory, every reason to be optimistic about the chances of winning something. The real puzzle is to select the most useful actors to approach, the best connections, the precise timing, the right balance between stick and carrot and much more that is important for successfully achieving one’s ambition. How to meet this challenge by study and prudence is discussed in the chapters to follow.

2 Factors to Use The number of relevant EU actors may be large, but it is modest in comparison to the sum of factors affecting their behaviour. Manipulating these factors is like farming by exploiting the natural conditions that enable 124

Chapter 3

Pushing the Buttons of ‘Brussels’

plants to grow and flower and, in the case of EU actors, more sophisticated and often more effective than approaching them directly. Increased competition among EU stakeholders produced this approach and made it a current trend. From the numerous factors, we present the following ones under four categories: cultural, formal, operational and decisional. 1 Cultural factors. Sooner or later, decision-making actors need a policy concept defining aspects of an issue, a policy value framing the decision in a specific direction and a regime value governing the process. Policy concepts often lack a clear definition, which has to be created. An example is that of ‘chocolate’, which was needed in 1999 when Belgian chocolate companies tried to get banned from their market the ‘rubbish’ of foreign (British) competitors that mislead the consumers. To the benefit of the Belgian producers the outcome became a strict definition (high percentage of cocoa) and, as policy value, labelling (to inform consumers). In the 1999 dioxin case (on polluted animal food), French farmers claimed that the sludge they used (as such forbidden, but not defined) was not ‘sludge’ but ‘fertilizer’. Other examples include the concepts of ‘precaution’ in food law [Szajkowska, 2010] and ‘tolerable risk’ in safety control (COM 2010/ 261). New policy values like health, solidarity, growth, sustainability, cohesion and security usually start without clearly defined meanings and thus invite interest groups to push one that gives a tailwind. Intelligent groups lobbying for subsidies try to fix their target to any new policy value and to get this adopted in the next framework-programme, thereby increasing their chances of ‘milking more euros’ thereafter. Cases of new regime values are subsidiarity, impact assessment, transparency, reciprocity, proportionality and better regulation. Clever lobby groups use them for dragging, like squirrels in the wood, their interests to a better arena. Then, for example, subsidiarity is used for slowing down EU policies, transparency for hindering efficient negotiations and reciprocity for retarding a more open gas or railway market. 2 Formal factors. As the skeleton of the EU can be a determinant of its flesh-and-blood, it provides potentially useful factors. Treaty texts offer much room for interpretation. For example, according to the treaties, taxation is (with a few exceptions like VAT) a national competence, but under the different names of levy or duty, a financial imposition may be an EU competence, as green groups discovered in 2008 in their fight against pollution in air traffic. In 2009 the Commission wanting to get control over ‘alternative investment funds’ (like hedge funds) but without clear treaty competence on this, took as the treaty basis the social article (now TFEU article 53) on ‘self-employed persons’ (COM 2009/ 207), so finally getting control not over the entities but (even more 125

Pushing the Buttons of ‘Brussels’

Chapter 3

effective) their managers. In 2011 the final issue between the EP and the Council on the draft-regulation on Food Information to Consumers (COM 2008/ 40) was about the procedure for taking specific measures: by this draft law (Council) or by delegated act (EP). It may also make a big difference for the outcome whether the process is based on the one or the other Treaty article, falls under the one or the other procedure, holds the EP under OLP or SLP, requires from the Council unanimity or QMV, or is run by the one or the other Commission DG, EP committee or Council working group, to mention a few potential factors. By linking them as switches and tracks, one gets repeated opportunities for blocking or pushing. Then a lobby group feeling hurt by a Commission proposal for legislation can use the EP to appeal against the Commission, the Council against the EP and the Court against all of them, as the German government did in the 1993 Banana Case [Pedler, 1994] and in the 1999 Tobacco Case. A change of the Treaty, which is the mother of all procedures, may provide many threats or opportunities, depending on a lobby group’s interest position. After nearly twenty years of stalemate, the first directive on works councils for multinational companies came through in 1994, thanks to the 1993 Maastricht Treaty’s provision of Council QMV on this matter, and the anticipation on this by the Commission, the EP and the trade unions federation ETUC [Van Rens, 1994]. In 1998, the EP decided to delay its work on the Water Framework Directive until after the Amsterdam Treaty’s date into force. The 2003 Nice Treaty that expanded codecision to food and transport policy and moved civil law issues from Pillar III to I was anticipated by lobby groups in these fields. In 2005 lobby groups, including EU officials, anticipated the possible adoption of the (failed) New Treaty of Rome, by trying to speed up or slow down secondary dossiers. In 2008, green lobby groups delayed a proposal on poultry meat that was pending in the Standing Committee on Food Chain and Animal Health (SCoFCAH, part of comitology), as they expected more scrutiny power from the new Lisbon Treaty to their befriended EP. In 2009, one day before this treaty went into force and would shift competences on matters of justice and police from the Council to the Commission and the EP, the Council issued its common position on various pending issues. The Commission (and the EP) took it as read and soon published its own work programme (COM 2010/ 171). 3 Operational factors. In their daily work, EU officials are very dependent on resources such as staff, budget, information and support. The underresourced Commission has no alternative to opening its doors and sourcing-in numerous excellent and inexpensive experts, which gives interest groups the opportunity to deliver specific information and sup126

Chapter 3

Pushing the Buttons of ‘Brussels’

port. In 1995, DG Social Affairs, unit V, had to call off meetings of its advisory committee (comitology) on Safety, Health and Hygiene, due to lack of budget at the year’s end; its members from organized management, fearing more labour laws, welcomed this and northern unions regretted it [Daemen and Van Schendelen, 1998, 135]. Imbalances between workload and resources may exist in the EP, Council and Court. The EP’s Friday session in Strasbourg, having a low turnout of MEPs and therefore abolished in 2000, was a case of underload that eased the adoption of resolutions by simple majority, to the pleasure of active lobby groups. Overload by volume or content distributes threats and opportunities. The Commission’s proposal on Reach (COM 2003/644) regarding ‘dangerous chemicals’ got in the 2005 EP the overload of about 1,700 proposals for amendment, which helped to annul their large majority, to make the proposal friendlier to industry and to put the responsible DG ENVI in ward of DG ENTR. In 2012 many more Commission officials suffered physical or mental stress due to the overload of work coming from the many urgent and complex crises and accepted help from stakeholders and consultancies. A problem for one is an opportunity for another. Three operational factors deserve special attention. One is friendship. In the multicultural EU, the daily operations are often quite dependent on informal sympathy [Suvarierol, 2009]. Through friendship one can promote a desired outcome. A DG civil servant may then be prepared to discuss an issue with a friend in another DG, an EP committee or a Council working group; and an MEP may mediate with party friends in the Commission, the COR, the Council or even at home. Secondly, there is the personal ambition of an actor. They want to score and to further their career. By paying a person and his principal a compliment, a favour may be received in return. Thirdly there is language, which, in the multi-lingual EU, is a sensitive issue. Any binding decision is prepared in a working language (usually English) and published in every official language, but no two languages have identical semantics, as a few examples show. In the euro zone there is no common word for the English ‘vigilance’ that the ECB can exert in special situations. The word ‘control’ has in English near-synonyms like scrutiny, surveillance, supervision and oversight, which many other languages lack. Since 2002, the semantics of ‘packaging’ have been under continuous redefinition. In 2008 the Court had to solve the question whether leaked oil falls under ‘waste’ or not (C 188/07). A clever interest group makes use of language differences and always checks the various translations (on the EUR-Lex website) according to the work language. In 2006, a Dutch trade group in fruits and vegetables found that in the Council directive on Harmful Plants and Plant 127

Pushing the Buttons of ‘Brussels’

Chapter 3

Products (EC 2002/89), composed in English, the ‘operational overheads’ of the phytosanitary inspection to be paid by importers (Article 13d) had been translated into Dutch (and German) as ‘fixed costs’, which is a much higher bill. The DGs SJ (legal) and DGT (translation) are the places to lobby for a better translation. 4 Decisional factors. On every issue some stakeholders and officials are in favour of a common decision and others are against. Thanks to these bureau-political fights, there is always room for lobbying. As long as no side is dominant, every outcome is a matter of giving and taking or decision-making by compromise. For the negotiations at least two factors are essential. Firstly, one’s demand regarding an issue at stake must come on the agendas inside the EU’s dealing rooms, as otherwise it cannot be taken into account. Hence, ambitious groups lobby for a position in, for example, an expert group that gives some control over the agenda. Secondly, one has to offer something of interest to important officials and stakeholders, as otherwise one will be neglected. By issuing blue, green and white papers, the Commission has developed sophisticated ‘market research tools’ for discovering the demand side of stakeholders and making its supply side more attractive. Interest groups also have to know the nightmares of other stakeholders, as this enables them to offer cheap remedies or to threaten more. In early 2000, the NGO Oxfam found that pharmaceutical MNCs refused to provide cheap medicines against HIV and Aids to poor countries out of fear that the low-priced products would be re-imported to the rich countries. The remedy became a regulation with a re-import prohibition (EC 953/ 2003). The effective use of factors clearly requires much study and prudence.

3 Vectors to Create The number of useful factors may be countless, but the number of potential vectors is as unlimited as human creativity. While in the factor approach the interest group makes use of existing factors that can influence officials and stakeholders, it creates new ones in the vector approach. Vectors are newly constructed carriers, comparable with the use of chemicals or GMO technology by a farmer. The conditions under which stakeholders and officials operate are constructed so that they increase the likelihood of achieving a desired result. Of all three approaches, the vector method is by far the most sophisticated and becoming a trend that is followed by most intelligent players. Many factors were once created as vectors. Major cases of cultural, formal, operational and decisional vectors are the following. 128

Chapter 3

Pushing the Buttons of ‘Brussels’

1 Cultural vectors. A newly developed concept or indicator can result in a trophy. In 1994, the wealthy Dutch province of Flevoland managed to receive a subsidy of more than €150 million from the regional fund ERDF by excluding from its poverty indicators its (wealthy) commuters. In 2008 the Commission, wanting to receive less annual criticism of the Court of Auditors about its expenditures, proposed a new concept of ‘tolerable risk of error’ (COM 2008/ 866). Fearing an open beer market in the 1990s, the German beer industry, referring to its Reinheitsgebot of 1516, first promoted the value of public health and, once this failed, that of environment by demanding that every empty bottle or tin should be repatriated to the country of origin, so increasing transport costs for suppliers from abroad. In the early 2000s, the European textile producers, organized as ELTAC (now EuraText), promoted the value of child rights in order to undermine cheap textile imports from the Third World that relies on child labour. Also pressured by interest groups, the Commission often uses an OCM for europeanizing a policy domain that is weakly anchored in Treaty text, such as spatial planning or taxation. This way it gradually catches opponent governments and stakeholders in its EU net. In the 1990s, the UK government promoted enlargement (‘widening’) of the EU as means to hamper the trend towards a more supranational Union (its ‘deepening’). As Council President in 2008, the French president Nicolas Sarkozy pleaded for a more transparent European Central Bank (ECB), so hoping to increase political control over it, an old French desire. At the 2011 start of the fight over the 2014-2020 budget for agriculture, France proposed a fresh budget for ‘transition regions’ that need reconversion, thereby hoping to serve its farmers. The treaty forbids the protection of a national interest but not that of ‘general interest’, which may be allowed. Reframing a concept or value [Daviter, 2011] can catapult it towards a desired outcome and is salesman’s talk (which will be discussed further in chapter 6). 2 Formal vectors. Ranking top is the creative (re)interpretation and reframing of formal procedures and powers, as laid down in treaty texts that usually refer to only general areas like agriculture or health and seldom to specific issues, such as a new type of margarine that lowers cholesterol. Every fight over an EU decision is, however, not general but specific and may be won by moving the specific issue to a different general label with different procedures, powers and people. The food producers Unilever, seeking market admission for a new margarine in 1999, got it from DG SANCO in 2004 under the health frame. Regional interest groups make creative use of the tension between EC Treaty article 10, keeping national governments responsible to the EU 129

Pushing the Buttons of ‘Brussels’

Chapter 3

and the Court’s ‘Constanzo’ ruling 103/88 that attributes responsibility to decentralized government. In the early 2000s, the German government, being against a tobacco ban, challenged the treaty base of the Commission’s draft Tobacco Ban Directive, which position the Court confirmed under addition how to make the ban legal, thus leaving Germany with a battle won but a war lost (EC 2001/37). Until the new 2011 Comitology Regulation, any Commission proposal to a regulatory (now: examination) committee that refused its support, had to be forwarded to the Council and, if it has had a codecision position before, to the EP, which institution(s) then could overrule the Commission by (qualified) majority within a short time; otherwise the Commission would be the winner. An ingenious group thus could win its case by getting the Commission’s proposal on its line and, if the committee is not positive, by creating strong divisions inside the Council and, if necessary, the EP, as French groups managed to do in relation to the free trade of transgenic maize in 1996 [Bradley, 1998]. 3 Operational vectors. Many an interest group seeks to meet the need of EU officials for special expertise, reliable information and solid support, and dreams of becoming the almost sole supplier. Green groups are smart on overloading, by volume and content, the Commission and making their experts available for exchanging information and support. Friendship, ambition and language are helpful vectors too. Friendship in political settings is a diplomatic show of sympathy and charm that most receivers appreciate, whether they are partners or opponents. Background information about the hobbies and lifestyles of high officials can be very helpful. The personal ambitions of an official one can serve by, for example, helping with career opportunities or good publicity. In the EU Babylon, language is a crucial vector. In 1991 Dutch Frisia Dairy was upset by an EU decree prescribing in the Dutch text ‘metal’ instead of ‘carton’ packaging for milk products, thus obliging it to change the machines in their factories. Finally the Commission qualified this to be a mistranslation of the word ‘boîte’ in the French working text, which was meant to refer to any sort of packaging. According to the Frisia Dairy, however, it was a trick played by the French, their main competitors on the dairy market. Interest competition can heat up a fight over words. At the Council Summit of December 2007, the UK claimed that trade agreements be based on ‘reciprocal’ instead of ‘mutual’ benefit, so hiding the real fight that the former word forbids protection of domestic industry and employment and the latter does not forbid and thus allows it. 4 Decisional vectors. A great vector is the creation of a strong and preferably cross-sectoral coalition. In order to get a subsidy, the steel regions 130

Chapter 3

Pushing the Buttons of ‘Brussels’

managed in 1995 to bring together local authorities, trade unions and steel companies under the umbrella of Caster (now Eira). A variant is the set-up of a GONGO (or BONGO), such as the 2004 EuroFed of Purple (PeriUrban Regions Platform Europe) of regional authorities thirsting after subsidies for their areas. Setting up a real or fake NGO is often useful, as it gives a finer licence to lobby and, for example, to get from the Commission subsidies and projects that cover some fixed costs, which enable to outbid commercial competitors on tenders by charging mainly variable costs (the fixed ones being covered already) and to influence the Commission by its report. Alert interest groups like to inflate rapidly rising issues to credible drama, such as terrorism in 2001 and food shortage in 2007, as this may bring new budget, legislation and organization. In 2003 DG TREN (then: transport and energy) was opposed in the Energy Council for its proposal on Gas Market Security (EC 2003/55) and as a compromise it got an advisory committee under comitology that allows measures in emergency times. The reverse, deflating a drama, can work too. In 2004 the chemical industry, organized in the ad hoc Alliance for a Competitive European Industry and fearing the Reach-proposals from DG ENVI, lobbied successfully for an impact assessment that would delay the dossier. To find policy room, the Commission often uses its mass-survey Eurobarometer, as in 2007 on the use of cannabis that revealed majority support against it in the EU-25 [COM, Eurobarometer, 2007, 66] and, under the Lisbon Treaty, became part of its agenda (COM 2010/ 171). In order to achieve a better compromise, an interest group can majorate its demand, pay concessions, push or block an exemption, create a package deal or split a dossier, in short manipulate parts of its demand and supply side. Examples of dossier splitting are the Treaties of Amsterdam, Nice and Lisbon, each containing leftovers from their predecessor. In order to bring an influence ambition to success, the political engineering of actors to approach, factors to use or vectors to create requires ever more proactive study and prudent application. This starts best with proactive thinking regarding actors, issues, time and many other conditions, as briefly indicated at the end of chapter 1 (figure 1.2 from activity 5 to 1) and to be detailed in the next chapters. This PAM work is not something that can be done by amateurish groups that score high on nonchalance and thus risk, but only by professional ones that are familiar with the possible actions and prudent regarding the many possible consequences of any action. This particularly applies to the following meta-game that integrates actors, factors and vectors.

131

Pushing the Buttons of ‘Brussels’

Chapter 3

III The Meta-Game of Triple P Well-known is the pious wish for free and open competition on the market, where every producer or merchant dreams of becoming the monopolist. As long as all merchants acting on freely accessible markets have the same ambition, they all end up as competitors indeed. However, thanks to their smarter buying and selling, some may be successful in gaining a stronger position in their market, by which this becomes less open. On the EU political market, the behaviour of players is not much different. Many interest groups piously claim to love a level playing-field, but silently they almost all dream of a most unlevel playing-field, with themselves uphill in a sort of citadel and their opponents in a weak position downhill. The really level playing-field is only their next-best preference, much better than a downhill position. The game of Triple P intends to make a playing-field more lopsided by getting, like pickets in the field, the friendliest persons in the best positions in the most beneficial procedures, so gaining a position in an uphill citadel filled with powerful sticks that save on delivering carrots. As long as the EU process law is still in development there is much room for this special game. The game must be played far before the match on substance (the ball) starts and thus is only open to early informed and proactive groups. Making the playing-field more comfortable to their interests in advance, they can easier score with the ball once the real match begins. It is like playing billiards by bricole and comes close to the old technique of encapsulation. Every treaty formation is in essence a Triple P game that places early pickets on policy fields and anticipates the future fighting on substance. Commission officials and French interest groups often win fame in this game. When others start to argue over the substance of an issue, they have already made the playing-field comfortable to their interests and limited the other players’ radius of action. The meta-game of Triple P requires a systematic and coherent selection of vectors to create, factors to use and actors to approach, to be connected in a series and/or parallel setting. Its rich menu is regularly enriched with new discoveries (figure 3.1). The game can be played in every arena, being the EU (an institution or part of it), among stakeholders (inside a EuroFed) or domestically (at the home front). Here we shall focus on the EU. The First P (procedures) comes close to our category of formal and decisional factors and vectors, as far as they prearrange or rearrange a field. By affecting a procedure so that it is benign to one’s interest, one automatically gets a distribution of formal powers that are fairly likely to protect it. The benign procedure can vary from secondary with the EP having code132

Chapter 3

Pushing the Buttons of ‘Brussels’

The Meta-game of Triple P P 1: Procedures and Principles, such as - Consultation or Codecision of EP - Unanimity or (Q)MV in Council - Comitology, Subsidiarity, OCM - Definitions, Review Clauses, Deadlines P 2: Positions, such as Inside - Committees, Work Groups, Inter-groups - Com DG, EP Committee, Council SG - Council Chair, Think Tank, EuroFed - Inspection Agency, Policy Consultancy P 3: People, such as - Befriended Commissioner, Cabinet People - Friends and Staff In EP, ESC, COR - Court Members and Referendaries - COM Officials (Statutory, Secondment)

Figure 3.1

cision powers or not and the Council voting by unanimity or QMV, to delegated procedure or comitology, an OCM or whatever else. The interest has to be reframed in advance in order to make feasible the benign procedure. For example, national market protection, in itself forbidden, might be preserved under the flags of health, security, environment or general interest and be followed by different procedures. One can also target the form of a decision as sketched in the menu of possible choices (chapter 2II-2), such as between hard or soft law and by regulation or directive. Even a seemingly small detail may result in another procedure and outcome. In 1990 French groups in the meat trade managed to get in the tender conditions for food supply to Russia a definition of can sizes for potted meat that are used in France but not known to their Dutch rivals, and they won the tender. The Second P (positions) is a mix of formal and operational factors and vectors. The interest group tries to get control over crucial positions in a procedure and to prevent opponents from doing the same. The best position to target is frequently not the top of the formal hierarchy but one that confers daily control over the issue, for example that of chef de dossier, expert, rapporteur or staff member. French groups often have an appetite for positions of chairman, and thus are formalistic rather than pragmatic. Their northern competitors often feel happier with low-key positions as 133

Pushing the Buttons of ‘Brussels’

Chapter 3

apparatchiks who control the work. A position close to a consultancy firm contracted by the Commission can also provide control over the advice given. The reframing of an issue can be part of the Second P game. In the 2000s, green interest groups got many issues related to agriculture, transport and trade reframed as issues of sustainability and so obtained positions close to their friends in DG ENVI that prepares the proposals. The Third P (persons) is a mixture of mainly cultural and operational factors and vectors regarding actors found to be crucial. A friend in a relevant position in a beneficial procedure can give the finishing touch to the ball in the real match. The friend may be brought in from outside, for example as expert or civil servant on secondment, but it is more efficient and often more effective to use somebody on the inside. An old friend who already shares your values and interests is highly valuable. New friends can be won by stressing common values and creating common interests, strengthened by personal characteristics such as policy preference, background or career. A good friend in a relevant position can distribute resources such as staff, budget and information in the desired direction, for example by promoting that a friend becomes chef de dossier in the Commission or rapporteur in the EP. When the real match starts, other stakeholders must come along and offer a deal. The Triple P meta-game usually has a bad reputation among the players who are in a downhill position. They consider the covert framing of an arena before the match starts as unfairly aggressive, Kafka-esque and sneaky. They grumble over devilish manoeuvrings regarding such abstractions as procedures, positions and people in charge and say they prefer to play the real game on substance directly and openly. Most interest groups making such complaints are sore and bad losers. Once they understand this meta-gaming, they often start to apply it and to appreciate its sophisticated efficiency and effectiveness. The ‘aggression’ is then reassessed as a professional skill, the ‘Kafka-esque’ quality as an interesting component and the ‘sneakiness’ as just a smart way to gain an advantage. In fact, there is nothing new about the Triple P game, as it is classical behaviour that can be found in, among other places, the politics of Rome, the Catholic Church, the Ottoman Empire and the Florentine republic [Loewenstein, 1973; Reese, 1996; Wright, 2004; Stout, 2005; Machiavelli, 1513]. At the EU, the game is applied in only new forms. One paradox, encouraging to losing players, is that the higher the number of groups playing this meta-game is, the lower the chance that a single group can prearrange the playing-field solely and fully for its own benefit. Another paradox is that most groups should feel familiar with the Triple P game, which is quite common in their usually less open domestic systems. Anyhow, an unlevel playing-field is as much caused by clever groups that got it as by nonchalant groups that 134

Chapter 3

Pushing the Buttons of ‘Brussels’

did not prevent it. The latter pay by the higher risk of losing the substantial match that follows and should blame themselves for this outcome, not the players being cleverer than them. Many examples of the Triple P game can be given. One is the 2002 European Food Safety Regulation (EC 178/2002), adopted on the eve of the fifth enlargement and in Central Europe widely seen as trick to keep their ‘less safe’ food products from the Western markets. However, the regulation, which is about 24 double-column pages long, does not contain any substantial decision on food safety and is, as a real Triple P text, only about definitions, objectives, principles, procedures, delegation of powers, bodies in charge and people to be recruited, including the set-up of the EFSA. After its adoption by the Council and the EP, the substantial game started under the Commission’s delegated powers, supported by the EFSA’s Management Board that included persons from the French retailer Carrefour and the Anglo-Dutch producer Unilever, two of the interest groups having pushed this regulation. Another example is the 2001/37/EC Tobacco Ban Directive, proposed by DG SANCO in closed shop with anti-tobacco groups, connected in series to both the EP’s Health Committee and the Health Council, and having as its main contents the empowerment of a regulatory committee (comitology) consisting of almost entirely ‘health freaks’. A third example is the Implementation of the Aarhus Convention (COM 2003/622), proposed by DG ENVI as an outcome of ‘green’ lobbying on the 1998 United Nations ‘Aarhus Convention’ [Bugdahn, 2008] and giving only ‘entitled’ (green) groups privileged access to information and participation, thus an uphill position. Opponents, being sharp this time, secured a milder version of the text from the EP and the Council, but the fight went on, reaching the Court in 2012. A fourth example is Giscard d’Estaing’s presidency of ‘his’ Convention that resulted in the 2005 predraft Constitution for Europe. As the composition of the Convention (third P) had been decided by others, he could only play with the first and second P and he did this ingeniously [Tsebelis and Proksch, 2007]. A final example is the 2008 EU agenda-setting regarding ‘the protection of unaccompanied minors in air traffic’, engineered in closed triangle by DG JFS (now DG Just), police organizations and the Belgian NGO Child Focus, the latter presenting on this theme sponsored research findings and precooked proposals to a televized meeting with invited stakeholders, so legitimizing official EU support to most of their proposals. To all those having to fight from downhill position an uphill battle for some time, we have two consolations. The one is that, by standard EU practise, a law or act often undergoes review after five to seven years, which offers a chance to recover and underlines that usually both wins and losses are temporary. The other is that no downhill position is necessarily 135

Pushing the Buttons of ‘Brussels’

Chapter 3

hopeless as one can start an uphill fight by Triple P. Due to the difficulty in conquering a citadel surrounded by open fields by direct attack, one first has to organize the downhill area by positioning friendly troops at strategic places around the stronghold. Then one can try to get control over the citadel’s supply lines and launch new mobilizing values, such as consumers’ freedom and mental health in the tobacco case. Attacking indirectly, one might arrange the playing-field more to one’s advantage and isolate the citadel. The battle then is a fight between two different games of Triple P, both inspired by PAM.

IV Structural Trends of Managing EU Public Affairs On the EU playing-field the public and private interest groups have in common the ambition to win or at least not to lose a game relevant to their interests. They try to achieve this by applying different methods regarding actors, factors or vectors. One structural trend is to combine them in parallel or in series, such as by Triple P. There are more such trends, to be presented here. The major cause is the increased intensity and amount of competition. To understand its logic we refer to Akens’ metaphor in the previous chapter: the more the many groups smell relevant ‘flowers and trees’ in the EU policy gardens (supply side), the more they will act as ‘birds and bees’ trying to gather nectar (demand side) that is scarcer than in nature (competition) [Hauser, 2011]. The inflow of non-EU stakeholders and the recent enlargements make the competition even stronger. Stimulated by their strong desires, sufficient resources, irresistible compulsions and seductive invitations [Milbrath and Goel, 1977], ever more interest groups participate in the EU. Since some time they overcrowd the playing-field, particularly at the Commission and the EP. Many groups feel that their lobbying is losing efficiency and effectiveness (the 2E of PAM). Some groups search for and develop PAM innovations that, as usual in sports life, are followed by some others and can be considered structural trends. In this section we describe the three main innovations: self-reliance, collective action and tailor-made PAM.

1 From ‘National Co-ordination’ to Self-reliance In the past the interest groups wanting to influence EU decision-making preferred to take, as a matter of course, a national route to Brussels. At the micro-level, they contacted their national association (meso-level) that stood at the crossing of routes to both their EuroFed and central government (macro-level). The latter, receiving all sorts of domestic interests, co136

Chapter 3

Pushing the Buttons of ‘Brussels’

ordinated them through a procedure with a central office and defended the selected interests at its Council level. On the national playing-field, every ministry thus had a triple position: as public interest group of its own (micro), as a sort of public association for interest groups in its policy domain (meso) and as participant in both its central government procedure and at Council level (macro). For a long time its third role made sense, as most binding EU decisions came by the secondary procedure with, until the 1987 SEA, a veto position for the national minister in the Council. Except some MNOs and agricultural associations, most groups lacked experience and infrastructure for self-reliant cross-border influence activities and looked for help from their government. This pattern is disappearing in the fifteen old member states, but still exists in the new member states. True enough, all central governments still have, at the highest level (Cabinet, Prime Minister or President), a procedure and office for national co-ordination [Kassim and others, 2000 and 2001; Wessels and others, 2003], which has become a ‘quagmire’ [Peters and Wright, 2001] or a ‘Byzantine mechanism’ [Schneider and Baltz, 2005] that involves many players, interests, conflicts and time at home. Since the early 1990s the procedure loses much effectiveness [Van Schendelen, 1993]. It is now mainly tried in four situations: (a) if for upcoming Council meetings the national voting position has to be formalized, (b) if the national parliament is pressing for a common position on some issue, (c) if there are heavy inter-ministerial conflicts over an issue and (d) if the government has veto power in the Council. The ‘co-ordination’ often results in coherency of views on paper rather than co-ordinated action, thereby remaining symbolic. The coordination is not really ‘national’, as it mainly involves the central government and hardly the rest of society. The reality is usually as follows. The first responsible ministry frames the voting position for the Council long before its meeting. The parliament receives symbols of coherency like solemn statements and promises rather than proofs of co-ordination. Conflicting ministries want their interest not to become co-ordinated centrally. In case of a Council veto position, the co-ordination is effective in blocking an undesirable proposal, but not in pushing a desired one. This blocking would also be possible without co-ordination and anyhow costs a price to be paid on another dossier. The growing new practise can be summarized by the catchword selfreliance [Van Schendelen, 1993]. Public or private pressure groups may like to co-ordinate others actively, but they abhor being co-ordinated by others passively. They increasingly prefer to act on their own [Knodt and others, 2011]. This is the case with all sorts of interest groups, even those that falsely preach co-ordination. Many trade associations wishing to co137

Pushing the Buttons of ‘Brussels’

Chapter 3

ordinate their members prefer to bypass their national umbrella, and units of ministries often find their way to Brussels long before informing their central office. Ministries of Foreign Affairs, which in most countries are charged with the dual task of both central government EU co-ordination and their foreign policy-making, frequently act in the EU self-reliantly regarding their second concern. In the early 1990s, even in France (formally the most centralized state), more than 100 interest groups, including units of ministries, escaped their central co-ordinating office SGCI (under the Foreign Ministry, since 2005 as SGAE under the Prime Minister), by setting up their own lobby office in Brussels [Legendre, 1993], discretely called ‘information bureaus’. Everywhere, ‘national co-ordination’ is now almost only urged by the national parliament wanting to bind the government formally, by the leading ministry wanting to control other ministries and by mass media sensing a conflict. The trend of self-reliance is a rational response to the need to win and not lose on one’s interests. The EU offers a strong pulling factor as an interest group that applies PA directly in Brussels might get its interest protected by a binding EU decision that overrules the domestic arena. This logic is found among all sorts of groups [Fontana, 2011; Tatham, 2010]. The domestic arena offers, besides, strong pushing factors that justify that rational approach, such as the following. 1 Domestic divisions on every EU issue are the natural state of affairs in a member country, nicely phrased as ‘pluralistic’ and nicer as ‘democratic’. The country always contains many more different interests than those of the state alone and for the latter more than those of only the central government, which is also internally divided. In the absence of a ‘national’ interest, defined as an interest ‘shared by all at home’, no domestic interest group can trust that its central government shall back its interests. By denying the pluralism at home, a central government willy-nilly stimulates many domestic stakeholders to look after their EU interests self-reliantly. By excluding them from participating inside, the central government even legitimizes their self-reliance. 2 The benefits of ‘national co-ordination’ can only exist if the Council plays a crucial role in the EU process and if the central government can make the difference there. As shown in chapter 2, the Council now usually plays a formally decisive role in only the last phase of about 15% of the EU’s binding decisions. Under QMV a single member state rarely makes a real difference anyhow. On the drafting of the EU legislation and on delegated and implementing acts the position of a central government is weak. 3 The costs of ‘national co-ordination’ can be enormous. An interest group relying on it runs all the risks of, for example, losing its case in 138

Chapter 3

Pushing the Buttons of ‘Brussels’

the national capital, mobilizing opponents early, diluting its interests with many different ones, being too late in the Commission phase or missing a rising European coalition. 4 The alternative benefits of self-reliance usually outbalance both its costs and the cost-benefit ratio of ‘national co-ordination’. At EU level every interest group can find befriended stakeholders, feel compelled to follow competitors already active in the EU, get access at the Commission and the EP and, in short, promote its interests. Self-reliant Route Mapping and Navigation This quadruple logic of self-reliance, based on 2E considerations of efficiency and effectiveness, applies to all domestic levels, from the central government and associations to single groups. In the open and free European countries every interest group can choose from a menu of routes to the EU, take a suitable one and bypass its overarching structures. Rarely can the parent or sister organization prohibit this effectively. For example, the central government has few means to keep domestic groups at home, because private groups are free to move around and decentralized governments (regions, cities, agencies) usually possess some legal autonomy. Figure 3.2 gives, from the perspective of an interest group at micro-level, a helicopter view of the three main routes from home to the EU, with arrows indicating the upstream direction. 1 The national route runs via the meso-level of domestic associations to the central government, from where the interests are transported into the EU. This route is still used for the few dossiers falling under Council unanimity and by poorly informed or desperate groups that ask part of their government or even parliament for some help. 2 The transnational route to the EU runs via the domestic association or directly to a European federation (EuroFed) or one of its variants (confederation, network, ad hoc group) and from there into the EU. Federations have a rather homogeneous interest profile and confederations a heterogeneous one. Ad hoc groups and networks are more informal, the former for a short-term cause and the latter for a long-term one. A Brussels office is for watching on the spot. Except national ministries, domestic groups intensely use the transnational route. 3 The international route runs via organizations in different EU countries (like France, the UK and Poland), foreign ones (like US, Japan, Brazil, Russia, India or China, the last four being ‘BRIC’) or global ones (like WTO, WHO, OECD). This route offers early back doors into the EU [Wouters and others, 2012; Costa and Jørgensen, 2012]. Multinational organized organizations (MNOs) can take the international route direct139

Chapter 3

Pushing the Buttons of ‘Brussels’

Interest Groups and Routes to EU National

EU/Transnational

Foreign, e.g.

Cabinet Ministries Parliament Agencies Court

Council Commission Parliament ESC, COR COURT

FR, UK … USA, Japan ... ‘BRIC’ ... OECD, IMF ... WTO, WHO …

Pri/Pu/Ngo:

Pri/Pu/Ngo:

Sectoral/ Regional Organizations

Confederations Federations Networks Ad-hoc issue groups Brussels office

Pri/Pu/Ngo: Small/Medium-sized Big/Multinational Pri = private / Pu = public / Ngo = non-governmental

Figure 3.2

ly through their subsidiaries, various national associations and/or their EuroFeds (or variants). Having resources and acting global, they are active users of this route for building up supported positions via other countries and international platforms, such as WTO on trade issues and WHO on food and health issues [Klüver, 2010]. The road-map in figure 3.2 offers many lanes, junctions and roundabouts and, if connected in a series, numerous options for intelligent road-mapping or navigation. Most interest groups that leave the old national route take the transnational one first, usually by becoming via their domestic association an indirect member of a platform there. Here they get within reach the international route. The menu allows every group to optimize its road-mapping. In daily life, the interest groups’ navigations show the following trends. 1 Multiple routes. All main routes are used. The national route too but seldom exclusively and usually selectively for shopping for information and support in the national capital, supplementary to the transnation140

Chapter 3

2

3

4

5

Pushing the Buttons of ‘Brussels’

al route. Many multinational groups use all main routes, in series and in parallel, thus also the many junctions and roundabouts. They practise multi-level PAM: in their home country, in different countries, at the global level and the EU. Deviant governments. Central governments still tend to go directly from their national capital to Brussels and particularly their Council building. Single ministries mainly rely on their officials in the PR and in expert groups, working groups and comitology, and they hardly use other domestic stakeholders, such as other ministries, decentralized governments and private groups, even when these hold strong positions around and inside the Commission and the EP. Seldom do they have a sort of EuroFed for the transnational aggregation of their interests and sometimes rely for this on informal bilateral meetings and largely on the various Council meetings. Only a few take the international route to the EU. Reversed arrows. Running up is the reversal of some arrows in figure 3.2. The one at the upper left, between the central government and the EU, is, in reverse, the route to squeeze the government at home by trying to get from the EU benign laws and acts that overrule unpleasant domestic ones. The arrow between domestic associations and the world of EuroFeds increasingly shows an also reversed use, as the former tend to become subsidiary to the latter. Particularly MNOs, if direct member of such a EuroFed (see next section), cause this reversal. Road mapping. The general factors of participation in the EU (chapter 1-IV) differentiate for road-mapping as well. The desires, capacities, compulsions and invitations can all lead to a specific and personalized choice of navigation. This partially explains why the best established and often multinationally organized groups, usually scoring high on all factors, tend to use so many routes in parallel and in series, and why so many groups from new member states (still) largely (have to) rely on their national route. Intelligence. As part of any effort to influence the EU, road-mapping should not be done at random, but as the outcome of much windowout intelligence (I2) or preparatory homework. Only then can one identify the best routes having the least obstacles, navigate efficiently and effectively and arrive safely at the desired destination on time. Clever navigation is like intelligent shopping and should be prepared at home. Most groups, however, are still nonchalant on road-mapping and simply take the route they see or are used to.

141

Pushing the Buttons of ‘Brussels’

Chapter 3

2 From Individual to Collective EU Action Every interest group would love to be the single player on the EU playingfield, but this is a daydream. Even in the old Community of six member states, interest groups usually had to compete with others and the more so in the current EU with its high amount of diverse interest groups, unparalleled at any domestic level. This strong competition compels every interest group to search for ways to make its influence efforts more effective and efficient (2E). The creation of critical mass (the M of our formula) by co-operating with other stakeholders (shared-listing) is one necessary way (the other being I2). A group may fly alone from home but, arriving near the EU playing-field, it has to be part of a flock of birds. Competition and collective action have thus become two sides of one coin. A single group may still contact people around or inside the institutions on its own, but such action is done for supplementary window-out monitoring and window-in negotiations on interests that differ from the common ones. The latter are pushed by collective action through a platform with common ears, eyes, face and tongue. An intelligent group anticipates this by earlier shared-listing, so getting higher returns for longer. Collective action is part of the logic of influence [Schmitter and Streeck, 1999] that is related to the size of a political system. From the local to the regional, national and European level every interest group needs ever more critical mass (M), as a multiplier of its potential influence. Also those at the receiver’s side of the EU want broader and anyhow cross-national mass from it. (The full-page political advertisement of three German interest groups on free sugar trade (European Voice, October 4, 2012) is an anachronism.) Commission officials want to interact with interest groups at low costs of overload and thus prefer aggregated rather than individual interests, and they may even subsidize group formation [Mahoney and Beckstrand, 2011; Greenwood, 2011, 16; Broscheid and Coen, 2007]. The optimal composition of a platform is a puzzle to every interest group and needs a subtle balance between homogeneity of interests and control of the process. With similar stakeholders (homogeneity) it is easy to find common issues, but difficult to arrive at common positions as market or power positions of the members are at stake too. Distrust always lays in wait and needs control. ’Single issue’ NGOs often feel this difficulty, as is exemplified by the fights between ‘green’ and ‘third world’ NGOs on the use of biofuels (the former in favour for the sake of environment, the latter against in order to avoid food shortages) or the NGOs for birds versus fish (the former eating the latter, never the reverse). With different stakeholders (heterogeneity) there is little community of interests, by which the formation of a common agenda often results in endless waffling or long-listing with142

Chapter 3

Pushing the Buttons of ‘Brussels’

out good chances of success. Most groups prefer homogeneity on usually a sectoral basis, with their direct competitors around the table, and if helpful with cross-sectoral extensions. The model for collective action is the rather homogeneous EuroFed [Greenwood, 2011; Beyers and others, 2008], although many EuroFeds have got more diversity by the enlargements from EU-15 to EU-27. As shown in the middle of figure 3.2, groups with interests in common meet together in a transnational umbrella close to the EU. In addition to the variants mentioned already, there are special umbrellas, such as the ‘European House’ where, as in the Renewable Energy House, stakeholders share logistical facilities and the ‘National House’ or the ‘Regional House’ like Austrian House, Czech House and Dutch Provinces House that provide common offices to their groups. A cross-border variant is the Palmerston Group of Danish, German and other regions around the Baltic Sea, residing in Brussels on Palmerston Avenue. Most EuroFeds (and variants) also have a basically similar organizational form. There is a General Assembly having as members the national associations and sometimes also major individual organizations. Groups from outside but active inside the EU, such as US companies and Turkish associations, can usually be members. The Assembly appoints the members of the Executive Board, being supported by an Executive Committee and usually a small Secretariat or Bureau (mostly about six people). The Committee and Bureau tend to be the main players, prepare the agendas, manage the internal working groups and push for common positions [EurActiv, 2009-B]. Many umbrellas suffer, however, from a fragile vitality, caused by insufficient homogeneity, experience, leadership, representativeness, focus, following, prestige, stability, cross-nationality or resources [EurActiv, 2009-B; Greenwood, 1995, 4447]. The internal management of a EuroFed requires the highest PAM skills. As evidenced by case-studies [Pedler, 2002; Greenwood and Aspinwall, 1998; Wallace and Young, 1997; Greenwood, 1995; Pedler and Van Schendelen, 1994] and surveys [EurActiv, 2009-A-B-D], the collective platform can create multiplied added value of both I2 and 2E to its members, as it: – provides a cross-national setting for meeting, watching and listening to each other, in order to find out ‘what is in it for me?’ – acts as radar that monitors EU initiatives, dossiers, processes, events, people and anything else of potential relevance – promotes a common agenda by aggregating the different interests of the members coming from different countries up to a long-list – makes common decisions regarding the agenda, particularly the downsizing to and the target-setting for both the shared-list and the short-list – creates a more European face, without which the lobby at the Commis143

Pushing the Buttons of ‘Brussels’

– – – – –

Chapter 3

sion is only a case of national attire or folklore and thus almost doomed to failure gives organizational structure to its economic sector or policy domain and thus a title of representation at other stakeholders and EU officials divides the work of PA by its use of a common PA desk (and usually technical experts provided by its members), thus enhancing the 2E of PA serves as shopping centre for finding allies, precooking decisions and, if necessary, for setting up an ad hoc and/or cross-sectoral new platform in sum, multiplies the potential influence of the individual members as collective action outperforms individual action, and allows other business to be done, as competitors inside are also useful for making a market deal or policy initiative unrelated to the EU but interesting at home.

Functional Trends among EuroFeds Of course, there is much functional variation among EuroFeds and their variants [Eising, 2009]. For example, the more heterogeneous a crossnational platform is, the less it functions as decision-maker for influence actions and the more as a shopping centre for an ad hoc coalition. Like in air transport, most groups use such a main port or hub as means to get a more efficient and effective connection to their final destiny. Collective platforms have some disadvantages too, such as dilution of one’s specific interest, internal dissent followed by stalemate, dependency on the platform for the quality of lobbying and, not least, distrust that at the end the gain is for a competitor and the pain for oneself. Thus, only naïve members of a common platform fully rely on only one platform. Intelligent groups safeguard their interests additionally, for example by being member of various umbrellas, using multiple routes, taking a strong position in the Committee or Bureau and/or installing a Brussels office for control on the spot. With these safeguards they consider the positive functions or advantages of EuroFeds as outbalancing the negative ones or disadvantages. Anyhow, they see no better alternative. Commission officials require an interest to be cross-nationally and preferably cross-sectorally aggregated. Single MEPs and Council people may be willing to support a domestic interest, but they can only push this through if they get their colleagues from other countries to support the same interest, thus by making it crossnational as yet. Among Commission officials and MEPs some criticism of collective platforms is rising however, particularly as a result of unreliable promises of information or support, which is usually caused by internal instability [Warntjen and Wonka, 2004, 48]. The EuroFeds and their variants are today, as stated (chapter 1-XII), 144

Chapter 3

Pushing the Buttons of ‘Brussels’

a voluminous phenomenon of more than one thousand Brussels-based entities [Transparency Register; EPAD; Stakeholder.EU]. Companies and associations in pharmaceuticals meet in EFPIA, chemical ones in CEFIC, small and medium-sized enterprises (SMEs) in UEAPME and food and drink producers in FDE. NGOs (including GONGOs and BONGOs) have platforms that range from the consumers’ group BEUC, employers’ umbrella BusinessEurope and trade union confederation ETUC to all sorts of social movements, from child care and environment to citizenship and health [Lahusen, 2004]. Regional groups have their EU umbrellas [Rowe, 2011; Greenwood, 2011; Antalovsky and others, 2005] like the steel regions (EIRA) and the big cities (EUROCITIES), as have professional groups like architects (ACE) and journalists (AEJ). National police organizations have their FIEP, agencies on shipping inspection ParisMou and those on car registration EReg. Seldom, however, does a EuroFed represent the whole sector. There are, for example, about 15 EuroFeds on construction, 20 on packaging, 40 on environment and 75 on chemicals. Many of them are not broadly European, but limited to, for example, only southern or Teutonic countries or are area-specific, such as for only one sort of chemical, bird or region. In short, almost every domestic group can find an EU platform – even voluntary groups on breast cancer for which there is EuropaDonna. The big exception are national ministries, which are almost entirely absent in the world of EuroFeds. By coincidence, the ministries of Defence, insofar as member of the NATO located in the north of Brussels, (can) use this platform for EU action. Since the mid-2000s, a few ministerial ad hoc platforms have emerged from informal Council settings, such as the ‘Sparing Nine’ on budget reform, the CEMT on transport and the Baltic Group on their regional interests. Even units of ministries now start to develop a transgovernmental network sometimes [Thurner and Binder, 2009]. Sooner or later they may develop into full-fledged EuroFeds. The world of collective platforms is, not surprisingly, full of change. One major trend from the recent past is the rise of EU BONGOs and GONGOs, such as the EACF (a consumer’s platform set up by air-transport companies) and TISPOL (transport police, sponsored by the Commission). Another one is their supply of technical experts to the Commission in return for seats in groups and committees. In addition, four new trends can be observed. 1 Going cross-sectoral. Many interest groups search for a broader critical mass than one that covers only their sector. Finding their EuroFed too heterogeneous now, they use this increasingly as a shopping centre for allies with more specifically similar interests, where after they invite governmental groups (central and/or decentralized) and/or private ones (business and/or NGO). The Commission welcomes this interest aggregation beyond sectoral borders. An early example is EACEM (now 145

Pushing the Buttons of ‘Brussels’

Chapter 3

EICTA), the EuroFed of the consumer electronics industry, whose members (like Siemens, Thomson and Nokia) pushed ad hoc coalitions with trade unions (on labour conditions), publishers (on intellectual property) and customers (on standardizations). A recent example is the 2012 Coalition for Energy Savings, a platform of about 25 EuroFeds. The costs of making an ad hoc coalition may be high for a short time but offers a high return on investment [Pijnenburg, 1998]. 2 Direct membership. Most EuroFeds have national associations as parents but, according to a sample of 170 EuroFeds, one-third of them now accept (for financial reasons) major individual organizations as direct members [EurActiv, 2009-B], being usually MNOs in the fields of business and NGOs, which buy, so to say, more influence inside the EuroFed. According to a survey, most large companies are member of two EuroFeds and almost half of them of more than five [EurActiv, 2009A]. Examples of direct membership are FDE (food and drink products), ECF (construction industry), Eurocommerce (retail), EFPIA (chemicals) and EPEE (energy-saving products). The old parents usually resist this trend and sometimes, like BusinessEurope and CEA, with success. 3 Subsidiarization. Another effect of direct membership is that via their EuroFed the direct members can instruct the national associations on their domestic agenda, thereby reducing them to a branch office or a subsidiary, which is another example of how the home country can be influenced via the EU and often easier than by PAM at home [Fontana, 2011]. Aside, consultancies show similar logic as the old trend of domestic ones searching for EU business is reversing into EU consultancies patronizing domestic ones. In the EuroFeds case, the small and medium-sized enterprises (SMEs) being dependent on their national association are on the losing side, about which Commission and EP show mixed feelings. 4 Multi-level lobbying. Many MNOs are increasingly embedding their EU lobbying in a multi-level approach that integrates the international level with the domestic and transnational ones [Eising, 2009; Van der Heijden, 2006]. They try to push or block EU decisions along different routes and levels. An example is the World Business Council for Sustainable Development (WBCSD), formed in 1995 by about 200 companies from 20 sectors, in order to influence the UN decisions on Climate Change, like the Kyoto Protocol 2005 and the planned post-2012 Climate Change Agreement. By implication, the earliest phase of EU legislation then starts much earlier and farther away than with a Commission’s proposal, so taking less alert stakeholders by surprise.

146

Chapter 3

Pushing the Buttons of ‘Brussels’

3 From Ready-made to Tailor-made Action Until the 1980s, the single interest group hardly showed much variation in its efforts to influence the EU. Acting relatively ready-made, it took the traditional route to its domestic umbrella and/or ministry and relied on their linkages to the EuroFed and the EU. It hardly differentiated its approach for different issues, stakeholders and dossiers. From the side of the EU, there was equally little variation in responses to interest groups that by number were small as a village. At that time, most legislation was secondary and based on Council unanimity; the word of comitology had just been coined. At EU level the interest representatives reflected the characteristics of their domestic association (sector, size, composition) or government, such as type of regime (presidential or parliamentary), structure (central or federal) and size (large or small). Government-related differences of behaviour still exist, particularly in the new member states that have lower scores of both self-reliant behaviour and membership of EuroFeds. Increasingly they become replaced by country-related differences that show more variety, such as rich versus poor countries, ‘Catholic’ versus ‘Protestant’ administrations, industrial versus agricultural economies and collectivist versus individualistic cultures [Barron, 2011]. The new trend is that interest groups reflect in their EU activities less the traits of their state and country and more their specific organizational interests [Barron, 2011]. Increasingly they are pushed by both their selfish interests and pulled by their mutual EU competition, which drive them towards tailor-made actions rather than ready-made ones. They try to optimize routes, coalitions, EU access points, actors, factors and vectors, to mention only these ingredients for success. Increasingly they reflect not only their organizational traits, such as size (large versus small), focus (profit-oriented or not) or status (private versus public), but particularly their specific interests. This trend makes that two food producers or two ministerial food-policy units (one from Ministry of Health, the other from Industry) from the same country, behave differently at the EU level, in spite of similarities of state, country and food domain. As such they behave like the self-reliant chess player Gary Kasparov, who does not play in a standardized and thus predictable way, but always varyingly and surprisingly. Because the EU playing-field offers countless options for influence, it requires high selectivity and a tailor-made approach. Even lobby styles become tailored, so producing more lobby variety. For example, a formal request to a Commission official or MEP is increasingly precooked by semi-formal activities through expert groups or intergroups and informal lobbying before. Desired information and support is best gathered informally, safeguarded by semi-formal engagement, and 147

Pushing the Buttons of ‘Brussels’

Chapter 3

formalized at the successful end; reversing this order will bring bad results. Support from somebody else, who is usually led by the question ‘What’s in it for me?’, is not gained by one’s demand side (better be kept secret for some time) but by one’s supply side adapted to the other’s demand side, even if it is only a showcase of charming flattery. The intelligent and prudent group tailors its own profile, usually beginning not with a high profile, but low-key and quietly. This style is less upsetting to the desired partner and causes less alarm to competitors. The high profile is kept in reserve for emergency situations. An unfamiliar person or body is not approached directly but indirectly, such as through a trusted contact, a friendly coalition, an authoritative mediator or another effective U-turn. All this tailoring (and more) requires much study and prudence in advance and during any game. It serves the ambition of escaping the overcrowded arena of competitors and pushing the buttons of Brussels more efficiently and effectively, in short to achieve more success at lower costs. In the next chapters we fully focus on this professionalization of EU lobbying. Now we present short country profiles of diverse EU lobbying first.

V Extra: Country Profiles of EU PA and Lobbying The many variables of the EU machinery and the new trends among stakeholders allow a high degree of EU lobby variety. Characteristics of a member state or country have become subordinate to the individual drive to save or promote a selfish interest. Generalizations about, for example, ‘typically’ French or Spanish EU lobbying can hardly be made. Thorough comparative research on national styles of EU lobbying hardly exists. Only a few studies give some insight. In the mid-1960s, Altiero Spinelli [1966] qualified the styles as ‘the easy and smoothly operating Belgians, not making too much trouble’; ‘the most difficult and obstinate Dutch, distrusting their southern neighbours and niggling about the most minute details’; ‘the opportunistic Italians, with their fine sayings and lazy doings, and more obsessed with Rome than Brussels’; ‘the dedicated Germans, wanting to reacquire esteem and acting with a great sense of primordial duty’; and ‘the ambivalent French, being antagonistic and constructive, sending their best officers, and holding them to co-ordinated instructions’; the modest Luxembourgers he overlooked. In fact, Spinelli based his qualifications on the behaviour of the six member-state governments at Council level. Another study [Van Schendelen, 1993] described the trends towards selfreliance and tailored actions for public and private groups in different countries. Studies that compare countries on their adaptation to the downstream of EU decisions provide supplementary insights useful for our 148

Chapter 3

Pushing the Buttons of ‘Brussels’

upstream focus [Bulmer and Lequesne, 2005; Kohler-Koch, 2003; Wessels and others, 2003; George and Bache, 2001; Kassim and others, 2000 and 2001]. Besides there are studies based on interviews, surveys or memoirs and single country studies, to be mentioned below. Due to the lack of systematic research, the summary of country profiles below can only be a tentative sketch, to the best of our knowledge. It is limited to the EU countries and thus leaves aside the EU lobbying from outside the EU, for example from candidate countries such as Croatia and Turkey [LaGro and Jørgensen, 2007], non-EU European countries like Switzerland and Norway that are otherwise linked to the EU [Church, 2006; Archer, 2004] and the US [Winand, 1998]. After a short sketch of the basic national context, thus leaving aside short domestic turbulences, we focus on how interest groups from a country behave in the EU and reflect the structural trends above. Successively we present the five old larger countries, the ten old smaller ones and the twelve new ones of Central Europe.

1 The Five Old Larger Countries French interest groups have an old reputation of acting under national coordination, in particular by the SGAE office that succeeded the SGCI in 2005 and now falls under the Prime Minister. Most of this reputation is dubious. The large-sized SGAE (in 2007 employing 183 people) is partially an internship programme for young graduates from the hautes écoles on their way to the ministries, agencies and Brussels PR. Its main product is not co-ordinating EU actions but making different policy views more coherent on paper and so instructing the French PR, out of the belief that effectiveness will follow [Guyomarch and others, 1998, 53-60]. This bureaucratic approach to EU affairs takes much time, which is rationally managed by planning early and storing memory, two outstanding qualities of SGAE. However, conflicts between the Elysée (President), Matignon (Prime Minister) and Quai d’Orsay (Foreign Affairs) often weaken coherency, especially in times of cohabitation when the first two have a different party-political colour. The French public and private interests groups often assess their interests at the EU level differently and, paying lip-service to Paris, go to Brussels self-reliantly. The main exceptions to this pluralist pattern are the issues falling under the President, such as security, treaty reform and a few others. The old power patterns inside France are rapidly changing since the mid-1990s, due to constitutional change and trends of decentralization and privatization that provide more EU lobby room to decentralized governments and private sector groups [Brouard and others, 2009; Culpepper and others, 2008]. Common to many French groups is that they are driven by 149

Pushing the Buttons of ‘Brussels’

Chapter 3

well-organized selfish interests [Drake, 2005; Grossman, 2007]. Agriculture, with its many small farmers, is run by strong umbrella associations. Industry, from chemicals to (private) utilities like energy and water, is tightly controlled by multinationals. A pantouflage (circulation) of leaders enables major companies to get close connections with ministries in Paris and often even a desk and funding there, although this is in decline now. NGOs and trade unions largely lack such a privileged position. At EU level, most French interest groups prefer positive rather than negative integration and specific rather than general interventions. Interest groups in external trade are frequently biased in favour of mercantilism (market closed for imports and open for exports). Centralized political planning, with strong input from experts and led by a hierarchy, is still popular, but now at the lower organizational level. Inside the EU machinery, originally of French design, the French usually turn the Triple P game into an art form, particularly regarding the formal variants of actors, factors and vectors. A special unit of SGAE is in charge of getting French persons nominated in procedures engineered before, treating Brussels as another Paris. But often the French mistake this meta-game for the end game, as they attach more value to the formal setting than to the actual process. Critical reports about French lobbying in the EU have been published by the National Assembly (under number 1594) in 2004, the Chamber of Commerce of Paris (report Karpeles) in 2005, the Ministry of Economy, Finance and Industry (report Dassa) in 2006 and the State Council (document 58) in 2007. This self-criticism regards failures at the EU output-side, such as loss of power, declining subsidies and Court procedures, often due to nonchalance at the EU input-side, where many French groups rigidly stick to their introverted plan of action and neglect the extroverted basics of PAM. Their behaviour is often formal, demanding and, in the final phase, loud and direct and thus ineffective and unlike their old antichambre. They tend to apply traditional influence techniques like encapsulation and coercion rather than modern PA techniques, although the latter are slowly arising and promoted by new PA circles like the AFCL. British interest groups are marked by self-reliance. There is a Cabinet Office for EU Affairs, but it only has a small staff, circulating among the ministries. In addition to its standard preparation of the UK voting position in the Council, it works mainly for settling interdepartmental conflicts and special interests of the Prime Minister. The number of such conflicts is usually small, because the last thing quarrelling ministers want is interference by the Cabinet. The special interests of the Prime Minister are often inspired by party politics, not to say the survival of the party as the EU is a sensitive issue-domain at home. Strongly voiced Cabinet demands usually reflect domestic values such as sovereignty, subsidiarity, rebate 150

Chapter 3

Pushing the Buttons of ‘Brussels’

(return on financial contribution) and open market, all indicating the low interest of this ‘awkward’ country in European integration [Heffernan and others, 2011; George, 2000]. The practise of self-reliance was encouraged under the Thatcher cabinets, as a side effect of her ideology of privatization, which left interest groups to be responsible for their EU affairs. The central government applied self-reliance also to itself by requiring from its civil servants experience in Brussels (‘otherwise no career in London’). This privatization of EU management prompted many UK interest groups to apply in the EU modern PAM they had discovered at home in the 1970s and to develop their own EU routes, coalitions, agendas and entries. When the UK government abstained from the EU Social Protocol and the EMU, it found British trade unions and financial organizations (the City) among the EU interest groups that opposed the UK and supported the EU’s social and monetary policies [Giddings and Drewry, 2004; Bache and Jordan, 2006]. Many UK companies, regions and NGOs have given birth to new EuroFeds, and Birmingham has been the first city ever to establish a lobby office in Brussels. Usually the British prefer negative rather than positive integration, general rather than specific legislation, and hard rather than soft law, often bringing them in opposition to French desires. They stimulated the birth of the EU agencies and regime values as ‘better regulation’, the latter officially for improving EU policies by impact assessments but in reality also for re-launching issues and delaying policies. Their lobby style is often partially tailor-made: informal and indirect during agenda-building, but visible and demanding during negotiations. Even if having no interest in an issue at all, they like to create nuisance value on it in order to get a compenzation for their acceptation of the compromise (which is an unpleasant word in English). Partially due to their feelings of discomfort with the formal institutions and the semi-formal work floors, the British lead in PAM consultancies. German interest groups in the EU often reflect the imprints of their decentralized state (called ‘federal’), their sectoralism and the economic weight of many of their organizations [Green and others, 2011; Maull, 2006]. The country has not one government but sixteen (the regional Länder) plus the federal one in Berlin, all fully equipped with their own ministries, parliaments and courts. Like a smaller-sized EU, the country is a community of laws rather than values, formalistic more by effect than by cause and accustomed to multi-level relationships. This makes Germans feel at ease in the EU, all the more so since their federal court in Karlsruhe acts as watchdog over the EU’s use of Treaty competences. Since the start of the EU the federal government has followed a pro-integration course and, probably due to its memories of past wars, it did so in a highly compliant way, keeping down its economic and demographic weight. Since the post-war 151

Pushing the Buttons of ‘Brussels’

Chapter 3

generation took power (starting with the election of Gerhard Schröder as Chancellor of Germany in 1998), it has been more selfish in its behaviour, like other governments in the EU [Wittlinger, 2010]. The regional governments, fearing the loss of their strong positions in ‘Berlin’ to a stronger ‘Brussels’, are often less EU-minded (except for subsidies) than their federal one and less than most regions in other countries. All have a Brussels office that they also use for watching each other and the federal PR there. Through their Bundesrat, the Länder-controlled federal ‘Upper House’, they take the German seats of the Council working groups for dossiers falling under their domestic competence. The governments cherish their Ressortprinzip (strict competences) and Expertenkultur, controlled by much Juristerei. Due to its disagreements at home, Germany is often represented by several ministers in Council meetings or has to abstain from voting, to reject a proposal, or to resort to the Court. National co-ordination is more a daydream (or nightmare) than a reality. The implementation of EU decisions in the Länder is another problem, resulting in many infringement procedures. In short, the upstream and downstream to and from the EU are full of domestic veto points [Dyson, 2003]. Germany has been the trend-setter of multi-level collective action. Decentralized and private interest groups first enter regional platforms that through federal platforms are connected to EuroFeds, where the MNOs increasingly have direct membership. The private groups score highest on membership of a EuroFed [Eising, 2009, 76] and at the EU institutions they exceed even the French [Quittkat, 2009]. Many have an office in Brussels. They usually prepare their EU positions self-reliantly and thoroughly, drafted by many experts and signed by a hierarchy of chefs, much like their ministries. Due to all these deliberations among experts and chefs internally and with the members of their platforms externally, they often arrive late on the playing-field, fall short of tailor-made proactive lobbying and thus are dependent on more formal approaches in the end-phase of EU decisionmaking. Their lobbyists usually have a limited mandate, meaning they have to wait for new instructions from their headquarters while the EU match is going on. They have difficulty with British-style negotiations; the German word for ‘offer’ means ‘sacrifice’ and not ‘supply’. Demanding and vociferous lobbying is not their style, which is less the result of a tailor-made approach than of their fortunate middle position between the French and the British on many EU regime and policy values. Since the early 2000s, the country shows growing interest in PAM and the setting up of PA platforms. For Italian interest groups, self-reliance is less a choice than a necessity, because since 1946 the country has had a new Cabinet almost every year on average and thus lacks a stable government [Bull and Rhodes, 2007; Cotta and Verzichelli, 2007]. The relatively young state – dating from 152

Chapter 3

Pushing the Buttons of ‘Brussels’

1861 – is fragmented by volatile party politics (partitocrazia), separatist regions and antagonisms among big companies. Its fragile legal and administrative system, corrupt practises and protectionist clientelism earned it the name of ‘Europe’s sick man’ [Mammone and Veltri, 2010]. Its politics and politicians enjoy very low esteem at home. Its government, mostly the forefront of twenty regional republics gaining in power with every change of the constitution, is kept together by the bureaucratic formalism of ‘Rome’. Here the President’s Council, the Ministry of Foreign Affairs and the Ministry of European Affairs cover EU affairs officially, and so no authority does really. Other ministries try to run their own EU agendas, as do regional governments, major associations and most multinationals. Finding that the central government takes too much time with formalities and hardly achieves EU successes, they all prefer to be selfreliant, out of necessity. Private interest groups usually operate through their domestic umbrellas, like Confindustria for business, linked to EuroFeds. In Brussels exist about 120 Italian front-offices (antennas) of Italian interest groups [Raffone, 2006]. On the input side of the EU machinery the Italian groups and particularly the ministries have a reputation for dealing in words rather than deeds, except for collecting subsidies. They are frequently absent in the EU drafting phase, applaud whatever the EU decides and, if working inside the institutions, tend to be focused on their network (apartenenza) in Italian politics. The country’s record of implementing EU laws is one of the poorest in the EU, as is their record of correctly spending EU subsidies. Since the early 2000s there has been, however, change in the air [Franchino and Radaelli, 2004]. Tighter EU inspections have begun to make Italians on the EU input side more alert. Yet, the quality of those 120 Brussels offices is assessed as mainly poor: locked in domestic rather than EU logic, weak on strategy and tactics, fragmented on interests and amateurish. Some regional and sectoral groups from northern Italy set the example of professionalization now. They play the EU game with charm, informality and secrecy, so reflecting domestic styles. They engage in clever alliances, such as the public-private coalition between the Piedmont region and the car company Fiat or, as it is called, the ‘horizontal axis’ running from the Lyons region in France to the state of Slovenia with the region of Lombardy in the middle. In 2004 the Commission concluded a Tripartite Agreement with the Italian central government and Lombardy on the implementation of transport mobility rules, which was the first-ever agreement between the EU and a regional government. Spanish pressure groups in Brussels reveal their own mixture of domestic features. Coming from centralism under Generalissimo Franco until 1975 and thanks to the 1986 third enlargement, Spain became one of 153

Pushing the Buttons of ‘Brussels’

Chapter 3

the greatest EU success stories of modernization [Magone, 2009; Closa and Heywood, 2004]. It is now a German-like federal state of seventeen moreor-less autonomous regions, all with a facility in Brussels. Both party politics and bureaucratic formalism like in Italy characterize the central government in Madrid. As in France, for promoting their interests in the EU most groups rely on their experts and on approvals from chefs, with coherencia on paper as the criterion of success. The main example of active EU lobbying is the establishment of a Brussels office by the region of Catalonia, the first-ever region to do so. New multinational companies in sectors like banking and telecom have also become active in the EU. The countless small-sized interest groups mainly operate through numerous domestic umbrellas that take the routes to EuroFeds and EU institutions gradually more self-reliantly. Spanish interest groups are notorious for arriving late on the playingfield in Brussels, particularly at the Commission level. Like the Germans, they spend too much time at home. Unlike the Italians, they remain committed players, keeping their words more in line with their deeds. When discovering proposals for legislation at a late stage, they tend to resort to the next-best game of the exemption lobby, once a largely Spanish innovation in EU lobbying that has become common practise. Not wanting or able to say ‘no’ to a proposal, they push for a ‘yes, provided that…’ and ask for an additional article giving Spain an exemption to the rule for a number of years; an early example is the 1997 Telecom Liberalization Directive. Like the Italians, they are rather averse to complying with EU law but, unlike them, they manage to make this vice legal by resorting to exemption articles. Another Spanish speciality and an exception to their slowness, is the subsidy lobby. Many regional and sectoral groups tailor their actions in this area. Acting low-key, indirect, charming and informal, they try to influence the Commission on the objectives and conditions of the next subsidy programmes by Triple P, so increasing their chances of milking more euros. The Spanish central government negotiators in the Council often exhibit the opposite styles that make them unpopular here.

2 The Ten Old Smaller Countries Small-sized states usually have in common a two-fold inclination to stay out of the way of the large-sized ones and to play the mixed tactics of taking a low profile and binding others by their lobbying, mediation and entrepreneurship [Grøn and Wivel, 2011; Steinmetz and Wivel, 2010]. In return for a loss of formal autonomy they try to get positions and influence around the EU tables. Their domestic groups often have more influ154

Chapter 3

Pushing the Buttons of ‘Brussels’

ence at the EU level than in their national capital, thus they can even win on real autonomy. How are they lobbying the EU? In the past, major interest groups of the Netherlands believed that their small-state position is best protected by the Commission rather than the Council and that their welfare with an export ratio of about 55% of GDP is most dependent on an open market, and hence they held a utilitarian preference for more supranational and negative integration [Van Schendelen, 2002]. Shortly after the Maastricht Treaty most members of parliament became critical of EU integration, which affected the Cabinet (only fragile coalitions), central government and the citizens that rejected their first-ever referendum on the 2005 Constitutional Treaty. At home many interest groups are bound to corporatist-like wheeling and dealing via their numerous umbrellas. The EU offers opportunities to escape this. The Dutch groups are active members of many EuroFeds, in which they feel at ease and like to act in the secretariat, the EuroFed’s information centre. Most act self-reliantly, many have a Brussels office and a minority are direct members of EuroFeds. Trendsetters are MNOs like the companies of Unilever and Philips and NGOs on environment and labour, followed by decentralized governments and SME umbrellas. Dutch Commission officials show low nationalism [Suvarierol, 2007]. Although the country is (after UK) second in Europe on adopting PAM, most groups have difficulty with tailoring their EU approaches. Many see their demands as generous offers to Europe, thereby earning the reputation of being direct, loud and blunt. They often lobby semi-formally on a long-list with a free mandate, in short as amateurs. Coming from ‘the first republic of Europe’ (Montesquieu), they see courtesy as intrigue. The commuting lobbyists often turn the fact that they live about two hours from Brussels into a disadvantage, by arriving in the morning and leaving in the evening, thereby neglecting opportunities for informal networking. In 1994 the PR was moved to an eccentric (but cheap) Brussels suburb near the highway leading home, and in 2010 it was moved near to Place Schuman. The Belgian and Luxembourger interest groups belong to the young states. Belgium has a weak sense of nationalism, a fragmented state structure (five decentralized governments and parliaments) and an unstable federal government [Deschouwer, 2012; Swenden and others, 2006]. Due to their high export ratio (about 65% of GDP), the two countries usually give strong utilitarian support to open market integration, except Luxembourg for its (protected) banking. Their behaviour can best be characterized as ‘pragmatism’, the academic word for improvization. They can do so because, living around the EU’s main lobby city Brussels [Van Schendelen, 2008], they can hardly escape informal networking and enjoy the overrepresentation of countrymen at almost all Commission levels, in all about 155

Pushing the Buttons of ‘Brussels’

Chapter 3

one-fifth of personnel. Due to their language issue at home, most Belgian groups are organized at only the regional level (Flanders and Wallonia) and are linked to EuroFeds along this line. The regional governments have their own people inside the Belgian PR and their ministers rotate in Council meetings. The two countries have French-styled party-politicized bureaucracies, driven by experts and formalistic chefs and full of internal distrust that makes private groups distrustful too. In the EU the Belgians seem to act tailor-made: informal, silent, indirect and charming. They are famous for free-riding on stronger foreign groups, which usually comes from their ability to identify even the smallest escapes out of any difficulty rather than overcome it with PA skills. In 2006, the Flemish government established the public-private desk VLEVA to support its interest groups in the EU. In 2010 the Belgian caretaker government earned respect for its outstanding (rotating) chairmanship of the Council. The Nordic countries (Denmark, Sweden and Finland) score high on nationalism, decentralization and open market orientation, but Finland has fulfilled this profile only since it became free from Soviet control [Raunio and Tiilikainen, 2003]. Their interest groups have many conditions in common. Brussels is far away, which is a hindrance for frequent and intense informal networking. Lobbying at the EU level is at home under critical debate, so only partially accepted. Their few multinational firms (such as Carlsberg, Scania and Nokia) apply PA methods. The many smallsized groups go to Brussels indirectly via their domestic associations, public-private networks and their PR. The Nordic parliaments initiated the parliamentary power to instruct their government on the national vote in the Council and the Danish parliament the now almost common practise of an own desk in Brussels. Their faith in parliamentary democracy often conflicts with achieving EU success, as an instructed minister can hardly negotiate. To solve this problem, the new trend is that the government drafts the parliament’s instruction. Coming from wealthy and frugal countries, the Nordic groups inside the EU usually support careful spending and both environmental and social policies. Their ‘Protestant’ culture of transparency and accountability makes them supporters of these regime values in the EU and means that their capitals often leak information from inside the Council, otherwise the oyster of Brussels. By keeping their words in line with their deeds, they have earned a reputation for predictability, which may reduce their lobbying effectiveness. Tailor-made behaviour comes from only the few multinational companies and some national umbrellas. Interest groups from central government often excel in direct, formal and demanding behaviour, especially at Council level. Irish, Portuguese and Greek interest groups have, of course, many idiosyncrasies. Not least among them is the Irish reputation for saying ‘no’ to 156

Chapter 3

Pushing the Buttons of ‘Brussels’

Treaty change, which is stimulated by its domestic constitution that prescribes a referendum whenever such a change is proposed. Many Irish groups have strong interests in the policy areas of industry, trade and open market [Laffan and O’Mahony, 2008], while those from Portugal and Greece want their agriculture and small enterprises to be protected. Irish groups act highly informally and via public-private networks, often led by their few MNCs. In Brussels they like to meet in their Irish pubs that function as an informal PR for all Irish people. Portuguese and Greek groups come from more formalistic societies. Their governments like to control their ministries and to intervene in the private sector, which by the absence of MNCs lacks counterweight. Without sufficient size and other resources, most private groups can hardly act self-reliantly. For the remote control of the EU, they are dependent on their umbrellas and domestic government [Pari, 2003]. Their governments put more effort in the production of coherent position papers at home than in lobbying in Brussels, except for the Council level. The three countries have, however, much in common too: they are small-sized and located far from Brussels; clientelistic party politics and rising regionalism fragment all three; their interest groups usually play a marginal role on the input side of the EU and rely on their few nationals in the Commission and EP and on their government’s Council position; like the Spanish they lobby mainly for exemptions and subsidies and have earned a reputation for milking structural funds; their records on implementing binding EU decisions and allocated subsidies correctly are weak; and, not the least, since 2010 the three countries were the first under budgetary control by the EU. More than the Portuguese and Greek the Irish are politic in tailoring their EU actions and in acting informally, charming and low-key, as they do at home [Murphy and McGrath, 2011]. All three countries are largely unfamiliar with PAM. Austrian interest groups are stigmatized by national corporatism, the academic euphemism for being held together by distrust. Believing in formal authority, they see the Council as decisive and many interest groups hold – in order to watch each other – positions inside the Austrian PR, such as officials from the federal level, nine Länder governments (like in Germany, all fully equipped), organizations of employers and workers, and major business groups like the Chamber of Commerce and the banks. This unique PR entitles its members to sit in Council working groups, an Austrian innovation on EU influence practises, though it is hardly followed by other countries. Some groups have their own office in Brussels or sit in the Austrian House. Various government units at both federal and Länder level try to co-ordinate all Austrian public and private groups, thus nobody really does. These units may get instructions from their formally powerful parliament, which causes frictions between domestic democracy and EU 157

Pushing the Buttons of ‘Brussels’

Chapter 3

effectiveness like in the Nordic countries. Broader group formation at home is difficult and time-consuming by mutual distrust. Consequently, many groups often arrive late in the EU processes. Since many have their links with the EU work floors and EuroFeds in Brussels, there is a rising trend toward self-reliance, particularly among the regions. Tailor-made actions remain exceptional, but Austrian groups often belong to the winners on many EU dossiers. The main reason is that, like the Germans, they often hold centre positions that lie close to the final compromise. An example is the EU agricultural policy, at issue between countries like the UK and France, but under the flag of Sustainable Agriculture coming close to the Austrian way of farming. There are no MNCs to set the trend of modern PAM, but interest in it is growing.

3 The Twelve New Central and Eastern European Countries Systematic observations of influence behaviour in the EU are still scarce for the twelve Central and Eastern European Countries (CEECs), from the Baltic States in the north to Cyprus and Malta in the south, which joined the EU in 2004 (10 states) and 2007 (Bulgaria and Romania). The following is based on general books [Olson and Ilonszki, 2011; Henderson, 2009; Epstein and Sedelmeier, 2008; Grabbe, 2005; Schimmelfennig and Sedelmeier, 2005], specific country studies, articles, lobby cases, reputations at the EU level and background data [Circa, 2008]. Easy generalizations cannot be made. The countries show a lot of differences, of course, such as by size (the village-state of Malta versus big Poland), language (eleven different ones), age (old empires like Bulgaria, Hungary, Lithuania and Poland versus the 19th-century Romania and the 1991 Slovenia), recent history (the 1956 uprisings in Hungary and Poland, the 1968 one in (then) Czechoslovakia, the 1993 separation of the Czech Republic and Slovakia), regionalism (arising in Poland and Hungary and reflecting patterns existing before 1939), ethnic composition (like the 30% Russians in Latvia and the diasporas of Hungarians and Roma), growth of civil society (high in Slovakia, low in Lithuania), economic sectors (industrial Hungary and Slovakia versus agricultural Poland and Lithuania), foreign conflict (Cyprus with Turkey) and political system (parliamentary Estonia versus presidential Poland) [Börzel, 2010]. Yet most new members also have common characteristics. Except Malta, they were all occupied, first by Germany and then (except Cyprus) by the Soviet Union, from the late 1930s until the late 1980s. Except Slovenia, they all have had some form of independence before, even if only for a short time (Baltic States) or as protectorate (Slovakia). Their citizens show a strong sense of national sovereignty and cultural differences 158

Chapter 3

Pushing the Buttons of ‘Brussels’

between the generations. According to socio-economic data (DG ESTAT) and compared to the old EU-15, all countries score high on the EU poverty standards, dependency on family support, cheap labour, unemployment, reliance on agriculture, lack of MNCs and low exports to the EU. Their economies are relatively informal (untaxed payments) but undergo rapid transformation towards Western practises. The countries have high levels of educational quality and (until the 2008 financial crisis) rapid economic growth; many countries prepared themselves for the adoption of the euro, and some already have. In spite of progress made, many countries score poor on the EU criteria of ‘good governance’, as they have weak rule of law and much corruption [Sissenich, 2010]; for this reason the Commission froze EU subsidies to Bulgaria and Romania in 2008. The political systems of the ten formerly communist countries have experienced rapid transitions from totalitarian to limited government, one-party to multi-party rule, dependent agencies and courts to independent ones, old to new policy and regime values like privatization and decentralization, and old to new political personnel. Personalized party-politics, volatile electorates and weak coalitions (with, for example, until recently a new Polish Cabinet almost every year) all indicate the fragility of these transitions, but signs of stability are seen around the 2010s. The CEEC’s europeanization is rapidly proceeding, as shown by the first country studies [Szczerbiak, 2011; Marek and Baun, 2010; Laurestin and Vihalemm, 2009; Hrebenar and others, 2009; Sepos, 2008; Papadimitriou and Phinnemore, 2008]. In their lobby behaviour at the EU level the CEECs have also much in common: – Ambivalent about the EU. After regaining independence from foreign occupation, the countries are ambivalent on europeanization (‘no Brussels instead of Moscow’). Both their citizens and politicians often express a preference for subsidiarity rather than the EU laws and even, except regarding subsidies, for sovereignty rather than subsidiarity. – Dependent on national capital. As their business landscape is very flat (hardly big companies) and both their decentralized governments and civil society are in an adolescent phase, most interest groups are dependent on their central government for their influence on the EU and use their parliament for influencing it [Fink-Hafner, 2011]. – Council focus. Having still only few nationals inside expert groups and comitology, the central governments are focused on the EU Council and its Brussels PR; by 2012 their share of Commission officials has become fair. If the Council decides unanimously, that focus on it can result in last-minute obstructions, as can be seen in the actions of the Polish government under the Kaszynski twins during 2005-2007. Often 159

Pushing the Buttons of ‘Brussels’

Chapter 3

the central governments have to play second-best games in the Council, such as on exemptions and subsidies, and to fight an uphill battle against Western stakeholders that got their interests included in Commission proposals. Some governments retaliate, partially for domestic use, with a lot of demanding, noisy, direct and formal exposure, thereby losing goodwill. – Unstable national positions. Due to unstable domestic politics, their policy positions in the Council are often unstable too, as ministers tend to come and go and their policy positions with them. – Common interests. The countries have a few common interests, such as agriculture (less food safety law and more subsidies), free movement of workers (export of cheap labour) and security (their borders with Russia, Belarus and Ukraine); in terms of the latter interest, some states, such as the Baltic ones, have set up a regional alliance. In early 2009, on the eve of the Council Summit, nine of their heads of state met informally to join hands on more financial assistance to CEECs. The twelve CEECs have a past without a developed public affairs culture and are now becoming aware of this. Particularly under the pressure of EU and foreign investors, their old influence practises such as favouritism, policy cartels or dirty lobbying have become increasingly scandalized at home and in the EU and have made way for the new practises of PAM [Harsanyi and Schmidt, 2012]. The trends of lobbying as mentioned for the EU-15 are already becoming visible in the twelve new ones. Interest groups from decentralized governments and businesses increasingly travel to Brussels and use a domestic consultancy or their own office there. The Polish airline LOT has been an early trendsetter of self-reliance, while the regional governments of Hungary took early trainings in ‘how to escape Budapest’ in 2001. In 2008, the Czech Republic, Hungary, Poland and Slovakia each had about forty interest groups present in Brussels, making a total of 160 groups, 40% of which represented business groups and the remainder (in fairly equal portions) decentralized governments, NGOs and professional groups. The national co-ordination of domestic interests is more being set at issue at home and probably over its peak. Many more interest groups have begun to attach value to modern (PA framed) EU lobbying (although its name is, due to domestic tradition, often mistaken as corruption) [McGrath, 2008], become critical about the weak support from their government, joined EuroFeds [Pleines, 2011; Eising, 2009, 76] and lobbied self-reliantly for financial assistance and supportive coalitions [Borragán, 2003]. Inside and via these EuroFeds they become familiar with both the tailor-made styles of PAM and the EU back doors (on which ten countries received unasked pre-training in their com160

Chapter 3

Pushing the Buttons of ‘Brussels’

munist past). They are encouraged by the Commission to participate on its work floors, attend PA trainings in Brussels or at home [Schneider, 2007] and are eager to get professionalized and to get influence. Those from the Baltic States and the four countries mentioned are, in particular, earning a reputation as runners-up in the EU’s ‘PA league’, as they succeed in changing EU policies, legislation and financial programmes more to their desires.

VI Professional EU Public Affairs Management This chapter shows that interest groups wanting to influence the EU can apply many sophisticated approaches such as creating vectors and Triple P, that they undergo structural trends of change at EU level such as collective action and tailored actions, that they are increasingly preparing themselves to act self-reliantly in the EU and that they are less interested in reflecting the imprints of their country or state. Arriving in Brussels, they show growing convergence towards modern PA and lobbying. As Machiavelli pleaded above, the EU offers every interest group from any country much room to ‘push its buttons’. The precondition is that one has to gather expertise on how the EU really works and can be influenced so that successes and not boomerangs follow. Our catchword is professionalism, which stands in contrast to amateurism and should not be confused with the full-time lobbyist or consultant, who may behave amateurish as well. Both have ambitions, but the professional carries out careful preparation and prudent actions, while the amateur acts like a tourist (‘flying-in and flying-out’, or FIFO), who returns home with souvenirs and maybe scandals instead of trophies from Brussels. The professional PA manager at the EU level sees more options for behaviour and is capable of selecting optimal combinations that score better on effectiveness (goal realization) and efficiency (cost-benefit ratio). The PA expert fine-tunes the management of actors, factors, vectors and, if possible, the meta-game of Triple P. The expert also takes into account the current EU trends among stakeholders and the situation in their home country. All this knowledge is only a necessary means, as the objective (dependent variable) remains the winning of a desired game or, as nextbest pay-off, an encouraging compromise, respect and backing. The single most important factor (independent variable) of lobby success, PAM expertise, can always be improved. Of our formula 2E = MI2 both the M and the I2 are always in change and subject to PA research and development. Expertise is never a property, but always a quality to be earned every day. Increasingly the stronger competition among interest groups requires 161

Chapter 3

Pushing the Buttons of ‘Brussels’

ever better expertise and gives birth to a ‘premier league’ of the most capable players, something that Darwin might have called ‘survival of the political fittest’. Regarding the fieldwork, intelligent interest groups puzzle over the question, ‘What is better to do?’ For example, is it better to have co-ordinated or self-reliant units in the home organization, to politicize or depoliticize an issue, to enter a standing or a flexible alliance, or to behave actively or passively? The amateur player hardly considers such dilemmas or solves them dogmatically or at random, while the professional’s answer to the question is: ‘It depends’. The next question, then, is: ‘On what?’ This is a matter of preparatory homework and prudent fieldwork regarding many variables that can determine the outcome. These key variables can be set in a new series of questions about both oneself and the other players, such as ‘Who acts why, for what, to whom, on what, how, where, when and with what result?’ They apply to all interest groups, including the EU officials. These questions of professional EU public affairs management are summarized in figure 3.3, with a breakdown for the key activities of homework and fieldwork. The figure can be seen as modules for learning. The interest group at Kindergarten level has only an emotional approach to some perceived event (or non-event) and only one question: ‘how to promote or to prevent that event?’ Without further thought, it is fascinated

Professional Public Affairs Management/Lobbying Questions

Homework: Analyzing

Fieldwork: Organizing

1. Who Acts?

the internal organization

improving the organization

2. Why?

threats and opportunities

choosing the strategy

3. For What?

menus and options

determining the targets

4. To Whom?

finding crucial actors

building relationships, networks

5. On What?

selecting issues, dossiers

bargaining

6. How?

methods and techniques

LOBBYING

7. Where?

route planning

bringing partners together

8. When?

time and agenda-phase

timing, agenda-building

9. Result?

process evaluation

learning, improving

Figure 3.3

162

Chapter 3

Pushing the Buttons of ‘Brussels’

by the last catchword on the sixth line, ‘lobbying’, and immediately tries to lobby other people on a corridor and to open a door, usually a revolving door. Then the learning process starts. At primary school level the group begins to wonder about reality and to pose a range of questions (the first column of the figure). Advancing to secondary school, the adolescent group believes it now understands the world and knows how to score there. Quick answers are provided (third column). If a clear strategy or a network is needed, it is made within three months; their quality is less important. The group becomes a doer, full of energy, spending many resources, but hardly scoring. A few groups go on to university level. They learn to think about all options for playing and about the better choices to make (middle column). Other groups often take them as an example to follow. The various questions in the figure have been placed in the sequence that most interest groups follow in practise. They start with questions regarding their own organization (Q 1-3), followed by those regarding the arena, including the issues and stakeholders involved (Q 4-5), continue with questions related to precisely tailored lobbying, locating and timing (Q 6-8) and between times pose the question whether the desired result is coming nearer (Q 9). This sequence is not necessarily logical or politico-logical, not to mention that it may be wiser to try to understand the world around before taking a position and to make the sequence dependent on the specific situation. In daily PAM life all questions are often on the table simultaneously and require integrated puzzling. In our educational approach here we shall discuss the questions from the perspective of not what is usually done but what is usually wiser to do. In the next chapter we shall focus on the questions regarding any EU arena’s stakeholders and issues (Q 4-5), in chapter 5 on the EU public affairs agenda inside the home organization (Q 1-3 plus 9) and in chapter 6 on the external fieldwork, including the optimal choice of lobby styles, location and time (Q 6-8).

163

Chapter 4 Getting Grip on an EU Arena Those wanting to transform a republic should first study its nature at any time, as otherwise they cannot apply the proper methods and will harm both the republic and themselves. What does this view of Machiavelli (Discorsi, book 3-8) imply for one’s ambition to influence the EU?

I Whom to Lobby in an Arena and on What? Following up on the end of the previous chapter, we dedicate this one to the extroverted window-out scanning and the management of what happens in a particular arena. An arena is not a physical place, but the virtual collection of stakeholders, including EU officials, together with their interests-at-issue with regard to a specific dossier at a specific moment. Usually a dossier is a paper or proposal from the Commission, but it may also be an issue forwarded by another institution or a stakeholder’s platform. As in the EU every dossier is controversial to some extent, a ‘dossier’ and ‘arena’ can be used as synonyms. Both are time-specific, as they can change by content or composition any time, in short they are situational. After the general introduction of arena-analysis below, we shall outline the method of scanning an arena and present the ‘quick scan’ for getting an elementary overview of either a single dossier or the long-list of dossiers assessed as relevant. Because this preparatory homework is only a necessary means for PA success, we subsequently anticipate chapter 6 and present as appetizers ‘better lobby practises’ based on such homework, and also a case-study. In chapter 5 we question how the interest group can downsize its long-list into a shared-list and a short-list and can get better organized for window-out and window-in activities. Our leading question here is whom to lobby in an arena of stakeholders and issues, and on what? The simple-minded interest group, at no higher than ‘secondary school’ level and still amateurish, will regard this question as superfluous. It knows what it wants to get from EU officials and finds their names and addresses in an EU directory. Then it approaches them as directly as possible and tries to convince them that their best response is to comply with its demand as soon as possible. If it gets a refusal, it will probably make a lot of noise at home or in the EU and launch a second strike, now joined by a number of ‘friends’ gathered in the meantime. It may also get advice from a semi-amateurish or semi-professional group 165

Getting a Grip on an EU Arena

Chapter 4

that already have had some experience on the EU playing-field and follow some of their rules of thumb, such as that it should first approach some national PR officials and nationals in the EP and/or in the responsible Commission DG, and that it should invite them to a free lunch or something similar. The professional group, perhaps observing that lunch, may consider this more charming, informal, quiet and indirect approach as preferable to the confrontational one, but also as poorly engineered for its composition (only domestic friends) and sequence (making trouble first). It knows that in the open and competitive EU arena the better way to lobby always depends on the arena’s situation. It investigates, first of all, ‘the situation’ and breaks this down into its four main components, being the stakeholders, their interests-at-issue (in short: issues), the time dynamics and the arena boundaries. By collecting information (I) on them and setting its brains to work, it gathers useful intelligence (I2) by which it can identify its friends and enemies (or partners and opponents), the issues at stake, the time aspects and the differences between the insiders and the outsiders. All this is a matter of window-out preparatory work or, in shorthand, homework. Then it may come to know to whom to lobby on what issues, and how to do it most effectively and efficiently. Then it can apply the better practises of managing an EU arena and its four components, in order to build up sufficient window-in support for its targets. Even a formal, demanding, noisy and/or direct approach might then be appropriate for a specific situation. All this puzzling over an EU chessboard may be seen as logical, not to say self-evident. However, in daily life most pressure groups behave in an amateurish rather than a professional manner. Apparently they consider as not so essential the quest for useful knowledge that, in the past, took three waves of PA thinking before it became established in theory and among professionals.

II The Predecessors of Arena Analysis The first wave may be called ruler analysis. It is the oldest and most classical one and among many interest groups reflected by the use of the management title ‘government relations’ instead of ‘public affairs’. In this old approach a group or, at the lowest level, a citizen, may have a strong interest in the outcome of some decision-making process and may want to influence the authority that is formally in charge of that decision. To this end, it may try to collect useful information about the highest official in charge, in order to increase the chance of success. In the old days the King or the Prince was the highest ruler to be approached. In his Il Principe, 166

Chapter 4

Getting a Grip on an EU Arena

Machiavelli gives not only free advice to his ruler, Lorenzo dei Medici, on how to survive politically but also, more implicitly, provides a code of conduct to his fellow citizens on how to successfully approach a ruler. More recently, until the mid-1970s, US scholars studied interest group behaviour for its impact on the institutions that formally rule, such as the Congress and the President [Key, 1964], or on the members of Congress alone [Milbrath, 1963; Scott & Hunt, 1965]. Interested in the safety of private investments in foreign countries, other US scholars studied the political risks presented by changes of rulers there [Coplin and O’Leary, 1976 and 1983; Gigerenzer and others, 1989]. Another US scholar applied the ruler analysis to the EU [Gardner, 1991] by equating ‘effective lobbying’ with knowing the idiosyncrasies of its highest authorities. This type of knowledge is certainly still necessary, but it is also insufficient. Its weakness is the assumption of political hierarchy or central rule, neglecting the presence of both negotiable issues and stakeholders intervening from outside. In modern jargon it is only actor analysis, limited to the top officials. Many French interest groups still belong to this first wave of PA. The second wave is described as issue analysis. It started in the late 1960s, as a branch of ruler analysis, under the flag of policy studies. Authoritative institutions and officials were placed in a dynamic perspective, namely as producers of public policies. They make their choices, called ‘policy’, from a surplus of more or less contested specific facts, values, instruments and/or solutions, in short from a menu of issues [Heath, 1997, 44 and 84]. Some US scholars linked this study to that of interest groups from business that want to get friendlier policies [Bauer and Pool, 1960; Bauer and others, 1963]. Others broke the concept of policy down into issues and presented their issue analysis as a tool, called ‘issue management’, to interest group managers [Kingdon, 1984; Buchholz, 1990; Frederick and others, 1996; Heath, 1997]. They also made a key distinction between ‘opportunities’ and ‘threats’, which are assessed as issues having respectively a positive and a negative effect on the interests of the specific group. US-educated people within the EU [Andersen, 1992; EC Committee, 1994] were soon promoting this approach. In many cases the issue analysis remained closely linked to ruler analysis, resulting in the study of issues inside, for example, the US Congress [Hojnack and Kimball, 1998] and the EU Commission [Emerson and others, 1988]. One theme of issue analysis has become very popular: the ‘dossier life cycle’ inside an institutional arena. A dispute over a specific fact, value, instrument and/or solution may start out silently or unnoticed. At some point it may emerge into the institutional arena and attract public attention in the form of a bell curve: rising boisterously, achieving some settlement and sinking to slumber or death. A similar biography may apply to 167

Chapter 4

Getting a Grip on an EU Arena

a set of connected issues, called ‘policy dossier’. The dossier life cycle helps us to understand the usual dynamics of an issue and is summarized in figure 4.1, which outlines its different phases, separated by barriers, and presents at the bottom-line for each barrier the most important parts of the EU machinery. It begins with the rise of a ‘social problem’ (phase 1) that may attract mass public attention and, passing the first barrier, can achieve the status of a political problem (phase 2), whereupon it may pass the second barrier and be placed on the official agenda (phase 3) and, overcoming the third barrier, become subject to decision-making (phase 4), after which barrier the decision may be implemented (phase 5), inspected for effectiveness (phase 6) and finally get stronger enforcement (phase 7).

Getting an Issue on (or off) the EU Agenda High Barrier 4 Political decision?

Barrier 3 On official agenda? yes Public Attention

Barrier 2 Political problem?

yes

yes

GOVERNMENTS COMMISSION EP

Low

Barrier 5 Implementation?

no: non-agenda no: non-decision

no: non-issue Barrier 1 Social problem?

yes

yes

no: nonimplementation

Barrier 6 Inspection? yes

no: non-inspection no: non-problem

COM EP GOVTS

Barrier 7 Enforcement? yes

no: non-enformcement COM COURT COUNCIL

Time

COM EP, ESC, COR COUNCIL

COM GOVTS COMITOLOGY

COM AGENCIES COURT

High

Figure 4.1

This life cycle has a stylized order here, but in reality it may run highly erratically and include feedback loops. An example is the new problemsetting in 2010 of the (weak) enforcement of the monetary criteria for national budgets that appeared to have been violated seriously, which 168

Chapter 4

Getting a Grip on an EU Arena

indicates the gradual europeanization of the life cycle. Anticipating chapter 6, the real life of the cycle may be like human life and thus, for example, suffer an early death, never get wide recognition or be gone with the wind. Viewed in retrospect, however, every established policy has more or less followed the life cycle and survived like an éminence grise. All this knowledge helps one to grasp the ups and downs of an issue or dossier inside an institutional arena. However, it is not sufficient, as it takes the life cycle as an institutional affair among, for example, EU officials. Other stakeholders it takes as only senders of messages from outside, so neglecting their dynamic interventions. However, many stakeholders from outside the institutions are in reality active participants inside. The third wave of research is stakeholder analysis, which took hold in the US during the mid-1980s [Phillips, 2011; Phillips and Freeman, 2010; Huxham, 1996; Alkhafaji, 1989; Carroll, 1989; Freeman, 1984]. A stakeholder is broadly defined as any organized body that can affect or is affected by the achievement of the objectives of one’s organization [Freeman, 1984]. In a pluralist society, every interest group must cope with a variety of such players that hold a stake or position in an issue area. The study of their behaviour is, technically, multiple-actor analysis. Stakeholders want to push their own interest and to block any opposing interest. They intervene in both the policy process and in the relationships between competitors and officials. Adopting a position during the life of the dossier at stake, they may become either a supportive partner and even a friend or a reluctant opponent and even an enemy, and join a coalition. Others remain ambivalent or indifferent, as a result of either mixed feelings or calculated negotiation behaviour [Savage and others, 1991]. There may even be simulating stakeholders, who give the impression of having an interest in the matter but in fact only want to build up nuisance value to be used on a different matter; many English interest groups have earned a reputation for this skill. At EU level, the stakeholder analysis is applied both by many practitioners and in a few EU case-studies [Pedler, 2002; Wallace and Young, 1997; Greenwood, 1995; Pedler and Van Schendelen, 1994]. This knowledge is useful and necessary, but its strength is also its weakness, as it suggests that the decision outcome is essentially produced by the interplay among stakeholders only.

III Arena Analysis: Going Window-Out Because all three aforementioned types of analysis are considered both necessary and insufficient, they deserve to be synthesized into one more integrated approach. This fourth wave is called arena analysis and is part 169

Getting a Grip on an EU Arena

Chapter 4

of public affairs [Parg, 1981]. An arena is, as said before, the virtual collection of all the stakeholders, including the officials, together with their issues regarding a specific dossier at a specific moment. Taken as a whole, it means ‘the situation’ on which depends the answer to the permanent question, ‘What is the better thing to do?’ Issues and stakeholders are not the sole criteria for the better action, although both are relevant. The dynamics of time and the arena boundaries are crucial too. Time provides opportunities and constraints, thus the concept of the life cycle is appropriate here. Every arena always has a particular composition of stakeholders and issues, which can change as soon as old stakeholders and issues leave and new ones enter by crossing the arena boundaries, which thus must be researched carefully as well. Of course, it is all the perception of both the situation and one’s own interest that drives behaviour. In order to avoid incomplete and false perceptions, an extroverted study of the situation is necessary, as free from preconceived ideas as possible. Adults being framed by their values, experiences and perceptions from the past usually have greater difficulty with this than youngsters with their open minds and curiosity. The extroverted discovery of the situation, reflected by the metaphor of ‘going window-out’, is indispensable for arriving at an overview of the situation that is outstanding by quality (reliability, validity) and quantity (completeness). Only then can an interest group see in advance the many different stakeholders in the arena, their many different interests in the same dossier, the opportunities and constraints of both time and boundaries, and particularly its own usually fragile position in this complex and dynamic arena. With such organized intelligence from the onset, it may be able to organize sufficient critical mass for its interests and to manage the arena with good chance of efficient effectiveness, thus to realize the formula 2E = MI2. Its difficulty is shown by the many EU interest groups that are usually driven by their preconceived ideas of the situation that only include their selfish interest and a few befriended stakeholders, so neglecting most of the situation. Arena analysis thus aims at acquiring realistic insights regarding a specific arena, not as a goal but as a necessary means for achieving better arena management and final success. It brings together all window-out intelligence (the upper arrow of figure 1.2), not performed once like taking a photograph but repeatedly like filming the battlefield before and during the fight. The description of the arena should precede the analysis that produces the intelligence needed for any action. All this preparatory work for every arena each time usually takes a lot of energy. Simple games with low complexity and dynamics seldom exist. A level playing-field is difficult for all, as stakeholders, issues, time and boundaries are not fixed but flexible. An unlevel field may not be so difficult for the insiders uphill but 170

Chapter 4

Getting a Grip on an EU Arena

is difficult to maintain in the longer run, while those in a downhill position may have a reversed view of difficulty. Even a seemingly simple game that falls under Council unanimity decision-making is usually complex, as many stakeholders from other institutions and from outside try to intervene informally via back doors or side-dossiers. In contrast, not all arenas of secondary legislation are difficult, as the 1992 Slots Case shows [Van den Polder, 1994], in which the national air carriers like Lufthansa and Air France could, after their proactive arena study, mobilize ‘their own’ ministers in the Transport Council in order to get protected their private distribution of landing rights (slots) against EU distribution as demanded by new carriers like Virgin and Lauda Air. Arena analysis requires a lot of work and resources and thus can best be organized in a shared-list approach that divides costs among many and enhances effectiveness. Lobbying without conducting an arena study beforehand seldom results in success, unless by sheer luck. As in equally complex and dynamic games like chess (or, indeed, warfare), proactive investigation and critical reflection are the best investments. Usually they make the difference between winners and losers.

IV Describing an Arena Determining ‘what is better to do in lobbying’ depends on the situation. Thus, the studious interest group wants to know in what sort of arena it is before it starts acting. By going mentally and physically window-out and setting its radar to work, it tries to collect useful information about the four main components of any selected arena: the relevant stakeholders, issues at stake, time dynamics and arena boundaries.

1 Stakeholders The serious interest group really wanting to win an EU game starts with drawing up an inventory of stakeholders that may intervene, including the officials on the matter and, of course, itself. In this chapter we assume that the group has cohesion and stable objectives; in the next chapter we will challenge this assumption. The inventory must not be biased to take in only friends or enemies or only formal entities. It is like military reconnaissance: all warriors on the field are taken seriously for their weapons and thus respected (as can be seen in the etymology of the Latin verb respectare, ‘looking around at armed people’). On a specific dossier a group of Greek farmers, a Lithuanian ministry, the French President and a Commission expert may all be equally relevant. Even non-EU players may be 171

Getting a Grip on an EU Arena

Chapter 4

relevant, such as the Thai exporters on the tapioca dossier, the Central American countries on bananas and the WTO on the steel trade. Groups from Switzerland, not an EU member but bound to many EU decisions, are often important stakeholders in policy areas like transport, food and banking [Sciarini and others, 2004]. All this may sound trivial, but in daily life most interest groups already fail on this first step of arena monitoring. There are two common mistakes. One is being biased when making the inventory. National ministries often view an EU arena as an intergovernmental field of battle and neglect local governments and private interest groups. Many a business group neglects the intervening role of a national ministry or decentralized agency. NGOs are often fairly good on unbiased scouting of an arena. Secondly, many interest groups deify stakeholders by putting them in the inventory as, for example, ‘the British position’, as if such a stakeholder ever exists [Thomson and others, 2012]. Every country is, in fact, pluralistic and thus is always divided regarding every dossier. At best, there may be a statement from the Cabinet or Prime Minister, which gets lip-service from other ministries that take their own way to Brussels for saving their interests there. The same holds true for the position of, for example, Greenpeace, Siemens or the Commission. Accurate scouting requires that stakeholders be inventoried at the lowest relevant level, even if this is as a unit or an individual. In the early phase of a dossier, a solid (unbiased and specific) inventory often results in a long list of stakeholders. In later phases its length is usually reduced to a very few coalitions. If waiting for this later moment, one usually must tackle compromises precooked by these coalitions. In any case, information of potential relevance for a later phase should be saved from the start. Subsequently, the identified stakeholders must be assessed in terms of relevance. A relevant stakeholder is one who is expected both to intervene actively and to possess sufficient capabilities to become influential. The stakeholders are placed along two crossing continua: one ranging from active to passive and the other from strong to weak [Mitchell and others, 1997]. Regarding the first continuum that covers the key variables of participation (chapter 1-IV), it is not sufficient to include only the stakeholders who are visibly active. Many others may act in an indirect or hidden way or may become active soon. The second continuum covers their resources for influence, such as expertise, skills, reputation, networks, budget and other assets (to be discussed in the next chapter). They indicate the stakeholder’s potential to become a winning warrior. Potential influence should, as stated above, never be equated with real influence, as a prepared army with many resources will not necessarily be victorious. Finally, a rank order of stakeholders should be made and kept up-todate. Most relevant are the primary stakeholders, identified as prone to 172

Chapter 4

Getting a Grip on an EU Arena

action and potentially highly influential. Those being assessed as remaining passive and/or falling short of sufficient potential influence can be taken as stakeholders of secondary importance. They should not be deleted but placed on a memory list as they may join the primary group in a next dossier phase, and the reverse may happen as well. The drawing up of the inventory of relevant stakeholders clearly requires a lot of information and intelligence. The cleverest stakeholders will try to play both passivity and impotence, like a cat seemingly sleeping on the roof but really watching the birds. The most assertive stakeholders, suggesting they can hardly refrain from strong action, may be like barking dogs that do not bite. One can even make a further typology of stakeholders and assess them as, for example, dormant, demanding, dominant, risky, dependent and definitive [Mitchell and others, 1997]. The outcome of all this homework is merely a reasoned estimate, but this is much better than nothing. Below we shall show that it enables taking prudent measures with regard to every major stakeholder [Hunger and Wheelen, 1998, 112].

2 Issues All stakeholders are driven by their own interests, which are chosen positions on issues arising in their environment and usually caused by new facts, values, instruments and solutions to problems. The positions represent daydreams and nightmares regarding the arena at stake. A chosen position always contains a confrontation between, on the one hand, cherished values, instruments and solutions and, on the other hand, perceived or expected facts. How stakeholders come to a chosen position or to the definition of their interest, and how they can arrive at a better choice or definition, will be part of the next chapter. The focus here is on the identification of the interests of the stakeholders in an arena. In pluralistic societies an asserted fact, value, instrument and solution is always contested by others, as they contain subjective elements. A sample study of online consultations shows that the Commission received much diverse responses from almost 2,700 interest groups on 56 issues [Klüver, 2011]. Positions and issues are thus almost synonymous in the field of PAM. In the EU practise of decision-making by negotiations and compromises, a professional group has to know which different positions there are on a dossier and for which conflicting interests they stand. With this information, it can take solid sides by getting primary stakeholders interested in its position and so become part of the final compromise. This information is necessary but never sufficient for achieving a desired or satisfying outcome. Without it an interest group has, however, a low chance of such an outcome and remains dependent on sheer luck. 173

Getting a Grip on an EU Arena

Chapter 4

Again, all this may sound trivial and self-evident. Yet, most interest groups fail to acquire this type of knowledge. Introverted and living in their own world, they often make two big mistakes. One is arrogance: believing that their interests are perfectly right, they expect that sooner or later other stakeholders with different interests shall join their ‘excellent’ position. They may even become proselytizing and enforcing, two variants of a Holy War to which there is usually no successful end. The other is naïveté: others not sharing their position must be wrong. Other stakeholders may, however, be perfectly right on their different positions, for example, because they adhere to different values or facts. The controversial draft-directive on Laying Hens (COM 1998/135), promoted by NGOs on animal welfare and aimed at allocating more space for laying hens, was seen by industrial egg producers as bringing additional costs, by traditional farmers as a comparative advantage, by suppliers of better cages as new business, by trade unions as worsening labour conditions and by northern governments as an expenditure to support their industrial farmers, and these are only five out of many more issues in this dossier. If all the players remain introverted, the result would be akin to an ancient Greek tragedy in which all the heroes are both justified and end up dead. The professional group respects the relative or subjective backgrounds of any issue and wants to identify and understand them in order to develop an interesting supply side before the negotiations. The amateurish group, lacking respect for the positions of others, hardly makes the effort, because it does not understand how this would help it to get its demand side accepted. The identification of current issues is not at all easy. Their volume is high in the early phase of a dossier but rapidly decreases once coalitions align into formation, thus making early shared-listing rational. It is particularly difficult to identify the real positions of primary stakeholders. Their official statements can never be taken for granted. The cleverest players disguise their internal divisions, remain silent or cover their real interests. The amateurish stakeholders seldom have a solid position, as it is based on weakly reasoned values and poorly perceived facts. The discovery of real positions requires a great deal of knowledge about stakeholders. As in diplomacy, one has to monitor their internal arenas at home, their dominant values, their perceptions of reality and the forces behind (and that may change their position). From this information the intelligent interest group may deduce their real interests or positions. Once knowing that a stakeholder, as usual, has more than one real interest on a dossier, it can find even more room to manoeuvre. By assessing their various interests as friendly, hostile or indeterminate, it can seek to build a coalition with the friendly ones, engage a fight with the hostile ones and take time or give 174

Chapter 4

Getting a Grip on an EU Arena

cheap support to the indeterminate ones. It can even make a trade-off between friendly and hostile interests and so come to a package deal. Such preparatory homework can even be used for other dossiers and thus create room for even more package dealing. The five stakeholders mentioned on the Laying Hen Dossier were all involved in other EU dossiers, so it might have been possible for them to come to terms in a broader package, thereby settling the current issues dividing them. It is all a matter of knowledge first.

3 Time Every EU arena continuously changes with time. The professional interest group is conscious of the relevance of time analysis. From general experience it knows that in the early phase of a dossier’s life, normally many stakeholders and issues are present; that thereafter most stakeholders become part of a larger coalition and most issues part of a package; and that in the end-phase of an EU decision process usually only two groups remain, these being the proponents and the opponents of the proposal, and only a few issues are left, such as ‘the envelope’ (who pays the bill), the date it takes effect and any exemption demanded. The professional group also knows that the dossier’s life is only partially regulated by procedures setting deadlines for, for example, consultation, codecision, inspection and review. The next phase starts once a barrier has been passed (figure 4.1). In its own rhythm, any irritating difference in Europe can be elevated to the EU, adopted by officials on their agenda, challenged by other stakeholders, mixed up with different issues or dossiers, processed into the EU machinery and, finally, decided upon, implemented and inspected - or not all that is happening. During the dossier’s life, some stakeholders act under time pressure; others act as though they have all the time in the world. Those in a losing mood want to delay, those in a winning one to speed up the process. Time, in short, distributes constraints and opportunities. All this is general wisdom that the professional group demonstrates when it, for every new situation, studies the time dynamics precisely in order to ‘fine-time’ its activities. Most interest groups are, however, nonchalant about time in an arena and thus make many mistakes and blunders. They enter their arena when the dossier comes into the public eye, which is usually halfway through its life cycle, or they miss the delegated or implementing dossiers that hardly get any public attention. Thus, they are usually too late to seize upon a growing opportunity or to stop a growing threat at its birth and only notice other stakeholders and issues after taking their position or shape. If taking a photograph of the 175

Getting a Grip on an EU Arena

Chapter 4

arena at all, they mistake it for the film. They are hardly aware of arising coalition building or package dealing and hope to meet formal deadlines at crunch time or to be accepted even thereafter. In short, they can hardly be proactive and effective on their interests. Delaying or speeding up the process is beyond their capacity. Their lobbying is like fighting one fire after another, directed at preventing only the growth of current threats, not its occurrence or break-out. In contrast, professional groups have within reach updated protocols for immediate and flexible responses to various types of serious threats, such as Unilever applied to the 2008 financial crisis. The calculation of time is clearly not easy. It takes a great deal of work, as it linked to the outcomes on stakeholders and issues, and must be regularly updated. Early shared-listing helps to get this done at lower costs. Even then difficulties remain. Competitors playing Triple P can make the early phase full of implicit decision-making that is subsequently binding. The early framing of the Laying Hen Dossier as a case of animal welfare strongly determined its outcomes. The phases of implementation and inspection may be full of substantial decisions at the detailed level. The professional group considers them to be highly relevant as ‘both the devil and the angel are in the details’ and thus remains alert and prudent. Keeping its inventories of stakeholders and issues updated, it links them to formal deadlines and crucial moments and reassesses any new situation in terms of time constraints and opportunities. By checking crucial information it may avoid being misled by, in fact, its own insufficient intelligence. In its time management it tries, where helpful or necessary, to delay or to speed up the process. Time knowledge is based on a reasoned estimate only, but this is much better than nothing. It enables intervening more at the right moment.

4 Boundaries As an arena is permanently changing with time, it is also changing in its composition. The studious interest group thus explores the arena boundaries. Seldom are they fixed or closed. Usually, every stakeholder can introduce new issues and every issue can attract new stakeholders. Triple P arrangements may limit the openness and the composition of an arena, but once a dossier attracts wider public attention, as often happens in the legislative phase of the EP and the Council, new stakeholders and issues can always enter. One example is the initially closed shop on the Implementation of the Aarhus Convention (chapter 3-III) that was opened in this phase. Another is the Laying Hen Dossier that similarly changed from almost a single stakeholder and issue (animal welfare) into multiple stakeholders and issues. Delegated and implementing dossiers usually attract 176

Chapter 4

Getting a Grip on an EU Arena

low levels of public attention as most MEPs and journalists let them pass by without remark, and their arena boundaries tend to be fairly stable. For many reasons stakeholders can also leave an arena and take their interests with them. Such volatility of arena boundaries is normal for an open society, and no different from any garden where animals searching for food come and go across boundaries and fences. Most interest groups neglect the volatility of arena boundaries. They consider the composition of an arena to be static like a photograph, and hardly pay attention to stakeholders and issues that may disappear out of or step into an arena. Apparently, they hardly care about the causes and consequences of this boundary traffic, even though it might affect their own position. Due to their nonchalance, they can hardly anticipate future coalitions and issues that may take them by surprise. Proactive management of boundary traffic, aimed at attracting supportive new stakeholders and issues or at removing the opponent old ones, is almost beyond their comprehension. In both the Aarhus and the Laying Hen case, the uphill groups had blind trust in their closed shop and forgot to anticipate the possible entrance of the groups downhill. The professional group focuses on the volatility of the arena boundaries by monitoring those who enter or leave, including their interests. From this traffic information it tries to deduce useful intelligence for improving coalitions, framing issues and keeping track of the time. Proactively it may try to bring on board new supportive stakeholders and issues and to outbalance or get rid of the old or annoying ones. It also broadens its homework by including adjacent dossiers or arenas, as any overlap is useful for creating cross-traffic or roadblocks, in short for pushing or blocking another combination of stakeholders or issues. Knowing that growing public attention to a dossier is the single most important determinant of attracting new stakeholders and issues, it therefore prudently wants to keep silence if the current arena is promising. Only in a dreadful situation will it make noise in order to enjoy the immediate effect of new stakeholders and issues that enter and delay the current process, so preventing an immediate loss. All such fine-tuning is dependent on the quality of the boundary analysis. Even if based on only reasoned estimates, it is much better than nothing as it can differentiate between a win and a loss.

V Gathering Information for an Arena Description Two short and contrasting cases first. One is the 1992 Dossier on High Definition Television (HDTV). In the early 1990s, the European consumer electronics industry, organized as EACEM (now EICTA) and including compa177

Getting a Grip on an EU Arena

Chapter 4

nies such as Thomson (France), Philips (The Netherlands) and Bosch (Germany), demanded from the EU both a legal standard and a big subsidy for R&D on HDTV [Verwey, 1994; Cawson, 1995]. Soon they received personal support from the DG RTD Commissioner Filippo Pandolfi, once a board member of Olivetti, and the promise of a subsidy of 850 million ecu (now euro). But in 1992, they appeared to have ignored opposing stakeholders, such as broadcasters (fearing the costs of rebuilding their studios), satellite companies (offering a better alternative) and consumer groups (against compulsory purchase of new television sets). The mass media, smelling fire, finally helped to bury the dossier, the standard and the subsidy. In hindsight, the companies regretted their lack of proactive arena study. The other case concerns the 2004 development of the Radio Frequency Identification Device (RFID), a new high-tech barcode with great efficiency potential. Before going to lobby for the introduction of RFID, some European food retailers like Carrefour (France) and Ahold (The Netherlands) first made an arena study and found that the new device would raise many issues such as on environment (waste, hazardous substances), privacy (data protection), health (electro-magnetism), competition (cost of RFIDs prohibitive to SMEs) and unemployment (around 50% of retail personnel) and, of course, would mobilize those interest groups to oppose the introduction of the RFID. With foresight, they decided prudently not to lobby for the RFID and to leave this to others, such as the producers. The description of the four components of any arena requires a lot of information that is neither readily available quantitatively (completeness) nor qualitatively (reliability, validity). By shared-listing this work can be divided among many and result in a higher quantity and quality of ‘data mining’, like President Obama’s 2012 electoral campaign. In the EU there is no shortage of useful sources for collecting specific information on our leading question here: ‘Whom to lobby in an arena and on what?’ According to an online survey, interest groups rely, in descending order, on their own monitoring, media, commissioned research, informal meetings and EU sources, with some variations for type of group [Chalmers, 2011]. A more detailed list of sources is mentioned in the upper part of figure 4.2 and explained as follows. 1 Desk research. The internet is a major tool for carrying out research. The EU website, being more open than most national government sites, provides information about much that is in the EU pipelines, but not everything by far. Sites of national governments may provide information about their own positions and, thanks to domestic transparency, some governments (like that of Sweden) can be more open about, for example, the Council than the EU site is. Academic sites offer datasets on dossiers from the past. National newspapers provide additional 178

Chapter 4

Getting a Grip on an EU Arena

information and can be easily consulted through websites like eufeeds.eu. Many established stakeholders have their own websites. Search engines like euractiv.com collect many of their documents. A goldmine of information is memory, as many arenas are recurrent and many stakeholders and issues with them. Corporate memory held in archives can be most useful, but the materials are usually poorly organized, except in France where information from past fights is highly valued as useful for future ones. The SGAE is famous for doing this. 2 Debriefing friends. Every interest group can make use of other groups that share its interests and, once it has identified them, exchange information. Collective platforms find one of their major functions just in the sharing of costs and benefits of monitoring. After some time, every group can find informants inside EU buildings and among stakeholders, including media, thanks to a match of interests and/or backgrounds (like passport, education or hobbies). For its own information needs, the Commission makes wide use of numerous in-sourced experts and, additionally, academics and consultancies. All groups can try to debrief the befriended people among them, as well as their own technical experts. 3 Visiting the corridors. Much detailed information, however, is not available in sufficient quantity or quality from the sources mentioned above, and has to be gathered or checked by going around the corridors, the mediaeval form of ‘lobbying’. Usually the most important place is the Commission, as it does not publish most pre-drafts of its texts, particularly not those under the non-legislative procedure, and thus has to be visited directly or through mediators indirectly. On legislation, the PRs acting on the crossroads between their national ministries and all other national capitals can be very informative, but only a few are willing to assist non-nationals. Journalists often get access to high EU officials competing for ‘good publicity’ and can, if trusted, be used to ask or check information, in exchange for a scoop later. Relevant are also the conference meetings in rooms of hotels in the EU quarter of Brussels, where stakeholders discuss current dossiers, often together with EU officials. The corridors around such rooms are the modern ‘lobia’. An ambitious interest group will not only search for information within the frame of the dossier at stake, but go beyond and try to broaden its mind by an open search that uses techniques mentioned in the lower part of figure 4.2 and briefly clarified here. 1 Comparative study. By studying a more or less similar (homomorphous) dossier it can find new solutions to issues at stake. For example, on an issue like working hours, which falls under social affairs, interest 179

Getting a Grip on an EU Arena

Chapter 4

Major Sources and Techniques of Arena Monitoring Targeted Search: 1. Desk Research Regarding: - Websites: EU, Countries, Stakeholders, Services - Memory (Personal and Corporate) - Newspapers, Mass Media 2. Debriefing Befriended Key People in: - Collective Platforms - Institutions and Agencies - Committees, Intergroups, Working Groups - EU Assessments (Academics, Consultancies) - One’s Own Organization (Technical Experts) 3. Visiting the Corridors of: - EU Officials - Stakeholders - Journalists Open Search: 1. Comparative Case-study 2. Evaluation: Getting Hindsight Wisdom 3. Piloting Conferences, Seminars and Media Debates 4. Launching Balloons and Rumours 5. Role-playing Simulation Game 6. Open Networking 7. Basic EU Knowledge

Figure 4.2

groups in different sectors like hospitals, retail, road haulage, fire and rescue, and manufacturing can all learn from each other’s dossiers on this matter, as this usually reveals useful similarities and differences for influencing their own case. 2 Hindsight wisdom. By evaluating a recent game and posing the question ‘which surprise could we have seen coming?’, an interest group can gather hindsight wisdom, useless for the past game but most useful for improving the monitoring of a next game. Well-organized memory yields a real reward here. There is always room for improvement and, as the costs of shortcomings have already been paid, this is gratis, as the HDTV companies learnt. 3 Pilots. A popular technique is to organize, at invitation, a conference, round-table or seminar for the sake of getting varied insights on an issue area, such as on ‘cisgenic fruits’ or ‘district heating and cooling’. Variants of this are open debates in newspapers or online, which have the advantage of producing even more varied insights, but also the dis180

Chapter 4

4

5

6

7

Getting a Grip on an EU Arena

advantage that they can be controlled less and so can early mobilize undesired opposition. Balloons. The launch of balloons or rumours may be informative. Such an activity could be carried out through, for example, a befriended MEP posing a question or a mass medium publishing a report. The objective is to solicit reactions from officials and stakeholders who take them as serious. This technique requires the anonymity of the real sender, much fine-tuning and great prudence. Simulation. Every interest group can organize internally a role-simulation game, in which the participants play very different roles that include EU officials and devil’s advocates and during which every player wants to win. The outcomes are often spectacular, not only for their wealth of information but also for the better understanding of even the greatest enemy. Open networking. The technique of just going around, like in classical diplomacy and lobbying, for watching and listening without any specific target in mind, is still an enlightening technique and usually brings new insights that are useful. Theoretical knowledge. The seemingly complex and dynamic EU machinery with its many arenas can be understood and anticipated better by using theories about its causes, trends and implications (previous three chapters). By its critical questioning of easy answers, theoretical knowledge is a search machine for better information.

An interest group may consider the exploitation of all these sources and techniques, aimed at getting a useful early description of the four components of an arena, to be costly and troublesome preparatory homework. By shared-listing its costs are much lower but may still be found to be a burden. If so, it should make better calculations of the effectiveness and efficiency of its operation, such as the following. – Comparative costs. Lobby action without engaging early on homework usually creates the biggest troubles, as one may miss relevant stakeholders, crucial issues, better moments and boundary traffic. The comparative costs of losing the game usually exceed those of homework. – Savings. If after early homework the interest group finds that the arena is less complex and dynamic than believed before, it may save on the costs of further homework and actions. – Comparative advantage. If this homework reveals the opposite, it can give the interest group the edge over competitors that are nonchalant about their homework as it knows better how to manage the arena with a good chance of success. Usually one can win more in a difficult arena (or market) than in a simple one. 181

Getting a Grip on an EU Arena

Chapter 4

– Time reduces costs. As time goes by, all the homework becomes easier. At the end of a fight there are usually only two coalitions, one in favour and the other against, and only a few last issues remain, such as ‘the envelope’, the date of entry into force and a possible exemption. Time also reveals who is merely a comedian, a passive cat on the roof or a barking dog. – Earlier shared-listing. The sooner an interest group starts its homework, the earlier it may find stakeholders with similar interests and willing to join forces for shared-listing and burden-sharing. They usually compete not on gathering information (window-out) but on using it (windowin). This last hurdle can be solved by negotiating a common position that also enhances effectiveness. – Short-list burdens the most. As stated in chapter 1-X and elaborated on in this and the next chapter, the most intense homework has to be done on the short-list that for reasons of 2E should remain really short. – Spin-offs. The gathering of information among stakeholders and officials can produce great spin-offs, particularly trusted relationships with them, by which one can more easily go from debriefing to briefing, from window-out to window-in and from information to influence.

VI Optimal Rather Than Minimal or Maximal Homework No interest group will ever have an overview of an arena that is, in terms of quantity and quality, perfectly complete, valid and reliable at any moment. As in warfare or in a game of chess [Simon, 1979] this is beyond human reach. No information source or technique is watertight or totally confirming and not contradicting another one. An arena may be so volatile that the interest group always suffers from lack of time and other resources. Thus, the criteria of completeness, validity and reliability block any useful outcome if they would be maximized. Neither is it rational to minimize them, as this would result in an arbitrary trial and error, followed by blunders and boomerangs. As usual in life, one should optimize rather than maximize or minimize. The main exception to this is a crisis situation in which one’s survival is at stake, which justifies maximal ‘war room’ efforts for getting more information for ever better intelligence that, however, remains optimal at only a higher level and is anyhow disadvantageous to one’s many other interests. Seeking full completeness is a utopian goal as there will always be events, incidents or coincidences that are not foreseeable or observable. Even in describing the past that is easier to do than the present, no historian can ever achieve a complete description. Validity, meaning that one finds what one seeks, is always a problem. For 182

Chapter 4

Getting a Grip on an EU Arena

example, the potential influence of a stakeholder remains potential and not yet proven by real impact. Perfect reliability, so that an observation can be taken as fully true or factual, is beyond human capacity. Some stakeholders always willy-nilly mislead the observer. A correct observation today can be outdated tomorrow, such as when a stakeholder redefines its interests in the arena. Due to all these obstacles to perfection, every interest group must be very ambitious to collect the best possible insights, but must also accept that the results and thus any of its PA decisions can only be optimal and thus continuously needs supplementary alertness and prudence, all based on the ‘science of muddling through’ [Lindblom, 1959] or our ‘art of lobbying’ (figure 1.2). Hence we phrased our leading question on arena management based on the arena description not as ‘what is best to do?’ but as ‘what is better to do?’ The optimized description has at least two bottom-lines. One is the reasoned estimate, this being the educated guess regarding the composition of an arena, to the best of one’s possible knowledge. This is not a licence to take it easy, as the level of ambition should meet that of the best competitors, so to say, one’s best friends in ambition. The reasoned guess protects an interest group against the two extreme alternatives, on the one side searching for what is beyond reach and on the other side acting blind or at random. The second bottom-line, still at a high level, is the repetition of the homework by one or more other observers in parallel setting for the same moment in time (duplica) and/or in series for a later moment (replica). The duplica gives the better photo and the replica the better film. By setting several observers to work on the same duplica or replica one will usually end up with, thanks to their intellectual competition, much better insights, as they will deliver different observations and their good reasons for them. By bringing their differences into critical discussion they all will be able to achieve a higher degree of collective wisdom. Therefore, a small team that in a shared-list operation is always available can better first be split into individuals each doing the same arena study and then be brought together for critical discussion. The replica is essential for observing changes in the arena. When observing that a new stakeholder or issue is entering (or an old one leaving), one should question whether this has been caused by always slightly unpredictable arena dynamics or by shortcomings of the previous homework. In the latter case there is room to improve the use of information sources and techniques.

183

Getting a Grip on an EU Arena

Chapter 4

VII The Quick Arena Scan Due to the interdependency among the four components of an arena, its study for the sake of successful PA management requires an integrated approach. Its simple form is the quick arena scan that can be applied to any dossier of the short-list or shared-list. Its essence is a two-dimensional matrix that has on the vertical axis the observed stakeholders and on the horizontal axis their issues involved (figure 4.3). The time dimension is included by regular repetition of the quick scan, which reveals the traffic of stakeholders and issues crossing the boundaries from or to the arena. On top of the form is mentioned the dossier under investigation, the date and the name of the observer. All together the exercise is no different from what people do in their minds when they think about an arena in a realistic and open way. Whether the major players and the key issues are pleasant or not, is irrelevant as long as they are real. The mind has, however, a limited capacity for considering many stakeholders and issues, for storing the collected insights and for doing all this in an unbiased way. It is better to write it all down on a piece of paper, but this method offers limited space. The best is to put the results of the quick scan onto a computer spreadsheet or grid with enough space and capacity for memory that allows to observe changes over time. The logic behind this is not different from the plan to enter a challenging place like a battlefield or jungle: one should first know its state and nature. Without a general overview first, no insight can follow. The quick scan often produces a sufficiently complete, valid and reliable helicopter view of the arena, from which (see next paragraph) can be distilled the best ways to proceed with PAM at a calculated risk or a reasoned estimate. On the vertical stakeholders axis there is endless room for listing all kinds of stakeholders, even with breakdowns made at the lowest relevant level, for example a unit or individual. The criterion is whether they are or can become relevant warriors in the arena. The scanning interest group can put itself on top; in the next chapter we shall question whether its position is sufficiently clear or needs an internal arena study, to compare with the external one here. The other stakeholders can be grouped into three categories. First, the domestic stakeholders that have an interest in the same dossier. They may be ministries, agencies, private companies, NGOs, trade unions, regional groups and any others, like organizations, units or even individuals. As interested players, they may try to intervene on the dossier at home or in the EU. Second are the foreign stakeholders, who may come from both EU and non-EU countries, including global platforms like the WTO; those from outside the EU can be as active and effective as those from inside. Third, the many stakeholders inside the EU ma184

Chapter 4

Getting a Grip on an EU Arena

Dossier Homework Dossier 1 : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (2 and ff = agenda analysis) Moment t : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (t–x …. t+x = time analysis) Analyst A : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (B and others for control)

Stakeholders

Issues involved 1

2

3

4

5

6

7

8

n

Own organization National -

Public and Private Sectoral Cross-sectoral Regional Others: media, etc.

Foreign -

Governments Sectoral Cross-sectoral Regional Others: media, etc.

Select items for trade-offs

Transnational -

COM (DGs, agencies) Committees EP, ESC, COR COREPER, Council Court EuroFeds, NGOs Others: media, etc.

Assess Their Actions + Potential Influence = Relevance

Identify coalitions for or against

Assess Relevance of Issues and Make for unfriendfly columns: loss scenario friendly columns: win-win/free ride scenario undetermined columns: scanning scenario

Other Dossiers and Players

Figure 4.3

185

Getting a Grip on an EU Arena

Chapter 4

chinery, ranging from Commission officials and experts to EP rapporteurs and Court referendaries. On the horizontal issues axis there is endless room for registering the interests or positions of the identified stakeholders regarding contested facts, values, instruments and solutions that are at issue. For every different issue they are grouped by column, which is at the top sharply defined for any issue in dispute. For reasons of efficiency it is best to take for this the positions of one’s own interest group on top as a starter, as this allows later on (see next paragraph) a quick overview of one’s chances to win a specific interest; if stakeholders have an issue that oneself does not have, one can take a position as yet, as this may give negotiation or nuisance value later on. If necessary, every issue can be split into more issues, which only gives even more columns. On the HDTV dossier, the issues on top might have been labelled as ‘EU standard’, ‘subsidies to the producers’, ‘studio compenzation’ and so on. The issues must be observed not introverted from one’s own position but from that of every stakeholder, this being the crux of extroverted monitoring. In the cells, where the two axes intersect, one can register for every stakeholder the estimated position on any issue in a simple and digital way, by putting in the cell, for example, a plus sign (+) if a stakeholder is assumed to support the issue as defined at the top; a minus (-) if one observes that the issue is perceived as a threat; a plus-minus (+/-) if one believes that a stakeholder is wavering or ambivalent on it and thus can go in both directions; a zero (0) if a stakeholder has really no position; and, if no position is observed, the cell is left empty. One may also apply scores that range from, for example, +5 to -5, thereby expressing the intensity of a stakeholder’s preference. This basic arena description for identifying potential partners and opponents on one dossier at a specific moment is indeed not substantially different from what every person can do in his or her mind or on paper. However, the computerized description is superior to the mind or paper, especially in terms of the following advantages. – Memory. It helps to save data for a long time and enables the inclusion of updates in order to study trends in the arena. The outcomes of replicas (at another moment, second line on top of matrix), which replaces the still photograph by moving film, can be easily added. – Checks. By organizing one or more duplicas (by other observers, third line on top), one can identify their differences and, by critical discussion among them, get better information. – Key stakeholders. To every stakeholder on the vertical axis one can add as measures of relevance their probable action and potential influence, which allow a closer study of primary stakeholders without losing information on the others. 186

Chapter 4

Getting a Grip on an EU Arena

– Key issues. On the horizontal axis one can mark the issues assessed as probably real and prominent, and not fake and fading, and also make a closer study of this hard core, without losing potentially relevant information on the other issues. – Blank areas. Empty and zero cells, indicating that a stakeholder seemingly or truly has no interest, which is often soon forgotten when working with the mind or on paper, remain within reach as sooner or later they may yet play a role. – Regulars. Many stakeholders and issues are regular parts of an interest group’s relevant dossiers, even for many years, as they are linked to its core business that is often fairly constant. These regulars it can easily download from its data files (corporate memory) and encapsulate in special networks and approaches. – Broader views. To the two axes one can easily add unexpected findings from the areas beyond the arena boundaries (the dotted lines at right and at the bottom of figure 4.3). They regard (horizontally) for every stakeholder its interests outside the dossier at stake but maybe useful for a trade-off or package-deal soon, and (vertically) the non-observed stakeholders that might or should have an interest in the dossier and thus may have to be mobilized or blocked soon. Often a game is won by this use of findings beyond the boundaries.

VIII From Information to Intelligence For every arena that is scanned as described before, figure 4.3 results in a useful data basis or file that one can use like Kasparov studying the chessboard and calculating his optimal moves. Many interest groups and consultancies produce, however, a much simpler arena study that, for example, is pre-framed after only institutional phases (formal time) or otherwise neglects important parts of the reality of any arena and thus falls short of PA study (and probably ambition and prudence) [Hardacre, 2011]. The making of a realistic arena scan that gives a useful data basis is a core expertise of PAM. It enables the transformation of information (I) into intelligence (I2) that is needed for both constructing a platform with optimal critical mass (M) and for identifying the better tactics for influencing others in an efficient and effective way, so making the formula 2E = MI2 real. The transformation of information to intelligence goes as follows. By vertical analysis of the stakeholders, the studious interest group can rationally arrive at one of three assessments of the situation it is or going to be in; this we shall do now by using the metaphor of wind and weather or political meteorology: 187

Getting a Grip on an EU Arena

Chapter 4

– Tailwind. Its own position on an issue (column), either being plus or minus, may be widely shared by most (primary) stakeholders. Here it has much room for shared-listing. For its interest the wind is blowing friendly and promising as a tailwind. The chance is high that its preference will come through during the competition with other stakeholders and EU officials and be approved. – Headwind. On an issue (column) most (primary) stakeholders may have an opposite preference and thus not share its position. Now it finds only small room for shared-listing. It feels an unfriendly and cold headwind blowing. The chance that its preference will prevail is low to zero. It has, so to speak, a nearly dead horse in the column, unless it finds new stakeholders with the same preference beyond the arena boundary (at the bottom). – Unclear wind. The interest group may observe that in a column most (primary) stakeholders hold positions that are indifferent (0), ambivalent (+/-) or still blank. Now, unclear side winds are blowing or there is no wind at all. The column is indeterminate, the game open and the outcome unpredictable, unless new supportive stakeholders are found beyond the boundary. With a few befriended stakeholders the interest group can share-list some interests and prepare for the tailwind or headwind that sooner or later shall rise. The horizontal analysis of the rows reveals the (maybe greatest) surprise that every stakeholder is always negotiable. In its row it has pluses (daydreams) that it wants to score and minuses (nightmares) it wants to get rid of. The 100% full opponent or friend is fiction, not a reality. The interest group has to secure the friendlier ones, of course. With an opponent it can make a deal by supporting, or threatening to oppose, it on one of its interests and, for possible package-dealing, by raising another interest as found in its area beyond the boundary at right. A group complaining that it cannot negotiate with other stakeholders has usually failed to make a good data file. In the HDTV case, the producing companies could have accommodated the opponents if they would have known their different interests beforehand. The costs of making a good arena analysis are much lower than those of misperceptions. An intelligent group earns the highest negotiation pay-offs on issues beyond the arena boundaries at right that are not crucial to itself, as it can cheaply offer highly valued support on it (like a cigar from the other man’s box) in return for support to its own interests. By making a diagonal analysis one gets the helicopter view. As both the horizontal and the vertical axis have nominal order, they may be reordered. By bringing both the pluses and the minuses together into aggregates, one sees the potential coalitions in favour or against a single issue or, summed 188

Chapter 4

Getting a Grip on an EU Arena

up for all issues, the full dossier. All stakeholders will see these coalitions at a late moment when they take shape, but the studious group can see the potential coalitions far before and anticipate them by clever actions. These coalitions are of high quality, as they are based on common interests and thus more solid than those based on, for example, only common language, tradition, passport or location. The helicopter view provides three further pieces of intelligence: – Fragile coalitions. No (potential) coalition appears to be perfectly unified. Thus there is always some room for undermining an opposing coalition, just as there is always a need for strengthening one’s own. – Mixed coalitions. Every (potential) coalition is always a mixed collection of various public and private interest groups from different countries. Through them and their networks one can try to make use of many routes and doors into the EU for one’s own interests and to erect road-blocks for opponents. – Shared-listing within reach. On any EU dossier no interest group is ever alone. The others are (potentially) supportive, opposing or indifferent. With friends it can enter a coalition or platform and, from then on, organize most PAM by shared-list that gives critical mass (M) and 2E. Finally, one can add the diachronic analysis, which is an assessment of trends over time or, to use our metaphor, the film instead of the photograph. On a computer grid one can easily store the history of the arena and acquire two benefits. Firstly, one can check one’s previous observations, such as regarding boundary traffic, coalition formation and package dealing. The more surprised, the more one can and should improve the data basis, as any surprise comes from failing I2. The second benefit is insight into the next development of the dossier’s life cycle. During the process most stakeholders will take a clearer position that allows the observer to fill in their previously ambivalent or empty cells. Some will come together, settle mutual issues and combine interests somehow. Others may fail to do so and become vulnerable to isolation. As time goes by, the two axes tend to become shortened, but the friction between the remaining groups of stakeholders and issues tends to intensify. This is normal once the dossier is at the end of a cycle’s phase, for example on the eve of the Commission’s proposal, the EP’s report, the Coreper’s A-points or an inspection procedure. The interest group with diachronic insight can, if required, reposition itself in plenty of time, not reactively but proactively.

189

Getting a Grip on an EU Arena

Chapter 4

IX Pro-Active EU Arena Management: Window-In The integrated and research-based study of an arena is, again, not a goal but a means to better action on the playing-field, to lobby window-in more successfully. A data file that is as reasoned and precise as possible gives great comparative advantages over the competitors that neglected this homework and seek gold without inspecting the ground first. Thanks to its steady work on the data basis, the professional group can always see reasonable options for trading-off issues, forming coalitions and getting the edge. Many an interest group may be embarrassed by the luxurious paradox of embarras du choix: seeing so many attractive options, it struggles to decide what is best to do. This dilemma is solved better by optimizing rather than maximizing the choice. In a difficult wood one route to the targeted destination may be the best in theory, but identifying it takes in practise so many resources that choosing one of the many found next-best routes is more rational. The really worst situation is one without much choice at all, such as in the simple arena of a fight with one’s spouse over the single issue of the next holiday’s destination, when he wants to go north and she to the south, for which the solution comes from lengthening the two axes of the matrix by inviting more people to join and by raising more issues among them. The proactive, research-based management of the EU arena can be described now for both the short and the long term. The only difference is the room to manoeuvre, which increases if there is more time.

1 Short-term EU Arena Management The principles of arena management are based on the four components of any arena, each being full of dilemmas and thus in need of criteria for choosing. Here we present for each component an exemplifying dilemma with the general question ‘what is the better choice?’ and with as (popular) choice criteria ‘low costs’ and ‘low risks’. (In chapter 6 we present a fuller menu of dilemmas.) 1 Stakeholder management. Should one exploit one’s own strengths and the opponents’ weaknesses or reduce one’s own weaknesses and the strengths of the opponents? In an EU arena the defensive strategy is usually easier and less risky than the offensive one, as one survives an ambush better by improving one’s shelter than by launching an attack. But by remaining under shelter one may get a World War I-style trench war, where nobody wins the battle. 2 Issue management. Is it better to push the opportunities to oneself and the threats to the opponents or to block the threats to oneself and the 190

Chapter 4

Getting a Grip on an EU Arena

opportunities to the opponents, thus to play either offensive or defensive again? In the EU arena, the blocking strategy is usually easier and less risky than the pushing one, just as one controls a car on a slippery road better with the brakes than the accelerator, but this may cause a slowdown in a queue. 3 Time management. Here a dilemma is whether to speed up or to delay the decision process. The delaying strategy is usually easier and less risky than the accelerating one, just as it is easier to keep one’s balance on a slow-moving rather than on a fast-moving treadmill. However, on a slow one more competitors can leap on as well. 4 Boundaries management. Is it better to widen or to restrict the arena boundaries? Here, the narrowing strategy is usually easier and less risky than the widening one, as one may negotiate better in a small group than in a mass meeting. However, the stakeholders and issues left out may become disruptive at a later stage. Easiness and low risk are, however, not necessarily the best criteria for deciding about lobbying ‘whom in an arena and on what?’ Often they result in poor rewards too, as their possible consequences above show. The interest group accepting big trouble and high risk may become the winner taking almost all or the victim of its own greediness by losing successively the desired outcome, respect from others and backing at home. Except in a crisis situation threatening their survival, most EU interest groups prefer moderate troubles and risks rather than zero or high ones. They prefer the good chance of winning little bits rather than the bad chance of losing almost everything, even if the latter also holds the small chance of winning it all. The rationality behind this moderate and prudent behaviour is usually their interest in many more dossiers than just one. The sum of many small but easily and safely obtainable trophies they see as more attractive than the one big prize with great trouble and high risk of full loss. The question remains, however, whether such avoidance of trouble and risk is rational. The blocking, defensive, delaying and restrictive practises are often nothing more than rules of thumb. Once a dossier or issue area is considered relevant, the rational choice of acceptable troubles and risks should not depend on popular rules, but on the arena’s situation. Above we distinguished three types of arena: friendly or promising, unfriendly or endangering, and indeterminate or unpredictable. These labels should not be taken as absolute but as relative because every arena can present varying degrees of difficulty and risk for different issues. Thus, for example, a friendly arena is one perceived as having many more friends than enemies, benign rather than difficult issues, good rather than bad moments 191

Chapter 4

Getting a Grip on an EU Arena

and less unpleasant than the outside world. The ambition to win and not to lose in whatever arena requires, after its study, prudent management of its stakeholders, issues, time and boundaries, with intelligent tactics and, finally, fine-tuned fieldwork (see chapter 6). The better tactical approaches for each main type of arena situation can be derived from the logic of PAM and be summarized as follows (figure 4.4).

The Better Tactics of Arena Management If the Arena Is

Friendly

Unfriendly

Indeterminate

keep status quo

change situation

influence components

Then, Better Tactics Are: General Management

Triple P

Stakeholders

secure support

divide opponents

maybe argumentation

Management

take free ride

approach waverers

negotiate

Issues

keep issues high

compensate the loss

reframe the issues

Management

block others

reframe the issues

try a balloon

Time

speed up

delay

wait and see

keep restricted

expand the arena

wait and see

Management Boundaries Management

Figure 4.4

– In a mainly friendly arena (or section of it) the wind is turning into a tailwind or is already blowing supportively. The general strategy here is to keep the situation as it is and to harvest the fruits. The interest group can confine itself to actions that secure the situation. The friendly stakeholders having primary status it should bring together, bind to the overlapping interests and set to common action in order to optimize the M in MI2 [Knoke, 1996]. Out of ‘intelligent laziness’ it should stimulate that those with an even stronger interest in the outcome take the leadership of this network or platform that, from here on, acts as common interest group and thus brings 2E closer to all members by 192

Chapter 4

Getting a Grip on an EU Arena

lowering the costs and increasing the rewards of influence actions that may follow. The opposing coalition that always starts fragile too, can (thanks to the data basis) be undermined by pumping up some internal issues and by seducing some primary members to defect. To keep the tailwind blowing, the common platform should regularly (but without overkill) publish new evidence that supports its positions and/or blocks counter-evidence. As long as the wind blows nicely, time should be managed for speeding up the decision process. The arena boundaries should not be widened but kept restricted to the friendly stakeholders and issues that dominate already. In this situation it is better to lobby very silently, as sound tends to attract new stakeholders who may bring in new and maybe less friendly interests, which would delay the process. If, however, on the eve of a crucial moment in the decision process, even more support is observed among the mass public, publicity can be used to force a great victory. – In a mainly unfriendly arena (or section of it) the wind is unpleasantly turning into a headwind or already blowing cold. A clear case is the downhill position in an arena with the opponents uphill. Here the general strategy is to stimulate a change in the situation. The strong coalition of opponents can best be divided by stirring their internal issues as revealed by the data basis. This also shows which opponents are wavering and may be open to argumentation, persuasion or negotiation. They might leave their coalition and thus help to turn the wind. On the unfriendly issues one might try to find compenzation for those assessed as probably lost, as there is no good reason to lose with empty hands. Even a ‘dead horse’ may have market value, especially among amateurs who believe it is still alive. One might also start to lobby for an exemption, to avoid a loss in the short run. More advanced is the reframing of a difficult issue (salesman’s talk: chapters 1-IX and 6) so that it will attract more support. Time management must be directed at delaying the decision process, for example by urging an extended impact assessment and public consultation. Any delay may temper the cold wind and retard the expected loss. The widening of arena boundaries can also make the tide friendlier. New issues and new stakeholders delay the process and give more options for combining issues and creating new coalitions. Here it may be rational to make some sound that can attract new stakeholders and issues at low costs, provided this is done through indirect channels that hide the sender. Even an uphill fight is not a lost cause. One only has to play differently, just as trained sailors can proceed under a strong headwind. Easier than the army uphill, the troops downhill can move around, build up an army, attack the lifelines of the enemy and regain a better position. 193

Getting a Grip on an EU Arena

Chapter 4

– A mainly indeterminate arena (or section of it) has increasingly or primarily either side winds (mixed positions) or no wind at all (empty cells). The latter case usually occurs for a short while in the earliest phase of a fresh dossier, when interest groups are still blank on it. Its initiator, for example, a Commission DG or an interest group, may have a selfish interest in keeping the arena indeterminate and to monitor possible support. A proactive initiator may, however, have engineered by Triple P an unlevel playing-field before. In both cases the general strategy of other groups is to keep the indeterminate arena open (no Triple P, unless one can join the uphill bastion) and to get the wind blowing clear and friendly in the direction of their own interests. It is crucial here to proactively manage the stakeholders and issues that may soon become manifest. Most stakeholders in this arena still look for information and support. Many shop around and, for getting more in return, feign support to others and thus have to be checked for their real position. By better reasoned or reframed arguments and by early offered negotiations and deals, one might get some primary stakeholders onto one’s side. Preferably indirectly via befriended stakeholders, mass media or any other U-turn, one might launch a trial balloon or spread rumours for testing reactions. There is no good reason to speed up or to delay the process as this depends on the wind that is still to get up. The same is true for either the widening or the narrowing of the arena boundaries as there is still plenty of room for PA and lobby inside the indeterminate arena, unless the data basis indicates that its widening would bring in mainly friendly issues and stakeholders, or that its narrowing would repel mainly unfriendly ones. In general, the better tactics here are to wait and see alertly and to invest the saved energy in preparing the two scenarios of tailwind and headwind, one of which will arise sooner or later. This appetizer on the principles of managing an arena, with one dilemma for each of its four components, demonstrates the high added value of preparatory homework. The data basis, the store-file of PA-intelligence, allows to assess the arena situation and to apply thereafter the best possible or better tactics, as the latter always ‘depend on the situation’. (In chapter 6 we continue with many more dilemmas and include other choice criteria, such as for crisis management.) The better tactics are much different from the popular rules of thumb that advocated blocking, defensive, delaying and restrictive actions. In a friendly situation, the action should be directed at pushing one’s coalition and positions offensively, speeding up the process and keeping the arena restricted. The first three rules of thumb are better not applied here, but only in an unfriendly situation. The fourth rule on 194

Chapter 4

Getting a Grip on an EU Arena

boundary management should in this latter situation be applied reversely, for getting in more stakeholders. In an indeterminate situation it is frequently better to wait and see. The popularity of those rules of thumb among EU interest groups suggests that many of them consider both the arenas and the areas beyond the boundaries to be unfriendly and that they feel to have a low sense of efficacy. It may also be that they have neglected their data basis and, without reflection, are simply relying on popular but unjustified rules of thumb. The two explanations are not mutually exclusive. The more unprepared an interest group is, the more uncertain it may be and the more it may feel itself to be in an unfriendly situation. Referring to the criteria of easiness and low risk, the amateurish groups suffer, paradoxically, the greatest troubles and highest risks. Rational risk avoidance should be based on a serious risk assessment only.

2 Long-term EU Arena Management The long-term management of an EU arena has both an early start and a late finish. The same should be the case for the preparatory work. The earlier one starts with it, the sooner one can get useful insights for managing it intelligently and proactively. This sounds self-evident, but is often hard to realize in practise. There is usually an imbalance between (scarcely) available resources and (manifold) pressures from other arenas in the EU, with stakeholders and at the home front. Early preparation as method of fire prevention can contribute to its rebalancing, as it reduces at least the risk of unforced errors and so saves on resources. As essential tool to realize 2E it can even bring trophies that increase resources and decrease distorting pressures. Any arena has to be monitored continuously and this applies even after its formal ending, when the decision has been taken. In fact, the arena is never completely ‘over’: the losers may take revenge; a win may turn into a loss in disguise; unforeseen spillovers may open new arenas and the phases of implementation, inspection and review still have to come. The long-term approach to EU arena management gives three special advantages. 1 Planning and fine-tuning of dossier phases. In every phase there is usually a different demand for some PA activity, which can be planned. When, for example, an issue in Europe is suddenly irritating many interest groups, the EU officials develop a special need for information that helps to define the underlying problem. By early preparation, an interest group might be in a position to supply this information and thereby influence how an official problem is defined, which will bring the resulting solutions closer to its own interests. In the decision phase, 195

Getting a Grip on an EU Arena

Chapter 4

both the officials and the other stakeholders usually have an appetite for support rather than information. Again, early preparation and a good data basis helps an interest group to be prepared to deliver efficient and effective support. In its long-term management the group should consider an integrated process-oriented PA approach rather than a phased one and, following figure 1.2, it should think backwards from the (possible or desired) future situation into the present and act forwards from the situation ‘here and now’ into the future. An example is the Air-Transport Slots case above, with the national air carriers acting early at Council level backwards against the Commission proposals. 2 Use of general influence approaches. As outlined in chapter 3-II-1, and there applied to primarily the EU officials, the actors’ approach that is directly aimed at those who really make a decision, can be extended to every stakeholder, such as a EuroFed and a member inside. The sophisticated approaches of factors and vectors can create desired influence effects in more indirect and subtle ways. Their cultural, formal, operational and decisional variables can be equally relevant for any stakeholder, issue, phase or boundary. For example, the vector of issue reframing may turn an enemy into a friend, shorten a phase and move a boundary. The Triple P game, with its time-dependent prefabrication of procedures, positions and people, can be applied to broader arenas that include collective platforms and make the future arena friendly from the onset. These approaches require covert sticks and overt carrots and a lot of poker play in disguise, and therefore an early and continuing study, permanent prudence and proactive management. 3 Scenario management. This third advantage of an early PA start places the following question at the centre: ‘What can happen if X, Y and/or Z occur and what then is the better course of action?’, thereby stimulating mental PA mapping that is useful for coping with unforeseen futures that always may happen, particularly in crisis times [Bertrand and others, 1999; Ringland, 1998; Van der Heijden, 1996]. Shell company initiated scenarios as part of PA in the early 1970s, made profitable use of it during the 1973 oil crisis and employs a core staff of about 30 people on it in 2012. A scenario holds logical relationships between imaginary but possible causes or independent variables X, Y, Z on one side and possible outcomes or dependent variables on the other. They must meet the tests of relevance, consistency and possibility, but not of probability. It is like the chess player’s short-term thinking about what may happen if the opponent moves one or another piece, but now for the long term. The X, Y and Z can be anything that is manageable, for example conflict among officials, another procedure, a reframed problem or a WTO intervention. The intermediate variables remain stakeholders, issues, 196

Chapter 4

Getting a Grip on an EU Arena

time, and boundaries, and the matrix of figure 4.3 remains useful for making a data file for the imaginary situation. The outcomes are a few scenarios or ‘possible futures’. To each the core question of PA management is posed: If this happens, what then is the better tactics? More specific questions are useful too: What is the better tactics, if… (for example) this or that stakeholder disappears or changes his position?;… if a particular issue is traded-off or reframed?;… if a potential coalition achieves or loses cohesion?; … if the arena becomes widened or restricted?;… if the process is speeded up or delayed?;… if, regarding a stakeholder, a particular (cultural, formal, operational or decisional) factor is influenced or vector is created?;… if a Triple P game is played differently? The answers to them are useful in any arena situation but particularly in an indeterminate one with high stakes, like in crisis times. Scenarios give the interest group different windows into possible futures and tools for managing them.

X Extra: The GMO Food Arena Food is permanently at issue in the EU. A few issue dimensions out of many more are traditional food versus industrialized (processed), fat (causing obesity) versus light, fast versus slow (‘tasteful’), regionally protected versus open-market produce, subsidized versus non-subsidized, healthy versus harmful (diseases) and produce containing genetically modified organisms (GMOs) from biotechnology versus those without. Every dimension has its variants with different underlying interests. For example, GMO has variants called red (medical, health), white (industrial enzymes) and green (food), and the latter, on which we focus here, differs for transgenic (different genes) and cisgenic (same genes). In the field of food, ranging from research and production to trade and consumption, the EU always has been so full of irritating differences that most current food laws operative in the member-countries are EU laws. In this issue domain many Commission DGs are involved, such as on general matters COMP and MARKT, and on specific issues AGRI for subsidies, restitutions and quota, SANCO for food health, ENTR for industrial processing, ENVI for environment, RTD for research, TRADE for global trade and WTO affairs, REGIO for land use and DEVCO for food support to poor countries, to mention only these. The EP and the Council play their role on legislation and agenda-building. The Commission is the key player, as it drafts the legislation and manages the delegated and implementing decisions on food. The milestone is the adoption of the 2002 Food Safety Regulation (EC 178/2002) and the 2003 start of the European Food Safety Authority EFSA. The indicators of the 197

Getting a Grip on an EU Arena

Chapter 4

relevance of the EU food domain are that this holds relatively the largest proportion of EU spending, expert groups, comitology committees and interest groups. Now we shall focus on that part of the GMO arena that Prince Charles of the UK once called ‘Frankenstein food’. In February 1998, the Commission’s DG Environment, at that time also in charge of consumer protection, proposed to the EP and the Council to revise the old Directive 90/220/EC on the deliberate release into the environment of GMOs. Its objective was a stricter regulation, but not a full prohibition of the research, production and trade (both external and internal) of GMOs. In June 1999, the ministers of the Environment Council reached an informal, non-binding agreement on mainly procedure: any acceptance of GMOs for experimental or market releases must be based on both risk assessment and a decision by a competent authority and must meet special standards of risk management, monitoring and public information (such as labelling). For the protection of its traditional farming, France wanted a full moratorium on new admissions of products with GMOs (after 18 admissions so far). The Council agreed upon ‘no further admission, unless no risk’, with the UK, Ireland and Portugal abstaining. In October 1999 the EP, being less rabid than the Council and acting under codecision, adopted 39 amendments, all directed at regulation without full prohibition of GMOs; the Commission accepted almost all of them. In December 1999, the Council included four other principles, being precaution (no admission, if risk), traceability (regarding the origin of GMO products), authorization period (a maximum of ten years) and the need for global agreements (‘Biosafety Protocol’), and it formalized its previously informal position accordingly, now with France, Ireland and Italy abstaining. On the second reading, the EP and the Council kept disagreeing, after which by conciliation procedure a compromise included the principle of labelling (‘let the consumer decide’) and the lifting of the moratorium, and so the proposal became law in March 2001 (EC 2001/18). As usual, the text held many provisions for detailed delegated and implementing acts, such as on admissions, traceability and labelling. As the devil (or the angel) is usually in the specific details of acts rather than in the general text of laws, new conflicts lie in wait. Indeed, the next conflict came as early as May 2001, when DG ENVI (not in charge of consumer affairs anymore and headed by Margot Wallström) presented its drafts on traceability and labelling and its new policy value of environmental liability (not limited to GMOs), and proposed to set low levels of permitted GMO immediately. The new DG SANCO (in charge of consumer affairs and headed by David Byrne) claimed the competence on labelling and took the position that allowable levels should be set by comitology in due time. In the meantime, the Commission said it would consider ending the moratorium. 198

Chapter 4

Getting a Grip on an EU Arena

Between 1998 and 2001, the GMO dossier showed many cleavages inside and among the three institutions that primarily functioned as forums of conflict between numerous different stakeholders from inside and outside, and hardly as bodies to promote cohesion. The Commission was divided by its DGs, their Cabinets and even inside them. The first responsible DG ENVI was attacked by DG Trade (settling GMO issues with the US), DG Industry (economic growth) and DG RTD (new technology). DG AGRI was divided between traditional and modern farming and DG Consumer Affairs (the forerunner of DG SANCO) between consumer benefits (price, quality) and safety (health). In the EP the ‘green freaks’ controlled the dossier in the committee on the Environment and, except the Greens, all political groups were divided. In the Environment Council the ministers disagreed among each other and quarrelled about principles and procedures, their resort of Triple P. At home many ministers had to cope with cleavages among ministries, industrial groups and NGOs. Not surprisingly, many issues were not settled and, as part of the conciliated law, were delegated to the Commission that won this Triple P game. The politicized dossier collected a mass of interest groups from outside, each connected to different stakeholders inside the three institutions. To cut a long story short, these groups soon established two main coalitions: the antiGMO and the pro-GMO one. The anti-GMO movement was led by Greenpeace and included interest groups from health, religion, ethics, animal welfare and development aid. It got support from traditional farmers (averse to new technology) and retailers (fearing consumer protests). Nestled within DG Environment and the EP’s committee, the anti-GMO movement had enjoyed a tailwind before. The 1990 adoption of the old directive had turned out to be a hard pill to swallow for industry. It became a classic Greek drama, full of ideological conflicts and power-plays between different Commission DGs, EP committees and Councils, but gradually with the ‘green friends’ getting the driver’s seats [Gottweis, 1999; Patterson, 2000]. The 1996 decision to admit GMO maize proved to be a Pyrrhic victory for the traders [Bradley, 1998], with Italy, Austria and Luxembourg refusing to implement it. The 1998 Biotechnology Patent Directive (EC 1998/44) was narrowly adopted, after fierce opposition from inside the Commission (DG Environment), the EP (across parties), and the Council (from Denmark, Austria and the Netherlands). Opposing this directive, Greenpeace had set up the European Campaign on Biotechnology Patents (ECOBP) and soon it could use this EuroFed as leverage against the 1998 proposal on the release of GMOs. To strengthen the new tailwind, it framed GMO as high risk for both environment and health [Daviter, 2011], and mobilized public opinion at national and EU level. Mass media smelled a controversy and fanned the 199

Getting a Grip on an EU Arena

Chapter 4

fires of public fear over the potential dangers of ‘Frankenstein food’. The anti-GMO coalition got its edge when the dossier came under control of their friends in DG ENVI, the EP and the Council. It applied and enjoyed the Triple P game. Seeing that the arena was turning friendly, it tried to speed up the process and to keep the arena restricted to its issues and allies. The better tactics of arena management were applied rather professionally. The pro-GMO coalition lacked a cohesive platform but was led by Europabio, a EuroFed of companies interested in biotechnology, particularly the red variant for pharmaceuticals (like GlaxoSmithKline). Europabio was not against regulation of GMOs, but wanted it to be based on sound risk assessment that permits research, enables consumers to choose and keeps trade competitive. Major food companies (like Unilever) shared this approach and, as their EuroFed CIAA was divided on the matter, started to co-operate with Europabio. Soon they found that companies and patients in the health sector consider GMO risk to be an opportunity rather than a threat and gradually got an isolated position. Stakeholders in the food chain (retail, farmers) and in other trades using GMO (like chemicals, beer, meat and feed) hardly supported the food companies. Only a few DGs (Industry, R&D and Consumer Affairs) and some MEPs (Socialists and Christian Democrats) gave support. Most national governments took shelter from this issue area. Noisy and aggressive support from US seed producers like Monsanto and Cargill proved to be disastrous. Mass media fanned the flames. The headwind calmed a little bit in 2000, when the EP plenary rejected the most contentious amendments, the Commission presented its package deal including the lifting of the moratorium, and the Council agreed upon a delegation of powers to the Commission. Unilever considered this to be insufficient: in May 2000, it decided to stop producing and trading products containing GMOs in the EU. Three months later the Swiss giant Novartis decided to do the same. These food companies had become weary of their downhill position on an unlevel playing-field. However, they should not have been surprised. The previous EU decisions on the 1990 Directive, Transgenic Maize case and Patent Directive were early warnings of a mounting headwind that turned the Commission, preparing its 1998 proposal, into an open forum of conflicts. All this was widely known and should have warned the food companies to be better prepared for the headwind. Instead of the exit option, they could have applied the following better tactics for an unfriendly arena. – A good stakeholders study could have shown that the opposing coalition was far from unified. For example, the green movement was divided into a polemic EU faction and a moderate Third World one that considered GMOs as the poor man’s ‘cheap fertilizer’, better than 200

Chapter 4

Getting a Grip on an EU Arena

chemicals. The health groups, partly organized in the European Public Health Alliance (EPHA), was divided into a ‘caring’ and a ‘curing’ faction, the latter supporting the benefits provided by GMO to seriously ill people. Other opponents had taken ambivalent positions, such as the consumer organization BEUC that was more concerned about consumer safety and price-quality ratio than about GMO. Traditional farmers were divided between orthodox growers averse to new technology and small farmers fearing stronger competition. Multinational food retailers, such as Carrefour (France), Sainsbury (UK) and Delhaize (Belgium), were divided by market competition with both each other and alternative retailers like health food stores. Not least, the Commission, the EP and the Council were notoriously divided over the whole dossier. In short, the advocates of GMOs had many enemies and (potential) friends everywhere and could have strengthened their own position by undermining the fragility of the opponent coalition, seducing some opponents to indifference or even defection and creating a better collective platform of their own, without aggressive stakeholders like Monsanto. – Proactive issue analysis could have provided many clues for all that. At an early stage, the food companies could have pushed the controversial issues among their opponents, such as the poor man’s fertilizer, medical benefits, price-quality benefits for consumers, crop benefits for small farmers, and market benefits for health food retail. So they would have restricted their opponents’ room to build a broad coalition on the values of environment and safety only. Many issues they could have reframed as, for example, global nutrition, agricultural development, public health, consumer welfare, economic growth, employment and environmental savings. They also might have found better loss-compenzation than the lifting of a moratorium that had no legal basis at all, for example EU research subsidies for low-risk GMOs or a simplified admission procedure for new food products. Shortly after the GMO game Unilever pushed the latter successfully, not for its GMO interests but for its GMO-free new margarine that lowers cholesterol. In the new 2003 EFSA it even managed to secure the chair of the Management Board. – By better time analysis of this unfriendly arena the food companies could have consumed the rich menu of ways to delay the process, from actors to approach, factors to use and vectors to create. On the alleged GMO benefits for the Third World they could have urged DG DEV directly and, via developing countries (where many make investments), indirectly to start on these alleged benefits research and consultations that delay automatically. They could have pushed patient groups to 201

Getting a Grip on an EU Arena

Chapter 4

lobby for exemptions that retard the process. They also could have elicited new research findings to confuse some opponents, challenged the Treaty basis of the Revised Directive or evoked an early complaint from the WTO, not so much for their own merits but as means to win time. With anticipation, the companies could have created a more cohesive platform of their own in good time, thereby blocking companies like Monsanto to speed up the decision process (to get their GMO products admitted), to divide Europabio and to help the enemy. – By boundary analysis the companies could have created their last safe resort. Their arena they should have widened. By provoking new issues and stakeholders, they might have won at least more time and at most more benign ones, as mentioned before. They also could have reactivated the old issues of stagnating employment and global competitiveness that need innovative technology, as this might have pushed to their side the largely passive trade unions and socialist parties that silently fear a loss of employment to US producers. DG Social Affairs (now EMPL) and DG Industry might have joined their coalition actively. The companies could have pushed that the precaution principle (if risk, no admission), if adopted, should also apply, as a matter of legal fairness, to traditionally made products in different sectors, such as transport, chemicals, energy, equipment and agriculture. Interest groups in these sectors would certainly not have been pleased but some might have joined the fight against this principle that otherwise could spill over to them. The food companies could even have linked their interests to the Agenda 2000 (decided in March 1999) with employment and growth as central themes. The foregoing recommendations are hindsight wisdom, useful for learning and illustrating the room to manage an unfriendly arena. Since 2002 the arena underwent some changes. The EP kept its divided but moderate position. DG SANCO (David Byrne) took charge of the food domain, prioritized traceability and labelling and based its new proposals more on research. DG AGRI (Franz Fishler) wanted controlled experiments with GMO in agriculture rather than full prohibition. DG RTD and DG ENTR joined the troops on research and production. In order to open the EU market to US GMO crops, the US government increased its pressure on the EU via the WTO and directly on governments in the Council. Soon the Commission sent to the EP and the Council its mild proposals on GMO food and feed authorization (EC 2003/1829) and traceability and labelling (EC 2003/1830). True enough, its Standing Committee on Food Chain and Animal Health (SCoFCAH, comitology) rejected most Commission proposals to allow new GMO crops, so enabling the EP and the Council to 202

Chapter 4

Getting a Grip on an EU Arena

overrule the Commission formally but not really as, by deep internal divisions, they largely failed to create a majority in favour or against. According to the rules of comitology then the Commission is formally the winner, but usually prefers not to push the matter to this end. Since the early 2010s, the wind is blowing slightly friendlier, thanks to (among more) the EFSA’s scientific approach to GMOs, the decline of the ECOBP alliance, reframes like innovation, stronger positions of GMO protagonists inside the EU, more pressures by the US and the BRIC countries in the WTO, the 2007 issue of ‘food shortages’ that by GMO might be eased, milder public opinion and rising fear that, as GMO production is embraced elsewhere in the world, the EU will fall behind in this new technology and market [Burns, 2012; Heslop, 2005]. In 2010 DG ENVI lost its biotechnology unit to DG SANCO. In 2012 the Commission’s Chief Scientific Adviser Anne Glover declared that GMO is not a risky technology and its food not riskier than conventional food. For a while, Frankenstein seems to have been exorcised.

XI Arena Analysis: Necessary Means, No Sufficient Goal The short exercise on the GMO food arena is no magic show but an example of the utility of thoughtful arena analysis for managing an arena more successfully. It also shows that systematic descriptive and analytical homework produces a data basis for transforming information into intelligence that gives better chances of PA success. Professional interest groups consider this preparatory work as necessary in order to find solid answers to the question ‘whom to lobby in an arena on what?’ They can see the better tactics for window-in lobbying in both the short and the long term. Semiprofessionals limit the homework to studying a few stakeholders and issues and perhaps some moments, and take the arena boundaries as set. Amateurish groups that usually consider little more than a few EU officials, their own desires and a late decision moment, hardly have a chance of winning. Yet, like in any sport, preparation is only the means and not the objective or, in academic words, it is a necessary but not sufficient condition for success. Even after the best preparation, at least four factors may contribute to losing a game regularly, a set sometimes, and perhaps a match. 1 Uncertainty. Inherent to every complex and dynamic situation or process is a degree of uncertainty, caused inevitably by a less-thanperfect quantity and quality of information and thus a dependency on next-best solutions. All players decide to the best of their knowledge, but this level differs significantly between amateurs and professionals 203

Getting a Grip on an EU Arena

Chapter 4

and thus creates very different outcomes of 2E. Like a few balls in the tennis tie-break, the smallest margin of expertise can be decisive. The management of uncertainty is thus the challenge to organize the best possible knowledge. 2 Bad luck. Every player can have some bad luck, even if this is only the good luck of a competitor. It may come from special information or support. In the 1992 HDTV case, Commissioner Filippo Pandolfi, personally supporting this technology, was the embodiment of luck: good for the producing companies and bad for the opponents, and when he left in 1994 the luck reversed. However, good or bad luck is seldom a message from heaven or hell. Often it can be (or could have been) foreseen by a proactive management of MI2 for both the short and the long term. 3 Chance to win. The best prepared player on the downhill side of an highly unlevel playing-field may have much less chance of winning than the amateur player in an uphill bastion. In the 1998 GMO case, the food companies stood downhill and so had small chances of winning. Chance, however, is not a metaphysical category, as better preparatory homework can enhance it, even from a downhill position. 4 Home front. An otherwise professional interest group can lose a game and even its reputation, if its home organization is disorderly or in bad shape. When all preparatory homework for an arena has been perfectly carried out but is not taken at home as input for strategy and mandate for action, then it cannot create a winning performance. In general, when the home organization has no sufficient capacity for arena management, lacks a clear strategy, leaves its targets undefined or does not learn from its mistakes, whether forced or unforced, then it will bring even its best professional to despair instead of success. This is the theme of the next chapter.

204

Chapter 5 Managing the Home Front Getting and maintaining a strong position in a foreign country presupposes either skilled ingenuity or fortune, but the more one relies on the former rather than the latter, the better one succeeds, according to Machiavelli (Prince, chapter 6). Which skilled ingenuity is needed at home for success in the EU?

I Who Is Lobbying, Why, for What and with What Result? In this chapter, the interest group is the unit of analysis. It may be the MNC of Siemens, an NGO like Animal Welfare, the EuroFed Eurima (isolation products), the Warsaw city government, the Regional Affairs Ministry of Italy, the AmCham in Brussels, the Commission or any other public or private interest group inside or outside the EU. They all are formally established entities, but also have semi-formal and informal networks and groups that can each act as interest groups. Examples are a company’s division, a ministry’s bureau, a Commission DG, a DG’s unit, an EP inter-group, a faction inside an NGO or the junior staff of an MEP. Alike formal entities develop external networks and platforms, so too do semi-formal and informal groups by creating an ad hoc coalition, an interface between industrialists and officials, an Irishmen’s club, a meeting of Central European young officials or a regular dinner of Baltic ministers. Each can act as a sort of interest group. Thus, a formal entity has many parts inside and faces outside that make up its complex character, which in this chapter our focus from PA perspective. As sketched in figure 1.2 (left side) and summarized in figure 3.3, our leading question is ‘who is the character that is acting, why, for what and with what results?’ This relates to the entity’s home affairs, such as its internal organization, strategy development, setting of agendas and targets and its performances. However, what are its parts inside and faces outside to which the questions apply and what are the answers? The simple-minded interest group answers questions before they are posed. It takes its own group as autonomous entity, its motivation as selfevident and its objectives as clear enough. It divides the results into two categories: the losses that are blamed on others and the gains that come from its performance. In daily life it will continually blame others and whitewash itself. In contrast, the professional group knows that its inter205

Managing the Home Front

Chapter 5

nal organization is always an incomplete puzzle, its motivations uncertain and its objectives full of dilemmas. The causes of its losses and gains it attributes to its own behaviour and/or outside actors or factors. It aims to strengthen the causal relationship between its home organization and the results of its PA by continuously improving its home affairs. Many PA officials spend more than half their time and energy on improving this, the remainder spent on the EU and stakeholders. The most difficult arena they consider to be their own organization, called the ‘home front’, and requiring much ambition, study and prudency. Our approach is, once again, not normative but advisory. There is no good reason why an interest group or even a citizen may not be simpleminded or amateurish, as it may live in a situation that offers no alternative. If, however, it wants to create real chances to influence its challenging European environment, then it must europeanize and take the EU arenas as sources of inspiration for its action planning at home and, if necessary, for the early improvement of its home front. For this reason, chapter 4 on managing the EU arena preceded this one. No interest group can ever get a desired outcome from the EU by an introverted focus on its own inner world, as this would block its going window-out and -in. It has to adapt its internal organization, strategies and agendas permanently to external constraints and possibilities and install in its home organization the variability and flexibility that is required for scoring. The ambitious interest group is also eager to learn all the time from both its own experiences and the performances of others. These differences between amateurish and professional PA behaviour remain our interest here.

II Organizational Assets for Public Affairs Management From general readings [Harris and Fleisher, 2005; Schlozman and Tierney, 1986; Kobrin, 1982; Milbrath and Goel, 1977] and the aforementioned case studies of EU interest groups, one can distil four main assets or qualities that strongly differentiate between amateurish and professional entities: sufficient cohesion, useful knowledge, basic resources and skills, and a good image. Of course, they are distinguishable but not separate from each other. An ambitious interest group permanently improves them and does so by, again, optimization rather than maximization. An intelligent group also benefits from the assets of the partners on its shared-list.

206

Chapter 5

Managing the Home Front

1 Sufficient Cohesion In the previous chapter we assumed, for the sake of simplicity, that an interest group acting on the divided EU playing-field is internally cohesive. Cohesion we define here as the existence of both coherency of preferences and co-ordination of actions. The assumption of cohesion is usually false. Under normal circumstances every group is internally divided, with its members arguing on their chosen facts, values, instruments and solutions. In short, it has its internal arenas of stakeholders and issues that are usually manifest at its informal, semi-formal and formal layers. 1 The informal level holds the pluralism of the employees or members that have different values and interests, which usually increase with the group’s size. Within a multinational company, NGO, agency or EuroFed, people usually show their pluralism of, for example, nationality, career, managerial style, education, religion, ambition, language and expertise. In a Commission DG too the civil servants differ likewise. A regional or local interest group often has its differences inside and among both administrative and political units. Such pluralism can fragment the larger organization and transform parts of it into self-reliant groups that act externally on their own. 2 At the semi-formal level the variety of role expectations is often the main cleaving factor. Staff and line people tend to have different expectations regarding each other’s contributions, which may cause conflicts between, for example, the legal affairs staff and the public relations staff, or the people in one product line versus those in another. A company like Philips, a DG like External Trade, an NGO like Greenpeace or a EuroFed like ETUC is, at the semi-formal level, normally an amalgam of different sets of role expectations, based particularly on functional and territorial divisions. The amalgam can easily and centrifugally cause self-reliant behaviour by the constituent parts, including external networking. 3 Even at the formal level of an established organization, internal cohesion can normally never be assumed. The formal division of tasks and duties makes the organization already internally divided. The Commission’s DG AGRI and DG DEVCO have real conflict of interests regarding meat exports to poor countries like those in the African Sahel, as the one DG is in favour of (subsidized) exports by European farmers and the other against them in order to protect the Sahel farmers. In an NGO, the steering committee, the bureau and the plenary differ by formal positions that easily conflict. In most countries, the ministry of Transport is formally responsible for both railway and road transport and is thus internally divided about the values of public versus private transport. 207

Managing the Home Front

Chapter 5

All such internal divisions are so normal for a living organization that it is an abnormal or unnatural state of affairs for them to be absent. If there is full and spontaneous cohesion at all layers, then something must be seriously wrong inside due to, for example, a crisis emanating from outside, internal despotic rule or simply a malfunctioning of people and units neglecting their different values, roles and duties. The natural state of internal divisions can even be seen as richness. Its pluralism can make the organization better adapted to the pluralism of the European environment, its variety of role expectations can generate useful dialectics internally, and the different tasks can create strong stakeholders inside for every formalized value of the organization. Even their centrifugal forces are potentially healthy for external operations, as they may increase the organization’s capacity for variable and flexible responses to challenges, for cost-saving self-reliance by the constituent parts, and for getting prospective antennae and protective shields. The richness of internal divisions should neither be minimized nor maximized, but optimized. The economy or efficiency of influencing the EU starts at home. Minimizing the divisions by strict orders to obedience creates high risks of losing the rewards from internal divisions and of becoming vulnerable at the top. Only in abnormal, crisis-like situations, when all hands need to be on deck, may this be justified. Under normal circumstances such an enforcement of both coherence and co-ordination makes the organization introverted, takes up a lot of scarce resources, and puts a premium on obstruction. Maximizing the richness of diversity should be avoided too, as a deeply divided organization cannot act effectively outside. Competitors having done their preparatory homework well might fan the flames of the internally dividing issues and paralyze the organization, and EU officials cannot take seriously an interest group with cacophonous or even contradictory voices, such as those that came from the government of Malta regarding its membership request (in favour in 1993 and 1998, but against in 1996). Only ‘sufficient’ cohesion is needed, but this optimum is, inevitably, always a delicate balance between internal support and external effectiveness and it has to be found anew each round [Kobrin, 1982]. The big question, of course, is how to achieve sufficient cohesion. Here too, preparatory homework should drive the internal fieldwork. The internal arena can best be monitored in the same way as the external one (as described in chapter 4 and summarized in figure 4.3). For any EU dossier, the internal stakeholders can be listed vertically, for example those in staff and line offices, main executives, key experts, informal leaders and, if applicable, those in the country units. Their issues and preferences are put horizontally. This data basis provides cues for I2-based better internal tac208

Chapter 5

Managing the Home Front

tics. This homework can be done by one PA officer, but the better result comes from the duplica method described in chapter 4-VI. Many MNCs apply a variant of the Siemens method, by which the PA unit of this German company regularly questions key people in the product divisions and country units about their views on new EU dossiers and so collects information for managing the internal arena better. In government organizations, many people spontaneously show their differences in written or verbal form, which is less efficient, as positions on paper are hard to change and spoken ones are easily confused and confusing. This homework has to result in the building of sufficient cohesion. Its formation must balance the assessed internal and external arenas and is the starting point of strategy development (to be discussed below). There is, of course, a great deal of variation by type of organization (to which we will return in the extra section below). Domestic culture can still be a factor. If cohesion is constructed top-down, as often happens in German and French groups, then the cohesion is highly formal, requires many signatures and much time and tends to be weakly internalized [Nutt, 1999]. The French even follow a presidential style, as the top management tends to consult its key staff after it has made its decision rather than before. Dutch and Danish organizations, by contrast, often take a bottom-up approach, which includes staff and line units and even work councils. The more bottom-up the approach, the better the cohesion that can be anchored at the informal level, but the costs of internal compromise and time are the price to pay. In British organizations the board often gives the mandate for creating cohesion to mid-level managers. The factor of domestic culture is, however, weakening due to the rise of other factors, such as sectoral culture, the organization’s specific interests and, not least, its European environment. Whatever the style, internal PA is always required, both windowout for monitoring and window-in for getting commitment. The real test of sufficient cohesion lies in the outside world and is the capacity to deliver what has been promised to stakeholders and officials there. Sufficient cohesion is necessary but it is often not achieved. An organization can remain seriously divided on an EU dossier, even after a Teutonic top-down intervention. In the case of enduring dissent the organization can best remain passive on the dossier, in order to avoid any boomerang effects. For this reason, heterogeneous EuroFeds like BusinessEurope usually have to abstain from taking positions on specific dossiers on which their members have conflicting interests. If dissent develops into paralyzing cleavage, then the dramatic step of really splitting the organization should be considered. In the mid-1990s, the Philips Company became internally deeply divided on the EU dossier regarding the copyright levy on compact discs (CDs) that its music-producing unit Polygram wanted but its CD-rewriters 209

Managing the Home Front

Chapter 5

production unit opposed, and in 1998 it sold Polygram to the entertainment company Seagram, so resolving the internal cleavage. In an (always voluntary) EuroFed, dissatisfied members usually solve such a drama by starting an ad hoc coalition or new EuroFed. In 1990, most European car manufacturers left their EuroFed CCMC because the French Peugeot group PSA continued to block a common position regarding the EU dossier on car imports from Japan. They formed the new ACEA in 1991 (with, ironically, PSA joining in 1994) [McLaughlin, 1994].

2 Useful Knowledge A second quality of professional PAM is the possession of knowledge useful for getting the desired EU outcomes. The amateurish group is easily satisfied with its own impressions and perceptions of reality, or it follows the wisdom of its leader or its rules of thumb. Of course, the professional group is not an epistemic or academic faculty as it sees the acquisition of knowledge not as objective but merely as essential means for realizing its EU targets [Lindblom and Cohen, 1979]. As this knowledge is a core theme of this book, we can be short here. The most critical category of knowledge is self-knowledge, which requires a balanced insight into one’s desires and fears on the one side and one’s capacities on the other. The desires and fears must have a better basis than the random collection of daydreams and nightmares. In a self-conscious interest group, they are assessed after much window-out monitoring (chapter 4), followed by a critical match with internal preferences that must be in balance with the capacity for action. The group has to know the real state of its internal variety of informal values, semi-formal role expectations and formal tasks and duties, which all exist behind its official façade of unity. It must also be informed about the state of its resources and skills and about its image among stakeholders. The matter of self-knowledge is so essential for PA strategy that we return to it in the next section. The second category of required knowledge is arena knowledge, as discussed in the previous chapter. For every major phase in the life of the dossier at stake, the interest group must indentify at least the primary stakeholders (including officials) with their core issues and preferences. It must be familiar with the arena boundaries, because otherwise it cannot exploit new stakeholders and issues outside the arena. Every time it must assess the precise nature of the arena and how the wind is blowing there or not. With the help of a PA expert, it can collect all this ‘software’ knowledge from the sources mentioned in figure 4.2, and add to this the ‘hardware’ of the EU decision process, such as the relevant institutions and their components (chapter 2) and the manageable actors, factors and vectors 210

Chapter 5

Managing the Home Front

(chapter 3). This type of knowledge can most efficiently be gathered by shared-listing its relevant dossiers. In spite of all efforts to collect useful knowledge, even the most professional group has to act under limited knowledge, as concluded in chapter 4. The required information about one’s own organization is, like that about the external arena, never fully complete, valid and/or reliable. Uncertainty is everybody’s lot in life, but of the amateur more so than of the professional. The latter has the ambition and expertise to improve both the quantity and quality of information continuously, but is also aware of all the costs to get them closer to maximal. Therefore, it wants to optimize by doing its utmost to get the best possible knowledge for its advanced muddling through [Lindblom, 1959]. Conscious of the relativity of its desires, it is more able to reconsider its strategy and to identify a new option or opportunity. It knows that the competitors and officials have limited knowledge too and that often the marginal difference of knowledge is most crucial and the final test of optimization. Wanting to conceal this margin, paradoxically, the professional group tends to overexpose its uncertainty and the amateur one to camouflage it by a show of self-confidence.

3 Optimal Mix of Resources and Skills Resources are the tools that are useful for survival. They are not fixed assets in permanent possession, but have to be constantly acquired, maintained and renewed. A tool can be anything that contributes to effective and efficient participation in EU politics. Sufficient cohesion, useful knowledge and a good image, all taken here as indicators of professionalism, could be presented as important resources too and the same for the strong ambition to become influential or the nuisance value acquired during an influence process. It is more common to limit the resources to the entity’s capacity for action [Schlozman and Tierney, 1986, V], irrespective of action factors like desires, compulsions and invitations. The following capacities are prominently mentioned in the literature, including case studies and field research: expertise, networks, external positions and financial means. PA based on shared-listing can greatly enhance their quantity and quality. 1 Expertise comprises not only the useful arena knowledge mentioned above, but also the self-knowledge and the expertise of managing an arena. The latter has a technical and a political dimension. In the EU system of compromising decision-making, every big fight, for example regarding health conditions at the workplace, tends to be rapidly split up into many technical issues, such as regarding the noise or the radiation coming from a machine. An interest group must have the technical 211

Managing the Home Front

Chapter 5

expertise to deal with such issues within reach. The political or, more precisely, the PA expertise is required for coming to a final deal, which is primarily a matter of lobbying for support. Most interest groups have plenty of technical experts within reach, but fall short of PA experts. The Commission and the EP usually show a reversed imbalance, as they excel in political expertise but fall short of technical expertise, and hence are in a permanent state of having to insource the latter via expert groups and inter-groups or by inviting experts to come along. 2 Networks are in the pluralistic and fragmented EU area crucial for meeting the different stakeholders. Most important is the informal network, as this is the precondition for the development of a stable semi-formal one, such as an ad hoc coalition, which in its turn is a precondition for the formation of a formal entity such as a EuroFed. A formal platform without a stable semi-formal layer falls short of authority and can only survive artificially, for example thanks to a privileged legal position. Many a Council session is such an artificial platform providing a privileged position to the national government. A semi-formal network without an informal layer easily lacks stability, as is the fate of many a Commission’s expert group or committee [Van Schendelen, 1998]. The larger an interest group, the easier it can enter EU networks and get involved [Bernhagen and Mitchell, 2009]. 3 External positions are seen as crucial because they link an interest group to the networks of stakeholders and to the EU machinery. The position may be that of an ordinary member of a EuroFed or an EU expert group, who sits around the table, monitors the others, voices a few comments and uses the seat mainly as a means to improve its self-reliant plan of action. It may also be that of an informal leader, such as a big multinational in a fragmented EuroFed full of small and medium-sized enterprises (SMEs). A formally empowered position is not necessarily better than a merely informal one. It is not formal power but influence that matters, and the former, with its obligations of accountability, can even hinder the latter. Often one can better be the secretary running the apparatus invisibly than the formal chairman attracting all attention. The accumulation of positions is most useful. Some interest groups have managed to position the same person as a secretary of a EuroFed, an expert on a committee, a member of an inter-group and/or a representative in a working group. By contrast, civil servants of the Commission seldom hold functional positions on the outside. 4 Financial means are often said to be a very important resource as they enable the acquisition of more expertise, networks and positions and even may help to improve the organization’s cohesion, knowledge and 212

Chapter 5

Managing the Home Front

image, so substituting money in any other useful asset and creating better chances of success [Dür and Mateo, 2012; Klüver, 2012; Eising, 2009; Greer and others, 2008; Coen, 2007]. Others are sceptical about this popular claim that attributes most influence to wealthy interest groups [Beyers and Kerremans, 2007]. In any case, information about PA budgets at the EU level is incomplete and, if available, much varied. Almost half the EuroFeds of any kind have an annual budget of less than €100,000, enough for a small staff and office space, while those from business are estimated at a mean budget of €1.5 million [Eising, 2009]. NGOs range from multi-millions of euros for Care International to less than €10,000 for many from Central Europe [Greenwood, 2011; Carmin, 2010; Belou and others, 2003, 18]. Regional offices have an average annual budget of around €200,000 [Marks and others, 2002, 11]. A survey among 140 PA practitioners shows that in 2007 their budget increased by 10% on average [ComRes, 2008], but flattened in the recession years thereafter [ComRes, 2011]. At the expenditure side the definition of costs is nowhere specified and may vary from full to marginal and from gross to net costs. Some cost items, such as for expert-group work, are often reimbursed or sponsored by others, such as the Commission. Many groups have a joint office and staff, and divide the costs among them. Hence, the concept of wealth is not easy to measure. Shell looks rich and Greenpeace poor, while both are rich if one capitalizes all the volunteers of the latter [Maloney and Jordan, 1997]. In 1998 Greenpeace, shortly after its Brent Spar fight, was able to spend €70 million on mainly political action [Financial Times, 20 September 2000], more than the PA budget of any known EuroFed or MNC. A popular question is, what is the most important resource? Answering this question is not easy and, anyhow, the answer would be different for each situation and even each interest group. The main difficulty is that the four resources may be interdependent and substitutive indeed. For example, expertise can be derived from one’s network, be the result of acquiring a position or be bought from a consultancy. A strategic position inside the EU may make networks or finances less important. Usually, however, expertise, positions and networks are bound to a particular organization or person and not easily substitutive or for sale. Thus, one cannot conclude that one single resource is absolutely the most important. PA expertise may, however, be relatively the key resource, as it produces the best possible 2E of any PA campaign, and the lack of it the worst possible. An example of the latter are the northern pension funds that get their big income from obligatory savings and in 2007 were at first nonchalant and then upset about the Solvency II Directive (COM 2007/361 and 2008/119) 213

Managing the Home Front

Chapter 5

requiring higher solvability of private financial institutions. Their nonchalance came from the absence of any reference to pension funds and their upset from discovering at second reading that their funds were not excluded. In fact, they lacked the basic PA expertise to properly interpret EU texts and to come to I2. By intensely and loudly informing the Commission how different they are from all other financial institutions, they got the exclusion in the end. Soon the Commission published its specific green paper on Pensions (COM 2010/365) with similar proposals on solvability and new ones on many more sensitive issues. The funds had won a battle and got a war for the years to come, due to their lack of even a basic expertise, which made their other resources useless. Because of the latter, many professional EU interest groups shift their PA investments from resources to skills that can be defined as meta-capacities for developing the real capacity for action, thus as meta-resources. It is therefore not the possession of expertise, networks, positions and budgets which is seen as crucial, but that of the skills needed to acquire them at the right moment, in the right place and with the optimal mix. An example is Unilever’s Crisis Prevention and Response Toolkit, developed in the mid-2000s for improving its managers’ ability to anticipate and cope with any sort of downturn. It soon proved useful during the financial crises of autumn 2008. The logic behind this emphasis on skills is two-fold: skills give quality to the resources and are more cost-efficient. The skill of, for example, fine-tuned networking results in a better network without the high costs of maintaining an old network. By homework an interest group can update its optimal mix of resources and skills that is needed for the action. The final test takes place, again, in reality. Two different skills deserve special attention. One is the ability to remain cool and calm during the game, in short unagonized, in spite of all the tricks played by opponents. Whatever happens during it, an interest group must always remain concentrated on its target in the arena, like the football player challenged by various tricks (like spitting, scolding, tackling) must keep watching only the ball, as otherwise the ball is lost. This difficult skill of remaining unagonized gives highly competitive advantage. The second special skill regards the PA research and development (R&D) for useful new resources. Many resources are so general and widespread that they hardly make a difference. The art is to develop new ones that surprise competitors and attract EU officials, thus truly making a difference. In theory anything, even making a virtue out of necessity, can be used as a new tool. Recent real-life examples are networking through electronic media (like blogs, YouTube and Facebook), ‘inspiring’ (influencing) the firms that work for the Commission, multi-level lobbying via the US or the WHO and establishing a fresh BONGO or GONGO. From its side, the 214

Chapter 5

Managing the Home Front

Commission has developed new Triple P tools, such as the online consultation and the impact assessments. Of any new tool the effectiveness has to be measured and awaited, of course. New tools may be found by luck or imitation, but when discovered by R&D the interest group can be the first and take the lead. They are ultimately the product of PA expertise.

4 Good Image The fourth asset of a qualified interest group is a good image among officials and stakeholders. They feel more easily attracted to groups with a good rather than bad image. The notion of image has two main dimensions, the first being its general reputation, which stands for both importance and trustworthiness. By being seen as important, the group can avoid being overlooked and can have an anticipated influence on the others, even if it remains passive [Dahl, 1991]. There are, of course, different forms of importance. It may be the power of a veto position (such as at Council level) or a strong nuisance value (a blocking minority in a EuroFed), but also high sympathy among the wider public as enjoyed by green NGOs in the 1990s. In between power and sympathy stand variants like authority, prestige and respect, as attributed to the European Round Table of Industrialists (ERT). The image of trustworthiness is the necessary addendum to importance. The interest group must be seen as one that can be trusted in making a deal, as it has a reputation for credibility in terms of what it says, reliability in terms of its promises and predictability in terms of its subsequent actions. The best trust is based on a common interest, but cultural traits can solidify this [Grant, 2003]. Without trustworthiness, even the most powerful or sympathetic group will have low appeal, as it is seen as a monster or comedian to be distrusted. After its 1995 campaign against the Brent Spar, Greenpeace was distrusted because it had provided false information. In 2008, German car manufacturers earned the Worst Lobby Prize, awarded by NGOs like Spinwatch, for their ‘biased information’ on CO2 emissions. However, trustworthiness without importance is even worse, as it makes the interest group a helpless lamb for the wolf. The professional interest group always cares about its general reputation. It wants to be perceived as important and trustworthy, as the resulting goodwill decreases its costs of PA and increases its effectiveness. The more anticipated influence it enjoys, the less it has to invest in creating impacts. The more respect or sympathy it enjoys, the easier it can obtain a place around a table. The more it is seen as trustworthy, the easier it can come to a deal. A good general reputation is no gift from heaven but the product of careful reputation management based on preparatory homework. Solely launching solemn statements, subsidizing social needs or 215

Managing the Home Front

Chapter 5

preaching social responsibility is insufficient [Mahon and Wartick, 2012]. They need evidences of political and civil engagement and, if these are absent, they can return as boomerangs, which is the risk run by every GONGO or BONGO that covers a selfish interest of government or business. A reputation for corruption or other dirty lobbying practise [Kowalczyk-Hoyer, 2012] is almost a death sentence at the EU level. By studying and measuring reputations and their shifts [Mahon and Mitnick, 2010; Fombrun and Van Riel, 2007], one may find room for looking more important and trustworthy than others. Competitors doing their homework well may unmask such impression management as hypocrisy, the old Greek word for theatre. The second dimension of a good image is the specific supply side. The amateurish interest group easily mistakes its demand side, laid down in a position paper that advocates its preferences, as an attractive supply to stakeholders. The professional group, by contrast, tries to meet the demand side of the primary stakeholders and to supply what makes them feel happy and willing to make a deal. Most stakeholders want to receive support, as they feel weak on the EU playing-field. EU officials have an appetite for detailed and reliable information regarding facts and trends in reality, and private interest groups, often having this information edge over governments, can supply this. Commission officials also want to get different interest transparently aggregated by the way of a EuroFed or cross-sectoral coalition. This saves them from doing the work of aggregation themselves, while the transparency unveils both the views behind and clues to divide and rule, if needed. In return, these officials have to supply items such as a nicer draft proposal, financial favours, better procedure, a changed deadline or a position inside. High-ranking officials and politicians have an insatiable appetite for good publicity through the mass media, in return for policy information and leniency. All such items may help but the best items for supply are attuned to the issues and the stakeholders in the relevant arena at the right moment. They differ for the friendly, unfriendly and indeterminate situations. The clues come from the preparatory work.

III Strategy: Why PA Management? The ultimate objective of every organization is to achieve greater happiness by scoring on its core targets in the outside world. In companies this may be in terms of the improvement of the balance sheet, turnover or profit; in NGOs the acquisition of members, budget or respect; and in governments the increase of authority, legitimacy or power. So they might 216

Chapter 5

Managing the Home Front

keep their licence to operate in the future. In addition to this general strategy, the PA strategy must add value by managing successfully external events, circumstances and trends that pose challenges or risks to the sense of happiness. Perceived problems and expected threats may be called negative challenges, while perceived blessings and expected opportunities can be seen as positive ones. Put simply, challenges or risks contain both nightmares and daydreams, and as such they form two sides of the same coin. In ancient Greece, this coin was called a ‘problem’, literally a rock that was thrown into the sea by playful gods and that could endanger ships on one side and provide shelter on the other. In Chinese, the word for ‘risk’ means both threat and opportunity. Today, a problem or risk no longer refers to the two-sided coin of danger and opportunity, but to only the dangerous side. The dual objective of PAM in EU is still inspired by the ancient Greek idea: to solve the negative challenges and to save the positive ones, insofar as they are related by cause or consequence to EU processes [McKellar, 2010; Kobrin, 1982]. The assumption is that this can be done. However, what, precisely, is a problem and what is a blessing, and what is a threat and what is an opportunity? The amateurish group will find the question superfluous, because it considers the answer to be self-evident. The professional knows that the answer can only come from the confrontation of a perceived fact (something that ‘is or is not’) with a chosen norm or value behind the norm (something that ‘ought to be or not to be’). Logically, the two are independent from each other and are only connected by the formula of ‘if X and if Y, then Z’: ‘If this is the fact and if that is our norm, then we conclude that this is the challenge’. Every problem is hence, indeed, ‘so-called’. It also implies that good (or bad) values or morals do not necessarily cause good (or bad) consequential facts. Good intentions can cause disasters and bad ones blessings. In any case, the objective identification of facts and norms is difficult, as they are embedded in one’s Pavlovian body full of strong beliefs. The only certainty is that, in the last resort, a fact has an absolute character: it either exists or does not, it is present or absent. An ultimate norm or value, such as better profit (companies), membership (NGOs) or authority (governments) may look absolute at abstract level, but once it is made concrete it becomes, in contrast to a fact, in the last resort a relative matter, as it is a subjective choice or value judgement. To quote Shakespeare: ‘There is nothing either good or bad, but thinking makes it so’ (Hamlet, Act 2, Scene 2). Thus, every interest group, including one’s strongest enemy, is free and right in its final choice of norms and, if derived logically and consistently, its targets. In coming to its choice, however, it can make many mistakes in the identification of the real facts and its best desired norms and so it is, then, the quality of their confrontation that delivers the challenge. This 217

Chapter 5

Managing the Home Front

The Challenges of Public Affairs Management I The Nature of a Challenge Facts Present Desire

Norms Fear

Blessing Opportunity Daydream Problem Threat Nightmare

Absent

A

B

C

D

Problem Threat Nightmare Blessing Opportunity Daydream

II The Strategic Management of a Challenge Facts Present

Absent

Desire A

B

C

D

Norms Fear

Solving a Problem, Threat, Nightmare Saving a Blessing, Opportunity, Daydream Figure 5.1

work of defining the challenge is so strategic that every interest group should collect critical advice from insiders and outsiders but finally keep it in its own hands. This basic logic is summarized in figure 5.1, presenting, in the upper part, the four situations that are logically possible. Two situations represent a problem or a threat: in one the perceived or expected fact is not according to the desire (cell B), and in the other that fact is feared (cell C). Here the greater happiness must come from solving the problem. The two other situations amount to a blessing or an opportunity: in one the perceived or expected fact is according to the desire (cell A) and in the other the fact gives no grounds for fear (cell D). In half the sit218

Chapter 5

Managing the Home Front

uations of life one can feel happy already and even greater happiness can come from saving the blessing. Once the challenge has been defined, the solutions for creating (more) happiness can be derived, as is clarified in the lower part of figure 5.1. Because the number of possible solutions is logically twice the number of problematic situations and that of possible savings twice the number of blessed situations, the intelligent problemsolver or lobbyist can only be an optimist who never complains about a situation but develops one of the possible solutions or savings. This is the field of the strategic management of challenges. To solve a problem, one might change either the fact or the norm. The first solution is targeted at changing the factual circumstances or developments. The way to do this is to turn cell B into A and/or cell C into D. This requires an enormous amount of work and a lot of physical energy for a long period of time (and even then the outcome will always be uncertain). The second solution is to change one’s chosen norm and value behind and to opt for another one. Then one turns cell C into A and/or B into D. This is a matter of thinking, which might be done quickly and has the greatest effectiveness. It only takes mental and not physical energy to reassess a fact feared as yet positive for tomorrow, or an unfulfilled desire as yet a blessing. Marcus Aurelius, the Roman emperor who observed the fall of his empire around 180 AD, achieved this change of mind in his Meditations, and finally did not call it a problem. In 2011 the French MNC in waste management, Sita (part of Suez Group), found, at second thought, that the overcapacity of incinerators in Europe was not a problem that should be addressed by an EU law expelling ‘dirty’ incinerators, after it realized that most of its own incinerators in France would come on the list for closure. Paradoxically, most interest groups (and people) are, different from Aurelius or Sita, more doers than thinkers. They prefer to spend years of work on changing facts than a short time on reconsidering norms. Some years later they often regret an EU law they lobbied for or praise one they opposed in the past. By neglecting half the number of possible solutions to a problem or a threat, they are not professionals. Saving a blessing brings even more enduring happiness to the intelligent interest group. In half the situations (cells A and D) it has every reason to feel happy now, although this is usually at risk in the future. By checking whether one still feels happy with such a situation and by preserving its facts, it can enjoy longer-lasting happiness. Paradoxically, most interest groups hardly count their blessings and opportunities. Consequently, they forget to safeguard the benign facts and/or to validate their norms. By taking their current happiness for granted they put it at risk. Only the most professional groups invest energy in saving a blessing or opportunity. As they never complain about a problem but create a solu219

Managing the Home Front

Chapter 5

tion, they are never jubilant about a blessing but save it. They know the fragility of earlier happiness. However, most organizations spend more energy on solving a negative challenge than on saving a positive one, let alone on safeguarding an old blessing. While trying to solve a problem, they may lose a blessing; and while trying to prevent a threat, they may miss an opportunity. Strategic management targeted at getting greater happiness by solving problems and saving blessings requires a great deal of skilled ingenuity. As the identification of real facts and desired norms is always difficult at home and contested in reality, so is the choice of one or more solutions always at issue, first of all on the home front. By their incapacity to interpret legislative texts, the northern pension funds (section II-3 above) could not assess the real facts and redefine their wishful norms and acted without proper strategy. Many facts are, besides, not easily changeable or manipulable (manageable), at least not within the desired time period. For example, the variety of Europe with all its different interests will remain a fact of life for a long period of time. Changing one’s norm is then the more rational choice, but it has its limits as well. A company with assets in a crisis sector like textile manufacturing or shipbuilding cannot easily decide to move to another sector, although the Finnish company Nokia moved from wood pulp to telecom. An NGO on wildlife cannot easily shift to, for example, children’s care, but such a shift sometimes happens too. Local governments cannot easily move their city or people, but relocations sometimes occur. Every interest group must respect dominant social norms too, like those on cartels, pollution and crime. When both the facts and the norms cannot be changed, as is possible in theory but seldom happens in practise, then there is no solution to the problem and, if there is really no solution, then neither is there a problem. The best attitude then is to be as relaxed as the sovereign Aurelius. PAM in EU regards, in essence, the achievement of greater happiness in the EU, even if the happiness is only greater than that of an opponent or competitor. In the latter case one has to move the opponent from his or her cells of happiness (A or D) to one of problems (B or C) by changing either the factual or the normative conditions that apply. Whether hedonistic or warlike, strategic PA reasoning is generally applicable to any field of human concern [Rochefort and Cobb, 1994] and particularly to the EU with its ever changing facts and norms. Interestingly, both the EU institutions and the professional PA groups usually try to come to the best possible strategic decisions by applying ‘the science of muddling through’ [Lindblom, 1959]. This best possible (optimal) method is superior to the alternatives of either arbitrary trial and error (minimal) or the utopian search after complete knowledge (maximal). This parallel between the EU 220

Chapter 5

Managing the Home Front

and PA should invite interest groups to review regularly their current EU strategy and to put energy in critical strategy development [Hunger and Wheelen, 1998]. The amateurish group does this at random or in a shoddy way. The professional group has within reach a PA unit for doing it systematically. Its strategy development goes as follows.

IV Strategy Development: Long-List, Shared-List and Short-List Following figure 1.2, strategy development takes three main steps. It starts with the creation of awareness about both oneself and the wider environment, which is done by going window-out in both the inside world of one’s organization and the outside one of officials and other stakeholders, in order to monitor and assess the facts and values that in both worlds are observed as potentially relevant. To put it simply: the four cells of the upper part of figure 5.1 are filled with observed daydreams and nightmares, altogether constituting the long-list of challenges that are considered relevant now and in the future. The information comes from the sources and techniques mentioned in figure 4.2. It easily amounts to about one hundred relevant challenges or dossiers, which is only about 4% of the new binding decisions made by the EU every year, excluding anything else of relevance that may happen. This long-listing requires much extroverted window-out lobbying for information or, better phrased, intelligence, as described in chapter 4. It may look tantalizing, but helps to select one’s best possible PA agenda for the EU. The longer the long-list, the lower is the chance to miss anything of relevance. The PA unit (or person) drafts the long-list with the help of the staff and line at home, for example by the Siemens method mentioned above. Critical is the creation of self-knowledge and the assessment of what should be desired or feared. On the latter, every interest group is, due to the absence of an absolute yardstick for determining the norms, fully free to choose, except for both the logical deduction from its ultimate value and the testing of the consistency with other norms. The difficulty of this is illustrated by many a strategy for tomorrow that is based on the assessment of today’s facts by yesterday’s norms. As a second step the quick scan (figure 4.3) is given to the challenges of the long-list in order to bring more nuanced insights. For every selected challenge or dossier the main stakeholders and their issues and preferences are roughly identified by the PA unit, again with the help of the staff and line. The techniques of duplica and replica for making the observations more reliable and complete (chapter 4-VI) can be used here. The 221

Managing the Home Front

Chapter 5

quick scan is not a goal in itself but a necessary means to enable selectivity, as no interest group can play with a fair chance of success single-handedly (no M) on more than a very few EU chessboards simultaneously. MNCs like Thomson (France), Novartis (Switzerland) and BAT (UK) make such scans for obtaining more selectivity. An MNO like Greenpeace, capable of spending significant resources on PA, might play on a few chessboards more but has to be selective too, as the total sum of desires and fears always outnumbers its resources. Many interest groups find it difficult to be selective, as they (incorrectly) believe that only they themselves can realize their long-list full of relevance. Amateurish groups that stick to their long-list, usually collected at random, will gather another long-list of disappointments. The third step regards this selection that is conducted by breaking down the long-list into a shared-list and a short-list, the former being the list of dossiers (like the issues of a single dossier, chapter 4) that, according to the quick scan, can be lobbied through a collective platform and the latter that has to be lobbied single-handedly as no partners were found. The semi-amateurish interest group is on the one hand aware of the need to downsize the long-list, but on the other hesitant to entrust its interests to collective platforms, and hence it sticks to a very long list with only grades of relevance. The professional group now sets the trend of giving priority to the shared-list, as a short-list dossier usually carries high costs (no burden-sharing) and low chances of success (one stands alone), with a reversed balance for the shared-list. It wants to get the shortest possible short-list and the longest possible shared-list and it manages these shared dossiers through tens of collective platforms, each often covering more than one shared-list dossier. On these dossiers it prepays the price of compromising with partners, which in the EU sooner or later anyhow must be paid and tends to increase with time. By their early transport to a collective platform, being established or newly constructed, it prefers the high chance of many small wins on its shared-list rather than the low chance of a single big win on its short-list. By a proactive use of a platform it soon expands the M of the formula 2E = MI2. So it practises intelligent laziness, i.e. it abstains from pushing what others having more or less the same interest (identified by the quick scan) also want to push. It invests energy primarily in that early dossier transport and in the remote control of what happens thereafter. Its short-list is the remainder of dossiers that can not easily be shared with others but are relevant, such as in cases of infringement, a unique interest or mere survival. On second thought, however, such dossiers can often be shared with others who are in a similar position and have an interest in sharing, just as Philips earns a fortune by sharing its unique patents. 222

Chapter 5

Managing the Home Front

The intelligent player selects from its long-list the shared-list and shortlist in a somewhat paradoxical way. Unlike the more-or-less amateurish player, it does not select from the long-list the dossiers that are most qualified for the short-list, but those that are least and thus go to the sharedlist or are left to the winds. The professional player prefers the logics of 2E that drive to shared-listing. Thus it applies to the long-list not positive but negative selection and it moves to the shared-list all dossiers that score poor (low 2E) for a single-handed short-list approach. The remainder is this short-list or is left to the winds. As a matter of interest, this procedure can also be applied to the issues of a single dossier as revealed by the quick scan (chapter 4), so providing cues for 2E at the issue level. Standard criteria of negative selection include the following ten (figure 5.2).

Downsizing the Long-list • Monitor and Select Relevant Dossiers = Long-list • Apply Negative Selection for Shared-list • Remainder, If Really Short, Is Short-list If Dossier Is Assessed as:

Then It Is Better to:

1.

‘Not so Relevant’

Outsource to Partners

2.

Damaging the Home Front

Keep Hands off

3.

Poor ‘Cost/Benefit’-Ratio

Share with Others

4.

Hardly Urgent now

Outsource it Temporarily

5.

Good Non-PA Alternative

Consider the Alternative

6.

‘Winds Behind’

Take the Free Ride

7.

Headwind-situation

Seek a Coalition

8.

Out of Reach

Wait and See with Partners

9.

A Lottery

Wait and See with Partners

10. ‘No Point to Choose’

Leave it to Others

Figure 5.2

223

Managing the Home Front

Chapter 5

1 Low relevance. Not all challenges have equal relevance. Those having relatively low relevance can rationally be left to befriended stakeholders. For example, the 2006 proposals to open the market of postal services by 2009 (COM 2006/594) affects every post-sending organization but is crucial for only a few, such as the state-run post monopolies, private couriers and mail-order firms. The others can better take side with one of them. 2 Poor internal situation. If the interest group is internally poorly equipped for external action (section II above), it is better for it to leave the dossier to others or to the winds, as the alternative is to suffer a loss in the best case and a boomerang in the worst one. Confederations like BusinessEurope and ETUC are often internally divided and wisely do little more than handle internal disputes and launch abstract papers that may encourage others. 3 Low cost-benefit expectations. A great benefit from action may be expected, but the costs of acquisition can make it unattractive, just as the fine cherry at the top of the tree, which would take too much energy or be too risky to pluck, is left to the birds. By shared-listing the costs can go down sharply. The benefits have to be shared too, but the new balance sheet is better. 4 Low urgency. The interest group may conclude that a dossier like the white paper on Transport in the year 2050 (COM 2011/144) has plenty of time to go and that it should not much invest in it. To avoid the risk that competitors will act sooner, it can better move the dossier to its shared-list, thereby sharing the costs of getting prepared for the right moment. 5 Good alternatives to PA in the EU. Many an EU challenge can be solved or saved through private markets, domestic politics or orthodox influence techniques. Meat importers can shift the costs of veterinary inspection, as obliged by EU law, to the consumers. By making a scandal of the sale of industrial processed (so called) ‘plop-poultry’, NGOs on Animal Welfare got it banned in major retail chains in 2012. Ministries often shift the costs of adaptation to EU decisions to their decentralized governments. The dossiers remain relevant, but not for PA at the EU level. 6 Free ride. If the quick scan reveals that a dossier holds a friendly arena, the interest group can take a free ride with other groups with a strong propensity and capacity for action. Small-sized companies, NGOs and regional governments often take a free ride or back-seat with bigger or stronger groups. On air-transport liberalization, small air carriers like Virgin now enjoy, unlike in the early 1990s, an anticipated influence and can take an almost free ‘flight’ with the EP and Commission. 224

Chapter 5

Managing the Home Front

7 Low chance. Conversely to the former case, the quick scan now reveals an extremely unfriendly arena with few supporters, mainly unfriendly issues and many headwinds. Attempting to influence it single-handedly would likely not be successful. The more rational path is to gather a few friends to launche a joint uphill fight or a loss-compenzation. If, after all, there really is no solution, then there really is no problem (being, by definition, something that is manageable) either. 8 Out of reach. Some dossiers like the 2012 draft treaties on TSCG and ESM may be most relevant but practically out of reach for influence action as they are settled in closed meetings inside the ECB, the Court, the College or among a few EU Council ministers. The time spent on ‘waiting and seeing’ can be better used to prepare a shared-list of actions to take once these dossiers come within reach, for example in the implementation phase. 9 A lottery. A few dossiers may remain indeterminate, even after all the better tactics for the situation have been applied. The dossier on lottery and gambling itself is an example. In the 2000s, when the internet lottery and gambling industries boomed, it was at a standstill due to stalemates inside the Commission, the Council and the EP. Here too, the better decision would be to wait and see alertly by shared-list approach. 10 No point in choosing. Even if the interest group sees a challenge as relevant and feels it has a good chance of influencing the outcome, it may believe that the difference is hardly relevant, as it is like being bitten by the dog instead of the cat (a negative challenge) or pampered by the mistress rather than the wife (a positive challenge). Then it is more efficient to stick to the cat or the wife. The German government that claimed at the Court that the Tobacco Ban Directive (EC 2001/37) had the wrong Treaty basis and won the ruling that this had to be changed, which legal change did not cause a substantial change, could have remained passive.

V Setting the Targets and the Agendas Downsizing the long-list to a shared-list and a short-list can, as shown above, be organized as a rational method and even be computerized. By doing a quick scan, one can get a score based on the mentioned criteria for each dossier or challenge on the long-list, resulting in an informative matrix. If one considers the criteria to be of unequal importance, then one can add some weight to each of them at will. If, for example, an interest group feels rich, it can freely delete the cost-benefit criterion, just as it can give this all the weight when feeling poor. If it wants to act anyway, such 225

Managing the Home Front

Chapter 5

as for the purposes of publicity, then it can skip the free-ride criterion. If it feels that the EU may bring it into a crisis situation of mere survival, then it should give full weight to only the criterion of relevance. Under normal circumstances, however, it is better to consider not just one but many criteria. It may even be rational to give them all the same weight. The more a dossier meets the negative criteria, the less reason there is to attempt to single-handedly influence it and the more to put it on the shared-list. Thanks to the burden of proof by negative selection, most dossiers appear to be fit for the shared-list approach. It may happen, however, that a typical shared-list dossier is still put on the short-list for nonPA reasons. In governmental groups, high-politicians often want to have something of personal relevance on the short-list, even if this is inefficient or ineffective at EU level and only serves to bolster their image at home. Many a rotating Council Presidency puts on its ‘Agenda for Europe’ some items for only domestic consumption and election. For both lists of dossiers the setting of targets has to follow now, at first at the home front. In its ultimate search for ‘greatest happiness’, the interest group can start by choosing a dream target for each dossier, even if this is unrealistic, as this gives direction to the next-best choices soon to be made. In the reality of the EU, it can only lobby for ‘greatest possible happiness’, and so it has to define its next-best targets, which are acceptable to both itself and sufficient stakeholders and officials who want to have ‘something in it for them’. In the phase of window-in negotiations, it can best start with broad next-best targets that give room for trade-offs and fine-tuning during the process. Part of it should be target-setting for package-dealing by which it might compensate some possible loss. Finally, it must define targets at its bottom-line, which is equal to the current situation that might be preserved or won via the Court. Below this line it should not negotiate, except for loss-compenzation. Sooner or later it has to sharpen the targets by making them specific and precise, for example a provision in a Commission proposal, another Treaty basis, change of an annex, amount of subsidy under an objective, position for playing Triple P or another precisely defined item. Without precise targets it can never really win (or lose) and can only shop around for ‘souvenirs from Brussels’, as is the fate of amateurish groups that mistake their list of activities to be done for a target, instead of a means. With precise targets a group can direct its intelligence lobby ‘backwards’ and its support lobby ‘forwards’ (figure 1.2), thereby achieving better results at the lower price (2E). The quality of target-setting can be enhanced by the tool of scenarios, which are prompted by the question, ‘What might happen if XYZ occur and what would then be the better tactics?’ (chapter 4-IX-2). By conceiving a next-best scenario for each dossier, as well as a loss-compenza226

Chapter 5

Managing the Home Front

tion scenario and a bottom-line scenario, the interest group can anticipate possible situations more efficiently and effectively during the process. After the target-setting, the dossiers of the short-list and the shared-list should be managed as two different agendas for action. The short-list agenda is a short one with a few heavy matches, usually carrying high costs and offering low chances of success. On each dossier on it, the serious interest group has to create MI2 single-handedly and to spend many of its resources and skills on all preparatory work and fieldwork, this being a tremendous amount of work; the bad alternative is not to do it in a serious but rather in an amateurish way. The shared-list agenda requires a similar amount of work, which now can be divided among the partners or be done by the common platform’s PA desk. This efficiency also enhances the effectiveness of the PA campaign, as the needed critical mass is included. On this shared-list agenda, every group has, however, still some work to do. In the beginnings it has to join a suitable platform or, if not existing, to construct it; for the former it has to apply for membership and for the latter to get a snowball rolling. Usually an existing EuroFed fits more or less, as most of them are sectorally organized like most interest groups, can cover many more shared-list dossiers or, if necessary, can be used as hub for making an ad hoc coalition. The next work is to get one’s dossiers included in the common agenda and one’s targets in the common positions of its platform. This requires negotiations that anyhow have to be entered sooner or later, but now produce earlier savings and influence. In the end-phase it has to do the light work of reaping its fruits. In between the beginnings and the end it can enjoy the efficiency of shared-listing by restricting itself mainly to activities requiring remote attention and control, such as the following: – Monitoring. The shared-list dossiers need a regular monitoring of their progress and stability, as some facts or norms may change in the arena, on the platform or at the home front. – Correcting. Under pressure from their home front, opponents, officials or outsiders, partners on a common platform may develop different interests as time goes on. Corrective activities may then be necessary. – Repairing causes. If a group regrets a dossier had to be put to the sharedlist or left to the winds, for example due to the poor internal state (criterion 2), then it should address this cause in the meantime. – Handling consequences. A case like ‘low urgency’ (criterion 4) should be given some preparatory groundwork and possibly a safeguarding free ride (criterion 6), as otherwise a big problem may arise in the future. – Exploiting nuisance value. If the interest group has put a dossier on its shared-list for the sake of nuisance value or loss-compenzation, then it must be active on cashing it. 227

Managing the Home Front

Chapter 5

– Testing consistency. At various moments it is useful to check whether the various dossiers on the shared-list are still consistent with each other and not (unless so desired) conflicting or maybe overlapping. This logic of strategy development, resulting in two PA agendas, of course together forming a consistent one, may be seen as creating an unpleasant amount of studious work to be carried out. The group that has the ambition to win and not lose its interests cannot, however, escape it and will be rewarded afterwards. Only an amateurish group considers long-listing, quick scans and shared-listing to be empty catchwords and lobbies blindly, at the mercy of the waves. Its PA agenda is a wishful daydream or ‘greatest happiness’ scenario that only reflects its demand side. In contrast, the professional or intelligent group takes the quality of all preparatory work as most crucial for the final outcome at the EU level. It selects and regularly checks rationally the best possible strategy, agendas and targets, all as necessary means to success. On this work it is critical as it, if blindly believing in it, would not be a thoughtful player but a naïve bureaucrat. All work is valid only for a period of time and to the best of available knowledge. The better the quality of the work, the more it helps to achieve the desired outcomes.

VI With What Result? Take the following two cases. In the 1993 Banana Dossier, the then Dutch Minister of Agriculture and former Minister of Development Aid Piet Bukman supported in the first Council reading the Commission proposal (1992/359) to protect the ‘ACP bananas’ (from poor countries). The Council agreed by (very marginal) QMV. In the second reading he supported the opposite case of market liberalization for the ‘dollar bananas’ (from US MNCs and largely shipped through Dutch harbours), became part of the overruled minority and regretted not to have blocked the proposal at the first reading [Pedler, 1994]. He should have regretted his lack of a logically derived and consistent strategy. In 2012 the German association of transport companies VDV challenged publicly its European Rail Freight Association ERFA about its support to the Commission’s white paper (2010/474) on the Unbundling of Railway Infrastructure and Operations. VDV had supported this position before, but changed this under pressure of the new director of its main member, Deutsche Bahn DB, so showing that the company lacked a stable strategy. Such events may happen. An amateurish player wants to forget such mistakes as soon as possible. An intelligent interest group, in contrast, wants to learn from its mistakes in 228

Chapter 5

Managing the Home Front

order to avoid them in future. Such a lesson is free as its price has already been paid by the loss. As the value of rationally assessing the effectiveness or success of influence actions is beyond doubt, the question now is how to do this. The answer serves both the hindsight wisdom useful for success in the future and the accountability or maybe prestige on the home front. Whatever the reasons behind it, the answer is less simple than it may seem to be. Of course, the ultimate justification of PAM is that it did realize the targets, even if in daily life the interest group may feel happy with good compromises, enduring respect from stakeholders and stable support from home, as said at the start of chapter 3. The targets remain, however, our reference point for the measurement of success. The professional group wants to see a causal relationship between its actions and the EU outcomes, this being the dependent variable. Its actions are not necessarily a case of doing something. As with the indeterminate arena, it may be a case of not doing anything, and waiting and seeing instead. Whatever it decides, the group can be said to have influenced the EU if – and only if – it has more or less caused EU outcomes as desired. However, causality is always difficult to determine. Following David Hume [1748], one can prove that, for at least two reasons, strict causality in social life is impossible to prove perfectly. Firstly, during the process, many more players may have tried to influence the outcomes, by which multi-causality a single impact cannot be isolated. Secondly, the interest group can never know what the outcomes would have been if it had been absent. The perfect control situation, necessary for the hard proof of causality, is not available in social life [Exadaktylos and Radaelli, 2012; Henning, 2004]. All this may be considered true and disappointing. The alternative, however, is not necessarily that one sees a PA campaign to get desired outcomes as a tombola beyond any explanation. In terms not of strict causality but of plausibility, one can examine some dependency of EU outcomes on PA actions, by making use of four next-best methods, often used in some combination, for assessing influences or impacts [Dür, 2008; Van Schendelen, 1998, 13-17]. These methods only take into account the question whether influence had been exerted and not by which specific factor(s). In terms of validity their rank-order is as follows: 1 Before-after. Here influence is measured by comparing the initial targets of an interest group with the final EU outcomes and by explaining, as well and plausibly as possible, the latter by the former [Dahl, 1991]. Then one may see whether the group has plausibly acted as a factor or not. 2 Initial chances. Now the context of the group’s initial position in an arena and its inherent chance of success are taken into account, as 229

Managing the Home Front

Chapter 5

revealed by the arena analysis. A small impact in an unfriendly arena can be considered a greater achievement than a big one in a friendly arena. 3 Reputation. Here the group’s reputation of (potential) influence is measured among informed EU officials and stakeholders, by inviting them to give their frank opinion about its performance in influencing the outcomes. 4 Backing at home. Now the measure is the support for PA in the EU at the home front. If the group gets stronger support (like facilities and mandate) after a PA fight, even if the fight was not won, then it has yielded at least some success here. These four next-best methods clearly have their weaknesses. They all fall short on strict causality and the last two also on high validity and reliability. Only the first two are clearly linked to the EU arena and therefore appropriate for measuring the effectiveness of PA actions in the EU, while the third is linked mostly to the arena of stakeholders and the fourth to that of the home front. However, if the best method is impossible, the nextbest methods are the best possible. A few surveys show that, if effectiveness is measured at all, it is done in terms of proxies. The large majority of 67 corporate PA officials takes ‘cost-saving for the company’ as a proxy for the before-after method and ‘recognition by high-level decision-makers’ as a proxy for reputation [EurActiv, 2009-A]. A majority of 170 PA officials of EuroFeds relies on ‘feedback from (potential) members’ as a proxy for backing at home [EurActiv, 2009-B] and that of 75 consultants relies on ‘clients satisfaction’ as a proxy for backing and ‘increase in our turnover’ as a proxy for their before-after method [EurActiv, 2009-C]. For the sake of its PA improvement and accountability, the professional group, different from the amateurish one, invests in a critical evaluation of its PA activities in the EU. At dossier level it compares the outcomes with its initial targets, preferably for different phases (‘where did it go wrong or better?’). To this it adds the hindsight of initial chances. Any correlation between efforts and outcomes should be explained in plausible terms. Most interesting are the unforced errors, which can never be blamed on others. For example, being misled by a competitor’s clever Triple P game is not necessarily an unforced error, but not having reckoned with this certainly is. To make a useful evaluation, the PA desk involves insiders at home (board, staff and line experts) and outsiders (officials, stakeholders, watchers), as the former can add indicators of backing at home and the latter those of reputation. It even invites, like in professional sports, its strongest opponents as eliciting their views can be extremely informative and even result in a friendly relationship. 230

Chapter 5

Managing the Home Front

An efficient way of collecting basic information is to approach key people by open questionnaire or interview on crucial items of quality, taken from our previous and following chapters, such as on its window-out intelligence, downsizing to shared-list and short-list, setting of targets, supply side to others, internal cohesion, used platforms, PA skills, briefing and debriefing, performances on the playing-field and much more. To avoid ‘inspired results’ the professional group organizes the evaluation impartially (such as by an outsider), carries it out confidentially and involves many people from both inside and outside. In short, it reviews its PA process from start to finish critically. By doing this for various dossiers regularly, it gets more valid and reliable insights into its PA performances in the EU. By engaging in hindsight it can gain better self-knowledge and thus inspiration for improving its PAM, including ideas for new R&D in PA and for training on its weakest points. The reward is improved competitiveness in future, as most other groups are nonchalant when it comes to self-evaluation and thus cannot strengthen the plausible relationship between their efforts and the outcomes. An annual report on PA performances is, even if kept secret, among the large majority of interest groups in EU a real scarcity indeed.

VII The Multinational Organization’s PA in the EU No interest group has its PAM in a perfectly organized condition, but some are more perfect than others. In particular, some multinational organizations (MNOs), largely in the private sector and ranging from Unilever to Greenpeace, are trendsetters in professional PAM, both internally and externally. They have to be. Internally they must cope with an exceptional diversity of cultures, role expectations and formal duties and exploit this as potential richness for coping effectively with the diverse EU, as said before. Externally, they have to cope with EU proposals in different policy fields and with the many actors, factors and vectors that may cause their outcomes. Some even make part of the EU agenda. A famous example is the 1985 EU policy programme ‘Open Market 1992’ that was drafted by the PA desk of the Philips Company in 1982 and first discussed at the Round Table of Industrialists in 1984 [Verwey, 1994]. Most MNOs also possess adequate capacities to exert influence, such as sufficient financial means, technical experts, cross-border networks and relevant positions, in which respects the companies and NGOs among them are not much different [Jordan and Maloney, 1996; Jordan, 1997]. Not least, they often have a good image that includes reputation and supply and provides access to EU officials for exchanging information and support at work-floor level. 231

Managing the Home Front

Chapter 5

Every interest group might learn from the MNO’s methods of managing PA at their home front [Coen, 2007; Sietses, 2000]. Even if the main lesson is not to follow their example, then such a decision is at least better reasoned. Of course, the MNOs are far from all being the same. Some are more multinational than others, or are organized as country units rather than product or service units. More factors contribute to their variation, ranging from market position, economic sector and labour intensity to history, ownership, age, prestige and more. MNOs are frequently different by type of product or service (profit versus not-for-profit) and by internal authority pattern (top-down versus bottom-up). In spite of all diversity, we observe among them a basically common pattern of PA organization, which is pragmatically derived from the EU processes that have to be managed (‘internal structure must follow external processes’). By the very nature of the EU, these processes are full of dynamic challenges and thus cannot be managed by a standing bureaucracy but only by a unit that is lean and mean, operates directly under the top management and has close networks with the staff and line experts of both the specialized divisions and the country units. The three main parts of this basic structure, simplified and stylized in figure 5.3, are as follows: – The Central Board (‘headquarters’) is responsible for the general EU strategic management. It has its own expert staff, such as on finance, law, sales, public relations (PR) and public affairs (PA). A key member of the Board is directly responsible for the Central PA Desk that is seen as ‘Chefsache’ and between the two exists a dual relationship. On the input side of the Board, where the corporate strategy is decided, the PA staff delivers crucial information about relevant EU challenges (the long-list) and reasoned proposals for main targets and best ways (like the shared-list) to manage them. On the Board’s output side the PA staff receives the decisions as a mandate that includes the strategic PA agendas (shared-list and short-list), the main targets, the allocated resources and the time schedule for reporting on progress (often by semester). The mandate leaves to the PA Desk the choice of tactics and techniques to bring the agendas to success, but it may contain a few pickets that limit the room to manoeuvre, for example for reasons of reputation. The mandate binds the whole MNO, from the PA Desk and other staff-and-line units in the countries to the Board itself. This way the MNO can act as one body with one voice and one message, both internally (country units and divisions) and externally. The logic of binding also the Board to the mandate is to avoid the risk that its members cause blunders and boomerangs by lobbying on their own, just as the prudent Odysseus ordered his sailors to bind him to the mast as protection against the temptations of the Sirens. 232

Chapter 5

Managing the Home Front

The multinational model Multinational HQ/board

PAM

PR

Law

Sales

Divisions

Country Units

EuroFeds

Natl. Capitals

......

Ad Hoc Teams From Homework to Fieldwork and Backwards

Brussels Office The Playing Field Com

EP

ESC

COR

Council

Court

Figure 5.3

– The technical and other experts of the divisions and country units are part of the wide internal network that, jointly with the other staff and managed by the PA staff, contributes to the internal take-off and the external realization of the long-list. Internally, the long-list is managed by shared-list method too, which creates optimal internal efficiency and explains why the PA Desk can be lean and mean. The PA staff promotes that the experts get connected to useful external positions, networks and platforms and it coaches them on how to score (2E) the targets by creating MI2. It all happens via a mechanism of mutual briefing and debriefing. Its logic is that the technical experts having high expertise 233

Managing the Home Front

Chapter 5

on the substance of a dossier but little on the playing-fields and the game of playing and the PA experts having reverse qualities can only achieve success at EU level by working together as a team and thus by mutual briefing and debriefing. The mechanism is anchored in ad hoc teams, intranet settings or otherwise (like teleconferencing and email) at dossier level. – The Central PA Desk is lean and mean, as intelligent PA is most dependent on ounces of brains rather than kilos of body weight. Usually it has less than a handful of staff. Its real working force is its internal network of experts, whose ‘ears and eyes’ help to gather useful information and whose ‘hands and feet’ help to build support, all embedded in briefing and debriefing. The Central PA Desk is focused not only on the EU but also on the global and domestic levels as EU dossiers usually have either their origin or overflow there and thus may need a multi-level PA approach. To support its daily fine-tuning of window-out and windowin lobbying, it usually has a small local office in Brussels and major capitals inside or outside the EU. This Brussels office is also small-sized (one senior, a junior and/or secretary) and has a location and budget that allow basic hospitality. The Central PA Desk oversees the main PA activities on the shared-list and short-list and reports regularly to the Board member responsible for PA. Within its mandate, the Central Desk has the following three core jobs to do.

1 Dossier Level Tasks for the Central PA Desk At the level of the long-list dossiers its first task is to get all preparatory homework done, from the window-out organization of intelligence to the window-in one of sufficient support from outside (figure 1.2). A proactive PA staff starts to think ‘backwards’ from the last step (support lobby) up to the first one (information lobby), by which it might save much energy later on and improve effectiveness. It also does more detailed PA work, as summarized in figure 3.3, in order to get the organization ready for, inter alia, changes in the EU playing-field, external shared-list platforms, strategy review, scenario planning, corporate memory and evaluating past ‘games, sets and matches’. Due to the overload of preparatory work, it has to play ‘intelligent laziness’ internally (shared-listing), by entrusting parts of PA work to other staff and line experts, with internal co-operation as a by-product. In the mid-1990s, the PA Desk of Siemens set a new trend (‘Siemens method’) by regularly asking its division and country managers to provide email feedback about the relevance of a long-list of EU challenges and the actions needed to be taken in relation to it. The PA Desk has evidently a selfish interest in making the short-list agenda the short234

Chapter 5

Managing the Home Front

est possible and the shared-list one long, as the former leaves all PA work at home and the latter gives better MI2 at lower costs. The second task of the PA Desk is the internal implementation of the two agendas of the long-list and, if existing, the short-list. By the way of ad hoc teams, online procedures or otherwise, it acts like a chef de dossier by bringing together three sorts of relevant people. Firstly, members from other staff units, such as from Public Relations (for managing public reputation and the issue of ‘sound’, to be discussed in the next chapter), EU law (for legal aspects and the bottom-line scenario) and, in companies, Sales (as PA must contribute to the profits, a lesson learnt again by Shell during the Brent Spar affair). Secondly, technical experts from the divisions, for tackling the technical side of an EU issue such as biotechnology, standardization, environment or engineering. Often they hold or get a seat in an EU expert group or EuroFed, by which they can contribute to window-out and window-in lobbying. Thirdly, people from the country units are included for their maybe different interests, which can better be included than excluded, and their supplementary capacities, such as tapping information from their government, getting supportive mass in their country and contributing to its multinational orchestra. The mechanism of briefing and debriefing keeps it all together. The PA desk’s third task is to realize the external fieldwork of exchanging information and support with EU officials and other stakeholders. Its main tool for each dossier is the ad hoc team, whose members represent the MNO and communicate with stakeholders and officials. It often already has a relationship with them, thanks to both their line position as expert and their membership of a EuroFed or an EU expert group. In order to carry out the fieldwork well, the PA Desk uses both the mechanism of briefing and debriefing and its Brussels front office. The latter acts as its ‘ears and eyes’ and ‘hands and feet’ on PA matters there. Going window-out, it monitors competitors and officials and forwards information to the Central PA Desk. Going window-in, it fine-tunes the lobbying by, for example, coaching the technical experts on the spot (often five to ten per week), talking with Commission officials and stakeholders and engineering indirect links. It is, in short, the most direct tool for achieving remote attention and control the MNO has. In the multinational PA approach to EU affairs, the EuroFeds and their ad hoc variants play a crucial role. The MNO can usually gather support for its interests and add crucial weight to the M of 2E = MI2 here. To this end the PA Desk drags the dossiers selected for the shared-list as soon as possible to a suitable platform. It extends or moves its three main PA tasks (preparatory homework, implementation and fieldwork) from home to this level, where they get their substance. The PA infrastructure 235

Managing the Home Front

Chapter 5

at home, with its ad hoc teams and briefings, is extended to here too and for each shared-list dossier arranged differently. An MNO easily participates so in more than fifty different external platforms or consortia with useful stakeholders. The home organization always remains, however, the ‘war room’ that organizes the remote control (as detailed above) over the platform’s homework, implementation and fieldwork, to be checked by, for example, its own Brussels office or one of its experts there. Inevitably, the MNOs are much dependent for their final scores in the EU on the PA quality of their platforms. The risks and costs of this dependency they usually estimate as much lower than those of a short-list approach. If they are direct member of a EuroFed, they can manage the risks by role-playing inside, co-setting the common agenda, providing their technical experts, checking the effects of common actions, keeping their financial contribution at issue and, if needed, by shopping around for an ad hoc coalition. If not a direct member, they often play these games indirectly, via one or more of their national associations that altogether form the EuroFed. Almost all EuroFeds consider the influencing of the EU as part of their raison d’être (chapter 3-IV-2) and invest in both their good image and the PA quality of their EU affairs. They have a small PA office (‘secretariat’), ad hoc teams (‘work groups’) and standard procedures for arriving at a common position and targets. Technical experts they usually recruit from among their members, particularly the MNOs.

2 The Issues of Trust and Consultants Many MNOs face two special issues. One is the issue of trust between the Central PA Desk and the Brussels office. On the one side, the latter needs a clear mandate that gives room to manoeuvre as otherwise it cannot finetune its actions. On the other side, the Central Desk wants to keep control over what is done or left undone. British, Dutch and Nordic MNOs usually give their Brussels office a mandate for most window-out and windowin activities there, which are scrutinized regularly for the results. MNOs from the rest of Europe (and the US as well) usually treat their Brussels office as merely their delegate that is strictly instructed for going windowout rather than window-in, as they fear the local office will become a player apart from the headquarters. The price they pay for this distrust in their Brussels office is often corporate passivity (or even absence) on the EU playing-field. Waiting for new instructions from Frankfurt, Paris, Madrid or Washington, the local officer sees them arriving after the game. Some MNOs have addressed this issue by compromising on distance: part of the week (usually Mondays and Fridays) the Brussels officer visits the Central 236

Chapter 5

Managing the Home Front

Desk, which requires much time and travelling. New communications technology may soften the issue more. The second special issue inside MNOs (and many other interest groups) is the usefulness of hiring PA consultancy firms. A questionnaire among 140 Brussels-based PA officers, largely working in big companies, reveals strong disagreement about the usefulness of PA consultants, with almost 35% in favour, 55% against and the rest not having an opinion [ComRes, 2008]. Due to the lack of in-depth research, one can only speculate about the reasons behind the negative appraisals. One reason can be the (perception of) poor quality of many consultants, among whom all varieties of quality exist indeed. Many consultants advertise themselves as full-fledged PA ‘professionals’, even though they are, in fact, not really experts or are confined to only a PA niche, for example internet monitoring, image creation or contacting MEPs. Another reason may be the poor quality of the contract made with the consultant, for which the interest group should blame itself. A quality consultant can, however, certainly add value to a lot of preparatory homework, implementation and fieldwork, particularly on complex or strange dossiers and during rush-times [Spencer and McGrath, 2008, 73-77]. The person cannot, however, play at least three crucial PA roles. Firstly, the consultant can give advice or second opinions, but never bear responsibility for any important decision or choice. The person can, secondly, supplement the in-house PA Desk but, lacking position, network and inside knowledge, can never replace it credibly. Thirdly, the person can carry messages to and from stakeholders and officials, but can never represent the organization in an authentic way, let alone negotiate on its behalf [Warntjen and Wonka, 2004, 89]. The MNO model of PAM at the EU level shows many strengths. It is, inter alia, integrative by its inclusion of internal variety, studious by its preparatory homework, intelligent in terms of its internal shared-listing, efficient on account of its flexible structures and potent as a result of its early intake of external support. Its organization by the way of a Central PA Desk with flexible ad hoc teams and a Brussels outpost, all under supervision of a Board member, is like that of a flexible military intervention force. Thanks to its multinational background, an MNO can easily navigate along national, transnational and international routes, play prominent roles inside EU platforms and fine-tune its communications with the relevant people inside the multinational EU institutions, in short promote its interests more efficiently and effectively. The model is a stylized summary of the basic pattern of PAM in many MNOs and, as such, allows variety in reality, while some MNOs hardly match it at all. In the EU, many MNOs have the reputation of belonging to its ‘premier PA league’, as they are seen as scoring relatively high on achieving their targets, strengthening their 237

Managing the Home Front

Chapter 5

positions and enabling their people to gain access and influence. They are, however, more admired in theory than followed in practise. Most other EU interest groups seem to find the application of the MNO model difficult.

VIII Extra: the Multinational PA as Benchmark In this section we shall use the MNO model as tentative benchmark for assessing the PA of a few other categories of interest groups in the EU, as based on the facts and findings mentioned in previous chapters. Its objective is to get other groups to considering the usefulness of the MNO’s basic elements for their own organization. Their answer, when asked why they have not already applied it, is usually two-fold: ‘we are different from an MNO as we have neither many PA assets nor a multinational background’. The rejoinder can be that by early shared-listing they too can acquire the assets and the multinational mass like an MNO. Apparently they find it difficult to share their interests with other groups. How valid is this difficulty? Small-sized interest groups in the fields of SMEs, decentralized governments and NGOs are most in contrast with MNOs. They usually have a low sense of efficacy for EU action, few resources and skills, too parochial an image, weak common identity and a more domestic than European interest orientation. Their long-list is usually made introverted, their short-list long and their shared-list (if any) really short. Their targets are usually set in terms of activities to be done and not as precisely defined and concrete end-results. If wanting to get something from the EU, they tend to rely on their director or secretary and also on their national ministry or association. A small minority, largely consisting of decentralized governments (big cities, regions and agencies), have a Brussels office. Often their drive behind this is not PA, except for the case of EU subsidies, but getting visibility in the EU, thus public relations or PR. Their Brussels officer is usually a loner, who is weakly embedded in the organization and acts on the spot according to circumstances. Only few regions, cities, NGOs and associations, usually driven by domestic devolution and opposition [Tatham, 2010], are active on an EU platform. In short, the real reason why these interest groups are far away from the MNO model is not that they are no MNO but have hardly internalized PAM. The few exceptions among them demonstrate that, after this internalization, they can perform alike MNOs. Since the mid-1990s some regional governments and independent agencies did join a EuroFed, network or ad hoc group, for example EIRA (steel regions), AEM (mountain areas) and EReg (vehicle and driver registration), and so got leverage to improve 2E by better MI2 at EU level. By gaining positions on these plat238

Chapter 5

Managing the Home Front

forms, they became familiar with the EU work floors and enjoyed some success in influencing the EU [Scully and Jones, 2010]. Most decentralized governments are, however, still weakly organized on PAM at home. A major cause seems to be their government status (chapter 7-VIII). Agencies often fear supervision by the ministry and regions and cities that by their elected politicians and mass media, and so they all feel hindered in their ability to internalize PAM. In 2005, the Flemish government cleverly established VLEVA, its open PA desk for Flemish interest groups in the EU, as an independent agency, at a distance from parliament and party politics. One might expect that national ministries would have internalized most elements of the MNO model. Since the 1993 Treaty of Maastricht, which broadened and deepened EU policy domains, they are all dependent on EU outcomes now, have seen their power to make binding decisions shifting to the EU and (one would think) might want to influence them there. In the Council they possess an infrastructure of common facilities, meetings and work groups. The Commission invites them to take part in committees and agencies. Given such fine preconditions for influencing the EU, one might expect an MNO-like approach to their EU affairs. Paradoxically, their real approach is frequently almost the opposite. Ministries are usually focused on domestic values and interests rather than EU affairs, on which they mistake their (weak) EU power positions as already influential. Their EU long-list is usually an introverted wish-list and seldom split into a shared-list and short-list, and the latter, if existing, is seldom short. Only a few have an EU/PA desk for preparing, implementing and realizing their EU agendas. The majority manages its EU affairs through their standing bureaucracy, composed of policy experts under scant supervision and coordination by superiors, let alone briefing and debriefing with other staff and line members. Lacking PA expertise and sufficient resources, the policy experts limit their EU work to the (reimbursed) meetings of the Commission’s expert groups and committees and the Council’s work groups, so neglecting the informal build-up and the many other stakeholders. Ministries rarely have a Brussels outpost or mistake as such their nationals in the Commission and their staff in the Permanent Representation (PR). However, the latter is focused on Council negotiations (‘last phase of 15%’), hardly on all the rest before and after. For the early europeanization of their interests, ministries seldom have a sort of EuroFed or they mistake as such, again, their Council setting. In short, most ministries hardly invest in MI 2 and thus score poorly on 2E. If they score on a few prioritized dossiers sometimes, they do so at high costs (or by luck) and at the detriment of their other long-list dossiers. Not surprisingly, most domestic stakeholders with experience in the EU 239

Managing the Home Front

Chapter 5

are reluctant to entrust their interests to such ministries and prefer to act self-reliantly. The few exceptions to this general pattern can be found among small-sized Nordic ministries, dossiers that are politicized at home or fully fall under Council regime, some informal Councils and among a few ministries, such as Defence, Transport and Finance (chapter 3-IV-2) that have some European ad hoc network. In theory, the national governments, defined as the domestic central government executives, also meet many conditions for edging out the MNOs on PAM in EU. They claim to be strongly interested in the EU outcomes, have many privileges as laid down in Treaty texts and possess many resources, such as manpower, budgets and technical expertise. They have a procedure and structure for national co-ordination. The reality is, however, much different from the theory. Except for a few ‘statist’ countries like France and the UK, the national government usually has at the top not a Board or CEO as MNOs have, but an inner arena of competing ministries and parties. Their tensions are fanned by the pluralism of the parliament, civil society and mass media, all being always divided over any EU challenge. Their national co-ordination is neither national by its exclusion of most other domestic stakeholders, nor a co-ordination of activities (chapter 3-IV-1). Most domestic stakeholders prefer to act self-reliantly in the EU and, if necessary, against their government. The co-ordination procedure largely covers only Commission proposals for legislation, thus missing most of the processes before and the acts thereafter. Only a few governments have a sort of Central PA desk, like the SGAE in France and the EU Cabinet Office in the UK, but the French one lacks a serious mandate and the UK one largely manages only the short-list of the ruling party or parties. The PRs that are formally the government’s frontoffice for Council affairs, gradually act informally as also the local office of (in MNO terms) their ‘headquarters’ by doing some corridor lobbying in the Commission and EP, but this is hardly embedded in MNO-like PA arrangements [Panke, 2012]. Undergoing the europeanization of their administration, most governments react more by adapting their structures to EU requirements (the downstream) than by redesigning and using them for PA management in the EU (the upstream) [Van den Berg, 2011] and, if the latter, they are reactive rather than proactive. Only when a new Treaty or their Council chairmanship comes nearby or when a very special interest arises, they establish a high-level ad hoc team for listing priorities as their wish-list. In short, they behave as if they each are really powerful in the EU and not dependent on MI2 for creating 2E. The new informal Council meetings and the bi/multilateral ones (like those of Benelux, ‘Berlin-Paris’ and Baltic States) might, however, eventually develop into platforms for building-up MI2 for better 2E. 240

Chapter 5

Managing the Home Front

Continuing our ‘intermezzo’ in chapter 2-III-3 we also benchmark the European Commission. Its DGs function on the one side as open channels for the aggregation of public and private interests from the countries and on the other side as interest groups themselves that try to bring the imported interests to common EU decisions targeted before. The most manifest vehicles for this are the annual work programme and the specific proposals that must pass the College. The Commission looks like a national government, as it is internally divided by the DGs and their layers and is externally under pluralistic pressures. However, its management of these challenges looks like that of the MNO model. The main layers of a DG are the policy units that, falling short of in-house technical expertise, in-source experts from interest groups. The chef de dossier acts like a PA officer, who brings stakeholders together in more or less open expert groups that function as common platforms, so applying the shared-list method for itself and for the rest working on the Commission’s own priorities (short-list). Some striking differences between the workings of the Commission and the MNO-model exist too. The chef, firstly, usually wants to in-source diversity, thus not only potential friends but also opponents, all delivering varied expertise and support. Secondly, the chef can play Triple P in-house, while the PA officer of an MNO has to do this from outside. Thirdly, the chef gets the PA mandate usually from the head of unit, who is subject to the control of higher levels, and not from the highest level. Fourthly, with so many chefs each managing a few dossiers, all the chefs together make the Commission look like a PA factory with numerous PA desks. It makes the Commission a superb case of ‘intelligent laziness’. By its early internalization of external diversity, it can proactively play the ‘backwards lobby’. By blue, green and white papers on both its shared-list and shortlist, the Commission efficiently monitors proactively and permanently the positions of stakeholders and other officials, including the MEPs (who are the only ones absent in its expert groups). From this tentative benchmark we can draw the following main observations and lessons: – PA expertise is crucial. Interest groups without a qualified PA desk take the high risk of lacking the political expertise to monitor and manage the EU playing-field and the game of playing. Small-sized groups may consider its costs prohibitive, but not if they pool their limited resources and enter a common platform in their sector or region at home, (to be) connected to the EU. National associations, ministries and governments cannot seriously consider the costs of qualified PA to be prohibitive, as the costs of non-PA are often exponentially higher. – EuroFeds are essential. Any variant of a EuroFed can give strong lever241

Managing the Home Front

Chapter 5

age to the promotion of one’s interest. Without it one stands singlehanded on the EU playing-field and loses much PA efficiency and effectiveness. As part of a EuroFed, any domestic interest group gets a virtual MNO construction and thus greater MI2 and 2E within reach. The different country members are its virtual country units that allow more multi-level lobbying (from local to global) and fine-tuned PA (like varied routes and tongues). For almost every group there exists a sort of EuroFed, to be contacted via its association or directly. Most ministries and governments lack such a platform, to their own disadvantage. – Commission as example. All EU interest groups can learn from the intelligent way the Commission manages its PA. Even for use outside the EU or only on their home front, they might profit from applying the Commission’s ‘internalized shared-list method’ of in-sourcing external stakeholders, so getting excellent information and commitment at a low price. EuroFeds might follow this Commission example by in-sourcing more stakeholders from outside into some of their work groups. MNOs too might apply this method for their few short-list dossiers, by making their internal ad hoc teams more open.

IX From the Home Front to the EU Fieldwork Organizing one’s home front is clearly a difficult job. Indeed, it may take over half one’s energy available for PAM in the EU. It requires a lot of skilled ingenuity. Nobody can blindly produce cohesion, knowledge, resources, skills, image, strategy, targets, scenarios, evaluations and more. All these ingredients for cooking up success should be prepared carefully and thus selectively by preparatory work. A clear example is the determination of the shared-list, which requires negative selection from the longlist of challenges, identified by the strategic confrontation of perceived facts and chosen norms. Only after this, it is possible to know in which quantity and quality which ingredients are required. The interest group neglecting this homework will almost certainly be punished by poor results from its EU actions. The quantity and quality of preparatory homework are, again, usually less than perfect. If the costs of perfection exceed those of imperfection, this imperfection may be justified, as the best is the enemy of the better and optimization is the best possible. The time pressure caused by EU deadlines alone often ensures that the alternative to an imperfectly prepared action is not perfection but taking no action at all, which may be worse. If an interest group believes that its preparation is better than that of its best competitor, it should not feel satisfied however, as it cannot 242

Chapter 5

Managing the Home Front

count on always winning the final ‘tiebreaker’. If it sees that some prominent competitors are organizing their home front better, then it has to improve its own preparation. This logic, true for every competition, is indeed the hard reality of the EU playing-field. More than ever, many interest groups want to create a desired outcome and invest in intelligent preparation for this at home. The multinational model is surely not the last word on optimizing the home front. It looks well designed: integrative, professional and efficient. It earned a reputation of success. Few elements of it are widely followed now, particularly the settlement of a Brussels office. By consequence, the followers have many questions about internal organization to answer, covering everything from mandate and shared-listing to briefing and debriefing. Usually they follow the MNO model piece by piece, starting with a Brussels officer and, after many years and frustrations later, creating a lean and mean PA desk at home. It remains a question of ongoing research whether the reputational claim, that this MNO model of PAM in the EU creates significant success, is based on facts. Both its theoretical logic and its reputation among practitioners favour the claim, but this is less than empirical evidence. The tentative benchmark suggests that the model can be improved by incorporating some elements of the Commission’s practises. A model that may be necessary for success is, besides, never identical to one that may be sufficient. The MNO model is part of the preparation for the match and is not yet the real match. The EU fieldwork still has to come. This is the theme of the next chapter.

243

Chapter 6 Managing the EU Fieldwork In field operations there are three sorts of armies. One has ardour and discipline, the second ardour but no discipline and the third neither ardour nor discipline. The most important is discipline, as this stimulates ardour and finally victory. What are the implications of this view of Machiavelli in his Discorsi (book III-36) for the PA fieldwork in EU?

I Lobbying: The Final Link In chapter 1-IX, we referred to lobbying as an old effort to influence decision-makers by visiting some place (lobby, antechamber) belonging to them, or by ‘corridor behaviour’. We defined this behaviour technically as ‘the build-up of unorthodox efforts to obtain information and support regarding a game of interest in order to get a desired outcome from a power-holder’. The element of ‘unorthodoxy’ we characterized generally as indirect, informal, silent and charming behaviour. Modern insights from PAM tell that even such a visit to a power-holder, called officials and stakeholders today, requires a lot of study and prudence, although similar insights can be taken from, for example, Sun Tzu and Machiavelli. In the previous chapters we applied this sense of disciplined creativity to the EU playing-field, the routes and navigations to it, the preparatory homework and the management of the home front. The three main lessons from these chapters can be summarized as: countless factors determine success or failure; many of them are manageable to some degree; and this management requires much study and prudence. The third lesson is necessary but not sufficient for making the strategic, tactical and technical choices so that one is a professional rather than an amateur, intelligent rather than nonchalant and a winner rather than a loser. After the last chapter on strategy we focus here on the tactical and technical choices that have to be made for the activities on the playing-field, called corridor lobbying in the past and, more broadly, fieldwork now [McGrath, 2005-A]. The tactical choices, anticipated in chapter 4-IX, we continue here in combination with the technical ones, because any ambitious effort to influence whatever power-holder (public official or private stakeholder) requires one overall approach that integrates strategy, tactics and techniques. Their intended effects must come together in the activities on the playing-field. As stated in chapter 1-X, these activities include 245

Managing the EU Fieldwork

Chapter 6

much more than just corridor lobbying, as most PA-professionals only spend about (an estimated) 15% of their time in corridors [Spencer and McGrath, 2008]. The modern ‘art of lobbying’ is to create influence success even (almost) without visiting corridors at all, (for example by a silent precooking of a Triple P setting. Referring to figure 1.2, the fieldwork includes all activities for both the upper arrow that stands for the window-out intelligence lobby, and the lower one that represents the window-in support lobby. On the upper arrow the fieldwork contributes to the homework for organizing the intelligence, and on the lower arrow the fieldwork is, reversely, directed by this organized intelligence for the creation of sufficient support. Combined with studious homework, the two forms of fieldwork result in the highest possible and final value of 2E for our formula 2E = MI2. Engaging studiously in homework alone may sharpen the brains but remains useless in practise, and conducting only fieldwork is amateurish. Although the focus on fieldwork is presented in the second part of the book here, in daily PA practise it often comes much earlier. Amateurs usually take it as both their starting and finishing points of considerations, if they have any. Professionals anticipate the tactical technical choices by their simultaneously backward thinking (from step 5 to 1 of figure 1.2) on strategy, tactics and techniques.

II The Surplus of Unorthodox Actions For the creation of a desired outcome on the EU playing-field every interest group has really no shortage of potential actions. It can choose from an almost infinite menu of possibilities of going window-out and windowin. Firstly, it can use the orthodox ways and routes as mentioned in chapter 1-IX, for which it does not have to lobby at all. Secondly, it can fall back on traditional techniques of influence, such as coercion, advocacy and argumentation (same chapter). Thirdly, it can make use of the many unorthodox techniques that are discussed below. In fact, almost every variable is potentially instrumental. There are always many actors to approach, factors to use and vectors to construct. The options of ‘exit, voice and loyalty’ [Hirschman, 1970] or, in popular parlance, ‘flee, fight, flirt’, are only three out of many. Every technique can be used in one or another direction, for example as stick or carrot, thus posing dilemmas to think about, such as the appetizing one in chapter 4-IX-1. Figure 6.1 presents a selection of common dilemmas in EU fieldwork. The first two categories regard actors like EU officials and other stakeholders directly and the others the factors and vectors affecting them indirectly. The following exemplifies these common dilemmas of fieldwork. 246

Chapter 6

Managing the EU Fieldwork

Major Dilemmas In Managing The Fieldwork Regarding

Dilemma

Officials

- Support or Oppose - Provide Information or Not - Appease or Disquiet

Stakeholders

- Mobilize or Demobilize - Reward or Punish - Divide or Unite

Relations

- Formalize or Informalize - Stabilise or Destabilise - Use Old or Make New Ones

Procedures

- Simplify or Complicate - Accept or Litigate - Interpret Strictly or Loosely

Positions

- Recognize or Challenge - Take or Leave - Recruit By Merits or Spoils

Issues

- Politicize or Depoliticize - Broaden or Narrow Down - Combine or Split

Frames

- Keep Frame or Reframe - Up-Frame One’s Own or Not - Down-Frame the Opponent’s One or Not

Dossier Life

- Relieve or Hinder - Sustain or Transform - Push or Delay Time

Arena

- Enter or Leave - The One or the Other Coalition - Restrict or Expand

Negotiation

- Use Stick or Carrot - Overcharge or Not - Compromise or Oppose

Decision

- Accept or Appeal - Adapt to or Escape - Submit or Undermine

Lobbying

-

Supplying or Demanding Low- or High-key Direct or Indirect Formal or Informal

Figure 6.1

247

Managing the EU Fieldwork

Chapter 6

1 Officials. The interest group can support or oppose the position that officials prefer to adopt on an issue, so making their life easier or more difficult. It can favour or damage their private careers, for example by good or bad publicity. Some officials may tolerate practises of patronage and clientelism, and thus receive subtle favours like lunches and loyalties in return for compliance. All officials have an appetite for reliable real-life information, which the interest group can satisfy or not. It can also appease or disquiet the concerns of the officials. Making a drama out of, for example, global competition, pollution or recession is an evergreen technique of bringing them rapidly to new EU agenda-formation. Even more direct actions are possible regarding the officials, such as changing their prestige, competencies, budget, workload, organization or recruitment. Many officials frequently play these games against each other. 2 Stakeholders. One dilemma is whether or not to mobilize more of them. More friends give more support, but they take a larger share of the game too. Another dilemma is either to reward the friends and those who surrender or to punish the opponents and the defectors. It is also possible to promote either division or unity among groups of stakeholders. Slumbering stakeholders might be kept sleeping or be awoken. Those wavering over an issue can be ignored, convinced or seduced. One can deal with irritating players by either excusing or accusing them in public, perhaps thereby bringing them either to obligation or to compromise. In 1998 the Dutch Minister of Finance Gerrit Zalm, defending his practises of fiscal aid to companies (‘fiscal dumping’), forwarded to the Commission evidence that he had ordered from international accountancy firms and that showed even worse practises by other governments. This halted the issue for three years. 3 Relationships. The interest group can make a relationship either more formal, semi-formal or informal. If it wants to have all three variants, then it has to cope with the dilemmas of their balancing and timing. It can also make a relationship more or less stable. Conflict can bring a better return than friendship [Coser, 1956], as shown in the CCMC/ACEA case of the car manufacturers (chapter 5). Another dilemma is whether to stick to old relationships or replace them with new ones. Due to the continual turnover of EU officials and lobbyists, an interest group often has no choice and has to invest in new relationships anyhow. After the 2009 reshuffle of the Barroso II administration, many Commission people had to build up new networks, and many interest groups had to do the same. 4 Procedures. These are part of the Triple P game. They can be made more simple or complex. Before the 2002 Commission’s decision to allow Unilever to market its margarine Becel as ‘healthy’, the company lobbied 248

Chapter 6

Managing the EU Fieldwork

for a simple procedure and after that decision it preferred a complex one for keeping new competitors out. An interest group can either accept or challenge a procedure as well and, in the latter case, even go to the Court, as the German government did in 1998 by appealing the Treaty basis of the Tobacco Ban Directive. Another dilemma is about whether to lobby for a strict or loose interpretation of a procedure. Different language texts and daily practises (‘precedents’) often allow new interpretations and procedures. 5 Positions. As another part of Triple P, an interest group can either follow or challenge established bodies and positions, such as committees and agencies. It can decide to join or leave a body; the latter can indicate protest but also a desire for more freedom of action. On the recruitment of officials for key positions it can be in favour or against its basis on either merits or patronage (spoils), or for somebody from inside or from another DG, country, sector or political group. In 2000 the Commission decided to promote recruitment on merits basis. 6 Issues. One dilemma is whether to politicize or to depoliticize the facts, values, instruments and solutions that are arising. A standard technique of depoliticization is to present an issue as ‘technical’, which many take for ‘less relevant’. An example of this is an expert group on railway safety that routinely put a mathematical formula on the first page of its reports. Another dilemma is whether to broaden an issue or to narrow it down. If some problem cannot be cut down, it may be better to expand it. In spring 2000, Europabio pressed the extension of the proposed liability for GMO to all nutritional products, thus hoping to escape specific regulation and to broaden its coalition. A third dilemma is whether to combine various issues in a package, to keep them separate or to split them even more, which can also change the numbers and composition of stakeholders involved. The Commission and the Council often combine issues in order to satisfy many stakeholders a little bit. 7 Frames. Every interest contains one or more values that in pluralist societies stand at issue. Different from a fight over facts, a clash over values is not solved by best evidence but by strongest support. As the real and selfish interest behind a value does not easily attract much support, it should be (re)framed or presented in a way that can mobilize that support [Daviter, 2011]. One dilemma is to reframe a value or not and, if so, in which value direction. The allowance of novel foods on the market can be framed as health risks or liability, two sides of one coin. Another dilemma is to up-frame one’s interest up to a ‘general interest’ or to down-frame that of the opponents. Figure 6.2 gives EU examples of a selfish interest (first column) being up-framed to a high social value (second column) that secured wide support more easily, just as the 249

Chapter 6

Managing the EU Fieldwork

young lady Lorelei on her rock in the River Rhine caught many fishermen. Opponents may want to prevent or block such an up-framing, to downgrade it to its selfish contents and/or to replace it by their own upframe. Frequently a lobby fight is between two or more competitive upframed interests, for example liberty versus health on tobacco. An upframe is most useful for an unfriendly arena and requires much study of what important stakeholders, officials and audiences want to hear. It took the European textile industry, wanting to get rid of cheap imports, many years before it found the nicer frame of ‘child labour’. Many NGOs are experts on this Lorelei technique and hoisted such fine flags as ‘public safety’, ‘sustainability’ and ‘humane farming’. In 2005, trade unions

EU Examples of Up-framing The Core Interest

The Upgrade

CAP (old) CAP (new) Cheap textile import ban Tobacco smoking Tobacco ban Tax-free shopping Beer import limits Charging private transport Shorter truck-trailer coupling Hands-free car-phone ban Fur industry (against) Laying hen (against) Subsidies to southern Europe ACP bananas R&D subsidies Lower VAT (shoe repair etc) Open labour market (against) Open labour market (for) Chocolate definition up Subsidized food exports Subsidies to CEEC Biotechnology (against) Idem (for) Oil stocking SME support Intervention in ECB (for) Non-intervention in ECB Energy-saving products Novel foods

Food supply Sustainability Child labour Liberty Public health Employment Environment Sustainable mobility Public safety Road safety Animal welfare Humane farming Social cohesion Development aid Global competition Employment Social dumping Equality, Welfare Development aid Humanitarian aid Stability Public health, Environment Development Security Employment Economic growth Monetary stability Environment Public Health

Figure 6.2

250

Chapter 6

Managing the EU Fieldwork

of northern Europe campaigned against the Services Directive proposal (COD 2004/001) that would allow cheap labour to move from central to western Europe, down-framed this to ‘social dumping’ and up-framed their own position as ‘for a social Europe’. ‘Solidarity’ is frequently used as a Robin Hood up-frame to get EU budget funds transferred from the rich to the poor, from others to oneself. In 2010, popular up-frames are ‘climate change’ and ‘financial crises’. 8 Dossier life. The interest group can either help the passage of a dossier or make this difficult. An extreme example is the UK’s position on the BSE affair (1996), when it blocked in the Council more than fifty dossiers (approved by rubberstamp later). Another dilemma is whether to preserve or to transform the dossier contents. The latter, coming close to reframing, is an indirect technique of time management that can solve or create a stalemate and so push or delay the process. With the 1997 ‘Amsterdam’ decision to reframe the matter of monetary stability as ‘Stability and Growth’, a stalemate was overcome. There are countless techniques for pushing or delaying a process. The most direct one is by moving a deadline forwards or backwards. Usually it takes more energy to push than to delay. 9 Arena. The interest group can choose either to enter or to leave an arena, with wait-and-see options in between. Inside the arena, it faces the dilemma of whether to join one or another coalition. Often, every coalition has its own mix of advantages and disadvantages that can make marginal difference and thus offers a real choice. A strong dilemma (chapter 4-IX) is whether to restrict or to expand the arena boundaries. Changing them may result in a very different set of both issues and stakeholders. An interest group can also choose to make one arena overlap with or separate from another one. If the group takes multiple routes to the EU (figure 3.2), it encounters the arena dilemma of whether to connect them in series or in parallel. Every lane, junction and roundabout on its way to the final EU arena amounts to a prelude arena. 10 Negotiation. A classical dilemma is to negotiate by stick or by carrot. Modern wisdom deems it best to show the carrot and to keep the stick covered within reach. Withholding the carrot is the first stick, and punishment by the real stick the ultimate one. In the game of giving and taking, another dilemma is overcharging or underbidding or going straight to a fair deal. English negotiators often show disdain for those who cut straight to the chase. Related is the dilemma whether to persist in one’s position or to enter into a compromise. If the Council decides by unanimity, some governments sometimes play the game of reluctance until ‘five to midnight’, as part of a loss-compenzation game. In 2001, the Spanish government held up the Council decision regard251

Managing the EU Fieldwork

Chapter 6

ing the timing of enlargement as means to keep its subsidies from the Cohesion Fund. Another dilemma is whether to participate or, for blocking the process, to be absent. The latter choice is highly risky as the other participants can continue with an easier formation of consensus. 11 Decision. Regarding an EU decision at issue, the interest group can choose to accept the outcome or to appeal to, for example, the Court. A milder variant is to make use of the review clause that is part of many an EU law and thus to push an agenda for retaliation. A dilemma at home is whether to adapt new EU laws or acts fully or to allow some escapes, as often happens with directives (needing domestic implementation) and in multi-layered countries such as Germany, Spain and Italy. Another dilemma is whether to submit to the secondary decision or to undermine its implementation at the EU level (comitology) or at home. The costs and risks of appeal, escapes and undermining have become, however, high and legitimate the EU as a system ruled by law. 12 Lobby styles. Standard dilemmas in styling a lobby on the playing-field are whether to behave in a supplying or demanding way, low-key or high-key, direct or indirect, formal or informal, defensive or offensive, confrontational or appeasing, reactive or proactive and so on. The question of style is widely considered to be crucial. It is, in fact, the only one that is common to both the amateurish and the professional groups (figure 3.3), although their answers differ, as we shall see at length below. The theoretical dilemma over lobbying in a legitimate or illegitimate way is hardly one in practise. Every interest group should fear and thus avoid the fall-out from behaviour that is widely considered illegitimate and usually treated as scandalous.

III Coping with the Surplus The most popular question posed by the dilemmas of EU fieldwork is, ‘What is the best thing to do?’ The following five types of answers can be heard or seen in the world of PAM in EU. 1 By reflex. Amateurish groups tend to lobby by spontaneous reflex. Inspired by emotional feelings about an EU challenge, they go straight to an EU high official of their own tongue and present their case there in a prolix, loud and selfish way. They are FiFo lobbyists: flying in, flying out. Or they visit the national capital or association and try the same approach there. Driven by their emotions, they act almost Pavlovian or at random. Unfamiliar with (as we characterized before) the EU labyrinth constructed on a trampoline, they miss the many accessible back doors 252

Chapter 6

2

3

4

5

Managing the EU Fieldwork

and better moments. The number of dilemmas they consider to be a problem rather than a blessing. Rules of thumb. Many manuals on EU lobbying keep it simple too [Hardacre, 2011; Venables, 2007; Guéguin, 2007; Randall, 1996; Anderson, 1992; Gardner, 1991]. They take the EU to be a pyramid of dominant institutions and high officials connected by procedures, and with the work floors and stakeholders at a distance. They are formalistic and belong to the first wave of the study of PAM in the EU that is focused on rulers. In addition, they give a number of rules of thumb, such as ‘be to the point and on time’, ‘keep it low-key’ and ‘avoid over-lobbying’. Such general lobby caveats make some sense, of course, sometimes. At other moments, however, the opposites may be more effective, as we anticipated in chapter 4-IX (‘better tactics’) and shall see below. The manuals neglect the dilemmas. Receivers. According to the small amount of field research that has been carried out, many Commission officials [Burson-Marsteller, 2003 and 2005; Koeppl, 2001] and MEPs [Burson-Marsteller, 2001; Kohler-Koch, 1998; Katz and Wessels, 1999] want to be approached in a low-key, informative and friendly way at an early stage and not, as frequently happens, the opposite. Their consumer demands come close to the rules of thumb above. What is good for the receiver is, besides, not necessarily good for the sender, as the former wants to be in the driver’s seat and the latter to achieve a lobby target. Predictably, many MEPs say it is better to approach them than to try the Commission officials. These findings coming from limited research do not distinguish between specific situations and neglect most dilemmas mentioned above. Consultants. Commercial consultants [Kohler-Koch, 1998] show strong and barely varying opinions about the best ways and moments for lobbying the Commission and the EP. Almost everything submitted to them they consider highly important. This gives at least the suspicion of commercial self-interest. Of the highest importance they rank the informal contacts with Commission officials during their preparation of a proposal and with the EP’s rapporteur during the writing of the report, limited to only the secondary process. Other stakeholders and lobby styles remain unquestioned in this piece of research, let alone the many other dilemmas. ‘It depends’. The short answer of the thoughtful practitioners in EU lobbying to the question, ‘What is the best thing to do?’ is: it depends on the situation in the arena. In line with chapter 4 it is, of course, our position that, for meeting the yardsticks of 2E, a PA action must always be fine-tuned to the specific situation. The amateurish interest group usually feels unhappy with the surplus of options, as it must choose with253

Managing the EU Fieldwork

Chapter 6

out knowing how. The semi-professional at least takes account of the formal settings, officials and procedures (‘skeleton’) of the various decision phases and goes step-by-step. The real professional bears in mind what happens behind that (‘flesh-and-blood’), the stakeholders around and the PA thinking (methodology) for making choices. It considers the surplus of dilemmas to be a blessing rather than a problem. The higher their number, the more options it sees for solving or saving a challenge. Seeing small chances, it usually knows how to enlarge them. Thanks to the surplus of dilemmas and variables, being richer than in the game of chess, it can prevent or solve stalemates and setbacks. Its lobbying is neither at random nor based on rules of thumb, nor focused on pleasing the receivers or consultants, but on a choice that is as rational as possible in a situation. The professional group has, of course, no guarantee of influence success. As said before (chapters 4-VI and 5-VI), precise predictions of the effects of its actions are impossible for two reasons. One is that any effect can be full of multi-causality (same effect coming from different actions) and multi-finality (same action producing different effects), often with weak correlations. The other reason is that much is beyond the reach of complete, valid and reliable calculation. In terms of its fieldwork, the intelligent group will thus not search for the maximal best choice, as this goal is seen to be utopian and usually comes too late, but for the optimal best choice. Only in a crisis situation of mere survival it may strive for the maximal choice by putting all its available energy on this one alone, but even then the result remains optimal, but at a higher level. Full control over the chain effects of its field activities is, for the same reasons, also impossible. The professional group thus takes the complexity and dynamics of any EU arena as facts of life that require permanent study and prudence. It may even feel confident enough to increase the complexity and dynamics, for example by creating new vectors and by crossing arena boundaries. It considers the absence of guaranteed success as something impossible to solve and thus not to be a problem. This optimal choice is, again, an educated guess of the probable outcomes of possible field actions, given the specific situation. Although to the best of possible knowledge, it is made under much uncertainty about the real outcomes and as such is an example of the ‘science of muddlingthrough’ [Lindblom, 1959]. The ambitious interest group needs for this, next to engaging in a lot of preparatory study, to exercise prudence during the fieldwork, as this enables it to make all sorts of detailed corrections, like when driving a car on a slippery road. The prudence is based on solid knowledge about the playing-field and the variables, for example that it is 254

Chapter 6

Managing the EU Fieldwork

easier to move from a soft approach (informal, silent, by carrot) to the hard opposite than the other way around. It is applied by means of the PA methods mentioned before, such as regarding proactive window-out monitoring, the arena assessment and the home front’s organization. By such methodical prudence the interest group can come to a reasoned estimate of the costs and benefits of whatever choice it makes (and its probable effects), and even consider two smart games of fieldwork. One is combining the two extreme options of a particular dilemma, for example the use of both stick and carrot, which requires precise engineering for different phases, stakeholders, issues or whatever variables. Another game is faking the choice of an option by indicating an intention to, for example, join another coalition or ‘go to the Court if this deal is all there is’. These games have, of course, their own costs and benefits that require methodical prudence. Every interest group is, of course, free to decide that choosing its actions by reflex, rules of thumb, pleasing the receivers or following the consultants is no less rational than doing it by educated guess. It then saves the costs of its own preparatory work. The chance of achieving a desired target is, however, low to zero in practise. The group that acts by reflex becomes dependent on sheer good luck, on crumbs from the table. Acting by rules of thumb, it shall make (like an amateurish tennis player) many unforced errors. If led by pleasing receivers or consultants, it shall make them happier but probably not itself. Our approach remains advisory and not normative. There is no ultimate argument for why a group should not be allowed to do all this instead of making a rational choice according to the best of its knowledge. However, if the interest group seriously wants to solve or to save a challenge, then it should ground its actions on the most rational choice as possible, as this allows saving on the costs of otherwise inefficient and/or ineffective fieldwork. We shall now demonstrate this for the recurrent questions of the fieldwork, summarized earlier (figure 3.3) as ‘how to lobby whom, on what, where and when?’

IV Lobbying Whom? At any moment and on any dossier, every group can go to hundreds of stakeholders, including the officials having a stake in the dossier. However, no group can approach all of them in series, let alone parallel, and it should not want to do so, as maximal is the enemy of optimal. In its personalized lobbying it can better act selectively. The best cues come from the homework, particularly the data file of the specific arena (figure 4.3). If the wind is blowing friendly, the group can confine its efforts to keep255

Managing the EU Fieldwork

Chapter 6

ing the supporting stakeholders on board and to arranging a free ride. If, on the other hand, the wind is unfriendly, then it is better to try to divide a group of opponents and to seduce some waverers. The best opponents to approach are the ones assessed as primary (influential plus active) and as sharing the smallest number of preferences with other opponents. The most promising waverers are, by analogy, those having both primary status and high ambivalence. If the winds are absent or circulating and thus make the arena indeterminate, then ‘waiting and seeing’ is more rational. In the meantime, the interest group should prepare scenarios for any wind that may arise, place Triple P pickets, intensify its monitoring and offer to a few primary stakeholders some arguments that match their interests in order to test their responses. The group can read their names from the vertical axis in figure 4.3. It might do the same for other dossiers on its shared-list and short-list. Usually it will discover that some of the same stakeholders are involved in some of these dossiers as well, have useful relationships with even more stakeholders and EU officials, and thus can be mobilized indirectly. Among the many information sources (figure 4.2) it can use EuroFed(s) as an easy shopping centre for names, contacts and more, in order to get its ‘intelligent laziness’ organized. Stakeholders that take friendly positions on several dossiers deserve an intimate courtship. Those having mixed positions may be made interested in some compromise. Strong opponents can be tackled, preferably indirectly upon their life-lines of, for example, support and budget. The interest group can also work the other way around and identify dossiers not being on its long-list but of great interest to primary stakeholders on their long-list. Dependent on their position on the latter, it can support or tackle them on these dossiers, so repaying them almost free of costs. Next to these stakeholders, two other categories of people need extra attention. The first includes the staff and line people on the home front who have a special expertise, status or network that is relevant to primary stakeholders, as shown by the data file. Some may be connected to EuroFeds and in EU expert groups, committees or otherwise and be used for liaising the external network, on which they only need briefing and debriefing. In the MNO model of PAM they become part of the ad hoc team. If they are absent at home, the proactive group can create and position them in time inside or nearby EU platforms and befriended stakeholders. The second category includes the people working inside the EU machinery. Acting under some functional and personal pressure, they need specific information and support, and the more so if they are new to the issue. After any election the majority of MEPs are new and usually inexperienced. Commission officials circulate to a different directorate or 256

Chapter 6

Managing the EU Fieldwork

upwards. Experts sit temporarily in committees and groups. The sooner the interest group can serve them with information and support, the more they will be able to get an edge over their competitors. The answer to the question ‘lobbying whom?’ thus depends on the specific situation. If the interest group has knowledge about this, it can benefit greatly from the checklist in figure 6.1 by considering various more dilemmas. Let’s look at three examples. First, if it has found a number of new stakeholders to be often friendly, it can promote joint action, such as a Triple P game for safeguarding common interests. Second, if it has identified some stakeholders as frequent opponents, it can attack them in their own area. For example, northern agricultural interest groups that seldom interfere in EU olive committees of high relevance to southern groups that hinder them on subsidies for milk or R&D, might by interfering as yet create an excellent trade-off at a low cost to their own interests. Third, if it has found that on a dossier only a few people of the relevant Commission DG are on its side, then it can delay the process by lobbying them for retarding a deadline. To make all this personalized lobbying possible, the interest group must keep its EU data file updated and within close reach.

V Lobbying on What? No interest group can tackle all the issues in all its dossiers at the same time. It has to be selective. For its specific issue management the professional group heavily relies on its data file on any arena (figure 4.3), which provides clues about the better tactics to use. If, for example, the arena is essentially friendly, it knows that it must keep the favourable issues high up and block opposing ones. From the cells of the file it reads which friendly issues deserve supportive evidence and which unfriendly ones maybe counter-evidence. It can even personalize the mailing of messages, as it knows from the file which stakeholders are crucial on either its own or the opposing side. If the arena is mainly unfriendly, it can glean from the file on which issues and to whom it might lobby for loss-compenzation. The file can also reveal which unfriendly issues are most suitable for up-framing to an attractive Lorelei and on which issues it should lobby for narrowing or widening the arena boundaries so that it can get rid of opponents. In an indeterminate arena the data file reveals on which issues and related to whom it has to intensify its monitoring, create a reframing or try a balloon. If the interest group has prepared itself well, it can even go beyond the boundary of an issue, take one as the Achilles’ heel of an opponent, as every player has (the columns at right of figure 4.3), and provide free 257

Managing the EU Fieldwork

Chapter 6

sticks or carrots. An example of such offensive lobbying is the aforementioned case of EU committees on olives, in which oneself may not have an interest but in which an opponent may. In 2001, UK pharmaceutical companies that were irritated about (British) Oxfam’s politicization of their price-setting of AIDS drugs for South Africa, considered from their side to fan through local retailers the issue of Oxfam’s local trade (as BINGO) that was, to the irritation of these retailers, largely free from local taxes. The enemy of one’s enemy can be used indirectly as a henchman. The pharmaceutical companies decided, finally, to compromise on the issue of AIDS prices directly with both the EU and the South African government. Making a virtue out of necessity, they offered cheap pricing in return for EU approvals of new drugs and WTO agreements, so even improving their public image. Excellent homework can pay its way. The interest group can also practise meta-lobbying that deals with the actors, factors and vectors affecting an issue. The checklist of dilemmas can again be a source of inspiration. For example, the group can decide to challenge the position of an opponent, to join another coalition or to lobby for a different procedure. Such actions may change the issues at stake. More specific is the meta-game of Triple P, intended to rearrange the playing-field and to make the dominant issues friendlier. The group can get clues for this meta-lobbying from its data file, if including different arenas and general developments. For example, during the secondary procedure on the establishment of the European Food Safety Authority (EFSA) in 2002 the debate was mainly about its layout thereafter and hardly about the substance of food safety. The latter usually causes most competition of interests and can, from lobby perspective, better be moved to later on and, as happened, be settled under delegated and implementing procedures, positions and people that were managed by DG SANCO and influenced by proactively lobbying ministries, producers and retailers in the food sector. If the arena remains unfriendly or indeterminate, the interest group may have to return to its preparatory work regarding its home front. Maybe it has engaged in a mission impossible. Disappointing field experiences are useful input for checking the reliability, validity and completeness of the homework done. The group may have to conclude that it has to redefine its challenges by reassessing the facts and norms involved. In the most extreme case, it may have to conclude that its chances of winning something are close to zero or only exist at a disadvantageous cost-benefit ratio. If the dossier has been on its short-list, it can more usefully be moved to its shared-list or, if it was already put on this, left to the winds, as Unilever did in 2000 with its GMO challenge, after which it spent its saved energy on the more promising dossier of the one-stop EFSA.

258

Chapter 6

Managing the EU Fieldwork

VI How to Style One’s Lobbying Better? The question of ‘how to lobby’ is the oldest one on the subject, going back to the days when lobbying was confined to political corridors and antechambers. It is the only question that is posed by both amateurish and professional groups. No group lobbying for a desired outcome can escape it. The answers ultimately shape and style its behaviour. Specific dilemmas are whether it is better to be supplying or demanding, formal or informal, and direct or indirect. Very topical is the dilemma of whether it should act silently or noisily. Of course, it can always behave at random, follow some rules of thumb, simply please officials or act on consultants’ advice. All this would make most preparation indeed superfluous. The following shows that, if wanting to achieve targets, one should solve those dilemmas rationally, based on homework.

1 Supplying or Demanding Without a supply side no interest group can get attention and interest from stakeholders, and without a demand side it cannot get its interests included in the outcome [Hauser, 2011]. As far as a rule of thumb makes sense, the group should be more aware of its supply than demand side, as by almost natural instinct it usually has the latter so much in the forefront of its mind that it must try hard not to forget its supply side. Besides, when entering an arena, it must be welcomed, which is achieved more easily with charming and offering than demanding behaviour. Investing in its image of charm, a professional group wants to be seen as interesting, pleasant, friendly and trustworthy. It plays courtesy and diplomacy, as the EU arena is as risky on fortune and fate as the ruler’s court [Castiglione, 1528] or the stage of international politics. Behind its face of charm and its hands full of carrots, it hides its selfish interests and its sticks for better or emergent moments. As tone makes the melody, it up-frames any objection to a friendly ‘of course yes, but’. Window-out, it shows interest in the wellbeing of others, so trying to get information about their problems and agendas as this gives the best input for developing a supply side, which may solve their problems without hurting its own interests. Most stakeholders and EU officials usually desire expertise and support to satisfy their needs. The art of lobbying, embedded in professional PA, is to discover their appetite precisely. What they say they need, gives, if reliable, clues about their strengths and weaknesses. All this charming behaviour lays the ground for the crucial window-in practises, which must result in achieving the deals as desired. The rational group, in short, tries to satisfy its demand side by pushing 259

Managing the EU Fieldwork

Chapter 6

first its supply side. A good deal finally links up one’s own demand and supply sides with those of the primary stakeholders and officials. This friendly-looking style of lobbying comes close to political marketing, albeit not for mass markets but for niche markets like EU arenas [Baines and Egan, 2004; Dermody and Wring, 2001; Harris and others, 2000; Andrews, 1996; O’Shaughnessy, 1990]. Here too, desired outcomes such as sales and profit cannot just be imposed, tied-in, advertised or talked-up. Even for the current mass markets, they have to be prepared by careful research into what consumers may want and what may satisfy them. An EU arena is usually a niche market, with a lot of open competition among multiple stakeholders and with rounds of wheeling and dealing, eventually resulting in a sufficient consensus. As an aside here: organizations that have excellent PA expertise probably also excel in their commercial market or policy domain (and vice versa), as they behave in a substantially extroverted manner [Bennedsen and others, 2011]. Marketing techniques can help to improve one’s supply side, ranging from branding and merchandising to export licensing and ‘after sales’. Their parallels in the EU arena are reputation building, displaying attractive items, appointing an intermediary, and servicing partners after a deal. On the latter, lessons can be taken from former MEPs who frequently complain that, from the date of their departure onward, they are forgotten by many previously most friendly lobbyists. By telling this to their successors they silently punish these lobbyists for being amateurish. The data file provides the best possible clues for the design of a supply side. Contrary to a popular rule of thumb, an interest group can sometimes, particularly in a largely friendly arena (a buyers’ market), even demand more than it supplies. Its studious homework gives information on which issues it can take a free ride and from whom it can ask more. In a most unfriendly situation it might decide rationally, contrary to a rule of thumb again, to overcharge its demand side in order to create nuisance value, get more loss-compenzation or delay the process. Such aggressive lobbying it has to fine-tune to the proper issues and stakeholders, which it can distil from its data file. In a mainly indeterminate situation it has no reason to overcharge, because it still has to build up its position. A prudent group fine-tunes its style while keeping in view both its other interests and the near future, as it always fears possible boomerangs. If it overdemands according to crucial stakeholders or officials, it may get the bad image of a greedy player and be punished for it. If it oversupplies, it may be seen as a softie and be treated as such. It is all a matter of studious and prudent lobbying.

260

Chapter 6

Managing the EU Fieldwork

2 Formal or Informal Many interest groups tend to take the skeleton of the decision process as the real process and thus to behave formally. Focusing on the desired endresult of the lobbying process, they look for the signature of some high official or authority, such as the chairman of a EuroFed, a Commissioner or the Council, and approach them formally by way of an official letter sent by their chairman. Many EuroFeds and national ministries hold the formal belief that for getting a desired outcome, the formal support from EU officials, from the beginning to the end of the process, is both needed and even sufficient. They are amateurish, as neither is true. Many decision processes start at some lower level or even outside the EU machinery, and proceed upwards along higher officials that put comments in the margins and usually sign. If their desired outcome is for no decision to be taken at all, then approval is not needed either. If the desired approval has been granted, the required measures can still fail to be taken or implemented. A formal decision can produce the decided outcome with time, but this is never certain. The EU open market for labour and capital was decided by the 1987 SEA, but is still developing more than two and a half decades later. Those having opposed that decision have lost the treaty game but not (yet) their real interest. In at least the following four situations, the formal style of lobbying is even counter-productive: – Wanting no decision. The interest group may want to prevent any decision. This is a matter of blocking and delaying the dossier’s life cycle (figure 4.1), which usually can best be achieved by informal and semiformal techniques. – Going window-out. If the interest group has to go window-out, it usually gathers crucial information better through an informal chat or a semiformal meeting than by a formal demand. – Work floors. Usually more relevant than the boardroom are the work floors, where the mid-level experts, clerks and other apparatchiks can often best be approached in non-formal ways, as they are human beings with some discretionary powers. – Starting window-in. If the interest group wants to make a window-in deal with a supportive, opposing or wavering stakeholder or official, it can best start informally, as otherwise there will be no room to manoeuvre. The last case applies particularly to the situation in which formal approval is really needed. This should be built up step by step, from informal to semi-formal and finally formal. It works by creating trust. Informally, one 261

Managing the EU Fieldwork

Chapter 6

gathers the best possible information and network. Efficient tools for this first step are the quiet restaurant, a pleasant outing or social events, all framed by an attitude of respect. This way the informal relationship can be based on a common interest (‘something in it for both of us’) that engenders trust in its most solid form, as loyalty to the other become a selfinterest. The informal understanding may then get a semi-formal followup, for example by an ad hoc group, platform or working group. So far the prudent interest group keeps its position paper in its head or luggage and never puts it on the table, as otherwise it cannot negotiate (‘give and take’) and may lose face. Long ago the Commission often published its proposals straightaway, but now it usually issues first tentative (blue, green and white) papers for collecting views and positions while hiding its own position. The third step is to get the semi-formal position formally approved, which after so many rounds of ever better trust, understanding and negotiation is usually a matter of formality. From its prepared data file, the interest group can read which stakeholders it should approach informally and on which issues. To this it adds useful information about their personal idiosyncratic preferences, such as walking in a park, having a drink or visiting the theatre. The office of the German region of Bavaria once had on hand a private plane to take people to, for example, a Munich football match. One stakeholder or official likes to talk about family or career, another about politics or philosophy. One takes familiarity as a sign of friendship, another takes it as impoliteness or even as contempt. One loves a phone call, another sees it as disruptive. High officials tend to open envelopes only if hand-written and leave preprinted ones to their staff. Some people dislike being approached informally at all or allow this behaviour only after a formal start. Professional lobbyists highly value information about such idiosyncrasies and share it with befriended colleagues, as it greatly helps to create an informal and trusted relationship. Some consultants even offer it for sale.

3 Direct or Indirect Direct ways of lobbying are those that go straight from the interest group to the targeted audience, be it a person, committee, platform or wider public. Indirect ways are those by which the group makes use of U-turns as medium for its message. ‘Indirect lobbying’ is, strictly, a contradiction in terms, as the original or medieval meaning of ‘lobbying’ (corridor behaviour) is face-to-face or direct communication. The menu of both direct and indirect ways to lobby is richly varied, as shown in figure 6.3, which presents a selection drawn from aforementioned case and field studies and ordered after intimate, formal and noisy variants. 262

Chapter 6

Managing the EU Fieldwork

Examples of Direct and Indirect EU Lobbying Direct Ways

Personal visit, letter, phone, email Social invitation Committee membership Hearing participation Presentation of position Formal visit, contact, delegation Formal request, petition Folder or brochure Mass media participation Political advertisement Press conference Manifestation, demonstration Love-site, thanksgiving, faming Hate-site, boycott, strike, shaming Litigation, Court procedure

Indirect Ways

(Sub)national association (Sub)national government (Cross-)sectoral (Euro)Fed Foreign (non-EU) network Ad hoc coalition Affiliated interest groups Science: scientists, studies, seminars Working visits, trips, tours Well-known personalities Mid-level civil servants Caretakers and friends inside Brokers and consultants Cyber-lobbying Political parties Mass media mobilization, polls Undercover action

Figure 6.3

263

Managing the EU Fieldwork

Chapter 6

The direct ways show great variety. Some are highly informal and personal, such as the face-to-face and ear-to-ear (‘lobby by body’) personal visit, phone call and invitation to the opera. Others are more formal and functional, such as going to a meeting or a hearing, presenting a position paper or sending an official letter or delegation to a stakeholder or official. Some direct ways are low-key or silent, for example informal, face-to-face meetings behind closed doors. Others are high-key or loud, as they make use of public platforms, such as an advertisement, press conference or public demonstration. Some direct approaches are friendly, such as the invitation for pleasure trips, the launch of a brochure or a public thanksgiving to a group of MEPs. The lobby group for Animal Welfare did the latter shortly after the EP’s ‘improvement’ of the directive on Animal Testing in April 2001. Others are more confrontational, for example the hate-site, boycott and litigation. A direct way often has, by choice or effect, a mixture of characteristics. The choice for a press conference is automatically one for a formal, loud and possibly confrontational way. The indirect ways also have many variants, ranging from informal to formal, low-key to high-key and appeasing to confrontational. Widely used U-turns are domestic platforms, EuroFeds, think-tanks and foreign (nonEU) platforms like the WTO. Underused as a U-turn but potentially highly effective is the OECD in Paris, which is a high-expert producer of definitions, standards and methods consumed (and often ordered) by the Commission as input for its policy framing [Dostal, 2004]. Ad hoc coalitions are also useful for broader-based and fine-tuned communication. Scientific studies, scholars and ‘wise men’, if looking independent, give an aura of authority to the sender and the message, until they are seen as faked ‘junk science’. Seminars under the flag of a Commission DG or EP committee may have similar effects. Popular personalities may be useful, such as the Spice Girls, who effectively demonstrated in the EP against the 1998 draft Copyright Directive, much to the displeasure of the music producers. People inside the system, like mid-level civil servants and friends, are excellent for helping to plug a message. Political groups can be used for the politicization of an issue. Many groups use mass media for disseminating information off the record and for promoting an EU issue or agenda. Some (wealthier) interest groups, such as ministries and MNOs, order public opinion polls that have the image of being ‘the voice of the people’, and they publish the results if supportive and not if damaging to their interest. An interest group can create a new U-turn by, for example, parachuting a friend into a useful position (part of Triple P) and other undercover practises. Highly ingenious is the interconnection, in series or in parallel, of various U-turns. In this way the interest group can arrange a polyphonic orchestra that sends the same message through different sound boxes to 264

Chapter 6

Managing the EU Fieldwork

the targeted audience that so becomes easier convinced. A trend-setting example was provided in 1986 by consumer electronics companies like Philips and Thomson, when lobbying for their high-definition TV (HDTV) [Verwey, 1994; Cawson, 1995]. Through their different country units they, experts in political engineering as well, approached various national ministries, trade associations and other stakeholders in parallel setting (simultaneously) and connected in series (consecutively) the key people of these organizations to the relevant people (work floors) inside EuroFeds, Commission, EP and Council. In 1995, DG ENVI, in cooperation with both national ministries and NGOs in its field, campaigned similarly in favour of sustainability. Politicians can easily be used for such interconnections, as they belong to party-political families that connect local, regional and national levels to the EP, the COR and the Council officially and often to other EU institutions unofficially. The surplus of so many different ways of lobbying is a richness that makes a reasoned choice possible and necessary, as every way has its own costs and benefits and also its own probable effects. Efficiency and effectiveness are generally used criterions for choosing. There is no rationality in applying rules of thumb and the like. In their daily lobbying, many interest groups show a bias in favour of direct ways. The reasons for this can be many, for example lack of capacity (such as expertise, relationships and resources) to manage indirect ways, or lack of time due to their late arrival in the process. Indirect ways require, indeed, investments made in the past and early preparations of games to come. An ad hoc group of stakeholders or scientists may require half a year of formation and two years before becoming effective. Some other risks or costs of a U-turn may be the return to be paid, disclosure of the sender, distorted transmission and lack of visibility on the home front. Finding the best balance of costs and benefits of going either directly or indirectly is a matter of studious consideration and contributes to the prudence of the choice. Often the costs of indirect ways can be lowered and their rewards increased, as the following shows: – Intermediaries helpful. If one has to lobby for information or support from an unfamiliar stakeholder or official, one may benefit from intermediaries (friends, PA colleagues), who can give clues on how to approach or make the connection. – Hands free. By going through another group one can gain invisibility and thus free hands for the moment. For example, launching a balloon in an indeterminate arena one does better from another platform than one’s own. – Generalized interest. By sending the same message through different channels, targeted receivers may take it as more convincing and repre265

Managing the EU Fieldwork

Chapter 6

senting a more general interest, on which product BONGOs and GONGOs are often experts. – Less backlash. If opposition arises, this does not immediately strike back at the real sender but only on the U-turn. In case of disclosure the sender can prepare its response in the meantime. The intelligent interest group selects from the menu the ways that, given its specific arena situation, can contribute to efficient and effective success of its final targets. If the situation allows, it usually prefers indirect rather than direct ways. One reason is that it gets from indirect ways, thanks to their lower visibility, better chances of success and another that, as case of prudence, it can easier move from indirect to direct than the reverse. Also acting out of prudence it usually prefers an informal rather than a formal path to follow, let alone one that draws much attention. It is, however, ‘the arena situation’ that should determine the decision to act directly or indirectly. If the interest group finds itself in a friendly arena, it can, for example, make direct informal visits to the friends it has identified and, after having secured their support, together engineer some U-turns for producing messages that keep the wind blowing favourably. In a chilly arena it can make use of, for example, scientific reports for reframing and mass media for up-framing the unpleasant issues. In an indeterminate arena, as is usually the case in an early phase of the dossier, it should invest in building up a strong platform or getting prestigious scientists on its side in order to impress wavering stakeholders and officials. Good preparatory study informs about the most efficient and effective ways to influence primary stakeholders and officials, to tap or inspire their networks, to encircle them with polyphonic sound and/or to deliver support in ways they like. The fieldwork can then be engineered backwards from the targeted receiver(s) to the interest group. This requires a lot of fine-tuning as otherwise the end-result may become a costly mess.

VII The Fine-Tuning of ‘Sound’ A challenging part of the big question of ‘how to lobby?’ regards the use of silence or noise in public, in short the sound button. As a variable on a continuum, the volume controlled by the button can go up or down. Many interest groups can often hardly resist to produce a lot of public noise on the EU playing-field. When feeling weak, they want to attract attention and if strong, to show their greatness. If seeing something they desire or fear, they easily voice their pleasure or anger. Such emotional noise is soon noticed by the numerous stakeholders and journalists that make Brussels 266

Chapter 6

Managing the EU Fieldwork

a transparent village. National ministries, business associations and NGOs cannot easily avoid publicity as they are often called by their parliament or members to account in public about their plans, actions and results. In short, many groups more readily produce noise than silence. Noise around an interest group can, however, also be the product of befriended stakeholders unable to avoid it or opponents who provoke it. A group may also find reason to engineer controlled noise. An example is the limited soundtrack directed at a selected audience of, for example, a EuroFed, EP committee or Commission unit. Some other groups want to create the broadest possible publicity or make intense use of a hate-site (with blackmail and guerrilla variants), press bulletin or advertisement. The most useful mass media for reaching people in the EP, Commission and Council are the few widely consumed UK media: the BBC, the Financial Times and the Economist [ComRes, 2011]. This outside lobbying by way of publicity campaigns comes mainly from grass-root NGOs and newly arriving US interest groups, such as (in the past) Philip Morris, Monsanto and the AmCham. In the EU, the amount of outside lobbying is, compared to the US, remarkably low [Lowery, 2008; McGrath, 2005-A; Thomas, 2004]. The explanation can hardly be a difference of pluralism in society and is largely related to three major differences between the two political systems. One is that the EU’s main product are laws and acts that by their nature (technical, time-consuming) are hardly suitable for publicity campaigns, while its budget is too small for much outside lobbying (except on its main expenditure: agriculture). In contrast, the US government has a wider variety and size of main products. Secondly, in the EU the Commission is almost exclusively the drafter of laws and acts, has only nominated staff and offers most interest groups direct or indirect positions inside (expert groups, comitology) and thus less reason to lobby outside. In contrast, in the US the Congress initiates legislation too, the administration hardly offers such positions to interest groups and both the legislators and some crucial administrators (recruited by ‘spoils’) have vulnerable grass-roots in their constituency and thus can easily be influenced by lobbying the voters outside. Thirdly, in the US the Court procedure takes place in public and offers a conventional direct way of lobbying, while in the EU the Court proceeds without mass media. The three differences with the US are, however, much smaller at the EU’s member-state level, where all parliaments can initiate laws, the elections usually go by some form of constituency and some Courts are more open to the public, thus making outside lobbying more usual than at EU level but still less common than in the US. In the past, many US interest groups entered Brussels poorly prepared for the much different EU. Being more silent now, they have apparently learnt a lesson. 267

Managing the EU Fieldwork

Chapter 6

The main reason why the PA management of sound is largely silence management comes from the logic of influencing. The window-out information lobby goes best by listening and watching, as information enters via the ears and eyes and never the mouth. As easy as it is to create noise, it is just as difficult to return the ghost of noise back into its bottle, to keep its effects under control, and to restore silence, if needed. Noise is, in short, technically an almost one-directional variable of manipulation. In the window-in support lobby, the main effects of public sound are usually that the arena gets wider boundaries, more stakeholders and issues enter and the making of compromises and decisions slows down, which is only rational in unfriendly arenas. Silence has the opposite effects of narrower arena boundaries and a faster decision process. As control freaks, professional interest groups usually find silence safer than sound, want to remain below their opponents’ radar and prefer the intimate meetings and networks held in conference rooms of offices, hotels and restaurants. Negotiating a common interest out of different ones is more easily achieved in silence than noise. More silence can, if needed, be created by, for example, applying Triple P, making a dossier more technical (‘low politics’) or getting one’s people on the work floor. However, once the outcome or decision breaks the surface like a submarine, the silent activities must stand up to daylight (chapter 7) and prudent groups behave accordingly. This issue of the sound button also explains why, in the MNO model of PAM in the EU, a member of the Central Board is responsible for its use, which is often at issue between the units of Public Relations and Public Affairs. The PR office may have a vested interest in making a lot of impressive sound, which yet remains sound, but the PA office may have strong reasons to be firmly against this and to prefer silence. The decision to turn the sound button up or down is then to this Central Board member, who has to solve the dilemma of whether sound or silence serves the interests of the organization the most. The decision to choose sound may damage its interests in the EU but promote them on the home front, the market or elsewhere, and choosing silence may have opposite effects. Sometimes the risks of choosing the one or the other choice are kept limited by internal compromise, for example on the timing or the contents of the sound, but even then the risk of undesired chain effects lies in wait. Avoiding the ghost of sound looks to be a rational rule of thumb, but this is not true for every specific situation. One is the unfriendly arena. If an interest group fears that a desired proposal will not pass one of the thresholds of the decision process (figure 4.1) or that an undesired one will pass the next threshold, it may feel its efforts to be doomed and rationally decide to cause some noise, preferably indirectly. If the often aggressive sound produced by some NGOs in the EU is rational indeed, then it may 268

Chapter 6

Managing the EU Fieldwork

imply that they have problems with thresholds; if not, they take high risks and may make mistakes in lobbying. Making noise is, however, no guarantee of a stay of execution or a recovery. The crowds may love to see a loser put to death, as did the tricoteuses when Marie Antoinette was executed on the Place de la Concorde in 1793. Making identifiable noises might even be worsening. If an interest group feels close to ‘hanging’ and wants to escape by noise, then it should up-frame its message and use only indirect media that hide its identity. Also in a friendly arena it may sometimes be better to apply some noise rather than full silence. The friendly issues have to be kept buoyant by orchestrating supportive sound indirectly and with up-framed contents, for example by providing new scientific evidence that credibly frames the desired outcome as positive for a high social value that benefits many and drives out counter-issues. The arena may then be won over for a speed-up, as done by the 1985 Round Table of Industrialists regarding the Open Market by 1990 (it became 1992). In the indeterminate arena, making a little bit of noise by trying a balloon or calling for support may also be rational, as the Commission does with its early papers and calls. Indifferent and wavering stakeholders may be put at ease and won over by some publicity and if they are dispersed over many countries or sectors, the interest group may have no alternative to using public channels and sound. In the paragraph on ‘supplying or demanding’ above we referred to some useful insights from political marketing for the niche markets of EU arenas. To this we can add at least four general uses of sound, which are not related to a specific arena and can be handled jointly by the units of PR and PA [Johnson, 2008]: – Trustworthiness. On this part of its general reputation every interest group is heavily dependent as otherwise it risks to lose its licence to lobby, which is ultimately granted or withheld by the mass public. Similar to ‘corporate imaging’ in marketing, the group can improve the public perception of its trustworthiness by disseminating evidence of its honest and sincere behaviour. This image management it should not carry out itself but, indirectly and unidentifiably, leave to others that look impartial. – Public thanksgivings. MEPs, ministers and high officials are usually sensitive to lyric applause in the mass media. An interest group should apply this negotiating chip only after having achieved the desired outcome and only in general terms unrelated to its specific interests, as the Eurogroup for Animal Welfare did in 2001. This is akin to ‘after sales’ in marketing and it is often (incorrectly) forgotten in EU lobbying. – Early up-framing. A proactive group may want to give an early up-frame to its new challenges and interests in order to get them well-placed at 269

Managing the EU Fieldwork

Chapter 6

the start of new EU agendas. For influencing public opinion, its sound must use language that is credibly linked to popular concerns, such as ‘sustainability’, ‘welfare’ or ‘security’ in the recent past and ‘stability’ and ‘recovery’ now. This is akin to ‘product positioning’ in marketing. – Sound for silence. In order to divert the attention of competitors away from their real interests, the interest group may rationally make noise about something else, for example its views on the distant future or a foreign area, just as in marketing it helps to launch publicly different vistas that hide what is in the current pipelines. There is always some room for theatrics, but the group must remain credible and thus be a professional player that is not unmasked as an amateurish comedian.

1 Tone Makes the Melody In the case of sound it is often the tone that makes the melody. In intimate meetings with stakeholders and officials, the prudent group knows how to strike a melodious note. It wants to be seen as charming, to show a sense of humour and to avoid any direct arguing over who is right or wrong, as this could easily result in a zero-sum game [McGrath, 2007; Warleigh and Fairbrass, 2002]. When some argumentation is asked (chapter 1-IX), it says not what it wants to say, but what the other wants to hear or accepts. Its language is not intellectual speak but salesman’s talk, adapted to a market where both sides want to become happier (variable-sum game). When it finds that this soft approach is not working well, it may decide to make its tone gradually more hostile. Due to the almost one-directional nature of tone, it knows that by going from hostile to friendly it must make an excuse first, which can only work sometimes and requires an honest performance. All this is no different for public sound. The Lorelei-technique of up-framing one’s selfish interest (figure 6.2) is the salesman’s pitch to lure followers. Between the making of sound in private circles and the public domain there is, however, one remarkable difference. The disarming power of humour is almost unused in EU public spaces. The messages sent to the mass media and mass publics are usually bitter, without irony or smile. Even cartoons usually have the form of humour but the contents of bitterness [Bal, 2011]. Only light entertainment is used sometimes, such as the Spice Girls at the 1998 EP. The weapon of humour may be easier used in intimate than in public settings, but at the domestic level it is often applied in public settings too. As humour is culture-bound, the most probable explanation for the near absence of humour at the EU level is that, so far, there is no European sense of humour, which is another indicator that Europe is not (yet) a community of common cultures. 270

Chapter 6

Managing the EU Fieldwork

Summing up, the fine-tuning of the sound button requires professional skill and even artistry. The interest group that makes spontaneous noise is like the brainless bird that twitters when seeing bread in the garden, so attracting competitive birds plus the cat. Given the one-directional nature of noise, the prudent group relies on silence management when starting its fieldwork, while it is free from a mindless phobia for the ghost of noise, as otherwise it would neglect the many rational exceptions. From that starting-point on, it wants to identify the rational exceptions and not to make unforced errors, this being a matter of study and prudence. Without this, there can be no fine-tuning of sound or silence.

VIII Lobbying Where? ‘Brussels’ is only the pars pro toto name for the EU machinery, as it is the headquarter of most institutions and offices. The relevant place for lobbying can, however, be anywhere. It can be in the EU buildings in Brussels, but also in the nearby restaurants or in the offices of interest groups. It can be anywhere EU institutions and agencies meet or are located, such as the EP’s sessions in Strasbourg (where MEPs usually have more time available), the Court in Luxembourg, the ECB in Frankfurt, the EFSA in Parma and EUROPOL in The Hague. It can also be a hotel along the road to Paris, the country holding the rotating Council chair, the office of a WTO panel or anywhere else where stakeholders or officials get together. The old lobia or antechamber is now often a quiet seating area close to a conference room in an office, hotel or restaurant. Most stakeholders and officials work and live elsewhere. MEPs travel a lot and have a life in their own country. Council ministers meet only incidentally in Brussels and their civil servants live at home. The same goes for the experts of Commission committees, Council working groups and EuroFeds. In short, ‘Brussels’ stands for a hub system with Brussels at the centre of a wide network full of routes and places [De Groof, 2008]. Most people can be met at many different locations and often in their home country. Only members of the staff of the Commission, Parliament and Council (including Coreper) live primarily in Brussels or nearby. On the matter of location, the Commission officials, in their role of lobbyists, hold the easiest position as most interest groups come to them spontaneously or by invitation, for example by their experts, lobbyists or letters. All other interest groups have to travel to a location. They cannot be everywhere at the same time and thus must be selective. The matter of locating the meeting may seem trivial, but without a meeting place it is hard to lobby, and without a satisfactory place it is dif271

Managing the EU Fieldwork

Chapter 6

ficult to be successful. Therefore, the professional group takes this matter seriously and selects its areas of operation rationally. First of all, if based in Brussels, it considers carefully the location of its own office there. It can learn from the Dutch PR that for almost twenty years kept office far away from the EU buildings around Place Schuman, on the road homewards, a detour most stakeholders are unwilling to make, thus compelling its staff to go to their offices (unwise for negotiations), a restaurant (difficult due to lack of budget) or their own space in the Council building (where others may see them and guess what is going on). Secondly, the prudent group wants to know where it might meet other stakeholders and officials. The data file on the earlier question ‘to whom?’ can provide the information, as most of them have regular locations and routine behaviour. The interest group only has to know their behavioural map or agenda. As most stakeholders prefer to keep a low profile, it must gather background information about their rhythm of life. Professional groups collect this as part of the vertical axis of arena study (figure 4.3) and keep it up to date in their directory. A few consultancies sell it. The choice of location can be optimized. For the physical window-out monitoring of stakeholders, Brussels is a fairly good place that attracts numerous interest groups and offices. Many relevant people travel to and from Brussels and can be watched or met here. Many other stakeholders identified as primary on a dossier are located elsewhere, however. Visiting them there can be a costly affair. Remote contacts such as by phone or email are efficient but only effective after personal trust has been established. Multinational groups might be contacted at some country office and national ministries through their attachés at the embassy. Every group, wealthy or poor, can ask consultants or befriended people elsewhere to assist in monitoring stakeholders located in their vicinity. By early shared-listing every group can get such local outposts within reach. This indirect use of local representatives is often both efficient in its cost savings and most effective, as these local people speak the local tongue, have local networks and do not look foreign, in short have a superior capacity for monitoring. The selection of primary and befriended stakeholders elsewhere is clearly a matter of good preparation. For its window-in meetings with crucial stakeholders, the professional group dovetails its behavioural maps (agendas and schedules) with those of other groups. Brussels may seem an efficient location for negotiations, but is usually full of social control, not neutral terrain and not at all relaxed, in which cases Lisbon or Budapest may be better places for meeting. At the most detailed level this optimizing is a matter of route planning and navigation (figure 3.2), including stops along the way. Similarly, the group can arrange multilateral meetings with primary stakeholders, which 272

Chapter 6

Managing the EU Fieldwork

sooner or later must take place for aggregating the interests up to a potentially winning (pushing or blocking) coalition. It is all a matter of dovetailing. With some stakeholders the interest group may already have established regular contacts, for example in a EuroFed or expert group. Then it can use the day before or after the planned meeting for holding an extra informal one. If the regular meeting concerns an expert group or committee, the Commission implicitly reimburses most variable costs of the extra day. Some members of expert groups and committees even lobby for more regular meetings in order to get more extra ones available.

1 Cyber-lobbying A topical question is whether a physical place can be replaced by a virtual one, in short by tele-lobbying or cyber-lobbying, as is seen to be happening recently in social spheres [Pitt, 2012]. The answers are varied. For the window-out information lobby, new tools such as websites, news alerts and search machines are very useful and much used. As interest group of its own, the Commission increasingly has been using e-communication tools and practises since 2006. Most interest groups monitor each other and the officials with cyber-tools. PA practitioners often spend most of their time behind their screens, like journalists. As applies to every source or tool, the quality of gathered information (reliability, validity and completeness) cannot be taken for granted and has to be improved by traditional means (by ears, eyes, feet). Useful for monitoring are chat-sites, online discussions, webinars, hate-sites and other variants, which may give a rapid and cheap downsize of a controversial value into different issues and frames. Sharing information with trusted others at home or in a EuroFed is much facilitated by tools like email, intranet, web-conferencing and web 2.0, but new problems affecting quality are security (hacking) and privacy (like the 2010 Wiki-leaks). For the window-in support lobby, new online tools like weblogs, twitter, rogue-sites, podcasts or buzzing have limited use. Most interest groups remain passive on the online tools that raise controversial matters, as they fear to become part of any controversy, except when they feel to have to fight an uphill battle. Then, the tools may be useful for carrying out a form of cyber-activism [Illia, 2003] that attracts a crowd of people and might result in, for example, mass mobilization, public debate or issue reframing in order to create a friendlier arena. This intended effect cannot be guaranteed, however, so such actions could result in a move into an unfriendlier arena. Anyhow, many interest groups inside an arena fear ‘crowd insourcing’ as this creates effects beyond one’s control. Cyber-activists from outside that mobilize mass protest against an established arena, like the 2011 273

Managing the EU Fieldwork

Chapter 6

Occupy Now protest against the financial sector, have the problem of taking the next step, as their mass of supporters is usually only united over some problem but not over its solutions. Their next step can only be traditional, i.e. to get organized by common body, voice and message, which are essential for any window-in lobby that has to result in making compromises and sufficient consensus. Not surprisingly, most Brussels PA practitioners are, according to surveys [EurActiv, 2009-A/B/C; ComRes, 2008], modest users of cyber-tools, except for gathering information online (particularly Wikipedia) and sharing it by email. MEPs lobbying for themselves largely use it for similar purposes and for contacting voters; the latter also by Facebook, Twitter and Myspace but mostly in domestic context only [Fleishman-Hillard, 2011; ComRes, 2012]. By online consultations (blue paper) the Commission launches open calls for interests but it urges the responding people to join existing groups or to establish new ones, if necessary with its help. Apparently, compromise and consensus cannot be made by the Wikipediamethod. Embarrassing incidents involving the use of cyber-tools in the recent past make prudent use of them even more important. In 2008 the European Banking Federation (EBF) emailed to the EP its critical position on the Commission’s proposal on Consumers Credit; unfortunately the Word document sent by the EBF included earlier drafts of wording that caused embarrassment. Our conclusion is that new media are useful for window-out rather than window-in operations and need to be supplemented by old practises and media. This may seem paradoxical in ‘the age of internet politics’ [Chadwick and Howard, 2008], but the acquisition of relevant information and substantial support remains a matter that depends on personal trust and personal contacts and, thus, of meeting places.

IX Lobbying When? ‘Lobbying when?’ is the last of our leading questions relating to the fieldwork. The rule of thumb in simple manuals urges the interest group to be early in the process, because then it can prearrange the playing-field and take the lead. An early player, however, is also visible at an early stage, spends a lot of scarce energy and can be bypassed by the competitors joining the match later. As in cycling, usually only amateurs and helpers rush ahead of the pack at the start. For professionals the proper timing is always a difficult balance between being not too early and not too late. Precise timing is an art. It requires the interest group to have a time culture, in particular an excellent memory (past time), a long-term perspective 274

Chapter 6

Managing the EU Fieldwork

(future time) and a great deal of alertness (any time) because, as described in chapter 2-IV on the EU dynamics, the pace of decision-making in the EU is fast and is often faster than in most national capitals. The leading question for lobbying is: when precisely is the best moment? The answer depends, of course, on the situation that is dynamic due to the dossier life cycle (figure 4.1). So far this cycle has been presented as a stylized model with seven phases and thresholds, from social problem formation to enforcement. This presentation falls short of full reality, however, as it has the following subtleties that, if neglected, easily result in mistakes in the determination of timing. – Feeding forwards and backwards. The move from the one to another phase is frequently not at all clear-cut. One phase often feeds forwards or backwards into another phase, as illustrated by the EU saying that ‘those who define the problem take half the decision’. Even the die of inspection may be cast long before its phase starts. – An open cycle. The cycle often has no fixed sequence of phases and is open to start anywhere. An expert group can initiate a dossier and then search for a social problem to be attached. A new dossier can be the spin-off from another agenda, a package deal or a finding from an inspection. An EU official may adopt a social problem as discretionary power for direct policy implementation. – No outcome. An outcome may never occur in reality. At any time, the officials can decide to terminate a dossier. Many a dossier fades away without a decision having been made and dies as, for example, a nonproblem or a non-output that has been caused by stakeholders opposing the dossier outright. – The attention curve. Public attention is often not at its highest during the formal decision phase and lowest during the phases before and after. Most parts of Treaty texts, Council decisions and delegated or implementing acts hardly attract any public attention that, by contrast, can be high for a green paper defining a problem, a white paper suggesting solutions or an inspection report criticizing a policy. – Mobile players. Stakeholders are not locked to positions in the early phases and officials not to those thereafter, as some suggest [Schaeffer, 1996]. In every phase an official can behave like a stakeholder and a stakeholder like being an official. Often, EP inter-groups and EuroFeds create agendas, experts from civil society make decisions and companies act as inspectors in competition. – Micro cycles. Every phase can even contain full cycles, so to say micro cycles. An EP rapporteur can shuttle from redefining the social and political problem to getting a formal decision from the plenary, and to scrutinizing the outcomes. In any phase the professional interest group 275

Managing the EU Fieldwork

Chapter 6

too feeds its homework and fieldwork mentally backwards and forwards to other phases.

1 Early, Late and Continual Activities The best moment for lobbying is, like anything in PA in the EU, clearly not easy to determine. True, it depends on the arena situation that for almost every dossier is a dynamic affair, as stakeholders and issues change during a life cycle that is full of subtleties. The formal procedure or published roadmap may give a handhold or clue for scheduling one’s lobbying [Hardacre, 2011], but only the co-decision procedure and a few other roadmaps like on elections and Council chairs give standardized deadlines [Marshall, 2010; Goetz and others, 2009]. In almost all other cases the deadlines are variable and subject to lobbying. The better moments for lobbying can only be selected by organized PA expertise, intelligence and reasoned estimate. The only rational rule of thumb on timing regards the following two activities, to be started at the earliest moment and continued thereafter: – Preparatory homework. This enables reasoned answers to the questions ‘who acts, why, for what, to whom, on what, how, where and with what result?’, leaving open only the question of ‘when’. The more proactive the preparatory work, the more time there is for an anticipatory action that can turn the factor of time from an independent into a dependent variable, and the question from the passive notion of time into the active one of timing. An ambitious government going to chair the Council starts its preparatory homework at least twenty months beforehand, the average time of the secondary process, plus the same length of time for monitoring the Commission’s draft phase, altogether three to four years earlier. Such anticipated timing is no different for a company, NGO, ministry or region that really wants to achieve success. – Anticipatory action. The preparatory homework may reveal that some anticipatory action is necessary or helpful and hence should be planned earlier. For example, an NGO having to abandon an EU challenge due to its internal weakness should swiftly address this. A Commission official observing that a playing-field is unfriendly should quickly start a Triple P game. A ministry observing that it lacks sufficient support from a network must start networking soon. A government wanting to score during its Council chairmanship must start its lobbying for information and support this chair years ahead. A company discovering that its interests are controversial should anticipate this by soon considering an up-frame and launching a campaign.

276

Chapter 6

Managing the EU Fieldwork

In contrast to these two early activities, the window-in activity of negotiations can usually be done better at a later time, as often the best bargains come not early but later. The interest group can find this better moment for deadline lobbying through its preparatory work, the data file. In an unfriendly arena and with nuisance value it can usually claim the best loss-compenzation just before the last minute. Important but wavering stakeholders it can approach early although it gets the greatest impact if it brings them on board at a late stage. In a friendly arena, the interest group that is seen as important can best act reserved, wait it out and show patience. In an indeterminate arena, wait-and-see tactics require the group to be slow on entering negotiations. Otherwise it will trade off some interests that it may have to re-trade later, once the arena turns friendly or unfriendly. If the homework reveals that the best deal can be made at the last moment, for example in a meeting of the Council ministers, it should not act earlier. The early preparations and some anticipatory actions will suffice here. So was it done by the national air carriers in the 1992 Air Transport Slot Case [Van den Polder, 1994] and by the French agricultural lobby in the 1998 Transgenic Maize Case [Bradley, 1998], mentioned earlier. While the preparatory work and the anticipatory actions should be done at an early stage and the negotiations at a much later one, the other fieldwork activities of PAM cannot be allocated any single best time and must remain adapted to the topical situation. Resetting the strategy, updating the shared-list, extending the network or playing Triple P can all have better or worse moments for action but, dependent on any situation, some moments are better than others. For example, a EuroFed in a friendly arena wanting to speed up the process should not necessarily arrange an early appointment with, for example, the EP rapporteur. In order not to cause opposition, its absence, late arrival or U-turn might speed up the process the most. This is clearly a matter of detailed homework and no different from the delaying and the wait-and-see strategies in, respectively, the unfriendly and the indeterminate arenas. The determination of a good moment to take action depends, like anything in PA, on the details. For example, the interest group that wants to reframe its issue can hardly do so in the Commission’s draft phase if the Commission’s official has a personal commitment to the policy line. It might consider the expert meeting stage, but may know from experience that experienced experts there tend to be cynical about reframing, as they are experts on this too. In this case the group can better wait for the decision phases of EP and Council if involved, and for now only prepare its actions. The dilemma of time and timing is not separate from the many other dilemmas of fieldwork mentioned before, such as those relating to lobbying styles. In the early phases of the life of a dossier and at the early stages of 277

Managing the EU Fieldwork

Chapter 6

its involvement, a professional interest group usually behaves in the most attractive, informal, indirect and low-key fashion. This helps it to become accepted as an interested stakeholder, to build-up relationships, to keep its hands free and to avoid early opposition. The early homework and the anticipatory actions should reflect those styles too, as they enable the group to gather better information and greater opportunities at lower costs. During the negotiations at the later stages, it may have to become more demanding in order to get secured its desired outcome and to become more formal in order to reach a real decision. The use of direct and noisy approaches remains subject to the many considerations mentioned above. The timing of actions not having an easily identifiable better moment can best remain linked to the other dilemmas, as every situation requires an integrated approach. The amateur lobby group neglecting this runs the high risk of managing its EU public affairs in the wrong ways at the wrong moments.

X Extra: The Ideal Profile of the EU Public Affairs Expert Increasing the chance of getting a desired outcome from the EU is clearly a matter of fine-tuned and well-timed fieldwork and so a matter of special expertise. This raises the question about the ideal profile of the expert in EU public affairs, working for whatever interest group or as a self-employed consultant. The work to be done can be distilled from figures 1.2 and 3.3. The person must, ideally, be a threefold master: a Socrates in posing fundamental questions for every case of lobbying, a Max Weber in developing rational answers by studious work, and a Niccoló Machiavelli in synthesizing ambition, study and prudence. The expert performs all homework and fieldwork systematically and on time. Superficial work completed on time is as useless as perfect work done after the deadline. Altogether, mastering PA is a matter of disciplined creativity. The specific skills needed for the preparatory homework are at least the following ones, the first three being scientific skills and the last three applied ones: – Research-minded. The PAM expert is research-minded, with strong descriptive and analytical capacities, and with a good appetite for valid and reliable observations or, as they may be called, facts. – Value relativism. The expert holds a strong sense of value relativism [Brecht, 1959] when assessing the facts, as no fact is good or bad by itself but only in relation to an independently selected value. A phlegmatic expert can even turn a threat into an opportunity and maintain an unagonized attitude when faced with challenges. – Critical. All this is done with a critical or discriminating mind. The 278

Chapter 6

Managing the EU Fieldwork

expert is aware of, inter alia, the limits of knowledge, the alternatives to every option, and the difference between shadow and substance. – Familiar with the EU. The expert is familiar with the developing EU and its main actors, factors and vectors at play and with the relevant positions, places and ways to enter the system. Thanks to this, the expert can draft menus for action on the playing-field. – Rooted in the home front. While having one foot in the EU, the PA expert stays firmly within their own organization, this being the source of every interest. This requires an overall knowledge of its internal dynamics, main processes and position in society. – Pragmatic. The expert has a sense of pragmatic efficiency, as constraints always exist. Maximizing any part of PAM may bring an academic award, but will nullify one’s position in the EU. The expert optimizes the search for the best possible lobby. For the fieldwork two specific skills are most essential. First of all, the PAM expert must be able to make a two-pronged connection between the homework and the fieldwork. From the one side, the expert is able to take fieldwork experiences as an inspiring source of inductive creativity to enable a different approach to the homework. Even seemingly trivial information may indicate an important change of issues or stakeholders that requires additional arena analysis. On the other side, the expert is able to connect by methodical deduction the homework to the fieldwork. The two-pronged connection between homework and fieldwork gives the best possible solutions to all dilemmas of lobbying. The second specific skill falls in the area of classical diplomacy [De Callières, 1716] and covers personality requirements [Hondak and others, 2010; Castiglione, 1528]. In direct contacts with stakeholders and officials, the expert should be like a diplomat: communicative, reasonable and pleasant according to the various standards of the others. The person also masters leading languages, including body language and informal codes, and is during negotiations seen as reliable, generous and flexible, possessing a great deal of patience and tenacity for coping with long and difficult matches. A sense of humour and optimism is important, not only for its positive image but also for one’s own mental survival. Due to the paradoxical necessity of remaining both unagonized and yet looking human under all circumstances, the expert can engage in a bit of acting: smiling in stormy weather and grumbling when the sun comes out. As a diplomat, the person remains charming and respectful to others who have different preferences, simply because the others are warriors with nuisance value, despite how irritating their preferences may be. The expert behaves like a sportsman who is eager to score but in a courteous way, because there is always another match coming. All of these skills the 279

Managing the EU Fieldwork

Chapter 6

expert demonstrates in an unspectacular and quiet way. In addition to these specific skills, the PA expert possesses two general attitudes. Firstly, the person has a broad curiosity for any development related to the EU playing-field. This is not limited to the EU machinery but includes seemingly marginal developments such as a new political culture in Italy, devolution of the French state or the rapid industrialization of Turkey. With some time lag all these developments may affect the EU playing-field in the near future. Curiosity functions as the engine for updating one’s understanding of the varieties and particularly the irritating differences in Europe, and helps to anticipate the EU method of living together peacefully and the PA challenges that arise from it. Secondly, the expert has a strong commitment to his or her interest group and acts as a warrior on behalf of it. The PA expert wants to win, or at least not to lose. This dedication gives a licence to operate as the representative lobbyist. It also feeds back to the interest group, in the form of a flow of useful information, creative thinking and realized targets. The expert is capable of, for example, defining a challenge, saving a blessing or solving a problem and turning a threat into an opportunity. If advertised, this ideal profile of the expert on EU public affairs management will certainly attract a number of applicants. Many of them hold the self-confident belief that they are naturally gifted and have the right talents for the job. They say they have the ability to professionally manage EU public affairs at their fingertips. Almost certainly only a very few applicants will, however, come close to the profile. Even among the many PA practitioners working in Brussels, only a very small minority are widely considered to be professional and all-round players. The majority are often nonchalant when they carry out homework and fieldwork and/or play only limited roles such as desk researcher, representative or platform manager. Expertise, being a match of talents and skills for one or more specific roles, is always a scarcity. To some degree talents can be developed and brought out through training, although there are always people who hardly benefit from any training, which clarifies the difference between talent and training. Training without much talent can make a person a good player, but not a Kasparov. Talent without skills shows high potential but is still far from playing at the top. Talent plus skills creates art. The development of skills is easier than that of talents. So far, most people working in the field of PAM in the EU did not receive any special training for it. They have developed their abilities largely through their own éducation non-permanente. Most seniors rarely train juniors up to an advanced level. Only the happy few received instruction from their predecessor (‘from father to son’), including the inheritance of mistakes. In 280

Chapter 6

Managing the EU Fieldwork

most interest groups the learning takes place just by doing, with much trial and error. Most persons in EU public affairs are self-made people. Coming from a different job, frequently in line management, they are dropped without a parachute into the work of PAM in the EU. Lobbyists from government organizations (central and decentralized) usually have this type of curriculum vitae and tend to confuse their policy expertise with PA expertise. Those from MNOs often have received some minor training from a ‘godfather’ or a centre outside, the latter being offered by a few institutes and consultancies. PAM, in short, is still in an early phase of institutionalized professionalization [McGrath, 2005-B]. The PA official is, of course, not necessarily a person but can also be a small team, as exists in MNOs like Siemens or Greenpeace. The team is usually supported by other staff, including experts on EU law, line people for technical expertise, colleagues from country units for local information and maybe specialized consultants. In a team, all the PA work can be divided according to the talents and skills of the members, but even this is more theory than practise. Most small teams are formed by chance rather than by design. By the latter method it is possible to organize the team as a small chamber orchestra with professionals in limited roles and the whole team acting both professional and all-round. With a team one can also profit from the scarce (US) academic research on the profile of ‘best lobbyist’, according to those that are lobbied. The findings suggest that the best ones are full-timers, women and government officials [Nownes, 1999]. Full-timers are seen as more experienced and expert, women as more patient and empathetic and government people as driven by a more general interest. These different virtues are provided more easily by a wellorganized small team than by a single person. There is much room for improvement of PAM. The easiest thing to organize and to improve is internal training. Any Board should allow its PA staff to spend about 10% of working time on maintaining and improving their PA expertise by study, reading and training. A senior should invest in a junior, even if the latter might overtake him/her or depart to a competitor. Regular sessions with staff and line personnel involved in EU affairs can create a common concern for shortcomings at the home front and so a common agenda for improvements. The evaluation of any game played is, as said in chapter 5-VI, a great but highly underused source of learning and training. PA people invited from outside, even if they lack great expertise, may add value to internal training as they can stimulate critical comparisons. Recognized experts from outside can be asked to perform peer reviews. Another way of learning is by going out and meeting informally with PA colleagues from different interest groups, as happens in various countries (chapter 1-XII). The British monthly PA News has a 281

Managing the EU Fieldwork

Chapter 6

network of subscribers that exchange information on PA. In many countries and in Brussels only a few institutes, colleges and consultancies offer short EU training sessions with a chapter on PA, with very different qualities however. A university programme on PA is absent in Europe. Only a few countries have universities at which master’s programmes or elective course refer to lobbying and/or PAM, such as the École Libre des Sciences Politiques (Sciences Po) in France, the University of Ulster and Brunel University in the UK, the universities of Bonn and Erfurt and the Fachhochschulen in Bremen and Berlin in Germany, and the College of Europe and the settlements of Kent University and the United Business Institutes (UBS) in Brussels in Belgium. In short, in Europe the field of PA is still young. As the field develops, we will see an increase in the number of university programmes, academic research, published findings and much more [Fleisher, 2007; McGrath, 2005-B], until it finally reaches a state of mature professionalization. If a full curriculum were to be created, it should be interdisciplinary [McGrath and others, 2010; Fleisher, 2005]. The major courses recommended here are the following five: science and methodology for understanding value-relativism and research, political science for learning the flesh-and-blood of the EU system, EU law for knowing the EU skeleton, psychology for comprehending the human side of PA, and economics for mastering the logics of interests, scarcity and efficiency. Minor courses should be devoted to European cultural variety, European history and geography, negotiations in complex situations, comparative country studies and special EU policy areas. European languages would have to be studied as options. Other electives are information retrieval techniques, communication and maybe, as being useful for the basics of political engineering, chemistry (how to control an effect), construction (how to make a linkage and U-turn) and mechanical engineering (how to make it work). Finally, the students would have to complete, individually and in groups, some exercises, case studies, case-related homework and essays and to participate for some time in a PA department of an EU-oriented interest group. Such a university programme would, of course, not be a panacea. It would merely help to develop the talents people may have and so to avoid the costs of trial-anderror learning of people starting untrained for a PA job.

XI From Potentials to Limits of EU Public Affairs Management At the end of the third chapter, we explained our choice to start the presentation of EU public affairs management with the homework for both 282

Chapter 6

Managing the EU Fieldwork

the external arena and the home front, all before the fieldwork. This design is justified from an educational point of view. The homework should be the basis of the fieldwork. To avoid an egocentric or introverted approach towards the EU playing-field, it is better to start with the external arena rather than with the home front. The alternative design, starting with the fieldwork, would only produce questions and no guiding answers. As emphasized before, the various activities of the preparatory work and the fieldwork, all distinguished analytically, are however closely interwoven in practise. All the questions and the activities of figure 3.3 lie, in daily life, simultaneously on the table of every interest group, whether it is conscious of this or not. Most groups are not conscious or at least do not show in their behaviour to be conscious of it. Only the small ‘premier league’ of the best professionals systematically interweaves the homework and the fieldwork by (figure 1.2) thinking backwards and acting forwards and by integrating strategy, tactics and techniques. The paradox is that the more one keeps those various questions and activities separated analytically, the better one can interweave them for the creation of a desired EU outcome. Without analysis, there can be no synthesis. The greatest skill lies in interconnecting all questions and activities systematically, thereby acquiring new insights for both the fieldwork and the homework regarding both oneself and the arena. To do this perfectly always remains a dream in practise, but without the ambition towards it nobody can ever join the ‘premier league’. The more experience with homework and fieldwork one gathers, the easier one makes the connections between the two and the more one grows in expertise. The reward is a better choice of the next activity to be done for the fieldwork and homework, so bringing the interest group closer to the desired EU outcome. Like in any strong competition, it is the margin that makes the difference. By this interconnection one gets both homework-based fieldwork and fieldwork-based homework, which is the perfect match. The preparatory work, then, is as much a source of inspiration for playing the game, as playing the game is for preparatory work. Through this process, one also creates a real R&D of PAM in the EU, resulting in even more options, tools, menus and dilemmas and, in addition, more manageable factors and constructible vectors. The surplus of potential actions gets even larger and is, theoretically, unlimited. For the professional this is not an embarrassing problem but a pleasant opportunity that provides more roads to EU success. What is unlimited in theory can, however, have many limits in practise. That will be the theme of the next chapter. To jump to a paradoxical conclusion: the professional, already taking greater advantage from the surplus of potential PA actions than the amateur does, suffers less than the amateur does from the limits of PA in practise. 283

Chapter 7 The Limits of EU Public Affairs Management According to Machiavelli (Discorsi, book II-29), in any course of affairs many events happen and many misfortunes come about. Therefore, it is better to bend with fortune rather than to oppose it and to act in accordance with this truth from history rather than to neglect it. What are the implications of this for PA management in the EU?

I From Tantalus to SCARE Our preceding chapters on the management of EU public affairs are full of buoyant spirits. The attractive flowers and trees of the EU playing-field seem to be within the reach of every interest group. Cultivating them may not be easy as this requires much homework and fieldwork but can be carried out if one is careful and energetic. Many groups, from governments to business and NGOs, hold the optimistic belief that, after their digging, fructifying and pruning, sooner or later a lot of flowers and fruits can be brought home as trophies. Many see room for planting even better varieties. Tantalus, in Greek mythology the hero unable to pick the grapes in Hades, would have loved to go to Brussels. In PAM, however, the sky is not the limit, but the players, the playing-field, the issues, the game, the audiences and all other features of a situation set the limits. Most of these limits are variables that can even be manipulated or managed to engineer a desired outcome, but only to some degree. The limit that is both most crucial and within reach, or manageable, is intelligence, being the summary variable of all preparatory homework regarding the three arenas of home front, stakeholders and officials and giving clues for the optimal support lobby for getting success. This summary variable is dependent on many factors that can pose limits to success and thus deserve closer attention here. The various limits can be classified under different headings. One distinction is between limits that are endogenously linked to an interest group and those that exogenously come from outside. Another one differentiates between structural limits caused by patterned behaviour and cultural ones set by the individual and the collective minds. More concrete is the typology that is based on categories of the communication process, which underlies any process of influencing others. We label them here as 285

The Limits of EU Public Affairs Management

Chapter 7

categories of SCARE: limits coming from the Sender or interest group, the Channels that transmit messages, the Arena where all information is exchanged, the Receivers (such as EU officials and other stakeholders) and the Environment conditioning the other four categories. In this approach, all limits of PA in the EU are viewed from the perspective of the interest group trying to influence others. Its message can have substantial form and contents but also be a non-message; for example, silence or absence meant to convey a position. We shall examine this SCARE typology and identify for each category a few typical limits. Their identification mirrors that of potential actions, shown in the preceding chapters, in two respects. Firstly, like potential actions, limits cannot be listed exhaustively. Theoretically their numbers are infinite. We confine our description to some major ones existing in the daily life of PA management in the EU. Secondly, for the management of these limits we shall discriminate again between amateurish and professional approaches. The catchword for the amateur is nonchalance and for the professional it is study and prudence. More than the amateur, the professional is scared of making unforced errors and of thereby becoming one of his own strong opponents. As a prudent player, the professional remains conscious of the many possible limits, tries to escape or to remove them in advance, and respects the unmanageable limits. The assessment of a limit remains a matter of expertise on realizing the formula 2E = MI2 and basically not different from that of a challenge, discussed in chapter 5-III. Every interest group must, firstly, establish whether or not a limit exists in a specific situation. The same channel, arena and receiver may set limits to one stakeholder, but give free passage to another, as limits are relative and not absolute. The group has, secondly, to determine whether a limit is a problem or a blessing. If an existing limit is considered helpful for the group’s targets, then it is a blessing that should be saved by efforts to keep it. If it is seen as harmful, then it is a problem that can be solved either by removing the limit and its consequences, or by reconsidering its harm. By analogy, the absence of a limit, if considered a challenge, can be managed as either a blessing to be saved or a problem to be solved. The professional group does this sort of homework not only for its own organization, but also for primary officials and stakeholders. They all have their own variable collection of limits.

II The Sender’s Mental Limits PAM is a highly human activity. The person or team in charge has to search continually for the next optimal move. Brains, enriched by talents 286

Chapter 7

The Limits of EU Public Affairs Management

and skills, are crucial. Inevitably all sorts of human shortcomings can play a disturbing role. For example, due to some personal drama even one’s private life can become an obstacle and undermine one’s PA performance. Someone’s character may be so full of arrogance and vanity that it turns the person into his own worst enemy on the playing-field. The PA official may be mentally obsessed with one’s own interest, an opponent, an issue or something else that is part of the playing process. These examples at the individual level can easily be transferred to the organizational level. Then, the interest group holds, for example, stereotypes and clichés about stakeholders from different markets, regions, countries and sectors. As a result, it has difficulty with compromising and finally making a deal with the other stakeholders en the officials [Gutmann and Thompson, 2010]. Here we discuss three cases of mental limits at organizational level: emotions, dogmas and myths.

1 Emotions The list of potential emotions is infinite. An interest group can become imbalanced by, for example, the threat or the opportunity attached to a challenge. While carrying out his cost-cutting operation Centurion in 1992, Philips CEO Jan Timmer emotionally blamed the Commission for its delaying of his great HDTV mission. He succeeded in only attracting more opposition. A group can also exhibit voracity, as a result of which it cannot reduce the long-list to a manageable shared-list and shortest possible short-list. Another frequent emotion is that of political (in-)efficacy, being the belief that it is one’s fate to become the winner or loser. Many emotions crop up during the game, for example resulting from the perceived unfair behaviour of another stakeholder. Many a group often feels disturbed and thus acts agonized. In the Biotechnology Dossier, the opponents of GMO promoted the norm of ‘zero risk’ and publicly blamed the proponents for putting mankind at maximum risk. Part of the food industry behaved, indeed, as if it were emotionally imprisoned by this blackand-white debate. An interest group can also have the emotional belief that some important stakeholders, such as Commission officials, MEPs or NGOs, are autistic, biased, unreliable or in some way vicious. From their side, many EU officials display emotions of distrust regarding what interest groups say and want. An emotional sender easily resorts to voicing annoyance, even if this hurts its interests, thus throwing the tomahawk instead of unaffectedly showing its peace pipe. A frequent emotion arises from uncertainties. The inevitable lack of useful information can be upsetting, especially when both the stakes and the time pressures are high. Emotions are a strong variable in the double sense that they can both 287

The Limits of EU Public Affairs Management

Chapter 7

strongly affect lobby behaviour and exhibit strong variation. Regarding the latter, there even exist specific nation-wide emotions, such as the British allergy to EU federalism (the ‘F-word’), the French aversion to takeovers by foreign companies and the Polish distrust of EU food law that limits their exports. Emotions can be bound to the type of organization too. SMEs frequently hold the emotion of inefficacy, while many an MNO regularly expects that it will be one of the winners. Issue-driven NGOs usually have an emotion anchored in their central policy value, such as ‘green environment’ or ‘safe and healthy labour’. National ministries may attach emotional value to a specific EU dossier when this has priority at home, for example in their parliament or mass media. Such a priority is seldom based on sound preparatory work and can soon create the problem of having to save face. The same can happen to a national government chairing the Council and being urged by its parliament to bring order to the EU. In 1991 the Dutch government launched an overambitious campaign for a European Political Union (EPU), much alike the Lisbon Treaty almost twenty years later. In a few weeks it had its ‘Black Monday’ and became so bewildered that in 1997 it publicly gave up any ambition for its new chairmanship. Such emotions are part of human life or nature, like earthquakes. Professional groups, however, know that, to some degree, earthquakes can be foreseen and some of the most devastating effects can be controlled for. This awareness is the precondition of prudence and a matter of good preparation. It is not that professional players have, necessarily, fewer emotions than amateurs have, but they filter the emotive stimuli better and react in a more reserved way than amateurs do. They are disciplined about heeding their homework and remaining concentrated on their targets. The emotions of other players they take as useful indicators of their feelings of weakness (unless they are clever comedians), and any unfairness from them they return as free boomerang or treasure up until later. In the MNO model the professionals are able to remain cool by such measures as the critical ad hoc team, the regular update of the long-list and the construction of scenarios. Even so, they remain susceptible to emotional behaviour, but less so than amateurs. This difference can be crucial for the desired outcome.

2 Dogmas A dogma is a strongly held belief in being right and those who hold it tend to be resistant to critical debate. Policy experts often believe that their policy analysis is unsurpassed and during meetings with colleagues from other groups and countries they easily engage in dogmatic clashes. Party288

Chapter 7

The Limits of EU Public Affairs Management

politicians frequently claim their ideology to be the best too and preach that mankind’s salvation lies in following their message. Issue-driven NGOs tend to have great difficulty with any compromising on their core value. Many interest groups are inclined to stick to a strategy once it is set and, seeing a problem, prefer to deal with the unpleasant facts rather than their fixed norms (figure 5.1), thus losing half the number of solutions. In monitoring a dossier arena (figure 4.3) they may have strong beliefs about who the primary stakeholders are and which issues are at stake, thus neglecting reality. Most national governments embrace dogmas of national sovereignty and coordination, both pushed by their national parliament, the mass media or even the constitution and allowing limited room for both European compromises and domestic pluralism at Council level. Dogmatism easily irritates others as it denies pluralism and values relativism [Galston, 1999]. It heats up emotions and hinders negotiations. There seems to be some national variation in dogmas. Interest groups from the Protestant areas of Europe have the reputation of selling their dogmas as expertise. The French are frequently dogmatic on central planning, the Dutch on accommodative compromising, the British on rule of law and the Romanians on flexible implementation. Some variation by organizational type seems to exist as well. Governments, usually lacking a PA unit and using their policy experts as their agents, frequently take dogmatic positions, particularly if their elected body instructs them to do so. The same pattern can be found in private organizations if policy experts set the tone. Many NGOs take their mission as dogma and easily launch a ‘fanfare doctrinaire à Bruxelles’, as portrayed by James Ensor. Many companies do so when their core product or service and ultimately their licence to operate are at stake. Amateurish groups adopt dogmas more easily than professionals do, as the latter are more critical about their own strategic norms and values. Therefore, in the MNO model the PA experts must counterbalance the policy experts at home. Proactively they include the dogmas of other stakeholders in their plan of action. Professionals are more inclined than amateurs to stay away from public platforms like press conferences and mass media that function as consuming markets for dogmas. They act publicly in only a few specific situations, mentioned in the previous chapter, for example when launching an upframed value (Lorelei) that should mobilize public support. They know that the negotiations that are part of the decision process proceed more smoothly when they take place in inner rooms, far away from public platforms. Even a professional, however, is seldom free from dogmas and may hold a sacrosanct value. The amateur is only less free, and this difference is important for the outcome.

289

The Limits of EU Public Affairs Management

Chapter 7

3 Myths A myth is a cherished preconceived image of reality. It is taken as true, but in fact it is merely fiction formed by, for example, tradition, prescripts, folklore, propaganda or tabloids. Like every system, the EU too is full of political myths [Della Sala, 2010]. Figure 7.1 exemplifies this through ten popular myths about the EU, all taken from the headlines of mass media and with their meaning put in italics below. Most have a large following, even when they partially contradict each other, such as the claim that the EU is both corporatist and dominated by governments.

Ten Popular EU Myths 1. 'Brussels': A Big Bureaucracy The number of EU civil servants is extraordinarily high. 2. Centralist Union 'Brussels' governs top-down. 3. Formal Powers = Real Influence Those who have the formal powers are the influential ones. 4. Corporatist Union The employers, trade unions and officials run the EU together. 5. Elitist Union Those running the EU form a closed group. 6. Dominant Governments The national governments have the final say. 7. Democratic Deficit The people and the Parliament are politically marginal. 8. Paris-Berlin Axis France and Germany run the EU together. 9. 'Not so Relevant for Us' The EU is much ado about nothing. 10. 'Towards One Hotchpotch Union' All varieties are cooked down to one bland average. Figure 7.1

The ten myths are as follows. ‘Brussels’ is said to be a big bureaucracy, but in real terms of budget or number of civil servants it is miniscule. The EU is seen as centralist, but a large part of its decision process takes place on work floors and lower levels and is in reality partially in the hands of both insourced experts and lobbyists from public and private interest groups. The skeleton of formal powers is mistaken for the flesh-and-blood of influence. Interest groups of organized management and labour, interwoven with EU officials, make the EU look like a corporatist system, but in reali290

Chapter 7

The Limits of EU Public Affairs Management

ty they rarely amount to a closed policy cartel. The EU is seen as elitist, but usually it is an open decision-making system or, as it is sometimes called, a polyarchy [Dahl, 1971 and 1989] of many competitive groups, which is the opposite of an elitist system. National governments are seen as dominant, as prescribed by the Treaty of Rome, but in reality they are each only one player among many and, acting as the Council, they approve only about 15% of the total of the bindings acts adopted by the EU every year. With regard to the EP, it is claimed that the EU has a democratic deficit, but in many respects and compared with many national systems it has more of a democratic surplus (chapter 8). The allegedly strong ‘Paris-Berlin axis’ is in reality a red-hot junction between two rivalling capitals amidst now 25 other ones. In particular, small-sized interest groups often attach little relevance to the EU, but in fact they only feel small in a largely relevant system. A hotchpotch Union has often been predicted, but goes counter to growing regional and cultural differentiation. In fact, all ten popular myths are incorrect ‘strong stories’ (the old-Greek meaning of ‘mythos’) indeed. Every person or group holds some myths and is thus misled by preconceived images of reality, but some are more under the influence of myths than others. There seems to be some national variation. Most of the myths are widely believed in the United Kingdom. The democratic deficit myth is popular in Scandinavia. The myth of formal powers is cherished in rather formalistic countries like France, Poland and Spain. The myth of little relevance is, except for subsidies, widely held in Greece, Romania and Italy, where EU legislation is often considered a recommendation rather than binding law. Some myths are typical for a type of organization. National governments frequently hold the myth of their own importance, except when they are defeated on important dossiers. Ministries of Home Affairs and Justice, being latecomers at the EU level, still tend to believe that their influence equals their formal powers. Newly arriving member states and NGOs often believe that ‘Brussels’ is a big bureaucracy, full of centralism and elitism. Among many SMEs the myth of little relevance is popular and used as an excuse for passivity. Amateurish groups hold many myths. They follow the popular stereotypes, gossip and newspapers. Then they go to Brussels when they should really be somewhere else or they put trust in their ministry in the Council when the Commission makes the real decisions. Professionals may have their myths as well, but usually in lower quantity and with less certainty. When faced with challenges from the EU, they act by conscious reflection rather than myth-bound reflex. The discipline of preparatory work gives them the best possible protection against a misleading EU mythology. For example, their dossier-bound inventory of stakeholders is an open-minded 291

The Limits of EU Public Affairs Management

Chapter 7

exercise regarding a wide range of stakeholders in the EU. In the MNO model, the professionals enjoy safeguards such as the critical mass of colleagues, ad hoc teams or peer review. The generalization of a single experience is a risk to all as it may generate a post-conceived image of reality with mythical contents. The dynamics and complexities of reality mean that no one is completely free from belief in myths. The professionals may be only slightly freer and less bound than the others, and this difference can tip the balance towards winning or losing.

III The Sender’s Organizational Limits An interest group’s mental situation as well as its state of internal organization can set serious limits for its PAM in the EU. Staff and line people may hinder each other on the EU playing-field, when the public relations (PR) staff thwarts the PA unit on the use of sound, the legal staff does so on litigation and the technical experts push their dogmas. In a voluntary association such as a EuroFed, some members can weaken the common position. In an MNO the various country units may choose their own way to Brussels. Many a group falls short of important resources, such as expertise, good reputation, strong leadership and sound strategy. NGOs constructed by government (GONGO) or business (BONGO) can be unmasked as fake ones and lose credibility, and those seen as not representative may suffer the similar fate [Warleigh, 2003]. In short, all organizational variables can be in a state that limits the 2E of PAM in the EU. Here we shall focus on three typical cases: dissent, scarcity and lack of leadership.

1 Dissent The lack of sufficient cohesion is one of the most common and most serious causes of poor performance in PA. Every interest group has a certain amount of internal pluralism, indicated by different values, role expectations and duties (chapter 5-II-1). To some extent this is a blessing, as it promotes internal creativity and external adaptability. This richness should be optimized, as maximal internal dialectics can hurt external performance. If a group sends different messages to the same audience within a short time, then it may soon lose credibility as a stakeholder and, if it does so as part of a comedian diversion, it may lose reliability. Every group has some record of serious dissent observed by other stakeholders, who can exploit it to their advantage. In its lobbying on the 2001 Tobacco Ban Directive (EC 2001/ 37), the market leader Philip Morris profited from the weak cohesion of its main competitor British American Tobacco (BAT). If a 292

Chapter 7

The Limits of EU Public Affairs Management

group has grass-roots layers, as happens in many voluntary associations, the display of dissent on major issues is a normal part of its internal life and easily noticed externally. The same holds true for vertically segmented dissent, such as between divisions of an MNO, ministries of a national government and DGs or agencies of the Commission. Settling an internal cleavage costs a lot of time, compromise and management. Then the group might even better abstain from playing and leave its interest to the winds. Dissent may be a regular feature of every interest group, but it is highly variable in frequency, intensity, form, contents, methods of solution and other aspects. There seem to be some patterns to the variations. One is national. In Denmark and the Netherlands, many interest groups have grass-roots layers taking up much internal debate and other costs. In France and Germany, internal dissent is often embedded in procedures and formalities, and is solved by a decree from the top to the rank and file being trained in paying lip service. In Italy internal dissent tends to follow lines of clientelistic factions, with patrons seeking accommodation. Organizational variation is another pattern. Government organizations show much dissent due to pressures from mass media and politicians. Their opponents can easily exploit their internal cleavages. Inside NGOs, the voluntary members frequently voice their different positions on an issue publicly. Confederations like ETUC and BusinessEurope are frequently paralyzed by internal dissent, more so than rather homogeneous EuroFeds like EMF (metalworkers) or Europabio (biotech firms). MNOs often have strong dissent between both their divisions and country units and SMEs have this frequently inside their domestic associations. The most striking variation exists between the amateur and the professional interest group. By its preparatory work, the latter has an early warning system for any internal dissent arising on an issue, thus enabling it to anticipate it by, for example, discussion, compromise or internal deal. If it fails to bridge the gap, it can rationally decide to remove the issue or full dossier from its agenda and to remain passive, thus avoiding any further waste of energy. By its preparatory homework it can also profit from the observed dissent inside any other stakeholder group. A relevant friend can thus be helped, a wavering group won or an enemy split. Such a result, of course, has a good chance of success but is not guaranteed. The amateur group, however, does not even get the chance as, due to its neglect of the homework, it lacks the early warning system.

2 Scarcity Important resources such as expertise, networks, positions and financial means are always scarcities, either by quantity or by quality. Only groups 293

The Limits of EU Public Affairs Management

Chapter 7

without ambition have no scarcity of resources. Even scarcer are the skills of acquiring expertise, networking, positioning and financing. Knowledge about the EU generally or about the arena specifically has limited reliability, validity and completeness. An interest group may have the finest reputation, but this is always less than sacrosanct. Time is almost always scarce. At the operational level of EU public affairs, qualified people are an utmost scarcity, and it takes a long time to get talented people well trained. Once they are qualified, they may switch to another position or organization, taking with them their qualifications. Often the successor has to start almost from scratch. Every asset, in short, is hard to obtain and can soon be lost. Scarcity is a fact of life, making the economy or efficiency of PAM a must. There are some indications of national variations in scarcity. Groups from the new member states still have rather weak networks, positions and budget that may be helpful for success in the EU, but increasingly they find their ways through their skills of building up networks and positions and by acting efficiently. Interest groups from the northern (Protestant) countries are often more rigid about informal networking than those from the southern (Catholic) areas. Since the 2008 financial crisis the southern groups suffer more from lack of budget. Some variation in scarcity by organizational type exists too. The most notorious case of resource scarcity is the Commission with, compared to other bureaucracies, its extremely small staff and budget. Governments are well-resourced, but often fall short of PA skills as they mainly rely on technical experts without PA back-office support. Many established NGOs have skilfully built up resources like nuisance value, popular image and, if they are not dogmatic, interesting supply sides. New NGOs often compensate for their lack of resources by having great ambition, studiousness or creativity. MNOs regularly suffer from their reorganizations that hurt or halt their PA work in the EU, as shown in the 2008 crisis [ComRes, 2009]. SMEs often have weak resources and skills, but get from the Commission a more preferential treatment since the mid-2000s. There exists a clear difference between amateurs and professionals. The latter manage their scarcities more prudently and are more selective in, for example, adopting a dossier for their short-list, developing a network, mapping a promising route or negotiating over a target. Their selectivity comes from their homework. Through this economic behaviour they can save on their scarce resources and skills or exploit them more successfully. By identifying their allies at an early point, they can arrange cost-sharing of the monitoring, networking and other fieldwork for influencing the EU. Amateurs, in contrast, behave more wastefully and are exhausted sooner. This difference can be dramatic for the desired outcome. 294

Chapter 7

The Limits of EU Public Affairs Management

3 Lack of Leadership Management deficits can take many forms. No management is ever fully free from emotions, dogmas and myths. It can be emotional about a competitive stakeholder, dogmatic about its strategy once set and mythical about its position on the EU playing-field. It is also never free from dissent and scarcity inside. All this is normal, but the special duty of management is to control and to relieve all these mental and organizational limits. Here management meets its own limits: its decision on strategy is often not based on a regular and open-minded assessment of preferred norms and observed facts, but on indolent compromise between, for example, the old and desired strategy, one or another emotion or two rivalling board members. Frequently it starts its consensus management not by stimulating the rich internal dialectics, but instead by leaping to formally imposed consensus. Even if it starts in the right way, it may fail to transform the dialectics into better synthesis. Many managers have a poor span of control, thus leaving the various units, including those in PA, to drift and resulting in an uncontrolled short-list, wasteful fight with PR over sound, and imbalance between targets and means. The variations in management deficits are even nation-specific. In countries where some emotions, dogmas and myths persist as mentioned above, the managers usually have great difficulty in overcoming them. In rather hierarchical countries such as Germany, France and Portugal, interest groups often fail to create consensus dialectically. In the Netherlands, managers are, for the opposite reason, often indolently compromising on setting strategy. In Protestant countries, managers frequently undervalue the utility of wining and dining. In most new member states, coming from a centralist past, there is no strong tradition of building broad consensus. More specific are some organizational variations. The management of government organizations often fails to develop a realistic next-best scenario and to keep its staff and line co-ordinated. Smaller-sized regional and local governments show much variation but some, like the German region of Bavaria and the Spanish one of Catalonia, have a reputation for exercising strong leadership over their PA in the EU. EuroFeds and other NGOs are frequently characterized by the weak position of their management that is often selected, replaced and pressurized by their grass-roots. In MNOs with vertical segmentation (divisions, country units), the top management often shows limited leadership over the organization’s PA. The management of SMEs has to cope with the leadership of the domestic and European associations to which they belong and on which they are dependent for success and remote control. Amateurishly organized interest groups have clearly more manage295

The Limits of EU Public Affairs Management

Chapter 7

ment deficits than professional ones. They are affected by mental limits and by internal dissent and scarcities more than the latter. Their managers are more doers than thinkers and more led by reflex than by reflection. They are, in short, highly dependent on good luck and thus often suffer bad luck. The management of professional groups, in contrast, oversees its EU public affairs more or less as outlined in the multinational model (chapter 5-VII). It has its mental and organizational limits better under control, thanks to its early warning system developed through homework and enabling it to anticipate limits. It usually has less numerous management deficits, but it still encounters limits to its ability to manage its PA in the EU successfully. However, these are less the result of its own deficits and more of external circumstances, as discussed below.

IV The Limits of Channel Management A channel is a transmission system for sending messages. There are many different types of channels. Figure 7.2 presents a selection of the analytical variety. By combining them in series or parallel, the number and variety of channels can be much increased. A channel may be one-way by either sending or receiving a message, or two-way (or more) for interactive

Variety of Channels for Public Affairs Management • One Way or Two/More Ways • Single or Multiple (Network) • Connected or Dead End • Parallel or in Series • Direct or Indirect • Open or Closed • General or Specific • Natural or Artificial • Permanent or Revolving • Established or Newly Developed Figure 7.2

296

Chapter 7

The Limits of EU Public Affairs Management

communication. There may be just one channel or a multiple channel system, forming a network. A channel can be connected either to a receiver or to another channel, or it can have a dead end. Channels may go directly or indirectly to some place, similar to the direct and indirect ways of lobbying a stakeholder (chapter 6-VI-3). A channel can be open or closed. In the former case, it is free like a public road and in the latter case it requires a fee to be paid, a ticket to be obtained or a password to be given, as applies to a toll road, a party or an intranet. The channel can also be general for any member of the public or specific to only a particular group. Some channels are natural like a river, while others are constructed or engineered like pipelines or canals. A channel can be either permanent or temporary, being there irrespective of its use or only for as long as it is used, like a moving vehicle. Some channels are old, established and tested, while others are newly developed and experimental. The channel metaphor is inspired by the three phenomena of water, sound and electricity. Through cybernetics and systems analysis it has made its imprint on political science [Deutsch, 1963; Easton, 1965]. In applied political science, with PAM in the EU as chapter, the key question of every sender is how to get a message, in the desired form and with the desired contents, delivered only to the desired place and at the right time? Much can go wrong with this objective. During the channel process, the form can be affected and the contents changed. The message may be delivered to the wrong place and at the wrong time. These four problems of form, content, place and time can also occur in combination, mathematically resulting in 24 different problem cases. The message can even be lost, disappear or not be delivered at all. If the transmission is good, it may have been delivered not exclusively but widespread, resulting in rumours going around and maybe politicizing the interest of the sender. All these problems may not be due to either the sender or the receiver but be channel-related. Many causes are possible. It may be that the chosen channel is not correctly connected to the desired place, has limited capacity that delays delivery, contains a filter that affects the contents of the message, or produces its own noise and echo that make the message widespread. All these and more channel properties are, of course, variables. To some degree they can be manipulated or managed by both the sending interest group and every stakeholder or official. The former wants to get the message delivered as desired, while the latter may wish to distort, change, delay, misplace, obstruct or politicize the transmission. Both have many options, but any option has its limits as well. Next we shall discuss three typical cases of channel limits: power, quality and configuration.

297

The Limits of EU Public Affairs Management

Chapter 7

1 Power Some channels are more powerful than others, just as there are turbulent rivers and calm waters. EuroFeds, if used as channels of information and influence, may be powerful with regard to, for example, Commission officials but powerless to other EuroFeds. Ministries can have some nuisance value at the Council level, for example by linking different dossiers in a complex way, but be helpless in the EP. EuroFeds are seen as authoritative as long as they aggregate the interests of their members from different countries, thus saving the Commission officials a lot of hard work, but as useless if not. National ministries may be useful channels for domestic interest groups but dried-up ones for MEPs or groups from different countries. For communication with the members of an expert group or a working group, scientific reports are usually more powerful vehicles than mass media, but for communication with the mass public their status is almost the reverse. For making public a political demand or an offer, a press conference in central Brussels is a useful channel, but for the pursuit of inside negotiations it is usually disastrous. The power of a channel depends partially on the relationship between the sending and the receiving side. It is strongest if there is balance of interests between the two sides, by which the message can indeed be delivered to the right place, at the right time and with the planned form and contents. The two sides have to be connected, of course. This can happen in many different ways, as shown in figure 7.2. Some ways are more powerful than others, but they are usually less so by their properties than by specific circumstances related to the arena and the life of the dossier. If there are many opposing stakeholders, it may be better to use, for example, channels in parallel rather than in series, so decreasing the chance of failing transmission. Even better, in case of such opposition, may be to combine parallel and indirect channels, so concealing one’s identity as the sender. Opponents may try to reverse the current, for example by issuing a counter-message through the same channel, so causing severe confusion. Or they may try to change the EU playing-field, for example by playing Triple P and installing veto points, thereby putting the sender’s desired place and time out of reach. The possibility of manipulating the power of a channel is limited as this partially belongs to the channel rather than to its users. For example, a general channel like the mass media is extremely difficult to keep under control. Even a specific and closed channel like an intranet is not easy to keep fully closed in practise. The limited room to manage the power limits requires good preparatory work. This helps to identify which channels are most powerful and how their power can remain under some control and 298

Chapter 7

The Limits of EU Public Affairs Management

perhaps be made immune to opposing actions. For obvious reasons, outlined in the previous chapters, even this homework has its limits. However, any difference in study and prudence can make a lot of difference for the desired outcome. The amateur group neglecting this homework uses the power of a channel like a novice sailor. Wanting to transmit its message to ‘Brussels’, it goes, for example, directly to the mass media or to the national government, with the risk of becoming overpowered. As a trained sailor, the professional group prudently tries to avoid such outcomes. It fears falling at the mercy of the waves and usually prefers a channel with power and safety in balance. However, the professional has no guarantee, but only a better chance of not being sunk.

2 Quality Some channels keep a message intact during transmission, while others do so much less. In the perfect case both the form and the contents remain unchanged and the message is delivered only at the planned place and time. Frequently, however, the transmission is of poor quality. Several variants of disturbed transmission can be distinguished. One is that the channel adds signals that affect the form and/or the contents of the message. Mass media produce a lot of channel noise. A message can be distorted or changed when going via central government, due to bureaucratic conflicts or, euphemistically called, national co-ordination. A second variant is that parts of the form and/or the contents are lost during transmission. A channel usually contains some filters, locks and lock-keepers. For example, in an EP session, the rules of order, the timetable and the clerks can be selective in recording messages. A third variant regards the delivery by place and time. The message may end up at a different address by going through side-channels. Many a company or NGO trying to send some idea through its EuroFed to the Commission sees it falling off the table, not reaching the targeted official or being consumed by someone else. In a fourth variant the transmission is kept intact, but not exclusive. The channel may have a memory system, which can be tapped by others. A formal letter to or from the Commission may arrive perfectly at its destination, but due to its registration it might be noticed, identified and read by other stakeholders and journalists. Channels have a large amount of quality variation. The informal faceto-face contact in, for example, a quiet restaurant is the channel with frequently the highest potential for quality. The time and place are fixed, the form is determined and the contents are directly transmitted from mouth to ears. But the meeting can be leaked to or be noticed by other people, such as the secretary, the chauffeur, an opponent or a journalist passing by chance. Even some traces of the meeting, such as the invoice or a scribble, 299

The Limits of EU Public Affairs Management

Chapter 7

can be found by others. All other channels contain even more risks of distorted transmission. In contrast to the quiet restaurant, a press conference is like shouting out the message. Journalists attending the conference can interpret the message as they want. Every channel poses limits by its risks of faulty rather than intact transmission. No interest group on the EU playing-field can operate without channels, however. Every group is dependent on others and has to communicate. To some degree an interest group can manage the quality limits and make them slightly wider or narrower. The sender can take measures to check the quality of transmission during the process and afterwards, but doing this in advance is more difficult. It is much easier for an opponent to disturb a transmission than for the sender to prevent this, as it is easier to spread rather than to stop rumours. Amateur groups suffer more from channel limits than professionals. The latter have a better knowledge of the usual quality of a channel than the former. For many situations they know that they should prudently choose a less risky channel, take some measures of quality control and maybe widen a few limits. When playing the role of opponent they know how to narrow the limits of a channel used by the enemy that sends a message. They are aware that such an act of channel management must be managed prudently, as otherwise it may return a boomerang.

3 Configuration Seldom does a channel connect only two or more places. Usually it is part of a wider configuration of transmission elements created by nature or mankind. The main ones are capacities, flows, regulators, collectors and sites, as in the following examples. The Commission’s free capacity (what is left after the current workload) to channel new political demands is very limited due to its small staff. The volume or contents of new messages easily overload it. Traffic jams at the Commission and/or overflows to other channels are the usual consequences. The EP is renowned for its multiple and parallel flows to and from institutions, countries and interest groups. Some flows it accelerates through an inter-group like that on animal welfare, others it delays, as happened to the GMO dossier in the late 1990s and yet others it converts into a circular flow like that on workers’ codetermination in the 1990s. The multiple and parallel flows can easily collide inside the EP, which then may adopt contradictory resolutions that, if noticed, shock MEPs and the mass media. Every channel has regulators that propel, slow down or redirect flows. Some are inherent to the flow, such as the French Council initiative on the ‘financial crisis’ in autumn 2008 that at first proceeded forwards at high speed, then was slowed down by traffic-jam300

Chapter 7

The Limits of EU Public Affairs Management

ming counter-proposals and finally was redirected by multiple circular discussions. Others are constructed, such as the secretariat of a EuroFed and the white-books of the Commission. Channel collectors are, among many others, the chef de dossier of the Commission, the rapporteur of the EP and the PA official of an interest group. Virtual collectors are the memory storage, the bureaucratic pipeline and the policy dossier. The sites of a channel refer to the areas around, such as expert groups around the Commission. They connect with other channels, provide room for overflows and feedback loops and receive all sorts of more or less fertile residues. All such elements, together forming the channel configuration, may be considered limitations or facilities. On the one hand, they put limits on perfect transmission, as much can go wrong due to the channel system. A local government, a Commission unit or a multinational NGO, wanting to send a message to an EU stakeholder, may end up in a traffic jam, circular flow, dried-up channel, pipeline or overflow. On the other hand, all these elements are three-fold facilities as well. Firstly, they enable one to monitor the channel process, because every element is a potential indicator that, after homework, can contribute to better fieldwork. Secondly, all limitations also exist for the opposing stakeholders. One has seldom any reason to fear that their messages will go through whatever channel smoothly and perfectly. By monitoring the configuration one can observe their bad or good transmission and respond accordingly. Thirdly, as partially manipulable variables, the elements provide some room for configuration management and engineering. An example is the Triple P game, which creates a new configuration by the early instalment of lock-keeping persons, positions and procedures. The configuration elements place limits on every sender, but fewer on professional interest groups than amateurish ones. Professionals have better management skills. By monitoring a configuration carefully and watching the streaming closely, they can take advantage by getting wider limits for themselves and narrower ones for their opponents, or the reverse. Even a small change, for example regarding a deadline or destination, can strongly improve the outcome they want. Professionals can, however, never fully solve all channel-related limits and their consequences, as many fall outside their control.

V The Limits of Arena Management The four main components of every arena (chapter 4-IV), are issues, stakeholders, dossier life and arena boundaries. Each of them can pose limits to every interest group every time a new conflict arises. As contested facts, 301

The Limits of EU Public Affairs Management

Chapter 7

values, instruments or solutions, the issues are claimed by many stakeholders. All stakeholders want to become part of the winning coalition and not the losing one. The life of a dossier is always unpredictable and thus largely falls outside the control of a single group but not necessarily of a collective platform. The boundaries of an arena are normally open, thus allowing entry and exit of issues and stakeholders. Each interest group, always just one among many, has every reason to feel overwhelmed more by the limits than by the opportunities of EU arena management. From its many limits and continuing their discussion in chapters 1 and 2, we select the following three: pluralism, complexity and dynamics.

1 Pluralism Europe is full of irritating differences regarding perceived facts, preferred values, proposed instruments and cherished solutions. Many interest groups believe, like common people, that both the differences and the irritations come from outside. Any irritating difference or issue is, however, embedded in the public and private interest groups themselves and fought among and inside them, as can be found for EU institutions, domestic governments, NGOs and companies. Their competitions come close to a civil war, no matter how civilized. Each group can have only marginal grip on the war’s development and outcomes. It may try, of course, to push or to block the passage of an issue or a dossier. Many competing stakeholders will try as well, with the result that the dossier’s life frequently ends up like a truck with many drivers behind the wheel, steering in different directions. It is also an open truck, easily permitting passengers to jump on or off. Its trailer-load of stakeholders and issues is seldom fixed. In this pluralistic setting many stakeholders are eager to develop their own networks, coalitions and cartels, in order to get a better grip on the arena and its outcomes. Even the most corporatist arena allows the insiders only limited control, as there are always challengers inside and many more outside. The pluralistic character of the EU poses significant limits to every interest group. An arena is seldom inviting: only the stakes may attract, as the grapes did Tantalus. The interest group nonchalantly entering an arena runs all the risks of becoming isolated or even victimized, so it must behave prudently. Any issue can be a threat or opportunity, every stakeholder can become a friend or enemy, every dossier can come to a good or bad end, and any boundary can become wider or narrower. The professional player, behaving prudently, tries to figure out all these possibilities beforehand. Clues for drawing conclusions can be taken from the preparatory homework, which may also reveal which stakeholders are acting prudently and therefore have to be taken seriously. The professional group 302

Chapter 7

The Limits of EU Public Affairs Management

can discover that, thanks to the pluralistic nature of the EU, there are always some beneficial issues and friendly stakeholders, both inside and outside the EU institutions. It only has to discover them by doing its homework, where after it can more or less manage them by fieldwork. In this way, it can even contribute to an increase or a decrease of the arena’s pluralism, for example by launching a new issue or by compromising on an old one. But it can never dictate the desired outcome, and neither can any competing stakeholder. This is the essence of pluralism. The professional understands this better than the amateur and has a better chance to achieve a desired outcome.

2 Complexity Every arena is like a labyrinth and even a seemingly small issue is often complex. The contested facts, values, instruments and solutions are related to different interest positions, each being right in their own way. Even such a black-and-white dossier as the Tobacco Ban Directive (2001) is as complex as an ancient Greek drama, with all parties fighting for noble but irreconcilable values: one for liberty and pleasure and the other for safety and health. Most stakeholders have complex structures as well. The Commission is full of both horizontal and vertical segmentation, not to mention its assistant bureaucracy of a few thousand expert groups. Almost always an interest group has a formal skeleton that is different from its real flesh-and-blood. This is also true for the relationships among interest groups. They range from rather formalized EuroFeds to informal ad hoc networks, and frequently they cut across both institutional and territorial borders. This makes it difficult to identify their membership and has resulted in such a vague concept as governance [Kohler-Koch, 2003; De Schutter and others, 2001; COM European Governance, 2001]. The dossier’s life is no less complex. It can suffer a premature death and it always has its mini-loops during institutional phases. EU officials and stakeholders may take part in any different phase of a dossier. Comitology even looks like ‘a dark glass’ [Bradley, 1998]. Combining dossiers by package dealing may create even more complexity. Last but not least, the arena boundaries are seldom fixed, rigid and concrete and are usually variable, open and abstract, in short complex. They almost always cut across those of other arenas, thus facilitating even more combining of issues and dossiers. The amateurish interest group can easily lose its way in the labyrinthlike arena and plead for reduced complexity. The professional player, however, understands both the limits and the opportunities caused by complexity. Assuming that every labyrinth has a structured pattern, however irregular this may be, it tries to discover this. Thanks to the complexity it 303

The Limits of EU Public Affairs Management

Chapter 7

at least enjoys the extra room for playing created by those who lost their way. It discovers that there is always more than one front door, back door, hopper window, passage, connection, bypass, exit or escape. The expert can even try to increase the complexity for others and especially for the opponents, for example by engineering a fake entry or another dead end. The professional group remains conscious about the limits of all advantages from complexity and is prudent about their use. It may lose its way as well or fall into its own trap. Complexity sets its limits to all, but to some more than to others. This margin can be crucial.

3 Dynamics To some extent an arena is dynamic like a trampoline. Newly chosen values, perceived facts, proposed instruments or preferred solutions may cause issues to change rapidly. A main actor in this is the Commission. It repeatedly launches new decision values, research reports and policy proposals, such as those on financial stability in 2009, or it reframes old ones, such as the shift from production-oriented to sustainable agriculture. Interest groups have changing agendas too, in the best case due to the reconstructions of their long-list, shared-list and short-list. Their organizations change permanently under countless pressures from both outside and inside, ranging from reorganization, growth and decline to reputation and turnover of personnel, which altogether change their capacities and performance. NGOs may beget a GINGO or a BINGO. All stakeholders can break off an engagement with others and enter a different one. The life of a dossier is full of dynamics, as stakeholders try to push or to block its course and do this in different directions. By influence techniques like negotiations and reframing they can kill a dossier or bring it back to life. Many a dossier spills over into different policy areas. The arena boundaries are flexible too. Old issues can fade away and new ones arise, as happened with the Biotech dossier in the late 1990s, when its core issue changed from product improvement to human safety, which also changed its composition of stakeholders. In every arena the interest group has reason to fear the trampoline, as it can easily trip or fall off it. Expertise is, of course, rewarding here too. Thanks to the dynamics, the professional group, recognizing some patterns of change, can survive longer and may even make a leap forwards. It has comparative advantage over the amateurs, who often ask for a more static arena. It might try to make the arena even more dynamic, for example by pressuring for earlier deadlines or provoking new issues and stakeholders to enter the arena. The professional group that creates scenarios is prepared for change and can thus anticipate events. Even in bad times it 304

Chapter 7

The Limits of EU Public Affairs Management

knows that a better moment will always arrive. By its management of time it hopes to promote the desired outcome. In spite of all these potential rewards, even the most professional group is limited in its possible actions by the arena dynamics. It has to respect the natural forces of any boisterous and wild arena, but through prudence it can manage them better than the amateurs can.

VI The Limits on the Receivers’ Side Receivers include not only stakeholders (such as EU officials), but also outside groups and the mass media. The outsiders, not being on the ‘mailing list’ of the sender, may have got the message by accident, echo or leaking channel. They may turn into stakeholders and limit the effectiveness of a message. If a message has been sent, channelled and gone through the arena perfectly, it may have been poorly received by the targeted receiver, due to, for example, its internal affairs, confusion or indifference. Or it has been received correctly but interpreted wrongly, in which case it assumes a different urgency, procedure or meaning. Even if received and interpreted correctly, it can still fail if the receiver overreacts, does not react at all or does the opposite of what is intended. All this can happen not only with a message that has form and content, but also with a non-message, such as a non-verbal action or the silence of the sender. To put all this in terms of communication theory: the message may have been incorrectly registered (syntactic filter), understood (semantic filter) or taken into account (pragmatic filter) [Cherry, 1966]. This threefold distinction runs parallel to that between the input side, throughput and output side of any body. From the many receiver-related limits, we select three here: inattention (syntactical), satiation (semantic) and neutralization (pragmatic).

1 Inattention Just as a sender can have his mental and organizational limits, so can a receiver have his emotions, dogmas and myths that cause syntactical filters and lead to selective perception of a message, by which some signals are neglected and others over-registered. A receiver suffering from internal dissent, scarcity or lack of leadership can be equally inaccurate in his reception of a message. The syntactical case of inattention may have many causes. For example, an EP rapporteur with a bias in favour of a particular outcome is easily blind to the counter-message provided by an opponent sender. A national ministry that believes itself to be perfectly right on its policy position, will for long remain deaf to any warning from outside. A 305

The Limits of EU Public Affairs Management

Chapter 7

national parliament that holds the myth that it is the Council that decides, shall neglect what happens before and after the Council phase. Small-sized interest groups can be inattentive due to their limited resources and their belief in the myth that the EU is not so relevant. Many groups have the experience that both their strong opponents and close allies hardly take in their messages, as the former may not be at all receptive and the latter may feel glutted. Most inattention is probably caused by a poor internal organization, such as internal dissent, reorganization or a PA function that is scattered over various units and people, not linked to the Board or simply unmanned due to vacancies or staff holidays. Every receiver can have at its input side a syntactical filter of inattention. Amateurishly organized receivers tend to be more inattentive than professional ones. Their antennae are poorly attuned to the stakeholders that are potentially relevant to them. Often they lack qualified people for monitoring or they are unaware of their mental and organizational limits, which make them deaf or blind to a message. By their poor intake they limit the effectiveness of their own possible response. The way the Commission usually organizes its input side, namely by launching public calls to interest groups and keeping open its doors for them, makes it a most professional receiver of messages. In contrast, some parts of DG SANCO and DG ENVI lose professionalism when inviting only ‘entitled groups’, being their allies on health and environment. Amateurs irritated by an inattentive receiver may be inclined to either send a stronger message or renounce the receiver altogether. Through preparatory work, the professional senders, in contrast, try to anticipate the attention limits of targeted receivers or to take advantage of these. They might even send fake messages indirectly, in order to test a receiver’s attention. When regretting the receiver’s inattention, they try to repeat the message prudently, for example, indirectly and quietly. But even the best professional cannot prevent or remove all causes and cases of inattention among all targeted receivers. Nobody can be seen or heard by those who remain blind and deaf.

2 Satiation A regular case of the semantic filter on an incoming message is that the receiver may soon have had ‘enough of it’. Most received information has a rapidly decreasing marginal utility or, to put it differently, an increasingly marginal irritation value. This explosively outgrowing indifference for ever more information exists in politics as well as economics. Receiving a message twice, in the same form and with the same contents, is in the best case taken as an indicator of disorganization on the part of the sender and in the worst case as an insult to the receiver. MEPs frequently complain 306

Chapter 7

The Limits of EU Public Affairs Management

about repetitious lobbyists, as the latter often do about the former. Most receivers feel bored with senders who believe that only they are perfectly right on an issue and who send their argumentation again and again. The reiteration of the same findings of research is often counterproductive by its provocation of counter-research. The nicest Lorelei labels used for upframing an interest (figure 6.2) tend to become obsolete after a few years. Any over-lobbying easily activates the filter of satiation at the receiver’s end and easily causes decreasing interest in both the message and the sender and increasing irritations. Satiation, in short, always lies in wait. Satiation can always become active during the throughput of a message. Amateurishly organized receivers are most vulnerable to this. When they receive a message, they may read it only cursorily, thereby missing its points, or they believe to have received it already. Feeling satiated, they can easily feel fed up and irritated, which weakens their attention capacity even more. Professional receivers, however, take even the reiteration of a message as information and like to know whether it is caused by, for example, nervousness or disorder on the part of the sender or a channel’s property, such as an echo. Then they can benefit from these findings by soothing the sender, exploiting his disorder or using the channel’s property. The senders can also react in different manners to the satiation of targeted receivers. Amateurish senders may react in irritated (and irritating) ways, such as by repeating the message louder or blaming the receiver. Professionals, in contrast, prefer to send their message again in a new form, through a different channel or even ‘polyphonic orchestra’ or via a platform, such as a BONGO, a federation or an ad hoc coalition. In their R&D on PAM they search for new vehicles. They also gain free advantages from the irritations of opponent receivers, for example by indirectly making public their amateurism. However, they cannot fully manage the risk of falling victim to the satiation of a targeted receiver. All receivers have their moments of satiation and indigestion.

3 Neutralization No sender is the only one sending a message sent to a targeted receiver. On every dossier, the latter receives different messages from mutually opposing senders. Usually the receiver also has pragmatic filters that allow different options for response, going from doing nothing to meeting the request or doing the opposite of what is suggested. This freedom of choice is, of course, limited by, for example, its own interests, current commitments or power positions, but may be increased by the paradox of crosspressures: the higher the number of contradictory messages, the lower the impact of each tends to be and the more freedom of choice. The many dif307

The Limits of EU Public Affairs Management

Chapter 7

ferent messages produced by competing interest groups on a single stakeholder or official should, in mathematical terms, not be added together or multiplied as is often done by anti-lobby NGOs [Burley and others, 2010] and naïve academics [Klüver, 2011], but divided by each other as they give the receiver room to manoeuvre. The Commission’s chefs de dossier and the EP’s rapporteurs often enjoy this paradox. The 2011 fights over the draft regulation on Food Information to Consumers (2008/ 40) produced almost 700 different amendments that enabled the rapporteurs to decide at will, after which there appeared not to be clear winners or losers as all groups had both won and lost something. Stakeholders too receive messages from many others inviting them to take their side. If one offers support to a targeted receiver, another stakeholder may overbid. If one presents some evidence, another may show it to be false and refute it. Research is always vulnerable to neutralization, as delivery of counter-evidence may take only three months. If a message gets public attention, then it may become neutralized in mass debate. The lobby to limit the import of cheap textiles from third-world countries by the up-frame of ‘child labour’ was soon neutralized by the mass media, which reported that those children had in fact experienced an even worse fate, as they went to work in crime or prostitution instead of factories, let alone attend schools. To some degree, every receiver has a pragmatic filter of neutralization at its output side. Professional receivers may feel happy with cross-pressures that allow room to choose. Under the flag of ‘democratic consultation’, many EU officials are eager to stimulate more stakeholders to send their views, as this not only saves them much monitoring work but also gives them, thanks to incompatible replies, more freedom of choice. So did the EP in 2011, when it successfully proposed to establish Finance Watch as an EU GONGO to neutralize the bank lobby of EBF. Amateurish receivers, however, often feel unhappy with cross-messages and find it difficult to exploit the room of choice. Neutralization may bring freedom or embarrassment to the receiver but also to the sender. Amateurish senders, dreaming of being the only one, usually regret the neutralization limit, except when their opponent is incidentally neutralized. Their message being refuted within three months makes them upset and maybe aggressive, thereby causing new mistakes. Professional senders, however, consider the neutralization limit as normal for any pluralistic system. As part of their arena study, they explore the probable cross-pressures in advance. Their homework helps them to immunize their own messages against cross-messages and to produce their own cross-messages against opponents in good time. To achieve better bargaining positions, they may simulate or stimulate new messages that cross the counter ones. On the GMO dossier (chapter 4-X), the counter-messages from groups like Europabio 308

Chapter 7

The Limits of EU Public Affairs Management

finally legitimated the Commission in its decision not to closely follow Greenpeace’s line. But no sending interest group can always or fully control the neutralization limit, as this belongs to the receiver.

VII The Limiting Environment Seldom is a PA communication system a closed one that only contains senders, channels, arenas and targeted receivers. Usually it has openings and gaps both to and from the outside. A sender may fail to keep his message in a closed circuit, for example when it, feeling weak, looks for outside support or when a journalist smells smoke. Channels can produce all sorts of echoes and noise that attract attention from outsiders. Stakeholders and issues can catch the eye of the outside when crossing arena boundaries. Senders can be accountable to an open forum, as happens to Commissioners in the EP and to Council ministers in their home parliament. There is also reverse openness, from the outside to the inside. Outside stakeholders that hear about a dossier, may want to interfere, in order to influence or to cash in on nuisance value. Interest groups that are weakly established, such as many SME alliances, citizens’ groups and those from new member states, often do so at a late stage loudly. Journalists from the mass media or the professional press love to enter an arena and nose around, to please their public. Domestic politicians, being sensitive to such reports, often urge their government to interfere in the matter. The outsiders bring in their own facts, values, instruments and solutions that may come to life inside the arena. Then they push or block new agendas, decision practises or policy lines, so contributing to the widening of europeanization. We now present the following three examples of the limit-setting environment: reputation, prescripts and outside groups.

1 Reputation The reputation of a person or a group is not a stable property, but a temporary assessment by others according to their current norms and morals. Mass media and the public play major roles in making or breaking a reputation. Different from stakeholders and officials (chapter 5-II-4), they take reputation as trustworthiness only and not, sympathizing with David rather than Goliath, also as an indication of importance. Many interest groups have learnt to care about their reputation, as that can have a severe impact on a group’s licence to operate in the EU, as other stakeholders and officials are reluctant to deal with a scandalized player (such as once with the controversial tobacco producer Philip Morris). It is far easier to dam309

The Limits of EU Public Affairs Management

Chapter 7

age than to establish a good reputation; after all, having a good reputation largely depends on the values and views of outsiders. Common people attribute an aura of ‘general interest’ first to government groups, secondly to NGOs and least of all to companies [Eurobarometer, 2000, 98]. This aura may, in objective terms, be false as, for example, retailers may contribute more than NGOs and governments do to a general interest like the supply of healthy food, but the subjective view matters here. Therefore many companies set up a BONGO and many NGOs a GINGO, so improving their image. A good reputation is helpful but never sufficient for getting a licence to operate, however. An interest group that does not respect, for example, procedures, deadlines or demand sides is easily vilified and neglected, as happened shortly to Unilever in 2007 when it had missed deadlines of the regulation on Health Claims (COM 2003/165). The fastest way to get a bad reputation is to lobby in ways that are (or may be) seen as dirty, sneaky or tricky. Since 2005, NGOs like Spinwatch and Friends of the Earth have organized online their ‘Annual Worst EU Lobbying Award’; in 2007 it was awarded to German car producers. To neutralize this, a few PA groups started the ‘EPAD Awards in PA Excellence’ in 2008; in 2009 it was won by AmCham. NGOs can be scandalized too, as happened to Corporate Europe Observatory (CEO) for its ‘one-man-band show’ without citizens. Reputations are now the targets of PA and lobbying and thus contested. Amateurish interest groups tend to be nonchalant about the limits of reputation. If they have a good reputation, they tend to treat it as a fixed asset and, if they have a bad one, as if ‘nobody cares’. When wanting to pick a pretty flower along a ridge, they take high risks of reputation as they may topple into the gorge of fallen names. This fate befell Greenpeace after it had provided false information about the Brent Spar oil platform in the mid-1990s, and to Alter-EU after its biased study in 2008 [Burley and others, 2010] that blamed the Commission for its expert groups as being ‘on the side of businesses’. Groups and people that apply domestic practises of lobbying in the EU sometimes discover – painfully – that what works at home may cause scandal in the EU. In the ‘corruption index’ for Europe, most countries score much worse than the EU [Kowalczyk-Hoyer, 2012]. In 2011 the British Sunday Times caught three MEPs willing to be bribed and soon all three had to leave the EP. Professional groups know that outsiders can watch them and thus they prudently allow any flower on the ridge to stay. They respect the various values and morals, maybe not with conviction but at least out of fear of being caught and scandalized if offending them. Their Praxismoral, as it is called in Germany, is based on preparatory work and today up-framed as Corporate Social Responsibility [Mahon and Wartick, 2012; Phillips, 2011; Baines and Harris, 2006]. They know that lobbying styles allowed in one country, sector or arena can 310

Chapter 7

The Limits of EU Public Affairs Management

cause scandal in another. They manage their reputation, for example by up-framing their contested interests, outsourcing risky behaviour to others or creating an image of openness. If they are swept up in a scandal, they soon seek to present the face of integrity in order to attract friendly attention, as Shell did after Brent Spar in its 1999 ‘Listening and Responding’ campaign, which aimed to establish a better dialogue with stakeholders. Prudent groups try to build up an even higher standard in meetings with selected ‘good guys’. However, even with the most prudent anticipation and management no interest group can fully control its reputation, as this is ultimately set by others.

2 Codes of Conduct In 1991, the Dutch socialist MEP Alman Metten privately published The Ghost of Brussels, in which he reported that during three months he had received almost 150 letters from interest groups, 90% of them from groups representing trade and industry. He concluded that lobbying is neither transparent nor balanced and he proposed both a register of lobbyists and a code of conduct. His action resulted in much mass-media attention and, in 1992, the first EP hearing on lobbying (Doc En/CM/118767), followed in 1997 by the first EP rules on a register and code of conduct for lobbyists. The Commission followed the EP by issuing codes of conduct for its personnel and registration forms for outsiders. In 2005 the Commission proposed its European Transparency Initiative (ETI), aimed at tighter registration and conduct rules for interest groups and lobbyists, and it invited the EP and Council to join it. In 2008 the College adopted the ETI and the EP established an even stricter version (the Council abstained). In 2011 the Commission and the EP agreed upon a common public register, a code of conduct and a complaint mechanism (OJ/ L 191/29, 22-07-2011). The agreement defines lobbying as activities ‘carried out with the objective of directly or indirectly influencing the formulation or implementation of policy and the decision-making processes of the EU’. It covers most interest groups but explicitly not governments, intergovernmental platforms, diplomatic missions and, under some conditions, entities like political parties and regional and local authorities, so exempting almost all lobbying by groups and people from public government (in spite of all their similarities with those from private society). The register is formally voluntary, but practically obligatory as it is published online and watched by other interest groups, NGOs on transparency and mass media, for example the British State Watch and the German Lobby Control. The code of conduct (figure 7.3) primarily requires openness on interests and accountable behaviour; the EP has added a few codes more. The 311

The Limits of EU Public Affairs Management

Chapter 7

Code of Conduct (2012 COM + EP) In their relations with the EU institutions and their Members, officials and other staff, registrants shall: a. always identify themselves by name and by the entity or entities they work for or represent; declare the interests, objectives or aims promoted and, where applicable, specify the clients or members whom they represent; b. not obtain or try to obtain information, or any decision, dishonestly, or by use of undue pressure or inappropriate behaviour; c. not claim any formal relationship with the EU or any of its institutions in their dealings with third parties, nor misrepresent the effect of registration in such a way as to mislead third parties or officials or other staff of the EU; d. ensure that, to the best of their knowledge, information which they provide upon registration and subsequently in the framework of their activities within the scope of the register is complete, up-to-date and not misleading; e. not sell to third parties copies of documents obtained from any EU institution; f. not induce Members of the EU institutions, officials or other staff of the EU, or assistants or trainees of those Members, to contravene the rules and standards of behaviour applicable to them; g. if employing former officials or other staff of the EU or assistants or trainees of Members of the EU institutions, respect the obligation of such employees to abide by the rules and confidentiality requirements which apply to them; h. observe any rules laid down on the rights and responsibilities of former Members of the European Parliament and the European Commission; i. inform whomever they represent of their obligations towards the EU institutions; Individuals representing or working for entities which have registered with the European Parliament with a view to being issued with a personal, nontransferable badge affording access to the European Parliament’s premises shall: j.

comply strictly with the provisions of Rule 9 of, and Annex X and the second paragraph of Article 2 of Annex I to, the European Parliament’s Rules of Procedure; k. satisfy themselves that any assistance provided in the context of Article 2 of Annex I to the European Parliament's Rules of Procedure is declared in the appropriate register; l. in order to avoid possible conflicts of interest, obtain the prior consent of the Member or Members of the European Parliament concerned as regards any contractual relationship with or employment of a Member's assistant, and subsequently declare this in the register.

Figure 7.3 312

Chapter 7

The Limits of EU Public Affairs Management

complaint mechanism can lead to various sanctions, such as removal from the register and loss of access badges. The agreement can be seen as a collection of semi-formal prescripts in response to both social demands and the desires of the Commission and the EP to get rid of irritating ways of lobbying [Burson-Marsteller, 2003 and 2001], but not of lobbying itself, which is explicitly praised as a positive contribution to EU decision-making. The prescripts are much like those that exist in a few member states (Germany since 1972 and Hungary, Lithuania and Poland since the early 2000s) and in countries outside the EU [Chari and others, 2010; Bertók, 2008; McGrath, 2008]. In addition, most EU institutions have their internal regulations regarding their own (former) officials and politicians at the receiver’s side of lobbying [Demmke and others, 2007]. Thus, both sides of lobbying have been made more transparent, so stimulating social control, the ‘mother of all control’. Some platforms of lobbyists, such as EPACA and SEAP, have made their own code of conduct as is requested by the agreement but also works as a tool to create an even better image. Amateurish interest groups may believe that everything not forbidden or prescribed is permitted and may conclude that the prescripts are mildly and loosely formulated, thus leave much lobby freedom. Acting nonchalantly, they may easily expose themselves to scandal and even lose their licence to be in the official arenas. Professional groups interpret the prescripts strictly, monitor their application closely and respect them prudently. They consider them as a blessing (in disguise or not) and some even ask for stricter prescripts [ComRes, 2008], as the prescripts help to gain a more respected and seemingly legal status and to keep amateurish groups at a distance. Of course, they try to influence the prescripts, which are always open to reformulation. They regularly renew their own informal codes and offer these to the institutions. This sort of lobbying is no different from influencing whatever new regulation comes along and is closely watched by NGOs concerned about transparency. The prescripts have become a new category of limits. They exist for all interest groups, but the professionals apply them internally better than the amateurs. This margin of prudence can also make a crucial difference for the desired outcome.

3 Outside Groups There are always newly arising interest groups that sooner or later enter the playing-field. Usually they start via their national platform with an indirect representation at a suitable EuroFed, after which they may open a direct facility. In the early 1980s only very few MNCs had a facility in Brussels. Soon they were followed by more facilities opened by NGOs, region313

The Limits of EU Public Affairs Management

Chapter 7

al and city governments and, gradually, national agencies. Interest groups from outside the EU area were almost completely absent, without direct or indirect representation, and most mass media organizations had only a part-time reporter in Brussels, to cover Belgium, NATO and the EU altogether. A few decades later, numerous interest groups from decentralized governments and civil society have both a network of indirect linkages and a direct presence, either part-time or permanently. Hundreds of nonEU interest groups now have their facility in Brussels and a few mass media have set up a permanent desk to cover the EU alone. Since the mid2000s, more new interest groups enter via their domestic platform being part of a EuroFed, such as from SMEs, local governments, domestic agencies and social groupings like the elderly and even the immigrants. Even more new outside groups can be expected in the future. Any EU enlargement brings in new interest groups with their different issues of, for example, religion, literacy, trade, income and tradition. Following the Lisbon Treaty, in 2011 the European Citizens Initiative (ECI) was adopted, by which regulation (OJ, L 65/2, 11-03-2011) at least one million citizens from at least one-quarter of member states are entitled, if meeting the procedure’s requirements, to submit a legislative proposal, so breaking the Commission’s privilege on this [Garcia and Greenwood, 2012]. After an early 2010 initiative of Greenpeace calling for a GMO freeze, the first accepted ECI came from the platform F2020 that calls for spending 10% of EU funds on cultural mobility, like the Erasmus programme on education. The ECI may bring more new interests of ordinary citizens into the EU. Like Greenpeace, established groups may push them, for example NGOs (like on health or work), companies (on food or pharmaceuticals) and regions (on public services or regional autonomy). Even without ECI, new interest groups will always enter the EU and may change current balances. The newcomers may pose new lobby limits for the insiders. At the very least, they want to be taken into account. Amateurish groups inside the system, seeing EU decision-making as a zero-sum game and thus fearing to lose what another group receives, consider newcomers to be intruders. Before the fifth enlargement in 2004, many Spanish interest groups were worried about losing their EU subsidies to applicant states, and many member states feared having to give in on voting points, EP seats and other privileges. Professional groups inside the EU are more prudent. Considering decision games to be variable-sum, they take their old subsidies and privileges as bargaining chips for other desired outcomes. New interest groups, such as immigrant shopkeepers, retired people or applicant states are assessed as potentially useful stakeholders and included in their preparatory work and anticipatory lobbying. In the late 1990s, Dutch interest groups on road transport, including the 314

Chapter 7

The Limits of EU Public Affairs Management

ministry, supported the Hungarian trucking sector, then their main competitor outside the EU, so making early friends for after enlargement. Professionals also try to exploit the issues promoted by new groups by using these as leverage for blocking, pushing or changing current EU policy lines. If they feel to be in an unfriendly EU arena, they may embrace newcomers that can change the arena elegantly in their favour. Being more proactive than the amateurs, they may harvest opportunities brought in by outside groups and turn imported problems into virtues.

VIII Extra: The Limited National Government To every interest group one can apply the typology of SCARE and thus to the national government too. Defined narrowly, this is a country’s central government as composed of cabinet, parliament and ministries, in short the national capital. The broad definition also includes forms of decentralized government such as (territorial) municipal and regional governments and (functional) more-or-less independent agencies. By written or unwritten constitution, the exercise of power is divided among several institutions, which makes every part of national government in Europe almost synonymous with limited government [Friedrich, 1974]. Here we shall follow the narrow definition and focus on limits of the national capital in its relationships with the EU. For each capital the assessment varies, of course, with time, place and other circumstances, as it does for every interest group. Some general observations can, however, be made. When sending messages to the EU, many a capital and particularly its parliament holds the dogma of national co-ordination and the myth of Council decision-making. The capital’s internal organization by conglomerate structure, multi-layered operations and internal pluralism limits its stable sending of clear messages and, when not briefing and debriefing its policy experts on the EU work-floors [Schneider and Baltz, 2005], it has no channel at all. Instead, it often uses open and noisy channels that start in parliament and mass media at home and are out of control thereafter. Its complex and dynamic pluralism at home is rarely secret to competitors, such as foreign capitals (embassies), EU institutions and other interest groups. At the receiver’s side it finds its messages often falling on satiated soil in the Council and it is seldom warmly welcomed by the Commission and the EP. It has to cope with countless domestic interest groups, including decentralized governments, that easily challenge and bypass it by going to the Commission and the EP directly. It faces three disadvantageous limits that are typical for any national government that is democratically formed: transparency, pressures from elected politicians and the dogma of general interest. 315

The Limits of EU Public Affairs Management

Chapter 7

1 Transparency National capitals score, willy-nilly, high on transparency. Their internal processes and workings may largely remain behind closed doors and be much less transparent than those of the Commission, but in every country many organizations are able to download the information from inside successfully. Journalists are eager to gather the information by professional window-out lobby techniques. Academic researchers are often permitted to investigate behind the scenes. Interest groups have in the ever divided capital always friends (and foes) inside the ministries who inform them off the record. They ask members of parliament (MPs) to gather, as ‘assistant lobbyist’, information and wider support. The MPs of opposition parties have a selfish interest in enhancing domestic transparency, also on EU affairs, as they want to oppose the government’s positions in the EU too [Kumlin, 2011]. In short, many facts, values, instruments and solutions relating to past, current or planned decisions and policies are leaked out and disputed openly. The national capitals thus have an awkward position as both sender and receiver of messages, by which confidential information easily becomes public. In comparison, private groups such as companies and NGOs have an easier position on transparency. Most strategic and operational decisions do not have to be made public. Outsiders are seldom legally entitled to demand inside information. Shareholders and other stakeholders normally receive only summary information through yearly reports and official statements. Some information may leak out or be obtained by journalists, but its volume is modest in comparison to that from the capital. The tragedy of government transparency is that it gives the capital, compared to private groups, a disadvantage in the EU. Intelligent interest groups can rather easily get informed about its internal divisions and positions and thus anticipate and exploit them. MNOs can do this for many capitals efficiently through their country units. From their side, the private groups, being rather well-informed about their capital, are hesitant to provide crucial information to it, as this functions as an almost open and often noisy channel to the wider public and thus to their own competitors in the EU as well. If providing such information at all, they want to receive in exchange some substantial support in the EU or at home. The capital can, however, not deliver much to them via the EU, as (long as) it is fixated on the Council, where it is one of 27 that decide in ‘about 15% of the legally binding proposals’, which is one of the reasons why it is increasingly bypassed by the domestic interest groups.

316

Chapter 7

The Limits of EU Public Affairs Management

2 Pressures from Elected Politicians The pressures from elected politicians are a second comparative disadvantage. In most countries the MPs have rather unstable positions and short horizons, and thus hardly encourage the cabinet to take the longer view that is necessary for influencing the EU. Due to their high turnover, most MPs hardly have basic EU knowledge, memory or experience. They are largely uninformed about the flesh-and-blood of the EU machinery, its delegated and implementing acts and how these acts usually bypass their parliament. Their party-political competition causes politicization of domestic issues for the EU rather than EU issues for domestic debate. They like to hang on to the myth of an EU led by national governments based on parliamentary sovereignty. In most north-eastern countries they managed to get formal powers to impose on their government any domestic demand for its Council positions [Karlas, 2011; Knutelska, 2011; Olson and Ilonszki, 2011]. When using these powers, they often deny the pluralist nature of the EU by taking uncompromising positions, at the loss of its government’s room to negotiate effectively. A similar case happened in 2011 in Germany, where at the request of a few MPs the federal court in Karlsruhe decided that the Bundestag has veto powers regarding the European Financial Stability Mechanism (EFSM) and its Fund (EFSF), so bringing Chancellor Merkel of Germany into stalemate in the EU Council. The north-eastern example is gradually spreading out over other countries [O’Brennan and Raunio, 2007; Auel and Benz, 2005], also in the southern ones, where the government is often more authoritarian and the parliament less critical of EU integration [Raunio and Wiberg, 2000]. In, for example, the French Assemblée Nationale and the Spanish Congreso, more MPs than ever now push their demands for the EU to their government, want information about Council sessions and politicize home issues for action in the EU. Following the 1991 Danish example, all parliaments except Malta and Slovakia have a small desk, located in the EP, for collecting early information. In comparison, private groups rarely have an openly elected and representative body that discusses in public and instructs or questions the Board on its lobbying in the EU. A few NGOs have a council that is said to represent the many members, but it is usually composed by selection and not accountable to the public. In companies and institutes, the workers and shareholders tend to regard the management of EU affairs as an executive task. Their workers’ council usually meets in closed session and only interested in labour issues. The tragedy of parliamentary democracy at home is that it can bind the capital rigidly before, during and after its negotiations in the Council and on the EU playing-field at large, thus limiting its much-needed room to manoeuvre there. Any new election can 317

The Limits of EU Public Affairs Management

Chapter 7

change these limits in any direction, so making the government’s position inside the EU unstable. Parliamentary procedures and practises take much time and energy from the government. Parliamentary democracy sets, in short, limits to both the effectiveness and the efficiency of the capital as an EU interest group.

3 The Dogma of General Interest The dogma of general interest creates a third comparative disadvantage to the national capital. In a pluralistic society, no interest is shared by all. Yet, the government has to consider all sorts of interests at home and not only those of, for example, particular NGOs or MNCs, working or retired people and one sector or another. It is also held responsible for collective goods, which are consumer goods to be delivered free or at subsidized price, such as infrastructure, public health and public transport. The capital cannot, however, be everybody’s friend. Regarding its (as this is called), general or national interest, it always has to make choices, which make this interest less general or not so national. The choice is made informally by negotiations behind the scenes and, regarding the Council, formally by a national co-ordination procedure (chapter 3-IV-1) that makes different ministerial positions at least coherent on paper. Usually only some ministers, sometimes the national parliament or leading MPs (and, incidentally, some invited domestic interest groups) are involved in such processes. The outcome always is that many interests at home are not really included, although they may be formally or pro forma by procedure. The bypassed interest groups often voice their protest at home, freeze their loyalty to their government or take the exit route to Brussels. Due to the dogma of general interest, the capital has to manage at home the inevitable gap between trustful expectations at its input side and distrustful disappointments at its output side, which is a job for Tantalus. In comparison, private groups can cope with their different interests inside much easier. Their specific interests are usually flexible. The Board can even close or sell a division, as Philips did with its Polygram division in 1998, but a government cannot easily or fully close its department for elderly people or sell it to the private sector. The tragedy of the general interest dogma is that the capital always has to manage an overload of different specific interests. When choosing, it always has to follow a complex procedure or to make a complex compromise. It cannot easily be selective in public. If it publishes a sort of long-list, this is either unmanageably long or highly controversial by the choices made. For every dossier on whatever list, it almost never defines clear targets. It usually has too many priorities, adopted under domestic pressure. Its problem is, in short, that 318

Chapter 7

The Limits of EU Public Affairs Management

its real home front is the whole of the country, which is always pluralistic and thus divided. Any attempt to define the general interest is, in fact, a mission impossible.

4 Solving the Limiting Tragedies The three disadvantages or tragedies of the national government have a strong common denominator. In all EU countries, the capital is largely driven by domestic politics. The elections are largely dominated by domestic issues and parties [Pennings, 2006; Poguntke and others, 2006]. The MPs feel accountable to domestic citizens. The cabinet and the ministers are accountable to this parliament and need stable majority support from it. Also at the lower levels of ministries, the political winds usually blow from the inland rather than from the EU. Contained in introverted structures and processes at home, the cabinet and ministries should none the less be extroverted when wanting to perform in the EU, from where the binding decisions overrule the capital’s ones. They can only feel torn by the very different demands from their two playing-fields. Their disadvantages at the EU level, caused by domestic politics, are voiced by the MPs in particular. The national parliament, with its parliamentary parties acting in behalf of the citizens, is in the EU the most national institution and at farthest distance from the EU. It is said to have become ‘the loser’ or ‘victim’ of EU integration [Mattila and Raunio, 2012; Ladrech, 2010]. True enough, all parliaments have an EU committee [O’Brennan and Raunio, 2007] and most north-eastern ones even a scrutiny committee that can propose to instruct the cabinet on its Council position but their real impact on the EU remains limited [Auel, 2007]. To only a few per cent of all EU legislative proposals, these bodies pay some attention that slightly varies with the parliament’s powers at home. On the new subsidiarity procedure under the 2009 Lisbon Treaty (chapter 2-II), the Commission received from the national parliaments 211 (2010) and 67 (2011) opinions on 82 (2010) and 28 (2011) legislative proposals, none meeting the threshold of at least nine full parliaments (both chambers) and hence all ineffective (COM 2011/345 and 2012/373). The first effective case happened in May 2012 to the Monti-II draft-directive on (limits to) workers’ rights to strike against cheap labour mobility (COM 2012/130). Opposition to it was orchestrated by activist trade unions and resulted in the withdrawal of the proposal in September. The national party delegates in the EP are hardly an influential channel, as most form an only tiny minority in their political group there. Missing EU expertise and networks, most MPs rely for EU information on their ministries, the mass media and befriended interest groups. If 319

The Limits of EU Public Affairs Management

Chapter 7

they come to majority position in front of the cabinet at all, the latter can easily pay lip service and report afterwards that its mission has been overruled in the EU. In times of crisis politics at the EU Council level, such as the post-2007 euro crisis, some MPs are swept up in the optical illusion that via their cabinet they really (can) make a difference in the EU. The three tragedies do not mean that the capital is like a tragic hero found in ancient Greek drama. In the following five respects it may, as Machiavelli would recommend, bend with misfortune, instead of opposing or neglecting it, and find its ways. 1 Comparative advantages. Compared with decentralized governments, NGOs and SMEs at home, the capital has some outstanding resources, such as formal powers, budget and manpower. It can easily shift costs, re-order its troops and influence domestic stakeholders, also by traditional techniques such as coercion by law and encapsulation by subsidy. Regional governments and big cities suffer from the three tragedies too, but can easily escape to the EU. Domestic agencies can have the better of two worlds, as they are free from the three tragedies, have formal powers and many resources and, when being appointed as ‘competent national authority’ in the EU, are ‘double hattedly’ falling under both national and EU supervision. Private groups have more limited resources for taking action at the EU than their capital, but many MNOs compensate for this by engaging in intelligent PA there. 2 Tricks and games. By playing some legal and legitimate tricks and games, a cabinet can control the parliamentary factor of the three tragedies to some degree and thus lose less in the EU. By keeping its real EU long-list, partially or temporarily, secret it may get enough time for confidential talks in the EU. By delegation it can empower its administration and agencies, by decentralization the more independent agencies, regions and cities and by privatization the market organizations and civil society to manage their EU dossiers, which recent trends put the parliament at a distance. By overloading the MPs with documents it may get them confused at a higher level and make them more submissive. By leaving the definition of a general interest to an open procedure with ‘calls for interests’ and insourced stakeholders, like the Commission does, it may gain paradoxical room to choose at will. 3 Virtues out of necessities. The cabinet can also make virtues out of the tragedies. By its internal divisions it tells all stakeholders that they have friends (and enemies) inside and thus it can easily attract allies. In unfriendly EU arenas, it can make use of the publicity and noise. Usually it has fixed control over the majority of MPs and can order sufficient support. The notion of general interest it can elegantly use for upframing its choice for any domestic interest, as all cabinets regularly do. 320

Chapter 7

The Limits of EU Public Affairs Management

If one cannot get rid of disadvantages, one must use them. The Danish government, often eager to proclaim its own virtues, used the Danish negative referendum on the 1992 Maastricht Treaty to get concessions by a protocol to it; and by its drafting of the instructions from the Folketing to itself, it creates pressure tools on the Council. 4 Change in the air. Every capital is always in flux. Failures in the EU push capitals to (consider to) adapt their internal organization better to the EU, a lesson learnt earlier by many other interest groups. To its Brussels PR that feels the rift between domestic politics and EU reality intensely, many a capital allows more room for negotiations, so positioning it as front office with some mandate there. A few capitals (London, Paris, Berlin) brief their MEPs regularly now and all have a small EP-unit inside their PR. Some more ministries, wanting to prevent domestic turmoil over an EU decision, brief and debrief their experts on the EU work-floors earlier and better. Some try shared-listing with ministries elsewhere, informal Council meetings or a sort of EuroFed. More parliamentary parties use their parliament like members of a trade association (chapter 3-IV-2) that join the majority inside if it suits them but shop around for partners outside if it does not. In the latter case they join, on an issue for the EU, forces with befriended groups in civil society, such as in the EU-15 the green parties do with green interest groups; in the CEECs the parties cannot easily do so as their civil society is still taking off [Fink-Hafner, 2011]. In 2006, the Flemish parliament, wanting to bypass its federal parliament (with Walloons), considered to create a PA desk for such party-political shopping. 5 Professionalization. Tragedies are, like any challenge, largely manageable for their causes and consequences (chapter 5-III). Transparency may be a fact of democratic politics, but not necessarily to any degree and at any moment. MPs have the voice of parliamentary democracy, but this is not necessarily a synonym for cacophony. Really general interests do not exist, but they can be defined and explained by procedure, just as a company’s board defines and defends its strategy before its shareholders and stakeholders. All capitals can learn from the Commission, that must cope with the causes and consequences of its much higher transparency on legislation, its pressurising MEPs and its challenges from countless interest groups that want their interests included in its proposals. The Commission exploits its comparative advantages, plays clever tricks and games, makes virtues out of necessity, uses change in the air and, above all, usually behaves professionally by studying and managing the arenas of the EP and the Council in advance and by in-sourcing interest groups from outside, so gaining room to manoeuvre. Most capitals are laggards in such professional PA on their home front. 321

The Limits of EU Public Affairs Management

Chapter 7

5 Can Capitals Become More Successful? The question remains whether capitals can realize basic preconditions for success. They often lack PA expertise for the EU, to be positioned in their structures and procedures as in the MNO model (chapter 5-VIII). Introverted by their obsession for domestic issues, they operate in the EU without sufficient proactive homework and attuned fieldwork far before and after any Council meeting, so neglecting the Commission, the EP and the primary stakeholders. Their office of national co-ordination is more for soothing their parliament and the mass media than for preparing EU matches. Increasingly, their PR in Brussels takes over some of the capitals’ functions and now even enters the Commission and the EP. More ministries try to lobby the Commission and the EP as well [Panke, 2012]. They do so largely by mediaeval corridor visits of policy experts, as if the experts on substance can replace PA experts on influencing the EU. They may have a few talented PA people but seldom give them sufficient training, resources, career perspective and proper position in the home organization. They seldom make use of the influence channels and capacities of interest groups from decentralized governments and private organizations, at home via their EuroFeds, which are often better connected to the EU. They seldom make a shared-list and short-list with clear targets, thereby behaving largely single-handedly in the EU and returning home with few trophies and many souvenirs. Many capitals show, however, a rising awareness of such shortcomings and of the five solutions to the three tragedies. Like any interest group, a capital (cabinet, parliament, ministry) may never be able to overcome all limits of SCARE but yet discover that many limits are variable and partially manageable. There is, however, no reason why a national government must behave like Tantalus. Similarly, one can pose the question whether the EU capital is capable of overcoming all the limits to its efforts to get Europe better integrated. Its main challenges today are the various post-2007 crises (chapter 2-III-6) that already caused hot public debates about the possible disintegration of the Euro-zone and even the EU’s ‘community method’ (chapter 1-II) [Bongiovanni, 2012; Dehousse, 2011; Finke, 2010]. Such debates are far from new. Since 1952 (ECSC) the decline of the ‘European project’ has been often predicted. In the mid-1980s the catchword was ‘euro-sclerosis’ (‘euro’ then standing for Europe). As key player on the current financial crises, the EU Council must cope with serious challenges of SCARE. As a sender it is internally divided by the different interests of the heads of state, who are increasingly driven by unstable and EU-sceptical home politics [Brack and Costa, 2012], and it lacks sufficient resources and treaty powers. Its main channels are with the capitals that show their nuisances, distortions 322

Chapter 7

The Limits of EU Public Affairs Management

and leaks. The real arena includes, besides, many more divided and dividing interest groups than the capitals alone, such as international bodies like the IMF, EU institutions like Commission, Court and ECB, watchdogs like the German court, and countless European interest groups like companies, decentralized governments and banks. As receivers, the capitals are more attentive to the unfriendly winds blowing at home than to the EU Council’s proposals that many of them dislike and neutralize, as they can do in the EU’s ‘non-state statehood’ [Neyer and Wiener, 2011]. The environment is more than ever full of anxious citizens pushing their parliaments to promote national solutions to the crises [Eurobarometer, 2012, 77.2]. Due to the limits of SCARE, the EU Council’s President, the ECB President and the German Chancellor, the three main people, rationally follow an incremental course by going from the one summit or feasible step to the other, as the alternative is no common solution at all. Of course, political systems come and go, even in recent Europe (USSR, Yugoslavia). So far, the EU has always reconstructed itself. For the 2012 situation there is an over-supply of ‘Leitbilder for the EU’ [Brincker and others, 2011]. The redesign of the Euro-zone started in 2012 with draft treaties on economic governance (TSCG) and financial stability (ESM) that after adoption and together with the EU Council’s new proposals (29 June 2012) may further a redesigned EU. The main difference with previous EU crises is that the current ones are intensely and widely discussed by all sorts of domestic interest groups, political parties and citizens. Paradoxically, this shows that the integration of Europe has come closer to where it should be: at mass level, far beyond the circles of the political leaders. Most domestic groups and citizens have altered their so far benevolent indifference to the EU into attitudes of supportive, opponent or wavering engagement. From a PA perspective, the national leaders should take this more engaged home front as an opportunity and manage it by professional PA. For example, depending on their own position on the EU and the distribution of attitudes at home, they can manage it as a largely friendly, unfriendly or indeterminate arena (figure 4.4). The EU Council president too can use the more engaged domestic societies as opportunity. Not being in the position to intervene directly, he should arrange a free ride on the Commission, where directly or indirectly most domestic interest groups come along. So far, most of them play charm in Brussels and get their interests included in compromises, but back home many of them remain silent about the benefits they receive from the EU. The Commission can, as part of its negotiations, demand from any of them that they make these benefits publicly well-known at home, so indirectly creating good chances to turn waverers into supporters and opponents into waverers. This short example shows that, by applying the disciplined creativity 323

The Limits of EU Public Affairs Management

Chapter 7

of PAM, the EU and national leaders too can, to some degree, manage their SCARE limits even during crises. As usual, any margin can make a big difference.

IX The Limits of PA Management in the EU: Problem or Blessing? All the limits discussed above may seem self-evident. Yet they are largely neglected both in literature and by the buoyant zealots of PA management. They are, however, much discussed by PA professionals working in the EU. As a reminder, these people are not necessarily those with full-time or commercial PA jobs, but those who carefully prepare and prudently do the fieldwork, whatever their job position is. These professionals regard the understanding of the limits as most useful knowledge for various reasons. They want to respect the limits that fall outside their control, as otherwise they would not only be inefficient and ineffective but also do damage to their reputation. They also like to be aware of the limits that apply to their rivals in order to profit from this knowledge and maybe their rivals’ nonchalance. Knowing that the many possibilities of PAM in the EU have their limits, they also know that the limits can provide opportunities, at least for gaining an edge over nonchalant players. So they can avoid unforced errors, enjoy virtues out of necessities and improve their room for manoeuvring prudently. All the various limits clearly do not have an absolute but a relative status. The same limit can exist for one interest group, but be absent for another. The one group may consider the presence or absence of the same limit a problem and another to be a blessing. This assessment should, of course, not be based on instinctive reflex but on careful reflection upon the specific situation. Therefore, the professional group always tries to identify both its comparative advantages and disadvantages. The former it wants to save as blessings and the latter to solve as problems. It may even try to engineer more disadvantageous limits for rivals. In the preceding chapters we have viewed such practises, possibilities and limits of PA and lobby management in the EU from the perspective of an interest group. In the next and final chapter, we shall review them from the broader perspective of democracy and examine the question of whether this PA and lobby management, and, in particular, its professional variant is good or bad for EU democracy.

324

Chapter 8 Public Affairs, Lobbying and EU Democracy It is a good thing that anybody anxious to serve the public is able to propose a law, in regard to which everyone can speak either in favour of it or against it, prior to a decision being reached. A civilian is more prudent, more stable and a better judge than a prince. What do Machiavelli’s views on democracy (Discorsi I-18, I-58) imply for the current EU?

I Democracy as a Criterion The views of Machiavelli above may be most surprising to many people but not to the informed few [McCormick, 2011]. In his private role of citizen the man from Florence was always most critical about rulers in the past and present, while in his public role of civil servant he advised his ruler, Lorenzo dei Medici, ‘how to survive’ in domestic and foreign affairs. His answer, given above, was that the ruler should institutionalize the greater qualities of the people by participatory forms of democracy, as otherwise he shall lose position at his home front and thus in foreign affairs too. If Machiavelli could have foreseen the future, he would have praised the phenomenon of countless active interest groups at EU level. From his side he would be surprised to hear that lobbying in the EU is nowadays often criticized for its ‘damaging effects’ on the democratic functioning of the EU. At the 1992 EP hearings on lobbying three lasting criticisms were launched. Firstly, industrial MNCs were said to create, by their activism and dominance, imbalances of decision-making, to the disadvantage of weaker groups such as workers, consumers and small enterprises. Secondly, much lobbying was said to take place behind closed doors, thereby creating a lack of transparency that frustrates competitors, mass media and even officials. Thirdly, many of those who engage in lobbying were accused of abusive and immoral practises, such as stealing documents, blackmail and bribery, which should be forbidden. Today, particularly in countries recently discovering the phenomenon of lobbying the EU, many mass media voice strong criticisms and fears, as has happened in Britain in the 1980s, Scandinavia in the 1990s, Germany in the early 2000s and in the Central and Eastern European Countries (CEECs) recently. Such initially normative concerns may be seen as a positive development because, so far, they always paved the way for a wider public acceptance and a more 325

Public Affairs, Lobbying and EU Democraty

Chapter 8

prudent application of EU lobbying. The evaluation of PAM (and lobbying) at the EU level within the frame of democracy is, of course, only one public choice out of many alternatives. Different frames or core values may make sense too, such as the following: – Better integration? Is an active participation of many interest groups good or bad for the integration of European countries? The short answer here is that it brings the settlement of irritating differences closer to the meso-level of decentralized and private organizations, where most irritations are voiced and need to be resolved. – Better EU decision-making? Does the wide involvement of interest groups positively or negatively affect the quality of EU decision-making? The preceding chapters allow the short answer here that most Commission officials and MEPs welcome this involvement and its effects of better information and broader support. – Better welfare of citizens? Have the active interest groups in the EU a positive, negative or no effect on socio-economic welfare? Here the short answer comes from the interest groups on socio-economic welfare themselves, which have made their view clear by continuing to increase their presence in the EU. For two reasons the short answers do not imply that, under any frame whatsoever, interest group activity should be maximized. One reason is that maximizing it for the sake of one value may be detrimental to another value, as shown by the well-known tension between democratic and effective decision-making. Only fanatics giving full priority to one core value are maximizers, but the countless many that cherish a wide range of values can only be optimizers. Secondly, next to the quantity of participation its (professional) quality may be equally important (to which we will return at the end). Because democracy is at least one of the leading core values in Europe, we focus here on the impacts of interest groups at the EU level on EU democracy. In academic language, their behaviour is the independent variable (‘cause’) and democracy the dependent one (‘effect’). Are the effects of their efforts to influence the EU positive, negative or indifferent for the state of EU democracy? The positive and negative effects are the most relevant for our focus here. In order to assess these effects, we shall have to define what is meant in Europe by the word ‘democracy’. But before we can do this, we must put aside the following four different but closely related and equally interesting questions. 1 Domestic notions. As we shall observe, the EU countries often hold different notions of democracy and rankings of them. Many groups and people take these national notions as their standard for the EU or a part 326

Chapter 8

Public Affairs, Lobbying and EU Democraty

of it, such as the EP. We shall leave the old and still relevant question of whether it makes sense to apply national notions of democracy to the EU [Goodhart, 2007; Hix and others, 2007; Weiler, 1995] and simply pursue the current public debate in which they are applied. 2 The best concept. We shall also not enter a discussion about the question of what the best concept of democracy is. As we shall see below, there is no generally accepted criterion to help determine this. Groups and people continue to disagree about this, both in the EU and in their home country. 3 Is the EU democratic? Neither shall we discuss the question whether the EU is democratic or not [Andersen and Eliassen, 1996]. To this question as many answers are possible as different notions exist. For example, the view that ‘the EU has a democratic deficit’ [Lord, 1998] is contradicted by both that it has a ‘democratic surplus’ [Meunier-Aitsahalia and Ross, 1993] and that in Europe the ‘real problem of democracy’ lies at the national rather than the EU level [Schmidt, 2006; Zweifel, 2002]. 4 Reversed focus. We also leave the reversed relationship between democracy (now as ‘cause’) and interest group behaviour (as ‘effect’). Democratic phenomena like an open society, active citizenship and limited government can, indeed, be excellent preconditions for the flowering of interest groups [Popper, 1945]. Our single question remains: do interest groups have specific impacts on EU democracy?

II Notions of Democracy in Europe In Europe (and elsewhere) not just one notion of democracy exists, but a container full of different notions that are all popular and disputed to some degree somewhere [Eriksen and Fossum, 2012; Coppedge and others, 2011; Pollack, 2000]. For example, in semi-presidential systems like France or Britain, the notions of accountability and rule of law are popular rather than those of pluralistic competition and consensual decision-making that are widely supported in parliamentary systems like Italy and Sweden. In countries with strong and volatile political parties like Poland and Belgium, the notions of parliamentary representation and discursiveness tend to be popular, while those of corporatism and responsiveness are more prevalent in countries with well-developed interest groups, like Austria or the Netherlands. In countries that have a layer of well-established interest groups, like Spain and Germany, next to many weakly organized SMEs and NGOs, the former tend to attach more importance to the notions of direct channels and limited government than to those of competitive elections and common identity, as is the case in Portugal and Hungary. 327

Public Affairs, Lobbying and EU Democraty

Chapter 8

The different notions have a common denominator, as they are negatively conjoined by their contrast to the (equally broad) notions of tyranny, despotism and the like [Held, 1996; Dahl, 1989]. The notions of democracy differ largely on what they stand for positively and are variously presented as core ideas, preconditions, elements, indicators, factors or outcomes of it. Some are value-related, such as ‘freedom’, ‘tolerance’ and ‘legitimacy’, while others are process-related, such as ‘elections’, ‘majority rule’ and ‘responsiveness’. Sometimes they differ only seemingly, as different words can be used for the same fundamental idea. For example, ‘liberty’, ‘freedom’ and ‘autonomy’ are often used as synonyms for the personal capacity to act as one wants. The reverse of using the same word for different ideas happens as well, sometimes even by the same author who defines ‘democracy’ on one occasion as legitimate authority and identity, and on another as only an element of the then wider notion of legitimate authority [Lord, 1998; Beetham and Lord, 1998]. A scientifically sound criterion to determine the single best notion of democracy is beyond reach as, in the science of knowledge, every value has not an absolute but a relative status [Brecht, 1959]. This idea of relativity opens the door to manifold competitive and newly arising notions that all may make sense, gather some support and, as appears below, come together at the EU level [Pinder, 1999]. As both at domestic and EU level the many notions compete with each other, optimization rather than maximization of any notion is the only rational solution here too. This results in the paradox that any single notion may be imperfectly realized but the many living notions together may be perfectly optimized. For our purpose here it is not only impossible but also unnecessary to decide upon the best definition of the concept of democracy with regard to the EU. For the construction of our dependent variable, it is sufficient to assemble the most popular notions of democracy as they feature in the public debate about the EU. They are listed in figure 8.1, subdivided into four categories that refer to the EU’s input side, its throughput of decision-making, its output side and its feedback-loop from civil society [Norris, 2011; Easton, 1965]. We shall briefly explain them first.

1 Input Notions The general idea of the input concept of democracy is that the decisionmaking system must offer openness to all sorts of people and groups wanting to achieve a desired outcome. The openness is not discriminating or selective, but gives to every desire regarding the outcome of a decision an equal and fair chance of permeability. A desire may involve anything: one or another decision or no decision at all regarding proposed laws, policies, 328

Chapter 8

Public Affairs, Lobbying and EU Democraty

Popular Notions of Democracy Input Notions Openness, Equality, Permeability Pluralistic Competition Competitive Elections Direct Channels Representative Channels Throughput Notions Representation Majoritarian Decisions Consensual Decisions Polyarchy, Opposition Legitimate Authority Discursiveness Transparency Output Notions Legitimacy Limited Government Rule of Law Accountability Responsiveness Feedback Notions Citizenship, Basic Support Tolerance Identity Freedoms and Rights Linkages to Input Figure 8.1

Public Affairs, Lobbying and EU Democraty

Chapter 8

allocations, implementations, recruitment, sanctions or whatever outcome. Openness, equality and permeability do not necessarily have to be real for every desire at any moment but, after some time, every type of demand, information or support should find its way into the system. The permeation of competitive civil pluralism into and inside the political system needs, of course, vehicles. At least three of them have become specific notions of democracy on the input side. One is the notion of regular competitive elections, based on equality and fairness, for the purpose of distributing formal positions of power. A second is that direct channels must exist for transporting the desires, and these channels may range from corridor visits to referenda and street protests. Finally, indirect channels that represent civil desires must exist by way of political parties, interest groups, mass media, bureaucracies or any other platform with an intermediary capacity. All these general and specific input notions have some varying degree of popularity in the European countries, which is another way of saying that they are contested. People and groups appear to differ in their preference for one or another input notion and for one or another vehicle. For example, the openness of the French government to employers’ organizations rather than to trade unions is at issue there. Everywhere the electoral system is regularly under critical debate, such as for its capacity to create equal effects of votes (proportionality system) or stable government (district system). Direct channels are popular in countries like Denmark, Greece and Poland, regions like Flanders and Bavaria, and concentrated sectors like consumer electronics and the green movement. Examples of strongly supported indirect channels are political parties in Italy, the mass media in Britain, bureaucracies in the Netherlands, central governments in Scandinavia and the Baltic States, churches in Austria and trade unions in Germany. These different input notions have, to some extent, all been put into practise at the EU level, where they often clash, are subject to compromise and even cause a degree of europeanization at home. The EU shows a substantial and remarkable openness to old and new desires that come from the countries, regions and sectors and permeate into the EU. Caused by the ‘size of democracy’ [Dahl and Tufte, 1973], its openness is wider for organized interests than for individual ones, which makes interest groups the proxy for ‘civil society’ [Saurugger, 2008]. By its active and massive use of expert groups and consultations, the Commission has become a unique case of ‘insourced democracy’. Some groups, such as those on banking and the environment, may act as interest cartels for a while and take the lead, but they are usually soon countervailed by SMEs, NGOs and regions, which indicates a remarkable degree of equality for those entering after 330

Chapter 8

Public Affairs, Lobbying and EU Democraty

some time. The main institutions have, according to a sample of lobbyists, a different accessibility, as the EP is found to be most accessible, followed by the Commission and the Council the least [ComRes, 2008]. The pluralistic competition inside member countries has an imperfect but real continuation at the EU level, as indicated by their interest groups. There are direct elections only for the EP, which are held on a national basis. The main direct channel into the EU is that of Brussels-based interest groups, with a few national referenda and street protests at a distance. The main indirect channels are interest groups and civil servants, now with political parties and mass media at a distance. Most political parties are weakly organized on transnational basis [Priestly, 2010], suffer from declining party membership at home (EU average at 4.6% of electors) [Van Biezen and others, 2012] and have problems of becoming europeanized [Külahci, 2012]. Most mass media cover EU news from home and not by Brussels correspondents that, if present, have a ‘communication deficit’ with EU officials [Martins and others, 2012] and act more like sports reporters covering scandals and scores than as informers about substantial processes. Whatever their quality, the many domestic input notions of democracy have come to competitive life at the EU level.

2 Throughput Notions The general idea here is that a government must be representative of the people’s desires [Brito Vieira and Runciman, 2008]. This idea of representation is another container-like notion, revived in the 1960s. A group of officials is called representative if, for example, it mirrors the social features of the people (such as age, sex and ethnicity), their opinions, their interests and/or their votes [Birch, 1971]. All that may apply to an elected body (‘electoral representation’) but also to a nominated committee, corporatist platform or bureaucracy (‘functional representation’) [Krislov, 1974]. The representatives are assumed to act on behalf of the larger population, instead of whom they govern. In their ‘acting for the people’ [Pitkin, 1967 and 1969] they may behave as either delegate or trustee [Pennock and Chapman, 1968], but the one role does not necessarily preclude the other. Regarding elected politicians there exist the formalistic notions that, so to say by definition, the elected people are representative due to the ballot box and their institution due to the constitution. Some notions of representation even jump ahead to the output and feedback sides of government and emphasize the responsive allocation of desired outcomes [Eulau and Prewitt, 1973] or the promotion of civil society [Pateman, 1970]. Two groups of throughput notions of democracy may be seen as more specific. The first is focused on methods of government. One variant is the 331

Public Affairs, Lobbying and EU Democraty

Chapter 8

notion of majority rule, usually with an amendment for a qualified majority if there is an inequality of either size or passion [Dahl, 1989]. A second variant replaces this amendment by the notion of consensual or even consociational government that continually accommodates different preferences by new compromises [Lijphart, 1977; Taylor, 1996]. A third variant of this first group emphasizes the method of polyarchy, a system that is permanently open, highly competitive, full of opposition and without any group being always in either a majority or minority position [Dahl, 1971]. The second group stresses specific values of the government process, such as legitimate authority [Friedrich, 1963], discursive or deliberative considerations [Dryzek and Niemeyer, 2008] or transparency [Deckmyn and Thomson, 1998; Bunyan, 1999], to mention only the most popular values. All these throughput notions of democracy have experienced some popularity in European countries [Schmidt, 2006]. For example, in Britain and those CEECs that in the early 1990s got party-political competition, a threefold strong belief in elected representation exists, based on party-political delegation, majority decision-making and the value of transparency. Popular in Germanic countries are the notions of functional representation by experts acting on behalf of their segment of society, consensual decisionmaking and the value of discursiveness. In France, the preference for functional representation is largely limited to experts from the employers’ side, while the value of transparency is supported after rather than during the decision-making process. In Italy, there is broad support for electoral representation on a party-political trustee basis, for consensual government and for polyarchy and opposition. In the EU area, one single dominant throughput notion is absent. The various notions compete with one another [Pollak, 2007]. All throughput notions undergo europeanization and have, to some degree, as an outcome of compromises, been put into practise at the EU level [Mair and Thomassen, 2010]. All institutions have a demographic composition that somehow reflects the different nationalities involved. At the administrative level this is, although declining, informally arranged (‘fourchette’) and at the political one it is formalized through seats, voting points and positions. Through their institutions, the political groups (EP), interest groups (ESC), regions (COR) and member states (Council) import into the system a wide variety of opinions and interests. By the way of expert groups, comitology and agencies the Commission holds a lot of functional representation, although the representativeness of their members for their ‘constituency’ is at issue [Borragán and Smismans, 2012; Heard-Lauréote, 2010]. Electoral representation is directly rooted in the EP and indirectly in both the COR (politicians from decentralized councils) and the Council (governments based on national elections). The specific 332

Chapter 8

Public Affairs, Lobbying and EU Democraty

notions of throughput democracy have also got a degree of EU design. All main methods of government have been put into practise to some degree. Variants of majority decision-making that go counter the dogma of national sovereignty exist in the Commission’s College, the Parliament and the Council. Polyarchy and opposition are vested in formal power distributions and informal conflicts. Indicators of the latter are the disagreements over nominations, policy objectives, regime values and constitutional change. The values of legitimate authority, discursiveness and transparency are hardly at issue as such, but intensely for their application in reality. Examples are the often contested authority of the Commission, the discursiveness of the EP and the transparency of the Council. These issues cause both public debate and new proposals, which indicate that the values are taken as serious notions of EU democratic government. Compared to national parliaments [Schmidt, 2006; Zweifel, 2002] the EP, being the institution that internalized most throughput notions, now shows a ‘democratic surplus’ [Meunier-Aitsahalia and Ross, 1993] rather than a ‘democratic deficit’. In short, the domestic notions of throughput democracy have come to competitive life in the EU.

3 Output Notions The general idea of output notions is that the government’s products should be legitimate or widely considered acceptable. There are various potential sources of legitimacy. One is satisfaction with the government’s outputs that meet social needs (such as health, employment, education and housing). The throughput process can be a source of legitimacy if widely regarded as fair and just. Then even a disappointing output is accepted as ‘all part of the game’, and a compromise among different groups as ‘the best possible result’ that comes the closest possible to the ‘general interest’. Another source of legitimacy is the personal authority or charisma of responsible people. Margaret Thatcher, Felipe González and Helmut Kohl (former ‘heads of state’ of Britain, Spain and Germany, respectively) have had their moments of speaking ex cathedra and getting what they wanted. Legitimacy can also be provided by prestigious vehicles such as tradition and science. If an outcome links up nicely with a famous past or is based on scientific evidence, it may be more easily accepted. More specific output notions of democracy are the following, two of which come close to being methods. One is limited government, which from the perspective of effective government is often considered a vice (chapter 7) but from that of democracy a virtue. This notion has at least three desired variants: checks and balances among the institutions, decentralization of government and restrained interventions in the private 333

Public Affairs, Lobbying and EU Democraty

Chapter 8

spheres, and sectors. The other method regards rule of law: outputs of government should be based on formally binding decisions, produced through formally correct procedures, approved by a formally representative platform, enforced by the state’s institutions and be open to court appeal. Specific output values exist as well. One is that the officials that produce outputs must be accountable for what they do (or do not do) [Lord, 1998] and that nominated officials are accountable to the elected ones and the latter to the electors. Another popular value is that the outputs should be responsive to the desires imported into the system at an earlier stage [Eulau and Prewitt, 1973]. This may imply that the government delivers what it has promised before to all or to specific regions, sectors or groups [Eckstein, 1971]. The output notions of democracy have variable levels of popularity at the country level, and thus adverse reactions as well. The outputs of government are usually considered legitimate, as is indicated by relatively high levels of civil obedience and low levels of civil protest in most countries. If protests occur, they are limited to the special issues of regional separatism (such as in the Basque region, Corsica or Northern Ireland), sectoral decline (like in farming, trucking or mining) and failing government performances (such as in France and Hungary). All countries have some form of limited government in one or more of its three variants (checks and balances, decentralization, private freedoms). Most notable are Germany that scores high on all three and the formerly communist CEECs that rapidly create all three. France is seemingly at the opposite end of the spectrum, but behind the formal show of central state dominance exist such realities as conflicts between the Matignon (Prime Minister) and the Elysée (President) (chapter 3-V-1), public decentralization and sectoral privatization. The notion of rule of law as being hard law is very popular in Britain, while in the continental countries it includes, as outcome of consensual decision-making, also soft law (semi-formal agreements), resulting in, as it is called, the rule of consent. In all countries, the value of accountability is in some way contained in the formal arrangements between representative councils and electors, and in the basic freedoms of the mass media. Every government is regularly evaluated for its responsiveness to public opinion through the ballot box. In countries like Italy, Belgium and Bulgaria, clientelism that pampers selected socio-political groupings is part of this. The different domestic notions and practises of output democracy increasingly undergo europeanization and vie with each other at the EU level. They all exist in some form to some degree. Legitimacy is rooted in legal and administrative principles [Smith, 2012], in popular policy values ranging from open markets to healthy food and in special vehicles such as 334

Chapter 8

Public Affairs, Lobbying and EU Democraty

charismatic leadership (like that of Walter Hallstein or Jacques Delors), tradition (from the old Idea of Europe to the traditions of representation) and science (as by expert groups). Protests are sometimes launched by a few interest groups (farmers, harbour workers) and by national political parties feeling caught between domestic pressures and EU decisions. The notion of limited EU government is practised by institutional checks and balances and, regarding the domestic levels and the private spheres, by rules for subsidiarity. As a union of common laws (chapter 1-II), the EU’s main form of government is by rule by law, including soft variants such as OCM. Discretionary decisions regarding subsidies and procurements must have a legal basis. Binding decisions are formally devised according to procedures based on treaties (laws) or on secondary laws (acts) and, unlike domestic arrangements in some member states, they are all open to appeal at the Court. The law enforcement that is vulnerable by its dependency on national co-operation, is remarkably effective, although more for civil societies than for central governments. Accountability [Curtin and others, 2010] is found in three developing formal practises: EU officials become more accountable to the EP, Council ministers to their home parliament and both MEPs and ministers to their ballot boxes. Interest groups and the mass media are becoming more accountable to their members and publics. Responsiveness to demands can be seen in both general policies and specific support to selected regions, sectors and groups.

4 Feedback Notions The general idea here is citizenship. The people are not just subjects of government, but are seen as citizens capable of internalizing the values of government selectively (civic spirit) and to behave accordingly (civic behaviour). This applies not only to individuals but also to groups and organizations (‘corporate citizens’) that include NGOs and companies and together act as a proxy for civil society. As government should be open and permeable to citizens’ demands, so the citizens should be open to demands from government as well. By civic spirit they consider the fit or misfit between the desired and the perceived functioning of government. By civic behaviour they express their views that may range from full support to intense protest with scepticism in between. No democracy can exist without a civil society [Fullinwider, 1999]. There are also specific feedback notions, two of which concern values. Firstly, citizens must be tolerant of each other and allow peaceful solutions of any irritating difference [King, 1976]. Secondly, to avoid a breakdown of society, citizens must have a stable sense of common identity that is based on, for example, social indicators (like age or ethnicity), belief sys335

Public Affairs, Lobbying and EU Democraty

Chapter 8

tems (like religion or ideology), regions (‘we, the Flemish’), sectors (like ‘German industry’ or ‘Italian food’), recent drama (formerly communist CEECs) and/or national past and pride (‘La France’). The more the various identities cross-cut each other, the more they produce broad and stable cohesion. Two other specific notions involve methods. Firstly, citizens must have substantial freedoms and rights, in both the negative form of a private domain outside (‘free from’) government and the positive one of opportunities to participate in (‘free for’) government [Friedrich, 1963]. Secondly, the civil cultures and structures should be linked to the input side of government. The feedback channels must start in civil society and go via political parties, interest groups and mass media back to government [Luttbeg, 1968]. Without feedback channels the government cannot meet the general notions of openness, representation and legitimacy and be called democratic. All feedback notions of democracy have some degree of popularity at the country level and thus are also contested [Bale, 2008; Almond and others, 2002]. In behavioural terms, in every country only a minority of individual or corporate citizens displays high levels of civic spirit and behaviour, with the majority being rather passive and tractable. In most countries the large majority basically supports the constitutional and regime values of their political system, but is rather sceptical about policies, performances and politicians. Tolerance looks to be a strong social value, as indicated by majorities disapproving open discrimination against, for example, immigrants, women or the elderly. New or deviant values and activities are often accepted as part of the pluralism of society, as long as they do not create disorder or irritate many others. Many group identities exist in the European countries, but in decline seem to be those based on belief systems (except in Scandinavia) and nationalism (except in CEECs). Regionalism is a strong common identity in countries like Belgium, Germany, Italy and Spain and taking off in some CEECs. Basic freedoms and rights have been institutionalized everywhere. Freedom from government has been promoted under the frames of privacy and privatization that diffused from the north-west to the rest of Europe, but a backlash on privacy occurred after the events of September 11, 2001, and on privatization since the 2007 financial crisis. Feedback linkages to government are widely supported. In northern countries, they shift from political parties to interest groups and mass media since the early 1990s [Klingemann and Fuchs, 1995]. The various feedback notions and practises show competition at the EU level. Like at the domestic levels, individual EU citizenship is the weakest realized notion here and anyhow difficult to realize in a union of about 500 million people. More effective is the feedback through its proxy of cor336

Chapter 8

Public Affairs, Lobbying and EU Democraty

porate citizenship, such as by interest groups acting for their workers, clients or members and by political groups acting in the EP for their citizens. A majority of about 55% of all citizens (say to) support EU membership and integration, a minority of about 15% to be against it and about 30% to be ambivalent, which support level is lower than in the past, varies per country [Gaxie and others, 2011] and includes variants of ‘EUscepticism’ about EU policies, performances and politicians [Leconte, 2010; Van Spanje and De Vreese, 2011]. Also a majority of EU citizens (56%) claim to be satisfied with EU democracy, and those in southern and CEEC countries value democracy in the EU above that at home [Hobolt, 2012; Munoz and others, 2011]. Most citizens have a moderate level of tolerance for people from other EU countries, as shown by cross-national trust scores (chapter 1-I). Corporate tolerance is indicated by most organizations’ willingness to settle their disputes and irritations by peaceful and compromising decision-making. A small majority of about 55% of citizens show some degree of EU identity (and 2% fully this identity), largely grounded on a sense of utility [Favell and Guiraudon, 2011] and often below their district, regional or country identity [Scheuer, 2005]. Among corporate citizens sectoral identity is the strongest. Under the Lisbon Treaty the notion of freedoms and rights got its constitution-like status, including the Charter of Fundamental Rights and the ECI referendum. Interest groups form the main linkage between civil society and the EU. Mass media and political parties mainly have a national (and some parties a nationalistic) focus and hardly a direct feedback into the EU [Brants and Voltmer, 2011; Binnema, 2009; Van Noije, 2007; De Vreese, 2007], but their messages may go indirectly via channels like domestic interest groups and officials in the EU, compatriots in EU institutions and the Commission’s outpost in any country.

III The Impacts of EU Interest Groups on EU Democracy All the different notions of democracy mentioned above continuously stream up from the member countries, indicate a broad popular demand for EU democracy and play some role at the EU level. The idea and practise of EU democracy show a unique amalgam of notions, each having its support basis at some places and times. It also has an impact downstream in the member countries, where groups and people appreciate notions that are part of the EU amalgam but hardly or not found in their country. Thus, on issues of democracy too, the EU functions as the main direction and source of europeanization. In both the EU and at home, the various notions are usually also contested, not for their noble ideas behind but for 337

Public Affairs, Lobbying and EU Democraty

Chapter 8

their consequences. Everywhere some groups and people have more interest in, for example, direct channels than in representative ones, while others take the opposite view. Still others differentiate further and want the direct channels to be open only to individuals, and the representative ones only to political parties. Some consider political parties and interest groups as two contradictory channels, while others see them as complementary. Not any notion gathers absolute or unconditional support from all people. In a pluralistic society the outcome is some form of optimal compromise with other notions democracy or different values like stability or effectiveness. The maximization of any notion or value might be fine for its supporters but makes trouble in society. For example, perfect transparency, rule of law or tolerance would seriously injure other notions of democracy and even other high values, such as integration and welfare. The notions of democracy need, in short, optimization rather than maximization. As discussed earlier, it is not our objective to assess the amalgamated state of EU democracy, where all notions have achieved some sufficient degree of realization. Our objective is to explore the dependency of EU democracy, with its variety of notions, on the behaviour of interest groups, which approach others begin to follow now [Greer and others, 2008; Karr, 2006]. Can results of EU democracy be explained by this behaviour? Or, do interest groups have a positive or negative impact on them, or none at all?

1 Input Impacts In comparative perspective, the EU is a highly open and competitive system. Three indicators, inter alia, are the constant influx of new issues resulting in new agendas, the regular permeation of new regime and policy values, and the massive in-sourcing of experts from interest groups. The strong competition between issues, values and experts reflects the pluralism of the countries. Interest groups play a significant positive role in keeping the system open and competitive. Evidence of this comes not only from their role in decision processes, as documented by case studies referred to before, but also from the imaginary control situation in which interest groups are absent. In that case, the EU would be a system populated only by officials of the Commission, members of the EP, ESC and COR, Council ministers and Court personnel, with only two linkages with the member countries: one being the direct election of the EP and the other the indirect linkages between the Council, ESC and COR on the one side and the national capitals and societies on the other. This virtual system would certainly be much more closed for many issues, values and peo338

Chapter 8

Public Affairs, Lobbying and EU Democraty

ple from private and even public interest groups, and altogether be much less open and competitive than the real system is now. On the vehicles of input democracy, interest groups have different positive impacts. There is hardly any evidence that they play a strong role in the direct elections for the EP, although some try to influence party manifestos, lists of candidates and public debates at domestic level, as happens in Britain. Interest groups dig their own channels, indirectly through EuroFeds and directly by sending their experts to the EU and opening a Brussels office. The Commission and the EP get the aggregated issues, values and people, thereby acting as a kind of representative bureaucracy and parliament. Most domestic political parties lack well-established direct links with the EU’s input side and rely on few indirect links, such as to their sister party in the EP, befriended interest groups as exemplified by the green parties and green movement, and, incidentally, party-political friends in the Council and COR [Lindberg and others, 2008]. With a very few exceptions, the mass media are almost never directly present on the EU input side. Interest groups may be called helpful but not sufficient for the EU’s input democracy. In terms of equality of entry, some types of groups can have better chances of permeation than others [Greer and others, 2008; Coen, 2007] and do have, according to some NGOs such as Alter-EU and Corporate Europe Observatory [Burley and others, 2010], groups that lobby against lobbying. Established groups like national ministries and multinationals (MNOs) can have a permanent ticket to enter their policy domain, but never have a master key entitling them to enter any room. Professional groups find the system particularly open and permeable, because they carry out a lot of preparatory work before lobbying. They are simply more clever than the amateurs and arrive early and at the right place. The electoral vehicle is potentially most open to NGO-like interest groups, including BONGOs and GONGOs, which claim to represent more general interests. Direct channels are mainly used by better-resourced groups that can afford the costs, such as MNOs, EuroFeds, national ministries and small-sized ones that know how to behave like an MNO. The representative channels are hardly discriminating, as they are open to every group that meets the minimum conditions of organization. There are, however, some weakly organized outside groups, such as retired people and immigrant entrepreneurs (to which we return in the section on feedback impacts). Interest groups can also have at least two negative impacts on the EU’s input democracy. One comes from those that try to make the access to the EU less open to competitors. This is the paradox of openness that can cause closure. Along the access routes to the EU, such groups form a coalition for 339

Public Affairs, Lobbying and EU Democraty

Chapter 8

getting priority treatment at the entrance and, once inside, they erect thresholds that keep their rivals outside and try to transform their coalition into a policy cartel that preferably includes EU officials. One example is the proposed Aarhus Convention Regulation (chapter 3-III), precooked by the green movement’s European Environmental Bureau (EBB) and limiting participation on green issues to ‘entitled groupings’ of its own family, thereby maximizing their core value of environment at the cost of democracy. Another example is the 2011 Commission’s effort to get the management of the debts of banks and states entrusted to only technocratic groupings like the Euro Working Group, at a distance from other interest groups and politicians (similar to what happens in many countries), thereby giving priority to effectiveness rather than democracy. The second source of negative impacts belongs to all interest groups together. Because so many groups seek to enter and to permeate EU operations, they can overload the system through either the volume or the contents of their demands, which is another case of the paradox of openness creating its own blockage. Some issues, values and groups are then held up, stockpiled, refused or otherwise not allowed entrance. Each of these negative impacts may result in an imbalance of interest groups on the input side. The possibilities of entry inequality and negative impacts are at least partially addressed by the following three system-linked correction mechanisms: – Openings from inside. Hungry for new information, support and demands, EU officials often open their doors, even for incipient, amateurish or penniless groups with an interest in the EU. Incumbent groups that want to turn a headwind inside their arena often open their doors too. Entry thresholds are often only temporary. – Scrambling groups outside. Interest groups that stand outside and want a position inside often knock successfully on closed doors and make the EU more competitive. Usually they receive a provisional ticket, start to behave more professionally and find ways to remain present or represented. – New input collectors. From inside the EU the overload problem is tackled by new collectors of inputs, such as new expert groups and the online consultations. No impact on the EU input side should, however, be equated with a final impact, as more impacts may come from the throughput, output and feedback that follow.

2 Throughput Impacts In many respects the EU decision-making system is fairly representative for the member countries, although this is not necessarily due to interest 340

Chapter 8

Public Affairs, Lobbying and EU Democraty

groups alone. Commission civil servants and MEPs can also act as agents of representation, for example by soliciting opinions, anticipating interests, recruiting people or distributing positions. The presence of interest groups is often a positive factor of improved EU representation. With these groups nearby, the officials can anticipate demands better. Each day they receive fresh information and indications of support from many different groups. The experts from interest groups make the work floors of the EU more representative of the various sectors, regions and countries. From their side, the groups that provide information stimulate, even if their information is selective, the representative and discursive discussion inside the institutions [Neyer, 2000], and by indicating their support they promote a decision process that can be more representative of the various interests. The methods of decision-making are also at least partially made more democratic by the presence of interest groups. This positive dependency is, of course, selective. Interest groups prefer the majority-vote method mostly in case it gives them the result they desire. Knowing, however, that not they but the officials formally control this method, they tend to prefer consensual decision-making rather than majority vote, as they see the resulting and adopted broad consensus to be a safer next-best method. In fact, they prefer the big chance of winning part of the game by consensus rather than the small chance of taking the full game by majority vote. Similarly, they are quick to communicate their opposition to what they dislike, thereby hoping to get a nuisance position, a better negotiating stance and a larger piece of the pie. If the arena is highly competitive and turning towards a stalemate, they usually push officials to cook up a fair compromise. The more the interest groups compete, the more they willy-nilly contribute to the polyarchic method of decision-making. Many interest groups enhance the democratic values involved in the throughput process. They do so not primarily for promoting legitimate authority, discursiveness or transparency, but for promoting their interests within the more or less democratic EU system, as otherwise they could be scandalized. Knowing that their interests are frequently indeterminate for some time, they often have next-best preferences for such values, as safeguards against conspiring practises by rivals. Groups that fear to lose a match usually demand that the authority makes the decision in a discursive and transparent way, which offers them new chances. Many improved democratic practises result from such next-best demands, as exemplified by the Commission’s authority relying on expert groups, the EP’s sympathy with popular views and the Council’s slowly improving transparency. The support from many groups for the throughput values is, thus, partially based on their self-interest, which is usually the most solid basis of support. 341

Public Affairs, Lobbying and EU Democraty

Chapter 8

Since the early 2000s, the campaign for more open access to EU documents that some Council members oppose, is pushed by NGOs like Statewatch, the EP, special interests (lawyers, academics, decentralized authorities in member states) and the country of Germany (only Belgium ranks higher, due to the many Brussels-based interest groups) (COM 2011/ 492). The activities of interest groups that deliver experts, information and interests to the EU officials are helpful but insufficient for making the decision process more representative. Their aggregate supply can be unbalanced if representing only some sorts of interests, and the officials can be self-obsessed or deaf to the other ones. If the interest groups strongly compete with each other and oversupply the officials, the latter can enjoy the paradox of neutralization (chapter 7-VI-3) that gives them room for choice and the position of a trustee with a free mandate to devise a compromise acceptable to most stakeholders. Many officials have indeed a selfish interest in competitive arenas, which on the one side deliver more free expertise, information and support and on the other side more freedom to decide. This freedom is like that of the cook making dinner with ingredients provided by many different guests bringing in their varied ingredients and making the kitchen more representative of what nature provides. Countless interest groups are similarly helpful. The interest groups can also have negative impacts on the throughput process. Given their primary drive to win and not to lose, they appreciate democracy rarely as a goal and only as a means to be used flexibly. Professional groups, unlike amateurs, want to respect the forms, methods and values of democratic decision-making but remain optimizers, not maximizers. When believing that they are on the winning side, they often dislike any change that would make the process more representative than it, from their point of view, already is. They may even fear the three methods, because the majority-vote holds the risk of becoming identified as a minority, the consensual one of having to make wider concessions and the polyarchic one of attracting new challengers. They may have little interest in furthering the values of authoritative, discursive and transparent decisionmaking. If being in a marginal position or losing mood, the professional group is usually no different in its instrumental approach to democracy, but only in its final preferences. Then it tends to be inclined to pledge for more representative, consensual and polyarchic decision-making and for stricter application of the three values. It may remain silent only on the majority-vote method, as this might reveal that it belongs to a defeated minority. The possible negative impacts are subject to at least the three following system-bound correction mechanisms: – Watching groups. No arena has fully closed boundaries. There are always 342

Chapter 8

Public Affairs, Lobbying and EU Democraty

groups that watch and may become active. By entering an arena, they can make it more representative and change the current decision process. The mass media happily report on what they see as undemocratic practises. – Self-interests of officials. The EU officials have a selfish interest in making the process more democratic, as this will bring them, ultimately, more firmly into the driver’s seat. Interest groups in a losing mood can have the same preference for more throughput democracy, as this may turn their mood to optimism. – Output anticipation. No final decision has been taken yet. Uncertainty is the fate of any interest group during the throughput phase. Next is the output side that requires legitimacy and may cast its shadow. More quality tests of democracy are coming.

3 Output Impacts In all member countries an absolute or relative majority of people supports their EU membership, this being the summary indicator of its legitimacy or acceptance, and in almost all countries similar majorities consider this membership to be more or less beneficial [Eurobarometer, 2008, 70; Scheuer, 2005]. The specific output notions, such as the methods of limited government and rule of law and the values of accountability and responsiveness, are also widely considered legitimate. These positive scores of output democracy are not necessarily the products of interest groups. Many national politicians, members of the EP, COR and ESC, Commission officials and Court judges contribute to the legitimacy of the EU too. Often, however, they do so under pressure from interest groups that take EU legitimacy perhaps not as an ultimate goal but as a useful means to promote their own interests. Likewise, interest groups once pushed policy values that now gather sympathy and acceptance, such as the values of the open market, healthy food, sustainability and social cohesion. They also helped to promote new regime practises that may make even undesired outcomes acceptable as being ‘all part of the game’, such as the practises of better regulation, open consultations and expert meetings on the work floors of the Commission. Some groups add their own charisma as the European Round Table did, adopt traditions like functional representation and act with a show of science and expertise. In short, groups from the private or public sector contributed to EU legitimacy. This positive contribution is no different for the identified methods and values of output democracy. Many interest groups have actively lobbied for limits on the EU’s government by campaigning for more openness, representation and legitimacy in it. Citizens’ groups have pressed for 343

Public Affairs, Lobbying and EU Democraty

Chapter 8

an EP with increased powers of checks and balances. Many groups are in favour of rule of law, as they find any alternative to be worse or because they come from countries with a legalistic tradition. In contrast, the British 2008 BONGO ‘Bruges Group’ strongly campaigns against any EU law that adds costs onto UK firms, thereby putting its commercial interests above the rule of law. Groups that suffer losses from EU decisions and appeal to the Court paradoxically strengthen its rule of law. British groups have pressed for more accountability in the EU. The general and specific responsiveness of EU policy outputs can only be well-explained by taking interest groups into consideration as a factor of this. The existence of interest groups is helpful but insufficient for the creation of EU output democracy. The officials that approve and sign the decisions may positively contribute to it as well, but can also negatively by being deaf or blind to lobbying efforts from outside and giving priority to their own values and interests, as happens in every system. As interested players, they can patronize selected interest groups by applying discriminating procedures to others, as DG ENVI tried at the Aarhus Convention. They may also undermine limited government by making deals and coalitions across institutional, territorial or sectoral boundaries. Officials outside the Court can weaken the supremacy of law by interpreting laws discretionary and falling back on supplementary acts. They can use procedures of accountability for highlighting their own performances and getting positive publicity. In 2008 the British NGO ‘Open Europe’ blamed the Commission for its costly campaigns full of ‘biased propaganda’. All this can happen in any system and thus in the EU as well. Interest groups can, of course, also have negative impacts on the EU’s legitimacy. The rationale behind it is their prime aim of coming the closest possible to their targets on a dossier. Professional groups are able to see opportunities for managing or manipulating methods and values that produce EU legitimacy but, as prudent players, they tend to feel restrained in using them. Amateurish groups frequently take higher risks on this by not respecting notions like limited government or accountability. For example, in the 2008 European Transparency Initiative (ETI) many NGOs were reluctant to account for their funding. The output side can, indeed, reflect shortcomings of legitimacy caused by interest group behaviour. The following three system-linked correction mechanisms, however, prevent such potentially negative impacts from easily becoming reality: – Internal controls. Even if inside the system some legal and social controls on the officials are not working well, many others remain effective, thanks to the system’s redundancy. A rapporteur being one soul with an interest group, has to get his report accepted by shadow rapporteurs, the EP’s committee, political groups and the plenary. Com344

Chapter 8

Public Affairs, Lobbying and EU Democraty

mission officials are subject to control from competitive offices, higher officials and other institutions. – External controls. More controls come from the numerous different interest groups that spend much energy on watching each other. The more professional, established or resourced groups are becoming less certain about these comparative advantages since more groups are acquiring such qualities. Groups that believe themselves to be on the losing side may blame the officials for lacking legitimacy or accountability. The mass media, when smelling a wisp of smoke, may report about a fire that burns an official. – Feedback anticipation. The feedback mechanism casts a long shadow. If an output is widely considered as falling short of democracy, it may evoke serious feedback reactions. The interest group aware of the risk of scandal prudently disciplines its behaviour beforehand. The group that is nonchalant on this easily gets the chance to discover the necessity of the virtue of prudence.

4 Feedback Impacts The basic unit of citizenship is the mass of individual citizens. In the largescale EU they are highly dependent on their proxy of intermediate organizations such as companies, NGOs, government offices, parties and mass media [Dahl and Tufte, 1973]. Whether these organizations are really intermediate and act as corporate citizens, depends on their representativeness for their members, clients or, in short, their following among the citizens. The feedback loop from citizens to the EU is in its turn fed by the flow of information about what happens in the EU. The topics can be anything and range from draft legislation and institutional changes to specific issues and scandals. The sources of information can range from EU statements and domestic comments to local media and gossip. Even if incomplete, distorted or misperceived this flow of information can push citizens to engagement and more EU citizenship. There are always some corporate or individual citizens that react to what happens (or does not happen) at the EU level. Citizenship creates a positive feedback loop of democracy, even if its message is sceptical or oppositional, like on the post-2007 crises. Citizens can react according to their civic spirit and/or civic behaviour. In the former case they develop new beliefs, values or judgements regarding the EU or confirm their old ones; and in the latter case, they react by engaging in activities that support or oppose what happens. Usually this feedback loop runs through the indirect channels of civil organizations that adopt some interests of their following and act for it at the EU level. Much EU lobbying has this history of a multi-layered feedback loop, which 345

Public Affairs, Lobbying and EU Democraty

Chapter 8

starts at some mass level of consumers, farmers or workers and takes shape through organizations that act as corporate citizens. The loop can positively affect specific values and methods of feedback democracy. Examples are the following. The extension of the internal market as pushed by MNCs and consumer groups has reinforced the value of tolerance for people, products and ownership from foreign countries. Intolerance meets more retaliation now than in the past, thanks to specialized NGOs at the EU level. Regional groups have promoted new EU ways and means for the value of identity, such as regional cohesion and subsidiarity. Private groups push more freedoms and rights, since they enjoyed the taste of market liberalization and free movement in the EU. Many groups have pushed the Commission to establish better linkage systems by inviting them to provide information and to sit in expert groups. Domestic political parties and the mass media in most countries are hardly a feedback channel and mainly a double dead-end route. On the one hand they inform their public about EU affairs in much lower quantity and quality than about domestic affairs and on the other hand they hardly deliver domestic information to the EU level. Only when the Council is engaged in a crisis, they devote more attention to EU affairs and often display their illusion that this can make a difference in the EU. Interest groups are helpful but insufficient for the promotion of feedback democracy in the EU. Citizenship has many more determinants, such as (lack of) welfare, education, free time and social networks [Milbrath and Goel, 1977]. Their unequal distribution may explain why there is so much variation of EU citizenship between corporate and individual citizens and within each category. Thanks to the EU, the values of tolerance and identity have acquired new meanings and forms, but their status in society is also dependent on much more than the EU and its interest groups alone. For example, respect for pluralism is also determined by the social situation in the neighbourhood that often is the source of tension. The formation of new identities, such as the regional ones, is mainly caused by regional factors such as language or religion, and at most catalysed by the EU. The proclamation of freedoms and rights through the Official Journal is not sufficient, as such values must be embedded in wider social demands and adequate infrastructures. EU stakeholders pushing for more and better feedback linkages can fail in their efforts, as the weak feedback streams of most political parties and mass media show. Interest groups can also have negative impacts on the feedback for EU democracy. The Commission exemplifies this by its management of both its 2010 OC on Tobacco Review in the report on which it denied the 96% of surveyed individuals wanting a milder regime (chapter 2-III-2) and the 2012 ECI that is full of thresholds to get a citizens’ initiative accepted 346

Chapter 8

Public Affairs, Lobbying and EU Democraty

(chapter 7-VII-3). If people come to believe that the EU is a sort of republican court with only lobbying groups and factions, they may lose civic spirit or behaviour and become indifferent or passive. Such scepticisms and dissatisfactions with EU democracy are, however, intensely channelled and also create a positive feedback [Neshkova, 2010]. Reversely, activist groups on tolerance and identity can cause the effects of rising social conflicts and cultural confusion. The lobbying in favour of equal freedoms and rights to foreigners was only partially successful due to the fanned tensions in neighbourhoods around job vacancies, house purchases or parking spaces. Counter-lobbies in the EU may challenge domestic freedoms and rights, such as the Dutch coffee shops selling soft drugs, the Spanish sport of bull-fighting, the British working hours and the CEEC’s appetite for unhealthy but good-tasting meat and wine. Organizations that pretend to be corporate citizens but in fact hardly act on behalf of their following, set EU feedback democracy at risk [Warleigh, 2001]. These potentially or really negative impacts remain limited, so far, by the workings of at least the following five correction mechanisms: – Official interferences. The EU officials, when getting feedback on negative lobby impacts, usually interfere and take some measures to deal with them. Commission officials welcome excluded interest groups, MEPs support weak special interests and Council ministers lobby under home pressure for saving domestic folklore as coffee shops and working hours. – Looking for support. Many EU interest groups search for new support groups and potential clients. In the field of DG RTD policies, big companies have made new alliances with SMEs, universities and NGOs. Groups acting in an unfriendly arena always search for new groups that can give support. – Protest groups. At the domestic levels new protest groups always arise. In many countries the prime example of this is now the national parliament, feeling itself marginalized on EU affairs. The concerned feedback of many parliaments on the same issue or dossier may receive preferential EU treatment. – New groups. Formerly outside groups have become better organized for EU action. The previously weak groups of consumers and workers are respected EU groups now. Groups of patients or retired people are growing in importance, and immigrant groups may follow. – Social control. Not the least, social control is working. If an interest group is widely seen as misbehaving according to some notion of democracy, it easily acquires a bad reputation and an isolated position among its citizens, members, stakeholders and EU officials. The fear of this boomerang is a self-correcting mechanism. 347

Public Affairs, Lobbying and EU Democraty

Chapter 8

IV Improving EU Democracy by EU PAM and Lobbying The relationship between interest groups as an independent variable and EU democracy as a dependent one is for two reasons more complex than could be described above. Firstly, there is the multifinality of interest groups. They create more positive or negative impacts than solely upon democracy, as shown by a few more examples. By their cross-border group formations they may strengthen the integration of Europe but by their competitions they may weaken it too. Their direct settlements (outside the EU) of irritating differences (such as on standardization) may contribute to EU stability, growth and welfare but also cause cartel-like effects. Their impacts upon the formation of special EU policies (such as on the environment, agriculture or R&D) may be in only their interest alone or also that of many others. Any impact assessment depends on what really happens and the value judgement about it. An impact on another value than democracy may even have positive or negative effects on an element of EU democracy. The interest groups are, again, usually driven not by the intention to create such impacts, but by their selfishness that has many side effects, good and bad ones, virtues and vices [Mandeville, 1705]. Secondly, there is the multicausality of EU democracy. Many more actors (and factors and vectors) than interest groups alone can have impacts upon it. Among them are the following three. Firstly, EU officials contribute to EU democracy. Positive examples are the contributions of the Commission to better stakeholders’ participation, of the EP to discursiveness, and of the Council (by any new treaty) to more limited government; in mid-2012 the EU Council’s President urged for an agenda to strengthen ‘the ground of public support on which we stand’. Secondly, the mass media contribute negatively rather than positively to EU democracy, as they report in low quality and quantity about the EU, thereby leaving their users without useful information relative to their EU citizenship. Cyber-democracy may become a third factor of EU democracy [Dai and Norton, 2007; Wilhelm, 2000] but its impacts are so far marginal and its assumptions about cyber-communication between citizens and officials doubtful [Hindman, 2008]. Our focus here remains, however, not how EU democracy can or should be explained or strengthened but the relationship between interest groups as ‘cause’ of EU democracy and the latter as its ‘consequence’. The observations above that these groups have many more impacts and that EU democracy has many more causes prevent simplification. We assume that interest groups remain a factor of the europeanization of notions of democracy in both the EU and the member countries. Some of their 348

Chapter 8

Public Affairs, Lobbying and EU Democraty

impacts may fit nicely with current notions of democracy and then be considered positive. Others are potentially or truly at odds with them. These negative impacts may be reduced or corrected by the system-bound mechanisms. Our question now is how EU democracy can benefit more from interest groups, even if these benefits are only unintended side effects of their behaviour. This big question can be downsized as follows. Can the positive impacts be strengthened and increased? Can the negative ones be reduced or blocked? Can the correction mechanisms be promoted and reinforced?

1 Strengthening the Positive Impacts Referring to figure 8.1, we conclude that interest groups have some positive impacts on all elements of democracy, with two exceptions. One is the input method of competitive elections and the other the throughput method of decision-making by majority-vote. Most groups keep their hands off these two. To the other notions they contribute, willy-nilly, also positively, although with strong variation by type of interest group. Clear enough, only the active groups can contribute positively. The passive ones are irrelevant to the positive impacts. In chapter 7, the established and professional groups emerged as being the most important among the active groups. The established ones, usually not falling short of assets, have not only the capacity and the drive but frequently also the invitations from officials to participate. They produce the high quantity of impacts on EU democracy, both positively and negatively. Professional groups that are not identical to established or commercial ones, produce the best-quality impacts on EU democracy. Thanks to their preparatory work they manage their PA prudently by promoting, for example, consensus, transparency and feedback in the EU. The positive impacts can, in consequence, be strengthened by the triple approach of activation, establishment and professionalization. The more the interest groups in the EU area are activated, well-established and professionalized, the more they can produce beneficial impacts on democracy. Only a few active, established and either amateurish or professional interest groups may be a big danger to democracy, but a surplus of highly professional groups is a great blessing to it. They keep the system open and competitive, make the process more representative and discursive, bring the officials into the driver’s seat of polyarchy and improve the feedback linkage, to mention only these effects. Time is already on the side of the triple approach. Increasingly, interest groups are active in the EU, build up established positions and gain in professional PA. Some do so under pressure from the perceived relevance of the EU and their competition 349

Public Affairs, Lobbying and EU Democraty

Chapter 8

with other stakeholders. Others are encouraged by Commission officials and MEPs that have a vested interest in active, established and professional interest groups. In some countries special platforms push interest groups to better performance in the EU, such as the German section of the 1948 European Movement that is an activating network of about 230 German interest groups.

2 Reducing the Negative Impacts The main negative impacts of interest groups on EU democracy, mentioned above, can be summarized as follows. Groups that enter and permeate the system may set up thresholds for others, form cartels and overload it at its input side. In the throughput phase the groups inside may hinder others to represent their interests, act as a closed shop, pay lip service to polyarchy and legitimate authority, and minimize discursiveness and transparency. On the output side, they may injure legitimacy and key elements as limited government, rule of law, accountability and responsiveness. At home they may dispirit citizenship, weaken basic EU support, disregard civil tolerance and common identity, discredit the freedoms and rights, and damage some linkages. These risks of PA and lobbying are worries among many citizens and journalists, being usually not the best informed about the EU. Of course, such risks may and do happen, at least now and then. Do they occur so often that they really endanger (an element of) EU democracy? The answer varies by type of interest group. The passive or absent groups are highly relevant to these risks or negative impacts as, maybe paradoxically, they leave more room for action to the active ones. In a basically open EU, a threshold or closed shop, erected by the active few, can only endure thanks to the many that are passive. Among the active groups, the semi-established ones create the high quantity of negative impacts. Falling short of assets like networks, stable position and good image, they want quick successes and take the risks of engaging in practises than can be seen as ‘dirty lobbying’ [Bennedsen and others, 2011]. In contrast, the non-established groups that are only occasionally active, seldom have any impact, while the best-established groups have much to lose when they are the talk of the town. Most negative impacts by quality come from the semi-professionals in PAM as they have limited skills in political engineering or apply them nonchalantly. They try to pick flowers on ridges by, for example, acting arrogantly, erecting thresholds to others, obstructing transparency or limiting the freedoms of others. In contrast, the real amateurs lack the skills to damage much more than their own interests, while the full professionals respect the limits of PAM in the EU prudently, as they fear the risk of boomerangs from scandals caused by allegedly undemocratic behaviour. 350

Chapter 8

Public Affairs, Lobbying and EU Democraty

The negative impacts can best be reduced by the same aforementioned triple approach. Firstly, the more domestic interest groups are activated to play roles in the EU, the less free room they leave to the active few. The best activation usually comes from the rank and file of the groups and ultimately from their engaged workers, members, clients or stakeholders at home that, even if they are few, can make a leveraging difference [Simonis, 2009]. The mass media and political parties, now barely part of the EU feedback loop, might contribute to that activation by making citizens more aware and concerned about their interests in the EU. Secondly, the better established the active groups become, the more they shall lose by creating negative impacts on democracy. Their greatest risk is of becoming scandalized and isolated and thus losing their ticket to go around. Often they feel encouraged to get more professionalized, which is the third key to the reduction of negative impacts. The more professional the interest groups behave, such as by engaging in skilled preparations and embracing a sense of prudence, the less they cause serious damage to democracy. Time is on the side of the triple approach. Increasingly, public interest groups are activated by their clients or citizens and private ones by their members or workers. Rank-and-file people push their boards to become more active and better established in the EU. Well-established groups want to avoid the pitfalls of nonchalant lobbying and to become more professional. EU officials and other stakeholders, fearing fall-out from any sort of scandal, require more prudent behaviour from any group.

3 Reinforcing the System-bound Correction Mechanisms There will always remain some negative impacts from interest groups on EU democracy. Nobody can prevent an interest group from wilfully or accidentally engaging in an undemocratic practise or seeking to distort a democratic one. Only a tyrannical system, unconcerned about democratic practise, can outlaw interest groups, but such a system is, paradoxically, always run by one interest group, for example by the military, clergy or a single party. In a democracy, interest groups are an essential part of the open society [Popper, 1945]. Their positive impacts on democracy can be enjoyed as benefits and strengthened and their negative ones be reduced by stimulating the interest groups to more activation, better establishment and greater professionalism. A third option is to reinforce the system-bound correction mechanisms, summarized in figure 8.2. We concluded already that these mechanisms are most dependent on the quantity and the quality of the active groups. The higher the number of active groups, the more of them enter the EU arenas, get established, increase the competition and search for new support from outside, thereby stimulating new entrants. 351

Public Affairs, Lobbying and EU Democraty

Chapter 8

The higher the number of better-established groups, the more of them will create these effects continually, as they care about their acquired positions and want to be PAM professionals. The better the quality of interest groups, the more they behave adequately, proactively and prudently, driven as they are by the double desire to win a current game and to keep their licence to play in future games. The active, established and professional groups may attract more idle, wandering and amateurish groups too, but as long as the latter behave as followers of the former and not as cat-burgling leaders, they hardly endanger democracy. Most newcomers take, indeed, the more active, established and professional groups as examples to follow. Any effort to strengthen the correction mechanisms should, thus, focus on increasing both the quantity and the quality of the active interest groups. The first is a matter of stimulating the quantity of all sorts of interest groups at the EU level. There is hardly any need for a kind of official EU stimulation policy on this. Many interest groups already have strong

Self-Correcting EU Lobby-Democracy Input-Side • Openings from the Inside • Scrambling Groups Outside • New Input Collectors Throughput • Watching Groups • Officials’ Self-interest • Output Anticipation Output-Side • Internal Controls • External Controls • Feedback Anticipation Feedback • • • • • Figure 8.2

352

Official Interferences Looking for Support Protest Groups New Groups Social Control

Chapter 8

Public Affairs, Lobbying and EU Democraty

desires to participate, caused by perceived threats and opportunities. Many also feel sufficiently compelled by their competitors in the market or policy sector. The invitations from inside the EU are at most a temporary problem, as many officials and groups inside want to get new groups on board. Most obstacles to participation relate to insufficient capacity at home, as manifested by internal dissent, lack of knowledge, poor mix of resources and skills and/or poor image. Their improvement is up to these groups, maybe with some help from stakeholders on their shared-list, including the Commission that may give temporary financial support to weakly organized groups with new interests. The second focus regards quality improvement, which is also strongly dependent on the quantity of interest groups. The higher the number of active and established groups, the stronger will be both their competition (which stimulates homework) and their social control (which stimulates prudence). This competition and social control are the two main factors of their professionalization in PAM, which is in their own hands and hardly needs specific EU policy. To keep the system open and competitive, only a few watchdogs are needed. DG COMP gives the great example here. It rarely develops specific competition rules for different trades but relies on general rules to keep all markets open and competitive and it intervenes when receiving information from interest groups that feel hurt by market failures. A similar approach is suitable for the political market of the EU. Open entry and fair competition are the two preconditions for strengthening the inbuilt correction mechanisms that promote EU democracy. Of the two preconditions the former is more important than the latter, as an open entry to a market that is found relevant automatically attracts many stakeholders that produce fairness by competition. The current state of high EU competition among countless interest groups, which in any dossier is much higher than at any domestic level, meets to a great extent the two preconditions. A few groups may try to close off an entrance or play dirty tricks, but the efforts usually fail soon, thanks to correcting pressures from inside and outside. The two best possible watchdogs are, firstly, as in the case of the EU’s competition policy, the other groups that feel themselves to be injured by imperfections of entry and competition and, secondly, the mass media, being the two main variants of social control, the mother of all control. This can work even better once the mass media that are largely absent, weakly established or insufficiently professional (but also facilitated by the 2008 ETI), start to function as serious watchdogs in the EU. If they (and their citizens) are really worried about lobbying in the EU, they can make a profitable business model out of reporting on it. Time is on the side of the two notions of openness and fairness. Although still differing by arena, sector and country, they are now under europeanization. Many 353

Public Affairs, Lobbying and EU Democraty

Chapter 8

more interest groups take safety margins and become more prudent now. The idea of a basically self-regulating lobby democracy is far from new. The views of Machiavelli expressed at the start of this chapter illustrate his trust in participatory democracy as being optimal to both the people and (even) the ruler. The core idea was further developed by the US founding fathers in their Federalist Papers [1788], a misleading title as their federalism means centralist rule being the opposite of the older continental European definition of decentralized rule (which makes the F-word confusing in the EU). In Paper 10, James Madison developed the following logic. An interest group, called a faction, he considered to be a normal phenomenon of free human life. If such a selfish group is seen as devilish, then there are two possible solutions. One is to remove the causes. This can only be done either by tyrannical rule or by an equalization of interests, which are respectively undesirable and impossible. The other solution is to reduce the negative effects of factions. This can best be done by allowing into the system the highest number and the highest diversity of factions. Thanks to their different interests and mutual competitions, they will check and balance each other and all together they prevent the situation in which one or a few faction(s) can dominate the others, as every faction shall always win or lose something in every game. In fact Madison considers openness and competition as twins. Quoting Madison: ‘Extend the sphere and you take in a greater variety of parties and interests; you make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens; or if such a common motive exists, it will be more difficult for all who feel it to discover their own strength and to act in union with each other.’ In short: one or a few devils can best be exorcized by letting in as many different devils as possible. Then they will not become saints, of course, but human beings controlled by other human beings. This ingenious vector rather than factor approach and without a ruling actor is the solution Madison recommends. In addition, he pleads for a limited government, with many formal checks and balances, in order to prevent the state becoming a devil.

V Extra: Improving EU Democracy by the Study of EU PAM Our proposition here is that the study of PAM at the EU level can greatly contribute to EU democracy, as measured by the various notions of democracy. The better developed this study is and the more widespread its knowledge and application are, the smaller the chance of closed shops and unfair practises in real EU life. These two conditions are inseparable. A well354

Chapter 8

Public Affairs, Lobbying and EU Democraty

developed science can be used for better or worse. Those who master it exclusively have the choice of both. But once this knowledge is widely spread among many, it loses its potential to overpower others. Then, the many others can counterbalance any single master and all together they, willy-nilly, respect each other and abstain from abuses. Any diffused knowledge must, however, have a sufficiently sound scientific basis. If people take their beliefs or myths as knowledge or neglect chain reactions, they may cause a lot of damage, maybe not by criminal intent but by stupid mistakes, which is even worse according to the French politician Charles de Talleyrand-Périgord (1754-1838). In short, the promotion of a wider diffusion of empirical knowledge about professional PAM on the EU playing-field may enhance EU democracy. What is its current state? In chapter 2-V we described the academic state of knowledge about the EU as once widely considered unsatisfactory: many different grand theories and big concepts being normative rather than empirical and lacking explaining power and predictive value. Three new trends started: empirical field research, inductive reasoning and mid-level theorizing. The old pretension to tell the grand story was replaced by the modest ambition to make mid-level statements that are reliable and valid for at least some variables or parts of the EU, such as a policy sector, type of arena or influence approach. Mainly younger and often American scholars promoted these trends but were not alone. The growing family of PA practitioners who, due to their jobs, have to know how the EU really works, criticized the old body of pretentious knowledge silently, by preferring to rely on their own trials and errors and, in a few advanced cases, their own R&D on PAM in the EU. They discovered, for example, the work floors of the EU, the nonlegislative acts and the functions of EuroFeds, and they developed their own forms of homework, Triple P, fine-tuned lobbying and more. Only few practitioners shared their findings with others and only few academics familiar with them saw their findings as potentially relevant and valid. In short, very few scholars and practitioners walked and still walk over the fragile bridge between societas and academia and have joint meetings, activities and outlets (chapter 1-XI). Between the two a perfect match is possible. Practitioners have real-life information based on experiences and want to get useful knowledge, while scholars want to gather empirical information and to deliver valid and reliable knowledge. Together they have a common interest in improving the knowledge basis of EU PAM. If kept exclusive, such knowledge might be used against EU democracy too. However, it is already disseminated among many people, particularly by four channels of diffusion, anticipated in chapters 1-XI and 6-X. Firstly, more results from research on EU PAM are spread through academic books and journals like the Journal of Public Affairs (although it is 355

Public Affairs, Lobbying and EU Democraty

Chapter 8

read by few practitioners). Secondly, some consultancy firms like ComRes and Burson-Marsteller publish outcomes of questionnaires on PAM in the EU and consultants working for different clients disseminate their insights among many clients. Thirdly, more private institutions and postgraduate schools offer training courses on EU PAM on a more or less commercial basis, with experienced practitioners and researching academics on it as lecturers. Finally, there are the few university programmes with a course on interest groups practising PA in the EU. Their students get a basic taste of it and maybe an appetite for more. Thus, the academic knowledge and the practical insights, as collected in this book, get wider diffusion in still largely separated circles of academics and practitioners. The demand for knowledge and insights on PAM in the EU largely comes from the practitioners. As consumers, many of them have a strong appetite for three products. 1 Results. They want to see results of research, questionnaires and the like. As consumers in a restaurant, they are less interested in the cooking than in the meal. They want to know the current facts, for example, about the EU work floors and the changes among stakeholders rather than backgrounds (causes and consequences) behind, which would enable them to better interpretation, anticipation and use. 2 Checklists. The practitioners are most interested in ready-made checklists regarding the design of, for example, a long-list, the better tactics in a particular arena, the fine-tuned lobbying or the improvement of PA at their home front. They have much less interest in the methodology behind the creation of such checklists, although it could enable them to enhanced PA performance. 3 Prescriptions. They want quick prescriptions for what is better to do or not to do, in order to reach their targets better, easier and sooner. Holding such a short-term view on 2E means that they neglect the need to conduct a diagnosis beforehand and to review the chain reactions afterwards. By following this path they score poorly on the efficiency and effectiveness of their PA.

Room for improvement The currently diffused information on PAM in the EU is, altogether, focused on ready-made applicable knowledge. This may satisfy many practitioners but hardly the scholars, to the disadvantage of both. Thus there is plenty of room for further improvement, and two priorities can be suggested. First of all, the diffusion that now goes through channels that are much separated from each other and consumer-driven needs two improvements. 356

Chapter 8

Public Affairs, Lobbying and EU Democraty

1 Ramifying the channels. The current channels are mainly connected to their own users, respectively academics, established interest groups, trainees and students. Interest groups of central, regional and local governments and small-sized private businesses and NGOs are hardly connected. In the southern and eastern countries most of them still lag behind in the awareness and practise of EU PAM, let alone its body of knowledge. The channels need more ramification. Efficient channels are electronic means (e-learning) and visiting scholars (travelling faculty). 2 Balancing the contents. The contents of produced knowledge can be improved by a better match between the demand sides of both practitioners and academics. If the former would develop an interest in more than just results, checklists and prescriptions and get interest in more substantial knowledge, they could inspire the researchers to deliver more useful knowledge and so be rewarded. If the academics would become seriously interested in the experiences and insights of practitioners, they might inspire them to become sharper EU watchers and be rewarded with even more inductive information from the EU playingfield. If both groups would meet in joint sessions, they could even stimulate and tap each other’s brains. Secondly, the channels of diffusion need to be fed with PAM research agendas inspired by practitioners and academics jointly [Andersen and Eliassen, 1995]. For the sake of this chapter we propose the following three, all related to EU democracy. 1 The limits of EU PAM. The current knowledge about these limits comes from logical thinking based on observations and experiences in practise rather than from empirical research in academia. The practitioners have more experiences with and thus information about these limits. In particular, those set by the arena, receivers and environment are most relevant for the state of EU democracy. The negative effects coming from the possible practises of closed shops and unfair competition are just one example. If the results of research refute such practises, they are useful for unmasking popular myths. If not, they will certainly alert mass media acting as (helpful) watchdogs of EU democracy and hence strengthen correction mechanisms. Once publicised, tricks simply become less effective. 2 The laggards in PAM. Why do so many interest groups in the EU fall short of activism, establishment and/or professionalism, the three features of their performance that have the strongest positive impacts on EU democracy? The practitioners are a rich source of information for getting valid answers to the question – the laggards on their problems and the leaders on their solutions. The results of research can help the 357

Public Affairs, Lobbying and EU Democraty

Chapter 8

stakeholders that are passive (immigrant entrepreneurs), unfamiliar (SMEs) or nonchalant (ministries) to become more active, established and professional. 3 The notions of democracy. What are the issues and the possible outcomes of the current europeanization of the various notions of democracy? Are they also a case of irritating differences to be settled by EU method? Most throughput and output notions have been tentatively settled since the start of the EU. The input and feedback notions increasingly undergo europeanization now, which is another indicator of deepened integration down to the level of civil society. The ETI and ECI are just two recent efforts to get more openness of input and feedback from citizens. This agenda from classical political science, fed by information from practise, might help all stakeholders to get better adapted to new norms of EU democracy. The two general priorities of better diffusion and more research are justified here for their contributions to the value of EU democracy, but are useful for other values too. As said before, the science of PA management in the EU can also contribute to other core values like integration, decisionmaking and/or socio-economic welfare in Europe. This may lead to a different optimization of democratic practises, as already happens in the current EU, which has to cope with various serious crises on the inside and the outside. It is, however, not necessary to have any other justification for whatever study than simply ‘wanting to know’ or ‘art for art’s sake’.

VI Final Reflection before the Action The EU is a most ambitious experiment in living together peacefully in Europe, in a way that partially or fully replaces the old practises of accommodation by resignation, leniency, war, imitation and ad hoc negotiation. The machinery of common EU decision-making has acquired a reputation for high relevance, thanks in particular to the priority of EU law over domestic law. Increasingly, all sorts of interest groups coming from the pluralistic member countries are trying to intervene in the workings of the machinery that offers a great deal of openness and permeability. The professional interest groups often get decisions that they accept as more or less satisfying and legitimate. The EU officials inside the machinery widely welcome the arrival of new issues and stakeholders that give more relevant contents to their machinery. Their interest is to get the domestic and the European level and also the public and the private sectors better integrated into one larger whole. The EU has become a major forum for political action. 358

Chapter 8

Public Affairs, Lobbying and EU Democraty

The new influence technique of public affairs management, and particularly its professional application, is better adapted to the realities of the EU than the traditional techniques are. It is essentially based on respect for the many (frequently irritating) different values and stakeholders in any arena that have the power of retaliation, which respect is not different from that of fighting soldiers, wanting to survive a risky battle, for their armed enemies. This respect gets substance by careful preparation for the battle and by prudence during it. The EU interest groups that take their PA seriously want to survive likewise and behave accordingly. They search for better ways within the labyrinth and better moments on the trampoline of EU decision-making. Unlike soldiers and politicians living in the era of Machiavelli, they cannot hope to use deadly force against their opponents forever. Going window-in with stakeholders and officials, they lobby for peace by searching a compromise and negotiating a deal. This approach to any EU arena, euphemistically called the ‘playingfield’, requires a lot of reflection to support the action. There are so many challenges, options, menus and dilemmas that the game of chess looks easy. On the home front, the interest group has to consider the real nature of challenges and check the preconditions, such as sufficient cohesion and a good image. The transformation of the long-list of EU daydreams and nightmares to a shared-list with realizable targets requires intense study of the playing-field. For every specific arena, one has to identify the issues at stake, the stakeholders involved, the time dimensions and the arena boundaries, and to reflect on their best possible management. The fieldwork is full of dilemmas, such as regarding styles of behaviour, the use of sound and the timing of activities. The numbers of potential actors to approach, factors to use and vectors to create are always too high for managing them all and thus require clever selection. The influence process is full of limits that require much attention too. Intelligent reflection upon action enhances the chance of success, but is neither necessary nor sufficient for this. When acting by reflex or haphazardly, an interest group might also, with much luck, achieve success. However, the chances of this happening are slight and almost absent in practise. If one wants to win or at least not lose one’s game, one’s respect from others and/or one’s backing at home, one should act after reflection. The better this is done, the higher the chance of success, although there are no guarantees. Reflection is always more or less imperfect, due to, for example, lack of information, time or support. Even with thorough reflection, the interest group may fall victim to bad luck caused by an unforeseeable event, like the 1998-99 fall of the Commission Santer, the 2001 issue of ‘terrorism’ or the 2007 banking crisis, to mention high-profile cases. As bad luck can always occur, the best responses to it are scenario planning and prudency. 359

Public Affairs, Lobbying and EU Democraty

Chapter 8

Reflection is not the goal, but a most useful means to increase the chance of success in a specific arena. This does not imply that influence activities always have to follow. After reflection an interest group may come to the conclusion that it can better remain passive, for instance when it considers the cost-benefit ratio of any further action as unattractive. Then it is at least mentally active by its awareness of the current situation and of changes that may come. If it decides to play actively, it must anyhow be prepared thoroughly, which ranges from organizing the home front effectively to analyzing the arena carefully. Its objective is to take optimal positions regarding issues, stakeholders, moments and boundaries that it sees as relevant. During the PA process, the activities will shift from window-out to window-in behaviour. Then the interest group has to deal with stakeholders and officials, as otherwise it cannot win a trophy. When looking for PA tactics and techniques that will create the best possible efficiency (cost-benefit ratio) and effectiveness (desired outcome), the interest group must make a substantial investment in collecting sound information about both its own position and the EU arena and, for gaining intelligence, reflect upon this. This investment is an ongoing activity, which should start before the action and continue both during it and afterwards. The interest group rewards itself with better decisions regarding its planned actions, its responses during the interactive lobbying and its improvements for future games. Its ‘better tactics’ are never permanently fixed but are dependent on its assessment of any specific situation, its strategy and the possible chain effects, which all are always in change. Useful lessons can be taken from different groups lobbying on the same dossier, people working on the receiver’s side, observing journalists, colleagues in PAM and academics researching interest group behaviour. In PAM, the real professionals are insatiable in their demand for enhanced knowledge and learning. They are also prudent for two reasons. Firstly, every arena holds certain risks. An interest group may become part of either the winning or the losing coalition. During the arena process it may lose old friends, get a hostile setting, be caught in issue reframing, be taken as a scapegoat, become divided on the home front or be faced with nightmares. Ultimately it may have to live for many years with a most undesired outcome from the EU. Therefore, like soldiers wanting to survive in a risky field, professional groups develop ambitious targets, studious preparations and prudent decisions. The second reason for prudency is that a clever interest group wants to keep safe its licence to operate in future EU arenas. The prudent group may sometimes lose part of its interest at stake, but maintain respect from others and backing at home. If it becomes part of a controversy or scandal due to its nonchalance, it may be out of any other arena at the same time 360

Chapter 8

Public Affairs, Lobbying and EU Democraty

or in the near future, thus almost certainly lose more interests. One general conclusion from all this is that professional PA expertise is the single most important factor of success in influencing the EU and contributes to widely supported values such as EU democracy. Thus, ‘more Machiavelli’ is better than less. Nonchalant groups make more mistakes to their own disadvantage and, disrespecting opponent interests and social control, contribute not at all or even negatively to EU democracy. They can learn from professionals or by trial-and-error and in due time contribute to both their own success and a bit to democracy (or another high value). The many groups that remain passive for whatever reason and complain about EU outcomes have the worst influence on EU democracy, as they leave more room to the lower number of active groups that sit around the tables and may put them on their menu. In EU’s participatory democracy they should not blame others but only themselves. The second general conclusion is that the flowering of the study of PAM, being highly dependent on open and competitive democracy, also contributes significantly to both the ‘art of lobbying’ and any higher value. Its results help many more groups to lobby the EU in an active, established and professional manner and to diminish the chance that only a few groups will win games, sets and matches. The EU’s participatory democracy is dependent on many groups and ultimately on many people knowing better how to participate and to influence professionally. The most important tool for acquiring this PA expertise, being in short our formula 2E = MI2, is organized mental capacity. Thanks to nature most people possess more or less the same amount of brains, the raw materials for this capacity. They are, of course, free to leave that capacity underused or unused and to remain poorly organized. Such people should not complain about undesired EU outcomes, but address their own ignorance and nonchalance. The prevention of more such complaints has been the educational objective of this book.

361

References

Abbreviated reference to journals APSR CEP CMLR EJM EJPR ELJ ELR EUI EUP JCMS JEI JEPP JLS JPA LSQ PA PAR PEPS PoP WEP

American Political Science Review Comparative European Politics Common Market Law Review European Journal of Marketing European Journal of Political Research European Law Journal European Law Review European University Institute European Union Politics (paper series) Journal of Common Market Studies Journal of European Integration Journal of European Public Policy Journal of Legislative Studies Journal of Public Affairs Legislative Studies Quarterly Public Administration Public Administration Review Perspectives on European Politics and Society Perspectives on Politics West European Politics

Alesina, A. and others (2002), What does the EU do?, Florence: EUI (paper RSC 61). Alkhafaji, A. (1989), A Stakeholder Approach to Corporate Governance. Westport: Quorum. Almond, G. and G. Powell (1966), Comparative Politics. Boston: Little Brown. Almond, G. and others (2002), European Politics Today. New York: Longman. Andenas, M. and A. Türk, editors (2000), Delegated Legislation and the Role of Committees in the EC. London: Kluwer. Andersen, C. (1992), Influencing the EC. London: Kogan Page. Andersen, S. and K. Eliassen (1995), ‘EU Lobbying: The New Research Agenda’, in: EJPR, volume 27, 4, 427-441. Andersen, S. and K. Eliassen, editors (1996), The EU: How democratic is it? London: Sage. Andrews, L. (1996), ‘The relationship of political marketing to political lobbying’, in: EJM, volume 30, 10, 68-91. Antalovsky, E. and others, editors (2005), Cities in Europe, Europe in the Cities. Vienna: Europaforum. Archer, C. (2004), Norway Outside the EU. London: Routledge. Armstrong, K. (2010), Governing Social Inclusion by Open Coordination. Oxford: University Press. Arnull, A. (2006), The EU and its Court of Justice. Oxford: Oxford University Press. Arregui, J. and R. Thomson (2009), ‘States Bargaining Success in the EU’, in JEPP, volume 16, 5, 655-676.

363

References

Auel, K and A. Benz, editors (2005), ‘The Europeanization of Parliamentary Democracy’, in special issue JLS, volume 11, 3-4, 303-421. Auel, K. (2007), ‘Democratic Accountability and National Parliaments’ Scrutiny in EU Affairs’, ELJ, volume 13, 4, 487-504. Avbelj, M. (2011), ‘Supremacy or Primacy of EU Law?’, in ELJ, vol. 17, 6, 744-763. Bache, I. and A. Jordan, editors (2006), The Europeanization of British Politics. Basingstoke: Palgrave. Bache, I. (2008), Europeanization and Multilevel Governance. Lanham: Rowman and Littlefield. Bainbridge, T. (2002), Penguin Companion to the EU. London: Penguin, 3rd edition. Baines, P. and Egan J. (2004), ‘Political Marketing’, in special issue JPA, volume 4, 3, 218-299. Baines, P. and Ph. Harris, editors (2006), ‘Corporate Social Responsibility’, in special issue JPA, volume 6, 3-4, 171-307. Bal, A., editor (2011), ‘Cartoons and Political Marketing’, in special volume JPA, volume 11, 3, 135187. Bale, T. (2008), European Politics, Basingstoke: Palgrave (2nd edition). Barran, A. (2011), ‘Impact of National Business Cultures on Large Firm Lobbying in EU’, in JEI, volume 33, 4, 487-505. Barron, A. (2011), ‘The impact of national business cultures on large firm lobbying in EU’, in JEI, vol. 33, 4, 487-505. Bauer, R. and I. De Sola Pool (1960), American Businessmen and International Trade. Glencoe: Free Press. Bauer, R. and others (1963), American Business and Public Policy. New York: Atheston. Bauer, M. (2008), ‘Reforming the European Commission’, in special issue JEPP, volume 15, 5, 625780. Baumgartner, F. (2007), ‘EU Lobbying: A View from US’, in JEPP, volume 14, 3, 482-288. Beach, D. and C. Mazzucelli (2007), Leadership in the Big Bangs of European Integration. Basingstoke: Palgrave. Beach, D. and T. Christiansen, editors (2007), ‘Political Agency in the Constitutional Politics of EU’, in special issue JEPP, volume 14, 8, 1163-1332. Beetham, D. and Lord, C. (1998), Legitimacy and the EU. London: Longman. Bellier, I. (1997), ‘The Commission as an Actor: An anthropologist’s view’, in Wallace and Young (1997), 91-115. Beloe, S. and others (2003), The 21st Century NGO. London: Sustainability.com. Bennedsen, M. and others (2011), Lobbying and Bribes, a survey analysis. Munich: Ces-Ifo. Bergström, C. (2005), Comitology. Oxford: Oxford University Press. Berkhout, J. and D. Lowery (2010), ‘The changing demography of the EU interest system since 1990’, in EUP, volume 11, 3, 447-461. Berkhout, J. and D. Lowery (2011), ‘Short-term volatility in the EU interest community’, in JEPP, volume 18, 1, 1-16. Bernhagen, P and N. Mitchell (2009), ‘Determinants of Direct Corporate EU Lobbying’, in EUP, volume 10, 2, 155-176. Bertók, J. (2008), Lobbyists, Governments and Public Trust. Paris: OECD. Bertrand, G. and others (1999), Scenarios Europe 2010. Brussels: European Commission (working paper Forward Studies Unit). Beyers, J. and B. Kerremans (2007), ‘Critical Resource Dependencies and the Europeanization of Domestic Interest Groups’, in Coen (2007), 128-149. Beyers, J. and others, editors (2008), ‘Politics of Organized Interests in Europe’, special issue WEP, volume 31, 6, 1103-1302.

364

References

Bindi, M. (1998), ‘The Committee of the Regions’, in Van Schendelen (1998), 225-249. Binnema, H. (2009), How parties change. Amsterdam: Free University (Ph.D). Birch, A. (1971), Representation. London: Pall Mall. Bisson, T., editor, (1973), Medieval Representative Institutions. Hinsdale: Dryden. Blom-Hansen, J. (2011), The EU comitology system in theory and practise. Basingstoke: Palgrave Macmillan. Bongiovanni, F. (2012), The Decline and Fall of Europe. Basingstoke: Palgrave Macmillan. Börzel, T., editor (2010), ‘EU Enlargement and Societal Mobilization in Central and Eastern Europe’, special issue of Acta Politica, volume 45, 1-2, 1-267. Börzel, T. and T. Risse, editors (2012), ‘From Europeanization to Diffusion’, special issue WEP, vol. 35, 1, 1-207. Borragán, N. (2003), ‘The EU organization of Central and Eastern European Business Interests’, in Greenwood, 2003, 213-225. Borragán, N. and S. Smismans (2012), ‘Representativeness in the EU’, in JCMS, vol. 50, 3, 403421. Borrás, S., editor (2004), ‘The open Method of Co-ordination and EU Governance’, special issue JEPP, volume 11, 2, 181-337. Boucher, S. (2004), Europe and its Think-tanks. Paris: Notre Europe. Bouwen, P. (2007), ‘Competing for Consultation: The Commission and Parliament’, in WEP, volume 30, 2, 265-284. Bouwen, P. (2009), ‘The European Commission’, in Coen and Richardson (2009), 19-38. Brack, N. and O. Costa, editors (2012), ‘Diverging views of Europe within the EU institutions’, special issue JEI, vol. 34, 2, 101- 204. Bradley, K. (1998), ‘The GMO Committee on Transgenic Maize’, in Van Schendelen (1998), 207222. Brandsma, G. (2010), Backstage Europe: Comitology, accountability and democracy. Utrecht University (Ph.D.). Brants, K and K. Voltmer, editors (2011), Political Communication in Postmodern Democracy. Basingstoke: Palgrave. Braun, D. and F. Gilardi, editors (2005), Delegation in Contemporary Democracies. London: Routledge. Brecht, A. (1959), Political Theory. Princeton: Princeton University Press. Brincker, G. and others, editors (2011), Leitbilder for the future of EU. Baden-Baden: Nomos. Brito Vieira, M. and D. Runciman (2008), Representation. Cambridge: Polity Press. Broscheid, A. and D. Coen (2007), ‘Lobbying activity and fora creation’, in JEPP, volume 14, 3, 346385. Brouard, S. and others, editors (2009), The French Fifth Republic at Fifty. New York: Palgrave Macmillan. Buchholz, R. (1990), Essentials of Public Policy for Management. Englewood Cliffs: Prentice Hall, 2nd edition. Bugdahn, S. (2008), ‘Travelling to Brussels via Aarhus’, in JEPP, volume 15, 4, 588-606. Bull, M. and M. Rhodes (2007), ‘Italy: A contested polity’, in special issue WEP, volume 30, 4. Bulmer, S. and C. Lequesne, editors, (2005), The Member States of the EU. Oxford: Oxford University Press. Bunyan, T. (1999), Secrecy and Openness in the EU. London: Kogan Page. Burley, H. and others, editors (2010), Bursting the Brussels Bubble. Brussels: Alter EU. Burns, C. (2004), ‘Codecision and the European Communities’, in JEPP, volume 11, 1, 1-18. Burns, C. (2012), ‘Historical institutionalist analysis of EU biotechnology policy’, in JEI, vol. 34, 4, 341-357.

365

References

Burson Marsteller (2001), A Guide to Effective Lobbying of the European Parliament. Brussels: BKSH. Burson Marsteller (2003), A Guide to Effective Lobbying of the European Commission. Brussels: BKSH. Burson Marsteller (2005), The Definitive Guide to Lobbying the European Institutions. Brussels: BKSH. Calder, G. and E. Ceva, editors (2011), Diversity in Europe. London: Routledge. Calori, R. and P. Lawrence, editors (1991), The Business of Europe. London: Sage. Caporaso, J. (1974), Structure and Function of European Integration. Pacific Palisades: Goodyear. Carmin, J. (2010), ‘NGO capacity and environmental governance in Central and Eastern Europe’, in Börzel (2010), 183-202. Carrol, A. (1989), Business and Society: Ethics and Stakeholders Management. Cincinnati: South Western. Castiglione, B. (1528), The Book of the Courtier. Edition C. Singleton (2004), Harmondsworth: Penguin. Cawson, A. (1995), ‘Public Policies and Private Interests: The role of business interests in determining Europe’s future television system’, in Greenwood (1995), 49-61. Cawson, A. (1997), ‘Big Firms as Political Actors’, in: Wallace and Young (1997), 185-205. Chadwick, A. and P. Howard, editors (2008), Handbook of Internet Politics. London: Routledge. Chalmers, A. (2011), ‘Interest, influence and information in the EU’, in JEI, volume 33, 4, 471-486. Chari, R. and others (2010), Regulating lobbying: a global comparison. Manchester: Manchester University Press. Cherry, C. (1966), On Human Communication. Cambridge: MIT Press. Christensen, J. editor (2010), ‘EU legislation and national regulation’, special issue PA, vol. 88, 1, 3-87. Christiansen, T. and others, editors (1999), ‘The Social Construction of Europe’, in special issue JEPP, volume 6, 3, 527-719. Christiansen, T. and S. Piattoni, editors (2003), Informal Governance in the EU. Cheltenham: Elgar. Christiansen, T. and T. Larsson., editors (2007), The Role of Committees in the EU Policy Process. Cheltenham: Elgar. Christiansen, T. and S. Vanhoonacker (2008), ‘The Council Secretariat’, in WEP, volume 31, 4, 731770. Christiansen, T. and C. Reh (2009), Constitutionalizing the EU. Basingstoke: Palgrave. Chryssochoou, D. (2008), Theorizing European Integration. London: Sage, 2nd edition. Church, C. editor (2006), Switzerland and the EU. London: Routledge. Cichowski, R. (2007), The European Court and Civil Society. Oxford: Oxford University Press. Circa (2008), Political and Economic Dictionary of Central and South-Eastern Europe. London: Routledge. Claeys, P. and others, editors (1998), Lobbying, Pluralism and European Integration. Brussels: EIP. Closa C. and P. Heywood (2004), Spain and the EU. Basingstoke: Palgrave. Coen, D., editor (2007), EU Lobbying: Empirical and Theoretical Studies. London: Routledge (also special issue JEPP, 2007, volume 14, 3, 333-488). Coen, D. and J. Richardson, editors (2009), Lobbying the EU. Oxford: University Press. COM (European Commission) – Eurobarometer: Public Opinion in the EU, bi-annual. – Europe in Figures, periodical. – General Report, annual. – (2001), European Governance, Brussels (white paper).

366

References

ComRes (annual survey), The State of PA. London: www.comres.eu.com. ComRes (2011), Which information sources do Brussels opinion leaders read? London: www.comres.eu.com. ComRes (2012), EU social media usage. London: www.comres.eu.com. Conway, M. and K. Patel, editors (2010), Europeanization in the 20th Century. Basingstoke: Palgrave. Coplin, W. and M. O’Leary (1976), Everyman’s Prince. North Scituate: Duxbury (2nd ed.). Coplin, W. and M. O’Leary (1983), Political Analysis through the Prince System. New York: PSA. Coppedge, M. and others (2011), ‘Conceptualizing and measuring democracy’, in PoP, volume 9, 2, 247-267. Corbett, R., P. Jacobs and M. Shackleton, (2011), The European Parliament. London: Harper, 8th edition. Coser, L. (1956), The Functions of Social Conflict. London: Routledge. Costa, O. and K. Jørgensen, editors (2012), The influence of international institutions on the EU. Basingstoke: Palgrave. Cotta, M. and L. Verzichelli (2007), Political Institutions in Italy. Oxford: Oxford University Press. Craig, P. (2010), The Lisbon Treaty. Oxford: University Press. Cram, L. (2012), ‘Implicit and explicit identification with the EU’, in JCMS, vol. 50, 1, 71-86. Cross, J. (2011), ‘Interventions and negotiations in the Council of Ministers’, in EUP, vol. 13, 1, 4769. Culpepper, P. and others, editors (2008), Changing France. Basingstoke: Palgrave. Curtin, D. and others, editors (2010), ‘Accountability and European Governance’, special issue WEP, volume 35, 5, 929-1164. Daemen, H. and M. Van Schendelen (1998), ‘The Advisory Committee on Safety, Hygiene and Health Protection at Work’, in Van Schendelen (1998), 129-147. Dahl, R. (1971), Polyarchy. New Haven: Yale University Press. Dahl, R. and E. Tufte (1973), Size and Democracy, Stanford: Stanford University Press. Dahl, R. (1989), Democracy and its Critics. New Haven: Yale University Press. Dahl, R. (1991), Modern Political Analysis. Englewood Cliffs: Prentice Hall (5th ed.). Dai, X. and P. Norton, editors (2007), ‘The Internet and Parliamentary Democracy in Europe’, special issue JLS, volume 13, 3. Daviter, F. (2011), Policy Framing in EU. Basingstoke: Palgrave Macmillan. Dean, J. and R. Schwindt (1981) Business, Government and Society: Reading Lists and Course Outlines. Durham: Eno River Press. De Callières, M. (1716), On the Manner of Negotiating with Princes. Edition University Press of America, Washington, 1963. Deckmyn, V. and I. Thomson., editors (1998), Openness and Transparency in the EU. Maastricht: EIPA. De Groof, R., editor (2008), Brussels and Europe. Brussels: ASP. Dehousse, R. editor (2011), The Community Method: obstinate or obsolete? Basingstoke: Palgrave Macmillan. Delhey, J. (2007), ‘Analysis of Trust between EU Nationalities’, in JCMS, Volume 45, 2, 253-279. Della Sola, V., editor (2010), ‘Political Myth, Mythology and the EU’, special issue JCMS, volume 48, 1, 1-190. Demmke, C. and others (2007), Regulating Conflict of Interest for Holders of Public Office in EU. Brussels: Commission (BEPA). Dermody, J. and D. Wring, editors (2001), ‘Political Marketing’, special issue JPA, volume 1, 3, 198280.

367

References

Deschouwer, K. (2012), The Politics of Belgium. Basingstoke: Palgrave, 2nd edition. De Schutter, O. and others (2001), Governance in the EU. European Commission, Brussels. Deutsch, K. (1963), The Nerves of Government. New York: Free Press. De Vreeze, C. (2007), The role of media and news in embedding Europe. The Hague: WRR. Dinan, D., editor, (2000), Encyclopedia of the European Union. London: Macmillan. Dinan, D. (2010), Ever Closer Union. Basingstoke: Palgrave, 4th edition. Dostal, J.(2004),‘Campaigning of Expertise: the OECD’in JEPP, volume13, 3, 440-460. Drake, H., editor (2005), French relations with EU. London: Routledge. Dryzek, J. and S. Niemeyer (2008), ‘Discursive Representation’, in APSR, volume 102, 4, 481-494. Dür, A. (2008), ‘Interest groups in EU: how powerful are they?’, in WEP, volume 31, 6, 1212-1230. Dür, A. and others, editors (2010), ‘Negotiations theory and the EU: state of the art’, special issue of JEPP, volume 17, 5, 613-757. Dür, A and G. Mateo (2012), ‘Who lobbies in EU?’, in JEPP, vol. 19, 7, 969-987. Dufresne, A. and others, editors (2006), The European Sectoral Social Dialogue. Brussels: Peter Lang. Duina, F. and P. Kurzer (2004), ‘Smoke in Your Eyes’, in JEPP, volume 11, 1, 57-77. Dyson, K. (2003), ‘The Europeanization of German Governance’, in Padgett and others (2003), 161-183. Dyson, K. and A. Sepos, editors (2010), Which Europe: the politics of differentiated integration. Basingstoke: Palgrave. Easton, D. (1965), A Systems Analysis of Political Life. New York: Wiley. EC Committee (1994), Issue Management Summary. Brussels: American Chamber. Eckstein, H. (1971), The Evaluation of Political Performance. London: Sage. Egan, M. and others, editors (2010), Research Agendas in EU Studies. Basingstoke: Palgrave. Egeberg, M. (2012), ‘The European Commission as a laboratory’, in JEPP, vol. 19, 6, 939-950. Eilstrup-Sangiovanni, M., editor (2006), Debates on European Integration. Basingstoke: Palgrave. Eising, R. (2009), Political economy of state-business relations in Europe. London: Routledge. Elgström, O., editor (2003), EU Council Presidency. London: Routledge. Emerson, M. and others (1988), The Economics of 1992. Oxford: Oxford University Press. EP/ European Parliament (2003), Lobbying in the EU. Brussels: AFCO 104. EP/ European Parliament (2009-A), Activity report 2004-2009 on conciliation. Brussels: EP/ Conciliation Committee. EP/ European Parliament (2009-B), Comitology Handbook. Brussels: EP/ Conference of Committee Chairs. EPAD (annually), European Public Affairs Directory. London: Dods. Epstein, R. and U. Sedelmeier, editors (2008), ‘Postcommunist Europe after Enlargement’, special issue JEPP, volume 15, 6. Eriksen, E. and J. Fossum, editors (2012), Rethinking Democracy and the EU. London: Routledge. Eulau, H. and K. Prewitt (1973), Labyrinths of Democracy. Indianapolis: Bobbs-Merrill. EurActiv (2009-A), Corporate Survey 2009. Brussels: EurActiv. — (2009-B), Federations Survey 2009. Brussels: EurActiv. — (2009-C), Consultancies Survey 2009. Brussels: EurActiv. — (2009-D), PA Memberships. Brussels: Euractiv. — (2009-E), What do EU actors think of ETI? Brussels: Euractiv. Eurobarometer: see COM. Exadaktylos, T. and C. Radaelli (2012), Research design in European studies: establishing causality in Europeanization. Basingstoke: Palgrave Macmillan.

368

References

Favell, A. and V. Guiraudon, editors (2011), Sociology of the EU, Basingstoke: Palgrave. Featherstone, K. and C. Radaelli, editors (2003), Politics of Europeanization. Oxford: Oxford University Press. Federalist Papers (1788). Edition by C. Rossiter (1961). New York: Mentor. Fink-Hafner, D. (2011), ‘Interest representation and post-communist parliaments over two decades’, in JLS, vol. 17, 2, 215-233. Finke, D. editor (2010), European integration and its limits. Colchester: ECPR. Fleisher, C. (2005), ‘Educating present and future PA practitioners’, in Harris and Fleisher (2005), 517-536. Fleischer, C. (2007), ‘Developing the PA Knowledge Body’, in JPA, volume 7, 3, 281-290. Fleischman-Hillard (2011), EP Digital Trend Survey. www.fleishmanhillard.com. Fligstein, N. and J. McNichol (1998), ‘The Institutional Terrain of the EU’, in Sandholtz and Sweet (1998), 59-91. Fligstein, N and A. Sweet (2001), ‘Institutionalizing the Treaty of Rome’, in Sweet and others (2001), 29-55. Flora, P. (1988), Growth to Limits: Unity and Diversity. New York: Walter De Gruyter. Flora, P. and others (2005), European Regions. Basingstoke: Palgrave. Fombrun, C. and C. Van Riel (2007), Implementing Practises for Effective Reputation Management. London: Routledge. Fontana, M. (2011), ‘How actors use Europe to modify domestic patterns of policy-making’, in JEPP, volume 18, 5, 654-671. Fouilleux, E. and others (2007), ‘Council Working Groups’, in Christiansen and Larsson (2007), 96-119. Franchino F. and C. Radaelli, editors (2004), ‘Europeanization and the Italian Political System’, special issue of JEPP, volume 11, 6, 941-1111. Franchino, F. (2007), The Powers of the Union. Cambridge: COP. Frederick, W. and others (1996), Businessmen and Society. New York: McGraw (6th edition). Freeman, R. (1984), Strategic Management: A Stakeholder Approach. Boston: Pitman. Friedrich, C. (1963), Man and his Government. New York: McGraw-Hill. Friedrich, C. (1974), Limited Government. Englewood Cliffs: Prentice Hall. Fullinwider, R. (1999), Civil Society. Lanham: Roman and Littlefield. Galston, W. (1999), ‘Value Pluralism and Liberal Political Theory’, in APSR, volume 93, 4, 769-778. Garcia, B. and H. Meier (2012), ‘Limits of interest empowerment in the EU: the case of football’, in JEI, vol. 34, 4, 359-378. Gardner, J. (1991), Effective Lobbying in the EC. Boston: Kluwer. Gaxie, D. and others, editors (2011), Perceptions of Europe. Colchester: ECPR. Geddes, A. (2004), The EU and British Politics. London: Palgrave. George, S. (2000), An Awkward Partner: Britain in the EC. Oxford: Oxford University Press. George, S. and I. Bache (2001), Politics in the EU. Oxford: Oxford University Press. Geuijen, K. and others (2008), The New Eurocrats. Amsterdam: Amsterdam University Press. Giddings, P. and G. Drewry (2004), Britain and the EU. London: Palgrave. Gigerenzer, G. and others (1989), The Empire of Change. Cambridge: Cambridge University Press. Goetz, K. and others, editors (2009), ‘The EU Timescape’, special issue JEPP, volume 16, 2, 180-336. Goodhart, M. (2007), ‘EU as Challenge for Democracy’, Perspectives on Politics, volume 5, 3, 567584. Gorges, M. (1996), Euro-corporatism? Lanham: University Press of America. Gornitzka, A. and U. Sverdrup (2008), ‘Who consults? Expert Groups in EU’, in WEP, volume 31, 4, 725-750.

369

References

Gornitzka, A. and U. Sverdrup (2011), ‘Access of experts’, in WEP, volume 34, 1, 48-70. Gottweis, H. (1999), ‘Regulating genetic engineering in the EU’, in Kohler-Koch and Eising (1999), 61-82. Grabbe, H. (2005), The EU’s Transformative Power. Basingstoke: Palgrave. Garcia, L. and J. Greenwood, editors (2012), ‘The European Citizens’ Initiative’, in special issue of PEPS, volume 13, 3, 251-384. Grant, W. (1989), Pressure Groups, Politics and Democracy in Britain. London: Philip Allan. Grant, W. (2003), ‘Trust: a sceptical view’, in Greenwood 2003, 197-204. Graziano, P. and M. Vink, editors (2007), Europeanization: New Research Agendas. Basingstoke: Palgrave. Green Cowles, M. and others (2001), Transforming Europe: Europeanization and Domestic Change. Ithaca: Cornell University Press. Green, S. and others (2011), The Politics of the New Germany. London: Routledge, 2nd edition. Greene, R. (1998), The 48 Laws of Power. London: Profile Books. Greene, R. (2001), The Art of Seduction. London: Profile Books. Greenwood, J., editor (1995), European Casebook on Business Alliances. Englewood Cliffs: Prentice Hall. Greenwood, J. and M. Aspinwall, editors (1998), Collective Action in the EU. London: Routledge. Greenwood, J., editor (2003), Change in EU Business Associations. Basingstoke: Palgrave. Greenwood, J. (2011), Interest Representation in the EU. Basingstoke: Palgrave (3rd edition). Greer, S. and others (2008), ‘Mobilizing Bias in Europe’, in EUP, volume 9, 403-433. Griffin, J. (2005), ‘The Empirical Study of Public Affairs’, in Harris and Fleisher (2005), 458-480. Griffith, S., editor (1963), Sun Tze: The Art of War. Oxford: Oxford University Press. Grøn, C. and A. Wivel (2011), ‘From small state policy to smart state strategy’, in JEI, volume 33, 5, 523-539. Grossman, E., editor (2007), ‘France and the EU’, special issue JEPP, volume 14, 7, 983-1150. Guéguen, D. (2007), European Lobbying. Brussels: Europolitics. Gutmann A. and D. Thompson (2010), ‘The mind-set of political compromise’, in PoP, volume 8, 4, 1125-1143. Guyomarch, A. and others (1998), France in the EU. London: Macmillan. Haas, E. (1958), The Uniting of Europe. Stanford: Stanford University Press. Häge, F. (2011), ‘The EU Policy-making Dataset’, in EUP, volume 12, 3, 455-477. Häge, F. (2012), Bureaucrats as Law-makers. London: Routledge. Hagemann, S. and J. Declerck (2007), Old Rules, New Game: Council Decision-making after 2004 Enlargement. Brussels: CEPS. Halman, L. and others, editors (2011), Atlas of European Values. Leiden: Brill. Hardacre, A., editor (2011), How the EU works and how to work with the EU. London: John Harper Harris, P. and others, editors (2000), Machiavelli, Marketing and Management. London: Routledge. Harris, P. and C. Fleisher, editors (2005), Handbook of Public Affairs. London: Sage. Harsanyi, F. and S. Schmidt (2012), ‘Creating a PA function in countries without a PA culture’, in JPA, vol. 12, 1, 86-97. Hauser, H. (2011), ‘EU Lobbying Post-Lisbon: an economic analysis’, in Berkeley Journal of International Law, vol. 29, 2, 680-709. Haverland, M. and M. Romeijn (2007), ‘The EU Transposition Deficit’, in Public Administration, volume 85, 3, 757-778. Hayes-Renshaw, T. and H. Wallace (2006), The Council of Ministers. London: Macmillan (2nd edition).

370

References

Hayward, J. and A. Menon, editors (2003), Governing Europe. Oxford: Oxford University Press. Heard-Lauréote, K (2010), EU Governance: effectiveness and legitimacy of Commission committees. London: Routledge. Heath, R. (1997), Strategic Issues Management. London: Sage. Heffernan, R. and others, editors (2011), Developments in British Politics. Basingstoke: Palgrave. Heisenberg, D. (2005), ‘Consensus in the European Council’, in EJPR, volume 44, 65-90. Heisenberg, D. (2008), ‘How should we best study the Council of Ministers?’, in Naurin and Wallace (2008), 261-276. Held, D. (1996), Models of Democracy. Oxford: Polity Press. Henderson, K. (2009), The EU’s New Democracies. London: Routledge. Henning, C. (2004), ‘Modelling the Political Influence of Interest Groups’, in Warntjen and Wonka (2004), 94-112. Héritier, A. and M. Rhodes, editors (2011), New Modes of Governance in Europe. Basingstoke: Palgrave. Heslop, L., editor (2005), ‘GMO Food Concerns’, special issue JPA, volume 5, 3-4, 193-345. Hindman, M. (2008), The Myth of Digital Democracy. Princeton: Princeton University Press. Hirschman, A. (1970), Exit, Voice and Loyalty. Cambridge: Harvard University Press. Hix, S. and others, editors (2003), ‘Fifty Years of Research on the EP’, special issue JCMS, volume 41, 2, 191-353. Hix, S. and others (2007), Democratic Politics in the EP. Cambridge: Cambridge University Press. Hix, S. and A. Noury (2009), ‘Voting Patterns in the Sixth EP’, in LSQ, volume 34, 2, 159-174. Hix, S. and B. Hoyland (2011), The Political System of the EU. London: Macmillan, 3rd edition. Hojnack, M. and D. Kimball (1998), ‘Organized Interests and the Decision of Whom to Lobby in Congress’, in: APSR, volume 92, 4, 775-790. Hobolt, S. (2012), ‘Citizen satisfaction with EU democracy’, in JCMS, vol. 50, s1, 88-105. Holzhacker, R. (2012), ‘Strategies of civil society organizations in Europe’, in CEP, vol. 10, 1, 2347. Holzinger, K. and F. Schimmelfennig (2012), Differentiated integration in the EU, in JEPP, vol. 19, 2, 292-305. Hondak, J. and others (2010), ‘Personality and Civic Engagement’, in APSR, volume 104, 1, 85-111. Hooghe, L. and G. Marks (2008), ‘European Union’, in WEP, volume 31, 1, 108-129. Hooghe, L. (2012), ‘Images of Europe: how Commission officials conceive their institution’s role’, in JCMS, vol. 50, 1, 87-111. Hosli, M. and others (2011), ‘Voting in the Council of EU after 2004’, in JCMS, vol. 49, 6, 12491270. Howell, K. (2000), Discovering the Limits of European Integration: Applying grounded theory. Huntington: Nova. Hrebrenar, R. and others (2009), ‘Developing interest groups in Lithuania’, in McGrath (2009), 235-260. Hudock, A. (1999), NGOs and Civil Society. Oxford: Polity Press. Hughes, J., editor (2010), EU Conflict Management. London: Routledge. Hume, D. (1748), An Enquiry Concerning Human Understanding. London: Cadell. Hunger, J. and T. Wheelen (1998), Strategic Management. Reading: Addison Wesley (6th edition). Hurka, S. and M. Keading (2012), ‘Report allocation in the EP after enlargement’, in JEPP, vol. 19, 4, 512-529. Hurwitz, L., editor, (1980), Contemporary Perspectives on European Integration. London: Aldwyck. Huxham, C., editor, (1996), Creating Collaborative Advantage. London: Sage.

371

References

Illia, L. (2003), ‘Passage to Cyberactivism’, in JPA, volume 3, 4, 326-337. Jackson, R. (2007), Sovereignty. Cambridge: Polity Press. Jans, J. and others (2007), Europeanization of Public Law. Groningen: Kluwer. Jansen, T. and T. Winzen (2011), ‘Legislative negotiations in the EP’, in EUP, vol. 13, 1, 118-149. John, S. (2002), The Persuaders: When lobbyists matter. Basingstoke: Palgrave. Johnson, D. (2008), Handbook of Political Management. London: Routledge. Jordan, A. (1997), The Protest Business. Manchester: Manchester University Press. Jordan, A. and W. Maloney (1996), ‘How Bumblebees Fly: Accounting for Public Interest Participation’, in: Political Studies, volume 44, 4, 668-685. Jordan, A. and others (1999), ‘Innovative and Responsive? A longitudinal analysis of the speed of EU environmental policymaking 1967-1997’, in: JEPP, volume 6, 3, 376-398. Jørgensen, K. (1997), ‘Studying European Integration in the 1990s’, in: JEPP, volume 4, 3, 486-492. Jørgensen, K. and others, editors (2011), ‘The performance of EU in international institutions’, special issue JEI, vol. 33, 6, 599-757. Judge, D. and D. Earnshaw (2008), The European Parliament. Basingstoke: Palgrave, 2nd edition. Kardasheva, R. (2009), ‘The EP’s influence in the consultation procedure’, in JCMS, volume 47, 2, 385-409. Karlas, J. (2011), ‘Parliamentary control of EU affairs in Central and Eastern Europe’, in JEPP, volume 18, 2, 258-273. Karr, K. (2006), Democracy and Lobbying in EU. Frankfurt: Campus. Kassim, H. and others, editors (2000), The National Co-ordination of EU Policy (Volume I). Oxford: Oxford University Press. Kassim, H. and others, editors (2001), The National Co-ordination of EU Policy (Volume II). Oxford: Oxford University Press. Kassim, H. (2003), ‘The European Administration: Between Europeanization and Domestication’, in Hayward and Menon (2003), 139-161. Katz, R. and B. Wessels, editors (1999), The EP, the National Parliaments and European Integration. Oxford: Oxford University Press. Keading, M. (2006), ‘Rapporteurship Assignment in the EP’, in JLS, volume 11, 1, 82-104. Keessen, A. (2009), European Administrative Decision. Groningen: Europa Law. Kelemen, R (2012), ‘Eurolegalism and democracy’, in JCMS, vol. 50, 1, 55-71. Kelemen, R. and A. Tarrant (2011), ‘Political foundations of the EU Agencies’, in WEP, vol. 34, 5, 922-947. Kelemen, R. and S. Schmidt, editors (2012), ‘The ECJ and legal integration’, special issue JEPP, vol. 19, 1, 1-139. Keohane, R. and S. Hoffmann, editors (1991), The New European Community: Decision-making and Institutional Change. Boulder: Westview Press. Key, V. (1964), Politics, Parties and Pressure Groups. New York: Crowell. King, P. (1976), Toleration. London: Allan and Unwin. Kingdon, J. (1984), Agendas, Alternatives and Public Policies. Boston: Little Brown. Klingemann, H. and D. Fuchs, editors (1995), Citizens and the State. Oxford: Oxford University Press. Klüver, H. (2010), ‘Europeanization of lobbying activities: the DOHA round’, in JEI, vol. 32, 2, 175191. Klüver, H. (2011), ‘The contextual nature of lobbying’, in EUP, vol. 12, 4, 483-506. Knapp, A. and V. Wright (2006), The Government and Politics of France. London: Routledge (5th edition).

372

References

Knodt, M. and others, editors (2011), ‘Territorial and functional interest representation in EU’, special issue JEI, volume 33, 4, 349-521. Knoke, D., editor (1996), Comparing Policy Networks. Cambridge: Cambridge University Press. Knutelska, V. (2011), Parliamentary scrutiny of EU matters in Czech, Poland and Slovakia’, in PEPS, vol. 12, 3, 320-339. Kobrin, S. (1982), Managing Political Risk Assessment. Berkeley: California. Koeppl, P. (2001), ‘The acceptance, relevance and dominance of lobbying the EU Commission’, in: JPA, volume 1, 1, 69-80. Kohler-Koch, B. (1998), ‘Organized Interests in the EU and the EP’, in Claeys and others (1998), 126-158. Kohler-Koch, B. and R. Eising, editors (1999), The Transformation of Governance in the EU. London: Routledge. Kohler-Koch, B., editor (2003), Linking EU and National Governance. Oxford: Oxford University Press König, T. and others (2006), ‘Quantifying European Legislative Research’, in EUP, volume 7, 4, 553-574. König, T. (2008), ‘Analysing the Process of EU Decision-making’, in EUP, volume 9, 1, 145-165. Kowalczyk-Hoyer, B. (2012), Transparency in corporate reporting. Berlin: Transparency International. Krasner, S. (1999), Sovereignty: Organized Hypocrisy. Princeton: Princeton University Press. Krislov, S. (1974), Representative Bureaucracy. Englewood Cliffs: Prentice Hall. Külahci, E., editor (2012), Europeanization and party politics. Colchester: ECPR. Kumlin, S. (2011), ‘How government and opposition frame Europeanization’, in EUP, vol. 12, 4, 575-595. Kurpas, S. and others (2008), The European Commission after Enlargement. Brussels: CEPS. Ladrech, R. (2010), Europeanization and National Politics. Basingstoke: Palgrave. Laffan, B. and J. O’Mahony (2008), Ireland and the EU. Basingstoke: Palgrave. LaGro, E. and K. Joergensen, editors (2007), Turkey and EU. Basingstoke: Palgrave. Lahusen, C. (2004), ‘Social Movement Organization at the EU’, in Mobilization, volume 9, 1, 5571. Larsson, T. (2003), Precooking in the EU. Stockholm: ESO. Larsson, T. and J. Murk (2007), ‘The Commission’s relations with expert groups’, in Christiansen and Larsson (2007), 64-95. Laursen, F., editor (2011), The EU and federalism. Farnham: Ashgate. Laursen, F., editor (2012), The Making of the EU’s Lisbon Treaty. Berlin: Peter lang. Laurencell, S. (1979), Lobbying and Interest Groups: A Selected Annotated Bibliography. Washington DC: Congressional Research Service. Lauristin, M. and P. Vihalemm, editors (2009), Estonia’s Transition to the EU. London: Routledge. Leconte, C. (2010), Understanding Euroscepticism. Basingstoke: Palgrave. Legendre, A. (1993), ‘The State’s Power under Pressure’, in Van Schendelen (1993), 51-66. Leonard, M. (1999), Networks Europe. London: Foreign Policy Centre. Lewis, J. (2000), ‘The methods of community in EU decision-making and administrative rivalry’, in: JEPP, volume 7, 2, 261-289. Lijphart, A. (1997), Democracy in Plural Societies. New Haven: Yale University Press. Lindberg, L. (1963), The Political Dynamics of European Economic Integration, Stanford CA: Stanford University Press. Lindberg, L. (1970), ‘Political Integration as a Multidimensional Phenomenon’, in International Organization, volume 24, 4, 649-731.

373

References

Lindberg, B. and others, editors (2008), ‘Party politics in EU institutions’, in special issue JEPP, volume 15, 8, 1107-1265. Lindblom, C. (1959), ‘The science of muddling through’, in PAR, vol. 19, 1, 79-88. Lindblom, C. and Cohen, D. (1979), Usable Knowledge. New Haven: Yale University Press. Linsenmann, I. and others (2007), Economic Government of EU. Basingstoke: Palgrave. Lintner, P. and B. Vaccari (2007), ‘Comitology and EP’s Scrutiny’, in Christiansen and Larsson (2007), 201-226. Loewenstein, K. (1973), The Governance of Rome. The Hague: Nijhoff. Lord, C. (1998), Democracy in the EU. Sheffield: Sheffield Academic Press. Loughlin, J., editor, (2001), Subnational Democracy in the EU. Oxford: Oxford University Press. Lowery, D. (2008), ‘The EU interest system in comparative perspective’, in WEP, volume 31, 6, 1231-1252. Luitwieler, S. (2009), The Black Box of Nice Treaty Negotiations. Rotterdam: Erasmus University (Ph.D. thesis). Luttbeg, N., editor, (1968), Public Opinion and Public Policy. Homewood: Dorsey. Mabbett, D. and W. Schelke, editors (2009), ‘Politics of conflict management in EU regulations’, special issue WEP, volume 32, 4, 699-865. Machiavelli, N. (1513), The Prince. Edited by A. Gilbert (1964), New York: Hendricks House. Machiavelli, N. (between 1513-1517), The Discourses. Edited by B. Crick (1970), Middlesex: Pelican Classics. Magone, J. (2009), Contemporary Spanish Politics. London: Routledge, 2nd edition. Magone, J. (2011), Contemporary European Politics. London: Routledge. Mahon, J. and B. Mitnick (2010), ‘Reputation shifting’, in JPA, volume 10, 4, 280-299. Mahon, J. and S. Wartick (2012), ‘Corporate social performance’, in JPA, vol. 12, 1, 12-28. Mahoney, C. and M. Beckstrand (2011), ‘Following the money: EU funding of civil society organizations’, in JCMS, vol. 49, 6, 1339-1361. Mair, P. and J. Thomassen, editors (2010), ‘Representation and EU Governance’, special issue JEPP, volume 17, 1, 1-148. Maloney, W. and G. Jordan (1997), ‘The Rise of Protest Business in Britain’, in Van Deth (1997), 107-124. Mammone, A. and G. Veltri, editors (2010), Italy today: sick man of Europe. London: Routledge. Mandeville, B. (1705), The Fable of the Bees: Private Vices, Public Benefits. London: Garman (1934). Marek, D. and M. Baun (2010), Czech Republic and the EU. London: Routledge. Marks, G. and others, editors (1996), Governance in the EU. London: Sage. Marks, G. and others (2002), ‘What do subnational offices think they are doing in Brussels?’, in Regional and Federal Studies, volume 12, 3, 1-23. Marks, G. (2012), ‘Europe and its empires: from Rome to the EU’, in JCMS, vol. 50, 1, 1-20. Marshall, A. (2005), ‘Europeanization at the Urban Level’, in JEPP, volume 12, 4, 668-686. Marshall, D. (2010), ‘Who to lobby and when in EP committees?’, in EUP, volume 11, 4, 553-575. Martins, A. and others (2012), ‘Information flow and communication deficit’, in JEI, vol. 34, 4, 305-322. Matthews, D. and J. Stimson (1975), Yeas and Nays: Normal decision-making in the US House of Representatives. New York: Wiley. Mattila, M (2009), ‘Role-call analysis in the Council after 2004 Enlargement’, in EJPR, volume 48, 6, 840-857. Mattila, M. and T. Raunio (2012), ‘Drifting further apart on EU: national parties and their electorates’, in WEP, vol. 35, 3, 589-606.

374

References

Maull, H., editor (2006), Germany’s Uncertain Power. Basingstoke: Palgrave. Mayes, D., editor (1992), The European Challenge: Industry’s response to 1992. London: Harvester Wheatsheaf. McCormick, J. (1999), Understanding the EU. London: Macmillan. McCormick, J. (2011), Machiavellian Democracy. New York: Cambridge University Press. McDonagh, B. (1998), The Paradox of Europe: An account of the negotiation of the Treaty of Amsterdam. Dublin: Europe House. McElroy, G. (2006), ‘Committee Representation in the EP’, in EUP, volume 7, 1, 5-29. McElroy, G. and K. Benoit (2011), ‘Policy positioning in the EP’, in EUP, vol. 13, 1, 150-167. McGowan, L. and D. Phinnemore, editors (2006), Dictionary of the EU. London: Routledge (3rd edition). McGrath, C. (2005-A), Lobbying in Washington, London and Brussels. Lampeter: Edwin Mellen. McGrath, C. (2005-B), ‘Towards a lobbying profession’, in JPA, volume 5, 2, 124-135. McGrath, C. (2007), ‘Persuasive communication’, in JPA, volume 7, 3, 269-281. McGrath, C. (2008), ‘Development and Regulation of Lobbying in New EU Member-states’, in JPA, volume 8, 1-2, 15-32. McGrath, C., editor (2009), Interest groups and lobbying in Europe. Lampeter: Edwin Mellen. McGrath, C. and others (2010), ‘The evolving discipline of PA’, in JPA, volume 10, 4, 335-352. McKellar, R. (2010), Short guide to political risk. Farnham: Gower. McLaren, L. (2006), Identity, Interests and Attitudes to European Integration. Basingstoke: Palgrave. McLaughlin, A. (1994), ‘ACEA and the EU-Japan Car Dispute’, in Pedler and Van Schendelen (1994), 149-166. Meerts, P. and F. Cede, editors (2004), Negotiating European Union. Basingstoke: Palgrave. Mendes, J. (2011), Participation in EU Rule-making. Oxford: University Press. Meunier-Aitsahalia, S. and G. Ross (1993), ‘Democratic Deficit or Democratic Surplus’, in: French Politics and Society, volume 11, 4, 57-69. Michalowitz, I. (2007), ‘What determines influence?’, in JEPP, volume 14, 1, 132-151. Milbrath, L. (1963), The Washington Lobbyists. Chicago: Rand MacNally. Milbrath, L. and M. Goel (1977), Political Participation. Chicago: Rand MacNally. Mitchell, R. and others (1997), ‘Toward a theory of stakeholder identification and salience’, in: Academy of Management Review, volume 22, 4, 853-896. Moravcsik, A. (2000), ‘Integration Theory’, in Dinan (2000), 278-291. Morgenthau, H. and K. Thompson (1985), Politics among Nations. New York: Knopf, 6th edition. Munoz, J. and others (2011), ‘Institutional trust and multinational governance in EU’, in EUP, vol. 12, 4, 551-574. Murphy, G. and C. McGrath, editors (2011), ‘Lobbying in Ireland’, special issue JPA, volume 11, 2, 71-134. Naurin, D. and H. Wallace, editors (2008), Unveiling the Council of EU. Basingstoke: Palgrave. Naurin, D. and A. Rasmussen, editors (2011), ‘Linking inter- and intra-institutional change in the EU’, special issue WEP, volume 34, 1, 1-179. Neshkova, M. (2010), ‘The impact of subnational interests on supranational regulation’, in JEPP, volume 17, 8, 1193-1211. Neunreither, K. and A. Wiener, editors (2000), European Integration after Amsterdam. Oxford: Oxford University Press. Neyer, J. (2000), ‘Justifying Comitology: The Promise of Deliberation’, in Neunreither and Wiener (2000), 112-128. Neyer, J. (2004), ‘Explaining the unexpected in EU decision-making’, in JEPP, volume 11, 1, 19-38.

375

References

Neyer, J. and A. Wiener, editors (2011), Political theory of the EU. Oxford: University Press. Nicolaides, P. and A. Suren (2007), ‘Rule of Law in EU: What numbers say’, in Eipascope, 1, 3339. Niemann, A. (2006), Explaining Decisions in EU. Cambridge: Cambridge University Press. Norris, P. (2011), Democratic deficit. New York: Cambridge University Press. Nowak, T. (2007), The Influence of the ECJ on the EU’s Legislative Process. Groningen: University (Ph.D. thesis). Nownes, A.J. (1999), ‘Solicited Advice and Lobbyist Power: Evidence from three American States’, in LSQ, volume 24, 1, 113-123. Nugent, N. (2001), The European Commission. Basingstoke: Palgrave. Nugent, N. (2003), Government and Politics of the EU. Basingstoke: Palgrave (5th edition). Nugent, N. (2010), Government and Politics of the EU. Basingstoke: Palgrave (7th edition). Nutt, P. (1999), ‘Surprising but true: Half the decisions in organizations fail’, in Academy of Management Executive, volume 13, 4, 75-90. O’Brennan, J. and T. Raunio, editors (2007), National Parliaments in the Enlarged EU. London: Routledge. Olson, D. and G. Ilonszki, editors (2011), ‘Post-communist legislatures in the second decade’, special issue JLS, vol. 17, 2, 115-255. Ongaro, E. and others (2012), ‘EU Agencies’, in Verhoest and others (2012), 400-410. O’Shaughnessy, N. (1990), The Phenomenon of Political Marketing. London: Macmillan. Padgett, S. and others, editors (2003), Developments in German Politics. Basingstoke: Palgrave. Page, E. (1997), People Who Run Europe. Oxford: Clarendon. Panke, D. (2012), ‘Lobbying institutional key-players: how states seek to influence the institutions’, in JCMS, vol. 50, 1, 129-150. Papadimitriou, D. and D. Phinnemore (2008), Romania and EU. London: Routledge. PARG (1981), Public Affairs Offices and Their Functions. Boston: University. Pari, I. (2003), ‘Members from the South in EU Associations’, in Greenwood (2003), 207-212. Pateman, C. (1970), Participation and Democratic Theory. Cambridge: Cambridge University Press. Paterson, W. and others, editors, (2009), Research Agendas in EU Studies. Basingstoke: Palgrave. Patterson, L. (2000), ‘Biotechnology Policy’, in Wallace and others (2000), 317-344. Pedler, R. and M. Van Schendelen, editors (1994), Lobbying the EU. Aldershot: Dartmouth. Pedler, R. (1994), ‘The Fruit Companies and the Banana Trade Regime’, in Pedler and Van Schendelen (1994), 67-92. Pedler, R. and G. Schaefer, editors (1996), Shaping European Law and Policy: The Role of Committees and Comitology in the Political Process. Maastricht: EIPA. Pedler, R. (2002), European Union Lobbying. Basingstoke: Palgrave. Peers, S. and M. Costa (2012), ‘Accountability for delegated and implementing acts after Lisbon , in ELJ, vol. 18, 3, 427-460. Pennings, P. (2006), ‘Empirical Analysis of the Europeanization of National Party Manifestos 1960- 2003’, in EUP, volume 7, 2, 257-270. Pennock, J. and J. Chapman, editors (1968), Representation. New York: Atherton. Peters, B. and V. Wright (2001), ‘National co-ordination of European policy-making: Negotiating the quagmire’, in Richardson (2001), 155-178. Peterson, J. (2001), ‘The Choice for EU Theorists’, in: EJPR, volume 39, 3, 289-318. Phillips, R. and R. Freeman, editors (2010), Stakeholders. Cheltenham: Edward Elgar. Phillips, R. editor (2011), Stakeholder theory. Cheltenham: Edward Elgar.

376

References

Pijnenburg, A. (1998), ‘EU lobbying by ad-hoc coalition’, in JEPP, volume 5, 2, 303-321. Pinder, J., editor, (1999), Foundations of Democracy in the EU. London: Macmillan. Pitkin, H. (1967), The Concept of Representation. Berkeley: California University Press. Pitkin, H., editor (1969), Representation. New York: Atherton. Pitt, L., editor (2012), ‘Web 2.0, social media and creative consumers’, special issue JPA, vol. 12, 2, 105-169. Pleines, H. (2011), ‘Representation of functional interest groups from new member states at EU level’, in JEI, volume 33, 4, 507-521. Poguntke, T. and others, editors (2006), Europeanization of National Political Parties. London: Routledge. Pollack, M., editor (2000), ‘Democracy and Constitutionalism in the EU’, in ECSA Review, volume 13, 2, 2-7. Pollack, M. (2005), ‘Theorizing EU Policy-making’, in Wallace and others (2005), 13-48. Pollak, J. (2007), ‘Contested meanings or representation’, in CEP, volume 5, 87-103. Popper, K. (1945), The Open Society and its Enemies. London: Routledge and Kegan Paul. Priestly, J. (2008), Six Battles that Shaped Europe’s Parliament. London: Harper. Priestly, J. (2010), ‘European political parties: the missing link’, in Notre Dame, October issue. Prins, C. (2008), Argued Consensus or Bargained Compromise. Groningen: University (Ph.D.). Proksch, S. and J. Slapin (2010), ‘Parliamentary questions and oversight in the EU’, in EJPR, volume 50, 1, 53-79. Puetter, U. (2012), ‘Deliberative intergovernmentalism on economic governance’ in JEPP, vol. 19, 2, 161-178. Quitkat, C. (2009), ‘Trade associations in French-German comparison’, in McGrath (2009), 124159. Quitkat, C. (2011), ‘The Commission’s online consultations’, in JCMS, volume 49, 3, 653-674. Raffone, P., editor (2006), Italian Lobbies in Brussels. Brussels: CIPI. Randall, H. (1996), A Business Guide to Lobbying in the EU. London: Cartermill. Rasmussen, A. (2008), ‘The EU Conciliation Committee’, in EUP, volume 9, 1, 87-113. Rasmussen, A. (2011), ‘Early conclusion in bicameral bargaining’, in EUP, volume 12, 1, 41-64. Raunio, T. and T. Tiilikainen (2003), Finland in the EU. London: Frank Cass. Raunio, T. and M. Wiberg (2000), ‘Does Support lead to Ignorance?’, in Acta Politica, volume 35, 2, 146-168. Reese, Th. (1996), Inside the Vatican. Cambridge: Harvard University Press. Richardson, J., editor (2001), European Union: Power and policy-making. London: Routledge. Richardson, J., editor (2005), European Union: Power and policy-making. London: Routledge, 3rd edition. Ringland, G. (1998), Scenario Planning. New York: Wiley. Rittberger, B and A. Wonka, editors (2011), ‘Agency governance in the EU’, special issue JEPP, volume 18, 6, 1350-1763. Rochefort, D. and R. Cobb, editors (1994), The Politics of Problem Definition. Kansas: Kansas University Press. Rosamond, B. (2000), Theories of European Integration. London: Macmillan. Rosenthal, G. (1975), The Men Behind the Decisions. Lexington: Lexington Books. Rowe, C. (2011), Regional representations in the EU. Basingstoke: Palgrave. Sabel, S. and J. Zeitlin, editors (2010), Experimentalist governance in EU. Oxford: University Press.

377

References

Sandholtz, W. and A. Sweet, editors, (1998), European Integration and Supranational Governance. Oxford: Oxford University Press. Saurugger, S., editor (2008), ‘Interest groups and democracy in EU’, special issue WEP, volume 31, 6, 1274-1291. Saurugger, S. and F. Mérand, editors (2010), ‘Mainstreaming sociology in EU studies’, special issue CEP, volume 8, 1, 1-159. Savage, G. and others (1991), ‘Strategies for assessing and managing organizational stakeholders’, in Academy of Management Executive, volume 5, 2, 61-75. Schaefer, G. (1996), ‘Committees in the EC Process’, in Pedler and Schaefer (1996), 3-24. Schaefer, G. and others (2000), ‘The Experience of Member State Officials in EU Committees’, in Eipascope, 2000, 3, 29-35. Scheuer, A. (2005), How Europeans See Europe. Amsterdam: Amsterdam University Press (Ph.D.). Schimmelfennig, F. and U. Sedelmeier, editors (2005), The Europeanization of Central and Eastern Europe. Ithaca: Cornell University Press. Schlozman, K. and J. Tierney (1986), Organized Interests and American Democracy. New York: Harper and Row. Schmidt, V. (2006), Democracy in Europe. Oxford: Oxford University Press. Schmitter, Ph. (1996), ‘Imagining the Future of the Euro-Polity with the Help of New Concepts’, in: Marks and others (1996), 121-150. Schmitter, Ph. and W. Streeck (1999), The Organization of Business Interests. Köln: Max Planck Institute (mimeo). Schneider, G. and S. Baltz (2005), ‘Domesticated Bureaucrats’, in Acta Politica, volume 40, 1, 1-27. Schneider, J. (2007), Lobbying and Interest Representation. Brno: Masaryk University Press. Sciarini, P. and others (2004), ‘How Europe hits home: Swiss evidence’, in: JEPP, volume 11, 3, 353-375. Scott, A. and M. Hunt (1965), Congress and Lobbies. Chapel Hill: North Carolina University Press. Scully, R. and R. Jones, editors (2010), Europe, Regions and European Regionalism. Basingstoke: Palgrave. Sepos, A. (2008), Europeanization of Cyprus. Basingstoke: Palgrave. Settembri, P and C. Neuhold (2009), ‘Achieving consensus through EP committees’, in JCMS, volume 47, 1, 127-151. Shaw, J. (1995), EU Legal Studies in Crises? Florence: EUI (RSC 95/23). Sietses, H. (2000), The Fragmented Interest Representation of Big Dutch Business in the EU. Rotterdam: University (mimeo Faculty of Management). Simon, H. (1979), Models of Thought. New Haven: Yale University Press. Simonis, C. (2009), Investing in democracy: engaging citizens in collaborative democracy. Washington: Brookings. Sissenich, B. (2010), ‘Weak states, weak societies’, in Börzel (2010), 11-40. Smismans, S. (2004), Functional Participation in Social Regulation. Oxford: Oxford University Press. Smith, A. (1776), The Wealth of Nations. New York: Modern Library edition (1937). Smith, M. (2012), ‘Developing administrative principles in the EU’, in ELJ, vol. 18, 2, 269-288. Snyder, F., editor (2012), ‘Financial market regulation and economic governance’, special issue ELJ, vol. 18, 1, 1-161. Spence, D., editor (2006), The European Commission. London: Harper (3rd edition). Spencer, T. and C. McGrath, editors (2008), The Future of Public Trust. London: Dods. Spinelli, A. (1965), The Eurocrats. Baltimore: Hopkins University Press. Stacey, J. (2010), Integrating Europe: informal politics and institutional change. Oxford: University Press.

378

References

Stakeholder.EU (annual directory for Brussels). Berlin: Lexxion (www.lexxion.eu) Steinmetz, R. and A. Wivel, editors (2010), Small states in Europe. Farnham: Ashgate. Steunenberg, B. and J. Thomassen, editors (2002), The European Parliament. Oxford: Rowman and Littleffield. Stout, H. (2005), The EU Papacy. Maastricht: Metajuridica. Sun Tze: see Griffiths (1963). Suvarierol, S. (2007), Beyond the Myth of Nationality. Delft: Eburon (Ph.D.). Suvarierol, S. (2009), ‘Networking in Brussels’, in JCMS, volume 47, 2, 411-435. Suvarierol, S. (2011), ‘Everyday cosmopolitanism in the Commission’, in JEPP, volume 18, 2, 181200. Sweet, A. and others, editors (2001), The Institutionalization of Europe. Oxford: Oxford University Press. Sweet, A. and T. Brunell (2012), ‘The ECJ, state non-compliance and the politics of override’, in APSR, vol. 106, 1, 204-223. Swenden, W. and others, editors (2006), ‘The Politics of Belgium’, in special issue WEP, volume 29, 5. Szajkowska, A. (2010), ‘The impact of the definition of precaution in EU Food Law’, in CMLR, volume 47, 1, 173-196. Szczerbiak, A. (2011), Poland within the EU. London: Routledge. Tallberg, J. (2007), Bargaining Power in the European Council. Stockholm: Sieps. Tanasescu, I. (2009), The European Commission and Interest Groups. Brussels: VUB. Tatham, M. (2010), ‘Territorial state-bypassing in EU interest representation’, in JEPP, volume 17, 1, 76-99. Taylor, P. (1996), The European Union in the 1990s. Oxford: Oxford University Press. Telò, M., editor (2001), European Union and New Regionalism. Aldershot: Ashgate. Thomas, C., editor (2004), Research Guide to US and International Interest Groups. Westport: Praeger. Thomassen, J., editor (2009), EU Legitimacy after Enlargement. Oxford: University Press. Thompson, J.D. (1967), Organizations in Action. New York: McGraw Hill. Thomson, R. and others (2012), ‘New dataset on EU decision-making’, in JEPP, vol. 19, 4, 604-622 (comments: 623-634). Thurner, P. and M. Binder (2009),‘EU transgovernmental networks’, in EJPR, volume 48, 1, 80-106. Timmins, G. and others (2007), Politics of the New Germany. London: Routledge. Töller, A. and H. Hofmann (2000), ‘Democracy and the Reform of Comitology’, in Andenas and Türk (2000), 25-50. Töller, A. (2010), ‘Measuring and comparing the Europeanization of national legislation’, in JCMS, volume 48, 2, 417-444. Trondal, J. (2004), ‘Resocializing Civil Servants: The Transformative Powers of EU Institutions’, in Acta Politica, volume 39, 1, 4-30. Trondal, J. (2008), ‘Balancing roles of representation in the Commission’, in Acta Politica, volume 43, 4, 429-452. Tsebelis, G. and S. Proksch (2007), ‘The Art of Political Manipulation in the Convention’, in JCMS, volume 45, 1, 157-186. Urban, S. and S. Vendemini (1992), European Strategic Alliances. Oxford: Blackwell. US/ Congress (1977), Senators: Offices, Ethics and Pressures. Washington: GPO. US/ GAO (1999), Federal Lobbying: Differences in Lobbying Definitions and their Impact. Washington: General Accounting Office.

379

References

Van Biezen, I. and others (2012), ‘The decline of party-membership in contemporary Europe’, in EJPR, vol. 51, 1, 24-56. Van den Berg, C. (2011), Transforming national bureaucracies for Europe. Leiden: University (Ph. D.). Van den Polder, R. (1994), ‘Lobbying for the European Airline Industry’, in Pedler and Van Schendelen (1994), 103-119. Van der Heijden, K. (1996), Scenarios. New York: Wiley. Van der Heijden, H. (2006), ‘Multi-level environmentalism and the EU’, in International Journal of Urban and Regional Research, volume 30, 1, 23-37. Van der Voort, W. (1997), In search of a role: The ESC in EU Decision-making. Utrecht University (Ph.D. thesis). Van Middelaar, L. (2012), The Passage to Europe. New Haven: Yale University Press. Van Noije, L. (2007), The democratic deficit closer to home. Amsterdam: University (Ph.D.). Van Rens, P. (1994), ‘Dutch Trade Union Federation and the EU Works Council’, in Pedler and Van Schendelen (1994), 283-300. Van Schendelen, M. (1984), ‘The European Parliament: Political influence in more than legal powers’, in Journal of European Integration, volume 8, 1, 59-76. Van Schendelen, M. and R. Jackson, editors (1987), The Politicization of Business in Western Europe. London: Croom Helm/ Routledge. Van Schendelen, M. (1990), ‘Business and Government Relations in Europe’, in European Affairs, volume 4, 2, 81-87. Van Schendelen, M., editor (1993), National Public and Private EC Lobbying. Aldershot: Dartmouth. Van Schendelen, M. (1996), ‘The Council decides: does the Council decide?’, in JCMS, volume 34, 4, 531-548 (erratum volume 35, 1, 171). Van Schendelen, R., editor (1998), EU Committees as Influential Policymakers. Aldershot: Ashgate. Van Schendelen, R. (2002), ‘At issue with EU’, in Dutch Crossing, volume 26, 1, 27-42. Van Schendelen, R. and R. Scully, editors (2003), The Unseen Hand: Unelected EU Legislators. London: Frank Cass. Van Schendelen, R. (2003), ‘Insourced Experts’, in Van Schendelen and Scully (2003), 27-39. Van Schendelen, R. (2008), ‘Brussels as the Place for Lobbying the EU’, in De Groof (2008), 197214. Van Spanje, J. and C. de Vreese (2011), ‘Motivations underlying the Eurosceptic vote in the 2009 EP elections’, in EUP, volume 12, 3, 405-429. Van Thiel, S. (2000), Quangocratization: Trends, Causes and Consequences. Utrecht: ICS (Ph.D.). Venables, T. (2007), Tips for the Would-be European Lobbyist. Brussels: ECAS. Verhoest, K. and others, editors (2012), Government agencies. Basingstoke: Palgrave. Verwey, W. (1994), ‘HDTV and Philips’, in Pedler and Van Schendelen (1994), 23-40. Vos, E. (1999), Institutional Frameworks of Community Health and Safety Regulation: Committees, Agencies and Private Bodies. Oxford: Hart. Vos, E. (2009), 50 years of integration, 45 years of comitology. Maastricht: Law Faculty. Vote Watch (2012), Agreeing to disagree. London: www.votewatch.eu. Waddell, H. (1952), The Wandering Scholars. London: Constable. Wallace, W., editor (1990), The Dynamics of European Integration. London: Pinter. Wallace, H. and A. Young, editors (1997), Participation and Policy-making in the European Union. Oxford: Clarendon. Wallace, H. (1997), ‘Introduction’, in Wallace and Young (1997), 1-16.

380

References

Wallace, H. and others, editors (2000), Policymaking in the EU. Oxford: Oxford University Press (4th edition). Wallace, H. (2007), Adapting to Enlargement: Institutional practises since 2004. Brussels: Tepsa. Wallace, H. and others, editors (2010), Policymaking in the EU, Oxford: Oxford University Press (6th edition). Wallace, C. and K. Stromsnes, editors (2008), ‘European identities’, special issue PEPS, volume 9, 4, 378-513. Warleigh, A. (2001), ‘Europeanising civil society’, in JCMS, volume 29, 619-639. Warleigh, A. and J. Fairbrass, editors (2002), Influence and Interests in the EU. London: Taylor and Francis. Warleigh, A. (2003), ‘Informal Governance: Improving EU Democracy’, in Christiansen and Piattoni (2003), 22-35. Warleigh, A. and L. Van Langenhove (2010), ‘Rethinking EU studies: the contribution of comparative regionalism’, special issue JEI, vol. 32, 6, 541-657. Warntjen, A, and A. Wonka, editors (2004), Governance in Europe. Baden-Baden: Nomos. Warntjen, A. (2007), ‘Steering the Union: The impact of the EU Presidency on legislative activity’, in JCMS, volume 45, 5, 1135-1157. Warntjen, A. (2011), ‘Measuring salience in EU legislative politics’, in EUP, vol. 13, 1, 168-182. Weiler, J. (1995), European Democracy and its Critique. Florence: EUI (RSC 95/11). Werts, J. (2008), The European Council. London: Harper. Wessels, B. (1999), ‘European Parliament and Interest Groups’, in Katz and Wessels (1999), 105128. Wessels, W. and others, editors (2003), Fifteen into One: The EU and its Member States. Manchester: Manchester University Press. Westlake, M. and D. Galloway (2004), The Council of the EU. London: Harper. Westlake, M. (2009), ‘The European Social and Economic Committee’, in Coen and Richardson (2009), 128-142. Whitaker, R. (2011), The EP’s committees. London: Routledge. Wiener, A. and T. Diez, editors (2009), European Integration Theory. Oxford: Oxford University Press. Wilhelm, A. (2000), Democracy in the Digital Age. London: Routledge. Williamson, D. (2006), ‘The conciliation procedure’, in JLS, volume 12, 1, 1-7. Winand, P. (1998), ‘The US Mission to the EU in Brussels’, in Claeys and others (1998), 373-405. Winzen, T. (2011), ‘Technical or political: officials in the EP committees’, in JLS, volume 17, 1, 2744. Wittlinger, R. (2010), German national identity in the 21th century. Basingstoke: Palgrave. Wonka, A. and others (2010), ‘Measuring the size and scope of the EU interest group population’, in EUP, volume 11, 3, 463-476. Wouters, J. and others, editors (2012), The EU and multilateral governance. Basingstoke: Palgrave. Wright, J. (2004), The Jesuits. London: Harper Collins. Yordanova, N. (2009), ‘The rationale behind EP committee assignments’, in EUP, volume 10, 2, 253-280. Zweifel, Th. (2002), ‘Who is without sin cast the first stone: The EU’s democratic deficit in comparison’, in JEPP, volume 9, 5, 812-840.

381

Index

References are to the beginnings of the key passages only. A Aarhus Convention case 135 Academics 16, 64, 113, 282, 354 Advisory committees 81, 91 Agencies 79, 109 Air Transport case 47 Antichambre 58, 150 A-point 104 Arena analysis, see Chapter 4 Boundaries 176 Definition 165 Home front, see Chapter 5 Issues 167, 173 Limits 301 Management 190, 251, 301 Meteorology 187 Monitoring 179 Predecessors 166 Quick scan 184 Stakeholders 169, 171 Time 175 Window-in/out, see there Austria 157 B Banana Trade case 126, 228 Belgium 155 Better tactics 192 BINGO 50 Blue paper 93 BONGO 50 Boundary analysis 176 Management 191 B-point 104 Brent Spar case 56, 213, 215 Britain, see United Kingdom Brussels method 95, 330 Brussels PA office 68, 234 Bulgaria 158

C Central PA Desk 234 Channel limits 296 Chef-de-dossier 89 Citizenship 323, 335 Code of conduct 311 Codecision, see Legislation OLP Codification 75 Cohesion on PA 207, 292 Collective action 142 College (Commissioners) 74, 78 Comitology 74, 81, 94 Commission, see European Commission Committee of the Regions COR 83, 102 Conciliation procedure 82 Constanzo Court ruling 102, 130 Consultancy 236, 253 Consultation, see Public Consultation Consultation procedure, see Legislation SLP COREPER 78, 103 Council of Ministers 77, 103 Crisis politics 19, 106, 322 Work groups 78, 104 Court of Auditors 109 Court of Justice 83, 108 Croatia 142 Cyber-activism 273 Cyber-democracy 348 Cyprus 158 Czech Republic 158 D Darwin’s Law (EU variant) 162 Decentralised governments 238 Decker-Kohl case 51 Decree 74, 80 Democracy, see Chapter 8 As criterion 325 By activation 348 By establishment 348 By professionalization 348

383

Index

Cyber-democracy 348 Deficit or surplus 333 Improved by PA 348 Notions of 327 Open entry 353 PA impacts on 337 Self-correcting lobby-democracy 351 Denmark 156 Diplomacy 59, 181, 279 Direct or indirect actions 262 Directive 74, 80 Directorate-General/ DG 78 Dogmas 288, 318 Domestic politics 18, 104, 114, 319 Dossier 165 Dossier life cycle 167, 251, 275 Down-framing, see Framing Downstream 45 Duplica method 183

Political groups 82, 98 Rapporteur system 82, 99 European Transparency Initiative ETI 67, 311, 344 European Union EU, see Chapter 2 Actors to approach 121 Characteristics 110 Country profiles 148 Crucial variables, see Chapter 3 Eu-ization 44 EU-method 35 Factors to use 124 Flesh-and-blood 72, 87 Political miracle, see there Skeleton 72, 73, 77 Vectors to create 128 Examination committees 81, 91 Expert groups 92, 94 Expertise 16, 54, 63, 211, 233, 242

E Economic and Social Committee ESC 83, 100 Emotions 214, 287 END 79, 90 Enhanced co-operation 85 Estonia 158 European Citizens Initiative ECI 314 EuroFed 143 European Commission 78, 88 As a lobby group 95 Benchmarked 241 European Convention case 135 European Council, see Council European Court of Justice, see Court of Justice European Food Safety Regulation case 135 Europeanization, see Chapter 1 Carriers of 45 Challenges 52 Creeping 51, 86 Defined 44 Downstream 45 Fuller story 49 Of member countries 43 Rolling 52 Upstream 45 Vectors of 45 European Parliament EP 81, 98 Committees 82, 98 Inter-groups 99

F Federalist Papers 354 Fieldwork of PA, see Chapter 6 Finland 156 Fonctionnaire 79 Formal or informal actions 261 Fourchette 79 Framing 55, 249 France 38, 149

384

G Germany 39, 151 GINGO 50 GMO Food Arena case 197 GONGO 50 Greece 156 Green paper 94 H Habilitation 89 Hard law 41, 334 Hierarchy of norms 74 High Definition TV case HDTV 177, 265, 287 High politics 92 Home front, see Chapter 5 Cohesion 207, 292 Defined 205 Organizational assets 206, 295 Reputation 215

Index

Resources and skills 211, 220, 293 Useful knowledge 210 Hume, David 229 Hungary 158 I Inattention limits 305 Influence challenges 52 Defined 87 Traditional techniques 55 Information 22, 59, 61, 177 Memory 149, 179, 186, 194 Insourced democracy 330 Insourcing (of experts) 92 Integration, see Chapter 1 Differentiated 86 EU method 35 Negative or positive 85 Traditional methods 36 Intelligence 22, 61, 97, 141, 187 Intelligent laziness 62, 97, 222 Interest group defined 53 General interest 318 Intergovernmentalism 73 Inter-group 99 Interservice 79 Ireland 156 Issue analysis 167, 173, 249 Defined 54, 167 Life cycle 167 Management 190, 257 Italy 152 L Latvia 158 Laying Hen case 174 Legislation: codecision (OLP) 75, 82 Consultation (SLP) 75, 82 Delegated acts 75, 81, 91 Discretionary 91 Implementing acts 75, 81, 91 Legislative acts 75, 80, 90 Non-legislative acts 75, 81, 91 OLP (codecision) 75, 82 Pace, speed of 112 Primary 73 Secondary 73 SLP (consultation) 75, 82 Supplementary 91

Lisbon Treaty 17, 75 Lithuania 158 Lobby, see also Public Affairs Budget 212 By body 264 By carrot and stick 124, 246, 251 By melody, humour 270 By reflex 252 By silence 268 Code of conduct 311 Collective action 142 Control of 62, 142, 254 Cross-sectoral 21, 48, 113 Country profiles 148 Cyber-lobbying 273 Deadline lobby 277 Definition 57, 245, 311 Democracy, see Chapter 8 Effectiveness 22, 62, 161, 203, 228 Exemption lobby 154 Female/ male 281 Individual 142 Limits of, see Chapter 7 Locations 271 Meta-lobbying 258 Multi-level 21, 141, 146, 214, 234 Offensive lobbying 258 Outside lobbying 267, 313 Over-lobbying 253, 307 Quality of lobby 63, 68 Quantity of groups 67 Rules of thumb 253 Styles 252, 259 Success measures 119 Surplus of choices 246 Timing 274 Limits of PA, see Chapter 7 Logic of influence 142 Long-list 61, 221, 223 Loss-compensation 201, 225 Low politics 92 Luxembourg 155 M Machiavelli, Niccolò 15, 63, 118, 134, 278, 325 Malta 158 Marcus Aurelius 219 Marie Antoinette 269 Marketing, see Political Marketing

385

Index

Mass media 58, 72, 180, 267, 298, 331 Member countries 43 Modification 75 Multinational PA Model 231 Myths 290 N National co-ordination 136, 240 National government 240, 315 National ministry 239 National parliament 317, 319 Navigation 139 Negative selection (long-list) 223 Netherlands 155 Neutralisation limits 307 NGO defined 40, 50, 238 Nomenclature 75 Norway 149 Nuisance value 169 O OECD 21, 124, 264 Online consultation OC 93, 346 Open Co-ordination Method OCM 38 Optimization 182, 208, 211, 243, 254, 328 Outside groups 313 P Paradox of Brussels method 96 Cross-pressures 307 Democracy notions 328 Embarras du choix 190 Europeanization 86 Neutralization 307 Openness 339, 340 PA analysis 283 Power and influence 109 Triple P 134 Paris- Berlin axis 291 Parliament, see European Parliament Participation 42 Pension Funds case 213 Permanent Representative PR 78 Pillars structure 75 Poland 158 Policy formation 84 Modes of 85 Political engineering 131, 265, 350 Political marketing 116, 260, 269

386

Political miracle 38, 72, 116 Polyphonic orchestra 265 Portugal 156 Power balance in EU 110 Praxismoral 310 Pressure group 54 Primary law 73 Professionalism 21, 64, 161, 278, 321, 349, 361 Public Affairs, see also Lobbying Academics, see there Arena management, see Chapter 4 Better tactics 192 Budget 212 Central PA Desk 234 Consultancy 236, 253 Definition 15, 60 Effectiveness 22, 62, 161, 203, 228 Fieldwork, see Chapter 6 Growth of 64 Home front management, see Chapter 5 Ideal profile of PA manager 278 Limits of, see Chapter 7 Multinational PA Model 231 Navigation 139 PA expertise 16, 54, 63, 211, 233, 242 Planning 276 Practitioner 16, 23, 66, 355 Problem analysis 217 Strategy 216, 228 Structural trends 136 Tailor-made 147 Trust 236 Useful knowledge 210 Public consultations 93 Public-private relations 40 Public relations and PA 268 Q Qualified Majority Voting QMV 73, 104 Quick arena scan 184, 221 R Radio Frequency Identification Device case 178 Rapporteur system 82, 99 Reframing, see Framing Regulation 74, 80 Replica method 183 Reputation 215, 309

Index

Resources and skills 211, 220, 293 Romania 158 Route mapping 139 Ruler analysis 166 S Satiation limits 306 SCARE 285 Scenario management 196 Secondary law 73 Self-reliance 136 Shared-list 62, 221 Shell 56, 196, 311 Short-list 62, 221 Siemens method 209, 234 Silent or noisy action 266 Slovakia 158 Slovenia 158 Small-and-medium-sized enterprise SME 238 Smith, Adam 117 Social Dialogue 101 Soft law 41 Spain 153 Spill-over 86 Stakeholders, see Chapters 4 and 6 Analysis 169, 171 Defined 169 Management 190, 248, 255 Primary 174 Strategy 216, 228 Subsidiarity 50, 77, 85, 319 Subsidiarisation (of EuroFeds) 146 Sun Tze 56, 63 Supplying or demanding actions 259 Supranationalism 74 Sweden 156 Switzerland 149

T Talleyrand-Périgord, Charles de 355 Target-setting 225 Textile Trade Liberalisation case 54, 129, 250 Time analysis 169, 175 Management 191, 274 Tobacco Ban Directive case 135, 346 Transparency 67, 311, 316 Trialogue, Triangle 82 Triple P 62, 132 Troika 19, 86, 103 Turkey 68, 149 U Unaccompanied Minors case 135 Unbundling Rail Freight case 228 Unilever 135,176, 200, 214, 310 United Kingdom 39, 150 Unorthodox actions 58 Upframing, see Framing Upstream 45 US-EU comparison 113, 267 V Variable geometry 86 Vectors of europeanization 45 Of influence 128 VLEVA 156, 239 W Weber, Max 278 White paper 94 Window-in 62, 190 Window-out 61, 169 Work floors 88

387