“The Europeanization of domestic legislatures. The empirical implications of the Delors' Myth in nine countries”
Sylvain Brouard, Olivier Costa and Thomas König (ed.) New York, Springer, 2012, 244 p.
MANUSCRIPT Table Contents
1. Delors’ Myth: The scope and impact of the Europeanization of law production 3 2. EU Legislative Activities and Domestic Politics 26 3. Measuring the ‘Europeanization’ of Austrian Law-Making: Legal and Contextual Factors 46 4. The Minor Impact of EU on Legislation in Finland 5. Are French laws written in Brussels?
71
95
6. Does Europeanization change Executive-Parliament relations: Executive Dominance and Parliamentary Responses in Germany 121 7. Leading governments and unwilling legislators. The European Union and the Italian Law making (1987-2006) 144 8. The Europeanization of domestic legislation in Luxembourg
175
9. Myths and Milestones: The Europeanization of the Legislative Agenda in The Netherlands 206 10. The Europeanization of law-making activities in Spain 238 11. So close, yet so far? The EU footprint in Swiss legislative production 273 12. Going beyond: causes of Europeanization
311
ACKNOWLEDGMENTS The “Delors’ myth project” is the result of four workshops organized in Bordeaux (November 2008), The Hague (June 2009), Barcelona (January 2010) and Mannheim (May 2010). The editors would like to thank Sciences Po Bordeaux (Centre Emile Durkheim), the Montesquieu Institute, the University of Barcelona and the Mannheim Center for European Social Research (MZES) at the University of Mannheim for their support. The editors would also like to acknowledge the Comparative Agendas Project, its initiators and its members. This research would have been much more difficult to achieve without this network as well as its common framework and focus. The research on the Austrian case was funded by the Austrian ‘New Orientations for Democracy in Europe’ (NODE) research programme. The authors are grateful to Ilse König and Martina Hartl of the NODE programme for their support, to their student research assistants in Mannheim and Vienna for devoted work. The French chapter is a joint output of the project “LEGIPAR: Parliamentary legitimization and democratic government in France and in the European Union” and of the project “AGENDA: the political agendas of the Fifth Republic”, both funded by the French Agence Nationale de la Recherche. The German chapter is grateful for support of the German National Science Foundation (1142 “Institutional Design of Federal Systems: Theory and Data”) and the Mannheim Center for European Social Research (MZES). The Italian team would like to acknowledge the financial assistance of the Italian Ministry for Research and Higher Education, Prin 2007 prot. scrwt4 “Legislative process and policy arenas. Games, vetoes and networks in the age of the Italian political alternation”. The Luxembourgish team would like to express its sincere gratitude to Daniel Andrich (Director, Service Central de Législation), Nicolas Henckes (Director, Legitech), Ben Fayot (Chairman of the Committee on Foreign and European Affairs, Defence, Cooperation and Immigration in the Luxemburgish Parliament), and Rita Brors (Officer, Chamber of Deputies) for their support and cooperation, which enabled them to accomplish this research. The Spanish chapter is an output of the project “The Politics of attention: West European politics in times of change: the case of Spain” (EUROCORES programme), European Science Foundation (ESF), and the Agencia de Gestió d’Ajuts Universitaris i de Recerca (AGAUR), Generalitat de Catalunya (SGR 536). The Swiss chapter is an output of the project “Agenda Setting in Switzerland” funded by the Swiss National Science Foundation (ref. 105511-119245/1), and part of the EUROCORES project “The Politics of Attention: West European Politics in Times of Change”, funded by the European Science Foundation.
1.
Delors’ Myth: The scope and impact of the Europeanization of law production
Sylvain Brouard, Olivier Costa and Thomas König
Since the beginning of the 1990s numerous theoretical and normative debates on European integration and the ‘democratic’ distribution of power among the European Commission, the Council of Ministers and the European Parliament have entertained scholars, politicians and opinion leaders alike. The ‘democracy deficit’ concept stems from the hypothesis that EU legislative activities have strengthened the power of governmental actors from the Council of Ministers at the expense of parliaments. While Moravcsik (2004, 2008) argues that these governmental agents are democratically elected agents of their national constituencies, the concerns about a democratic deficit have been reinforced by the progressive extension of qualified majority voting in the Council of Ministers since the end of the 1980s: theoretically, the possible exclusion of some governmental agents and the lack of transparency of Council deliberations made their effective control by their domestic parliamentary principals extremely difficult, as members of parliament even lack reliable information on the voting behavior of their governmental agents in the Council (Scharpf 1993; Hix 1999; 2005, Majone 1998; Moravcsik 1994, 2002; Lord 2004 ; Bartolini2005 ; Follesdal and Hix 2006; Siedentop 2001; Jachtenfuchs 2001; Rittberger and Schimmelfennig 2005; Follesdal and Hix 2006; Olsen 2007; Magnette and Papadopoulos 2008; König 2008). Without going into detail on the different normative views about the EU’s distribution of power, the controversies about the democratic deficit commonly state a weakness of parliamentary representation in the EU, in a system dominated by representatives of the executive and technocrats, and marked by the supremacy of law over politics. As a response, the participation of the European Parliament has steadily increased by treaty revisions and the co-decision procedure, which established a bicameral legislature between the Council and the European Parliament, has become the standard procedure since the entry into force of the Lisbon Treaty in December 2009. This, however, does not mean that the criticism on the democratic foundations of the EU has stopped. Almost simultaneously with the continuous increase in power of the European Parliament, scholars started to discuss the topic of domestic “deparliamentarization” because the Maastricht Treaty in 1993 –
3
the Treaty on European Union underlined the growing potential of European integration by formally extending qualified majority voting in the Council and transferring sensitive policy matters, including the introduction of the Euro, to EU legislative discretion (Norton 1995; Raunio and Hix 2000). The argument is that policy competencies are formally uploaded to the EU level by treaty revisions, negotiated and then presented as “take-it-or-leave-it” packages to domestic parliaments, who can hardly amend these packages. Thereby, it is assumed that the increased application of qualified majority voting makes it difficult for domestic parliaments to force their governmental agents to make credible ex ante commitments before adopting these packages at the EU level. Furthermore, through the technical nature of these packages, bureaucrats and interest groups rather than directly elected parliamentarians are expected to benefit from this kind of Europeanization. These formal developments by treaty revisions have shifted the focus of the debate on a democratic deficit from the interaction between the European Commission, the Council of Ministers and the European Parliament to the impact of European integration on domestic politics (Kassim et al. 2000; Hix and Goetz 2000; Crean Cowles et al. 2001; Featherstone and Radaelli 2003; Holzhacker 2007; König 2007). Unsurprisingly, the numerous theoretical accounts on European integration and countless interpretations of the power distribution in the EU lead to divergent and controversial conclusions on this point too. For some scholars, the Delors 80% myth of a super-state, that increasingly dominates domestic parliamentary legislatures by the expansion of EU legislative activities, has become reality yet (Herzog and Gerken 2007; Hoppe 2010). Similar to the discussion about the democratic deficit, these conclusions on the level of deparliamentarization are drawn from different theoretical and normative viewpoints on European integration rather than empirical evidence for a changing role of domestic parliaments through European integration in their legislatures. On closer inspection of domestic parliamentary legislatures of the member states, the role of their constitutive elements – the political parties in those parliaments –, the criticism addresses three factors induced by European integration: deprived of a part of their legislative powers by the EU, excluded from EU legislative decision making and subjected to growing normative obligations (Maurer and Wessels 2001; Wiberg 1997). Under this impression, the willingness to strengthen the role of domestic parliaments has, therefore, been expressed by several declarations and protocols of recent treaty revisions. Some domestic parliaments have already reacted to this by reforming their own institutional provisions, in particular in the Scandinavian countries where the parliamentary legislature is involved in the ex ante control of governmental agents when negotiations on Commission proposals start in the Council of Ministers (Bergman 1997; Laursen et Pappas 1995; Maurer and Wessels 2001; Norton 1996; Smith 1996; Raunio 1999; Raunio and Hix 2000; Wiberg 1997). Compared to the early years of European integration, some domestic parliaments have also established committees on European Affairs and special affairs, which strengthen their informational position vis-à-vis their governmental agents.
4
More recently, scholars have pointed to the dualist nature of Europeanization for members of parliament, which is both passive (Europeanization seen as something that simply happens to parliamentarians) and active (Europeanization considered as a process of adaptation or reaction of parliamentary behavior to European integration) (Saalfeld 2005; O’Brennan and Raunio 2007). According to Auel and Benz (2005), members of domestic parliaments are faced with a dilemma: a successful scrutiny can undermine the effectiveness of European governance, whereas permissive parliamentarians cause a legitimacy deficit. Other studies, for instance Zeitlin and Pochet (2005) and Duina and Raunio (2007), concentrate on new instruments such as the Open Method of Coordination (OMC)1 and call for a more nuanced analysis of the deparliamentarization-hypothesis. The OMC empowers governmental agents through executive federalism, and thus – it is feared to contribute to a further marginalization of domestic parliamentarians; but it also provides domestic legislators with opportunities that the traditional “Community method” cannot offer. A major deficit of these discussions and criticisms on the democratic deficit and deparliamentarization is the lack of solid data, which make a more robust evaluation of their empirical implications on the power distribution in the EU and the role of domestic parliaments possible. The conclusions on the democratic deficit- and deparliamentarization-hypothesis are almost exclusively drawn from the formal developments by treaty revisions, which have only increased the potential impact of Europeanization. The question however is whether and to what extent this potential has been used empirically – firstly as a necessary condition by legislative activities at the EU level, which produced Commission proposals with and without participation of the European Parliament, and secondly as a sufficient condition by the number of cases which suggest some impact of these activities for the domestic level. The empirical discourse on the evolution of these effects is still full of references to the supremacy of European law and to the case-wise restriction of the freedom of action for some (domestic) actors. Regarding the impact of EU legislative activities on domestic politics, a few empirical studies investigated the case-wise effect of some specific Commission proposals in distinct policy areas, such as the Europeanization by environmental (Knill 1998; Haverland 2000), transportation (Héritier et al. 2001) and social standards of the EU (Falkner 2001, 2005; Graziano 2005). In sum, there is little evidence for a more general cross-sectional and longitudinal effect of Europeanization on domestic parliamentary legislatures of the member states, which means that the most cited authors have no choice other than to refer to normative presuppositions or to quote the former President of the Commission (1985-1995) Jacques Delors. In 1988, Delors declared in front of the European Parliament that “in ten years 80 per cent of the legislation related to economics, maybe also to 1
Introduced at the end of the 1990s, the OMC is an intergovernmental mean of governance of the EU. The OMC privileges soft law mechanisms when the traditional EU instruments (directives, regulations etc.) cannot be used for legal or political reasons. With OMC, European institutions don’t vote on binding norms: the Council of ministers agrees on common objectives (policy goals) to be voluntarily followed by the member states. The Commission plays a role similar to a secretary, which at best supervises the progress made by each member state.
5
taxes and social affairs, will be of Community origin”(Bulletin No-2-367/157, 6 July 1988). This declaration has been repeated on many occasions by Jacques Delors in the following years and has been widely quoted or paraphrased by scholars alike (Hix 1999, 3; Hix and Goetz 2000, 4; Greenwood 2003, 6), reported and distorted by journalists, politicians and officials of the European institutions. Delors’ prediction lies at the heart of a persistent myth, according to which 80% of the activities of domestic parliamentary legislatures consist of a simple transposition of European norms – which is far from the initial declaration and by far more alarming. The idea that norms are almost exclusively written in Brussels comes up as soon as public debates and criticism of the EU begin, whether on the occasion of the ratification of a new treaty revision, European elections, EU legislative activities etc.. To some surprise, both opponents to European integration and representatives of EU institutions bring forward this mythical reference. Most recently, the reference that 80% of the activity of the domestic parliamentary legislatures consists of a simple transposition of European norms has largely been used by the candidates of the Eurosceptic lists ‘Libertas’ during the campaign for the European elections in June 2009, campaigning in ten member states, in order to denunciate an excessive European super-state. On the other side, the press service of the European Parliament insisted on the same number in order to convince the citizens of the importance of European elections, implying that many political decisions are taken in Brussels and no longer in the national capitals. But is there any empirical evidence for this 80% myth?
Understanding Europeanization The literature on Europeanization, which has become more popular in European studies since the beginning of the 1990s, attempts to provide empirical evidence for a broad set of Europeanization issues (Graziano & Vink 2007). Some studies examine the EU’s influence on member states (so-called “downloading” or “top-down” Europeanization), others the influence of member states on EU policy making (“uploading” or “bottom-up” Europeanization2), while others concentrate on the transfer of policy from one country to several others without direct EU involvement.3 This broad research program has led Robert Ladrech to adopt an encompassing definition of Europeanization as “an incremental process reorienting the direction and shape of politics to the degree that EC political and economic dynamics become part of the organizational logic of national politics and policy-making”4. Europeanization studies thus attempt to analyze the various characteristics of how EU level dynamics are gradually taken into account at the domestic level. Thereby, the top-down approach has been criticized for
2
Simon Bulmer and Claudio Radaelli, ‘The Europeanization of National Policy’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, p. 340 3 Claudio Radaelli, op. cit, p. 27 4 Robert Ladrech, ‘Europeanization of Domestic Politics and Institutions: The Case of France’, in: Journal of Common Market Studies, vol. 31, n°1, 1994, p. 69
6
reducing the role of the member states, their institutions and representatives to a purely passive one. It is questionable whether and to what extent the actors of the domestic level, in particular those in the domestic parliamentary legislatures, are only “takers” of European norms, models or pressures, or whether and to what extent they also actively “shape” those instruments. This mechanism can only become evident when relaxing the unitary actor assumption on member states which represent their preferences at the different stages of the Europeanization process. For example intergovernmentalism, which focuses on governmental actors, can go one step further by relaxing the unitary actor assumption and take a close look on the mechanism design on how these actors are challenged in domestic parliamentary legislatures, in particular from their coalition partners when “Europeanized” laws are introduced and adopted. In this vein, Europeanization can be understood as a useful means for increasing ministerial power when the coalition partner can hardly scrutinize ministerial drift in the parliamentary legislature. However, rather than deepening our understanding of this mechanism, Europeanization studies have widened their field of investigation: they are no longer limited to the EU’s impact on politics and policymaking, but also analyze effects on processes, policies and institutions (policy, polity and politics).5 On the one hand, the global definition of Claudio Radaelli attempts to incorporate all aspects of Europeanization: “Europeanization refers to processes of (a) construction, (b) diffusion, and (c) institutionalization of formal and informal rules, procedures, policy paradigms, styles, ‘ways of doing things’, and shared beliefs and norms which are first defined and consolidated in the making of EU public policy and politics and then incorporated in the logic of domestic discourse, identities, political structures, and public policies.”6 Such a definition encompasses the different tendencies in Europeanization literature by referring to both construction and diffusion, and to all subjects of influence. It promulgates the view that Europeanization should be understood in the broadest possible sense, whereby institutions can also refer to non-material concepts such as norms, ideas and beliefs. On the other hand, a more specific perspective on the actors involved is warranted for evaluating the empirical implications of the various factors of Europeanization, in particular those which are considered to affect the parliamentary legislatures in the member states. This specific perspective draws the attention to the causal mechanism where the formal developments by treaty revisions only establish a necessary but not a sufficient condition for the Europeanization of domestic parliamentary legislatures. Treaty revisions require the consent of both governmental and parliamentary actors, in particular in the ratification process of treaty revisions (König and Hug 2000, Hug and König 2002). Until now, these treaties provide the European Commission with the sole right to draft proposals. Hence, a second necessary condition is determined by the European Commission’s initiative to draft proposals, which are 5
Tanja Börzel and Thomas Risse, ‘Conceptualizing the Domestic Impact of Europe’, in: K. Featherstone and C. Radaelli (eds.), The Politics of Europeanization, Oxford University Press, Oxford, 2003, p. 60 6 Claudio Radaelli, op. cit., p. 30
7
binding and enforceable for the member states. In some areas, such as agricultural and common market affairs, the European Commission has been more active over time. In collaboration with the governmental agents in the Council, and to an increasing extent with the European Parliament, these proposals were almost always adopted. Some scholars have therefore spoken from a culture of consensus (Lewis 2003, Heisenberg 2005). Another explanation is that the Commission attempts to avoid conflict and preselects less controversial policies, which are often only in the Council and European Parliament. The Council itself has established a sophisticated committee system, which promotes consensus-building by facilitating logrolling across proposals (König and Junge 2009). This suggests that the formal development by treaty revisions and the European Commission’s initiative are two necessary conditions for the Europeanization of domestic parliamentary legislatures. They determine the potential for the Europeanization of domestic parliamentary legislatures, in which political parties pursue own interests in order to form governmental coalitions and to become re-elected. This aspect of party interests and competition has received little attention in the literature yet, even in those parts, which devotes special attention to the relationship between political parties and European integration. According to Peter Mair (2007), there are three strands in this literature, the first reflecting the emergence of political parties at the EU level (i.e., Pridham 1975; Pridham and Pridham 1981; Bardi 1994; Hix 1995), the second investigating the activities of political parties in the European Parliament (i.e. Attina 1990; Kreppel and Tsebelis 1999; Hix 2001; Hix et al. 2003), and the third collecting data on the positions of political parties towards Europe (Marks and Steenbergen 2004; Benoit et al. 2005). The few studies which have examined the impact of Europeanization on political parties and party systems come to the conclusion that the developments at the EU level are hollowing out policy competition between political parties at the national level (Ladrech 2005: 333, Mair 2007: 159). According to these findings, Europe has limited political parties in terms of policy space, capacities and policy repertoire. However, these conclusions are based on impressionistic accounts rather than robust empirical evidence. This does not mean that they go into the wrong direction but it is also possible that European integration empirically increases the policy space of political parties by adding another dimension, raises capacities by offering additional strategies of “take-it-or-leave-it” packaging and enriches policy repertoire of shaming and blaming of third parties such as the European Commission or, more generally, the EU. Instead of adding another impression about the impact of Europeanization on domestic parliamentary legislatures, our aim is to contribute to a more robust empirical evaluation of the degree of Europeanization in several member states. We will thus deliberately focus on the empirical analysis of that phenomenon, based on a comprehensive dataset and operational definition which will enable us to examine the “Delors myth” and other widespread hypotheses, such as an increasing level of EU legislative activities and their impact on domestic parliamentary legislatures, and the role of political parties in those legislatures etc.. The examination of such questions has often been limited to the most evident examples of Europeanization by treaties, international agreements of the EU, jurisprudence of the
8
European Court of Justice or the transposition of directives, which domestic parliamentary legislatures of the member states usually transpose into domestic law. For a long time, specialists – mainly lawyers – have concentrated on the transposition problems of implementing EU directives. The official statistics presented by the European Commission’s General Secretary suggest a high rate of successful implementation in every member state, even in the new ones (in May 2007, it was between 96.5% and 99.8%7). However, those results are partly undermined by a gold-platting phenomenon (Falkner and al. 2004). Put differently, even if member states are obliged to transpose EU directives into domestic law, they have notable discretion for transposition and may also fail to comply with EU directives. Political scientists are particularly concerned by the extent of, and reasons for, non-compliance. In addition to many case studies, a growing number of quantitative studies prove the existence of “opposition through the backdoor” – originally theorized by intergovernmentalists – and show that member states occasionally refuse to implement directives when they have failed to assert their interests during the EU decision making process (Thomson et al. 2007, König and Luetgert 2009). It appears that such situations also arise without prior opposition, and that non-compliance is still sometimes due to administrative shortcomings, interpretation problems, and issue linkage. It is worth noting that the transposition of directives is dominated by executive implementation measures, while parliamentary activities in domestic legislatures remain rare and selective in this process (König and Mäder 2008, Franchino and Hoyland 2009). There are, thus, very few studies that address the Europeanization of domestic parliamentary legislatures in a systematic way, with the exception of a recent issue of the Journal Public Administration (2010) and some contributions which are either largely theoretical or focused on a single member state (Töller 2006, 2007, 2008; König and Mäder 2007, König 2009, Raunio and Wiberg, 2010). Concerning the number of norms adopted through the transposition of directives, very little data is available. At the national level, comparative politics scholars and constitutional lawyers emphasize the constraining element by the norms of existing directives for policy makers, and especially for parliaments. However, they have mainly focused on the strategies developed by member states to avoid the pressure of these norms (active Europeanization), and have not collected data regarding the actual percentage of law resulting from directive transposition. Like EU specialists, they seem to take Jacques Delors prediction as given. Jean Maïa, for instance, comes up with the impressive estimation that 60% to 70% of all laws originate in Brussels (Maïa 2005). And even if so, it remains an open question whether and to what extent Europeanization affects parliaments versus governments and administrations – a necessary condition for proving the widespread hypothesis that European integration has weakened parliamentary legislatures and created a “democratic deficit” at the national level. Hence, our understanding of Europeanization is still hampered by a lack of solid empirical material and systematic
7
http://ec.europa.eu/community_law/eulaw. See also: European Commission (2006).
9
studies. We have therefore chosen to take an empirical approach to analyzing the phenomena of the Europeanization of domestic legislatures.
The empirical approach of the Europeanization of law production The analysis of the Europeanization of domestic legislatures is promising but particularly challenging in practice. In the academic field, there is a heated discourse on this question. Scholars recognize that it would be more than useful to dispose of more solid data on the degree of Europeanization of the domestic law production. But, at the same time, as soon as research proceeds in this manner, the usual controversies on the possibilities to capture social phenomena in a quantitative way emerge. Qualitative scholars immediately raise a whole range of concerns about the relevance of “counting” laws that presents such a variety of form, objectives and impact, or, the pitfalls when comparing domestic legislatures that are so different from constitutional, legal and political points of view. One could discuss at length about the right way to identify what a “Europeanized” law is. It is certainly strong assumption on independence and quality when just “counting” laws, but it is a necessary starting point of any empirical analysis of the respective features of the two categories of Europeanized and non-Europeanized laws. Hence, before qualifying laws, we are interested in counting the number of Europeanized and non-Europeanized laws. The main challenge of this study is to establish and apply an operational definition of Europeanization which is, at the same time, operational and not limited to only one aspect of Europeanization. For instance, taking account of only the laws transposing directives or binding acts would limit Europeanization to the ‘downloading’ dimension. On the other hand, a broad definition of Europeanization as a process which globally affects the societies of the member states would lead to the impression that all national laws are more or less linked to European integration, as they are elaborated in a juridical and political framework, which is already highly Europeanized. This would risk to result in a statement that everything is Europeanized and make it impossible to distinguish the impact of European integration from the one of other phenomena such as globalization or modernization. There would thus be no interest for any empirical study at all. In order to contribute to an empirical foundation of the debate on Europeanization we decided to use a reliable sample identification and indicators, which allow to reproduce our findings. Our research design takes the whole sample of national laws into account and identifies Europeanized laws as those statutes that explicitly mention the EU or one of its institutions or instruments in their texts. Ideally, we would have an explicit reference in each national law indicating its origin from Brussels. However, such a reference only exists in Germany. By taking into account all references to the EU or one of its institutions or instruments in their texts, we may capture indirect references and perhaps overestimate the level of Europeanization. For example, if a national law only mentions the European context of policy-making
10
and the need to be competitive, our measure will consider it as Europeanized. The same applies to a national law that makes the EU responsible for a controversial reform (blaming) without relating it explicitly to a European norm or decision. Of course, we excluded laws that mention the EU in order to emphasize that there’s no EU constraint or link – which is, for instance, very often the case in Austria. Conversely, the use of EU as a framework for a national law will be considered as Europeanization, whether the aim is to anticipate a decision of the EU, to adapt policies as a reaction to the EU or to overcome a national handicap or to get an advantage in competition within the EU. All of the participants in this research project agreed that there is no perfect methodology. Still, we believe that nothing is more biased than a study of a phenomenon that, in the end, would refuse any empirical analysis or try to draw general conclusion from a very small number of cases. We feel that the study of Europeanization in a systematic way will deepen our knowledge of this phenomenon, and we have chosen to face the numerous methodological problems and to best control them through our research design. We have also agreed that a qualitative analysis of a sample of texts was necessary in each national case in order to define unequivocal criteria to distinguish Europeanized and non-Europeanized laws. In spite of a clear and operational definition of Europeanization, several problems emerged and delineate the limits of this research. The comparative analysis raises further methodological concerns. First of all – and this is a frequent challenge for any comparative study – the political systems of the member states show important variation in the balance of power, the status of law and the dynamics of its making. More precisely, the modalities of the transposition of EU law differ strongly from one country to another. Even when the transposition of a directive takes the form of a law, the national differences remain. This task is more or less part of the parliamentary responsibility and can take various forms: law, legislative act delegated to the government, regulation act, simple circular, absence of any norm. The legal texts also have different forms and status: in some countries, the exposition of motives – which often mentions diverse aspects of the political context, amongst others the European integration – is a fully-fledged part of the law. In other countries the law is not preceded by any text setting the background or the rationale, which reduces the probability of a EU constraint being mentioned. In the same way, while in some countries the laws implementing directives mention this fact explicitly, in others this remains inexplicit. Moreover we observe that many laws are only partially Europeanized; no more than a part of the text, often a single article, can be linked to the EU. Also, in some countries laws are especially long and detailed and can include very verbose introductions, whereas in others, the texts are reduced to a strict minimum. Those differences have a strong impact on the possible perception of their Europeanization, the references to the EU being potentially more frequent in the first case than in the second. These differences also challenge our strategy to search for keywords in all countries. This is the reason why each national team was also free to adapt the common methodology to its specific case, in order to best identify the
11
presence of a Europeanized legislature. Also, we must be aware that there are “omnibus laws”, dealing with a whole range of topics related to diverse policies and sectors of public action, as well as laws focused on a single narrow question. The first ones are more likely to refer to EU than the second, and have generally a far greater legal and political impact. The present book is based on nine national cases: eight member states of the EU (Austria, Finland, France, Germany, Italy, Luxembourg, The Netherlands, and Spain) and one non-member (Switzerland). The eight countries of the EU present a large spectrum of situations, regarding the type of political system (parliamentary, semi-parliamentary with strong or weak presidential power), the partisan situation (bipolarization or no, left or right majority), territorial organization (federal or unitary) or the seniority as a member of the EU. We do not include a new member state from the last two enlargements in our sample. As regrettable as this may be, this absence is the result of two types of considerations. Firstly, the decision to analyze for each country all laws voted since 1986 excluded de facto the new member states; the exploitation of the data would also have suffered from a lack of temporal perspective. In a more prosaic way, it also seems that the data necessary for our research is not available in any of the countries which joined the EU in the last decade. We decided to include Switzerland as a control case and as a way to question Delors myth more critically. It indeed allows us to show that European integration not only impacts its member states, but also all its commercial partners and more specifically its neighbors. Thus, it challenges the idea that it would possible for a member state to escape from Europeanization by quitting the EU – unless it chooses to break all ties with its former partners. At four workshops, involving all the participants of the project (Bordeaux 2008, The Hague 2009, Barcelona and Mannheim 2010), we defined a common methodology taking into account our definition of Europeanization, our object of research and the methodological constraints mentioned above. As said, the limits of this research project are mainly linked to the specificity of national legislatures. In order to gather the most useful data, we have thus collectively defined a balance between the necessity to standardize the research design and to take into account of the particularities of each national case. Since we define the Europeanization of laws as a link with the EU, we have considered as Europeanized those laws which contain an explicit reference to it. After several tests, it became clear that the most reliable solution, above all for reasons of cross-country comparability, was to look in a systematic way for the presence in laws of twenty-one keywords and their acronyms in relation to the EU and its politics (Table 1).
Table 1. List of the 21 common keywords.
12
European Union
EEC
European Community
EC
European economic Community
ECSC
Common Market
EAEC
Single Market
Euratom
European Market
EMU
European Coal and Steel Community
EMS
European Atomic Energy Community
Directive
European monetary Union
Community law
European monetary System
European law
EU
Each team performed this task for searching these keywords in all laws voted between 1986 and 2008. The date 1986 has been chosen for four main reasons. First of all, the Single European Act, the first major revision of the Treaty of Rome, was signed in February 1986 and came into effect on 1st July 1987. This marked an important turning point in the normative production of the European Community: its powers were sensibly widened and the decision procedures modified (qualified majority voting at the Council, stronger implication of the European Parliament, restoration of the leadership of the Commission). 1986 stands also for the start of the legislative program which, by the implementation of 279 legislative measures, was at the origin of the Single market (European Commission 1985). The 1st January 1986 is also the date of accession of Spain and Portugal to the EU. Finally, the year 1986 corresponds in most of the countries with the creation of computerized databases; it is often difficult to get hold of the systematic data in electronic form before this date. Preliminary analysis using the raw keywords data showed very important discrepancies from one country to another due to the specificities of the national constitutional, legal and political contexts, and to methodological problems. For example, the very specific process of directive transposition in Italy, the tendency in France to hide the European origin of laws or the systematic mentioning of the EU in Austrian laws, which always state if they are or not connected with it. With regard to methodological difficulties, we can refer to linguistic problems in Finland (each of the keyword being potentially spelled
13
in numerous ways because of the declension system) or to the confusion in French between acronyms (EC, EU, etc.) and common words. At first, our objective was to set up a very strict common methodological frame. The first results however proved that it was not an adequate way to gather comparable data, mainly because of the high impact of national specificities. By using a highly standardized approach, results become incoherent, with a very high level of fake-hits in some cases, and important underestimation of Europeanization in others. We have thus decided to give more autonomy to each team, from a methodological point of view and regarding the definition of Europeanization they use. On that respect, this book is also a methodological reflection about the ways to study Europeanization in a comparative way. In order to overcome the difficulties induced by our research protocol, national teams have adapted it to their own cases. They were free to decide whether it was better to look for keywords in laws full text, legislative projects full text (including or not their introduction) or different kinds of summary. We have measured the impact of those choices by comparing the results of the different options on a sample of texts (one or two years). Several teams have also decided to control the “keywords” indicator with another one: a manual coding of the laws by searching for a reference to an EU link in the full text of statutes, an indicator of the number of laws transposing directives, or a measurement of the EU “impulse” – i.e. the norms resulting from an EU constraint. Over several workshops, we have ensured some fine-tuning between the research designs used by the various teams. At the end of the day, the methodology used to study Europeanization in the nine national cases varies along four dimensions: the type of texts considered (law, bills, statutes, proposals, legislative-decrees…), the type of search in addition to keywords (manual coding, transposition of directives, EU impulse…), the scope of the research (full text, abstract, keywords…) and the conception of Europeanization (link, contextual, legal, downloading…). In our view, the research design provided in this book is however the best compromise available between two objectives that are partially contradictory: a realistic estimation of the level of Europeanization, adapted to the specificities of each national context; a common research design, allowing transnational comparisons. In sum, here are the methodological choices made by each team:
Type of document
Type of research
Scope of the
Focus (wording)
research Austria
federal laws
keyword search
Contextual Europeanization
14
federal laws
manual coding
legal Europeanization
national council
keyword search
Contextual
bills national council
Europeanization manual coding
legal
bills Finland
statutes
Europeanization keyword search (3)
top-down Europeanization
France
statutes
keyword search
keywords associated
EU link
to each law by the official statutes
manual coding
Downloading Europeanization
Germany
legislative
EU impulse
EU impulse
initiatives
(constraining Europeanization)
Italy
Luxembourg
statutes
keyword search
Abstract
EU link
statutes
keyword search
full text
EU link
statutes
keyword search
EU link
legislative-decrees
keyword search
EU link
statutes
keyword search
statutes
transposition of
full text
EU impulse Transposition
directives Netherlands
statutes
keyword search
titles and preamble
Europeanization
(extended to some related treaties) Spain
statutes bills
Switzerland
legislative acts
keyword search
direct
+ manual coding
Europeanization
keyword search
direct
+ manual coding
Europeanization
keyword search
full text
direct and indirect footprint
legislative proposals
keyword search
Abstract
direct and indirect footprint
15
Beyond elements on Europeanization, all teams have collected a number of additional characteristics for each law: -
the type of the law (according to each national system) has systematically been coded, in order to
enable an analysis of the EU link on different types of norms; -
a thematic code has been attributed to each text, to allow comparisons by sector and between
countries.
We
have
used
the
coding
of
the
international
‘Policy
Agendas’
project
(www.policyagendas.com) in which the majority of the teams were involved, and which proposes a very detailed codebook; -
we have also collected information on the amendments of the laws. For each law we know
whether the bill was amended, and for most of the countries, how many times. This information enables us to test several hypotheses regarding the degree of parliamentary influence on the legislative process, for laws with or without EU link. One might assume that MPs develop more activity when faced with Europeanized texts, as they aim to control the process of Europeanization and because it represents a possibility for them to play an active role in the legislative process without giving the impression of countering government’s initiatives. The opposite hypothesis would be that the highly technical character of the transposition of European norms has a repellent effect on the MPs. The data on the amendments will help us to explore this debate. -
Several national teams were also able to collect systematic data on the fate of directives and,
above all, on the percentage of directives transposed by a law, an act of the executive, other type of act, or not notified. This data will allow us to measure both the degree of European constraint on parliaments and governments as well as the power balance among national institutional arrangements in this respect. -
Using the information on the Europeanization of domestic legislatures from the country chapters
we are also able to construct a dependent variable for the empirical analysis of the factors causing the share of Europeanized laws. To analyze the causes of Europeanization we use these data in a comprehensive pooled-time-series-cross-section format.
Aims and structure of the book The use of the EU link allows us to empirically measure the extent to which the laws voted on at the national level are Europeanized and to distinguish Europeanization from other phenomena. Furthermore, we relate these hits to the overall number of national laws and distinguish between different policy areas. This will allow us to compare the share of area-specific Europeanized laws in each country over time. One advantage of our approach is that we can address several types of questions and challenges regarding Europeanization and make country- and area-specific assertions about the Europeanization processes.
16
First of all, we are able to confront Delors myth of national legislation being comprised of 80% of Europeanized laws with real figures.8 In a sample of eight member states plus Switzerland, we contest this myth, which has dominated the public and scholarly debate on the impact of European integration on the member states for more than twenty years, and contribute to both robust quantification and reliable clarification.9 The collection of systematic data on the number of Europeanized laws allows us to contest Delors’ prediction on a very high level of Europeanization while observing and analyzing three trends: The national variation of the Europeanization of domestic parliamentary legislatures: despite the limits of such an exercise, the adoption of a common methodology allows us to observe significant differences in the degree of Europeanization between countries, due to the degree of misfit between domestic legislation and that of the EU, the respective influence of member states in the negotiation of treaties and the adoption of European norms, or specificities of law production and EU norms implementation in the different national contexts. Our data allows us to analyze the diachronic variations of the Europeanization of domestic parliamentary legislatures and to answer several questions: has the European impact gotten stronger over time? If yes, in what areas and what has happened over time? Is it possible to relate these variations to events of the European integration (EU enlargement, rejection of the Lisbon treaty…) or national political life (elections, changes of majority, constitution reforms)? With respect to the area-specific formal development by treaty revision it is important to compare the degree of Europeanization by type of law and by policy areas. By introducing a thematic code for each national law, we can answer the following questions: what types of law and policy areas are the most Europeanized? What are the diachronic trends of Europeanization per sector? Does a concordance exist between national policy priorities and European ones? Is there convergence among the various member states under study? Our cross-sectional dataset allows us to examine rivalry explanations for the area-specific share of Europeanization in these countries. One common explanation is that a (growing) number of EU legislative activities promote “deparliamentarization”. A second hypothesis comes from the literature on coalition politics and assumes that the relationship of coalition partners pursuing diverse interests is characterized by ministerial drift and parliamentary oversight. Legislative review is a means of coalition
8
Public Administration special issue (2010) deals only with 3 countries: Denmark (J. Christensen), Austria (W. Müller and M. Jenny) and the Netherlands (M. Boven and L. Yesilkagit, only for directives transposition). 9 We have already done so in France, by publishing an opinion column during the electoral campaign for European elections of June 2009. Brouard, Sylvain and Costa, Olivier, « Les lois françaises sont-elles écrites à Bruxelles ? », Le Monde, 29 May 2009. See also for Finland: Wiberg, Matti, « EU vaikuttaa vain harvoihin lakeihimme », Helsingin Sanomat, 28 February 2004, A5 ; « EU:n vaikutus lakeihimme vähäinen », Helsingin Sanomat, 28 November 2006, A2 ; « Normitulva: määrällisiä näkökohtia », Oikeus 39:2, 172-178.
17
partners to overcome their delegation problems, which means that coalition partners will scrutinize Europeanized proposals by legislative review when they expect ministerial drift. The quantitative analysis of the Europeanization of domestic parliamentary legislatures is thus a way to evaluate the impact of European integration on the political systems of the member states. Here again, the collected data enable us to answer several questions, which have already been discussed in the political science literature but never looked at from that angle: what is the impact of European integration on the balance of powers at the national level? Are Europeanized laws more or less controlled (i.e. amended and discussed) by national parliaments than the others? Do laws implementing European norms reach stronger consensus in parliament and between coalition partners? How about the relationship between political parties, which must often form coalitions for building government in the member states but pursue different interests in policy making? The widespread hypothesis stating that European integration weakens first and foremost the domestic parliamentary legislatures has never been refuted using solid data on lawmaking. Finally, in a more general way, this renewed perspective prompts re-examination of analyses regarding the interplay between the EU and national political systems. Therefore, the quantitative analysis of Europeanization is not an end in itself but an additional means to study the impact of European integration on political systems and domestic legislative production. The multiplicity of the European integration process, together with the empirical difficulty to distinguish them one from another have inspired our approach. Our goal is not to focus on a specific type or dimension of Europeanization or to weight the respective importance of different characteristics: it is to re-establish the order of sequences of the analysis of Europeanization. Our study aims at identifying the amount of Europeanized laws using reliable criteria and at evaluating the level of Europeanization in different political systems as well as its development through time. For this purpose, we also introduce into the data on recent EU legislative activities which are conceived as a necessary condition for the Europeanization of domestic parliamentary legislatures. This will allow us to estimate whether and to what extent there is a systematic relationship between the activities at the EU and domestic level. Furthermore, we will take a closer look at the domestic level by investigating the reasons for the parliamentary
scrutinizing
activities
in
the
domestic
legislatures.
According
to
the
“deparliamentarization” school of thought domestic parliaments are increasingly lacking power for scrutinizing the activities of the executive, in particular when European laws only offer a “take-it-orleave-it” decision on complex packages. Hence, we will evaluate whether the level of Europeanization of domestic legislatures is also determined by parliamentary responses to Europeanization. Domestic legislatures can respond to several impulses from the EU level: above all directives, but also treaties, international agreements of the EU, decisions of the European Court of Justice (ECJ). In some cases, Europeanization can be the result of more diffuse mechanisms, that have no constraining effect: requests of the European Commission (before an action for failure to fulfill an obligation), objectives
18
resulting from the OMC and other forms of soft law, reference to general principles of EU law, anticipation of EU reforms, importations of models (European or national) by the way of ‘mutual learning’, etc. In those cases, there is no direct measurable European constraint but the laws are nevertheless a consequence of European integration. From the domestic level, the strategic use of EU references can also impact the Europeanization of domestic legislatures. Some texts can refer to a European impulse without substantial but for strategic reason, only to justify unpopular reforms or to propose a “take-it-or-leave-it” decision. On the contrary, the advocates of a European motivated reform can conceal its supranational dimension and claim its authorship. Nevertheless, such strategies strengthen de facto the link between domestic legislatures and the EU and fully fledged parts of the multiplicity of the Europeanization processes (downloading, blaming, anticipating, framing…). The chapters gathered in this book also offer a deep insight on the consequences of Europeanization on the nine political regimes under study. Its impact on the balance of power and relationship between government and parliament is discussed for each country under study. Besides that, several phenomena are underlined. The chapter on the Netherlands describes two worlds of Europeanization in Dutch politics and society. The French chapter discusses the overall impact of European integration on policy-making, and argues that EU is a veto-player more than a dictator. The German chapter provides a cross-validation for the applied measure of Europeanization and furthermore examines the EU influence on governmental agenda setting and parliamentary oversight. The Austrian chapter deals with the notion of EU reference. The chapter on Italy explains why its government can be called ‘pro-European’. The one on Switzerland shows that Europeanization can also be a significant phenomenon outside the EU. Finally, as it is impossible to understand the Europeanization of domestic legislatures without taking into account the development of EU legislative activities, a chapter is devoted to this question. This background view is an essential precondition for the evaluation of the variations of Europeanization through time, by country and by policy sector. The EU normative production is one of the fundamental independent variables for the study of Europeanization and its variations have to get an influence on this phenomenon. Strangely enough, even these data are difficult to access: several research projects start to propose a global view on the EU norm-making process10, but the data are still relatively unknown and incomplete. Above all, EU institutions do not seem to be motivated to communicate reliable and systematic information on this topic. It was therefore necessary to propose our own global overview on the normative activity of the EU in order to interpret national results. On behalf of these results, we finally examine causes of the share of Europeanized domestic laws. Hereby, the common expectation is that the level of EU legislative activities influences the share of Europeanized domestic laws. However, as we show in the national chapters, this kind of policy growth 10
European Institutions Observatory in Paris, Sciences Po, Paris (www.portedeurope.org/spip.php?rubrique526); Policy Agendas project applied to EU (www.policyagendas.org)
19
differs across countries and policy areas. Furthermore, it remains an open question whether Europeanization only “hits” the domestic level without response of the targeted (parliamentary) actors. Looking at the share of Europeanized domestic laws from the domestic perspective, we ask whether parliamentary actors make use of the legislative review process for oversight of executive actors, in particular when the risk of ministerial drift is high in coalition governments. All these analyses on the Europeanization of domestic legislatures were made possible by a collective effort of nine research teams, which generated datasets on different legislatures, filtered their peculiarities and searched for those laws affected by Europeanization. It is important to mention that even though technological progress eases the downloading of legislative documents, and thus the quantitative analysis of legislative production, it has been crucial to be able to count on the expertise and qualitative input of the scholars involved in this project. Hence, the European and the country chapters provide the reader with detailed insight in our data generation process and the specificities of each national case. Our common methodology however allows for the replication of our findings in other countries. We hope that this will stimulate further research and more empirical than normative discussion about Europeanization in the future.
20
References Attiná, Fulvio (1990), 'The Voting Behavior of the European Parliament Members and the Problem of Europarties', European Journal of Political Research, 18 (5), 557-79. Auel Katrine and Benz Arthur, « The Politics of Adaptation: The Europeanisation of National Parliamentary Systems »,The Journal of Legislative Studies, Vol.11, No.3/4, Autumn/Winter 2005, pp.372–393 Bardi 1994 Bartoloni, Stefano (2005): “Restructuring Europe”, Oxford: Oxford University Press. Benoit, Kenneth, et al. (2005), 'Measuring National Delegate Positions at the Convention on the Future of Europe Using Computerized Word Scoring', European Union Politics, 6 (3), 291-313. Bergman T., ‘National Parliaments and EU Affairs Committees: Notes on Empirical Variation and Competing Explanations’, Journal of European Public Policy, 4/3 (1997), pp.373–87 Börzel, Tanja and Risse, Thomas, 2003, ‘Conceptualizing the Domestic Impact of Europe’, in: K. Featherstone and C. Radaelli (eds.), The Politics of Europeanization, Oxford University Press, Oxford, pp. 57-80 Brouard Sylvain and Costa Olivier, « Les lois françaises sont-elles écrites à Bruxelles ? » (Are French laws written in Brussels?), Le Monde, 29 May 2009. Bulmer, Simon and Radaelli, Claudio, ‘The Europeanization of National Policy’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, pp. 338-359 Green Cowles, Maria, Caporaso, James A., and Risse, Thomas (eds.) (2001), Transforming Europe: Europeanization and Domestic Change. (Ithaca/London: Cornell University Press). Duina Francesco and Tapio Raunio (2007), "The open method of co-ordination and national parliaments : further marginalization or new opportunities ?", Journal of European Public Policies, 14-4, p. 489506. European Commision 1985 European Commission 2006 http//ec.europa.eu/community_law/eulaw. Falkner, G 2001 The Europeanisation of Austria: Misfit, Adaptation and Controversies. European Integration online Papers (EIoP) Vol. 5 (2001), No. 13, . Falkner, G., Oliver Treib, Miriam Hartlapp and Simone Leiber (2004) Complying with Europe. EU Minimum Harmonisation and Soft Law in the Member States.(Cambridge: Cambridge University Press).
21
Falkner, G., Treib, Oliver; Hartlapp, Miriam; Leiber, Simone 2005: Complying with Europe. EU Harmonisation and Soft Law in the Member States. Cambridge / UK, Cambridge University Press (2005), 404 Seiten [with]. Featherstone and Radaelli 2003 Follesdal, Andreas; Hix, Simon (2006): “Why there is a Democratic Deficit in the EU: A response to Majone and Moravcsik. Journal of Common Marked Studies 44 (3): 553-562. Franchino, F. and B. Hoyland 2009 'Legislative Involvement in Parliamentary Systems: Opportunities, Conflict, and Institutional Constraints', American Political Science Review, 103,4: 607-621 Graziano 2005 Graziano, P. and M. P. Vink (2007). Europeanization : new research agendas. Houndmills, Basingstoke, Hampshire ; New York, Palgrave Macmillan.
Greenwood Justin, 2003, Interest Representation in the European Union, Palgrave Macmillan, Houndmills, Basingstoke, p. 6. Haverland, Markus (2000), 'National Adaption to European Integration: the Importance of Institutional Veto Points', Journal of Public Policy, 20 (1), 83-103. Heisenberg, Dorothee (2005), 'The institution of 'consensus' in the European Union: Formal versus informal decision-making in the Council', European Journal of Political Research, 44 (1), 65-90. Héritier, Adrienne, et al. (eds.) (2001), Differential Europe: New Opportunities and Restrictions for Policymaking in the Member States. (Lanham, MD: Rowman & Littlefield). Herzog, R. and Gerken, L. (2007), 'Revise the European Constitution to protect national parliamentary democracy', European Constitutional Law Review, 3 (2), 209-18. Holzhacker 2007 Hoppe 2010 Hix, S. 1995 ‘Parties at the European Level and the Legitimacy of EU Socio-Economic Policy’, Journal of Common Market Studies 33(4) 527-554. Hix, S. (1999): The political system of the European Union (Basingstoke: Macmillan). Hix, Simon and Goetz, Klaus H. (2000) « Introduction: European integration and national political systems ».West European politics, 23 (4). pp. 1-26. ISSN 0140-2382 Hix, S. 2001 Legislative Behaviour and Party Competition in the European Parliament: An Application of Nominate to the EU’, Journal of Common Market Studies 39(4) 663-688. Kreppel, Amie and Hix, Simon (2003), 'From "Grand Coalition" to Left-Right Confrontation: Explaining the Shifting Structure of Party Competition in the European Parliament', Comparative Political Studies, 36 (1/2), 75-96.
22
Hix, Simon (2005), 'Chapter 3: Legislative Politics', in Simon Hix (ed.), The Political System of the European Union (2nd Edition edn.; Houndmills: Macmillan), 72-110. Hug, Simon, und Thomas König. 2002. "In View of Ratification: Governmental Preferences and Domestic Constraints at the Amsterdam Intergovernmental Conference." International Organization 56 (2): 447-476 Jachtenfuchs, M. (2001), The Governance Approach to European Integration, Journal of Common Market Studies, 39:2. Kassim et al. 2000 Knill, Christoph and Lenschow, Andrea (1998), 'Coping with Europe: the impact of British and German administrations on the implementation of EU König, Thomas and Hug, Simon (2000), 'Ratifying Maastricht. Parliamentary Votes on International Treaties and Theoretical Solution Concepts', European Union Politics, 1 (1), 93-124. König, Thomas. 2007. "Discontinuity. Another Source of the EU's Democratic Deficit?" European Union Politics 8 (3):411--32. König, Thomas, and Lars Mäder. 2007. "The delegation of transposition powers in the member states of the EC: From a transaction cost perspective." Paper presented at the 4th ECPR general conference, Pisa, Italy. König, Thomas (2008), 'Analysing the Process of EU Legislative Decision-Making: To Make a Long Story Short..."', European Union Politics, 9 (1), 145-65. König, Thomas and Mäder, Lars (2008), 'Das Regieren jenseits des Nationalstaates und der Mythos der 80-Prozent-Europäisierung in Deutschland', Politische Vierteljahresschrift, 438-64. König, Thomas and Junge, Dirk (2009), 'Why Don't Veto Players Use Their Power?', European Union Politics, 10 (4), 507-34. König, Thomas/Luetgert, Brooke (2009), 'Troubles with Transposition: Explaining Trends in Member State Notification Failure and Timeliness', (39; Cambridge: British Journal of Political Science), 163-94. Kreppel, Amie and Tsebelis, George (1999), 'Coalition formation in the European Parliament', Comparative Political Studies, 32 (8), 933-66. Ladrech, Robert, ‘Europeanization of Domestic Politics and Institutions: The Case of France’, in: Journal of Common Market Studies, vol. 31, n°1, 1994, pp. 69-88 Ladrech 2005 Laursen F. and S.A. Pappas (eds.), The Changing Role of Parliaments in the European Union (Maastricht: European Institute of Public Administration, 1995), pp.43–60. Lord, Christopher (2004) “A Democratic Audit of the European Union”. London: Palgrave Lewis, Jeffrey (2003), 'Institutional Environments and Everyday EU Decision Making: Rationalist or Constructivist?', Comparative Political Studies, 36 (1/2), 97-124.
23
Magnette Paul and Yannis Papadopoulos (2008), “On the politicization of the European consociation: A middle way between Hix and Bartolini”, EUROGOV working paper No. C-08-01 Maïa, Jean, 2005, « La contrainte européenne sur la loi », Pouvoirs, pp. 53-71. Mair, P. (2007), 'Political opposition and the European Union', Government and Opposition, 42 (1), 1-17. Marks, Gary and Steenbergen, Marco (eds.) (2004), European Integration and Political Conflict (Cambridge: Cambridge University Press). Maurer, Andreas, and Wolfgang Wessels (eds.) (2001), National Parliaments on their Ways to Europe: Losers or Latecomers? Baden-Baden: Nomos. Majone, Giandomenico (1998), 'Europe’s “Democratic Deficit”: The Question of Standards', European Law Journal, 4 (1), 5-28. Moravcsik, A. (1994): “Why the European Union Strengthens the State: Domestic Politics and International Cooperation”, Center for European Studies, Working Paper Series 52. Moravcsik A. (2002), Andrew, “Reassessing Legitimacy in the European Union”. Journal of Common Market Studies, Vol. 40, pp. 603-624. Moravcsik, Andrew (2004), 'Is there a 'Democratic Deficit' in World Politics? A Framework for Analysis', Government and Opposition, 39 (2), 336-63. Moravcsik, A. (2008), 'The European constitutional settlement', World Economy, 31 (1), 158-83. Norton Ph. (ed.), National Parliaments and the European Union (London: Frank Cass, 1996) O'Brennan, J. and T. Raunio 2007, National Parliaments within the Enlarged European Union: from ‘Victims’ of Integration to Competitive Actors? Abingdon: Routledge Olsen, J. P. (2007): Europe in Search of Political Order (Oxford: Oxford University Press). Pridham, G. 1995 Transitions to democracy: comparative perspectives from Southern Europe, Latin America and Eastern Europe Pridham, G. and Pridham, P. 1981 Transnational party co-operation and European integration: The process towards direct elections Allen & Unwin (London and Boston) Public Administration (2010), 88-1. Raunio T. and S. Hix, ‘Backbenchers Learn to Fight Back: European Integration and Parliamentary Government’, in West European Politics 23/4 (2000) pp.142–68 Raunio T., ‘Always One Step Behind? National Legislatures and the European Union’, Government and Opposition, 34/2 (1999), pp.180–202. Raunio, Tapio and Matti Wiberg (2010): How to Measure the Europeanisation of a National Legislature?, Scandinavian Political Studies 33:1, 74-92. Rittberger, Berthold and Schimmelfennig, Frank (2005), The Constitutionalization of the European Union: Explaining the Parliamentarization and Institutionalization of Human Rights (Wien: Institute for Advanced Studies).
24
Saalfeld, Thomas, “Deliberate Delegation or Abdication? Government Backbenchers, Ministers and European Union Legislation”, The Journal of Legislative Studies, Vol.11, No.3/4, Autumn/Winter 2005, pp.343–371. Scharpf, Fritz W. (1993): “Versuch über Demokratie im verhandelnden Staat.” In: Verhandlungsdemokratie, Interessensvermittlung, Regierbarkeit, edited by R. Czada and M.G. Schmidt. Opladen: Westdeutscher Verlag. Siedentop, Larry 2001, Democracy in Europe London: Allen Lane. Smith E. (ed.), National Parliaments as Cornerstones of European Integration (London: Kluwer Law International, 1996) Thomson, R., R. Torenvlied and J. Arregui. 2007. The paradox of compliance. Infringements and delays in transposing EU directives. British Journal of Political Science, 37: 685-709. Töller, Annette Elisabeth (2006): “How European Integration Impacts on National Legislatures: The Europeanization of the German Bundestag”, Center for European Studies Program for the Study of Germany and Europe, Working Paper Series 06.2. Töller, Annette Elisabeth (2007). “Mythen und Methoden. Zur Messungder Europäisierung der Gestzgebung des deutschen Bundestagesjenseits der 80%-Mythos.” Helmut-Schmidt-UniversitätUniversität der Bundeswehr Hamburg. Töller, Annette Elisabeth (2008). ”Measuring the Europeanization of Public Policies – But How? A Research Note.” Helmut-Schmidt-Universität-Universität der Bundeswehr Hamburg. Wiberg, Matti (ed.) (1997): Trying to make democracy to work: the Nordic parliaments and the European Union (Stockholm: Gidlunds). Wiberg, Matti (2004). EU vaikuttaa vain harvoihin lakeihimme, Helsingin Sanomat, 28. February 2004, A5. Wiberg, Matti (2006).EU vaikuttaa vain harvoihin lakeihimme, Helsingin Sanomat, 28. November 2006,A2. Wiberg, Matti (2010). Normitulva: määrällisiä näkökohtia, Oikeus 39:2, 172-178. Zeitlin J., Pochet, P. and L. Magnusson 2005 The Open Method of Coordination in Action: The European Employment and Social Inclusion Strategies, Brussels: Presses Interuniversitaires Européennes-Peter Lang.
25
2. EU Legislative Activities and Domestic Politics
Thomas König (University of Mannheim), Tanja Dannwolf (Mannheimer Zentrum für Europäische Sozialforschung), Brooke Luetgert (Mannheimer Zentrum für Europäische Sozialforschung),
Much of the Europeanization debate revolves around the impact of Brussels on the domestic legislatures of the member states, in particular the dominance of their legislatures by the binding and enforceable activities of the EU. Because this impact is a direct result of the amount of legislative activities at the EU level, this chapter documents EU legislative decision making and explores the consequences of treaty revisions and enlargements, which have changed the goals and conditions for EU legislative decision making over time. The famous prediction of the former Commission president Jacques Delors, claiming that 80% of economic legislation, perhaps of taxation and social legislation, would originate in Brussels (6.7.1988. EP 2-367/157), was made at the end of the 1980s when EU legislative activities reached a peak in the attempt to complete the Common market by the effective application of qualified majority voting in the Council of Minister until the end of 1992.11 Ever since the EU has experienced several enlargement rounds with the accession of a large number of countries having diverse interests, which are considered to make the adoption of EU legislation more difficult. Moreover, the Nice treaty (2003) complicated the rules for qualified majority voting in many policy areas, which led to a higher risk of gridlock in the Council of Ministers. This prompts the questions whether EU legislative activities have slow down due to accessions and consensus rules in the Council of Ministers. On the other hand, Brussels impact may not only be determined by the amount of legislative activities, but also by the type of instrument used. For example, it is possible that Brussels still produces a large number of legal obligations via directives, which member states must transpose into domestic law. Hence, this chapter reports on the amount of legislative activities, the type of instruments and their area-specific application over the past 20 years. In general, the level of EU legislative activities is determined by two crucial factors, the extent of material competencies of the EU and the motivation or ability of the Commission, the member states
11
Under the presidency of Jacques Delors, the Commissioner with special responsibility for the Internal Market, Lord Cockfield, had a White Paper drawn up: a program for the completion of the Internal Market, proposing the total abolition of physical, technical and fiscal frontiers on the basis of a precise seven-year schedule and setting out the some 310 directives and regulations to be adopted to that end by the Council of Ministers.
26
and the European Parliament to use them. Regarding the material competencies, the EU has integrated a large number of policy areas through several treaty revisions. After the creation of the ECSC, the EEC and EURATOM in the 1950s, in particular the Single European Act (1987), Maastricht (1993), Amsterdam (1999) and Nice (2003) treaties have significantly raised the EU’s material power in several policy areas. At the same time, these treaties have increased the ability of EU actors to use these competencies by introducing and expanding the application of qualified majority voting in the Council of Ministers. However, it is commonly believed that the accessions of new member states in the 1980s, mid-1990s and beginning of 2000s have increased the diversity of interests in the Council of Ministers, which conventionally reduces the EU’s ability to act (König and Bräuninger 2004). Furthermore, the European Parliament has been equipped with veto power under the shadow of qualified majority voting in the Council of Ministers, which should also lower the amount of activities (Schulz and König 2000). These developments – the increasing level of material competencies on the one hand and the diversity of interests on the other hand, which are accompanied by a changing of the Council of Minister’s voting rules and the participation of the European Parliament – demand a closer inspection of the EU’s legislative output (König et al. 2006). In particular, the number and range of directives that must be transposed by the member states into domestic law, is an important indicator for the extent to which domestic legislation originates in Brussels. Hereby, member states have the discretion to implement directives by legislative or executive measures and vary considerably in their use of these transposition instruments (Franchino and Hoyland 2009; König and Mäder 2007). As a consequence. the level of Brussels executive influence is also homemade by the in- or exclusion of domestic legislatures in the implementation of directives. This chapter will take a closer look at the EU legislative activities from 1984 to 2007 using information from EURLEX. We will present the amount of binding legislation, which the EU adopts in the form of directives, regulations and decisions. These three types of legislation are adopted by the Council of Ministers under the consultation procedure, the Council of Ministers and the European Parliament under the cooperation and codecision procedures, or by the Commission alone (tertiary legislation). We will see how the use of these procedures has developed over time and across policy fields, which were concurrently complemented by treaty revisions and accessions. Furthermore, we will take a closer look at the application of qualified majority voting in the Council of Ministers and the participation of the European Parliament. In addition to these formal characteristics of EU legislative productivity, we also present indicators on the potential for conflict among the member states, which provide additional insight into the nature of EU decision making. Finally, we will pay special attention to directives, which can have a direct impact on domestic legislatures due to their required transposition into domestic law.
27
EU legislation over time
The history of EU legislative activities is characterized by distinct events of enlarging and deepening. As suggested in Figure 1, the treaty revisions of the past twenty years deepened the relationship among the member states by increasing the material competencies of the EU across various policy fields and expanding qualified majority voting in the Council of Ministers, while also steadily increasing the influence of the European Parliament in EU legislative politics since the Single European Act in 1987. During this period, the EU also experienced several accession rounds with the Western and Southern enlargements until the mid-1980s, followed by the Northern in the mid-1990s and the Eastern enlargement rounds in 2004 and 2007. The Lisbon treaty marks the latest treaty revision, which came into force in 2009 after a nine-year reform effort. This final event clearly demonstrated the variation in the interests of the member states in the design of qualified majority voting and the composition of the Commission. Both topics were heavily disputed among the political leaders and the public of large and smaller member states leading to several conflicts and crises, including a perennial reflection period (Finke et al. 2012, forthcoming). Figure 1 about here Do we expect that these events of enlarging and deepening affected EU legislative activities and hence Brussels impact on the domestic legislatures of the member states? The answer to this question is not easy because some events have the potential for contradictory effects: while it is commonly expected that enlargement would make decision making more difficult among the member states, the extension of qualified majority voting in the Council of Ministers could counter this enlargement effect (Schulz and König 2000). Similarly, the effects for the Commission remain an open question: on one hand, the expansion of material competencies might overload the Commission’s capacity to act, while it can also facilitate the proposal of package deals which find consensus among the actors involved (König and Junge 2009). To determine these effects empirically, we need to take a closer look at the binding legislative activities of the EU across policy areas and over time. Our numbers stem from the Eurlex database (http://eur-lex.europa.eu/en/index.htm) that is publicly accessible and run by the Publications Office of the Commission. Eurlex documents the history of decisions, directives and regulations. Compared to regulations and directives, decisions are directed toward specific member states. Hence, we expect the number of decisions to increase with membership size. In contrast to regulations and decisions, directives are only binding “as to the result to be achieved”. As a consequence, directives have to be transposed by member states into domestic law. This does not mean that regulations and decisions do not affect domestic legislatures (Töller 2008, 17). However, directives are the most direct measure for the impact of EU legislative activities on domestic legislatures, because they must be directly transposed and enforced under domestic law –
28
either by domestic legislative or executive measures. Compared to decisions, it is not easy to make a plausible prediction on the development of directives. If directives are more related to achieving common goals such as the completion of the internal market by providing more discretion to the member states, one would expect their heydays in the end of the 1980s. Another important distinction for the evaluation of EU legislative activities concerns tertiary and secondary legislation. Compared to secondary legislation, which is adopted by the member states in the Council of Ministers and sometimes with the participation of the European Parliament, tertiary legislation is delegated to the Commission (COM) and mainly comprises implementing and administrative matters.12 One reason for this delegation is the technical nature of tertiary legislation, which requires more expertise than democratic legitimization. However, once the Commission has received this delegation power, it becomes more difficult to overrule tertiary legislation when the Council of Ministers is blocked. Hence, we expect that tertiary legislation becomes more prominent when the level of conflict increases among member states. Within secondary legislation, we distinguish between legislation adopted by the Council of Ministers and the European Parliament (CM & EP) from those acts, which are only passed by the Council of Ministers (CM). Council legislation is not exposed to parliamentary veto power, which can make a difference when dominating domestic politics. Table 1 provides an annual overview of all binding European legislation adopted between 1984 and 2007. Table 1 about here According to table 1, there is an upward trend in the total numbers of decisions and regulations, while the annual adoption rate of directives has remained almost stable over time. We find a similar trend for secondary legislation whereby the annual number of decisions has increased and that of regulations and directives has decreased. The goal of completing the Common Market by the end of 1992 seems indeed to account for the high numbers of directives between 1988 and 1993. While we correctly expected a higher number of decisions due to the larger number of member states, the increase in the competencies of the Commission stemming from the Maastricht treaty as well as the Northern enlargement explains the high numbers of Commission regulations since 1995. Additionally, a shift in decision-making towards the Commission may have been encouraged by difficulties within interinstitutional decision-making arising from more conflict in the Council or more power of the European Parliament. As the deepening of the EU has been accompanied by an increase in power of the European Parliament and in the application of Council qualified majority voting, the decrease in the annual adoption rate of secondary legislation could be a result of these changes in formal rules for inter12
Basically, three types of Comitology procedures exist e.g. , however, we cannot distinguish among them in these data (for a list of committees and an overview of Comitology procedures please refer to the Annual Reports of the Work of the Comitology Committees).
29
institutional decision-making. Inter-institutional decision-making is best documented in the PreLex database, which provides more detailed insight into the involvement of the European Parliament as well as Council decision-making.13 PreLex documents legislative processes that are inter-institutional, but excludes tertiary legislation.14 Participation of the European Parliament in legislative decision-making has been brought forward as one of the remedies for the purported European democratic deficit.15 The involvement of the European Parliament, the application of qualified majority voting and conflict in the Council of Ministers has changed with treaty revisions and multiple enlargement rounds. Looking at the empirical evolution of parliamentary involvement over time reveals that the share of legislative acts adopted under cooperation and/or codecision has steadily increased since 1984 but has not yet crossed the 30% mark. Compared to parliamentary involvement, the application of Council unanimity voting has decreased considerably. Thus, the application of unanimity as well as decision-making without parliamentary involvement has decreased, but the results suggest that this has not occurred to the extent that the treaty reforms might have intended. Nonetheless, these developments might explain the drop in adopted EU legislation. Treaty revisions brought institutional reforms as well as an increase in the number of policy fields delegated to the EU level, and this might contribute to the observed patterns of EU legislative output. In order to enhance the comparability with the national chapters, we recoded the directory codes provided by Eurlex to 15 policy fields according to the scheme of the policy agenda project that are important for the EU.16 The different material focus of EU legislation is evident from the second largest category, “Macroeconomics”, which includes legislation relating to the Common Market whereas “Banking and Finance” only lists rules relating to banks and insurances. In the EU context, the policy field of “Government Operations” accommodates legislation on general affairs and institutional matters, e.g. institutional provisions. Legislation relating to “Judicial and Police Cooperation” is assigned to the category “law” of the policy agenda project. Table 2 shows the distribution of EU legislation across these policy fields and also indicates the type of legislation according to instrument and responsible institution. Table 2 about here
13
Since PreLex and EURLEX provide different information and have some differences , we use a sample that is cross-classified by the proposal number. 14 The data used here is an update of König et al. (2006) and comprises only those cases in PreLex that have a proposal that is also found in the EURLEX database. 15 While the title of legislation allows for the identification of legislation where the codecision procedure was applied, PreLex allows for a more detailed examination of reveals more information on the extent of parliamentary involvement. Accordingly, we define parliamentary involvement as the application of the cooperation or co-decision procedure. 16
An overview of the directory codes can be accessed online at the EURLEX homepage.
30
Differences between policy fields are considerable in terms of numbers, institution involved and type of instrument.17 Accordingly, the Common Agricultural Policy accounts for the largest share of EU legislation across all policy fields: 67.8 % of all binding EU legislation adopted between 1984 and 2007 deals with “Agriculture and Fisheries”. The distribution over time even suggests that agricultural policy has grown in importance over the period under investigation, for which enlargement is a logical and probable explanatory factor. The large share of decisions that are member state-specific further underlines this. However, the widespread use of tertiary legislation in this policy field also points to the technical and regulative nature of this policy area. The second biggest policy field is ”Macroeconomics”, which accounts for 15.7% of legislation. Judged by the legislative output over time, this policy field has decreased in importance. It also includes a rather high share of tertiary legislation as well as regulations as it largely falls under Commission competence. Further 7.3% of EU legislation belongs to the policy field of “Foreign Trade”. Similar to “International Affairs” and “Law” this policy field stands out by its absence of directives and a low share of codecision legislation. This is in striking contrast to legislation on “Banking and Finance” with many directives and codecision legislation. “International Affairs” and “Government Operations” each account for 2.5% of EU legislation and are very similar to “Foreign Trade” with low share of directives as well as low involvement of the European Parliament. The annual output of the smaller policy fields has increased steadily over the period of investigation. The EMU is also reflected in the growth of legislation in “Banking and Finance” and “Government Operations” since the end of the 1990s. First introduced by the Maastricht Treaty, Justice and Home Affairs differentiated into “Law” and “Civil Rights” presents the best example of a deepening in EU cooperation. These fields are characterized by a remarkable growth rate, whereas “Labor” and “Education” have stayed comparatively stable over time. ”Law” and to a lesser degree “Civil Rights” include few directives and a large share of Council legislation. By contrast, environmental policy has been identified as one of the most Europeanized policy fields. This is reflected in the continually increasing output of environmental legislation over the years and the inclusion of a high share of Commission legislation (68.2%). Furthermore, 20.2% of environmental policies are adopted in the form of directives. Empirically as well as formally, new policy fields have been included, which are likely to cause more conflict and reduce the legislative output. The legislative output suggests that the deepening dimension is reflected in an increase in legislation and addition of new policy fields (e.g. Law) as well as an increase in parliamentary involvement in the more established policy fields. Investigating the agenda of the Council, which is divided into A and B points can assess the degree of conflict in this phase of deepening and enlargement. A-points are deemed not to need further discussion and have been resolved by national representatives previously in COREPER prior to the ministerial meeting. 17
An overview of the development of the policy fields over time can be found in Appendix 1.
31
Proposals listed as B points could not be resolved in preparatory meetings by the national representatives (Rules of Procedure of the Council, Decision 2004/338/EC, EURATOM, Article 3) and can therefore be regarded as controversial. Table 3 shows the percentage of parliamentary involvement, the application of unanimity voting as well as Council conflict within the 15 policy fields. Table 3 about here The findings confirm that member states have protected their veto powers in some areas such as in Energy, Banking and Finance, Social Welfare and Health and that the European Parliament is represented in most legislation dealing with “Education”, ”Environment”, “Banking and Finance”, and ”Transport”. The numbers also suggest that the Common Agricultural Policy is contested with respect to the relatively high share of B-points. The same is true for Education.
Implementation of Directives Different types of legislative acts have different consequences for domestic legislatures and thus for their Europeanization. Directives stand out because they are used to harmonize national legislation rather than equalize it, allowing the member states to adapt their national legislation to European policies. Directives also require transposition, application and enforcement at the domestic level. The main data source to empirically evaluate transposition activities is provided by sector 7 of the Eurlex database (König and Luetgert 2009). Member states must notify the legal instruments they adopted in the course of transposing the directive to the Commission, and these are documented in the Eurlex online database.18 Even though this notification requirement only exists for directives, it allows us to investigate the level of Brussels impact on domestic legislatures from the viewpoint of EU legislative activities. For this purpose, we also disregard measures with an adoption date prior to a directive.19 Arguably, only an explicit reference guarantees legal accessibility for citizens. Figures 2a and 2b about here As a potential consequence of enlargement and the completion of the Common Market, the number of non-Commission directives has decreased slightly in the 1990s (figure 2b). At the same time, the use of regulations has also declined. Most likely reflecting the increase in conflict and the enlargement of the EU, the use of decisions within secondary legislation has increased (figure 2b). In contrast to policy fields that aimed at the completion of the Common Market, new EU competencies such as Justice and Home Affairs, Economic and Financial Affairs, and to some extent Education not only 18
Sometimes only information on the year of adoption can be identified because of differences in national reporting standards, in which case we coded the instrument as adopted on 1 July of the respective year. 19 Since 1990 member states are required to make reference to the directive in the transposing instrument, from which an obligation to change existing legislation can be inferred.
32
have low parliamentary involvement but are also dominated by decisions and regulations rather than directives. Another effect of the deepening is the increase of Commission legislation within directives, which was 25% in 1985 and constituted more than half of the adopted directives in 2006. In addition, this jump in Commission legislation, especially regulations, from 1995 onward may partially reflect the completion of the Common Market. For example, the Maastricht Treaty granted the Commission more competence in the area of competition policy. This pattern is even more striking for regulations that nearly doubled from 1995 onward. Insofar, the numbers confirm our expectations about tertiary legislation and decisions, which have increased for secondary legislation. Taking a closer look at the legislative process, as documented in Prelex, also reveals that directives stand out with regard to their conflict potential. First, they have the highest share of B points in the Council of Ministers. 16.8% of directives require continued discussion in the Council of Ministers compared to 6.3 % (decisions) and 13.4 % (regulations). Second, the average adoption speed of a directive takes far longer than that of other legislative acts. Third, directives experience the highest share of unanimity voting among adopted inter-institutional legislation. 35.8 % of directives fall under unanimity compared to 24.3 % of decisions and 18.8 % of regulations. This supports the notion that directives include the more contested issues and cover salient legislation even though they fail to constitute the majority of European legislation. As discussed earlier, the deepening of the EU has been accompanied by an increase in tertiary legislation and an increase of the power of the European Parliament that is most pronounced in the codecision procedure. Although the involvement of more actors on the EU level might lead to higher conflict and fewer adoption rates, another side effect could be easier transposition at the national level. The notion that legislation adopted by the codecision procedure faces fewer difficulties on the national level is supported by compliance research (e.g. Luetgert and Dannwolf 2009; Mastenbroek 2003). When looking at the share of directives that were transposed on time of depending on the actors involved in European policy making, König and Luetgert (2009) find that the technical Commission legislation is adopted with the fewest difficulties in most member states (Austria being the only clear counter-example). A second finding of their transposition study is that directives adopted by codecision face less implementation difficulties on the national level than directives where the EP does not play a strong role. Finally, they show that transposition patterns highlight two countries where the involvement of actors during EU policy making does not seem to make a difference: Portugal and Luxembourg. This might suggest that implementation difficulties are rooted in national administrative difficulties, limited resources or a lack of willingness. Regarding the impact of this transposition on the domestic legislatures of the member states, there is no prerequisite by the EU that directives have to be implemented by legislative measures. From a legal perspective, the requirement is that the measure be publicly accessible and enforceable for private subjects within the national legal system. Besides restrictions introduced by the limited time
33
prescribed by the deadline, there is no restriction to the involvement of a domestic legislature from the EU level. The extent to which domestic legislatures take part in the implementation process thus depends on institutional prerequisites, governmental preferences and practices of parliamentary decision-making. Empirically, cross-country as well as cross-sectoral differences can be observed (see König and Mäder 2007; Franchino and Hoyland 2009). Figure 3 shows the percentages of directives that have been adopted with at least one legal instrument that formally requires the involvement of the domestic legislature. Figure 3 about here Finland, Austria and Germany have the highest shares of formal involvement of their legislatures. Ireland, Italy, Portugal and the UK are on the other end of the scale. However, this does not capture the rights of national parliaments to influence EU legislative decision-making by exerting influence on their government via binding orders. For example, Denmark is well known for rather strong involvement of the parliament in the preparatory stage of EU decision-making (e.g. Bursens 2002) but only has a rather low share of the Danish legislature’s involvement in implementation. Even more importantly, this does not take into account institutional prerequisites some countries have taken in order to speed up the transposition process that might involve the national parliaments indirectly. For example, the Italian parliament delegates transposition to the government by the legge comunitaria (see chapter on Italy). Notably, this reform has succeeded in speeding up the transposition process (Borghetto, Franchino, and Giannetti 2006). As a consequence of the deepening of the EU, an increase in involvement of domestic legislatures could be expected due to the recent increase in new policy fields that more directly challenge issues of national sovereignty than regulative rules on agriculture. The deepening with respect to the policies decided on the EU level may also cause a rise in the importance and also the legal necessity of involvement of domestic legislatures. Looking at the share of the legislatures’ involvement in national transposition over time, we see that there is a minimal and gradual increase in their involvement in most member states. A clear increase is observable in Austria, Germany, France, the Netherlands, Belgium and Denmark. The only country evidencing a slight decrease is Finland. Ireland, the UK, Italy, Portugal and Luxembourg have maintained relatively stable rates of involvement of their legislatures. Turning to cross-sectoral differences, figure 4 illustrates that differences across countries are considerable. Policy fields are approximately sorted according to size so that small policy fields are displayed on the right side. Some countries stand out by few cross-sectoral differences, such as Finland and Austria on the upper and Greece, Italy and the UK on the lower end measured by shares of legislative acts. Despite these different patterns between countries, some policy fields stand out due to their relative proclivity for legislative involvement in the transposition of European directives. We
34
would expect that legislative involvement is higher in policy fields that touch on sensitive issues such as the welfare system, labor or civil rights as well as financial regulations. Figure 4 about here In line with these expectations, legislatures seem to participate relatively more often in the transposition of directives in the area “Social Welfare”, “Civil Rights” and “Banking and Finance” in all countries. Policy fields with relatively low legislative involvement in nearly all countries include “Macroeconomics”, “Agriculture” and “Transportation”, which also are more established policy fields that have a high share of technical Commission legislation. Earlier, we identified “Environment” as a policy field with many European directives as well as a high share of tertiary legislation. The patterns for “Environment” are quite different across countries. In stark contrast to the UK, Portugal, and Spain, especially the Austrian and German legislatures take part in the transposition process more than in other policy fields. Differences between countries may be due to different priorities as well as different institutional conditions.
Conclusion This chapter set out to present the developments of EU legislative decision making and the consequences of the process of deepening and enlarging of the EU. We have identified three key developments at the EU level: First, there is a decrease in secondary and a slight increase in tertiary legislation. Within secondary legislation, decisions have become more prevalent whereas all other instruments have decreased in numbers. With respect to policy fields, agriculture is responsible for the bulk of EU legislation. These numbers confirm our expectation that enlargement affected the legislative activities of the EU. Most obviously, the number of decisions increased due to a larger membership size, while the increasing weight of tertiary legislation may also result from the fact that decision making in secondary legislation has become more difficult by a larger number of member states with diverse interests. Second, the deepening of the institutional rules of the EU is reflected in day-to-day policy making. We find that parliamentary involvement has increased as well as the application of qualified majority voting in the Council of Ministers. Our findings also suggest that conflict in the Council of Ministers increased after central developments, such as the completion of the Common Market, the inclusion of Judicial Politics and the accession of new members. This confirms our previous result that decision making in secondary legislation has become more difficult. However, this development also depends on the type of instrument: the number of decisions slightly increased, while the number of regulations decreased and the number of directives remained almost stable in secondary legislation.
35
Third, directives remain the instrument with the highest impact on domestic legislatures. Our results suggest that directives are more prone to conflict, e.g. take longer to adopt and have a higher share of B-points in the Council of Ministers. When measuring the Europeanization of domestic legislatures, directives are particularly important because their obligations must be transposed either by those legislatures or executive measures into domestic law. We identified clear country-specific patterns regarding the involvement of domestic legislatures in the transposition process. Generally, Finland, Austria, Germany and Sweden are characterized by high formal involvement of their legislatures, although this differs across policy fields. As a result of the treaty reforms, we would expect the involvement of domestic legislatures to increase due to the inclusion of more significant policy fields. Investigating domestic legislatures’ involvement over time confirms this expectation only to some degree. Our results of EU legislative decision making and the consequences of the process of deepening and enlarging of the EU emphasize the need for considering area-specific and cross-country differences particularly in formal institutional rules. Agricultural politics, followed by macroeconomics, considerably dominate EU legislative output. Environmental politics, by contrast, is relatively low in numbers but usually adopted with directives. Hence, we would expect an area-specific level of Europeanization in the member states. However, when the type of instrument is decisive for Europeanization, the influence of Brussels on the domestic legislatures is also homemade because they have the discretion to transpose directives either by legislative or executive measures. From this point of view, we expect a country- and area-specific level of Europeanization which changes over time. The chapters on national policy-making take the next step and clarify our expectations.
36
Bibliography Borghetto, Enrico, Fabio Franchino, and Daniela Giannetti. 2006. Complying with the transposition deadlines of EU directives. Evidence from Italy. Rivista Italiana di Politiche Pubbliche 2 (1): 7-38. Bursens, Peter. 2002. Why Denmark and Belgium have different implementation records: on transposition laggards and leaders in the EU. Scandinavian Political Studies 25 (2): 173-195. Franchino, Fabio, and Bjorn Hoyland. 2009. Legislative Involvement in parliamentary systems: Opportunities, conflict, and institutional constraints. American Political Science Review 103 (4): 607-620. Hartley, T. C. 2003. The Foundations of European Community law. An Introduction to the constitutional and administrative law of the European Community. 5 ed.: Oxford University Press. König, Thomas, and Thomas Bräuninger. 2004. Accession and Reform of the European Union. A Game-Theoretical Analysis of Eastern Enlargement and the Constitutional Reform. European Union Politics 5 (4): 419-139. König, Thomas, Brooke Luetgert, and Tanja Dannwolf. 2006. Quantifying European Legislative Research: Using CELEX and PreLex in EU Legislative Studies. European Union Politics 7 (4): 553-574. König, Thomas, and Lars Mäder. 2007. Bürokratisierung oder Politisierung? Eine Untersuchung der Umsetzungseffekte von Europäischen Richtlinien in den Mitgliedstaaten der Europäischen Union im Zeitraum von 1985-2003. Edited by Deutsches Forschungsinstitut für öffentliche Verwaltung of Speyerer Forschungsberichte. Speyer: Deutsches Forschungsinstitut für öffentliche Verwaltung. Luetgert, Brooke, and Tanja Dannwolf. 2009. Mixing Methods: An Analysis of EU Member State Transposition Patterns. European Union Politics 10 (3): 307-334. Mastenbroek, Ellen. 2003. Surviving the Deadline: The Transposition of EU Directives in the Netherlands. European Union Politics 4 (4): 371-395. Töller, Annette Elisabeth. 2008. Mythen und Methoden. Zur Messung der Europäisierung der Gesetzgebung des Deutschen Bundestages jenseits des 80-Prozent-Mythos. Zeitschrift für Parlamentsfragen (1): 3-17. Wölke, Jonas. 2002. Die Umsetzung von Richtlinien der Europäischen Gemeinschaft. Eine rechtsvergleichende Untersuchung zum Recht der Bundesrepublik Deutschland, der französischen Republik und des Vereinigten Königreichs. Edited by Jürgen Schwarze. Vol. 270 of Schriftenreihe Europäisches Recht, Politik und Wirtschaft. Baden-Baden: Nomos Verlag.
37
Tables and Graphs Figure 1: Historical Development by Treaty Revisions and Enlargement Rounds
Enlargement Rounds: West `73=Accession of Denmark, Ireland and UK; South `81&`86=Greece (1981), Portugal & Spain (1986); North `95= Austria, Sweden & Finland; East `04, `07=Estonia, Latvia, Lithuania, Poland, Czech Republic, Slovakia, Hungary, Slovenia, Malta & Cyprus (2004), Bulgaria & Romania (2007) Treaty Revisions (year indicates its entry into force): EEC `57=European Economic Treaty 1957, EEA `87=Single European Act 1987, Maas `93=Treaty of Maastricht 1993, Amst `99= Treaty of Amsterdam 1999, Nice `03=Treaty of Nice 2003, Lisb `09=Lisbon Treaty 2009
38
Table 1: Type of Legislation and Instrument by Year of Adoption from 1984 – 2007 Decision Year COM CM 1984 453 81 1985 311 77 1986 378 171 1987 340 110 1988 396 105 1989 387 151 1990 349 144 1991 431 125 1992 387 134 1993 508 155 1994 648 173 1995 483 159 1996 513 141 1997 637 178 1998 449 221 1999 531 207 2000 504 205 2001 653 208 2002 503 204 2003 495 234 2004 470 298 2005 438 240 2006 513 237 2007 458 264 Total 11235 4222
Directive CM & EP Total COM CM 534 17 58 388 22 62 549 25 77 450 24 44 501 27 66 538 22 83 493 24 75 556 35 71 521 29 99 663 52 70 2 823 30 29 6 648 34 28 10 664 39 40 7 822 35 24 9 679 43 31 15 753 60 23 15 724 40 11 10 871 44 18 16 723 44 14 24 753 62 20 16 784 60 19 11 689 54 10 19 769 76 28 19 741 53 6 179 15636 951 1006
CM & EP Total 75 84 102 68 93 105 99 106 128 122 20 79 10 72 17 96 24 83 28 102 19 102 42 93 45 107 39 97 46 128 36 115 27 91 40 144 18 77 411 2368
Regulation COM CM CM & EP 911 426 907 500 1091 550 1121 480 1059 477 961 416 1115 411 1127 336 1137 391 1072 329 992 302 1 2821 267 0 2277 247 4 2405 242 1 2612 230 0 2594 176 14 2573 181 23 2440 140 16 2202 143 25 2052 159 34 1965 177 41 2026 109 19 1804 167 45 1419 122 22 40683 6978 245
Total 1337 1407 1641 1601 1536 1377 1526 1463 1528 1401 1295 3088 2528 2648 2842 2784 2777 2596 2370 2245 2183 2154 2016 1563 47906
COM=Commission legislation, CM=legislation adopted by the Council of Ministers (implying no EP involvement or the application of the cooperation or consultation procedure), CM & EP=legislation adopted by the Council of Ministers and the European Parliament (application of the codecision procedure)
39
Figure 2A:
Figure 2B:
40
Table 2: European Parliament (EP) Involvement, Council Decision Rule and Council Conflict (Adopted Legislation in PreLex) Policy Fields (AP)
No EP No. % Macroeconomics 1844 78.0 Civil Rights 47 79.7 Health 1 25.0 Agriculture 3041 95.2 Labour 74 54.8 Education 19 26.8 Environment 115 39.4 Energy 93 68.4 Transportation 116 42.0 Law 36 73.5 Social Welfare 57 74.0 Banking & Finance 30 22.6 Foreign Trade 2306 83.4 International Affairs 440 77.3 Government Operations 205 68.6 Total 8424 80.8
EP No. 453 12 3 54 59 48 154 34 154 11 17
% 19.2 20.3 75.0 1.7 43.7 67.6 52.7 25.0 55.8 22.4 22.1
Missing No. % 67 2.8 0 0.0 0 0.0 101 3.2 2 1.5 4 5.6 23 7.9 9 6.6 6 2.2 2 4.1 3 3.9
Total No. 2364 59 4 3196 135 71 292 136 276 49 77
QMV No. % 1831 77.5 30 50.8 1 25.0 2585 80.9 85 63.0 54 76.1 187 64.0 65 47.8 174 63.0 31 63.3 35 45.5
Unanimity No. % 533 22.5 29 49.2 3 75.0 611 19.1 50 37.0 17 23.9 105 36.0 71 52.2 102 37.0 18 36.7 42 54.5
Total No. 2364 59 4 3196 135 71 292 136 276 49 77
Item A No. % 2065 87.4 45 76.3 4 100.0 2370 74.2 99 73.3 45 63.4 215 73.6 116 85.3 211 76.4 42 85.7 62 80.5
Item B No. % 179 7.6 2 3.4 0 0.0 714 22.3 26 19.3 16 22.5 51 17.5 14 10.3 46 16.7 6 12.2 12 15.6
Written Procedure No. % 49 2.1 1 1.7 0 0.0 86 2.7 3 2.2 5 7.0 4 1.4 3 2.2 3 1.1 0 0.0 1 1.3
Not mentioned No. % 71 3.0 11 18.6 0 0.0 26 0.8 7 5.2 5 7.0 22 7.5 3 2.2 16 5.8 1 2.0 2 2.6
Total No. 2364 59 4 3196 135 71 292 136 276 49 77
100 75.2 231 8.4
3 2.3 133 76 57.1 227 8.2 2764 2435 88.1
57 329
42.9 133 95 71.4 20 11.9 2764 2489 90.1 68
15.0 6 4.5 2.5 181 6.5
12 9.0 26 0.9
133 2764
42
87
66
11.6 569
4.7
10 1.8
569
7.4
28 9.4 1400 13.4
15.3 569
503
88.4
66 22.1 299 197 65.9 600 5.8 10424 8289 79.5
510
89.6 27
22
3.9
102 34.1 299 233 77.9 18 6.0 32 10.7 16 5.4 2135 20.5 10424 8601 82.5 1199 11.5 396 3.8 228 2.2
299 10424
QMV=Qualified Majority, Simple Majority, and unclear; Item A=Item A on Council Agenda, Item B= Item B on Council Agenda, Written Procedure=Use of Written Procedure in the Council Meeting
41
Table 3: Instrument and Type of Legislation by Policy Fields (1984-2007)
Policy Fields (AP) Macroeconomics (n=10,392) Civil Rights (n=148) Health (n=13) Agriculture (n=44,483) Labour (n=216) Education (n=142) Environment (n=946) Energy (n=220) Transportation (n=638) Law (n=134) Social Welfare (n=136) Banking & Finance (n=225) Foreign Trade (n=4,875) International Affairs (n=1,690) Government Operations (n=1,651) Total (n=65,909)
Type of Legislation Decision Directive Regulation % % % 33.5 8.4 58.1 64.2 12.2 23.6 0.0 100.0 0.0 15.2 1.7 83.0 33.8 43.1 23.1 87.3 0.0 12.7 52.0 20.2 27.8 69.1 15.0 15.9 26.5 30.7 42.8 76.9 1.5 21.6 53.7 9.6 36.8 24.9 62.7 12.4 0.0 44.7 55.3 33.7 0.0 66.3 78.0 1.1 20.9 23.7 3.6 72.7
Type of Legislation according to Document Title COM CM CM & EP Total % % % % 71.5 25.7 2.8 100.0 10.8 81.8 7.4 100.0 23.1 69.2 7.7 100.0 91.6 8.3 0.1 100.0 26.4 59.7 13.9 100.0 28.2 45.1 26.8 100.0 68.2 19.2 12.6 100.0 30.9 56.8 12.3 100.0 49.2 35.9 14.9 100.0 12.7 79.9 7.5 100.0 42.6 44.1 13.2 100.0 29.8 39.6 30.7 100.0 45.9 54.0 0.1 100.0 48.2 50.6 1.2 100.0 21.2 75.9 2.9 100.0 80.2 18.5 1.3 100.0
42
Figure 3:
43
Figure 4:
44
Environ ment
Energy
Transporta tion
Law
Social Welfare
Banking & Finance
Foreign Trade
Internatio nal Affairs
Gov Operation
Total
No. 0 3 2 0 0 2 1 0 0 1 0 0 0 1 2 1 0 0 0 0 0 0 0 0 13
Education
No. 0 0 0 0 0 0 2 0 0 1 1 4 2 1 2 2 6 21 16 20 20 10 19 21 148
Labour
Civil Rights
No. 483 416 494 502 486 474 423 444 413 555 479 366 426 549 427 456 440 501 414 363 381 286 342 272 10392
Agriculture
Macroeco nomics
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Total
Health
Year of Adoption
Appendix 1: Policy Fields over time
No. 1069 1146 1432 1346 1356 1232 1327 1304 1396 1245 1300 2986 2409 2525 2656 2679 2594 2545 2241 2113 2023 2082 1946 1531 44483
No. 7 3 4 7 12 8 12 10 11 9 9 8 8 9 9 12 10 8 11 17 9 10 8 5 216
No. 2 5 2 3 4 6 3 10 11 2 5 6 2 3 6 7 7 1 2 14 11 8 12 10 142
No. 13 18 16 7 13 15 15 19 19 67 35 23 31 43 30 54 56 68 52 60 70 68 82 72 946
No. 11 15 0 4 9 7 9 7 7 7 15 7 21 10 14 14 8 5 7 7 9 4 18 5 220
No. 14 25 14 15 15 20 18 23 40 35 21 15 29 17 31 28 25 28 37 38 38 25 46 41 638
No. 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2 7 12 12 15 14 15 14 15 27 134
No. 4 15 7 4 4 14 9 5 5 4 2 4 3 4 5 4 6 5 3 4 8 3 6 8 136
No. 3 6 5 4 9 10 11 10 10 8 11 7 4 6 9 5 7 13 11 19 13 9 17 18 225
No. 285 188 260 170 166 168 218 229 205 162 179 176 150 195 214 173 207 207 239 241 263 200 215 165 4875
No. 29 15 36 36 19 24 30 31 28 41 57 119 127 117 118 121 111 93 73 114 86 103 87 75 1690
No. 26 24 20 21 37 40 40 33 32 49 83 87 75 73 98 76 105 67 68 102 136 112 116 131 1651
No. 1946 1879 2292 2119 2130 2020 2118 2125 2177 2186 2197 3808 3288 3553 3623 3639 3594 3574 3189 3126 3082 2934 2929 2381 65909 45
3. Measuring the ‘Europeanization’ of Austrian Law-Making: Legal and Contextual Factors Marcelo JENNY (
[email protected] Wolfgang C. MÜLLER (
[email protected]) University of Vienna
1. Introduction In this chapter we subject Austrian legislation to two different approaches of measuring the extent of Europeanization. We begin by outlining the political and legal framework of legal Europeanization in Austria. Next, we address the Europeanization of lawmaking, that is, the impact of EU membership on the national laws passed immediately before and since Austria’s accession. We show the degree of Europeanization of Austrian legislation as identified by (1) manual coding and (2) by automatic coding with a keywords approach. We show how the keyword measure fares when interpreted as a substitute for the former. We then turn to the legislative process. Given that Austria has one of the most Eurosceptical electorates we might expect that the fact of legislation being ‘Europeanized’ would make a difference in this process. Note that the EU membership period was characterized by stiff party competition, including five elections in 15 years (1995–2009), great levels of electoral volatility, the birth and death of political parties with very distinct opinions on the EU, and the change of the basic political orientation of government both in 2000 and 2006
2. What legislation can tell us about 'Europeanization'. Austrian EU membership was not uncontested, and the two-thirds majority in its favour in the 1994 referendum came as something of a surprise. Yet the ‘honeymoon’ between Austrians and the EU was short and today Austria has one of the most Euro-sceptical electorates among the member states. Indeed, the European integration dimension impacts on party politics. At the same time, the Austrian parliament has ensured that its position in the European decision-making process is stronger than that of any other parliament (Bergman 1997; Maurer and Wessels 2001). In practice, however, its direct impact has remained somewhat limited due to the fact that the government–opposition divide has dominated over the legislative–executive divide with the government majorities being able to maintain coherence (Müller 2000, 2001; Müller et al. 2001; Blümel and Neuhold 2001; Hegeland and Neuhold 2002; Neisser 2002;
46
Pollak and Slominski 2003). Over the time period covered in this chapter, two types of coalitions held government office. Initially, a ‘grand coalition’ government of the Social Democrats (SPÖ) and People’s Party (ÖVP), then a coalition of the ÖVP and Freedom Party (FPÖ) in 2000 (continued as an ÖVP-BZÖ cabinet after the FPÖ split in 2005), and finally a coalition of the SPÖ and ÖVP came to office at the beginning of 2007. While the ÖVP and SPÖ engineered EU membership, the FPÖ opposed it and has transformed itself into an increasingly Euro-sceptical party. Given parliament’s unwillingness to make much out of its considerable powers to steer EU policy and the solid pro-EU majority therein, there is not much genuine political incentive for the government to use secondary legislation and thereby bypass parliament. Conversely, the participation in government of Euro-sceptical parties (the FPÖ and its 2005 breakaway, the BZÖ – the Alliance for Austria’s Future) has provided the mainstream parties with an incentive to maintain the transposition of EU rules in the hands of the parliament rather than empowering individual cabinet ministers who may belong to Euro-sceptic parties. The background for this expectation is that neither the Chancellor nor the cabinet have the formal right to give orders to individual ministers, and although the Chancellor (and the minister’s party) can, in principle, exercise political pressure, this is not always feasible. For these reasons we would expect the parliament to play a significant role in the transposition of EU rules. Establishing the ‘Europeanization’ of lawmaking hence will provide us with a valid indicator of Europeanization in the Austrian case.
3. The Austrian legal system and process. With the exception of the budget, all types of draft legislation can be introduced in parliament by either chamber, the cabinet, or the people (i.e., 100,000 citizens supporting a People’s Initiative). In practice, government proposals and private members’ bills from (a minimum of five) Nationalrat MPs prevail.1 The government introduces its agenda to parliament mainly via government bills. Consequently, most Austrian laws originate from government bills. Yet, the government may use private member bills (sponsored by MPs of the government parties) as a ‘fast track’ alternative to government bills. In contrast to the latter private member bills do not require going through the established consultation procedure (with the Länder, interest groups and other stake-holders) in the pre-parliamentary stage. Consensusseeking of the government is another reason why a considerable share of laws is based on private member bills. Then, typically, a policy compromise forged with one or several opposition parties is introduced to parliament as a private member bill, rather than as an amendment to a government bill. The resulting private member bill is signed by MPs from all parties involved in the compromise. Some issues, for instance a reform of the chamber’s rules of procedure, are as a norm introduced as a private member bill.
1
Before 1988 the minimum number of MPs was eight. Bundesrat proposals in parliamentary practice are completely irrelevant. With few exceptions, the people’s initiative (Volksbegehren) has also been irrelevant with regard to actual law-making. However, it has been quite relevant as a means for party competition (Müller 1998, 1999b).
47
Committee bills constitute a quick ‘correction mechanism’ for the government. Deliberations and political negotiations at the committee stage of a government bill on a complex topic may make out that additional changes to the legal order (mostly touching only marginally upon the government bill’s content) are required. These changes then are the rationale of the committee bill. The audit office submits a yearly statement of account on the execution of the national budget. Bills emanating from the second chamber and bills based on people’s initiatives complete the available types of bills. Second chamber bills reflect regional policy wishes that are unable to attain national support. People’s initiatives that have attained 100.000 signatures have to be deliberated in parliament, yet typically do not have much chance of getting passed into law (Müller 1998). Any bill introduced in the Nationalrat is either passed as a law, turned down in a committee or plenary vote, or dies automatically at the end of the legislative period. Note, however, that bills that have died can quickly be resurrected in the next period simply by re-introducing them to parliament. Of course, this makes most sense if the parliamentary majority is interested in passing them. Opposition bills often die quietly in committee. To pass, ordinary legislation requires a quorum of one third of the MPs and a majority of the votes cast (50%+1). A referendum on a law is held before it comes into force if the Nationalrat so decides (Art. 43), which has occurred only once (a referendum on the use of nuclear energy was held in 1978) (Müller 1998). The process of budget making differs from normal law making in that it requires the government to submit a proposal at least ten weeks before the end of the budget year. MPs may propose a budget only if the cabinet fails to do so (Art. 51). In order to be enacted, constitutional amendments require a quorum of 50 per cent of the MPs2 and the support of a two-thirds majority of the votes cast. The same applies to the Nationalrat’s rules of procedure (Art. 30) and, since 1962, to school legislation (Art. 14 and 14a), which do not have constitutional status. Constitutional amendments are subject to a referendum if so demanded by one third of the MPs in either of the two chambers (Art. 43). This has, however, never occurred. In contrast, a referendum is obligatory for a ‘total revision’ of the Constitution (Gesamtänderung)3 (Art. 43). This has happened only once. In 1994 a referendum was held on the constitutional amendments required for Austria’s accession to the European Union. Although Austria has a two-chamber system, which reflects its federal constitution, it is a very asymmetrical one (Kathrein 1986; Schäffer 1999; Fallend 2000a). The upper chamber, the Bundesrat, represents the Länder. Its members are elected by the Land diets. The representation system in the Bundesrat takes into account the size of each Land’s population and party strength. With regard to ordinary legislation passed by the Nationalrat, the Bundesrat can issue only a suspensory veto. The
2
If not otherwise stated, all references to MPs refer to those of in the Nationalrat. The constitution itself does not indicate what constitutes a ‘total revision’. Constitutional lawyers and the Constitutional Court have interpreted this clause as referring to amendments that involve changes in one or more of the guiding principles of the 1920 constitution. These principles, in turn, are not explicitly listed but need to be extracted from the text of the constitution by means of interpretation. They unquestionably include the principles of democracy, federalism, the rule of law, the separation of powers, and liberalism (see Walter 1972: 101–13). 3
48
Nationalrat can override it with a quorum of 50 per cent of the MPs and a majority of votes cast. By vetoing legislation, the Bundesrat can delay it for, at most, eight weeks. With regard to constitutional amendments, the Bundesrat is more powerful for two reasons. First, since 1984, changes in the distribution of competencies between the national level (Bund) and the Länder become effective only if the Bundesrat accepts them with a quorum of 50 per cent and a two-thirds majority (Art. 44). Second, as already mentioned, it can enforce a referendum on any constitutional amendment, if this is demanded by a third of its members. In its own right, however, the Bundesrat cannot block constitutional amendments that do not fall under Article 44. Moreover, there are important decisions that do not involve the Bundesrat at all. In particular, it does not participate in budget-making and other financial acts (Art. 42). The Nationalrat and the Bundesrat are politically largely ‘congruent’ (Lijphart 1999: 207), in the sense that the same parties or coalitions held majorities in both chambers. This is true despite the fact that the representational systems are different enough to have produced different plurality parties in the two chambers (during 19.5 years in the period 1945–2000, including 11.4 years when two parties in the Bundesrat were of equal strength). From 1945 to 2000 the two chambers were politically ‘incongruent’ only for 5.1 years. Nevertheless, the Bundesrat’s limited political relevance is much more a consequence of its lack of formal powers than a result of political congruence. In practice, political parties and nonconstitutional bodies consisting of Land government representatives are much more important national channels of political influence for the Länder than is the Bundesrat (Luther 1986, 1997; Weber 1992; Fallend 2006). Constitutional and ordinary laws do not exhaust the universe of universally-binding rules. At the lowest level of such rules in Austria is the government decree (Verordnung). Such decrees are issued by individual ministers or, in rare cases, by the cabinet. Some of them require the consent of Parliament’s main committee. According to the Constitution (Art. 18), the content of a government decree must, as a rule, already be determined by law. However, the decree may reduce the degrees of freedom contained in the law. It is also possible to introduce new rules (which add to those already contained in the law) by decree, however, only if this is specifically authorised by constitutional law (Koja 1986: 141–64).
Federalism Austria is a federal state, comprising nine Länder. This chapter is concerned exclusively with rule-making at the federal level. Yet, federalism should have an impact on our object of study. Everything else being equal, federal countries should have fewer rules at the federal level than unitary states and fewer EU rules should affect the federation’s jurisdiction. Provided the tasks delegated to the federation are those that can be better served by centralization, we should also expect the federal level to be more affected by Europeanization than the state (Land) level. Delegating the tasks of the federation to the EU would only equate to carrying the economy-of-scale idea underlying centralization one step further. Note, however, that the optimal level of centralization may not always be the highest one. Moreover, concern for national sovereignty should weigh more heavily than concern for regional autonomy. This would suggest that
49
member states should be more ready to transfer sub-national competencies than those of the central government. Yet, what is under central or state jurisdiction differs between member states and shifting power to the EU level is a complex game involving many actors. Notwithstanding these caveats, we expect the federal level to be more affected by legal Europeanization than the state level. The Austrian Constitution has reserved the power to decide on all matters not explicitly vested in the federation (Bund) for the Länder. Nevertheless, their jurisdiction is extremely limited. The legislative power of the Länder is unconstrained with regard to area planning and related issues, tourism, nature protection, agriculture and forestry, hunting and fishing, sports, youth protection, and pre-school education. With regard to domains that include public welfare, hospitals, land reform, and electricity production and supply, the Länder have the power to fill in details on the basis of national framework legislation (Fallend 2006: 1029). The remainder falls under the exclusive jurisdiction of the federation. As the EU has some competencies that fall under the jurisdiction of the Länder domestically, Europeanization of legislation partly affacts Land legislation. Yet, given the overall modest nature of Land jurisdictions, this does not have a major impact on our results. Indeed, the few empirical studies returned an average of about 7 EU-related Land laws per year in the membership period (Bussjäger and Larch 2004; Jenny and Müller 2010).
Legal tradition Austria is a civil law country: legislation is the primary source of law and courts decide cases by identifying the general principles underlying the specific case (rather than drawing on preceding cases, as in the case law tradition). Civil law countries, however, do not form a homogeneous group but fall into three distinct categories: the French, German, and Scandinavian traditions (Merryman 1969; Hausmaninger 2003). Austria belongs to the German tradition and indeed has contributed much to the legal positivism prevalent with the writings and political impact of Hans Kelsen, the main author of its 1920 constitution (Kelsen 1997). Article 18 of the Constitution requires that all acts of the public administration be based on law. In simple terms, the legality principle of Article 18 means that all relevant contents of regulations need to take the form of laws. The addressee of legal norms must be able to learn everything that is important to their position under the law from parliamentary legislation. Likewise, the law courts must be able to check administrative acts against the law (Antoniolli and Koja 1986: 215–9, 141–64; Adamovich et al. 1998: 114). This leaves little room for delegated legislation for the purpose of introducing new norms. The major exceptions to this rule are specific empowerments of the administration by constitutional law (Antoniolli and Koja 1986: 153). However, enacting constitutional law requires a two-thirds majority in parliament and hence depends on building broad political consensus.
50
While other civil law countries would refer to the same principles, nowhere do they seem to have been taken as literally as in the Austro-German tradition. This may be largely due to the fact that constitutional review in its European variety was the invention of the Austrian Hans Kelsen. In any case, traditionally, the Austrian Constitutional Court has enforced Article 18 rigorously. Over the post-war period, the Court’s rulings and the training of law students who fill the ranks of the bureaucracy in strict legal positivism have certainly helped to largely uphold this intention of the Constitution. While the above remains largely true, in recent years the Constitutional Court has become slightly more permissive with regard to the interpretation of Article 18 (Michalitsch 1989). It has accepted legislation that contains ‘final’ goals to be achieved by administrative action rather than detailed prescription of administrative action. However, such rulings indicate a mere chance but no guarantee that such legislation will pass the court’s test. Indeed, constitutional lawyers agree that it is hard to predict the outcome of such court cases. The Constitutional Convention, aimed at a fundamental overhaul of the Austrian constitution and held between 2003 and 2005, made Article 18 a prominent agenda item. The government’s proposal was leaving the decision to parliament whether or not the laws themselves would incorporate all relevant details or whether the administration would be empowered to fill in substance by issuing decrees. The transposition of extremely detailed EU rules figured prominently in making the case for a greater amount of delegation to the executive. Nevertheless, the opposition considered this proposal to be a carte blanche for government by decree and consequently rejected it. This proposal would, in fact, have left it up to the then major government party, the pro-EU ÖVP, whether to transpose by means of delegated legislation or detailed parliamentary legislation (in a pro-EU parliament). The ÖVP probably would have opted for government transpositions with one of its ministers in charge and for parliamentary legislation in the case of legislation falling in the jurisdiction of one of the ministers from its Euro-sceptic coalition partner. The Constitutional Convention was also concerned with the government proposal of exempting at least those directives that already had been scrutinized by parliament in the ‘upstream’ process from the need for parliamentary legislation. Again, this idea was rejected (Constitutional Convention 2005: 78–9). Thus, legal positivism and conditional (rather than final) steering remain the dominant features of legislation, while the leeway for delegated legislation remains narrowly constrained and not always predictable when rules are put to a test by the Constitutional Court. Delegated legislation comes in the form of government decrees (Verordnungen). The relevant law spells out which is the body in charge of issuing decrees. Decree power mostly falls in the jurisdiction of individual government ministries, but the requirement of two or more government departments having to agree occurs relatively frequently. Occasionally, the entire cabinet is the body entrusted with issuing government decrees. Nevertheless, there is no hierarchy between the government decrees issued by the cabinet and those of individual ministers. This mirrors the lack of hierarchy between the various forms of constitutional law: Constitutional law comprises the Constitution in the narrow sense (the BundesVerfassungsgesetz), separate constitutional laws, and constitutional clauses in ordinary legislation. All
51
have the same status. Constitutional principles enjoy a superior status. They can only be derived from constitutional law by interpretation. . Over the decades, Article 18 has helped weave a dense web of detailed laws. This pattern of regulation in turn requires that a considerable part of new legislation must be devoted to the more or less regular ‘updating’ of existing laws. This includes minute textual revisions that do not exhibit any policy change. Technically, legal change often takes the form of omnibus laws that list the revisions of a number of existing laws in response to some common cause. The features of Austria’s legal tradition outlined here suggest that much transposition will take the form of parliamentary legislation. The complexity of legislation certainly constitutes a challenge to proxying measures of Europeanization.
4. The formal transposition process The extensive participation rights and special procedures that make the Austrian parliament formally the most powerful in EU affairs of all member states, relate to the upstream process of making rules in the EU (that is, to the period before the EU bodies make their decision). In contrast, the downstream process – the transposition of EU rules into national legislation – has not led to formal institutional adaptation (except for intra-government centralization in 2003; see further discussion below). The legal requirements for the transposition of EU directives into Austrian legislation are largely identical to national regulations that would carry the same regulatory content. The major exception is the fact that the rule-making process cannot end with a non-decision (that is, the decision to maintain the status quo) without violating EU obligations. As our review of the law-making rules has made clear, there are three possible ways to transpose EU directives: by constitutional laws, ordinary laws, and government decrees (Verordnungen). The type of legal instrument required is decided on a case-by-case basis. According to the Constitution, the choice of the legal instrument should be exactly the same for EU-induced regulation as for regulation exclusively triggered by domestic concerns. Hence a new directive may require the enactment of a new law. This is certainly the case if the subject matter addressed by the relevant European directive is at that time unregulated nationally. If there is already a specific law, several alternatives exist. First, the law (or laws plus further decree regulation) may already include the regulatory content of the European directive. In such cases, no further domestic action is required (though the relevant laws and decrees need to be notified to the European Commission and a technical reference to the EU directive needs to be incorporated in the already existing domestic legal acts). Second, if the law or even the Constitution contain regulations that conflict with those of the EU directive, amendments of the laws and/or the Constitution are required. Third, if a law exists that neither conflicts with, nor exhausts the relevant EU directive regulation by decree may be sufficient for
52
transposing the directive’s uncovered contents. For that purpose, it is required that the law provides for decree regulation and empowers the relevant minister. If such empowerment is not already contained in the law, it needs to be introduced in order to ensure that the executive can decide on the substantive issues. Wherever frequent technical changes in EU directives are anticipated (mostly in their appendices), the general strategy of the government is to empower the relevant minister so that future (technical) changes in the European rules can be transposed by the means of government decrees. The national division of tasks, as laid down in the Constitution and the Federal Ministries Law, establishes responsibilities for the implementation of EU directives. Experience has taught that being obliged to transpose newly enacted EU law is not sufficient to make individual ministries or Länder act. Hence, the 2003 amendment to the Federal Ministries Law (introduced in the context of the formation of the second ÖVP-FPÖ cabinet, Schüssel II), has introduced a new coordination competence for the Federal Chancellery. Based on this formal task, the Chancellery now takes an active part in establishing the domestic responsibilities for transposition and for monitoring that process. Austria has been a full EU member since 1995. Yet, Europeanization of legislation does not begin with formal membership. Indeed, formal membership requires that most adaptations to the acquis communautaire have already been made. We therefore begin our empirical enquiry of EU effects in 1992. This was the year when the EEA Treaty was ratified and large-scale adaptation to EU norms began at full steam. Note that these adaptations were made under special rules that were in force for the period when Austria participated in the EEA but was not yet a full EU member. The EEA Treaty was signed in May 1992 and ratified by the Austrian Parliament in September 1992. EEA accession had originally been planned for January 1993, but did not occur until 1994. Membership in the EEA not only required the ratification of the EEA Treaty but also the transposition of the relevant body of existing EU law (that is, the complete first pillar of EU legislation dealing with the ‘four freedoms’ – the free movement of goods, persons, services, and capital). According to an estimate by the head of the Federal Chancellery’s Constitutional Service department, about 60 per cent of secondary EU law had to be implemented at that stage (Okresek 1998). Generally, national laws were seen as the appropriate means for the transposition of EU directives. Nevertheless, the same constitutional law (Bundesgesetzblatt 115/1993) that subscribed to the principle of using parliamentary legislation opened a window for the use of delegated legislation. Provided that the newly created EEA Joint Committee had adopted EU directives and that these directives were sufficiently detailed (hence matching what was expected from a national law), the federal and Land governments were empowered to transpose these rules (Hummer 1994). Thus, in the EEA period before full membership, Austria had more permissive rules for transposition than during the membership period itself.
53
5. Measuring Europeanization by the Study of Lawmaking A growing literature provides some overall assessments on policy-making (for example, Falkner and Müller 1998; Falkner 2000, 2006), the national coordination of taking positions in EU bodies (Müller 2000, 2001, 2002), parliamentary scrutiny of EU affairs (for example, Blümel and Neuhold 2001; Müller et al. 2001; Pollak and Slominski 2003), and specific policy consequences for various sectors. Notwithstanding the importance of these contributions, there is no natural metric for their integration that provides us with a comprehensive account of Austria’s Europeanization. Hence, overall assessments of the EU’s impact on Austria often take the Delors’ estimate as their point of departure. Hearsay, however, has tended to inflate the expectations. While Delors was careful to limit his claim to economic legislation (with the possible extension to fiscal and social legislation), the popular reading is just ‘legislation’. Interestingly, this is also reflected in the perception of Austrian political practitioners. According to the long-serving chairman of the finance committee of the Austrian parliament Ewald Nowotny (1998), about 70 per cent of all laws passed by the Austrian Parliament are either directly or indirectly the implementation of or adaptation to EU directives (see also Wohnout 1999). Likewise, in personal interviews conducted in 1997–98, several other Austrian MPs, in their responses to open-ended questions, have estimated shares of up to 70 per cent (Müller et al. 2001: 479). MPs interviewed in 2005 provided similar estimates. However, not all politicians share such extensive interpretations of Europeanization. The then president of the Austrian Parliament, Heinz Fischer, in an interview with Austrian television (‘Hohes Haus’ in 2002), explicitly rejected these estimates. Without providing precise figures, his own estimate was a share of EU-induced legislation of well below 50 per cent. As the other chapters in this volume show, such ‘guesswork’, educated as it may be, is by no means confined to Austria. And it cannot substitute for empirical research. Thus, we really know very little about the extent to which the EU impacts the rules governing its member states. This is a highly relevant political question that, in turn, relates to issues such as the current degree of policy coordination, the potential for further integration, the popular response to ‘Europe’, and the relevant party political strategies. Although it is a simple question, providing answers requires both methodological rigour and painstaking empirical research. The different estimates made by the insiders referred to above need not necessarily be wrong. Indeed, they may simply reflect the time of making these estimates, the ups and downs in the regulatory output of the European Union, and the areas of special expertise of practitioners. It is well known that the EU produced fewer directives for some years after the completion of the Single Market programme (Pollack 2000: 529–37; Dehousse 2002) and that the number of directives increased again in the 1990s (see Figure 1). Moreover, no attempt has been made to properly operationalize the term ‘EU-induced’. Is a domestic law EU-induced only when it is exclusively devoted to the implementation of EU directives? Or, conversely, is a law EU-induced if it has any function of adaptation to EU rules? Obviously, such definitional issues can account for great variation in the number of EU-induced laws at the national level.
54
6. Measuring Legal Europeanization. In this chapter we employ two different measurement approaches for identifying EU-related legal norms. The first measure that establishes the existence of a relation of domestic legislation to EU law is based on traditional, ‘manual’ extracting of information from a set of different sources that indicate such relationships on a one-to-one basis. The second measure is based on automatic search of the bills’ texts and accompanying parliamentary documents with a set of EU-related keywords. We then take search ‘hits’ as evidence that a bill is related to EU law. We report the search results for the subset of successful bills (i.e. those that turned into laws). Thus we will be able to compare the results of the two methods applied to federal laws. Manual coding The first measurement approach collects information from several distinct sources. The combined evidence is used to code each law or decree as EU-related or not. We have drawn on the same multisource measurement approach used in a previous publication (Jenny and Müller 2010). For each of our EU-related national norms one or more of the following characteristics need to apply in order to code a national law as EU-related: a specific reference to EU rules in the introductory or final clause of the law; a reference to the Eur-Lex (formerly CELEX) number in the head information of the law when published in the official law gazette a reference to the need to conform to EU rules in the official materials accompanying government bills when they are introduced to parliament (as suggested by Page 1998, pp. 804–5); information extracted from the parliamentary debates (particularly relevant for identifying relations between national laws and EU primary law) and national authorities’ data indicating that the respective law implements EU rules and has been reported to the Commission; the transposition data from the Eur-Lex database (Sector 7) provided by the EU Commission.
For various reasons, no single source is perfect. Nor are the individual sources equally reliable for each period. The data sources 5 and 6, for example, should correspond like flipsides of a coin as only the keeper of the records differs. However, national records were quite unsystematic in the early years of membership. Reporting on implementation measures was largely left to the individual government department. This allowed different ministries to adopt different strategies, leading to inconsistencies in the reporting. This, in turn, caused some problems with regard to Austria’s early transposition record. In 2003, a change to the jurisdictions of the various government departments put the Federal Chancellery in
55
charge of coordinating the implementation of EU directives and of the reporting of the implementation measures (Bundesgesetzblatt 17/2003).4 Automatic computer-assisted coding The second measurement approach uses a computer-based search of the digital texts of the bills and their accompanying material. The digital texts are available through the national parliament’s website from the 20th legislative period (January 1996) onwards. The search terms were the list of keywords provided by the project group but we added some word flexions to them (see Table 1).
4
In previous work (Jenny and Müller 2010) we covered the 1992–July 2003 period and employed six indicators. When finishing the present chapter we still lack an update on the national authorities’ data. Therefore our count for the period August 2003–2007is based on a total of five (rather than six) indicators. This may produce a minor undercount of EU-related laws in the more recent period. However, each new indicator produces diminishing returns in terms of additionally identified EU links. Therefore we do not consider the data quality to differ significantly between the two periods.
56
Table 1. German keywords and abbreviations used as search terms EU keywords (German)
Abbreviation
EU keywords (English)
Europäische/r/n Union
EU
European Union
Europäische/r/n Gemeinschaft
EG
European Community
Europäische/r/n Wirtschaftsgemeinschaft
EWG
European
Economic
Community Einheitliche/r/n Markt
Common Market
Binnenmarkt
Single Market
Europäische/r/n Markt
European Market
Europäische/r/n Gemeinschaft für Kohle und Stahl/ EGKS
European Coal and Steel
Montanunion
Community
Europäische/r/n Atomgemeinschaft
EURATOM/
European Atomic Energy
EAG
Community
Europäische/r/n Wirtschafts- und Währungsunion / EWWU/ EWU
European monetary Union
Europäische/r/n Währungsunion Europäische/r/n Währungsystem
EWS
European
monetary
system Richtlinie(n)
Directive
Gemeinschaftsrecht/s
Community law
Europarecht/s
European law
57
7. Results. Let us first take a look at our universe of legislation. We begin by plotting the annual production of federal laws (as published in the official law gazette, the Bundesgesetzblatt) from 1986 to 2007 (Figure 1). During that period a total of 3001 laws were published. 68 per cent originated from government bills, 24 per cent from private member bills, and 7 per cent from committee bills. The annual audit office bills amount to about one per cent of the laws.
Figure 1. Laws by type of bill (1986–2007)
Europeanization of laws Figure 2 provides a long-term view on rule production at the national and EU levels from 1945 to 2007. It plots the annual law production and the number of European directives passed per year since the Treaty of Rome. Although there are ups and downs, overall law production at the European level by way of EU directives is still growing. In contrast, national law production looks rather stable. The federal laws’ alltime high of 191 (in 1993) fell in the busy period of preparations for EU membership.
58
Figure 2. National and European law production (1945–2007)
The number of national laws passed per year has declined slightly over the EU membership period (since 1995). Yet, this decline means a return to the national long-term level of output after the extraordinary legislative activity in the years immediately before and after EU accession rather than resulting from the transfer of national policy competences to the European level. There is a second domestic reason: three coalition governments in office in the membership period broke down only about a year in the legislative term. They triggered early elections that were followed by complicated coalition negotiations. We know that legislative output tends to be lower under such conditions. Figure 3 provides the share of national laws related to EU laws from the 1990s onwards. As already mentioned, data from August 2003 may slightly ‘undercount’ relations to EU rules.
59
Figure 3. Laws passed and Europeanized (1992–2007)
The highest share of EU-related laws was 44.5 per cent in 1993. This is far below the 80 per cent level generally associated with Delors and the related estimates of most politicians in Austria. Yet, 1993 was an extraordinary year – the year with the most legislative acts in modern Austrian history. Over the entire period, every fourth national law is related to EU law. The mean share of EU-related federal laws is 26.4 per cent for the period 1992–2007, and 24.5 per cent for the membership years 1995–2007.
Europeanization of law-making measured by keyword search Next we measure the extent of Europeanization of successful legislative proposals by the computer-based keyword search introduced above. For that purpose we draw on the texts of all bills and the materials supplementing them. There is one exception: We exclude audit office bills that have no policy content and are completely ‘retrospective’. Our electronically readable text collection covers 99.7 per cent of all bills (not counting audit office bills) in the four legislative periods between 1996 and 2008. Less than a dozen bills were excluded, due to faulty hyperlinks on the parliament’s website or unreadable files. The amount of supplementary materials provided by the bill’s initiators varies considerably by type of bill. Government bills are usually the best-documented ones as they have to be accompanied by introductory comments from the initiating ministry. The ‘explicating remarks’ that accompany government bills include a section on the bill’s relation to EU law. If a bill is entirely nationally induced, this section will either say so or state that the bill’s content conforms to EU law. Initial keyword searches repeatedly turned up bills that on closer inspection were not EU-related in any substantive way. To get rid
60
of these false ‘hits’ we deleted this text section from each bill’s text base prior to running keyword searches. The initial set of keywords employed in our search is common to all chapters in the book (Table 1). “Directive(s)” turned out to be by far the most frequent keyword found in Austrian bills and their accompanying texts. Yet, its German translation (“Richtlinie”) means “guideline” and is a word also in frequent use outside the realm of EU law. It refers to political goals and to specific policy or administrative guidelines. Due to its non-discriminating nature we excluded the keyword “directive(s)” from the search procedure. If any of the remaining ‘EU’ keywords turned up at least once, a bill was coded as EU-related. The automatic, computer-assisted coding procedure returns an extremely high proportion of bills which apparently have a European link. Compared with manual coding the share of Europeanized laws is much higher. Table 2 presents the discrepancy between the two methods. Table 2. Manual and automatic coding of EU-related laws compared (1996–2008) Type of bill
Manual coding
Automatic coding
Government bills
31
79
Private members’ bills
12
30
5
23
25
63
Committee bills All laws
Note: The time period covered is Jan 15th 1996–Oct 27th 2008. Automatic coding identifies 63 per cent of the Austrian laws as EU-related while manual coding gives a proportion of 25 per cent. The largest difference is among laws based on government bills (79 to 31 per cent) but private member’s bills (30 to 12 per cent) and committee bills (23 to 5 per cent) also show significantly higher proportions of ‘Europeanization’ when identified by automatic keyword search. Given the many indicators employed in the manual search we do not accept these results as answers to our substantive research question. Rather, they call for further examination. The two measurement approaches clearly differ in the sensitivity with which they pick up a linkage between a national law and Europe. The keyword search, at least as we have conducted it here – conforming to the common standards of our cross-national research cooperation –, is hyper-sensitive. A closer inspection of a small sample of laws that have been identified in the automatic coding but not in the manual coding as being connected to ‘Europe’ show that there is indeed no connection to legal documents at the EU level. Rather these documents make reference to Europe as the wider context of regulation.
61
Often legislative initiators use the European yardstick to claim necessity for their proposal (e.g. ‘Austria is falling behind’) or praise its merits by comparing it to the rules in force in other EU member states. There are two ways to proceed from here. One is to stick to the original aim of our enterprise: the study of legal Europeanization. Then we need to conclude that one single keyword hit with the above list of terms is not a good strategy to establish that kind of link. The logical next step would be the development of more robust measures of legal Europeanization via keywords. Such a research strategy could include using a larger minimum number of keywords that need to appear in the same document as a condition for classifying it as EU norms-related. Whether this research strategy can be developed into a robust measure, however, remains an open question. Methodologically, our research demonstrates that any such measure needs cross-validation with more conventional ways of identifying legal influences, at last for a subset of the universe of national laws. The other strategy would be to stick to the data generated by the keyword search and to make sense out of it in terms of interpretation. Clearly, ‘yardstick Europeanization’, as we have found it (see above), is an interesting phenomenon. Indeed, ‘Europeanization’ has many faces (Mair 2004; Olsen 2002) and the keyword search seems to speak to many of them. We would not have expected so much ‘Europe’ to have found its way into the law book and the documents accompanying the legislative process and 10 years earlier this may indeed have been the case. Our keyword count thus may be valuable in its own right as an umbrella measure for all (or many) kinds of Europeanization that affects national politics and law making in particular. Yet, we are quick to add a caveat: It might be an interesting indicator for a limited time period only. Once ‘Europe’ is so omnipresent that it finds its way into each of the documents covered here, the indicator will be of limited value only. In contrast, the hard indicator measuring (rather than assuming) a legal connection between domestic laws and European norms will remain in more narrow confines and hence continue to show a sharper picture that allows distinguishing facts from political rhetoric.
8. Europeanization of laws by policy area This section describes the extent of legal Europeanization by policy area. We use the policy areas classification provided by the Comparative Agendas Project and our manual coding procedure for identifying EU-related national laws. Table 3 gives the number of laws assigned to the 20 different policy areas that make up this scheme. As expected, there is a very uneven distribution of laws by policy area. The by far largest number of laws falls into to the policy area ‘domestic macroeconomy’. The policy field accounts for almost one quarter of the laws passed in the 15 years between 1992 and 2007. The second and third largest policy areas in terms of legislative output are domestic affairs (‘law, crime and family issues’) (10.9 per cent) and public administration (8.3 per cent). None of them ranks particularly high with regard to Europeanization. ‘Domestic macroeconomy’ has a share of 17.6 per cent Europeanized laws, ‘law, crime and family issues’ a share of 26.3 per cent, and ‘public administration’ 25.8 per cent.
62
Energy stands out as the most Europeanized policy field with a whopping 78.6 per cent of national laws exhibiting a link to EU law. At the same time, energy is one of the policy fields with least output in terms of legislation. On average, just one law is passed per year.
Table 3. Europeanization laws by policy area (1992–2007)
63
Policy Area
Federal Laws Europeanized
%
passed
federal laws
federal laws
passed
Energy
14
11
78.6
0.6
Environment
67
32
47.8
3.0
Transport
122
56
45.9
5.5
Public Lands & Water
11
5
45.5
0.5
Agriculture
66
29
43.9
3.0
Housing
41
17
41.5
1.8
Civil Rights & Liberties
5
2
40.0
0.2
Health
139
55
39.6
6.2
Foreign Policy
120
40
33.3
5.4
Labour, Immigration
130
38
29.2
5.8
Family 243
64
26.3
10.9
Law,
Crime,
Europeanized % of all laws
issues Public Administration
186
48
25.8
8.3
Banking & Finance
85
21
24.7
3.8
28
24.6
5.1
Science, Technology & 114 Comm. Foreign Trade
13
3
23.1
0.6
Domestic
541
95
17.6
24.2
169
28
16.6
7.6
2
10.5
0.8
Macroeconomy Education Culture
& 19
Entertainment Social Welfare
124
8
6.5
5.5
Defence
27
0
-
1.2
64
Total
2236
582
26.0
100
Note: Laws were classified into Comparative Agendas Project scheme using the individual bill’s parliamentary committee assignment and its official keywords. A very small number of bills could not be classified.
Figure 4. Laws passed and Europeanized by policy area (1992–2007)
Seven policy areas have shares of Europeanized laws ranging between 40 to almost 50 per cent (see Figure 4). Among these health and transport are intermediate sized policy areas in terms of national legislative output. Their share of total legislative output is 6.2 and 5.5 per cent, respectively. Environment and agriculture are prominent policy fields that were transferred more or less from national to European Union jurisdiction long ago. About every second national law passed in Austria from 1992 to 2007 in these two policy fields had a European link. However, environment and agriculture accounted for only three per cent of national law production each. The least Europeanized policy areas in Austria were defence (0 per cent) – not even containing a single Europeanized law –, social welfare (6.5 per cent), and culture & entertainment (10.5 per cent). Recall that overall the 20 large and small policy fields with their varying shares of Europeanized laws contain a share of slightly more than one quarter of the Austrian national law production related to EU law.
65
9. Conclusion In this chapter we have outlined the political and legal context of Europeanization of legislation in Austria and described the formal transposition process. We then turned to quantify the impact of the EU on national legislation by employing two different measurement approaches: manual coding based on multiple sources and automatic coding based on multiple keywords. The measurement approach using manual coding identifies about one quarter of the national law as Europeanized whereas the corresponding figure for the automatic coding in its internationally coordinated form is 63 per cent. The latter number, clearly, is inflated. Yet, it is based on an internationally agreed list of keywords that prima facie seem to make sense. To be sure, no keyword search will ever produce perfect results. To pass the test of usefulness a keywords-based indicator needs to satisfy two criteria: (1) it has to produce a number of cases reasonably similar to that obtained by more accurate (but also more expensive and time-consuming) hand-coding, (2) there must be a reasonable amount of matching of individual cases identified by the two methods. Obviously, except the most extreme distributions, meeting the first criterion is not a sufficient guarantee that the search has produced substantively meaningful results (i.e. a sufficient matching of individual cases). Yet, not-meeting the first criterion is sufficient to invalidate the results of a keyword search. This is what happened in the Austrian case. Obviously ‘contextual’ Europeanization has become an important feature of the legislative process that overlays legal Europeanization. We nevertheless consider our keyword search a valuable exercise, independent from the methodological lesson that it has told. In our understanding it results in a more general indicator of contextual Europeanization that can be meaningfully interpreted and compared to the indicator of legal Europeanization that we have cumbersomely generated from legislative process data. Unfortunately, keyword search is unlikely to do the latter job for us in the near future with a sufficient degree of accuracy. The history of science abounds of important discoveries that were not intended originally. While making such a claim would be much too strong for ‘contextual’ Europeanization, we nevertheless consider the resulting indicator a nice by-product of our methods-driven exercise.
66
References Adamovich, Ludwig K., Bernd-Christian Funk and Gerhart Holzinger (1998). Österreichisches Staatsrecht. Vol. 2, Staatliche Organisation. Vienna: Springer. Antoniolli, Walter and Friedrich Koja (1986). Allgemeines Verwaltungsrecht. Vienna: Manz. Bergman, Torbjörn (1997). ‘National Parliaments and EU Affairs Committees: Notes On Empirical Variation and Competing Explanations’, Journal of European Public Policy, 4(3): 373–378. Blümel, Barbara and Christine Neuhold (2001). ‘The Parliament of Austria: A Large Potential with Little Implications’, in Andreas Maurer and Wolfgang Wessels (eds.), National Parliaments on their Ways to Europe: Losers or Latecomers? Baden-Baden: Nomos, pp. 313–336. Bussjäger, Peter and Daniela Larch (2004). Landesgesetzgebung und Europäisierung. Innsbruck: Institut für Föderalismusforschung (Föderalismusdokumente 16). Constitutional Convention (2005). ‘Endbericht Österreich Konvent, No. 3’. Vienna: Parlament. Dehousse, Renaud (2002). ‘Misfits: EU Law and the Transformation of European Governance.’ New York School of Law, Jean Monnet Working Paper 2/02. Falkner, Gerda (2000). ‘How Persuasive are Euro-Politics? Effects of EU Membership on a New Member State’, Journal of Common Market Studies, 38(2): 223–250. Falkner, Gerda (2006). ‘Zur “Europäisierung” des österreichischen politischen Systems’, in Herbert Dachs, Peter Gerlich, Herbert Gottweis, Helmut Kramer, Volkmar Lauber, Wolfgang C. Müller and Emmerich Tàlos (eds), Politik in Österreich: Das Handbuch. Vienna: Manz, pp. 82–94. Falkner, Gerda and Wolfgang C. Müller (eds.) (1998). Österreich im europäischen Mehrebenensystem. Vienna: Signum. Fallend, Franz (2000a). ‘Der Bundesrat in Österreich’, in Gisela Riescher, Sabine Ruß, and Christoph M. Haas (eds.), Zweite Kammern. München: Oldenbourg, pp. 97–113. Fallend, Franz (2006). ‘Bund–Länder-Beziehungen’, in Herbert Dachs, Peter Gerlich, Herbert Gottweis, Helmut Kramer, Volkmar Lauber, Wolfgang C. Müller and Emmerich Tàlos (eds.), Politik in Österreich: Das Handbuch. Vienna: Manz, pp. 1024–1040. Hausmaninger, Herbert (2003). The Austrian legal system, 3rd edn. Vienna: Manz.
67
Hegeland, Hans and Christine Neuhold (2002). ‘Parliamentary Participation in EU Affairs in Austria, Finland and Sweden: Newcomers with Different Approaches.’ European Integration online papers 6, No. 10. Hummer, Waldemar (1994). ‘Demokratiedefizit im EWR? Mitwirkung der Legislative an Rechtsetzung und Vollzug von EWR-Recht in der österreichischen Rechtsordnung.’ Jahrbuch des österreichischen Parlaments 1994: 216–237. Jenny, Marcelo and Wolfgang C. Müller (2010). ‘From the Europeanization of Law-Making to the Europeanization of National Legal Orders: The Case of Austria.’ Public Administration, 88(1): 36–56. Kathrein, Irmgard (1986). ‘Der Bundesrat’, in Herbert Schambeck (ed.), Österreichs Parlamentarismus. Berlin: Duncker & Humblot, 337–401. Kelsen, Hans (1997). Introduction to the Problems of Legal Theory. A Translation of the First Edition of the Reine Rechtslehre or Pure Theory of Law (Translated by B.L. Paulson and S.L. Paulson). Oxford: Oxford University Press. Lijphart, Arend (1999). Patterns of Democracy: Government Forms and Performance in Thirty-six Countries. New Haven: Yale University Press. Mair, Peter (2004). ‘The Europeanization Dimension.’ Journal of European Public Policy, 11(2): 337– 348. Maurer, Andreas and Wolfgang Wessels (eds) (2001). National Parliaments on their Ways to Europe: Losers or Latecomers? Baden-Baden: Nomos. Merryman, J. H. (1969). The Civil Law Tradition: An Introduction to the Legal Systems of Western Europe and Latin America. Stanford, CA: Stanford University Press. Michalitsch, Martin (1989). ‘Die geänderte, realverfassungsändernde Rechtssprechung des Verfasssungsgerichtshofes.’ Österreichisches Jahrbuch für Politik 1989: 197–207. Müller, Wolfgang C. (1998). ‘Plebiscitary Politics and Party Competition in Austria.’ Electoral Studies, 17: 21–43. Müller, Wolfgang C. (1999). ‘Plebiscitary Agenda-Setting and Party Strategies: Theoretical Considerations and Evidence from Austria.’ Party Politics, 5: 305–317. Müller, Wolfgang C. (2000). ‘EU Co-ordination in Austria: Challenges and Responses’, in Kassim Hussein, B. Guy Peters and Vincent Wright (eds.), EU Policy Co-ordination: The National Dimension. Oxford: Oxford University Press, pp. 201–218.
68
Müller, Wolfgang C. (2001). ‘Ministerial Government at the European Level: the Case of Austria’, in Hussein Kassim, Anand Menon, B. Guy Peters, and Vincent Wright (eds.), The National Coordination of EU Policy: The European Level. Oxford: Oxford University Press, pp. 229–255. Müller, Wolfgang C. (2002). ‘EU-Mitgliedschaft und Regierungshandeln’, in H. Neisser and S. Puntscher-Riekmann (eds.), Europäisierung der österreichischen Politik. Vienna: WUV Universitätsverlag, pp. 101–132. Müller, Wolfgang C., Marcelo Jenny, Barbara Steininger, et al. (2001). Die österreichischen Abgeordneten. Vienna: WUV-Fakultas. Neisser, Heinrich (2002). ‘Die Funktionen des Parlaments im Wandel’, in Heinrich Neisser and Sonja Puntscher Riekmann (eds.), Europäisierung der österreichischen Politik. Vienna: WUV Universitätsverlag, pp. 133–152. Nowotny, Ewald (1998). ‘Europa und wir.’ Wiener Journal, September 1998, 10–11. Okresek, Wolf (1998). ‘Schwierigkeiten bei der Anpassung der österreichischen Rechtsordnung an das Rechtssystem der EU’, Lecture given at the 1998 convention of the European Community Studies Association Austria. Vienna: Wirtschaftsuniversität Wien. Olsen, Johan P. (2002). ‘Many Faces of Europeanization.’ Journal of Common Market Studies, 40(5): 921–952. Page, Edward C. (1998). ‘The Impact of European Legislation on British Public Policy-making: A Research Note.’ Public Administration, 76(4): 803–809. Pollack, Martin (2000). ‘The End of Creeping Competence? EU Policy-Making Since Maastricht’, Journal of Common Market Studies, 38(3): 519–38. Pollak, Johannes and Peter Slominiski (2003). ‘Influencing EU Politics? The Case of the Austrian Parliament’, Journal of Common Market Studies, 41(4): 707–729. Schäffer, Heinz (1999). ‘The Austrian Bundesrat’, in Ulrich Karpen (ed.), Role and Function of the Second Chamber. Baden-Baden: Nomos, pp. 25–55. Schambeck, Herbert (1997). Regierung und Kontrolle in Österreich. Berlin: Duncker & Humblot. Scharpf, Fritz W. (1999). Governing in Europe: Effective and Democratic? Oxford: Oxford University Press. Schefbeck, Günther (1997). ‘Der Nationalrat zwischen Rechtssetzungsorgan und Tribüne.’ Journal für Rechtspolitik, 5(1): 117–135.
69
Sickinger, Hubert (2000). ‘Die Funktion der Nationalratsausschüsse im Prozess der Bundesgesetzgebung.’ Österreichische Zeitschrift für Politikwissenschaft, 29(2): 157–176. Stadler, Gerhard (1991). ‘Autonomer Nachvollzug von Europarecht durch den österreichischen Gesetzgeber’, in Fritz Schwind (ed.), Österreichs Weg in die EG. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, pp. 111–141. Tálos, Emmerich and Kittel, Bernhard (2001). Gesetzgebung in Österreich. Vienna: Wiener Universitätsverlag. Weber, Karl (1992). ‘Macht im Schatten? (Landeshauptmänner-, Landesamtsdirektoren- und andere Landesamtsreferentenkonferenzen).’ Österreichische Zeitschrift für Politikwissenschaft, 21(4): 405– 418. Werndl, Josef (1984). Die Kompetenzverteilung zwischen Bund und Ländern. Vienna: Braumüller. Will, M.R. (1991). ‘Autonome Rechtsangleichung in Europa’, in Fritz Schwind (ed.), Österreichs Weg in die EG – Beiträge zur europäischen Rechtsentwicklung. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, pp. 53–109. Wohnout, Helmut (1999). ‘Die Mitwirkungsrechte des österreichischen Parlaments an der Willensbildung in der EU.’ Die Union, No. 1: 69–80.
70
4. The Minor Impact of EU on Legislation in Finland Matti Wiberg (University of Turku) & Tapio Raunio (University of Tampere) Introduction
The impact of European integration and the European Union (EU) on national parliaments first received serious political and academic attention in the mid-1990s. Both members of parliament (MPs) themselves and scholars became more interested in whether and how national parliaments controlled their governments in European matters, particularly as in a broader range of policies and national laws were overtaken by EU laws. The first decade of the new millennium has seen the completion of several comparative research projects on the role of national parliaments in the EU (Wiberg ed. 1997; Maurer & Wessels eds. 2001; Auel & Benz eds. 2005; Holzhacker & Albæk eds. 2007; O´Brennan & Raunio eds. 2007; Tans et al. eds. 2007; Barrett ed. 2008). But despite this proliferation of research, we lack empirical studies on the extent to which national parliaments have in fact been Europeanised, i.e. how much and in what ways the EU impacts national parliaments. And more specifically, we lack agreement on how to measure or operationalize the Europeanisation of national parliaments.1 The unicameral parliament of Finland, the Eduskunta, had good reasons to be concerned about the impact of EU on its policy influence. Recent constitutional reforms have quite radically altered the Finnish political system, with the government and the prime minister emerging from the shadow of the president as the leaders of the political process (Raunio & Wiberg 2003, 2008). The new constitution, which entered into force in 2000, completed a period of far-reaching constitutional change that curtailed presidential powers and brought the Finnish political system closer to a standard version of parliamentary democracy. While the executive branch still dominates national politics, at least now that executive is accountable to the Eduskunta. The Eduskunta is thus constitutionally stronger than ever before. Hence membership in the EU presented a challenge for the Eduskunta, as it feared seeing its position weakened as a result of the political dynamics of the EU policy process. While European integration is undoubtedly a major external constraint on national political systems, EU membership has arguably contributed to the strengthening of parliamentary democracy in Finland by consolidating the political leadership of the government and the prime minister. Most notably, with the exception of Treaty amendments, EU matters belong almost exclusively to the jurisdiction of the cabinet. The government dictates national EU policy, with the president intervening mainly when questions related to Finland’s foreign policy are on the agenda. The prime minister is the primary representative of
1
For reviews of literature on national parliaments and European integration in general, see Goetz and MeyerSahling (2008) and Raunio (2009). For a suggestion as to how the Europeanisation of a national parliament should empirically be measured, see Raunio and Wiberg (2010).
71
Finland in the EU, but the president also participated in most European Council meetings. However, since the Lisbon Treaty, it appears that prime minister will be the only representative of Finland in the European Council, thus ending the era of ‘dual plates’ in the summits. (Raunio 2008a) Hence, the constitutional reforms were an essential prerequisite for establishing parliamentary accountability in EU affairs.
Examining the impact of EU membership on the Eduskunta itself, we note that while the Eduskunta has lost policy-making competencies to the European level, it has subjected the government to relatively tight scrutiny in EU matters. Comparative studies indicate that the Eduskunta has one of the strongest scrutiny systems of the national parliaments among EU countries (Raunio 2005b). The scrutiny model of the Eduskunta has four main strengths: the position of the parliament is regulated in the constitution, the Eduskunta gets involved relatively early in the processing of EU legislation, the parliament enjoys unlimited access to information from the government, and the responsibility for monitoring European matters is delegated downwards to specialised committees (Jääskinen 2000). The Grand Committee is responsible for coordinating the Eduskunta’s positions on EU issues, while the Foreign Affairs Committee is the main body dealing with the foreign and security policy matters handled at the European level. The minister appears in the Grand Committee in person before and, when required, after the Council meeting. The standing committees are closely involved in the scrutiny of EU matters from an early stage in the policy process, and the final position of the Grand Committee is based on guidelines from the standing committees. The constitutionally regulated access to information from the government reduces informational asymmetries in EU affairs. The rules encourage the government to provide the Eduskunta with information of its own accord, without any specific requests by MPs. Moreover, the active scrutiny of European legislation has improved the overall dialogue between the government and the Eduskunta. The regular appearance of ministers before the Grand Committee has led to improved policy coordination within the cabinet and between the ministries, and has forced ministers to study the issues more thoroughly than might otherwise be the case. An often-mentioned feature of the EU policy process is bureaucratisation, the shift of power from directly elected office holders to civil servants. In Finland, civil servants perform a central role at all stages of the process, from the initial formulation of the national position in the ministries in Helsinki, to negotiations among the permanent representatives in Brussels. However, the autonomy of civil servants is at least partially counteracted by the active scrutiny of the Eduskunta in EU matters. (Raunio 2007) In terms of the party system, European integration has caused divisions within most Finnish political parties. Despite the divisiveness of the 1994 EU membership referendum (in which 57 % voted in favour of joining the Union; turnout was 74 %), the shape of the party system has remained intact, with the same set of parties contesting elections as before integration entered the domestic political agenda. More
72
difficult to measure, but arguably increasingly important, is the way in which European integration (and in general the opening of borders) has reshaped patterns of partisan contestation. While the left-right dimension continues to be the main cleavage in Finnish politics, Europe as a political issue has reinforced the role of the centre/periphery cleavage in political competition. Indeed, since the early 1990s the rural– urban/centre–periphery divide has become the second main cleavage in Finnish party politics, largely because the EU and globalisation issues have emerged on the political agenda (Nousiainen 2000, 265268; Paloheimo 2005, 2008; Reunanen &d Suhonen 2009). Parties that represent either peripheral areas or occupational groups are more likely to suffer as a result of increased economic competition and have thus faced serious internal divisions over integration. The office-seeking tendencies that dominate Finnish party politics have produced broad ideological consensus over Europe, with the inevitable consequence that parties are not representative of their voters on the EU dimension. Indeed, a broad partisan consensus has emerged for national European policy that can be characterized as flexible and constructive and has sought to consolidate Finland’s position in the inner core of the Union. Finland is the only Nordic country that belongs to the Economic and Monetary Union (EMU), and the single currency was adopted without much political contestation. Underlying this pro-integrationist stance is the conviction that a strong and efficient EU can best protect the rights and interests of smaller member states, as intergovernmental processes tend to favour larger member states (Raunio & Tiilikainen 2003). The commitment to integration which prevails among the elites is not shared to the same extent by the Finnish electorate. There is a notable incongruence between the citizens and the political parties. In fact, according to the European Election Study survey carried out in 2004, Finland had one of the lowest levels of party-voter congruence on integration matters, with only Great Britain, Hungary, and Luxembourg recording bigger differences (Mattila & Raunio 2006). Hence the EU has been a difficult issue for most political parties ever since European integration first entered the domestic political agenda in the early 1990s. Overall, Finnish parties have kept a fairly low profile in integration matters, and the rules of the national EU coordination system – based on building broad domestic consensus, including often government and opposition parties in the Eduskunta and its Grand Committee – have contributed to the depoliticization of European issues (Raunio 2005a, 2008b). This has applied particularly to major EU decisions, such as EMU membership in the late 1990s and Treaty reforms. Given that parties are internally divided over the EU, it was not surprising that the main parties showed little interest in subjecting the Constitutional Treaty or the Lisbon Treaty to a referendum.
In this chapter we start by discussing the theoretical and methodological challenges involved in measuring the effect of the EU on national parliaments, particularly regarding the Europeanization of law production. Our empirical analysis covers mainly Finland’s EU membership, from 1995 to 2009. The data set consists of all national legislation (N=8 719). We shall focus in particular upon the quantitative impact of EU on national legislation. We show that the actual share of EU influenced laws, contrary to statements by prominent political and administrative figures in Finland, has been rather small. This
73
finding is probably explained by the fact that the bulk of Finnish legislation concerns areas where the EU does not have legislative competencies.
The impact of Europe on domestic laws
Any meaningful study of how European integration impacts on domestic policy-making must start with a definition of Europeanization. After all, the selection of empirical indicators should be driven by the choice of the theoretical concept. In this chapter we are primarily interested in the ‘top-down’ dimension of Europeanization and hence we stick to the original definition of Europeanisation by Ladrech (1994, 69), according to which Europeanization is ‘an incremental process re-orienting the direction and shape of politics to the degree that [EU] political and economic dynamics become part of the organizational logic of national politics and policy-making.’2 This definition is useful as it contains the fundamental question we are interested in, the extent to which national law production is influenced by European integration.3 Before proceeding to examine how much the EU impacts Finnish domestic legislation, it is worth highlighting some common arguments and misperceptions related to the influence of Europe on national law-making and discussing some measurement problems related to our research question. The famous ‘80 %’ prediction made by the Commission President Jacques Delors in the late 1980s about the share of legislation that would flow from Brussels has proven to be overestimated. Previous research has shown this share to be much lower, even when including domestic laws that were in some way ‘inspired’ by the EU. For example, Töller (2008) demonstrates that while the share of EU-inspired legislation enacted by the German Bundestag has more than doubled since the mid-1980s, in the 15th election period (20022005) 39 % of the laws were influenced by a ‘European impulse’. König and Mäder (2008) also show that the share of German legislation with an ‘EU impulse’ has increased over time, with 24 % of the laws enacted between 1976 and 2005 having such a European connection. According to Hegeland (2005) only 6 % of the legislation adopted by the Swedish Riksdag between 1995 and 2004 contained a reference to an EU law. Johannesson (2005) in turn showed that between 1998 and 2003 20 % of Riksdag’s legislation was related to binding EU legislation, with an additional 10 % in some way related to the Union. The 2
There are several competing or alternative definitions of Europeanization. See the volumes edited by Graziano and Vink (2007), Bulmer and Lequesne (2005), and Featherstone and Radaelli (2003). 3 Obviously one could also employ a more demanding definition of Europeanization. For example, according to an often-utilized definition, Europeanization is ‘processes of (a) construction, (b) diffusion, and (c) institutionalisation of formal and informal rules, procedures, policy paradigms, styles, ‘ways of doing things’, and shared beliefs and norms which are first defined and consolidated in the making of EU public policy and politics and then incorporated in the logic of domestic discourse, identities, political structures, and public policies.’ (Radaelli 2003, 30) Radaelli’s definition thus shares many elements with the original definition by Ladrech, but goes also deeper by emphasising more explicitly cultural, attitudinal or informal aspects of domestic politics. When applying this definition, scholars could examine whether engagement in the EU’s policy process alters the ways in which national laws are drafted or whether the ‘rhythm’ or time-space’ of EU politics impacts on when or how often domestic laws are enacted. However, while such questions would undeniably be highly interesting for future research, utilizing such an allencompassing definition of Europeanization might be very demanding in terms of both data and the reliability of findings.
74
most likely explanation for this relatively low share of EU-related domestic laws is that most of the legislation adopted by national parliaments deals with policy sectors where the EU has no formal competence to enact its own laws.4 One must also be aware of the measurement problems involved in attempting to conduct comparative research on the share of ‘EU-related’ legislation (see Töller 2007, 2008). First, the production of laws differs between EU member states, with some parliaments adopting considerably fewer laws than others. In some EU countries the adoption of legislation may be delegated more extensively to the government that issues decrees in place of laws processed by parliaments. Secondly, much of EU legislation consists of regulations that are directly binding and hence do not require parliamentary approval. But perhaps the most significant problem is measuring the indirect influence of European integration. Domestic legislation can be indirectly shaped or influenced by European integration, for example in the form of policy diffusion and peer pressure under the Open Method of Coordination (OMC) and other ‘soft law’ coordination efforts (Duina & Raunio 2007). And apart from pressure and policy diffusion resulting from OMC and other forms of intergovernmental policy coordination, governments may import policies from other EU countries or follow the EU’s recommendations without this being explicitly acknowledged in the text of the initiative. This applies particularly to policy sectors – most notably foreign and defence policies – where the outputs are normally not laws but other types of decisions. Turning to the debate in Finland, both Europhiles and Eurosceptics share a common estimation of the massive legislative impact of the EU on the national legal orders of the member states. Quite commonly, as a matter of fact, on a routine basis, it is claimed and moreover believed, that the EU produces some 80 % of new laws in the EU member states. Many relevant actors have their opinions on the quantitative impact of the EU on domestic law production, based on reliable and systematic empirical evidence – or, as it more often seems to be the case, on something other than empirical evidence. In the political debate many different estimates have been presented on how much the EU has influenced Finnish national legislation. Estimates vary widely from ‘more than 80 %’ (Stubb 2003) to ‘at least 65-70 %’ (Suominen 2004). ‘Brussels already decides half of Finnish legislation’, claimed Ville Itälä, a Finnish member of the European Parliament (MEP), some years ago in a local newspaper (Turkulainen 28.9.2005). The Permanent Secretary of the Ministry of Justice Kirsti Rissanen claimed that ‘approximately one third of governmental proposals is linked either to the implementation of EU-obligations or international commitments.’ (Helsingin Sanomat 13.11.2006) The Finnish MEP Carl Haglund claimed in a radio interview on the 7th of August 2009 that ‘70 % of our law production comes from the EU’ (Sommarprat Radio Vega 7.8.2009). There are also those that insist that the Finnish parliament is not just a rubber stamp for EU legislation (Jääskinen 2000). It is no exaggeration to claim that too many of those claiming some estimate have been far too sloppy in their arguments. What is obvious is that all seem to adhere to the formula ‘EU something impacts national 4
See also Page (1998), Bovens and Yesilkagit (2004), and Blom-Hansen and Christensen (2004) and other chapters in this volume.
75
something’, but both the ‘EU something’ and the ‘national something’ have been left unspecified. At least the following possibilities are possible both as a source of impact and as the object of impact: legal order (system), legal norms (collection of regulations), political system, political culture, operational procedure of decision making and policy diffusion. Ideally we should have a measure of how the EU’s legal order impacts the national legal order norm by norm at every step in the norm hierarchy. This is, unfortunately, not feasible as there is no universally accepted categorization of the legal order neither of the EU nor of the national political system. Another complication should also be addressed: are we really making claims of causality or mere correlation? It is not possible to establish with certainty whether the association between the EU and national law production is causal. After all, all co-variation is not causal in its nature. A further complication is that the EU does not work in isolation from national inputs – that is, laws of the EU are obviously often based on or strongly influenced by national positions and laws. One plausible operationalization of the influence of the EU would consider whether there is a reference to some EU norm in the title or in the motivations for a bill. It should be noted that this selection criterion probably overestimates the EU’s influence on national legislation as the influence in question could cover only one paragraph or section of the bill in question. In the following analysis all national bills passed by the Eduskunta are classified into two categories: those influenced by the EU and those not influenced by the EU. Given that consecutive Treaty amendments enacted since the late 1980s have extended the policy reach of the EU, and in line with previous research discussed above, we hypothesise that the share of EUrelated laws has increased over time. However, prior to Union accession in 1995, Finland had adapted her national legislation to the acquis communautaire. And as Finland joined the EU just after the completion of the EU’s law-intensive internal market, we expect this increase in the share of EU-related laws to be relatively small.5
Empirical analysis
We turn first to the total amount of formal regulation in Finland. Since 1945 over 70 000 items have been published in the Statutory Book (Säädöskokoelma), comprising altogether over 174 000 pages of legal texts. These items are of five different types: Laws, Decrees, Decisions of the Council of State, Ministerial decisions and Other regulations. According to these figures Finland is one of the most industrious regulators in Europe (Figure 1: the enactment of some 22 000 laws, 28 000 decrees, 8 700 Decisions of the Council of State, some 8 800 Ministerial decisions and over 3 100 other regulations (Figures 1 and 2) has occurred in the post WWII period.
5
It is a well-established fact that Finland was a frontrunner in implementing EU directives since becoming an EU member. However this no longer seems to be the case, as Finland is now in the middle group. Without the autonomous Åland islands, which often lags behind in implementation, mainly for language reasons, Finland’s record would most likely be better. See http://ec.europa.eu/community_law/directives/directives_communication_en.htm.
76
FIGURES 1 AND 2
Let us now take a closer look at the number of laws. There are three peculiar aspects of Finnish law production that merit mentioning: almost all passed laws are initiated by the government, they are rarely voted upon, and they are typically short, mostly one paragraph modifications of existing laws. For the period 1945-2003 almost all successful bills were initiated by the government. Less than two percent of all bills were initiated by MPs. Law production is very consensual with a very low levels of conflict: only 10,4 % of laws were voted upon on the floor. Finnish laws are typically very short: for the period 19452003 altogether 39,5 % of laws were modifications of one paragraph only.6 From January 1995 until the end of 2009 the Eduskunta passed 8 719 bills were passed by the Eduskunta. Of these 8 614 (98,8 %) were introduced by the government, only 102 bills (1,2 %) were introduced by the individual MPs, three were introduced by the Speaker’s Council (puhemiesneuvosto) and one as a budget initiative (talousarvioaloite). Only 1 025 of these laws (Figure 3) or 11,8 % of bills enacted during Finnish EU membership have contained a reference to the EU.7 This borders on minimal proportions, not anywhere near Delors prediction. Although the share of EU-related laws remained around 10 % until 2003, since 2004 over 15 % of the laws approved by the Eduskunta have contained references to EU legislation (last year being an exception). The peak was in 2004 when 25 % of the laws were related to the EU. It should be noted that our measure in fact overestimates the impact of the EU as it does cover not only direct references to EU laws in the strict legal sense, but also includes all references to all types of regulations which have stemmed from the European Community, the European Economic Community, or
6
Mean length is 6,7 paragraphs, the median is 2, mode is 1, and maximum is 301 paragraphs. (Wiberg 2004b) The data was originally coded by research assistant Olli Aaltonen at the political science department of the University of Turku, Finland. All laws were taken from the electronic database of Finnish legal acts (www.finlex.fi) and were coded as EU-related only when they contained an explicit reference to at least one of the following: the European Community (EY), European Economic Community (ETY) or Official Journal (EYVL). This was operationalized by searching semi-automatically for EU legal terms (EU, ETY, EYVL) in the legal texts proper in their entirety. Keyword searches in the Finnish language are problematic as there are so many cases. The word European Union (Euroopan unioni in Finnish), for instance, takes the following forms: Euroopan unioni (Nominative) Euroopan unionia (Partitive) Euroopan unionin (Genitive) Euroopan unionin (Accusative) Euroopan unionissa (Inessive) Euroopan unionilla (Adessive) Euroopan unionina (Essive) Euroopan unionista (Elative) Euroopan unionilta (Ablative) Euroopan unioniin (Illative) Euroopan unionille (Allative) Euroopan unioniksi (Translative) Euroopan unioneineen (Comitative) Euroopan unionein (Instructive) Euroopan unionitta (Abessive) 7
77
have been published in the Official Journal. Hence our measure even includes references to soft law declarations and other non-binding EU documents.
FIGURE 3
We now turn to take a closer look at the characteristics of Finnish laws. As we have not coded all laws for the membership period 1995-2009, the following analysis is restricted to the period 1995-2007. There is huge variation among ministries and committees, with the influence for any single ministry not larger than approximately 25 % (Ministry for Transportation and Communications). The Prime Minister’s Office and Ministry for Defense have the smallest share (Figure 4). And when examining the share of EU-influenced bills by the standing committees of the Eduskunta, we note that two committees stand out: the Commerce Committee and the Transport and Communications Committee (Figure 5).
FIGURES 4 AND 5
One could of course claim that even when the frequency of EU influence on national laws is not large, it could still be strong in some other way. For instance one could speculate that entirely new bills could be more influenced by the EU. This is clearly not the case, however. All bills enacted during 1995-2007 were classified into nine categories: 1. Regulation of a new issue area; 2. Replacement law; 3. Modification or amendment to already existing bill; 4. Concerns state property; 5. Ratification of international treaty; 6. Termination of law; 7. Continuation of law; 8. Concerns state offices and posts; and 9. Other. We note that the bulk of laws (83,5 %) were modifications or amendments to already existing bills. Only 4,4 % of the new bills were regulations of new issue areas.
FIGURES 6 AND 7
The bills were also coded according to the Comparative Policy Agendas scheme into the following topics:8 1. Macroeconomics; 2. Civil Rights, Minority Issues, and Civil Liberties; 3. Health; 4. Agriculture; 5. Labor, Employment, and Immigration; 6. Education; 7. Environment; 8. Energy; 9 IS MISSING; 10. Transportation; 11 IS MISSING; 12. Law, Crime, and Family Issues; 13. Social Welfare; 8
See the Comparative Policy Agendas project website (http://www.comparativeagendas.org). The project distinguishes among more than 200 political issues within 20 main categories of public action.
78
14. Community Development and Housing Issues; 15. Banking, Finance, and Domestic Commerce; 16. Defense; 17. Space, Science, Technology and Communications; 18. Foreign Trade; 19. International Affairs and Foreign Aid; 20. Government Operations; 21. Public Lands and Water Management; 22 IS MISSING; 23 IS MISSING; 24. State and Local Government Administration; 25 IS MISSING; 26. Weather and Natural Disasters; 27. Fires; 28. Arts and Entertainment; 29. Sports and Recreation; 30. Death Notices; 31. Churches and Religion; and 99. Other, Miscellaneous, and Human Interest. We conclude that the bulk of new legislation concerns welfare state regulations. And in terms of EU influence, we see that finance regulations are more strongly impacted by the EU than any other category (Figures 6 and 7).
Discussion
The purpose of this chapter has been to measure the Europeanisation of Finnish law production. Our analysis has some important lessons for future research. First, our analysis showed how the impact of Europe varies inside the legislature and to a great deal by ministry and topic. Secondly, we have seen that the actual share of EU-influenced laws, contrary to statements by prominent political and administrative figures in Finland, is rather small. Perhaps the main explanation for this relatively low influence of EU is the simple fact that the bulk of domestic Finnish legislation concerns areas where the EU does not have competence to regulate. Our evidence supports the hypothesis that the share of EU-related laws has increased over time.
However, we acknowledge that our indicators do not capture all dimensions of the potential impact of Europe on national law-making. Europe can leave its imprint on national laws without this being explicitly acknowledged in the text of the initiative. For example, national politicians may want to downplay or hide the influence of EU or other European countries in order to claim credit for (hopefully) successful laws. Nor can our quantitative approach adequately capture any policy diffusion processes taking place inside the Union. More detailed sector-specific studies are needed to investigate how such diffusion impacts on national legislation and public policies. Methodological concerns regarding both the operationalization of EU impact and coding errors should also be seriously considered in future research. But perhaps more importantly, parliaments do much more than simply approve or amend laws. Hence future research should pay more attention to how the work of national parliaments has in fact become Europeanized. Our explorative study on the Eduskunta showed how the impact of EU varies inside the legislature, with parliamentary committees most burdened by European affairs (Raunio & Wiberg 2010). The share of committee time spent on EU issues is considerably higher than the share of domestic EUrelated laws. Indeed, the Eduskunta (2005, 14) itself has estimated that nearly half of all items processed
79
by it are related to EU matters. That same report also noticed that the workload of committees, measured by the number of agenda items, had roughly doubled as a result of EU membership. For example, in 2003 the specialized committees (excluding the Grand Committee) processed 314 EU matters, 119 of which resulted in a report. In that same year the number of domestic items handled by committees was 244 (Eduskunta 2005, 15). There was considerable variation between the committees, with this variation primarily driven by the allocation of powers between the national and European levels. Committees on Education and Culture and Future were least burdened by EU matters, while in the Environment, Commerce, and Agriculture and Forestry Committees, European questions comprised half or more of the meeting time. These findings probably result from the constitutional obligation of the specialized committees to report on EU matters, and hence the situation is likely to be different in most other national parliaments. The extent to which EU matters are centralized to the EU committee may thus be a key variable in explaining how Europe impacts domestic legislatures. In legislatures where European matters are mainly isolated in the EU committee, the other committees should deal more infrequently with EU questions.
Future research could also explore the role of plenary debates and control instruments in the processing of EU affairs and European laws. In European affairs the plenary can become involved both before and after decisions are taken at the European level. The Speaker’s Council can decide that proposals for EU decisions be debated in the plenary, but in such cases the chamber is not entitled to make formal decisions. Plenary stage is also required when the implementation of directives requires legislation. The prime minister must provide either the plenary or the EAC information on European Council meetings both before and after the summits. The same applies to Treaty amendments, which require the Eduskunta’s approval. So far the plenary has performed a marginal role in European matters in Finland. Routine EU matters, such as the implementation of directives, are only very rarely debated on the floor. Farreaching political decisions such as the EMU, the EU’s multi-annual budgetary frameworks, Treaty amendments, and notably the development of EU’s foreign and security policy have inspired longer plenary debates. The Eduskunta (2005, 18, 40-41) itself has paid attention to the lack of plenary debates, recommending for example that such debates could be held in connection with European Council meetings or in conjunction with the Commission’s annual legislative programmes, or that specific question-times be held on European matters. The main explanation for the brevity of plenary debates is probably the role accorded to the Grand Committee, which coordinates parliamentary work in EU issues and speaks on behalf of the Eduskunta in such matters. Regarding control instruments, between 1995 and 2007 only three out of 46 interpellations (7 %) focused on European affairs, with two of them dealing with the EU’s agricultural policy and its impact on Finland. 467 out of 14 113 written questions (3 %)
80
were on EU matters. As for oral questions, that share was somewhat higher, 8 % (209 out of 2 663 questions). (Raunio & Wiberg 2010) However, perhaps the most important result of this chapter concerns the methodological difficulties involved in separating EU affairs from other matters handled by national parliaments. What actually constitutes an EU issue? While certain questions – such as Treaty amendments, enlargements, EU budget, or EU laws – can be rather uncontestedly categorized as European matters, more typical are cases where EU and domestic spheres become so intertwined that ‘isolating’ the EU dimension is very challenging. This applies particularly to policy-related questions (e.g. agriculture, economy, environment etc.), regardless of whether the matter is of European or national origin. Not only does an increasing share of matters formally decided at the national level have a European dimension, but also debates on EU laws or European level processes can be dominated by domestic issues. Considering the interdependence of EU and national agendas, it is likely that these kinds of problems will only become more serious in the future. Hence these methodological concerns should be taken into account in future research.
81
References Auel, K. & Benz, A. eds. 2005. ‘The Europeanisation of Parliamentary Democracy’, Journal of Legislative Studies 11(3-4) (Special Issue). Barrett, G. ed. 2008. National Parliaments and the European Union: The Constitutional Challenge for the Oireachtas and Other Member State Legislatures. Dublin: Clarus Press. Bovens, M. & Yesilkagit, K. 2004. ‘The Impact of European Legislation on National Legislation in the Netherlands’, Paper prepared for the 2004 NIGH Conference, Erasmus University Rotterdam, October 29, 2004. Duina, F. & Raunio, T. 2007. ‘The open method of coordination and national parliaments: further marginalization or new opportunities?’, Journal of European Public Policy 14, 489-506. Eduskunta 2005. EU-menettelyjen kehittäminen: EU-menettelyjen tarkistustoimikunnan mietintö. Helsinki: Eduskunnan kanslian julkaisu 2/2005. Hegeland, H. 2005. ’EG-rättens genomslag i svenska lagar och förordningar’, Europarättslig tidskrift 8, 398–9. Holzhacker, R. & Albæk, E. eds. 2007. Democratic Governance and European Integration: Linking Societal and State Processes of Democracy. Cheltenham: Edward Elgar. Johannesson, C. 2005. ’EU:s inflytande över lagstiftning i Sveriges riksdag’, Statsvetenskapliga Tidskrift 107, 71–84. Jääskinen, N. 2000. ’Eduskunta – Aktiivinen sopeutuja’, in Raunio, T. & Wiberg, M. eds., EU ja Suomi: Unionijäsenyyden vaikutukset suomalaiseen yhteiskuntaan. Helsinki: Edita. König, T. & Mäder, L. 2008. ‘Das Regieren jenseits des Nationalstaates und der Mythos einer 80Prozent-Europäisierung in Deutschland’, Politische Vierteljahresschrift 49, 438-63. Ladrech, R. 1994. ‘Europeanization of domestic politics and institutions: the case of France’, Journal of Common Market Studies 32, 69-88. Ladrech, R. 2007. ‘National Political Parties and European Governance: The Consequences of ‘Missing in Action’, West European Politics 30, 945-60. Mattila, M. & Raunio, T. 2006. ‘Cautious Voters - Supportive Parties: Opinion Congruence between Voters and Parties on the EU Dimension’, European Union Politics 7, 427-49. Maurer, A. & Wessels, W. eds. 2001. National Parliaments on their Ways to Europe: Losers or Latecomers? Baden-Baden: Nomos. Nousiainen, J. 2000. ‘Finland: The Consolidation of Parliamentary Governance’, in Müller, W.C. & Strøm, K. eds., Coalition Governments in Western Europe. Oxford: Oxford University Press. O’Brennan, J. & Raunio, T. eds. 2007. National Parliaments within the Enlarged European Union: From ’victims’ of integration to competitive actors? Abingdon: Routledge. Paloheimo, H. 2005. ’Puoluevalinnan tilannetekijät’, in Paloheimo, H. ed., Vaalit ja demokratia Suomessa. Helsinki: WSOY.
82
Paloheimo, H. 2008. ’Ideologiat ja ristiriitaulottuvuudet’, in Paloheimo, H. & Raunio, T. eds., Suomen puolueet ja puoluejärjestelmä. Helsinki: WSOY Radaelli, C. 2003. ‘The Europeanization of Public Policy’, in Featherstone, K. & Radaelli, C. eds., The Politics of Europeanization. Oxford: Oxford University Press. Raunio, T. 2005a. ‘Hesitant Voters, Committed Elite: Explaining the Lack of Eurosceptic Parties in Finland’, Journal of European Integration 27, 381-95. Raunio, T. 2005b. ‘Holding Governments Accountable in European Affairs: Explaining Cross-National Variation’, Journal of Legislative Studies 11, 319-42. Raunio, T. 2007. ‘The Finnish Eduskunta: Effective Scrutiny, Partisan Consensus’, in Tans, O., Zoethout, C. & Peters, J. eds., National Parliaments and European Democracy: A Bottom-Up Approach to European Constitutionalism. Groningen: Europa Law Publishing.
Raunio, T. 2008a. ’Parlamentaarinen vastuu ulkopolitiikkaan: Suomen ulkopolitiikan johtajuus uuden perustuslain aikana’, Politiikka 50, 250-65. Raunio, T. 2008b. ‘The Difficult Task of Opposing Europe: The Finnish Party Politics of Euroscepticism’, in Szczerbiak, A. & Taggart, P. eds., Opposing Europe? The Comparative Party Politics of Euroscepticism: Volume I, Case Studies and Country Surveys. Oxford: Oxford University Press. Raunio, T. 2009. ‘National Parliaments and European Integration: What We Know and Agenda for Future Research’, Journal of Legislative Studies 15, 317-34.
Raunio, T. & Tiilikainen, T. 2003. Finland in the European Union. London: Frank Cass. Raunio, T. & Wiberg, M. 2003. ’Finland: Polarized Pluralism in the Shadow of A Strong President’, in Strøm, K., Müller, W.C. & Bergman, T. eds., Delegation and Accountability in Parliamentary Democracies. Oxford: Oxford University Press. Raunio, T. & Wiberg, M. 2008. ‘The Eduskunta and the Parliamentarisation of Finnish Politics: Formally Stronger, Politically Still Weak?’, West European Politics 31, 581-99. Raunio, T. & Wiberg, M. 2010. 'How to Measure the Europeanisation of a National Legislature?', Scandinavian Political Studies 33, 74-92. Reunanen, E. & Suhonen, P. 2009. ’Kansanedustajat ideologisella kartalla’, in Borg, S. & Paloheimo, H. eds., Vaalit yleisödemokratiassa: Eduskuntavaalitutkimus 2007. Tampere: Tampere University Press. Stubb, A. 2003. Minne menet EU? Helsinki: Tammi. Suominen, I. 2004. EU vaikuttaa selvästi lakeihimme, Helsingin Sanomat 10.3.2004, A5. Tans, O., Zoethout, C. & Peters, J. eds. 2007. National Parliaments and European Democracy: A Bottomup Approach to European Constitutionalism. Groningen: Europa Law Publishing. Töller, A.E. 2007. ‘Measuring the Europeanization of Public Policies – but How? A Research Note’, FoJuS-Diskussionspapiere, Nr. 1/2007. Töller, A.E. 2008. ’Mythen und Methoden: Zur Messung der Europäisierung der Gesetzgebung des Deutschen Bundestages jenseits des 80%-Mythos’, Zeitschrift für Parlamentsfragen 39, 3-17.
83
Wessels, B. 2005. ‘Roles and Orientations of Members of Parliament in the EU Context: Congruence or Difference? Europeanisation or Not?’, Journal of Legislative Studies 11, 446–65. Wiberg, M. ed. 1997. Trying to Make Democracy Work: The Nordic Parliaments and the European Union. Stockholm: Gidlunds. Wiberg, M. 2004a. ‘Lainsäädäntömme EU-vaikutteisuus luultua oleellisesti pienempää’, Oikeus 33, 200206. Wiberg, Matti (2004b): Lainsäädäntötuotos Suomessa valtiopäivillä 1945-2002: Peruskartoitus. Valtioopillisia tutkimuksia 58, Turun yliopiston valtio-opin laitos. Wiberg, M. 2006. Miten paljon EU vaikuttaa Suomen normituotantoon? http://www.edilex.fi/lakikirjasto/asiantuntijakirjoitukset/2995/.
Ministries VNK Valtioneuvoston kanslia
Prime Minister's Office
UM Ulkoasianministeriö
Ministry for Foreign Affairs
OM Oikeusministeriö
Ministry of Justice
SM Sisäministeriö
Ministry of the Interior
PM Puolustusministeriö
Ministry of Defence
VM Valtiovarainministeriö
Ministry of Finance
OpM Opetusministeriö
Ministry of Education
MMM Maa- ja metsätalousministeriö
Ministry of Agriculture and Forestry
LM Liikenneministeriö
Ministry of Transport and Communications
KTM Kauppa- ja teollisuusministeriö
Ministry of Trade and Industry
STM Sosiaali- ja terveysministeriö
Ministry of Social Affairs and Health
TM Työministeriö
Ministry of Labour
YM Ympäristöministeriö
Ministry of the Environment
Standing Committees
SuV Suuri valiokunta
The Grand Committee 84
PevPerustuslakivaliokunta
The Constitutional Law Committee
UaV Ulkoasiain valtiokunta
The Foreign Affairs Committee
VaV Valtiovarain valiokunta
The Finance Committee
HaV Hallintovaliokunta
The Administration Committee
LaV Lakiasiainvaliokunta
The Legal Affairs Committee
LiV Liikennevaliokunta
The Transport and Communications Committee
MmV Maa ja metsätalousvaliokunta The Agriculture and Forestry Committee PuV Puolustusvaliokunta
The Defence Committee
SiV Sivistysvaliokunta
The Education and Culture Committee
Stv Sosiaali- ja terveysvaliokunta
The Social Affairs and Health Committee
TaV Talousvaliokunta
The Commerce Committee
TuV Tulevaisuusvaliokunta
The Committee for the Future
TyVTyö- ja tasarvovaliokunta
The Employment and Equality Committee
YmV Ympäristövaliokunta
The Environment Committee
85
Figure 1. Items and Pages in the Statutory Book by Calendar Year 1945-2009.
9000 8000 7000 6000 5000 4000
Items Pages
3000 2000 1000 0 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
86
Figure 2. The Number of Laws (Lakeja), Decrees (Asetuksia), Decisions of the Council of State (VNpäätöksiä), Ministerial decisions (Minpäätöksiä) and Other (Muu) regulations by Calendar Year 1945-2009.
1000 900 800 700
Lakeja
600
Asetuksia
500
VNpäätöksiä
400
Minpäätöksiä Muu
300 200 100 2008
2005
2002
1999
1996
1993
1990
1987
1984
1981
1978
1975
1972
1969
1966
1963
1960
1957
1954
1951
1948
1945
0
Figure 3. Number of enacted non-EU-impacted and EU-impacted laws 1995-.
1200
100
1000
80
800
60
EU-impacted
40
%
600 400
20
200 0
Laws
Linear (%)
0 1995 1997 1999 2001 2003 2005 2007 2009
Sources: Wiberg (2004, 2006) and own updated calculations.
87
Figure 4. Share of EU-influenced bills (%), by ministry, 1995-2007. vm om stm lm ktm sm mmm ym min tvm um opm vnk stv pm 0,0
10,0
20,0
30,0
40,0
50,0
60,0
70,0
80,0
90,0
100,0
Source: Own calculations.
Figure 5. Share of bills (%) influenced by the EU, by standing committee, 1995-2007 (%).
tav liv vav ymv mm stv hav tyv lav siv uav pev suv puv 0
10
20
30
40
50
60
70
80
90
100
Source: Own calculations.
Figure 6. Number of Laws by Topics 1995-2007.
88
5 15 1 12 3 14 20 10 6 7 19 4 13 2 16 21 18 17 8 31 28 24 30 0
200
400
600
800
1000
1200
Source: Own calculations.
Figure 7. Number of non-EU-impacted and EU-impacted Laws by Topics 1995-2007. 30 24 28 31 8 17 18 21 16 2 13 4 19 7 6 10 20 14 3 12 1 15 5 0
200
400
600
800
1000
1200
Source: Own calculations.
Data for Figures [Just in order to facilitate the type setting]
89
Figure 1 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
Items Pages 1346 2294 955 1647 1031 1635 1005 1580 894 1299 680 1195 699 1298 479 878 575 1187 540 1089 614 1211 706 1247 482 1042 565 1300 565 1186 558 1415 622 1329 702 1621 671 1294 712 1634 760 1626 787 1758 677 1533 831 2060 930 1898 917 1693 1016 1874 965 2033 1051 2070 1126 2022 1097 2093 1156 2052 1134 2242 1142 2392 1089 2434 1107 2404 1114 2103 1120 2116 1155 2211 1031 2068 1146 2175 1075 2146 1340 3281 1384 3244 1381 2516 1399 2982 1744 3460 1715 4076 1709 4796 1599 4695 1809 4190
90
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
1377 1421 1229 1378 1307 1568 1371 1394 1437 1287 1460 1510 1155 1853 70654
3982 4742 3462 3646 3564 4376 5634 5634 4622 4140 4468 5790 3604 7638 170926
Figure 2 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981
Lakeja Asetuksia VNpäätöksiä Minpäätöksiä Muu Säädöksiä Sivuja 157 329 287 536 37 1346 2294 122 214 200 399 20 955 1647 124 270 205 401 17 1031 1635 178 241 181 392 13 1005 1580 127 268 212 262 25 894 1299 128 259 167 110 16 680 1195 139 242 188 118 12 699 1298 100 161 99 106 13 479 878 101 216 115 103 40 575 1187 102 182 160 81 15 540 1089 172 231 119 77 15 614 1211 140 296 176 79 15 706 1247 103 184 98 81 16 482 1042 139 222 117 74 13 565 1300 123 258 106 69 9 565 1186 140 239 90 77 12 558 1415 125 300 99 76 22 622 1329 168 337 100 80 17 702 1621 139 317 112 88 15 671 1294 162 324 129 77 20 712 1634 209 299 133 109 10 760 1626 225 323 142 81 16 787 1758 170 297 123 75 13 677 1533 201 344 176 92 17 831 2060 284 347 176 108 15 930 1898 254 398 155 87 23 917 1693 270 449 172 101 24 1016 1874 203 411 192 129 28 965 2033 275 439 179 127 31 1051 2070 328 415 202 155 26 1126 2022 256 454 184 175 28 1097 2093 369 441 175 145 26 1156 2052 311 466 187 131 39 1134 2242 344 448 187 137 30 1142 2392 356 423 171 112 28 1089 2434 274 462 199 143 29 1107 2404 280 497 189 127 21 1114 2103
91
1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Figure 3 Laws 1995 790 1996 539 1997 464 1998 472 1999 595 2000 453 2001 590 2002 546 2003 589 2004 619 2005 521 2006 497 2007 655 2008 428 2009 962 Sum 8720
366 318 293 353 279 432 396 407 463 575 641 670 564 796 539 468 478 597 453 592 546 589 621 522 528 661 439 969 21853
439 504 447 488 462 553 620 605 559 743 616 549 518 499 359 443 313 357 654 848 698 644 682 636 758 718 610 788 28113
EU-impacted 43 5,4 44 8,2 48 10,3 49 10,4 40 6,7 37 8,2 25 4,2 52 9,5 58 9,9 154 24,9 83 15,9 88 17,7 117 17,9 75 17,5 112 11,6 1025 11,8
165 170 162 148 169 184 203 180 142 187 170 154 125 124 111 123 87 92 25 6 7 19 13 6 15 10 28 14 8741
123 127 104 122 131 126 133 136 179 190 241 281 294 279 257 273 230 196 58 11 0 4 8 3 6 7 5 2 8846
27 36 25 35 34 45 32 53 56 49 47 55 94 115 115 122 127 144 119 119 120 138 113 120 153 114 73 80 3126
1120 2116 1155 2211 1031 2068 1146 2175 1075 2146 1340 3281 1384 3243 1381 2515 1399 2981 1744 3459 1715 4076 1709 4796 1599 4286 1813 4190 1381 3982 1429 4742 1235 3462 1386 3646 1309 3564 1568 4376 1371 5634 1394 4622 1437 4139 1287 5523 1460 4467 1510 5789 1155 3604 1853 7638 70686 170399
%
Figure 4
92
pm stv vnk opm um tvm min ym mmm sm ktm lm stm om vm
Share % 0,0 0,1 0,1 1,4 1,4 3,7 4,3 6,6 7,2 7,4 9,2 10,5 13,0 15,4 19,6
Figure 5
puv suv pev uav siv lav tyv hav stv mmv ymv vav liv tav
Share % 0 0 0,9 0,9 1,7 5,8 6,0 7,0 7,2 7,6 7,7 9,5 11,8 33,9
Figure 6 Topic Laws 30 2 24 30 28 60 31 65 8 68 17 92 18 103 21 135 16 148 2 160 13 164
93
4 19 7 6 10 20 14 3 12 1 15 5
186 251 255 301 313 335 384 490 856 921 936 1075
Figure 7
5 15 1 12 3 14 20 10 6 7 19 4 13 2 16 21 18 17 8 31 28 24 30
Non-EU 983 720 816 774 460 359 315 251 295 194 226 156 160 149 145 126 86 70 50 63 59 29 2
EU 92 216 105 82 30 25 20 62 6 61 25 30 4 11 3 9 17 22 18 2 1 1 0
94
5. Are French laws written in Brussels? The Limited Europeanization of Lawmaking in France and its Implications Sylvain Brouard, Olivier Costa, Eric Kerrouche Centre Emile Durkheim – Sciences Po Bordeaux
Evaluating the power of national political systems in the European Union (EU) and the cross influence of national and European levels of governance strongly divides scholars, politicians and citizens. Since the beginning of the 1990s, public debate surrounding European integration has often featured questions concerning its impact on the national political system and especially on the balance of power between legislative and executive bodies. The “democratic deficit” concept stems from the hypothesis that reinforcement of the European Community was weakening parliaments and giving increasing power to governments (Moravcsik 1994) via the EC Council and European Council. On another dimension, the work of Majone (1996) on the “European regulatory state” follows the famous statement of Jacques Delors about the dominance of EU over national legislation. Nevertheless, in the academic field, there is still no general agreement on the identity of “winners” and “losers” in European integration, as evidenced in the various integration theories and their divergent conclusions. One of the reasons for these ongoing controversies is the lack of solid and systematic data on Europeanization, and especially Europeanization of national laws. Using data gathered over the course of the French Agenda Project (Baumgartner, Brouard and Grossman, 2009), this chapter provides an empirical investigation of the level of Europeanization of French law-making and its evolution between 1986 and 2007. We use four different indicators to account for the phenomenon: the number of laws citing the European Union, the number of laws implementing European directives, the number of European directives implemented by bills, and the proportion of EU directives implemented by bills. In this way, we explore the variation of Europeanization according to multiple indicators. This data enables us to evaluate the impact of the EU on domestic institutions and political systems, not by looking at constitutional or institutional rules but at effective institutional outcomes. The common claim is that domestic law making is severely constrained by the EU and 95
that consequently, national Parliaments, the French Parliament included, are rigorously weakened by the European integration process. The idea that laws are written in Brussels is more widespread than ever in France, as recently illustrated by the campaign for the European Constitutional Treaty referendum. Following this logic, the impact of French “rationalized Parliamentarianism” would be intensified by European integration, and thereby the legislative powers of the French Parliament drastically limited. The way in which growing EU legislative production constrains (or fails to constrain) domestic legislative production must be empirically evaluated in order to understand not just the potentially altered balance between the legislature and the executive in France, but also how domestic and EU political systems are articulated more generally. In the first part of the chapter, we briefly present the lawmaking process in France. In the third part, we will define and operationalize the concept of Europeanization. We then describe the growing but limited Europeanization of legislative production in France. Finally, we analyze what the pattern of Europeanization of French lawmaking means for French domestic institutions and their relationships to the EU.
I. Lawmaking, EU decision-making and Parliament under the fifth republic.
The 1958 Fifth Republican constitution, which was drawn up under de Gaulle’s influence, was intended to create a system that would prevent a recurrence of the instability of the Fourth Republic. A clearly defined specification of powers, especially those vested in the executive, was designed to play an essential part in ensuring a smooth transition. The success of the Fifth Republic is generally attributed to two main factors: First, there was the creation of a strong presidency, the foremost concern in the mind of its founder, Charles de Gaulle. Second, the ‘rationalization’ of parliamentarism was designed to fortify the government’s position and a variety of institutional arrangements (Huber 1996; Brouard 2011) were introduced to that end. This rationalization has deeply weakened the National Assembly. Huber writes that “the rules included in the Constitution to strengthen the French government against the legislature seem formidable. The Constitution contains provisions that grant control of the legislative agenda to the government, that limit the right of deputies to submit and vote amendments, that limit opportunities for deputies to gain information and expertise, and that even limit opportunities for members of parliament to vote on bills themselves” (Huber 1996). Until 2009, the number of parliamentary committees was limited to six in order to limit 96
legislative scrutiny. The two most known procedures are the package vote (vote bloqué; art. 44.3) and the confidence vote procedure (art. 49.3). The latter has the specific role of pushing through policy proposals while the former forces the assembly to have a single vote on an entire legislative proposal, rather than adopting it article by article. This closed rule is also systematically applied to the “law ratifying international agreements or conventions”. Accounting for 40% of the law production, this type of legislation is all but rare in France (Figure 1).
Figure 1: French legislative production by types of laws
Three other types of French laws exist. Constitutional laws deal with constitutional amendments. There are more constitutional laws in France than in most of the democracies, especially in recent years. The so called “organic laws” (lois organiques) implement institutional reforms in areas explicitly mentioned in the French constitution. Finally, ordinary laws are the most frequent type. There is another component to take into account regarding the way laws are adopted, as some laws are adopted by referendum. Beyond government control of the legislative agenda, the main peculiarity of Fifth Republic that must be underscored is the distinction between law (art. 34) and regulation (art. 37), which explicitly limits the domain of law. The remaining part is de jure the domain of regulation, controlled by the government. Thus, given
97
these various institutional settings, the 1958 constitution reinforced the predominance of the government compared to the two previous Republics. Since these rules were written into the Constitution, the Members of Parliament cannot easily change or get rid of them. These factors explain why, under the Fifth Republic, the National Assembly is often regarded as one of the weakest legislatures in any modern democracy. These constitutional rules also explain why both cabinet dominance and specialization have become two major features of the Fifth Republic. The National Assembly is perhaps even weaker in the field of European involvement. The arguments given to illustrate this fact seem definitive: Bergman estimated in 2000 that the French parliament was tenth out of fifteen national parliaments of the EU in terms of activity and influence (Bergman 2000). Other scholars (Raunio and Wiberg 2000) also positioned the two French assemblies in the category of “weak” parliaments regarding their European influence. During the first decades of the Fifth Republic, the Parliament paid very limited attention to European integration. The first direct election of the European Parliament (EP) in 1979, the Single Market Program (31th December 1992) and, above all, the progressive recognition of the superiority of EC legislation by the Conseil d’Etat (the highest administrative court) and the Constitutional Council were the triggers that finally caught the attention of deputies and senators. Before the Maastricht Treaty and the related amendment to the French Constitution, the power of scrutiny afforded to the French parliament was limited to treaty ratifications. Although European integration is now clearly different from other areas of international relations, it remains rooted in the area of diplomacy and foreign affairs (Grossman and Sauger, 2007).1
The first body that monitored French European policy-making at the parliamentary level, the Delegation for European Affairs, was established by the Senate in 1973 and by the National Assembly in 1979. It is composed of 18 members from each chamber. This is comparable to the organ established by the German Parliament in 1957. The design of the Délégation is also its a weakness because the number of permanent committees is constitutionally unchangeable. From its creation until the end of the 1990s, its role was extremely limited. The permanent committees were not very willing to make use of the expertise provided by these new Délégation pour les Communautés européennes, and so its performance has been more or less a failure.
1
Grossman and Sauger also underline that the European Affairs Minister has always been a sub-secretary or « ministre délégué » at the French Ministry of foreign affairs. Emiliano Grossman and Nicolas Sauger, op. cit.
98
A change took place at the beginning of the 1990s: in 1990 membership in the Delegation for European Affairs doubled, governmental information policy became more systematic, and the ministers for European Affairs gained an informed parliamentary forum to present governmental policy in European matters in periodic hearings.2 Because of the impact of European integration on national policy making, it is not surprising that national parliaments would try to increase their influence and their scrutiny powers in that field. This was exactly what happened in the French case. Each treaty ratification round has been linked to a substantial improvement in the scrutinizing powers of the two chambers (Grossman and Sauger, op. cit.). But the real shift occurred in 1992, with the constitutional reform prior to the ratification of the Maastricht Treaty, which had to pass through Parliament. A revised system of parliamentary scrutiny of European secondary law-making emerged based on the new article 88-4 of the Constitution. Since the reform implementing the constitutional revision for the ratification of the Amsterdam Treaty in 1999, this article states that3: "The Government shall lay before the National Assembly and the Senate any drafts of or proposals for instruments of the European Communities or the European Union containing provisions which are matters for statute as soon as they have been transmitted to the Council of the European Union. It may also lay before them other drafts of or proposals for instruments or any document issuing from a European Union institution. In the manner laid down by the rules of procedure of each assembly, resolutions may be passed, even if Parliament is not in session, on the drafts, proposals or documents referred to in the preceding paragraph." With this reform/ amendment the French Parliament had, for the first time in its existence, the constitutional right to be informed of, to scrutinize, and to intervene into the conduct of French European Community policy (Rizzuto, 1996). The ‘resolutions’ – which are only declarative and do not have compulsory effects – allow the two Chambers to share their opinions with the Government on acts resulting in French legislation (Rozenberg, Szukala, 2001). The scope of the article 88.4 has been extended over time (especially in 1992 and 1999), allowing the national Assembly to scrutinize the whole EU policy process in spite of government reluctance. As some scholars have indicated, scrutiny of EU affairs appears problematic in nearly all member states: most national parliaments have not just delegated the management of EU politics to their respective government; they have also given up on even attempting to merely influence the government on those issues (Grossman and Sauger 2007). Moreover, it is an understatement to write that French MPs no longer make intensive use of their right to adopt resolutions in EU
2
Loi N°. 90-385, 1990. Senate on an average 12-15 auditions per year; National Assembly 18-20 auditions per year (1993-2003). 3 Loi Constitutionnelle N°. 99-49, 1999. The article 88-4 was revised in 2005, but the implementation of the new provisions depends on the enforcement of the ECT.
99
matters: on average, only 2 to 5 per cent of the transmitted proposals are subject to resolutions and very few resolutions are tabled during public sessions. Nevertheless, this does not mean that European integration has automatically weakened the French Parliament. European integration is not the only factor responsible for this. The effect of European integration on Parliament depends on the specific domestic institutional context. In fact, in France, the transposition is a shared competence between government and parliament according to the distribution of roles outlined in article 34 and 37 of the 1958 Constitution. In other words, the government is supposed to implement measures that concern regulation, while the National Assembly and the Senate are responsible for legislative aspects. The distinction between law and regulation is made by the Conseil d’Etat. So, the transposition of European directives is not only constraining the French Parliament’s activities but also the French government’s activities. The extent to which each of them is effectively constrained by the EU is a challenging empirical question in the French case. Beyond the scope of Europeanization of French statutes, the way MPs effectively deal with Europeanized laws deserves specific investigation. Do MPs passively accept Europeanization? Or do they try to shape it when they have an opportunity during the legislative process?
II. The growing but limited Europeanization of the French laws The theoretical and empirical literature on Europeanization, which has represented a major trend in European studies since the beginning of the 1990s, addresses a broad set of issues (Graziano & Vonk 2007). It not only analyzes the EU’s influence on member states (so-called “downloading” or “top-down” Europeanization), but also the influence of member states and their components (administrations, regions, civil society organizations, companies…) on EU policy making (“uploading” or “bottom-up” Europeanization4), as well as the transfer of policy from one country to several others5. The top-down approach has been criticized in that it assumes a passive role for member states. Member states, however, are not only “takers” of European norms, models or pressures, but also active “shapers” of those forces. Europeanization studies have also widened their field of investigation: today they are no longer limited to the
4
Simon Bulmer and Claudio Radaelli, ‘The Europeanization of National Policy’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, p. 340 5 Claudio Radaelli, op. cit, p. 27
100
EU’s impact on politics and policy-making, but also analyze effects on processes, policies and institutions (policy, polity and politics).6
1. Operational definitions of the Europeanization of French laws Given the complexity of Europeanization, we have adopted an approach based on the plurality of the operationalization of the phenomenon. So, four ways were explored to build variables related to Europeanization.
Europeanization as a reference to the EU: the EU link First, Europeanization may be conceived as a link between the EU and a specific statute. While legislating, lawmakers may use the EU as a frame. We explore the extent of Europeanization by looking for EU related terms referenced in the laws. We have used the set of 21 keywords and acronyms described in the introduction of this volume in order to estimate the reference to the EU in French statutes. Unfortunately the full text of French laws is only available on the Internet after 1990. Nevertheless, each law is associated with a set of keywords in the official on-line French law database Legifrance (www.legifrance.fr), which covers legislation since 1986. We ran a word search of our selected European keywords among the keywords associated with each law. We then explored the differences between both searches. First, we discovered that the search engine of the official website Legifrance has some difficulties coping with the full text search. For example, there were no hits for six important keywords (EU, EC, ECSC, EAEC EMU, EMS), even though they appeared in the text. Briefly stated, the results of the full text search were not reliable enough. Second the hit rate was lower for the full text search than for the keywords search. In addition to being more reliable and expanding the time span, using the keywords search did not underestimate the scope of the Europeanization. Thus we have built our proxy for the EU link by searching the 21 keywords in the list of keywords associated to each French law. The EU reference keyword variable is a dichotomous variable indicating whether or not the keywords associated to a law contain at least one of the selected EU keywords.
6
Tanja Börzel and Thomas Risse, ‘Conceptualizing the Domestic Impact of Europe’, in: K. Featherstone and C. Radaelli (eds.), The Politics of Europeanization, Oxford University Press, Oxford, 2003, p. 60
101
Europeanization as a constraint Europeanization may be defined as the implementation of a binding European decision at the national level. Therefore, the dataset we use systematically links French laws and European decisions. For every law adopted by the French Parliament between 1986 and 2006, the EU constraint variable coded whether or not the law implemented a European decision, which could be a directive, treaty, or an international agreement. When a law was an indirect result of a European decision, we did not code it as a law affected by Europeanization. For example, in 2007, a French law reformed the Central Bank’s internal organization7. The Monetary Policy Council adopted in 1993 to improve the autonomy of the French Central Bank was abolished due to the implementation of the Maastricht Treaty and the establishment of the European Central Bank. The change was not compulsory. The internal reform was an indirect but not binding change related to European Integration. In fact, French MPs could have left this specific dimension of the internal organization of the French central bank unchanged, and they had done precisely that for several years. In this example, European integration gives opportunities or incentives to legislate but does not constrain lawmaking. It would be misleading to extend the EU constraint variable to these types of cases. A similar situation occurs when a law is adopted to adapt a national statute to a non-binding European norm. For example, in 2006, a parliamentary bill about plant variety rights added five years to the national property rights8. In fact, the new law allows the same length of property rights for plant variety that is acknowledged by the European Union. The text is explicitly about legal points that are not determined by European Law or international treaties. Thus there was no legal obligation to achieve the same level of protection of property rights. Protecting certain plant varieties and remaining competitive in this economic sector were the motives behind this bill. It could be argued that it is a form of Europeanization but it is not part of the narrow meaning of Europeanization we are interested in here. Accordingly, we have not included this modality of Europeanization, as it is, in our eyes, more a result of the competitive globalization process. We extend the EU constraint variable to another type of European constraint that includes directives, treaties and agreements. Some French laws are also motivated by the Court’s decisions, and by a desire to follow the lead of the European Commission (avis motivé). For example, the law about « concessions d’aménagement »9 (concessions of development) in 2005 7
Loi portant diverses dispositions intéressant la Banque de France n° 2007-212 Loi relative aux obtentions végétales n°2006-236. 9 Loi relative aux concessions d'aménagement n° 2005-809. 8
102
is an answer to the judgment C-324/98 Telaustria in 2000; the European Commission also addressed on December 7th, 2004 the incompatibility between French Law and European Law. In this example, we see the process of Europeanization that we are trying to capture with our narrow definition. Europeanization does not always dictate every part of a « Europeanized » law. First, it is possible for a law to be only partly dedicated to the implementation of EU legislation, while its other parts address domestic issues. Second, the measures of implementation can be adapted to the specificities of the country. But a specific legal frame is compulsory for any legal implementation. From this point of view, what matters is what is constraining in the process of Europeanization. Consequently, in our perspective, Europeanization happens only by an a posteriori transposition of European norms. Our perspective on Europeanization is a good way to measure European constraint because there are cases in which legislating is not necessary given the pre-existing laws or the institutional rules. In fact, some European norms only generalize the existent state of law in a specific country. When one analyzes the implementation of EU directives, for example, some cases may be found in which laws previously adopted by the country have already implemented the EU directives. For example, the 1992 EU directive 92-44 concerning the application of open network provision to leased telecommunication lines had already been partially transposed in France by an earlier 1990 law regulating telecommunications. In another example, a French law related to the employee’s advisor, voted in 1991, contains the legal basis for the implementation of an EU directive dedicated to safeguarding the rights of employees in the event of business transfers – this EU directive, however wasn’t enacted until 1998, seven years after the French law. Thus, the huge increase in EU norms does not automatically lead to a similar increase in Europeanized laws. The level and evolution of the discrepancy between both trends provide crucial insights into not just the relationships between EU and national institutions, but also into the impact that the EU has on domestic political institutions. What we measure with our approach to Europeanization is only the European constraint on lawmaking in a specific polity. Beyond those two main approaches, for all Europeanized laws, we have also coded the number of European directives, treaties, agreements or judicial decisions implemented by a given statute10. According to the same rule, we also used a dataset established by Emiliano Grossman11 10
The data about the Europeanization were collected from the reading of the law and from the use of the official French law database Legifrance available online. We attached to a specific French law the European texts that are either quoted in the law either appeared as references (« texte source ») in the official French law database. 11 E. Grossman, « La résistance comme opportunité : stratégies de l’exécutif français face à l’Europe », Revue internationale de politique comparée, 2008/4 (Volume 15)
103
that associates each directive with the presence, or absence, of at least one law among the texts of different domestic transposition instruments. Finally, both approaches of Europeanization are theoretically different but they also identify different set of laws. Contrary to our expectations, Europeanization as a constraint is not simply a sub-set of Europeanization as a reference. In fact 5.8% of the laws have reference to the EU without being constraining and 3.6% of the statutes are constrained by EU without containing one of the EU keywords. So it might be argued that both approaches are complementary in order to estimate the scope of the Europeanization of the French laws. This new dichotomous variable EU impulse combines the EU constraint variable and the EU reference keyword. The extent of Europeanization can be now easily measured with our four different variables: what is the scope of the EU impulse? What is the proportion of French laws with a reference to the EU? What proportion of French laws are constrained by the implementation of EU directives? Does the expansion of European norms lead to increasing constraints on the French legislative process? 2. The scope of the Europeanization of French Laws
From 1986 until the end of 2007, 369 laws out of 1968 (18.75%) had an EU impulse in France. The EU was referred to in 15.1% of the French statutes, and the implementation of EU legislation concerns 13% of the laws. Yearly variation aside, there has been an increase in Europeanization since 1986. In 1990, the average annual percentage of laws with an EU impulse was around 10%. From less than 7 % in 1986, the proportion of the total number of laws with an EU impulse grew to 18. 75 % in 2007. Furthermore, in 21 years there was no clear decrease in the global trend of Europeanization (see Figure 2). The trend is the same for Europeanization as a reference or as a constraint.
Figure 2: Europeanization of French laws (1986-2007)
104
The bulk of the Europeanized laws are the ordinary laws that accounts for 55.6% of the laws with an EU impulse. 40% of the laws referring to the EU are laws ratifying international treaties but the proportion reaches 48% when the EU is approached as a constraint. This pattern is part of a more general trend of the growing importance of laws ratifying international treaties in French law production. It is a sign of the impact of globalization on French law production. Regardless of the indicator, the most Europeanized part of French law production is the rarer type of law and the highest one in the hierarchy of norms: 41-2% of the 17 constitutional laws have an EU impulse. The lowest level of Europeanization is found among the organic laws. Only 3.3% of them have an EU impulse. Non-domestic statutes (18.59%) and ordinary laws (19.39%) have a similar occurrence of EU impulses. By using the policy content coding system of the Policy Agendas Project (Baumgartner and Jones, 1993), it is possible for us to analyze the variation of Europeanization according to the different topics. The greatest number of laws with an EU impulse (152) is related to international affairs. In fact, most of the Europeanized laws are dedicated to regulating the way the EU works and the way the EU interacts with others countries and regional organizations. While some of these international agreements introduce important policy consequences, they are still not the most constraining part of Europeanization (see figure 3). In proportion, international affairs, of which only 20% can be described as Europeanized, is not the most Europeanized policy sector. Unsurprisingly, the topics related to the establishment of a European free market are the most 105
Europeanized: Space, Science and Technology (45%), Banking, Finance and Domestic Commerce (39.8%), Labor and Employment (24%), Macroeconomics (27.5%) and two old European policies: Agriculture (36.7%) and Energy (20%). Policy changes in these five policy sectors are only partially driven by EU legislation. Environment (35.5%), Health (25%) and Immigration (20%) are new policy sectors that are significantly Europeanized. This is a sign of the new attention given to these issues at the European level. As for other policies, however, the impact of Europeanization is low: 7.6% for Justice and Crime, 3.9% for social welfare, 2.5% for Culture. French exceptionalism has not been as severely affected as expected by Europeanization. The prediction of Jacques Delors in 1988 at the British Trade Unions Congress that more 80% of economic and social legislation in European countries would be Europeanized in the next ten years is not supported by the data.
45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
1. Macroeconomics 2. Civil Rights 3. Health 4. Agriculture 5. Labor 6. Education 7. Environment 8. Energy 9. Immigration 10. Transportation 12. Crime 13. Social Welfare 14. Housing Issues 15. Finance &… 16. Defense 17. Technologie and… 18. Foreign Trade 19. International Affairs 20. Gouvernment 21. Public Lands 23. Culture
Figure 3. Proportion (in percent) of laws with an EU impulse per policy area
Thus, international affairs are the biggest component of the Europeanization of French laws. Around half of all Europeanized law involves domestic policy issues. Among them, for the last 21 years, EU norms have dictated a significant but limited part of the changes in French laws in nearly one third of the domestic policy spectrum. Even in the most Europeanized policy sector, the majority of the laws that the French Parliament votes on is neither implementing EU decisions nor referring to EU. So empirically, the situation is far from the prominent belief in France that the French Parliament is constrained to merely voting on texts already decided in Brussels. The way in which the impact of the EU on domestic politics is commonly depicted 106
should be drastically changed. The empirical results reverse the Delors’ myth: 80% of the French statutes are not Europeanized!
2. Dynamics of EU Directives and the constrained Europeanization of Domestic Laws in France After studying the level and nature of the Europeanization of French lawmaking, it is necessary to compare the evolution of the production of French domestic law and EU directives. The focus on directive production at the EU level is relevant in order to estimate the constraints induced by Europeanization – because the directives are the part of EU norms that require transposition. As the birth of the Fifth French Republic and the beginning of European integration happened roughly at the same time, the trends can easily be compared since 1959. The patterns of the evolution in the production of a) law in France and b) EU directives are clearly different. At an aggregate level, EU directives have been less numerous than French laws since 1959. The average number of EU directives per year (66.1) is lower than the average number of French laws (92.3). Nevertheless, since the first step of European integration in 1958 and the first directive in 1959, the number of EU directives has increased. Ten years after the beginning of European integration, there were only 20 EU directives per year. In 1979 more than 80 directives were enacted. Furthermore, after a period of global stability during the 1980s, a new wave of directive production in the early 1990s led to more than 100 EU directives per year. The number of directives even reached as high as 122 in 1992. Since then, the EU rarely produces less than 80 directives per year, and it established a new record in 2006 with 136 directives. This evolution underlines the growth of European integration and the different steps of this process. The trend for the production of law in France, however, is radically different. There are some important yearly variations. They are first of all linked to the relative length of the parliamentary sessions according to the presence of national elections or not. On average, there was a progressive increase in the production of law in France until the beginning of the 1980s. Then, a sharp decrease reduced the average production of laws to around 90 per year, excluding a three year spike in the second half of the 1980s. In sum, the ratio of domestic laws to EU directives is decreasing at a regular pace. Historically, the dynamics of French law production and EU directive production appear to be divergent and provide quantitative support for the hypothesis that the EU places increasing constraints on domestic lawmaking. Be that as it may, if the focus of the analysis took into account the Single European Act (1986) as the point of departure, the last twenty-one years (1986-2006) are characterized by a higher 107
level of similarity between the production of French laws and EU directives. Indeed, the average number of EU directives per year (98.9) exceeds the average number of French laws per year (88.4). With the Single European Act, the relative productivity of European and French institutions is reversed. Since then, the perceived dominance of the EU over domestic policy making has become a common trope, readily denounced in different ways (loss of sovereignty, rise of a European bureaucratic state, etc.). The empirical challenge is how to estimate the effect of these trends: is there an important and increasing European constraint on French law-making? Despite this dramatic change in the relative dynamics of domestic law production and EU directive production, the domestic laws do not present a different pattern and trend of Europeanization. There are important variations in the yearly percentage of Europeanized laws. Nevertheless, since the beginning of 1990’s, this indicator has seldom been under 10%. The last three years reached a particularly high level of Europeanization with around 40% of Europeanized laws in 2004 and 2005 and more than 20% in 2006. Those specific figures are the result of a conscious effort on the part of the French government to cope with delays in the implementation of EU legislation. Moreover, the mobile average proportion of Europeanized domestic laws has been growing since 1986. Every year, the proportion of Europeanized laws out of the total number of domestic laws since 1986 is higher, but remains moderate (13.2%). Since 1986, more than 86% of domestic laws do not implement EU legislation.
Figure 4: The level of EU constraint on French law-making (1986-2007) nb of europeanized domestic laws
Europeanization of domestic laws
% of europeanized domestic laws since 1986 yearly % of europeanized domestic laws
60
50
40
30
20
10
0 198619871988198919901991199219931994199519961997199819992000200120022003200420052006
108
Using another indicator - the number of EU directives implemented by domestic laws - the analysis shows a similar level and evolution of Europeanization. Both indicators yield highly correlated results. The main result is that beyond yearly variation, the level of Europeanization of domestic laws has increased since 1986. But, out of the nearly one hundred EU directives enacted each year, only 14.5 are implemented by a French law. So the same conclusion can be drawn: the level of Europeanization is increasing, but it remains limited. Not all EU directives result in a new law implementing it in France. In fact, among the EU directives adopted between 1992 and 2004, only a minority were transposed with a law. More than 90% of the directives were not transposed by a statute. In other words, the inflation of EU directives has not been associated with a sharp increase of domestic laws implementing EU directives in France.
Figure 5: The implementation of the 1992-2004 EU directives in France Nb of directives implemented by at least a law Implementation of the 1992-2004 EU directives in France
mean number of directives implemented by at least a law % of directives implemented without any law
100 90 80 70 60 50 40 30 20 10 0 1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
III. The pattern of Europeanization and French domestic institutions
109
The pattern of the Europeanization of lawmaking is a very useful indicator for estimating the effect of the EU on the domestic institutional balance and for understanding the distinction between French institutions and the EU.
1. The effect of Europeanization on domestic institutions: No double penalty for the French Parliament.
In this part, before looking at the effective impact of “downloading” Europeanization on law making in France, we will describe the institutional settings of the Parliament in the law making process and in the specific issues related to European affairs. First, we want to underscore that the Europeanization of French statutes is weak, or weaker than expected according to the Delors’ myth, because most transposition is assumed by the French government given its institutional supremacy in the regulation process. Although it considerably limits the role of Parliament, this process is, strictly speaking, independent from the European integration process. In fact, if the Europeanization of the French laws is correctly captured, their weak number and proportion in the production of law support the hypothesis of a limited Europeanization of lawmaking and invalidate the hypothesis of the French Parliament having been weakened by European integration. Whatever the origin of the policy changes, similar policy change would always be implemented through regulations. Hence, the French Parliament is comparatively weak because of the French Fifth Republic’s parlementarisme rationnalisé, but there is no double penalty induced by Europeanization. From our point of view, and contrary to Moravcsik’s argument (1994) and common wisdom, the government is paradoxically more affected by European integration than the Parliament: a growing part of the policy changes enacted by the French executive through regulations is dictated by the EU. Second, as in other political systems studied in this book, Europeanization is not only a constraint for French MPs. It also provides incentives for a more thorough scrutiny of the bills in Parliament. It is rational for MPs that fear the loss of power induced by Europeanization to check and control it. Amending the content of the Europeanized laws would thus be an obvious indicator of a parliamentary reaction to the Europeanization. Some Europeanized French laws may not be amended. In fact the laws ratifying agreements are adopted under closed rule. Nevertheless beyond these laws, Europeanization has also affected the most important part of the French law-making, the ordinary laws. Because an EU directive sets the objectives to reach, each member state must adopt a policy with the relevant means in order to reach them on its territory. 110
Hence changes through amendments are theoretically possible. Are they empirically observable? Do MPs react to the EU incentives by scrutinizing the Europeanized laws more thoroughly? Unfortunately, the data on the number of amendments per law in France is unfortunately only available after 1999. On average, the 97 Europeanized ordinary laws have 131 amendments whereas the 287 non-Europeanized laws only have 54 amendments. This result supports the idea of a strategic reaction of MPs in face of Europeanization. As shown in Figure 6, the higher level of the average number of amendments is consistent over time. The mean difference is also statistically significant (p < .000). Even if in the French system all amendments are not from MPs, amendments are a reliable indicator of parliamentary scrutiny. In fact, despite the rationalized parliamentarism that gives huge agenda control to the government, most of the adopted amendments (80% on average since 1969) are proposed by the committees. Furthermore, it could also be argued that because the government introduced the bulk of adopted laws (88% on average since 1986), government amendments are also the sign of an influence of the MPs as a compromise between government and parliamentary majority. The dominance of the executive is even more pronounced on the laws with an European constraint: 95% of them are government bills.
Figure 6: Average number of amendments per law 300 250 Europeanized Domestic laws
200 150
NonEuropeanized Domestic laws
100
Domestic laws 50 0 07 20 06 20 05 20 04 20 03 20 02 20 01 20 00 20 99 19
The number of amendments may also be affected by factors other than Europeanization. A first obvious factor is the length of a law. The longer a statute is, the higher the number of amendments is- provided that amendments are allowed. So the difference in the average number 111
of amendments may only be an artifact of the length of the law: on average Europeanized laws may be longer than non-Europeanized laws and consequently may have more amendments. In order to control for that fact, we run a OLS regression with the number of amendments as the dependent variable and the Europeanization variable and the size of the law (in number of characters) as independent variables. The coefficients of both independent variables are statistically significant. As expected, longer laws are associated with a higher the number of amendments. But more importantly, the Europeanization variable is still significant. Everything else being equal, Europeanization induces 8.7 amendments more on a given laws. On average, a law has 36.67 amendments. Thus, the Europeanization effect is far from negligible, bringing nearly a quarter of the average number of amendments on ordinary laws since 1999. Thus, there is no double penalty for the French Parliament. In a system with executive dominance, Europeanization constrains the government first and foremost. Furthermore MPs are not simply compliant with Europeanization. They react. Europeanization is a red flag that gives them incentives to scrutinize bills and amend them. 2. Europeanization and the domestic political system: behind the myth of a European policy dictator, the reality of a new veto player. Even if we only focus on the concrete and legal effects of Europeanization (i.e. institutional and policy changes), we must take into account the various degrees of the phenomenon. To analyze that aspect of Europeanization, Börzel considers the notion of “misfit”, described as a necessary condition for domestic change and an explanatory factor of its degree12: as long as there is no policy or institutional misfit between the national situation and what is promoted at the European level, there will be no pressure for change13. This does not mean that national change cannot occur if the European and national levels fit well, but that such change does not necessarily arise directly from EU influence. According to Börzel, Europeanization has five different outcomes depending on the degree of misfit: “inertia” or the lack of change, “retrenchment” or increasing misfit, “absorption” or the incorporation of the EU’s influence without substantial national change, “accommodation” or national change at the margin, and “transformation” or the full replacement of policies and institutions.14 The misfit approach is useful for analyzing situations where there is a clear EU template that has to be “downloaded” if we are in a situation of “governance by hierarchy”: circumstances where the supranational institutions have a considerable amount of power delegated to them (Bulmer and Radaelli, 2005, 343). It is less 12
Tanja Börzel, ‘Europeanization: How the European Union interacts with its Member States’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, pp. 49-60 13 Ibid., p. 50 14 Ibid., pp. 58-59
112
convincing when Europeanization is a more diffuse phenomenon, as with policies ruled through open method of coordination (OMC); in this case, the absence of a specific model, as well as the possibility for multiple horizontal policy transfers and for national discretion, make misfit an unlikely mechanism for creating change15. Nevertheless, our focus on the downloading Europeanization of the production of French law fits perfectly with the ideal condition of use of the misfit approach. The extent to which Europeanization takes place depends on several factors. As we saw, there first needs be a sufficient degree of “misfit” between national and EU-proposed policy. The misfit approach was initially applied to specific policies or policy sectors. The relative level of implementation by laws versus regulation in the French case is a way of dealing with misfit at a national level. The level of Europeanization of law production is a proxy for the degree of misfit. In the French case, a higher degree of misfit constrains a policy change implemented by law whereas a lower degree only implies transposition by regulation. So, despite the widespread belief in France that there exists a high degree of misfit between French policies and the process of European integration, the data suggest the contrary: the degree of misfit in France is low and probably concentrated on a narrow but sensitive set of issues, such as market regulation, for example. Thus the core of French policies is easily compatible with EU policies. There are two reasons for this paradoxical interpretation. France is one of the founding members of the European Communities. So it has had time to learn and adjust its policy-making process in an EU-compatible way. From this point of view, France stands in sharp contrast with the new member states that had to implement the entire aquis in a very short time. Consequently, consciously or not, an anticipatory behavior arose that limited the level of potential discrepancies between France and the EU. This is without any doubt another dimension of the process of Europeanization. For the same reason, and given France’s weight in the EU, a lot of EU policies have been either heavily influenced by French preferences or introduced by French leaders. In a lot of cases, they tried to and succeeded in generalizing their policies at the EU level. Probably, the nature of the EU has changed and the French ability to shape EU outcomes has dramatically decreased over the years. This trend would also explain why the French citizens today overestimate the level of domestic policy change brought about by the EU. While there have admittedly been some clear policy misfits with occasionally important consequences, the symbolic value of these episodes in France belies the actual, high level of congruence between EU and French legislation.
15
Simon Bulmer, ‘Theorizing Europeanization’, in: P. Graziano and M. Vink (eds.), Europeanisation: new research agendas, Houndmills: Palgrave, 2007, p. 52
113
The EU’s effect on a domestic political system can be characterized between two extremes: policy dictator or veto player. In the first case, the EU constrains numerous policy changes. This is sometimes the case with the new EU member states that were former Soviet states. Countries such as these with very different regulatory traditions need to vote in a whole new set of laws in order to be able to join the EU. In these cases, the EU narrowly dictates the policy changes. At the other extreme, the misfit between EU legislation and domestic law is null. In these cases, the EU never imposes any policy change, as the current domestic legislation already perfectly fits with European legislation. This occurs, for example, when EU norms copy domestic norms. In this case, there are no Europeanized laws. Nevertheless, change in domestic laws is conditioned by the compatibility of the new policy with the current state of European law, or by changes in relevant EU norms. Thus in this hypothesis, the EU is a veto player: its agreement is needed for any domestic policy change. Between the two poles, policy dictator or veto player, given our empirical results and their meanings, the EU’s behavior towards France has been that of a veto player, more than that of a policy dictator. The EU lessens the difficulty in changing a growing set of existing French laws that became European norms. The most famous examples of European constraint on French policy making clearly fit this pattern, from the decrease of VAT for restaurants to the end of the summer-time, to the changing of the opening and closing dates of hunting season. In each case, European rules generalized at the European level the former French rules, or they had been adopted upon French initiative. The main effect of the EU on French domestic policy is to constrain and not to dictate policy changes.
Conclusion
Europeanization is first of all an important component of the globalization characterized by the increasing number of Member States’ agreements. Half of the French laws dealing with EU issues are laws ratifying international treaties and agreements between the EU members about different dimensions of EU integration or between the EU and other States. Only half of all Europeanized laws concerns domestic policy issues. Among them, for the last 21 years, EU norms have dictated a significant but moderate part of the policy change. Even in the most Europeanized policy sector, a majority of the laws voted by the French Parliament are not related to the implementation of EU decisions. In domestic law, whatever indicator used– the number of laws implementing EU legislation or the amount of EU legislation implemented by laws – the result is similar and underscores an increasing but moderate Europeanization. For the last 21 114
years, the most important influence of the EU on France has not, contrary to popular belief, been the ‘forced’ adoption of new French laws. The EU has only dictated a small, albeit growing, portion of new French laws Our findings are far from the traditional intuition that 60 or 70 % of domestic French laws are induced by the EU (Maïa, 2005), and from the famous statement of Delors – which is repeated ad libidum by politicians and journalists at the occasion of each campaign for European elections. The EU is not primarily a top-down constraint on law-making. In fact, the results in the French case match those in the Danish case, leading to an obvious conclusion: the weight or power of the « European regulatory state » (Majone) should be examined more carefully, and with studies based on verifiable data as Levy-Faure convincingly advocated (2007). Given the role of the Parliament and of the government in France, Europeanization does not constrain the Parliament, but rather the government. In fact, given the inflation of EU directives, Europeanization is most constraining with regard to regulations. In the country of rationalized Parliamentarianism, despite the exclusion of parliaments from EU-level decision-making (Auel & Benz 2005), the current Europeanization of the production of law does not create a double penalty for the weak French Parliament. Conversely, Europeanization fuels parliamentary involvement in the drafting of the French statutes as shown by the fact that Europeanized laws are more amended (when allowed) than non-Europeanized ones. Both results have a very clear meaning about the process of Europeanization and the way the EU interacts with the French political system. For the last 21 years, a moderate level of misfit led to a low level of Europeanization in France. Conversely, the changes induced by the EU are overestimated in France because the episodes where there were conflicting policies occasionally had important consequences, leading to a symbolic value that exaggerated the actual influence of the EU on domestic French laws. Most of the time, EU and French legislation are congruent. Only the key changes of French policies necessitate a new law. When the misfit of a current domestic policy is weak, only some new adapted regulations are required. In short, the influence of the EU on the domestic political system can be characterized as falling along a continuum between two extremes: policy dictator or veto player. Between these two poles, given the empirical results presented here, the EU has acted more as a veto player than as a policy dictator. The EU makes it increasingly difficult to change a growing set of existing French laws that had already become European norms. The main effect of the EU on French domestic policy is not to dictate policy changes but to limit policy changes or decrease the scope of the alternative policies. 115
References Auel Katrine and Benz Arthur, « The Politics of Adaptation: The Europeanisation of National Parliamentary Systems », The Journal of Legislative Studies, Vol.11, No.3/4, Autumn/Winter 2005, pp.372–393 Baumgartner F, S. Brouard and E. Grossman (2009), “Agenda-setting dynamics in France: revisiting the ‘partisan hypothesis’ French Politics”, French Politics, 7(2), 75-95. Bergman T., ‘National Parliaments and EU Affairs Committees: Notes on Empirical Variation and Competing Explanations’, Journal of European Public Policy, 4/3 (1997), pp.373–87 Bergman, Torbjörn: “The European Union as the next step of delegation and accountability”, European Journal of Political Research, vol. 37, n° 3, May 2000, pp. 415-429. Börzel, Tanja, ‘Europeanization: How the European Union interacts with its Member States’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, pp. 45-69 Börzel, Tanja A. (2000a). ‘Improving Compliance through Domestic Mobilisation? New Instruments and Effectiveness of Implementation in Spain’, in Christoph Knill, and Andrea Lenschow (eds.), Implementing EU Environmental Policy. New Directions and Old Problems. Manchester: Manchester University Press, 222–50. Börzel, Tanja A. (2001b). ‘Non-Compliance in the European Union: Pathology or Statistical Artifact?’, Journal of European Public Policy, 8:5, 803–24. Börzel, Tanja A. (2002a). ‘Pace-Setting, Foot-Dragging, and Fence-Sitting: Member State Responses to Europeanization’, Journal of Common Market Studies, 40:2, 193–214. Börzel, Tanja and Risse, Thomas, 2003, ‘Conceptualizing the Domestic Impact of Europe’, in: K. Featherstone and C. Radaelli (eds.), The Politics of Europeanization, Oxford University Press, Oxford, pp. 57-80 Brouard, Sylvain, 2007, « The systematic institutional advantage of government in law-making », in Rash B. E. & Tsebelis G. (eds.), The Role of Governments in Legislative Agenda Setting, London : Palgrave, forthcoming. Bulmer, Simon and Lequesne, Christian, ‘The European Union and its Member States: An Overview’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, pp. 9-20 Bulmer, Simon and Radaelli, Claudio, ‘The Europeanization of National Policy’, in: S. Bulmer and C. Lequesne (eds.), The Member States of the European Union, Oxford University Press, Oxford, 2005, pp. 338-359 Bulmer, Simon, ‘Theorizing Europeanization’, in: P. Graziano and M. Vink (eds.), Europeanisation: new research agendas, Houndmills: Palgrave, 2007, pp. 47-58 116
Cananea, Giancinto della (2000). ‘Italy’, in Hussein Kassim, B. Guy Peters and Vincent Wright (eds.), The National Co-ordination of EU Policy. The Domestic Level. Oxford: Oxford University Press, 99–113. Chagnollaud D. and J.L. Quermonne, 2000, La cinquième République, Paris: Flammarion. Commission Européenne (2001). “Rapport sur l’application du droit communautaire par les Etats membres sur le contrôle de celle-ci par la Commission, contenant des recommandations en vue de les améliorer du point de vue de la gouvernance démocratique européenne”, Bruxelles: Secrétariat Général de la Commission des Communautés Européennes, http://europa.eu.int/comm/governance/areas/studies/index_en.htm. Cowles Maria Green, James Caporaso and Thomas Risse (eds.) (2001), Transforming Europe: Europeanization and Domestic Change. Ithaca, NY: Cornell University Press. Dimitrakopoulos, Dionyssis (2001a). ‘Incrementalism and Path Dependence: European Integration and Institutional Change in National Parliaments’, Journal of Common Market Studies, 39:3, 405–22. Dimitrakopoulos, Dionyssis (2001b). ‘The Transposition of EU Law: ‘‘Post-decisional Politics’’ and Institutional Economy’, European Law Journal, 7:4, 442–58. Duina Francesco and Tapio Raunio (2007), "The open method of co-ordination and national parliaments: further marginalization or new opportuinties ?", Journal of European Public Policies, 14-4, p. 489-506 Duina, Francesco G. (1997). ‘Explaining Legal Implementation in the European Union’, International Journal of the Sociology of Law, 25:2, 155–79. Duina, Francesco G. (1999). Harmonizing Europe: Nation-States within the Common Market. Albany: State University of New York Press. Elgie, Robert: Semi-presidentialism in Europe, Oxford, Oxford University Press, 1999. European Commission (2006), XXIIIrd Report on monitoring the application of Community law, COM (2006) 416 Falkner Gerda, Hartlapp, Miriam, Treib, Oliver (2006), “Worlds of Compliance: Why Leading Approaches to EU Implementation Are Only Sometimes-True Theories”, Working Paper of the Robert Schuman Centre for Advanced Studies, European University Institute, Florence Falkner, Gerda (2000). ‘How Pervasive Are Euro-Politics? Effects of EU Membership on a New Member State’, Journal of Common Market Studies, 38:2, 223–50. Falkner, Gerda, Hartlapp, Miriam, Leiber, Simone and Treib, Oliver (2004), 'Non-Compliance with EU Directives in the Member States: Opposition through the Backdoor?', West European Politics, 27:3, 452 - 473 Goetz Klaus H. and Simon Hix (eds.) (2001), Europeanised Politics? European Integration and National Political Systems. London: Frank Cass, 27–51.
117
Graziano, P. and M. P. Vink (2007). Europeanization : new research agendas. Houndmills, Basingstoke, Hampshire ; New York, Palgrave Macmillan. Grossman, Emiliano and Sauger, Nicolas, 2007, “Political institutions under stress? Assessing the impact of European Integration on French political institutions”, Journal of European Public Policy, forthcoming. Huber, John D. 1996, Rationalizing Parliament. Legislative Institutions and Party Politics in France, Cambridge: Cambridge University Press. Huber, J.-D. and Martinez-Gallardo, C. (2004) 'Cabinet instability and the accumulation of experience: the French Fourth and Fifth Republics in comparative perspective'. British Journal of Political Science, Vol. 34, No. 1, pp. 27-48. Jacobsson, Kersten and Schmid, Herman, ‘Real integration or just formal adaptation? – On the implementation of the National Action Plans for Employment’, in: C. de la Porte and P. Pochet (eds.), Building Social Europe through the Open Method of Co-ordination, European Interuniversity Press, Brussels, 2002, pp. 69-95 Journal of Legislative Studies, Vol.11, No.3/4, Autumn/Winter 2005 Kassim, Hussein, 2005, “The Europeanization of member state institutions”. In Bulmer, S. and Lequesne, C. (eds.) The Members States of the European Union, Oxford, Oxford University Press, pp. 284-316. Kerrouche, Eric, 2006, “The French Assemblee Nationale: The Case of a Weak Legislature? “, Journal of Legislative Studies, 12: 3-4, 336- 365. Knill C. and D. Lehmkuhl (2002), ‘The National Impact of European Union Regulatory Policy: Three Europeanization Mechanisms’, European Journal of Political Research, 41/2, pp.255– 80. Knill, Christoph (2001). The Europeanisation of National Administrations: Patterns of Institutional Change and Persistence. Cambridge: Cambridge University Press. Ladrech, Robert, ‘Europeanization of Domestic Politics and Institutions: The Case of France’, in: Journal of Common Market Studies, vol. 31, n°1, 1994, pp. 69-88 Laursen F. and S.A. Pappas (eds.), The Changing Role of Parliaments in the European Union (Maastricht: European Institute of Public Administration, 1995), pp.43–60. Lequesne Christian, Paris-Bruxelles. Comment se fait la politique européenne en France ?, Paris, Presses de Sciences Po 1993. Leuffen, Dirk: Europeanization and the Probabilities of Power During Divided Government in France, Paper presented at ECPR Joint Workshops, Turin/Italy, 22-27 March 2002, http://www.essex.ac.uk/ecpr/events/jointsessions/paperarchive/turin/ws19/leuffen.pdf, 20.2.2005 Maïa, Jean, 2005, « La contrainte européenne sur la loi », Pouvoirs, pp. 53-71. 118
Majone, Giandomenico, « A European regulatory state? », in Richardson (Jeremy), European Union : power and policy-making, London, Routledge, 1996. Maurer, Andreas, and Wolfgang Wessels (eds.) (2001), National Parliaments on their Ways to Europe: Losers or Latecomers? Baden-Baden: Nomos. Mendrinou, Maria (1996). ‘Non-Compliance and the European Commission’s Role in Integration’, Journal of European Public Policy, 3:1, 1–22. Norton Ph. (ed.), National Parliaments and the European Union (London: Frank Cass, 1996) Olsen, Johan P. (2002a). ‘The Many Faces of Europeanization’, Journal of Common Market Studies, 40:5, 921–52. Perrineau, Pascal et Ysmal, Colette (eds.): Le Vote de tous les refus. Les élections présidentielles et législatives de 2002, Paris, Presses de Sciences Po, 2003. Radaelli, Claudio, ‘The Europeanization of Public Policy’, in: K. Featherstone and C. Radaelli (eds.), The Politics of Europeanization, Oxford University Press, Oxford, 2003, pp. 27-56 Raunio T. and S. Hix, ‘Backbenchers Learn to Fight Back: European Integration and Parliamentary Government’, in West European Politics 23/4 (2000) pp.142–68 Raunio T., ‘Always One Step Behind? National Legislatures and the European Union’, Government and Opposition, 34/2 (1999), pp.180–202. Raunio, Tapio; Wiberg, Matti: “Does Support Lead to Ignorance? National Parliaments and the Legitimacy of EU Governance”, Acta Politica, vol. 35, n° 2, 2000, pp. 146-168. Rizzuto, Franco: ‘The French Parliament and the EU: Loosening the Constitutional Straitjacket’, in Norton, Philip (ed.), National Parliaments and the European Union, London, Cass, 1996, pp. 46-59. Rozenberg Olivier and Surel Yves (dir.) (2003), Politique européenne, « Parlementarisme et construction européenne », n° 9 Saalfeld, Thomas, “Deliberate Delegation or Abdication? Government Backbenchers, Ministers and European Union Legislation”, The Journal of Legislative Studies, Vol.11, No.3/4, Autumn/Winter 2005, pp.343–371. Schmidt, Vivien, ‘Europeanization and the mechanics of economic policy adjustment’, in: Journal of European Public Policy, vol. 9, n°6, 2002, pp. 894-912 Siedentopf, Heinrich, and Jacques Ziller, eds. (1988a). Making European Policies Work: The Implementation of Community Legislation in the Member States. London: Sage. Smith E. (ed.), National Parliaments as Cornerstones of European Integration (London: Kluwer Law International, 1996)
119
Spyros A. Pappas (ed.) (1995), National Administrative Procedures for the Preparation and Implementation of Community Decisions. Maastricht: European Institute of Public Administration. Szukala, Andrea: ‘The European Transformation of the French Model’, in: Maurer, Andreas; Wessels, Wolfgang; Mittag, Jürgen (eds.), Fifteen Into One. The European Union and the Member states, Manchester, Manchester University Press, 2001. Wessels Wolfgang, Andreas Maurer and Jürgen Mittag (eds.), Fifteen into One: The European Union and its Member States. Manchester: Manchester University Press, 271–97. Wiberg M. (ed.), Trying to Make Democracy Work: The Nordic Parliaments and the European Union (Stockholm: Gidlunds, 1997).
120
6. Does Europeanization change Executive-Parliament relations: Executive Dominance and Parliamentary Responses in Germany Thomas König & Lars Mäder, University of Mannheim
Introduction This chapter investigates the level of Europeanization of the German legislature and the consequences of Europeanization for the governmental-parliamentary relationship in the period from 1986 to 2005. Some authors fear an erosion of democratic control via Europeanization (i.e. Scharpf 1999; Zürn and Wolf 1999), and others quote the mythical number of an 80% influence of Brussels on the German legislature (Herzog 2007). In order to provide a comprehensive and reliable picture of the Europeanization rate of the German legislature, our empirical analysis includes all federal laws. It distinguishes between domestic and “internationalized” laws, which were introduced by different types of legislative actors (government, others) in the Bundestag. Focusing on domestic lawmaking, we provide two indicators for measuring and cross-validating the level of Europeanization of these laws, which cover almost twenty years of German legislative decision making in multiple policy areas. The cross-validity of the two indicators shall provide a reliable and comparable picture of the level of Europeanization, while the coverage of all policy areas should reduce the risk of a selection bias, which is common to areaspecific studies of Europeanization. In addition to measuring the Europeanization rate, we also investigate the consequences of Europeanization, which inherently create a principal agent-problem: when a growing number of member states pursue different interests and must send their governmental representatives – which often come from different parliamentary (coalition) parties and have divergent interests – to the EU in order to adopt common obligations which must be implemented into domestic law. In this delegation game, the governmental representatives play a prominent role because they are responsible for and have exclusive information on the negotiation, adoption and implementation of common obligations. Whether this informational advantage of governmental actors results in a severe principal agent-problem with parliament or their (coalition) parliamentary parties is hotly 121
debated in Germany. Most scholars argue that Europeanization promotes the empowerment of governmental actors (Moravcsik 2002; Majone 2005; Follesdal and Hix 2006). Most poignantly, Moracvcsik(1997:212) points to a European “executive cartel”, which redistributes national resources by international cooperation at the expense of the powers of national parliaments. During the period of this study from 1986 to 2005, the expansion of the EU’s legislative competencies by several treaty revisions and the effective application of qualified majority voting in the Council of Ministers have promoted Europeanization, suggesting an empowerment of governmental actors at the expense of the parliament. For example, the EU adopted almost 300 measures to complete the internal market, which should have increased the Europeanization rate of the German legislature because their obligations had to be implemented into German law. However, the empirical identification of the level of Europeanization is not a trivial task. Recent empirical studies find that the level of Europeanization – measured by the reference to an EU impulse in the German legislative database – is lower than expected (Töller 2008; König and Mäder 2008). However, it remains unclear why the number of EU impulses has increased, even though the number of European measures is decreasing (König and Mäder 2009) This chapter will provide further insights into the level of Europeanization of the German legislature and the consequences for the relationship between governmental and parliamentary actors. The major empirical questions are to focus on the extent to which the German legislature is influenced by Brussels, and how the relationship between the government and parliament has been affected by Europeanization. For a cross-validated measure of Europeanization, we provide an alternative indicator to the European impulse-indicator by using a keyword-search in the abstracts of each individual law. To analyse the consequences of this cross-validated level of Europeanization, we will investigate whether the adoption rate of proposals from governmental and other types of actors is changed when lawmaking takes place under the shadow of Europeanization. In addition to the adoption rate of proposals, we will explore the amendment rate of Europeanized and non-Europeanized laws. The amendment right endows parliamentary (coalition) actors with the power of legislative oversight, which can balance their relationship to governmental actors. Our results reveal that the level of Europeanization and the consequences for the relationship between governmental and parliamentary actors have been dramatically overstated in the past. According to our findings, the average share of Europeanized German laws is between 17.9% (EU keywords) to 26.3%, (EU impulse) ranging from 6.5 percent in 1995 to 40.3 percent in 2004. These numbers are cross-validated by two independent indicators, the first referring to a 122
reference of an EU impulse in the database of German legislation, the second resulting from a keyword search in the abstract of each individual law. Although our findings indicate an increasing level of Europeanization over time, the question is whether and how this development is changing the power balance between governmental and parliamentary actors in the German legislature. Our findings do not support the fears about an empowerment of the executive at the expense of the parliament. Because the governmental adoption rate is already very high in the German legislature, only other types of legislative actors can profit from Europeanization and increase their adoption rate. Furthermore, the amendment statistics suggest that Europeanized German laws are more frequently amended, which indicates more legislative oversight or at least more parliamentary attention to principal agent-problems in the event of Europeanization. In the following we provide a brief introduction to the German legislature and discuss the power balance between the actors involved. Hereby, we also provide a general statistical overview differentiating between different types of federal laws. We then present the numbers on the Europeanization level of the German legislature. Finally, we discuss our results of Europeanization for the executive-parliamentary power relationship.
The German legislature: Mandatory and Non-mandatory domestic laws The Germany legislature is organized along federal lines, in which the government, Bundestag and Bundesrat act as institutional players of the Federation. In the past, governments were always based on coalitions of political parties, which had the majority of seats in the Bundestag. The number of political parties increased over time from three to five. In the Bundesrat, the Länder are represented by their (coalition) governments, which cast their votes weighted by population size. With German unification in 1990, the number of Länder also increased from 10(+1) to 16. According to Articles 70-74 of the German Basic Law (GG), the legislatures of the Länder and the Federation share legislative competencies, but most legislative activities are assumed by the federal legislature. Article 76 provides the government, the Bundestag and the Bundesrat with the right to initiate bills, which must be adopted by the Bundestag. A special feature of the German federal legislature is that it distinguishes between two types of bills, mandatory and nonmandatory bills (Article 77). For non-mandatory bills, the corresponding majority of the Bundestag can overrule the Bundesrat, while mandatory bills require the approval by the majority of (weighted) votes in the Bundesrat, because the Länder have to implement these laws. 123
From a comparative perspective, the German legislature is often viewed as a two-chamber system, which consists of the Bundesrat and the Bundestag. According to (von Beyme 1997:96) “This classification is unavoidable in substance, but is nevertheless legally wrong: according to the Basic Law (Art. 77 Abs. 1 Satz 1 GG) the Bundestag is the only legislator”. The Bundesrat is neither part of parliament, nor a true second chamber. Apart from this terminological distinction, the Bundesrat only has veto power in about half of all federal legislation affecting Länder interests in mandatory legislation. For the analysis of the activities of the German legislature, we use a dataset which we generated from the GESTA-database of the German Bundestag. GESTA is part of the Documentation and Information System for Parliamentary Processes (DIP), which was established by the Bundestag and Bundesrat in 1972. It is a fulltext-database available in electronic form that records all legislative initiatives from 1976 to 2005.1 This database provides a rich source of longitudinal information on the German legislature. It allows for the extraction of information on each individual legislative proposal: the full-text contents, the source and identifying number, the initiator, the date introduced in the Bundestag, the date of approval and the date of enactment, a reference of whether or not the respective legislative proposal was subject to the mandatory or non-mandatory procedure (König and Bräuninger 2005). Table 1 provides a summary statistic of adopted mandatory and non-mandatory German laws in the period between 1986 and 2005. To study the level of Europeanization, we separate domestic from international laws because the international laws are exclusively negotiated by the government. Due to the EU’s increasing competencies in matters related to international law, in particular in the area of trade politics since the beginning of the 1990s, we concentrate our analysis on domestic laws.
Table 1 about here
According to Table 1, the German legislature produces a similar number of mandatory and nonmandatory laws in domestic lawmaking. Regarding the laws related to international treaties, the share of mandatory laws is significantly higher. Furthermore, the number of adopted laws drops significantly in the first year after elections and increases at the end of each legislative term.
1
Internet access: http://dip.bundestag.de/
124
Europeanization of the German legislature: A cross-validated analysis The GESTA database includes a reference to European impulse which documents the linkage between each legislative act and European Community law, including detailed information about the type of European impulse. It specifies whether the respective legislative act was originally initiated based on a directive, a regulation, a decision, a recommendation of the Commission, a certain Treaty provision, a judgement of the European Court of Justice or any combinations of these possible impulse sources. Although this specification enables us to distinguish the type of Europeanization, it does not provide information about the specific directive, regulation etc. in terms of a title, a number etc. which is responsible for the reference. In fact, the number of impulse references has increased over time, while the number of directives, regulations etc. has decreased (König and Mäder 2008, 2009). To cross-validate the GESTA identification of European impulses, we introduce a second measure of Europeanization, which is based on a comprehensive content analysis of the abstracts of the respective legal acts. Hereby, we search for specific European keywords in the abstract of each law and code the respective law as Europeanized whenever we find at least one of these keywords. Table 2 provides the complete list of these keywords and their respective English translation.
Table 2 about here
Using the information from both criteria for Europeanization we are able to cross-validate the percentage of Europeanized laws in the time period between 1986 and 2005 over time and across different policy areas. Table 3 displays both measures of Europeanization per year.
Table 3 about here
Both criteria yield quiet similar results. Regarding our first measure, the European impulse reference, we find a trend towards a growing Europeanization of German laws over time. Relative to 1986 when only 18.4 percent of all adopted laws originated in Brussels, this number 125
increased considerably to more than 30 percent in 1993. However, after the completion of the internal market the level of Europeanized laws dropped dramatically, reaching its lowest level of only 6.5 percent in 1995. Since 1998 the level of Europeanization passed again the 30 percent benchmark, reaching its highest level in 2004 with more than 40 percent of German laws having been induced by EU impulses. The second measure uses EU keywords, and we find an essentially identical trend until 1997 with a significant lower increase in the percentage of Europeanized laws thereafter. Figure 1 displays this trend for our second keyword measure graphically.
Figure 1 about here
One might criticize that we conduct our content analysis on the abstracts of the respective legislative acts. To measure the risk of an abstract-specific bias, we also conducted a keyword search in the full text for a subsample of all adopted laws between 1995 and 2005 where these full texts are available. Comparing the three measures of Europeanization with each other, we unsurprisingly find more Europeanized German laws for our third measure than for either of the former two. Nevertheless, all three measures of Europeanization are highly correlated2 and show a similar trend of an increasing influence of the EU on the domestic legislation towards the end of the studied time period. One possible explanation for this trend is that the EU has increased the level of competencies after several treaty revisions in specific policy areas. For example, the EU increased its legislative activities especially in the areas of Agriculture and Environment. Table 4 therefore takes a closer look at the area-specific changes. 3
Table 4 about here
As displayed in Table 4, we find considerable variation between these areas independent from our measures used. According to both of our measures, the policy areas with the highest rates of 2 The correlation between the measure of EU impulse and EU keywords (Abstract) is 0.69; correlation between EU impulse and EU keywords (Full text) is almost 0.87; finally the correlation between EU keywords (Abstract) and EU keywords (Full text) is 0.64. 3
The German laws were coded according to an adapted version of the policy classification scheme originally developed by Frank R. Baumgartner and Bryan D. Jones for the Policy Agendas Project.
126
Europeanized laws are Agriculture, Environment, Energy, Transportation, and Banking and Finance. Furthermore, we find that the influence of Europeanization is considerably lower in the areas of Social Welfare, Government Operations, Housing, and Public Lands. In all of these areas, the average share of laws caused by Brussels is not higher than 15 percent. With the exception of foreign trade, the average share of Europeanized laws has never reached the mythical 80 percent level in any area. This finding is also displayed in Figure 2 for our EU keyword-indicator where the bar charts display the absolute number of adopted laws and the absolute number of Europeanized laws, while the line graphs show the percentage of Europeanized laws across areas.
Figure 2 about here
Table 5 compares the results of our two indicators of Europeanization with the keyword search of the full text for the specified subsample of our dataset. Unsurprisingly, we find a higher level of Europeanized laws if we use the keyword search on the full text including considerations for minor references to the EU. The differences between our different measures are most severe in terms of absolute numbers in the policy area of Macroeconomics, Agriculture and Transportation. However, despite these differences, all three measures are extremely highly correlated with each other. The correlation between our EU impulse references and the full text keyword search is highest with almost .98, followed by a correlation of .96 between EU impulse references and abstract keyword search, followed by the correlation between both keyword search methods of .94.
Table 5 about here
Briefly summarized, our results suggest a lower level of Europeanization of the German legislature than the mythical fears of an 80% influence by Brussels. Both criteria reveal quite similar results with an area-specific trend towards Europeanization over time.
127
Consequences for governmental agenda-setting power and parliamentary scrutiny Regarding the power relationship between governmental and parliamentary actors in the German legislature, the most prominent cleavage exists when the government, based on the parliamentary majority of political parties, differs from the political party orientation of those of the Länder governments. Although this party configuration has often been described as a divided government configuration, we can distinguish three types of configurations in the German legislature: the corresponding party majority as a kind of unified government, the opposing party majority as a kind of divided government, and a mixed configuration, in which the government of the pivotal actor in the Bundesrat is formed by political parties from both party camps. In particular the divided government configuration is expected to constrain governmental activities and thus to lower the adoption rate of governmental proposals. In addition to this view on the adoption rate, the amendment right of parliament has recently attracted scholarly attention (Martin and Vanberg 2005; König and Bräuninger 2005). Even though coalition partners pursue different interests, they delegate ministers to the cabinet government, which has been the most successful legislative initiator in the past – independent from the configuration with the Bundesrat. However, when a minister pursues his/ her own interests at the expense of the coalition partner and introduces “hostile” proposals, the partner can amend these proposals in the Bundestag – an important parliamentary prerequisite that can balance governmental agenda-setting dominance and help overcome ministerial drift at the expense of a coalition partner. Due their informational advantage, the amendment rate of governmental proposals should decrease for Europeanized laws, in particular when these laws transpose the obligations of EU directives, which member states must implement either by laws or regulations. In the German legislature, the formal agenda-setting power is shared between the government, a group of parliamentary members of the Bundestag and the Länder represented in the Bundesrat. While all of these actors can initiate a bill, the government is by far the most active and the most successful initiator in the German legislature (Bräuninger and König 1999). Even in periods of divided government, i.e. when the governmental party majority of the Bundestag differs from the (party) majority of the Länder in the Bundesrat, governmental proposals have significantly higher adoption rates than initiatives by any other type of initiator (König 1999). At the same time, opposition initiatives are almost always rejected, and the few Bundesrat initiatives are only sometimes adopted. However, governmental bills are also almost always amended by the Bundestag and/or the Bundesrat (König and Bräuninger 2005). The question therefore is whether this pictures changes with an increase in Europeanization. 128
In order to analyse the impact of EU impulses on the relationship between governmental and parliamentary actors, we take a close look at the statistics over time and assess whether the reference to EU impulses changes these major characteristics of the German legislature, namely whether legislative initiatives and governmental bills have different adoption rates and different levels of amendment activities when EU impulses are notified or not.
Table 6 about here
Table 6 lists the adoption rate of all legislative initiatives and distinguishes between governmental and other proposals for all of our three measures of Europeanization in the period between 1987 and 2005 (11th to 15th term)4. Note that the 15th term was a shorter legislative period due to early parliamentary elections. Accordingly, we find an overall increase in the number of legislative initiatives since the 11th term, with more modest numbers for the shorter legislative period. Regardless of the measure used, the comparison of the adoption rates of initiatives with and without an EU reference shows that those with a reference have a higher adoption rate over all legislative periods apart from the type of initiator. The difference in adoption rates is most pronounced for the EU keyword abstract indicator with an increase from 33.0% to 59.2%, followed by the EU full text indicator and the EU impulse indicator. However, the differences in the adoption rates are much higher for non-governmental proposals. While governmental proposals generally have a very high adoption rate, the likelihood of adoption is not significantly higher for governmental proposals with an EU reference. Unlike governmental proposals, the adoption rate for other parliamentary initiatives drastically increases when they refer to the European level.
Figure 3 about here
Using our second measure of Europeanization, EU keywords in the abstract, Figure 3 illustrates the effect of the European reference on the adoption rate. Figure 3 plots the adoption rates of governmental and non-governmental proposals with and without European reference. This figure distinguishes between governmental proposals with and without European reference as well as 4 Due to the availability of the legislative full texts our third measure is limited to the time period between 1995 and 2005, the 13th to the 15thterm.
129
other proposals with and without European reference. According to Figure 3, the indication of an EU impulse has almost no impact on the adoption rate of governmental proposals, but it significantly increases the likelihood of legislative success for non-governmental proposals for almost all legislative periods. The EU impact was especially strong in the 11th and the 12th term and only modest thereafter. Regarding the amendment rates for legislative proposals with and without European reference, we also distinguish between governmental and nongovernmental proposals. According to our results, displayed in Table 7, legislative initiatives with a European reference have a higher likelihood of parliamentary action than initiatives without. This finding contrasts with our first hypothesis but does partly confirm our expectation from the parliamentary scrutinizing approaches. However the positive effect of Europeanization on the parliamentary amendment rate is constant for both types of initiators. Regardless of whether we consider governmental or nongovernmental proposals, the parliamentary amendment rates are considerably higher for initiatives with an EU impulse. These findings suggest that Europeanization peaks the attention of parliament, who then scrutinizes these initiatives more frequently.
Table 7 about here
Our explorative analysis does not confirm the literature on an empowerment of governmental actors in the event of Europeanization. However, Europeanization affects the German legislature. Regarding the adoption rate of initiatives, there is no evidence for a change in the power distribution between government and parliament. While governmental proposals always have a very high adoption rate in the German legislature, the adoption rate of nongovernmental initiatives increases for EU impulses. This suggests that other initiators may perhaps more benefit than the government from Europeanization. Similarly, legislative initiatives with a European reference have a higher likelihood of parliamentary action than initiatives without. This suggests that legislative oversight can stabilize the relationship between governmental and parliamentary actors, respectively that parliamentary actors are aware of the principal agentproblem which inherently arises with Europeanization.
130
Table 1: German Legislation between 1986 and 2005 Year of adoption 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Average per year
DomesticLaws Non-Mandatory 39 22 25 34 51 29 35 38 61 14 44 36 55 34 40 54 69 34 69 46 41,5
Mandatory 48 15 21 37 68 21 50 47 71 32 30 38 74 25 46 63 74 34 55 46 44,8
LawsRatifying International Treaties Non-Mandatory Mandatory 7 8 6 8 14 18 8 11 14 17 7 17 11 14 18 20 26 38 3 7 11 38 16 40 33 61 3 9 15 21 18 20 17 35 12 16 18 22 8 19 13,3 22,0
Total 102 51 78 90 150 74 110 123 196 56 123 130 223 71 122 155 195 96 164 119 121,4
131
Table 2: European keywords EU keywords (German)
Abbreviation
EU keywords (English)
Europäische Union
EU
European Union
EuropäischeGemeinschaft
EG
European Community
EuropäischeWirtschaftsgemeinschaft
EWG
European Economic Community
EinheitlicheMarkt
Common Market
Binnenmarkt
Single Market
EuropäischeMarkt
European Market
Europäische Gemeinschaft für Kohle
EGKS
und Stahl/Montanunion EuropäischeAtomgemeinschaft Europäischen Wirtschafts- und
European Coal and Steel Community
EURATOM/
European Atomic Energy
EAG
Community
EWWU/ EWU European monetary Union
Währungsunion/Europäischen Währungsunion EuropäischeWährungssytem
EWS
European monetary System
Richtlinie(n)
Directive
Gemeinschaftsrecht
Community law
Europarecht
European law
132
Table 3: Number and proportion of Europeanized laws per year according to EU impulse/ EU keywords (Abstract) 1986-2005 and EU keywords (Full text) 1995-2005
Year of adoption EU impulse
EU keywords (Abstract)
Total
EU keywords (Full text)
Total
#
%
#
%
#
#
%
#
1986
16
18,4
15
17,2
87
-
-
-
1987
6
16,2
4
10,8
37
-
-
-
1988
9
19,6
6
13,0
46
-
-
-
1989
15
21,1
12
16,9
71
-
-
-
1990
25
21,0
24
20,2
119
-
-
-
1991
7
14,0
5
10,0
50
-
-
-
1992
18
21,2
25
29,4
85
-
-
-
1993
29
34,1
19
22,4
85
-
-
-
1994
37
28,0
30
22,7
132
-
-
-
1995
3
6,5
3
6,5
46
11
23,91
46
1996
13
17,6
14
18,9
74
21
28,38
74
1997
15
20,3
11
14,9
74
31
41,89
74
1998
49
38,0
36
27,9
129
64
49,61
129
1999
14
23,7
7
11,9
59
19
32,20
59
2000
31
36,0
13
15,1
86
37
43,02
86
2001
40
34,2
17
14,5
117
69
58,97
117
2002
57
39,9
28
19,6
143
71
49,65
143
2003
24
35,3
14
20,6
68
33
48,53
68
2004
50
40,3
27
21,8
124
68
54,84
124
2005
37
40,2
22
23,9
92
59
64,13
92
24,8
26,3
16,6
17,9
45,0
50,6
Average per year
86,2 43,9
133
Table 4: Number and proportion of Europeanized laws per policy area according to EU impulse/ EU keywords (Abstract) in the time period between 1986 and 2005
EU Impulse
EU keywords (Abstract)
#
%
#
%
#
Macroeconomics
66
30,0
44
20,0
220
Civil rights
9
13,4
7
10,4
67
Health
30
29,7
24
23,8
101
Agriculture
77
58,8
54
41,2
131
Labour
36
20,5
25
14,2
176
Education
12
21,4
10
17,9
56
Environment
28
62,2
19
42,2
45
Energy
15
48,4
9
29,0
31
Transportation
43
35,2
32
26,2
122
Law
44
23,0
21
11,0
191
Social Welfare
13
8,0
11
6,7
163
Housing
4
10,3
3
7,7
39
Banking & Finance
81
45,0
51
28,3
180
Defence
16
18,4
9
10,3
87
Technology
10
40,0
6
24,0
25
Foreign Trade
0
0,0
2
100,0
2
International Affairs
3
60,0
1
20,0
5
Government Operations
7
9,3
3
4,0
75
Public Lands
1
12,5
1
12,5
8
495
28,7
332
19,3
1724
Policy areas
Total
Total
134
Table 5: Number and proportion of Europeanized laws per policy area according to EU impulse/ EU keywords (Abstract)/ EU keywords (Full text) in the time period between 1995 and 2005
Policy areas
EU impulse
EU keywords EU keywords (Abstract) (Full text)
Total
#
%
#
%
#
%
#
Macroeconomics
45
32,6%
26
18,8%
73
52,9%
138
Civil rights
5
16,7%
4
13,3%
16
53,3%
30
Health
20
27,4%
15
20,5%
38
52,1%
73
Agriculture
48
67,6%
30
42,3%
54
76,1%
71
Labour
27
24,8%
14
12,8%
40
36,7%
109
Education
7
20,0%
5
14,3%
15
42,9%
35
Environment
16
69,6%
11
47,8%
19
82,6%
23
Energy
14
53,8%
9
34,6%
15
57,7%
26
Transportation
29
38,7%
22
29,3%
43
57,3%
75
Law
30
25,2%
13
10,9%
44
37,0%
119
Social Welfare
12
12,9%
5
5,4%
18
19,4%
93
Housing
4
22,2%
2
11,1%
7
38,9%
18
Banking & Finance
53
52,0%
26
25,5%
66
64,7%
102
Defence
11
28,2%
5
12,8%
16
41,0%
39
Technology
9
50,0%
2
11,1%
12
66,7%
18
Foreign Trade
0
0,0%
1
100,0%
0
0,0%
1
International Affairs
0
0,0%
0
0,0%
1
100,0%
1
Government Operations
2
5,9%
1
2,9%
4
11,8%
34
Public Lands
1
14,3%
1
14,3%
2
28,6%
7
333
32,9%
192
19,0%
483
47,7%
1012
Total
135
Table 6: Adoption rates of legislative initiatives in the time period from 1987 to 2005 (11th to 15th term) Initiatives Legislative period
No EU impulse
EU impulse
No EU keywords (Abstract)
EU keywords (Abstract)
No EU keywords (Full text)
EU keywords (Full text)
11
41,3
84,6
42,6
78,9
-
-
12
41,5
69,0
42,0
70,9
-
-
13
36,7
58,6
36,5
70,8
35,6
51,7
14
39,7
71,0
43,8
75,0
36,1
68,9
15
37,1
61,4
39,4
74,1
33,6
57,9
Total
39,3
67,1
40,9
73,4
35,3
59,9
Government initiatives Legislative period
No EU impulse
EU impulse
No EU keywords (Abstract)
EU keywords (Abstract)
No EU keywords (Full text)
EU keywords (Full text)
11
81,1
93,9
81,6
93,2
-
-
12
86,7
86,3
87,1
85,1
-
-
13
84,5
83,8
82,4
89,5
85,7
82,9
14
84,7
93,1
87,5
91,0
86,7
89,4
15
67,4
82,9
68,7
92,1
69,9
77,0
Total
81,3
87,8
81,7
89,8
81,6
83,3
Other initiatives Legislative period
No EU impulse
EU impulse
No EU keywords (Abstract)
EU keywords (Abstract)
No EU keywords (Full text)
EU keywords (Full text)
11
20,3
56,3
21,6
30,8
-
-
12
22,8
40,8
23,3
41,7
-
-
13
23,3
27,1
22,8
37,5
23,4
24,8
14
23,5
30,0
24,3
23,8
21,9
34,6
15
24,3
23,9
24,3
22,7
21,8
30,4
Total
22,9
31,4
23,3
33,1
22,4
29,6
136
Table 7: Amendment rates of legislative initiatives in the time period from 1987 to 2005 (11th to 15th term) Initiatives Legislative period
No EU impulse
EU impulse
No EU keywords (Abstract)
EU keywords (Abstract)
No EU keywords (Full text)
EU keywords (Full text)
11
35,4
69,2
36,2
66,7
-
-
12
35,3
57,4
34,6
65,5
-
-
13
30,2
43,6
30,7
46,1
28,0
42,9
14
33,0
54,5
35,4
60,2
29,3
55,6
15
24,5
50,5
27,7
58,8
20,4
47,1
Total
31,9
53,3
33,0
59,2
26,6
48,9
Government initiatives Legislative period
No EU impulse
EU impulse
No EU keywords (Abstract)
EU keywords (Abstract)
No EU keywords (Full text)
EU keywords (Full text)
11
76,1
77,6
75,7
79,5
-
-
12
67,8
66,3
65,6
71,6
-
-
13
57,0
55,4
58,5
50,9
56,2
56,8
14
57,6
67,7
59,6
70,1
53,1
68,2
15
43,3
65,8
47,7
71,4
40,9
60,6
Total
61,3
66,0
61,2
68,5
50,6
62,6
Other initiatives Legislative period
No EU impulse
EU impulse
No EU keywords (Abstract)
EU keywords (Abstract)
No EU keywords (Full text)
EU keywords (Full text)
11
13,8
43,8
14,9
23,1
-
-
12
21,9
42,9
21,7
52,8
-
-
13
22,7
28,8
22,5
37,5
21,1
31,0
14
24,1
30,0
24,7
28,6
22,5
34,6
15
16,5
23,9
17,5
22,7
13,7
27,8
Total
20,4
31,4
20,8
36,3
19,9
31,1
137
138
Figure 1: Europeanized laws per year according to EU keywords (Abstract) (1986-2005)
139
Figure 2: Europeanized laws per policy area according to EU keywords (Abstract) (1986-2005)
140
Figure 3: Adoption rates of governmental and nongovernmental proposals with and without EU impulse across legislative periods (1987-2005; 11th – 15th term)
141
References Bräuninger, Thomas, and Thomas König. 1999. "The Checks and Balances of Party Federalism: German Federal Government in a Divided Legislature." European Journal of Political Research 36 (2):207-34. Follesdal, Andreas, and Simon Hix. 2006. "Why There is a Democratic Deficit in the EU: A Response to Majone and Moravcsik." Journal of Common Market Studies 44 (3):603-34. Herzog, Roman 2007. "Hat die EU zuviel Macht?" Die Welt, 31.03.2008. König, Thomas. 1999. "Regieren im deutschen Föderalismus." Aus Politik und Zeitgeschichte 49 (13):2436. König, Thomas, and Thomas Bräuninger. 2005. "Gesetzgebung im Föderalismus." In Speyerer Forschungsberichte. Speyer: Deutsches Forschungsinstitut für Öffentliche Verwaltung. König, Thomas, and Lars Mäder. 2008. "Das Regieren jenseits des Nationalstaates und der Mythos einer 80-Prozent-Europäisierung in Deutschland." Politische Vierteljahresschrift 49 (3):438-63. ———. 2009. "Antwort auf die Replik von Daniel Göler zum Beitrag "Das Regieren jenseits des Nationalstaates und der Mythos einer 80-Prozent-Europäisierung in Deutschland"." Politische Vierteljahresschrift 50 (1):80-5. Majone, Giandomenico. 2005. Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth. Oxford: Oxford University Press. Martin, Lanny W. , and George Vanberg. 2005. "Coalition Policymaking and Legislative Review." American Political Science Review 99 (1):93-106. Moravcsik, Andrew. 1997. "Warum die Europäische Union die Exekutive stärkt: Innenpolitik und internationale Kooperation." In Projekt Europa im Übergang? Probleme, Modelle und Strategien des Regierens in der Europäischen Union, ed. K. D. Wolf. Baden-Baden: Nomos. ———. 2002. "In Defense of the “Democratic Deficit”: Reassessing the Legitimacy of the European Union." Journal of Common Market Studies 40 (4):603-34. Scharpf, Fritz W. 1999. Regieren in Europa. Effektiv und demokratisch? . Frankfurt a. M.: Campus. Töller, Annette Elisabeth. 2008. "Mythen und Methoden. Zur Messung der Europäisierung der Gesetzgebung des Deutschen Bundestages jenseits des 80-Prozent-Mythos. Zeitschrift für Parlamentsfragen." Zeitschrift für Parlamentsfragen Heft 1:3-17. von Beyme, Klaus. 1997. Der Gesetzgeber. Opladen: Westdeutscher Verlag.
142
Zürn, Michael, and Dieter Wolf. 1999. "European Law and International Regimes: The Features of Law Beyond the Nation State." European Law Journal 5 (3):272-92.
143
7. Leading governments and unwilling legislators. The European Union and the Italian Law making (1987-2006) Enrico Borghetto, Marco Giuliani & Francesco Zucchini (University of Milano) Introduction If there is some truth in the old saying that “you don't marry someone you can live with but rather the one that you cannot live without”, one should not be too surprised by the firm resolve of the Italians (both at the elite and mass levels) to tie their country’s destiny to a strong and stable European Union. For most of its early Republican history, Italy’s governing coalitions considered European membership a sort of “insurance against the threat of democratic breakdown” (Cotta 1992, p.210) posed by extremist parties. Italy’s participation in the European community as one of the founding members and the rising interdependence linking the major capitals in Western Europe represented a political buttress upon which the major pro-European political forces (Christian Democratic, Socialist and Liberal) consolidated their electoral support. The symbolic force of Europe did not lose its raison d’être even after the “normalization” of the Communist party and its gradual acceptance of the principles underpinning European integration.1 Rather, it changed from a stabilizing instrument to a resource in the hands of policymakers to drive long-needed changes. The legitimization potential of “external European requirements” (what Dyson and Featherstone [1996] have referred to as a useful vincolo esterno) stem from the fact that short-term political costs of austerity measures were successfully outweighed by the long-term benefits promised by a significant development in supranational integration. Hence, the “European card” was regularly put on the table whenever the need for unpopular measures loomed on the horizon. Under the threat of “not keeping up with Europe”, Italy undertook reforms in a wide array of policy sectors, ranging from budget to social policy (e.g. Della Sala 1997; Ferrera & Gualmini 2004; Di Palma et al. 2000).2 The perseverance displayed by the Italian elite in making Italy one of the founding members of the European 1
In the mid-1970s, the Italian Communist Party formally accepted Italy’s membership in the NATO and in the European Community. The end of this die-hard ideological bias in the domain of foreign policy foreshadowed the PCI’s strategy of widening the distance from Moscow and favouring a gradual rapprochement with the European socialist left. 2 All in all, what distinguished Italy from other European countries has been the perception of Europe in most cases as the solution more than the problem (Radaelli & Franchino 2004, p.944). Others have argued that Italy was “rescued by Europe” (Ferrera & Gualmini 2004).
144
Monetary Union is the most visible example of the strategic importance of “European constraints” in the national political arena. In most cases, these constraints took on the “hard” legal form of EU directives and regulations.3 Over the decades, this ensemble of regulatory measures has progressively pervaded an increasing number of sectors, making the EU one of the main sources of primary law alongside national legislative institutions. The phenomenon has not failed to attract the attention of Italian political commentators and experts who, echoing Delors’ prediction, estimate that a large proportion (the figures differ, but are never too far from 80%) of national policies originate in Brussels. Most of the time, these statements are accompanied by two corollaries regarding the domestic impact of Europe. First, the sphere of action of primary laws adopted within the parliamentary circuit should be gradually restricted. This should be particularly apparent in the Italian case, where Article 70 of the Constitution envisages that the legislative power is jointly exercised by the two Chambers and no specific policy area is reserved to administrative measures. This provision – alongside the doctrine of the “centrality of parliament”, whereby “parliament rather than the executive was (should be) the central and supreme institution of the political system” (Cotta 1994, p.66) – contributed to making the Italian legislature one of the most prolific in Europe in terms of annual law production . If one conceives of European integration as a zero-sum game in terms of sovereignty, the supremacy of European law over national law is expected to place boundaries on the previously unlimited parliamentary mandate. In this sense, some authors understand the “democratic deficit” afflicting the EU as a phenomenon originating in the national political arena because of European membership, rather than as a deficit rooted in democratically unaccountable EU institutions (Katz & B. Wessels 1999; Schmidt 2006). Second, the process is expected to alter the balance of power between the executive and the legislative to the detriment of the latter (Goetz & Hix 2001; Maurer & W. Wessels 2001). National executives consolidate their role as gatekeepers between the supranational and domestic spheres in order to reinforce their privileged access to decision-making venues in Brussels and their control over key areas such as the provision of information (Moravcsik, 1994). Studies focusing on the Italian case reveal that these dynamics may also have affected the Italian parliament and undermined its traditional predominance in the law-making process.4 Giuliani 3
We do not dismiss the importance of “soft law” instruments, such as the various existing forms of intergovernmental policy coordination (e.g. in the framework of the Open Method of Coordination, the Common Foreign and Security Policy etc…) or the constraints imposed by membership in the wider EU market and monetary union. Nonetheless, our approach privileges de jure pressures over de facto ones. 4 Hine (1993), writing in the immediate aftermath of the Maastricht Treaty, still considered the Italian parliament as an exception to the “decline of parliaments” thesis.
145
(2006) argues that the involvement of the two chambers in defining EU policy, in coordination with executive representatives, has been by and large marginal. Italian legislators passed a variety of reforms to grant the legislature a more relevant role in both formulation and monitoring of European policies.5 But, the parliament lacked the willingness to use these powers, because it was in a mode of, what can be termed, “self-marginalization” (Giuliani 2006, p.177). The same asymmetry between formal powers and actual performance seems to apply to EU policy implementation too, the phase that underwent the most radical upgrading by the Italian policy-makers. In order to standardize and increase the periodicity of the intermittent attention being devoted to European issues by Italian legislative and administrative bodies, the legislator prescribed, toward the end of the 80s, that the standard transposition channel would in future be an annual Community Act (legge comunitaria).6 Such an act would be drafted every year by the executive and presented to the parliament before March 31st (now January 1st). The legge comunitaria, once discussed and adopted by the parliament following the ordinary process, provides, along the lines of a voluminous package law, for the transposition of those directives that are due to expire in the first half of the following year. The peculiarity of this act is that it does not accomplish, except for some rare exceptions, transposition directly.7 Rather, the legislator has to use it as an all-embracing delegation law (see below), establishing for every directive the preferred executive measure needed for its implementation: either administrative acts, regolamenti (secondary level regulations) in areas subjected to previous delegislation or, finally, decreti legislativi (legislative decrees). The last legislative option has been by far the most common. During the period under consideration, one out of every two directives was transposed through legislative decree.8 By devising this procedure, the legislator explicitly aimed at a dual goal: to allow the parliament to keep the transposition phase under its control without giving up on the swiftness and expertise of executive law-making. According to Fabbrini and Donà’s analysis (2003), the legge comunitaria procedure proved more successful in pursuing the second objective. Interviews with public officials revealed that the level of parliamentary scrutiny, during the adoption of the legge comunitaria, leaves a lot to be desired. Their
5
Among the most important reforms, one might single out the following: setting up of EU affairs committees; creation of a special session in the plenary working schedule devoted to the discussion of the executive EU policy; enhanced access (both in terms of timeliness and scope) to information flowing from Brussels; right to request for a temporary interruption in the Council debate of a policy so as to allow MPs to draw up a specific mandate for their governmental representatives. 6 Since the mid-80s, Italy has been among the worst performers in the rankings compiled by the Commission as far as the domestic application of Community rules is concerned. 7 This occurs mainly in two cases: when the object of the directive to be implemented is not very complex or when it is required for the abrogation or amendment of national provisions in contrast with Community Law. 8 That said, one must notice that the Community Act, though representing the primary transposition instrument in Italy, did not exclude recourse to the classic repertoire of regulatory measures.
146
reconstruction of the policy-making process concludes that “both the elaboration and approval stages of the legge comunitaria are under government control” (ibid. p.45). This study builds on this strand of research and contributes to previous Europeanization studies on the Italian case in three ways. First, it focuses on the Europeanization of Italian law-making, a topic which has been so far relatively overlooked in literature.9 By providing a first measurement of the extent of Europeanization across different policy sectors, it offers an insight into the progress and differential impact of Europe, which has yet to be properly understood due to a lack of data. Second, this study makes the first attempt to analyze the impact of European integration on the Italian legislature’s policy influence by adopting a large-N research design. Previous work has focused on changes at the level of institutional rules and on the context in which they operated (Giuliani 2006), or adopted a qualitative approach based on interviews with experts (see for example Fabbrini & Dona' 2003). This study develops a set of indicators aimed at capturing the concept of “legislative influence” and draws inferences from a large data set containing information on the processes and characteristics of the population of legislative acts adopted in Italy across 20 years (1987-2006). Third, the analysis has been extended to decreti legislativi, a type of primary legislative instrument that has become quantitatively and qualitatively important in Italy in the last two decades but is often overlooked in the literature (Vassallo 2001). The rationale for their inclusion in this study lies in their predominant role in the transposition of EU directives, which makes them a crucial case to test the Delors’ prophecy on the Europeanization of national law-making. This chapter is organized into five sections. The next section aims at illustrating the evolution of Italian legislative activities over the last 20 years by identifying two ongoing, interconnected dynamics from the available data. The second section presents a measurement of the extent of Europeanization in Italian law-making, both in the primary legislation passed in parliament and in the legislative decrees issued by the Government. The third section examines whether the Europeanization of legislative activities benefited the government and if so, what policy-specific factor accounts for it. The last section presents our conclusions.
1. The recent evolution of Italian law-making Studying changes in Italian legislative processes over the last 20 years by means of quantitative indicators is complicated because of at least three factors. First, diachronic analyses have to take into account the very different durations of recent Italian legislatures and their effect on law9
An exception can be found in Borghetto, Giuliani and Zucchini (2009). Interestingly, they detected evidence of Europeanization also in administrative measures.
147
making. Second, the political system as a whole has undergone substantial changes to such an extent that the post-1992 period is usually referred to as a “Second Republic” by political commentators (though it did not entail amendments to the Constitution). Inevitably, these changes impacted law-making too, and hence the variations observed in the number and type of laws must be considered with caution as their nature and length have been altered substantially over the years. Third, the proliferation of new countries as a consequence of the collapse of Soviet Union and the expansion of the globalization process has multiplied the number of international relationships to be ruled by laws, namely the number of ratifications of international treaties. The parliament reserves a few days in each legislature for these bills, which is considered at best to be a meaningless formality. Because the inclusion of these bills could bias our results, it was decided to exclude them from the rest of the analysis. Yet, once all of these factors are considered, two dynamics, partially interwoven, emerge clearly from the data (see Tab.1): a) the decrease in the number of laws b) the increasing importance of Government's role in the legislative process. [INSERT TABLE 1 HERE] 1.1 The decrease in the number of adopted laws In the 10th legislature the Italian parliament approved, on average, one law every two days, but in the 14th legislature, an average legislative process took twice as long. This change was caused by a variety of factors. First, the polarization and fragmentation in the last part of the first republic and during the second republic hindered the adoption of bills within committee, i.e. without reporting the proposal to the full floor (the so-called “sede legislativa”).10 The laws adopted through this procedure, also called “leggine”, used to be in general short statutes providing “pork barrel projects” and “financial benefits” to relatively narrow constituencies, geographically and/or functionally delimited. Second, the economic emergency in the early 1990s and the introduction of government alternation since the 1996 elections required ruling coalitions to focus their efforts on a few areas of reforms instead of spreading them over many incremental and marginal legislative changes (Zucchini 2010). Finally, the increase in the delegation of laws (see below) reflects that a growing proportion of binding legislative decisions has been made by the executive through legislative decrees, thus reducing the parliament’s legislative load (Vassallo 2001).
10
This decentralized legislative procedure is envisaged by the Italian Constitution and was very common during the First Republic (Zucchini 1997; De Micheli & Verzichelli 2004). Until the early 70s, over 75% of all legislation (on average) was adopted in committees
148
1.2 The strengthening of the government's law-making role If having the first and/or the last say in a decision making process can be taken as evidence of a powerful role, then the data available on legislative output indicate a strengthening of the Italian government in the legislative arena in the last two decades. This is evidenced by the following observations: a) a slight increase—from 61% in 10th legislature to 69% in the 14th legislature–in the proportion of laws sponsored by the government; b) a substantial, though irregular, increase in the delegation of laws and of corresponding legislative decrees. A delegating law (Art. 76 and Art. 77[1] in the Italian Constitution) is formally a law approved by ordinary procedure and, in rare instances, by decree law procedure (see below). This type of law contains at least one provision (“delega”, delegation) delegating the Executive the power to issue new primary legislation (called “decreti legislativi”, legislative decrees) according to more or less detailed criteria and within a specified period of time (both constraints have to be explicitly laid out in the delegation). Each delegating law contains a variable number of “deleghe”. The Constitutional Court can declare a legislative decree invalid if it crosses the limits provided by the delegating law. Once the government is granted these legislative powers, the parliament cannot take them back, and most policy-making processes will have to be undertaken exclusively in the government’s rooms. Thus, the Government has the last say in the decision making process. Delegating laws and legislative decrees were not very common during the First Republic (De Micheli & Verzichelli 2004). Mostly, they were associated with “technical” issues and only occasionally with important reforms. The recourse to this procedure has grown exponentially since the early 1990s, and both practitioners and scholars consider it to be the most apparent evidence of the increased role of the executive, vis-à-vis the parliament, in the law makingprocess (Vassallo 2001; Gianniti & Lupo 2004; Kreppel 2009). Some scholars refer to it as the determining factor explaining why “compared to the previous decades, the 1990s were a period of legislative activism and of real governance” (Capano and Giuliani 2001, 24) or describe it as “the most significant change in relations between government and parliament introduced during the 1990s and maintained since” (Vassallo 2007, 699). Indeed, the growth of delegations is impressive: if at the end of the eighties, on average, there was one delegation for every seven laws, in the 14th legislature there was one delegation for every two laws. Of course, this proliferation of delegations brought about a parallel increase in legislative decrees. In the last legislature (14th), the average number of legislative decrees per day was twice that of the 10th legislature. Decree laws (“Decreto legge”) should not be mistaken for legislative decrees. In the decree law procedure (Art. 77[2] Italian Constitution), the Executive can issue a decree in “extraordinary 149
cases of necessity and urgency” (in most cases, these conditions are taken for granted). Executive’s decrees come into force immediately and remain effective for sixty days since their issuance occurs without the need for parliamentary approval. If after this period, the parliament does not ‘ratify’ the decree, then the status quo is restored. A decree law cannot, therefore, be defined as executive legislation as the last say belongs to the parliament. Certainly, this procedure allows the Executive to push its priorities through the crowded parliamentary agenda. Because of the 60-day deadline for its ratification, the bill ratifying a decree-law is granted top priority both in the Committee and the Floor’s agenda. The other side of the coin (at least for the government) is that the parliament is not constrained in its right to amend this bill.11 If the executive opposes its amendments, MPs can always threaten to sink the bill. This situation has often resulted in the “swelling up” of ratification bills because of the number of additions to the final text. Even so, the number of decree laws has substantially increased over the decades. Whereas in the early 1990s they were about 20% of the laws approved by the parliament, in the 14th legislature they comprised more than 40%.
2. The Europeanization of the Italian legislative output The objective of this section is to provide evidence on the extent of Europeanization of the Italian legislative production. For this, a keyword search was carried out on all of the legal texts. A law is classified as Europeanized if it contains at least one of the key words selected in the framework of the Delors’ Myth Research Project (see Table 2). As laws and legislative decrees are equal in status according to the Constitutional hierarchy of norms, and the proportion of the latter has become quite substantial over the last few decades, the unit of analysis of this study comprises both. But, because merging of the parent acts (delegating laws) with the corresponding delegated measures (legislative decrees) under the same index amounts to counting the same normative content more than once, the analysis will have to be done in two steps. Therefore, the presence of European links was inspected first in parliamentary laws and then in legislative decrees.
11
Executives used to reissue law-decrees not converted into primary law by the parliament, in which case a decree loses validity from the day of its issuance. The latter situation became more common after the mid-70s and reached its peak during the 12th legislature, when out of more than 700 decrees issued (and subsequently reiterated) only 120 became laws before expiring. Confronted with this proliferation of “emergency measures”, the Constitutional Court could no longer hold itself back from coming down heavily with a famous sentence in 1996 that reiteration was unconstitutional. This decision brought about a drastic drop in the number of law-decrees issued (or, more precisely, re-issued) and, according to some authors (see Zucchini 2010), contributed to reinforcing the executive position visà-vis the parliament.
150
[INSERT TABLE 2 HERE] 2.1 Laws adopted in parliament In absolute terms, the number of laws with a European link has been steady over the study period, but it has been marginally increasing in terms of the overall legislative output (largely because of the contemporary decrease in the number of legislative measures). Whereas in the 10th legislature only 13% of the laws had a European link, this figure increased by only 4% in the last legislature (14th) (see Figure 1). According to the chosen europeanisation measurement, Delors’ predictions went largely unfulfilled. [INSERT FIGURE 1 HERE]
Furthermore, we expect that the distribution of European links should also vary remarkably across policy areas (see Figure 2). To answer this question, the classification of individual legislative measures was employed according to the main policy content developed by the US Policy Agendas Project and coded each law into one policy sector. To capture the varying level of EU involvement by policy area, all sectors were grouped into five ordinal classes, which resemble - with some adaptations - those of the classification proposed by Nugent (2006, p.388). [INSERT FIGURE 2 HERE]
Across the whole period, more than 50% of the laws relating to foreign trade have a European link, but the percentage shrinks to less than 4% for laws dealing with crime, legal codes and family. Predictably, agriculture (28%), environment (26%) and transport (22%) are the other Europeanized sectors. The distribution of laws with a European link (both in terms of absolute and relative figures) across sectors reflects quite faithfully the order suggested by Nugent.12 However, two important exceptions are found: “Civil rights” and “Science and technology”. The former is unexpectedly a Europeanized sector, and in the latter, Europeanization is weaker than expected.
12
If the category “Extensive EU involvement” is assigned 5, the category “No EU involvement” 1, and the intermediate categories other values in between, the correlation index between Nugent’s classification and percentage of laws with European link is 0.75
151
The dispersion of legislative measures, with or without a European Link across policy areas is more or less the same: the Duncan index is 25 and the Pearson correlation 0.592.13 Quite interestingly, this similarity increased over the years– albeit unevenly – and these figures changed from 30.5 and 0.382 in the 10th legislature to 23.5 and 0.669 in the 14th. This finding may originate from a gradual extension of EU competences as a consequence of EU treaty reforms, although it is not confirmed by data on legislative decrees.
2.2 Legislative decrees The soaring number of legislative decrees is a recent phenomenon and it is closely connected to the acceleration of the European integration process towards the end of the 80s. Indeed, laying out the juridical foundations of the European Union required every Member State to assimilate a large number of measures (mostly directives) in its domestic legislative framework over a tight time-schedule. As already mentioned, the Italian government has fulfilled its transposition obligations mainly by implementing the numerous delegations contained in the annual Community act and issuing legislative decrees.14 Figure 3 illustrates that, especially in the early 1990s, the existence of a EU link is strongly correlated with transposition duties. However, since the 13th legislature, an increase of norms is observed with only a European inspiration. Even so, the proportion of Europeanized legislative decrees over the total output is almost always above 50%, with peaks of 80%, thereby confirming, to some extent, Delors’ predictions. References to Europe, which are largely marginal in statutory measures, mushroom when the legislative decrees issued by the executive are taken into consideration. [INSERT FIGURE 3 HERE]
13
The
Duncan
Duncan index =
index is a measure
1 æ xi yi å ç 2 i èX Y n
of
similarity of
two distributions.
It is defined as:
ö ÷ 100 , where i = variable category, and xi(yi) = frequency of the category i ø
in the population X (Y). If the laws sponsored by the government (X) and the laws sponsored by the parliament (Y) were divided into only two categories, A and B, then the overall number of government’s laws is X= 300 and that of parliament’s laws Y= 200. The government’s laws of category A are xA=100 and the parliament’s laws of the same category are yA= 75. The Duncan index is
1 é æ 100 75 ö æ 200 125 ö ù êç ÷+ç ÷ ú 100 = 4.166 2 ë è 300 200 ø è 300 200 ø û
The Duncan index varies from 0 (maximum similarity) to 100 (minimal similarity) 14 Besides being the main gateway between the EU and the Italian legal systems, the Community Act has been the single greatest source of legislative delegations. If one counts one delegation for every directive transposed by legislative decree authorized by the Community Act, the number of Comunitaria-related delegations outnumbers that of the other types of delegations by 586 to 449.
152
If one looks at the range of policy areas regulated through this instrument (Figure 4), a clear-cut division of labour is discovered - as expected - between the EU and Italy. Legislative decrees with a European link (both transposition decrees and decrees with simply a European inspiration) insist on very different policy areas with respect to decrees not linked to Europe. The Duncan index calculated by comparing the two series is 51.73 and there is no significant correlation (0.13). The greatest amount of legislative effort has been put in traditionally EU-regulated areas belonging to the European Communities Pillar, namely agriculture, market regulation (Banking & Commerce) and environment.15 Similar to our analysis based on laws, the distribution of Europeanized legislative decrees also fits rather satisfactorily in Nugent’s classification of sectors according to EU policy involvement.16 The impact of EU directives is extensive – albeit allegedly in more recent times- in policy areas where policy responsibility is shared between the EU and member states, or EU policy involvement is limited in health sectors. Once again, European links are detected not only when the fulfillment of transposition obligations is in order, but also otherwise. The data show that the scope of Europeanization, for reasons not related to transposition, has been wide-ranging and Europeanization has touched every sector over the last twenty years. [INSERT FIGURE 4 HERE]
3. How the evolution of Italian law-making interacts with European link
In the first section, it is observed that the decrease in the number of laws and the strengthening of the executive role in the legislative arena are two important and uncontroversial aspects that characterize Italian law making over the past two decades. Are these trends connected to some extent with the variation in the presence of EU links in our pool of legislative acts? The two columns making up Table 3 suggest as follows: 1)
The proportion of executive-sponsored laws in the last three legislatures is higher among
the laws with the European link. 15
Both foreign trade and foreign policy are not normally regulated through legislative decrees, because acts in these domains appear mainly in international treaties, which - according to the Italian Constitution - require a parliamentary act to make them operative. 16 If one assigns 5 to the category “Extensive EU involvement” and 1 to the category “Virtually no EU involvement” and the other values in between to the intermediate categories, the correlation index between Nugent’s classification and the percentage of legislative decrees with European link is 0.73.
153
2)
Delegations (and consequently legislative decrees) are much more common when bills
have a European link. This is not surprising because many of them are contained in the Community Act (introduced right at the beginning of the period under consideration). These findings support the view that the gradual retrenchment of the Italian parliament in the legislative arena has been stronger when the bill originated from or was inspired by Europe. In the most recent legislatures, MPs rarely sponsor a bill with a European link. Moreover, when they do so, the odds are that they are authorizing the government to define the details of its normative content by means of legislative decrees. [INSERT TABLE 3 HERE]
3.1 Inspecting data on legislative amendments That said, even though the parliament is not a major promoter of legislation with EU links, this does not necessarily mean that it cannot lead the review and discussion of Europeanized bills and change, more or less radically, their content. The same holds true for delegating laws which, once amended by the parliament, can in fact drastically restrict the discretionary powers of the executive in the post delegation phase. Unfortunately, the present data on the parliament amendatory activity cover only the last nine years of the period under study (roughly the 13th and 14th legislatures). On the other hand, they provide both the number of approved amendments per type of law and identity of the proposer (Government vs. MPs). As the length of bills varies substantially, one has to consider these numbers very carefully: bills with a EU link are, on average, systematically larger than others and, therefore, they are more likely to be amended a priori, regardless of whether the parliament or the government is determined to correct and change their original content. The proportion of laws (see Table 4), with at least one amendment, is considerably higher among laws with a EU link, but this difference disappears once the totality of amendments and the length of the laws are considered. Moreover, if the focus is only on MPssponsored amendments, their number per 100 words is systematically lower when dealing with laws containing a EU link.
[INSERT TABLE 4 HERE]
154
This might suggest that the stronger role of the Government at the beginning (when proposing a bill) and end of the legislative process (by taking advantage of the numerous delegations), in the presence of a EU link, is not counterbalanced by a more proactive amending activity of MPs inbetween.17 A more in-depth investigation reveals that the number of approved amendments varies significantly across policy areas. As has already been done for laws and legislative decrees, the classification elaborated by Nugent was adopted for assessing the level of EU policy involvement in the coding scheme developed by the US Policy Agendas Project. Considering the overall period (1997-2006), it is found that some laws with a European link have been amended more exhaustively than those concerning the same issues but without a European link, while some others remain less amended (Tab. 4). However, division of the policy areas of the last 10 years in the two legislatures (respectively ruled by a centre-left and centre-right coalition) into policy areas more amended and policy areas less amended shows, not too surprisingly, a change between the two. Moreover, while in the 13th legislature there is an inverse correlation between the intensity of the amendment activity and the a priori level of EU involvement18, in the next legislature no such correlation exists. Therefore, policies that traditionally belong to the European domain have not necessarily become less controversial in the Italian legislative arena. Rather, the difference seems to have originated from policy orientation and party composition of the legislative arena, either in the EU political system or in Italy. Looking at the number of amendments per 100 words in the 13th legislature, assessing if a law is linked to Europe seems to lead to more amendments in sectors such as “Law & Crime”, “Social Welfare” and “Transportation”. In the 14th legislature, the same dynamic concerns “Education”, “Labor, Employment and Immigration” and “Foreign Trade”. Additionally, “Health” and “Banking & Commerce” have a similar level of amendments in both legislatures, regardless of the link with Europe. “Health” is less controversial when linked to Europe, and the reverse is true for “Banking & Commerce”. What these apparently confusing results suggest is that bills with a European link are not located in any ideological consistent dimension, from which the current parliamentary majority is systematically more or less distant. And, the distance changes according to policy areas.
17
This result holds true when considering only executive-sponsored legislation. If one assigns 5 to the category “Extensive EU involvement” and 1 to the category “Virtually no EU involvement” and the other values in between to the intermediate categories, the correlation index between Nugent’s classification and average number of amendments per 100 words is -0.388 in the 13th legislature and 0.13 in the 14th Legislature. 18
155
[INSERT TABLE 5 HERE] 3.2 Has Europe contributed to strengthen the Italian government? A tentative explanation The evidence presented so far points to a strengthened role for the government when the bills have a EU link. In many cases, Europeanized legislation was initiated and drafted directly by the government, often following a delegation by the parliament. What is more, legislative processes involving European issues do not exhibit a greater rate of approved amendments in comparison to other processes. Yet, one needs to explain what bill-specific factors may account for it. The objective of this section is to answer this with the available data by means of three a priori plausible hypotheses, the last two of which are not easy to distinguish empirically. Because of space and data limitations, the following analysis adopts an exploratory approach.
H1: Bills with a European link concern more controversial issues. This explanation is based on the level of conflict between government and opposition and within the government coalition. The propensity to delegate the bills that concern EU-related issues might be a strategic move by the governing coalition to take an issue outside the more exposed parliamentary arena and leave discussion on the details of the measure to the Council of ministers. Besides, the government initiative originates from the rational calculation of party leaders to prevent fragmentation and conflict inside the parliamentary majority. If bills with a EU link were more controversial, then one should find evidence of this conflict in the legislative process. Opposition MPs, as well as dissenting majority MPs, will try to put off their approval by using all delaying tactics permitted by the standing orders. The measurement of conflict followed here is the number of tabled amendments. This indicator might be taken as a sign of disagreement for two reasons: 1) it may signal a departure from the original proposal 2) it is a common device used to delay the approval of a law. Just as the approved amendments, the number of voted amendments cannot be considered independently of the length of the bills; therefore, the number of amendments per 100 words (Figure 5) was calculated. During the period 1997-2006, this index does not lend much support to the conflict-based hypothesis: it is just slightly higher for laws linked to Europe. Moreover, most of the difference must be attributed to the first five years of the period under consideration. [INSERT FIGURE 5 HERE]
156
H2: Bills linked to the EU concern highly technical and complex issues and, generally, MPs do not have enough knowledge and expertise to deal with them. This time, the decision to delegate the discussion and drafting of these norms to the government is based on the incapability of MPs. To explore this hypothesis, the pool of laws, not linked to the EU, was divided into two categories, based on whether they relate to the same issue addressed by Europeanized laws or not.19 If the issues concerning Europe were, in general, more technical or complex, one should expect a higher rate of government initiative, more delegations and fewer approved amendments among laws without European link, but belonging to sectors where most Europeanized laws are concentrated. The results are inconclusive (Table 6). While laws not linked to the EU but sharing the policy fields with Europeanized laws do not seem to be remarkably less amended than the rest, they were sponsored more often by the government and they had more delegations (albeit the figure is not comparable to that of laws linked to Europe). [INSERT TABLE 6 HERE]
H3: The topics of bills linked to the EU are not salient in the eyes of MPs’ Constituencies, and in any case individual MPs can hardly be considered responsible for their content. This hypothesis is based on the lack of interest of MPs. What plays an important role is, perhaps, not the complexity, but the nature of issues characterizing the laws with a European link: the prevailing role of the government can derive from a growing indifference on the part of MPs too. They prefer to ignore these bills, leaving the discussion and approval to executive actors. To gauge the importance that MPs attach to the issues addressed by laws linked to Europe, the distribution was calculated across three (non exclusive) categories of issues: laws sponsored by the government; laws sponsored by MPs; laws with European link. Then, the similarity index (Pearson correlation) and Duncan index of the three distributions were computed. Table 7 shows considerable difference between the laws sponsored by MPs and the laws linked to Europe. The latter group of laws does not clearly insist on the same issues as those of the laws promoted by MPs. Moreover, if one divides the whole period into three parts—First Republic (1987-1992), Transition (1992-1996), and Second Republic (1996-2006)—it can be seen that the interests of
19
The authors have considered as issues the micro-topics of the U.S: policy agenda projects.
157
MPs have become more different from those revealed by the legislative production with a European link.20 [INSERT TABLE 7 HERE]
The exploration of the reasons explaining the strengthening of the Executive in the legislative process when a bill is linked to Europe emphasized two factors: lack of interest and (maybe) capability of MPs. These findings fit rather well with the evidence presented above showing that the parliament does not seem to recover what it clearly loses in the amendment and discussion phase as promoter. They are also in line with the concept of “auto-marginalization of the parliament” as employed by Giuliani (2006) with reference to Italy-EU relations. It is speculated that the nature of Europeanized legislation tends to limit MPs from actually playing a more active role either because they do not have the necessary competence to play such a role or because such a role does not yield any short or medium-term electoral payoff. In the crowded legislative agenda, their interest may be more easily caught by localized or symbolic legislation.
Conclusions
In this chapter, we have contributed to previous analyses on the Europeanization of the Italian political system by focusing our attention on a subject which has received little attention: the EU impact on the law-making process. We have approached this phenomenon from two perspectives. First, we have tried to provide an evidence-based measurement of the extent of Europeanization in the Italian legislative output, thus empirically testing the often-reported “Delors’ prophecy”. The data collected on the population of Italian legislative processes over a 20-year period reveal that the 80% prediction is largely off the mark as far as ordinary laws are concerned. On the other hand, Europe appears to have entered the Italian legal system preferably through the “backdoor” of delegated legislation. This legislative option has been extensively used for the transposition of EU directives since the early 1990s, mainly by using the legge comunitaria as a large delegation law. Also, the difference between ordinary statutes and legislative decrees stands out when we look at the variation in Europeanization from a time- and sector-specific
20
Data available upon request.
158
angle. In comparison with ordinary legislation, the number of legislative decrees linked to Europe has steadily expanded over time affecting more clearly the policy sectors related to firstpillar issue areas, while largely neglecting those that pertain to second and third pillar policies. The second aim was to examine whether the deepening and expansion of EU legislative competence over an increasing range of policy areas impacted the executive-legislative balance in favor of the former. Isolating and weighting this specific effect is not easy as there are good reasons to hypothesize that other political factors (collapse of the First Republic’s party system, new electoral rules, government alternation) contributed to strengthen the government in the legislative arena. That said, the analysis suggests that Europe distinctly contributed to this ongoing process. Laws linked to Europe are more likely to originate from government initiatives, to contain delegations and to remain less amended. It is hypothesized that MPs’ passivity in EU affairs stems largely from the low importance these issues have for MP’s political career and electoral perspectives, and from their technical nature. A closer inspection of the amendatory activity across policy areas also reveals that what matters is not the presence of a EU link but rather the nature of the policy area. Therefore, by looking at the legislative process, it is not possible to talk about the presence of a “pro-Europe” or “anti- Europe” cleavage in Italy. To sum up, it is not argued here that monitoring tools available to parliamentarians are inadequate and that the executive branch will always end up on the winning side to the detriment of the parliament. Instead, it is argued that MPs’ impact can be substantial, but it strongly depends on the level of attention granted to the matter under discussion. MPs are more likely to become active, acquire the necessary knowledge and use their political leverage when issues manage to leave “technical” foreign-policy agenda and enter the domain of domestic politics. Ultimately, the “internalization” of EU affairs constitutes a critical variable for the rebalance of powers in the parliament-government subsystem. The traditional portrait of the 60-year old relationship between Italy and Europe as a happy marriage should not make one turn a blind eye to the existence of winners and losers or maybe just to the existence of leading governments and unwilling legislators.
159
Bibliography Borghetto, E., Giuliani, M. & Zucchini, F., 2009. Quanta Bruxelles c’è a Roma? L’europeizzazione della produzione normativa italiana. Rivista italiana di Politiche Pubbliche, 2009(1), 135-162. Cotta, M., 1992. European integration and the Italian political system. In F. Francioni, ed. Italy and EC Membership Evaluated. London: St Martin's Press. Cotta, M., 1994. The Rise and Fall of the “Centrality” of the Italian Parliament: Transformations of the Executive-Legislative Subsystem After the Second World War. In G. W. Copeland & S. C. Patterson, eds. Parliaments in the Modern World. Changing Institutions. Ann Arbor, MI: The University of Michigan Press, pp. 59–84. De Micheli, C. & Verzichelli, L., 2004. Il parlamento, Bologna: Il Mulino. Della Sala, V., 1997. Hollowing Out and Hardening the State. European Integration and the Italian Economy. West European Politics, 1, 14-33. Di Palma, G., Fabbrini, S. & Freddi, G. eds., 2000. Condannata al successo?: l'Italia nell'Europa integrata, Bologna: Il Mulino. Dyson, K. & Featherstone, K., 1996. Italy and EMU as a 'Vincolo Esterno': Empowering the Technocrats, Transforming the State. South European Society and Politics, 1(2), 272299. Available at: http://eprints.lse.ac.uk/22076/ [Accessed October 21, 2009]. Fabbrini, S. & Dona', A., 2003. Europeanisation as Strenghtening of Domestic Executive Power? The Italian Experience and the Case of "Legge Comunitaria". Journal of European Integration, 25, 31-50. Ferrera, M. & Gualmini, E., 2004. Rescued by Europe? Social and Labour Market Reforms in Italy from Maastricht to Berlusconi., Amsterdam: Amsterdam University Press. Gianniti, L. & Lupo, N., 2004. Il governo in Parlamento: la fuga verso la decretazione delegata non basta. In S. Ceccanti & S. Vassallo, eds. Come chiudere la transizione. Cambiamento, apprendimento e adattamento nel sistema politico italiano. Bologna: Il Mulino, pp. 225-246. Giuliani, M., 2006. Le politiche pubbliche in Italia. La politica europea, Bologna: Il Mulino. Goetz, K.H. & Hix, S., 2001. Europeanised Politics? European Integration and National Political Systems, London: Frank Cass Publishers. Hine, D., 1993. Governing Italy. The Politics of Bargained Pluralism, Oxford: Claredon Press. Katz, R.S. & Wessels, B., 1999. The European Parliament, the national parliaments, and European integration, Oxford: Oxford University Press. Kreppel, A., 2009. Executive-Legislative Relations and Legislative Agenda Setting in Italy: From Leggine to Decreti and Deleghe. Bullettin of Italian Politics, 1(2), 183-209. 160
Maurer, A. & Wessels, W., 2001. National parliaments on their ways to Europe: losers or latecomers?, Baden Baden: Nomos. Nugent, N., 2006. The Government and Politics of the European Union, London: MacMillan. Radaelli, C.M. & Franchino, F., 2004. Analysing Political Change in Italy. Journal of European Public Policy, 11(6), 941–953. Schmidt, V.A., 2006. Democracy in Europe: The EU and National Polities, Oxford: Oxford University Press. Vassallo, S., 2001. Le leggi del governo. Come gli esecutivi della transizione hanno superato i veti incrociati. In Parlamento e processo legislativo in Italia. Continuita' e mutamento. Bologna: Mulino, pp. 85-126. Zucchini, F., 1997. L'attività legislativa del Parlamento italiano: consociativismo? Polarizzazione? Rivista italiana di scienza politica, (3), 569–610.
161
Table 1: Italian legislative production (1987-2006) Legislature
Years
Laws without International Treaties(N.)
10th
1987 1988 1989 1990 1991 1992 Total (daily average)
11th
1992 1993 1994 Total (daily average)
12th
1994 1995 1996 Total (daily average)
13th
1996 1997 1998 1999 2000 2001 Total (daily average)
14th
2001 2002 2003 2004 2005 2006 Total (daily average)
42 132 174 195 204 114 861(0.49)
Executive initiative
Average number
(% over Total without I.T.)
of delegations
90.48% 69.70% 69.54% 56.41% 59.80% 38.60% 61.21%
34 151 64 249(0.35)
85.29% 69.54% 56.25%
64 95 22 181 (0.24)
93.75% 74.74% 81.82%
80 112 113 120 104 76 605 (0.33) 52 89 96 92 80 47 456(0.25)
68.27%
82.32% 78.75% 75.00% 73.45% 67.50% 63.46% 50.00% 68.60% 82.69% 76.40% 70.83% 68.48% 68.75% 38.30% 69.08%
Total Legislative decrees
Laws converting Decree laws (% over Total without I.T.)
per law 0.05 0.04 0.03 0.35 0.03 0.34 0.15
3 23 24 31 65 146 (0.08)
71.43% 28.03% 27.01% 13.85% 16.67% 10.53% 21.72%
0.97 0.14 0.55 0.36
48 57 11 116 (0.16)
55.88% 47.02% 43.75% 47.39%
0.02 0.26 1.50 0.33
9 26 16 51( 0.07)
82.81% 57.89% 63.64% 67.40%
0.25 0.40 0.50 0.68 0.30 0.18 0.41
28 89 79 114 76 69 455( 0.25)
58.75% 28.57% 17.70% 27.50% 26.92% 18.42% 28.76%
0.15 0.48 0.73 0.28 0.76 0.26 0.48
8 38 80 62 78 54 320 (0.18)
67.31% 48.31% 37.50% 45.65% 40.00% 25.53% 43.86%
162
Table 2: European keywords EU keywords (Italian)
Abbreviation
EU keywords (English)
Comunità economica europea
CEE
European Economic Community
Comunità europea
CE
European Community
Comunità europea del carbone e dell’acciao
CECA
European Coal and Steel Community
Comunità europea dell’energia atomica
EURATOM
European Atomic Energy Community
Diritto comunitario
Community law
Diritto europeo
European law
Mercato comune
Common Market
Mercato europeo
European Market
Mercato unico
Single Market
Sistema monetario europeo
SME
European monetary System
Unione europea
UE
European Union
Unione monetaria europea
UEM
European monetary Union
163
Figure 1: Europeanization of the legislative production in Parliament (1987-2006) 250
50% 45%
200
40% 35%
150
30% 25%
100
20% 15%
50
10% 5%
X
XI Not Europeanized
XII
XIII Europeanized
2006
2005
2004
2003
2002
2001
2001
2000
1999
1998
1997
1996
1996
1995
1994
1994
1993
1992
1992
1991
1990
1989
1988
0% 1987
0
XIV % Europeanized
164
Figure 2: Europeanization of the legislative production in Parliament in terms of policy areas (1987-2006) 350 300 250 200 150 100 50
Extensive EU involv.
Consid. EU involv.
Policy responsability shared between the EU and the member states
Not Europeanized
Limited EU involv.
Law&Crime
Housing
Government Operations
Civil Rights / Liberties
Social Welfare
Public Lands/Resources
Health
Education
Defense
Culture&Media
Labor&Immigration
Transportation
Science & Technology
International Affairs
Environment
Energy
Macroeconomics
Banking & Commerce
Foreign Trade
Agriculture
0
Virtually no EU involv.
Europeanized
165
Figure 3: Proportion of legislative decrees per year and type of EU link
120
90% 80%
100 70% 80
60% 50%
60 40% 40
30% 20%
20 10%
X
XI No EU link
XII EU link/Transposition
XIII Only EU link
2005
2004
2003
2002
2001
2001
2000
1999
1998
1997
1996
1996
1995
1994
1994
1993
1992
1992
1991
1990
1989
0% 1988
0
XIV % EU link
166
Figure 4: Number of Legislative decrees by policy area and type of EU link
140 120 100 80 60 40 20
Extensive EU Considerable involv. EU involv
Policy responsability shared between the EU and the member states
Limited EU involv.
Law&Crime
Housing
Government Operations
Civil Rights / Liberties
Social Welfare
Public Lands
Health
Education
Defense
Culture&Media
Transportation
Science & Technology
Labor&Immigration
International Affairs
Environment
Energy
Macroeconomics
Banking & Commerce
Foreign Trade
Agriculture
0
Virtually no EU involv.
No EU link EU link/Transposition Only EU link
167
Table 3: EU link in Executive-sponsored laws, delegations 1)Executive-sponsored laws EU link % over total laws with EU link
No EU link %over total laws without EU link
2)Delegations EU link Average number
No EU link Average number
10th
54.87%
62.03%
0.96
0.02
11th
53.33%
70.32%
2.20
0.11
12th
84.62%
82.14%
4.38
0.01
13th
73.58%
67.54%
1.88
0.09
14th
77.22%
67.37%
2.46
0.07
168
Table 4: Approved amendments EU link Years Percentage (Legislature) of laws with at least one amendment
No EU link
Approved
Approved
Percentage
Approved
Approved
Total Amend.
Parl. Amend.
of laws
Total Amend.
Parl. Amend.
per 100 words
per 100 words
with at least one amendment
per 100 words
per 100 words
1997-2001 (13th)
41.24
0.25
0.20
30.61
0.25
0.23
2001-2006 (14th)
77.22
0.20
0.18
42.71
0.27
0.25
Total
57.39
0.23
0.19
36.27
0.26
0.24
169
Table 5: Approved Amendments per policy areas.
Domestic Macroecon omic Issues Civil Rights, Minority Issues, and Civil Liberties Health
Tot Percent of Amendme laws with nts at least an per 100 amendmen words t La Law La Law ws s ws s wit with 80, 23,9 wit 0.1 with 0.1 h65 out 1 h 8 out E.L E.L. E. E.L. . L. 33,3 0.9 0.35 3 7
75 20
36,9 6 40,5 4 40
0.1 1 0.1 6 0.3 4
0.33
13th Leg. Percent of Amendme laws with nts at least an per 100 amendmen words t La Law La Law ws s ws s wit with 68, 25,9 wit 0.2 with 0.18 h75 out 3 h 8 out E.L E.L. E. E.L. . L. 66, 16,6 0.9 0.36 67 7 9
14th Leg. Percent of Amendme laws with nts per 100 Nugent’s at least an words Classifica amendmen tion* t La Law La Law ws s ws s wit with 93, 21,0 wit 0.1 with 0.06 3 h33 out 5 h 0 out E.L E.L. E. E.L. . L. 100 44,4 0.6 0.34 1 4 2
0.23
14, 29 20
0.22
50
0.25 0.29
16, 67 25
0.15
0
63, 64 71, 43
55,1 7 48,0 3
0.3 7 0.3 3 0.8 8 0.3 1 0.2 0
0.23
57, 14 33, 33
50
19,0 0.5 5 7 31,8 . 2
0.32
63, 64
36,3 6
0.3 1
0.17
60
33,3 3
0.3 0
0.17
66, 67
38,7 1
0.3 2
0.16
4
33,
32,8
0.1
0.21
33,
34,2
0.2
0.30
33,
30,7
0.0
0.11
2
Agricultur e Labor, Employme nt and immigratio n Education
35, 29 66, 67
Environme nt Energy
54, 55 50
28,2 1 35,7 1 50
Transporta tion Law, Crime, and Family Issues Social Welfare Communit y Developm ent and Housing Issues Banking, Finance, and Domestic Commerce Defense
50
-
0.34
50
0.08 -
27,5 9 26,0 9 35,7 1
0.0 8 0.0 5 0.2 9
0.28
33, 33 0.09 57, 14 0.19 100
52,9 4 64,2 9 43,7 5
0.1 5 0.2 4 0.5 3
0.41
2
0.49
5
0.26
3
20,8 3 23,5 3 66,6 7 53,8 5 45,0 7
0.0 5 0.2 8 -
0.28 100
40
0.18
2
0.33
4
0.14
3
0.3 5 0.5
0.14
0.6 2 0.3 8 0.9 9 0.2 9 0.1 3
0.41
3
0.40
1
33,3 3 20
0.5 7
0.38
0
-
2
0.01
41,6 7
0.20
1
0.27
71, 43 0.17 100 75
0.32 100
54,5 5 45,4 5 56,2 5 51,7 9
170
33 25
1 43,7 5
0 0.2 4
33 -
1 41,6 7
1 -
33 0.53 100
7 50
1 0.4 8
Space, 0.49 0.39 3 Science, Technolog y and Communic ations Foreign 50 25 0.2 0.16 33, 0 0.0 0.00 100 50 0.6 0.24 5 Trade 6 33 4 6 Internation 70, 0.2 0.13 55, 11,1 0.0 0.08 87, 31,0 0.1 0.16 3 al Affairs 59 9 56 1 4 5 3 7 and Foreign 21,4 Aid 3 Governme 66, 45 0.1 0.43 57, 29,5 0.4 0.4 72, 57,1 0.2 0.46 nt 67 5 14 5 5 73 4 8 Operations Public 46,1 . 0.56 57,1 0.40 33,3 0.88 2 Lands and 5 4 3 Water Manageme nt Culture 0 9,3 0.0 0.05 0 8 0.0 0.03 11,1 0.0 0.12 and 0 0 1 0 entertainm ent * 1= No EU involvement; 2=Limited EU involvement ; 3= Policy responsibility shared between the EU and the member states; 4=Considerable EU involvement; %=Extensive EU involvement
171
Figure 5: Voted amendments (1997-2006)
Laws without E.I Years and Legislatur e
Amendments Voted Amendments per 100 words
Amendments per 100 words
1997
269
0.26
909
0.49
1998
2037
1.45
3085
2.49
1999
2273
1.34
3039
2.16
2000
2773
2.55
4069
2.15
2001
1640
1.49
940
1.52
8992
1.42
12042
1.71
2001
1443
1.99
1920
3.54
2002
1580
1.63
3560
2.31
2003
1858
1.47
1239
0.95
2004
2248
2.16
2614
2.21
2005
1261
1.62
2287
0.98
2006
637
0.96
135
0.38
9027
1.66
11755
1.62
18019
1.53
23797
1.67
Leg.13
Leg.14 Tot
Voted Amendments
Laws with E.I.
172
Table 6: Laws without European link
Laws about the same
% sponsored
Average number
Approved amendments
by the government
of delegations per law
per 100 words
68.9
0.06
0.23
57.3
0.02
0.25
issues treated in the laws with E.L. Laws about issues not addressed in the laws with E.L.
173
Table 7: Similarity indexes in terms of issues: Laws sponsored by MPs, Laws sponsored by the government, Laws with a European link
Laws sponsored
Laws sponsored by MPs
Laws sponsored by the government
Laws with E.L..
Pearson Correlation
Pearson Correlation
Pearson Correlation
Dunca n index 1
0
Duncan index
Dunca n index
0.479**
44.67
0.158*
55.19
1
0
0.715**
38.98
1
0
by MPs. Laws sponsored by the government Laws with E.I.
174
8. The Europeanization of domestic legislation in Luxembourg
Patrick Dumont & Astrid Spreitzer Université du Luxembourg
On 28 July, 2009, on the occasion of his first speech as Chairman of the Chamber of Deputies, the Christian democrat Laurent Mosar stated that ‘70 % of the bills adopted by the Chamber were related to European directives’, and that for this reason national MPs would have to be more involved in the whole European legislative process rather than acting only at transposition time. He recalled that the Lisbon Treaty would give national parliaments more means to do so, and wished to this end that the government would inform the assembly as swiftly as possible about preliminary legislative texts discussed at the European level. During the 2009 campaign for the election of the European Parliament, two parties had even more explicitly referred to Jacques Delors’ so-called prediction in their electoral manifesto. In order to stress the importance of EP elections (in a country that has organized its national elections at the same time as European elections ever since 1979), the Liberals (DP) 1 stated that 80% of national legislation originated in the initiatives and rulings decided at the European Union level, the EU being nowadays the framework of reference for all policies. The Alternative Democratic Reform party (ADR)2, which was the party most critical of European integration represented in parliament at the time also referred to the 80% figure with a quite different intention, hereby denouncing the lack of involvement of citizens and of national parliaments in the vast majority of decisions that were taken at the European Union level. These recent examples highlight the fact that Delors’ claim dating back from the 1980s provides ammunition for both pro- and anti-European integration political camps in Luxembourg, as it does in other European countries. A renewed interest in European affairs and how they affect domestic politics has emerged since the referendum on the European
1 2
Demokratesch Partei (DP), member of the ALDE Alternativ Demokratesch Reformpartei (ADR), associate member of the UEN
175
Constitutional Treaty in 2005, and the new provisions of the Lisbon Treaty regarding the involvement of national parliaments have put this question on the electoral and parliamentary agendas. The reference is very handy to use in the political discourse of parties and politicians with quite different positions on the issue of European integration in general. Whether coming from the pro-integration Christian Socialists (CSV)3 and the Liberals (DP) or from the most sovereignist party (ADR) in Luxembourg, these statements were not backed by empirical proof because as of yet, no scientific analyses of the Europeanization of domestic legislations have been performed in the Grand Duchy. This chapter is a first attempt at filling this gap.
1. Background
1.1. Luxembourg and the EU
Ever since its creation, Luxembourg’s small size has heavily influenced its choices. First, it accounts for its unitary constitution. There is no regional political or administrative level, only national and local administration exist. Hence, both the national legislative process and the handling of EU policies – that is their preparation, decision making and implementation – are centralised. Second, militarily weak and dependent on foreign markets, Luxembourg has always participated in (or even initiated) international political and economic agreements, as participation to international regional integration was the only possible way for such a small country to be potentially heard on the international scene. The design of supranational institutions like those leading up to the European Union certainly provide Luxembourg with considerably increased leverage. The sheer size of the country as compared to neighbouring France or Germany further enhances its chances of having an advantage in the process of regional integration, because it is not seen as a threat by the powerful member states. To illustrate this with examples, the executive of the European Coal and Steel Community (the High Authority) was “temporarily” installed in Luxembourg up until the merger of all the three European communities in the mid-sixties; the Grand Duchy representatives then managed to receive compensations (mainly the installation of the Court of Justice and the 3
Chrëschtlech Sozial Vollekspartei (CSV), member of the EPP
176
Court of Auditors) in the negotiations about the locations of the new European institutions for this loss of the ECSC High Authority (the question of the seats of the EU institutions is still one of the few issues of vital national interest in Luxembourg); of the ten presidents of the European Commission, two came from Luxembourg (Gaston Thorn 1981-1985 and Jacques Santer 1995-1999), current PM Juncker was cited as potential best candidate for 2004, became President of the Euro Group instead and was close to becoming the first President of the European Council in December 2009. As a founding member of the ECSC and the EEC, and being a small Member State, Luxembourg has always favoured European integration and the Community method, even though the national implementation of EU directives and the expansion of the EU’s policy competence certainly constrain its government. The 1992 Maastricht Treaty is a striking example. The right of all EU citizens to participate in local elections in Luxembourg violated articles 52 and 107 of the Constitution.4 Neither of these was included in the June 1989 declaration of revision prepared for amending the Constitution and could therefore not be reformed before the Treaty was to enter into force unless the Chamber of Deputies was dissolved and anticipated elections organized. A technical solution was found in order to avoid an electoral campaign that would inevitably have been fought on the issue of these voting rights for foreign citizens.5 Luxembourg was, as a consequence, one of the first countries to ratify the Maastricht Treaty, which was adopted on 27th July 1992 by a large majority in parliament including its unconstitutional provisions after the three traditional parties agreed that the relevant articles would be declared open for revision in 1994.6 This episode marks one of the first breaches in the general political consensus formerly shared by the political class, the social forces of the country and its population regarding the participation of Luxembourg in transnational economic and political structures. Experts denounced the government’s actions in the press and associations were created to defend and promote respect for the Constitution. Apart from the claim to respect the Constitution, the 4
As the Constitution did not specify voting rights for the European elections, the Treaty’s measure giving EU citizens passive and active voting rights did not conflict with the national fundamental Law. 5 The Treaty mentioned the deadline of 31 December 1994 for the Council of the Communities to finalize the details of implementation. This was interpreted by Luxembourg’s Council of State as the deadline for amending the Constitution, as these modes of enforcement of the voting rights were needed for their actual exercise. Hence, it was possible to “wait” until the scheduled June 1994 elections. 6 The Constitution was duly amended in December 1994, one week before the deadline set up by the Treaty for deciding on the modes of implementation of the voting rights. Non-national citizens were not yet allowed to vote for the 1993 local elections.
177
question of European citizenship and the right for EU residents to vote in local elections were divisive in the population, as some saw these as potential threats for the national sovereignty and identity in a country where, at the time, about 30% (above 40% nowadays) of the residents were foreigners. After an economic slowdown in the early 2000s, many Luxembourgers, who increasingly felt threatened by the growing importance of foreign employment in the country (cross-border workers and foreign residents represent the large majority of the labour force), also started to worry about the prospects of European fiscal harmonisation and its potentially negative consequences on the crucial banking sector which would indirectly hurt the national economic and social way of life.7 These fears, together with the population’s massive rejection of future EU enlargements (mainly to the Balkan states and Turkey) of, materialized in the run-up to the referendum on the European Constitutional Treaty, which made up the country’s first electoral campaign focussed on European matters and the influence of EU decisions on domestic politics. The outcome of this referendum revealed a clear disjunction between the parties represented in the Chamber that massively favoured the adoption of the European Constitutional Treaty and the population who barely approved the text with a 56.5% majority. Traumatic for most representatives of the political elite, this occasion entailed a number of institutional reforms in order to improve citizens’ information on Europe, encourage the participation in European elections, and involve national MPs in European decision-making processes (see below). Overall, the consensus on European integration shared by the Luxembourg political parties and the population has not been tremendously altered: During the 2009 electoral campaign, all electoral programs - except the one by the Communists - for the EP elections described Luxembourg’s EU membership as a necessity and as a benefit for the country. The parties less enthusiastic about Europe, but willing to reform the EU institutions are the ADR, The Left (déi Lénk)8 and the newly created BiergerLëscht, a party centred around one outgoing MP who left the ADR during the parliamentary mandate. The latter opposed the adoption of the Lisbon Treaty and claimed to represent those who voted 7
Fears of fiscal harmonisation felt by domestic institutions actually date back to the 1980s, as evidenced by the quasi-Eurosceptic advice of the Council of State on the Single European Act arguing that European integration was pushed too far (Schmit 2006). Luxembourg was at that time successfully developing its financial place with new products therefore re-assuring its economic and political independence after the deep crisis of its steel industry. The following treaties creating and deepening the European Union were on the other hand welcomed by the Council of State. 8 Member of the European Left.
178
against the European Constitutional Treaty at the 2005 referendum. Altogether, these less Euro-enthusiastic political parties received the support of less than 14% of the voters in both the 2009 national and European elections. They gathered 8.5% of the Chamber seats but none of Luxembourg’s seats in the European Parliament.
1.2.
Luxembourg’s legislative process
The national legislative process is rather lengthy, as it involves a number of nonparliamentary actors that do not have sufficient personnel and expertise to allow for a swift production of advice formally requested for the adoption of new legislation. The Council of State is the government’s legal advisor and somewhat compensates for the lack of a second chamber9 by giving compulsory advice on each bill or amendment. Though the advice is nonbinding, voting in Parliament cannot take place until it is received. It can also decide on whether or not legislation must be put to a second vote. Article 59 of the Constitution stipulates, in effect, that all bills must pass a second vote in the Parliament no earlier than three months after the first vote. This second vote can only be avoided if plenary sessions of the Parliament and Council of State decide that it is unwarranted. The Council of State thus has a formal suspensive veto, which, in practice, is rarely used. The elected professional chambers, while not enshrined in the Constitution, play an important role on bills relating to particular professional interests, including appropriations in the national budget. Even though failing to consult them would not constitutionally nullify the law, in practice the Chamber of Deputies usually waits for their (non-binding) advice before holding a final vote. In contrast to laws, executive decrees (‘Grand Ducal decrees’ and ‘regulations’, see below) voted on without the advice of the professional chambers can be rendered null and void.10 Professional interests are also represented in the ‘Tripartite committee’ for crisis management, created in 1977 by means of a government bill. This
9
Political elites considered that a bicameral system was not adapted to such a small country as Luxembourg. Since 1966, the government may also consult the Economic and Social Council (comprised of the forces vives of the country) on broader economic, financial and social legislation, even though the latter plays no part in any compulsory phase in the legislative process. Its main function is to help reconcile divergent interests (e.g. conflicting advice from the professional chambers). 10
179
corporatist institution, composed of the government, employers and employees’ (trade unions) representatives, issues advices when one out of three well-defined unemployment thresholds is reached. Not only is this process compulsory, but in practice its recommendations are binding. The parliament does not take part in this interchange between labour and government representatives, but has to ratify agreements and formalise them into law. GRAPH 1 ABOUT HERE Graph 1 allows us to analyze the law-making activities of the government and MPs longitudinally. The dotted lines in the graph represent the start and the end of a legislative mandate (five years) that also corresponds to the duration of the successive coalition governments throughout the period. The 1994-1999 mandate marks the only change of Prime Minister in the period, as Jacques Santer became President of the European Commission in January 1995, after having fought national elections and formed a new government in 1994. He handed his position over to Jean-Claude Juncker who has kept it ever since. Their party, the Christian democrats, governed throughout the period in coalition with the Socialists (LSAP)11 except for the 1999-2004 term, when they ran the country together with the Liberals as junior partner. Across legislative mandates, we observe a positive trend in the number of bill projects, introduced by the government, compared to a stable amount of bill proposals, which are put forth by MPs. With only two exceptions, the number of bill proposals remains less than twenty per year throughout the whole period. MPs appeared to be slightly more productive during the three post-1989 terms than in the first and last periods under study here. Within those five successive legislative terms, we quite logically find a low number of projects introduced on the first and last of the five ordinary sessions of the Chamber, and a last parliamentary year before the elections also characterized by a higher number of adopted bill projects than projects introduced. GRAPH 2 ABOUT HERE
11
Lëtzebuerger Sozialistesch Aarbechterpartei (LSAP), member of the PES
180
Graph 2 presents the adoption rates of each type of legislative initiatives. The government clearly has its way most of the time, with an average adoption rate largely over 90 percent. MPs were at times also successful in having their proposal approved by the Chamber, with a record high during the 1984-1989 term. This was also the mandate with the lowest absolute number of bill proposals (only 27 in five years). The 1999-2004 term reveals an improvement of the adoption rate of bill proposals (that were also the most numerous in absolute numbers, 81). Correspondingly, a lowering of the dominance of bills emanating from the government can be observed and is reflected by the decrease of the adoption rate of bill projects compared to the previous parliamentary mandate. The following mandate resulted in a return to past levels of adoption rates for both bill projects and, to a lesser extent, bill proposals. The numerical importance of the opposition, the reduced dominance of the Christian democrats in government after a bad electoral result and a different view on the management of coalition politics with the Liberals compared to that with the Socialists during this legislative mandate may partly explain this outcome. Overall, even though the situation has improved in recent years, the relative lack of parliamentary resources (in terms of personnel, expertise and documentation) at the disposal of MPs to some extent accounts for the low absolute number of bill proposals introduced in each parliamentary year. The constraints of majority rule (coalition agreement, inter-fractional meetings that restrain initiatives of coalition MPs), and the risk of seeing one’s proposal blocked at the Council of State advice stage (because government bills are given priority) also contribute to this outcome. Finally, like in other European countries, most MPs in Luxembourg prefer to invest time to more rewarding tasks than technical talks in parliamentary committees. The highly personalized electoral system (Dumont and De Winter 2003) and the fact that most MPs also have local duties (Dumont, Kies, Poirier 2010) are factors that largely explain this behaviour in the Grand Duchy.
2
The Europeanization of the Luxembourgish legislation
2.1. The domestic management of EU policy
181
The Ministry of Foreign Affairs has always been, and still is in large part, the key body as far as formal relations with the EU and national institutions are concerned. Until 2006, when the European Commission started communicating all its legislative proposals and initiatives directly to the Chamber of Deputies, all EU written documents were channelled through the Ministry to ensure that all necessary information was communicated to ministerial departments, to the Conference of Presidents of the Chamber of deputies and its Foreign and European Affairs Committee. In the upstream process, the Ministry of Foreign Affairs asks the relevant technical ministries to analyze the relevant documents and draft their comments, before sending these observations to the permanent representation in Brussels, together with (or without) special advice from the Minister of Foreign Affairs. These ministerial evaluations constitute the national position for the matter at hand. In case of divergent evaluations, an inter-ministerial committee for European policy (CIPCE) discusses the issue.12 If and when the problems are not settled or national interests are at stake, the question may be debated in the Council of government. Since 2000 the relevant ministries are requested to write an introductory note for the corresponding permanent technical committees in parliament on a selection of potentially important EU directives. These documents are thus also treated in parliament, and committees may formulate their position. Finally, interest groups affected by a directive on the EU agenda are also consulted on an ad hoc basis,13 and so are partners of the Benelux (Bossaert and Vanhoonacker 2000), if needed, in the same form of informal consultation. In the downstream process, the role of the Ministry of Foreign Affairs is also crucial. The competent ministries analyze the technical and administrative consequences of the adopted directives and propose the instrument to be used for its transposition to the Ministry of Foreign Affairs, if a transposition is at all needed given the current state of domestic 12
CICPE stands for “Comité interministériel de coordination de la politique européenne”. This committee was set up in November 2005 and is chaired by the Ministry of Foreign Affairs. A less formalized body, with an ad hoc composition existed before 2005 and was referred to as Comité de l’Union. The first visible attempt at formalising coordination between ministries in the treatment of EU issues was the appointment, since 1993, of a European Correspondent in all ministries. These individuals are supposed to keep track of major EU matters and monitor the correct implementation of EU directives (Bossaert 2003). They form a network that is the main communication tool between EU institutions and domestic civil service. The CICPE gathers high civil servants of ministries concerned with current European debates as well as the permanent representative of Luxembourg to the EU. It was set up in order to complement this network of European correspondents and enhance coordination of European policy between European and domestic levels and across ministries at home. 13 For instance the Economic and Social Council referred to in footnote 11 is very much interested in EU affairs and calls for more formal consultation in order to safeguard the position it acquired in the national legislative process.
182
legislation. The technical ministry then drafts the legal text to be adopted, which is then submitted, after deliberation in the Council of government, to the Chamber of deputies in case the instrument chosen is either the law or the Grand Ducal executive decree on the basis of the loi d’habilitation spéciale (see below). A method for handling the transposition of European directives, initiated by the Minister-delegate for Foreign Affairs was established gradually from 2006 onwards to involve domestic actors at an earlier stage of the EU policy process. Since 2008, reports of delays in transposing directives are also evoked during Council of government meetings. At the end of the downstream process, all transposition measures are published in the Mémorial (official journal) and the Ministry of Foreign Affairs notifies the transposition to the EU Commission.14 As the central body of coordination, the Ministry of Foreign Affairs more generally formally monitors the follow-up of EU directives and reports to the Chamber on the progress of transposition rates. However, in accordance with the decentralised functioning of the government, technical ministries are competent and responsible for the correct implementation of EU directives.
The involvement of the Chamber of Deputies in EU affairs is, according to comparative analyses (see Bergman 1997 and 2000; Raunio and Wiberg 2000; Maurer and Wessels 2001), particularly weak with regard to the elaboration of the national position. The parliament’s lack of administrative and scientific staff (it now consists of about 70 civil servants), and the lack of clear and systematic procedures for the handling of EU policy at an early stage accounted, in part, for this result. Furthermore, there is no parliamentary committee exclusively devoted to European affairs in Luxembourg, despite repeated attempts to create a specific advisory body dealing only with European matters. Since December 1989 the name of the Foreign Affairs committee has simply been changed to Foreign and European Affairs committee. Since 1999, Defence competencies have even been added to this committee and nowadays this committee is responsible for ‘Foreign and European Affairs, Defence, Cooperation and Immigration’. Since 2009 this Committee consists of 12 members (altogether, there are 20 members but only 12 of them are allowed to take part of the Foreign and European matters). Most members of this committee prefer dealing with competencies that appear closer to their
14
The EU Commission can monitor transposition as projects of transposition are often transmitted to the Commission during the process for advice.
183
electorate. The six Luxembourg MEPs are invited to participate to the committee meetings,15 but their rate of participation is not very high. Since November 2003, the Standing orders of the Chamber of deputies specify that MEPs may also be invited (given the authorisation of the Chairman of the parliament) to any permanent technical committee dealing with the elaboration or transposition of EU law. They have no voting rights, nor can they ask questions to ministers if the latter are invited. They cannot participate in plenary meetings either. The trauma of the 2005 referendum in Luxembourg, the constant problems of the Grand Duchy in transposing directives in due time and the 2006 decision of the European Commission to send all of its documents directly to national parliaments as an anticipated and unilateral decision to implement parts of the provisions envisaged by the European Constitutional Treaty, echoed profound changes in the internal organization of the management of EU policy by the Chamber. First, the Chamber appointed a permanent representative to Brussels16. Second, the International Relations service of the Chamber is now in charge of classifying all documents emanating from EU institutions into two categories: 1) those without political, economic, legislative or financial interest for Luxembourg or those for which the process at the European level is too advanced to expect any influence of the national parliament anymore, and 2) those that call for a detailed analysis.17 Third, a ‘European dossiers’ point is on the agenda of any meeting of a parliamentary committee, the latter may invite members of the European Commission or European civil servants to obtain further explanations on the policy initiatives envisaged. Finally, a modification of the standing orders of the Chamber adopted unanimously in 2009 specifies the measures to be followed regarding the cooperation between the government and the parliament. In the procedure of control of the subsidiarity principle, the government may 15
These are usually scheduled on Monday mornings so that MEPs can participate. Hand in hand with the rotating Council presidency goes the membership of the respective national parliament in the COSAC-secretariat (Conference of Community and European Affairs Committees of Parliaments of the European Union). On this occation, national parliaments mostly install a staff position in Brussels, which normally is maintained even after the 18 months of membership in the COSAC secretariat as a liaison office. Luxembourg formally established its liaison office in January 2006. 17 Since the direct sending of documents of the European Commission to the Chamber in 2006, we have a clearer view of the proportion of such documents that are considered to call for Chamber action or analysis. For the years 2006, 2007 and 2008, out of more or less 800 documents sent per year, respectively 42%, 45% and 47% were judged as important enough to be sent for consideration to parliamentary committees. Communications of the Commission are the most numerous documents considered, followed by reports, decisions, rulings and directive proposals. The committee that received the largest proportion of these documents was the Foreign Affairs Committee, usually followed by the Economic, Finance, Transports, Environment and Justice Committees. The committees to which most directive proposals were sent were the Economic, the Environment, Foreign Affairs, Transports and Finance committees. 16
184
be asked to assist the Chamber in the research phase (by producing an evaluation of the impact of an European Commission proposal on domestic legislation) in order to craft a well informed position of the parliament on the dossier at hand. The government is formally obliged to present an annual report on European policy and an annual report of the transposition of European directives and the implementation of EU law. These may be followed by a plenary debate. The Chamber is also compulsorily informed about the national position envisaged by the government in the event of Treaty revision or enlargement, decided upon at the occasion of Intergovernmental conferences. The government-parliament cooperation concerning EU policies was therefore eventually formalized, whereas in the previous situation most of it was dealt with in an informal, ad hoc manner.
2.2. The transposition of EU law In Luxembourg, constitutional engineering to allow a (temporary) transfer of national sovereignty to international cooperation dates back to 1956, a few years after the adoption of the ECSC Treaty. For such an international Treaty to be ratified, it has to be signed by the government (officially the Grand Duke) and voted on by the Parliament if two thirds of the votes (blank votes excluded) are cast in support.18 Article 37 of the Constitution makes the government responsible for implementing treaties, but it has never been used as the legal basis for the transposition of EU directives. Recent advice of the Council of State on 2002 and 2003 proposals for the revision of this article of the Constitution point to the complexity of transposing EU directives in Luxembourg’s legal order in due time. They suggest, under conditions to be specified by law, to lift the constraint of excluding ‘matters specified in the Constitution as being exclusively reserved to the law’ from the field of executive decrees.19 This recommendation thereby opened the door for the transposition of EU directives through executive decree based on a revised constitutional provision in the future.
18
Up until the December 17th 2003 constitutional reform, Treaties had to be approved by parliament with the same quota (three quarters of all MPs had to be present) and decision rule as constitutional amendments. This quota constraint on the approval of Treaties has been lifted up. 19 These calls were relayed by the then Minister of Justice Luc Frieden at the 150th anniversary of the Council of State in November 2006.
185
In any case, a large majority of EU directives is transposed into Luxembourg’s legal order through executive decrees that do not require any, or just minimal, parliamentary participation. The most common executive instruments, especially used when no new specific obligations are introduced in the national order by an EU directive, are the “normal” Grand ducal and the ministerial decree taken according to the general executive’s power of implementing laws (articles 33 and 36 of the Constitution). This is the lightest procedure, as these instruments are simply deliberated in the Council of government (such a discussion does not even take place for ministerial decrees), signed by the Grand Duke and counter-signed by a minister, and finally published in the Official journal (Mémorial). In order to avoid waiting for the mandatory advice of the Council of State, the government may invoke urgency. It usually does so for directives that do not introduce fundamental new obligations or when the existing national legal order already covers the field of the EU measure.
A law from August 1971 especially designed for the implementation and sanction of Community decisions and directives in economic, technical, agricultural (including forestry), social and transportation matters gives the executive another legal basis (habilitation spéciale) for transposition instruments.20 It is used when the contents of the EU directive imply a more important transformation of the national legal order and thus necessitate more consultation. This requires a procedure involving the mandatory advice from the Council of State, which may refuse the use of this simplified form of transposition and insist on having the directive transposed by law. The process requires asking professional chambers’ advice and the assent of the Conference of Presidents of the Chamber, but allows the government to avoid taking the ordinary lengthy legislative procedure. Reserved competences specified in the Constitution, but also fiscal, customs and public health affairs are excluded from the rather large field. Normal laws are only used when a European directive introduces a new obligation or if the subject matter is one that the Constitution vests exclusively in the legislature. Apart from
20
Other potential executive instruments exist, such as the UEBL (Union économique belgo-luxembourgeoise) convention framework, that is however limited to the fields of customs only. Governments may also use executive decrees that refer to the annual loi d’habilitation through which parliament grants the cabinet with ‘temporary special powers’ (taking over the powers of parliament in order to speed up the process for specific issues), but the latter legal basis has so far never been used for the transposition of EU directives through executive decrees.
186
these objective criteria, the scope of legislative, economic and political changes included in the directive may be taken into account. There is no close one-to-one relationship between national law and EU directive when legislation is chosen as the transposition instrument, even though this may be sought by national MPs. Moreover, former pieces of national laws may be added to the content of one or several EU directives in order to result in a coherent bill and a somewhat rationalized legal order.21 On the other hand, a single directive may be transposed by more than one domestic legal instrument, sometimes of different types (a law and one or several executive decrees).22 The data provided by Luxembourg’s official journal for the 1984-2009 period that concern only laws and decrees still in force in 2009, show that laws are used about a quarter of the time when it comes to transposing EU directives, for three quarters of the time executives decrees serve this purpose. Given that some EU directives, or technical amendments to them, may be adopted and notified to the Commission without publication in the national official journal, with the exception of a simple reference to the publication of the act in the European official journal, this proportion surely overestimates the use of laws for transposition. The extent to which a single law may only partially transpose a given directive or, at the other extreme, transpose completely several ones, proves that quantitative analyses of the Europeanization of primary legislation on the basis of the transposition of EU directives are bound to face large methodological hurdles and seem unlikely to provide useful results for comparative research. It is quite difficult to clearly delineate new domestic laws from the transposition of European directives, as the former also often anticipate directives that are not already in force (especially when the topics are of high national interest or have tremendous consequences for the national order). Taking, therefore, such figures with all due caution, we see that among the 1,673 laws published from 1986 and 2006 and still in force in 2009, 140 (8.4%) transposed EU directives. The percentage of ‘Europeanized’ executive decrees is somewhat higher: 450 (11.55%) out of 3,896 Grand Ducal decrees transposed directives (only 1.19% of the 1,769 ministerial decrees produced in the period did). In the period between 21
Executive decrees may also merge two or more EU directives. In our dataset, for each type of instruments we find about the same proportion transposing more than one directive by national act (i.e. roughly 60% of the laws and executive decrees transposing one or more directives transpose one). 22 About half of the directives transposed in the period under study have been transposed by more than one domestic act. Directives transposed by only one act are much more frequent among executive decrees than among laws. Overall we can say that in Luxembourg we come closer to a one-to-one relationship between EU directive and national instruments when we deal with executive decrees than with laws.
187
2007-2009 the rate of laws transposing directives has increased (it now reaches 9.7% for the 1984-2009 period) whereas the rate for executive decrees remained stable.
Finally, Luxembourg is one of the EU’s worst pupils in terms of timely transposition of directives. Reports of the Commission on the comparative performance of member states in the transposition of EU directives pointing at the bad results of Luxembourg are commented in the press, and reports of successive Ministers of Foreign Affairs have ever since the 2000s clearly stated that the country had serious difficulties in the implementation of EU law. This bad performance was accompanied by an incapacity to cope with these delays and react quickly when the deadline had passed, often leading to the launch of infringement proceedings (Dumont 2005).23 Such results inspired the numerous reforms presented above, because the informal, pragmatic and flexible processes at hand had clearly shown their limits. The administrative shortage, typical of smaller countries, and its role on delayed transpositions of EU directives has already been studied in comparative analyses of performance of smaller countries (Thorhallson 2000; Falkner and Leiber 2003; Falkner 2003). Arguably, smaller States must mobilize comparatively more administrative resources to overcome EU complexity than larger ones. Even though the legislative route is more often than not self-chosen, the length of the law-making process and the level of interest of actors must be added to this chronic lack of personnel at the administrative and parliamentary level. For instance, government will is important when the legislative procedure is selected. In parliament, urgency is usually given to governmental bills on national matters that are more conducive to electoral rewards. Furthermore, MPs are also very involved in local affairs and thus parliamentary meetings almost never finish late at night. This means that the transposition of EU directives may be delayed for a marked lack of interest on the part of the government and often also the rest of parliament. Political delays may also be due to the degree of misfit (scope of changes to be undertaken in accordance to a given directive), so that the government prefers to allow more time to pass before introducing a bill in the
23
As in most of the cases infringement proceedings are abandoned and classified after a dialogue has installed between the Commission and the national administration, one might interpret this Luxembourg feature as a form of ‘pragmatic negligence’. But given the considerable additional workload occasioned to the – already overloaded – national administration by these infringement proceedings and the risks of being condemned and fined, it is more reasonable to point to the very administrative personnel and time shortage as priority is given to national matters (or sometimes very sensitive European issues).
188
Chamber. Moreover, the Council of State is short of staff and expertise to draft detailed reports on sometimes highly technical bill (or decree) projects of transposition and thus takes more time to execute its duty than for ordinary legislation. Anyway, most projects transposing directives have already passed their deadline for transposition when they are transmitted to the Council of State; not to mention the overall rise of acts to be examined by this body which increased for about 50% from 1994 to 2004. All of these factors together further increase the potential for late transposition.
2.2. Measuring the Europeanization of Luxemburgish legislation 2.2.1. Methods Two databases, maintained by the Chamber and the government (for the publication of the official journal) respectively provide basic search functions to retrieve the content of legislative documents and executive measures. The database of the Chamber24 enables searches for bills, Council of State recommendations and documents of the other actors involved in the process, as well as final texts voted on. The official journal (Mémorial A – Recueil de legislation25) holds published laws and executive decrees (the latter are not retrievable from the Chamber’s). Due to a lack of human resources, only a few basic statistics on the features of legislative acts are made available to the public: for instance no systematic entries exist in the Chamber database concerning the date of entry, date of vote, policy area, result of vote, amendments etc. Hence, coding to extract this information had to be done manually by consulting the homepage of the parliament and opening every single document leading to one of the around 2,000 laws published in the period from 1986 to 2006. Data collection, therefore, proved tricky and involved among other things the merge and comparison of pieces drawn from the two databases. A keyword search was conducted in both databases on laws published between 1986 and 2006, bringing up differing results, which were corrected for in a further step (see Table 1). The differences were due to differing logics of the two search engines. The degree of agreement between the two databases is about 60%,
24 25
To be found on the website of the Luxembourgish Chamber of deputies at www.chd.lu The database is available on the following website: http://www.legilux.public.lu/leg/a/index.php
189
with the dataset of the Chamber finding a much larger number of Europeanized laws. As a general rule, if a keyword was found in one of the two, it was coded as Europeanized.
TABLE 1 ABOUT HERE
This method of defining Europeanised laws in a rather generous way was justified by the fact that the Chamber search engine is not immune to flaws, and that the smaller number of Europeanized laws found by the Legitech instrument (due in large part to its case sensitivity) was on the other hand backed by solid evidence such as the frequency of occurrence of keywords in the texts. The keyword list common to all country contributions is based on about 20 French expressions. This list was used for an investigation of occurrences in the titles and texts of the laws voted on during the period (all Luxembourgish legislation is written in French even though parliamentary debates are held in Luxembourgish). A law was classified as Europeanized if its text body contains one or more of these common keywords. The saturation by keywords in the text bodies and titles ranges from one (zero for the ones not Europeanized) to more than 500 hits for the abbreviations ‘CEE’ and ‘CE’. A closer analysis of keywords found in domestic legislation reveals their uneven occurrence: whereas about half of these were not at all found in the titles and text bodies of laws, four keywords made for more than 80% of the occurrences in the 539 laws containing keywords. The keywords most often found in the Luxembourgish laws published between 1986 and 2006 were ‘CEE’, ‘Union européenne’, ‘CE’ and ‘Communauté européenne’. As we know from the context of this comparative research project, Europeanization operationalized in this way is of course a purely mechanical measure. For the case of Luxembourg, looking only at published laws is another limitation. As the sources and motives for legislating do not appear in the published texts, analyzing laws only further underestimates the influence of European templates on national legislation. For instance, the law establishing a Constitutional Court in Luxembourg in July 1997 does not contain any of the keywords used in our content analysis, whereas the bill proposals introduced by MPs clearly stated in their motives that the mechanism for the reference to the Court for a preliminary ruling was
190
inspired by the one adopted by the European Court of Justice (Gerkrath 2008: 4). Another example often cited by Luxembourg political personnel in relation to the Europeanization of legislation and the transposition of directives concerns the financial sector: contrary to those European legislations that do not touch upon national interest, directives in this sector were transposed and implemented earlier than European competitors (the 1988 law transposing the 85/611/CEE directive on Undertakings for Collective Investment in Transferable Securities) or were even anticipated by Luxembourgish domestic legislation (the 1983 law on Undertakings for Collective Investment, i.e. investment funds). In the latter case, reference to the preparatory work around the directive was made in the motives of the bill project, advice of the Council of State and of other actors, but does not show in the adopted law since the European legislation did not yet exist.
2.2.2. Results Table 2 gives the proportion of all laws published per year that contained at least one of our keywords. There is a clear trend towards an increased Europeanization of laws over the 21 year period, from 11.8% in 1986 to 49.15% in 2006, cumulating to an overall proportion of 28.8% of all laws adopted throughout the period. As a longitudinal comparison of Europeanization rates between laws still in force and abrogated laws reveals no bias, we focus in the next analyses on these laws that were still in force in 2009 (85% of all laws adopted in the 1986-2006 period). TABLE 2 ABOUT HERE
The policy areas (as defined by the Comparative Agendas Project) most Europeanized in Luxembourg are Banking and Finance, Civil rights and International affairs, where more than 50% of laws contain one or more of our keywords (see Table 3). Given the importance of the banking and finance sector for Luxembourg in particular, international affairs for small and open countries in general, and also referring to the Europeanization of 40% of macroeconomics legislation (all of these three policy sectors are in the top five of most intensively legislated sectors) one may conclude that the influence of Europe on the main and
191
arguably most crucial legislation for the national economy is far from benign, even though one falls clearly short of Jacques Delors’ prediction. The areas left out by this influence concern the areas Public lands/Water management, Immigration, Housing and Defence.
TABLE 3 ABOUT HERE An overwhelming share of Europeanized laws, 89.9%, was initiated by the government (Table 4). In total almost 30% of all laws initiated by government over the period were Europeanized, compared to 13% of those proposed by MPs or a parliamentary committee. This difference reflects the European burden put on government legislative initiatives, whereas MPs are freer to adapt to the growing influence of Europe on policy and introduce bills in policy sectors less heavily affected by upper-level decisions. TABLE 4 ABOUT HERE
Table 5 compares our “Europeanized” laws with laws adopted in order to transpose one or several EU directives. It shows that 23.7% of laws that do not transpose directives nevertheless refer to our chosen keywords, and are thereby influenced in some way by EU decisions. On the other hand, 87% of the 141 laws voted on and transposing European directives in national legal order contain such a reference. It is not surprising that this percentage grows as the number of directives transposed by the domestic instrument increases.
TABLE 5 AROUND HERE
We also found 19 laws transposing one or more directives, which do not contain any of our keywords indicating Europeanization. This could be due to two main reasons: First, the national legislator deliberately chose not to refer to the directive in the adopted text (but most likely refers to it in preparatory work), either because such mention in the articles of the law
192
was not deemed necessary (for instance when no article has to refer to an article of the directive), or for more political motives such as claiming credit for some measures that were actually required by EU legislation (it is no surprise to find a concentration of laws concerning welfare state benefits or modifications of laws regarding taxes taken in the high economic growth years – thereby increasing benefits and lowering taxes – in the list of laws transposing directives without mentioning our keywords, as well as environment protection regulations). Second, for a few cases, the laws transposing EU directives were taken before the latter were issued (in some cases the preparation of the European text may be referred to in the motives of the bill project). In order to analyze the potential effect of Europeanization on the initiation and adoption of amendments, we drew a sample of laws produced within five years (for each parliamentary term: we took the second calendar year after elections and the proportion of Europeanized laws in these years ranged from the second lowest to the highest levels recorded in the period, i.e. 11% to 48%). Table 6 describes the initiator of the 131 initiated amendments for these five years and reveals that parliamentary committee amendments are the most frequent and have the highest rate of adoption (almost 98%), closely followed by government amendments. The amendments initiated by MPs in plenary sessions come third, and their rate of adoption is almost negligible (7%). On the 383 laws under scrutiny, about 32% were amended, and as Table 7 shows there was no discernable temporal pattern with regard to the initiation or adoption rate of these amendments. In line with results for other countries in this comparative project, we find, as reported in Table 8, that compared to non-Europeanized laws, Europeanized laws are significantly more frequently amended, with a rate of 27% for the former for 42% for the latter. Moreover, all five years analyzed display consistently higher rates of amendments for Europeanized laws, with a rate twice as high for the year 1991. Even when we restrict our analysis to the Europeanized legislation that is actually transposing a European directive, the picture stays the same: laws transposing directives are more often amended than others. Even though we rely on a sample of laws, this appears to confirm that MPs tend to invest more efforts in the control of Europeanized laws in order to compensate for their lack of involvement in EU policy and the information asymmetry with government. At least, these findings indicate an intensified scrutiny of legislation stemming from the European level: The main difference in the adoption rates of amendments per initiator shows that Europeanized laws are more frequently amended by parliament at committee stage than
193
non-Europeanized ones26. Again, European legislation seems to constrain the government more than the parliament in Luxembourg.27 We can link these quantitative findings to those of our qualitative investigations. Ben Fayot, chairman of the Foreign Affairs Committee of the Chamber, stated in an interview that the degree of implementing discretion for national authorities awarded by EU directives has an impact on the likelihood of parliamentary involvement in the transposition process. Largely in line with the theoretical claim made by Franchino and Hoyland (2009), this practitioner observes political discussions in the Chamber being more intense when discussing EUinspired laws if the latter falls within the shared competencies (for instance the third pillar) of the EU rather than exclusive ones. More generally, when the government is awarded more discretion for implementing EU law, it often seizes this opportunity to maximize its room to maneuver by taking some liberties in transposing the European text. As a result, parliamentary scrutiny becomes more intense in reaction to this attempt by the executive to use its informational advantages to make the Chamber adopt policies closer to its own preferences. To some extent Table 9 confirms this expectation: We see within the Europeanised laws that sectors such as International Affairs and Environment display a lower rate of amendments than Law and Crime. Overall, we also note weak but positive correlations suggesting that the Chamber scrutinizes EU-inspired laws mostly when these belong to the most legislated (total number of laws in the sample years) and Europeanized sectors.
3.
Conclusions
On the one hand, the Europeanization of domestic legislation in Luxembourg remains modest. Less than 10% of the laws published in the two last decades have transposed EU directives. The more general view of EU impulse taken by this comparative project leads us to consider that almost 30% of laws voted on in the Chamber of Deputies are Europeanized, as defined by the occurrence of at least one keyword found in the body of their text. On the other hand, this Europeanization has clearly increased over time, reaching an average of almost 40% of laws since 2001. Furthermore, some sectors of vital importance for Luxembourg, in particular for 26
The differences in rates of government amendments and of plenary amendments are smaller. Another proxy for the level of contestation of different laws is the duration of the legislative process. Europeanized and non-Europeanized laws however differ only slightly (the latter taking a bit more time) with regard to the average duration of their production, and this difference is not significant. 27
194
its economy which is in large part a guarantee for its sovereignty, are obviously influenced by decisions made by European Union institutions. Overall however, in the case of the Grand Duchy, we do still fall short of Delors’ prediction. Yet other forms of Europeanization could well be analyzed in the context of this project, aside from completion of data collection and coding of laws, of bills (allowing for the analysis of the motives for legislating) and perhaps other parliamentary activities such as questions and interpellations. For instance, from 2001 to 2008, the Council of State issued 635 formal oppositions to bill projects, proposals and executive decrees. These oppositions were motivated mainly by reference to the Constitution, by other general legal principles but also sometimes by international law, which for the most part refers exclusively to EU legislation. Hence, the Council of State may have influenced domestic legislation at an early stage of the process by referring to EU legislation, causing revisions of the bills or delaying their adoption. Studying the impact of the Council of State certainly marks an area for future research. We further saw in this study that many aspects of the management of EU policy have been reformed in recent years in Luxembourg, not the least in order to cope with the Grand Duchy’s transposition deficit. These reforms mainly concerned the organization of government and the provision of information and cooperation to the Chamber. Combined with the finding that government bills are more Europeanized than laws initiated by the Chamber, and that MPs appear to exert more control through amendments in Europeanized matters, one can safely point to European integration as a growing constraint for the national government.
195
References
Bergman, Torbjörn (1997). ‘National Parliaments and EU Affairs Committees: Notes on Empirical Variation and Competing Explanations.’ Journal of European Public Policy 4: 373–387. Bergman, Torbjörn (2000). ‘The European Union as the Next Step of Delegation and Accountability.’ European Journal of Political Research 37: 415–429. Bossaert, Danielle (2003). ‘Luxembourg: flexible and pragmatic adaptation’, in Wessels, Wolfgang, Maurer, Andrea, and Mittag, Jürgen (eds). Fifteen Into One? The European Union and Its Member States. Manchester: Manchester University Press. Bossaert, Danielle and Sophie Vanhoonacker (2000). ‘Relaunch of the Benelux?’, in Pijpers, Alfred (ed.). On Cores and Coalitions in the European Union. The Position of some Smaller Member States. The Hague: Clingendael Institute, 157-170. Conseil d’Etat (ed.) (2006). Le Conseil d’Etat, gardien de la Constitution et des Droits et Libertés fondamentaux. Luxembourg: Conseil d’Etat. Dumont, Patrick and Lieven De Winter (2003). ‘Luxembourg: A Case of More “Direct” Delegation and Accountability’, in Kaare Strøm, Wolfgang C. Müller and Torbjörn Bergman (eds) Delegation and Accountability in Parliamentary Democracies. Oxford: Oxford University Press, 474-497. Dumont, Patrick (2005). ‘Luxembourg and the upstream and downstream relations to the European Union: Selectivity, stability, pragmatism and flexibility as responses to administrative shortage’ paper presented at the international conference Progress & Problems in Building a European Legal Order: Europeanization through the Transposition and Implementation of EU Directives, Luxembourg Chamber of Deputies, June 24-25. Falkner, Gerda and Simone Leiber (2003). ‘A europeanization of governance patterns in smaller European democracies?’. Paper for the Eighth Biennial International Conference, European Union Studies Association, USA, 27-29 March 2003, Nashville, Tennessee.
196
Falkner,
Gerda
(2003).
‘Comparing
europeanisation
effects:
from
metaphor
to
operationalisation’. Paper for the ECPR General Conference, 18-21 September, Marburg. Franchino, Fabio and Bjørn Høyland (2009). ‘Legislative Involvement in Parliamentary Systems: Opportunities, Conflict, and Institutional Constraints’. American Political Science Review, vol. 103, 4: 607-621. Gerkrath, Jörg (2008). ‘Compétence er recevabilité des questions préjudicielles’. In Gerkrath, Jörg (ed.). La jurisprudence de la Cour Constitutionnelle du Luxembourg, 1997-2007. Luxembourg: Pasicrisie Luxembourgeoise. Hoscheit, Jean-Marc and Malou Weirich (1988). ‘Luxembourg’, in Siedentopf, Heinrich and Ziller, Jacques (eds.). Making European Policies Work. The Implementation of Community Legislation in the Member States. 2 volumes. London: Sage. Maurer, Andreas and Wessels, Wolfgang (eds.) (2001). National Parliaments on Their Ways to Europe: Losers or Latecomers? Baden-Baden: Nomos. Raunio, Tapio and Matti Wiberg (2000). ‘Parliaments’ Adaption to the European Union’, in Peter Esaiasson and Knut Heidar (eds.), Beyond Congress and Westminster: The Nordic Experience. Columbus: Ohio State University Press. Schmit, Nicolas (2006). ‘Le Conseil d’Etat face à l’intégration économique et politique du Luxembourg sur la scène européenne’. In Conseil d’Etat (ed.). Le Conseil d’Etat face à l’évolution de la société luxembourgeoise. Luxembourg: Conseil d’Etat, 553-572. Thorhallson, Baldur (2000). ‘The administrative working procedures of smaller states in the decision-making process of the EU’. Paper for the Political Studies Association-UK 50th Annual Conference, 10-13 April 2000, London. Wessels, Wolfgang, Maurer, Andrea, and Mittag, Jürgen (2003). Fifteen Into One? The European Union and Its Member States. Manchester: Manchester University Press.
197
Graph 1: Number of bill projects and bill proposals, 1984-2009
Number of Bill Projects and Bill Proposals 1984-2009 140
120
100
80
Bill projects introduced Bill projects adopted Bill proposals introduced Bill proposals adopted
60
40
20
/0 9
/0 8
08
/0 7
07
/0 6
06
/0 5
05
/0 4
04
/0 3
03
/0 2
02
/0 1
01
/0 0
00
/9 9
99
/9 8
98
/9 7
97
/9 6
96
/9 5
95
/9 4
94
/9 3
93
/9 2
92
/9 1
91
/9 0
90
/8 9
89
/8 8
88
/8 7
87
/8 6
86
85
84
/8 5
0
Source: Chamber activities annual reports Note: We here only reproduce the activities during ordinary (full) sessions of the Chamber, and not the extra-ordinary sessions that take place for a limited amount of time between the first gathering of the Chamber after elections and the official start of the parliamentary year.
Graph 2: Adoption rates per legislative mandate
198
Adoption rates per legislative mandate 100 90 80 70 60
Projects
50
Proposals 40 30 20 10 0
1984-1989
1989-1994
1994-1999
1999-2004
2004-2009
Source: Chamber activities annual reports
Table 1: Europeanised laws in different databases 1986-2006
Nr of NonEuropeanized
Nr of Europeanized laws
Laws
Total
%
Government database www.legilux.lu
1,683
281
1,964
14.31
Chamber database www.chd.lu
1,450
514
1,964
26.17
both
1,399
565
1,964
28.77
Source: Legitech dataset, Chamber archive, own coding and calculations Correlation: Phi=0.584 (Sig. = 0.0)
Table 2: Number and proportion of Europeanized laws by year, 1986-2006 (Total of laws compared to laws still in force)
199
ALL LAWS
LAWS STILL IN FORCE
Europeanized Year
Europeanized
laws
Total
%
laws
1986
10
85
11.76
7
62
11.29
1987
7
94
7.45
3
65
4.62
1988
12
84
14.29
8
62
12.90
1989
16
94
17.02
13
68
19.12
1990
13
73
17.81
7
47
14.89
1991
16
92
17.39
10
62
16.13
1992
27
79
34.18
21
61
34.43
1993
22
93
23.66
20
82
24.39
1994
20
78
25.64
17
68
25.00
1995
24
66
36.36
20
55
36.36
1996
16
66
24.24
15
62
24.19
1997
38
93
40.86
31
82
37.80
1998
26
97
26.80
24
86
27.91
1999
27
103
26.21
26
99
26.26
2000
25
77
32.47
21
67
31.34
2001
49
97
50.52
38
82
46.34
2002
32
104
30.77
28
92
30.43
2003
35
131
26.72
32
125
25.60
2004
52
126
41.27
50
120
41.67
2005
40
114
35.09
39
111
35.14
2006
58
118
49.15
55
115
47.83
566
1964
28.80
485
1673
28.99
Total
Total
%
Source: Legitech dataset, Chamber archive, own coding and calculations
200
Table 3: Europeanized laws by policy area 1986-2006 (laws still in force)
Nr of NonEuropeanized Topic
Laws
Nr of Europeanized laws
Total
%
Macroeconomics
80
52
132
39.39
Civil rights
12
13
25
52.00
Health
38
15
53
28.30
Agriculture
14
9
23
39.13
Labour
31
20
51
39.22
Education
102
12
114
10.53
Environment
71
26
97
26.80
Energy
4
3
7
42.86
Immigration
91
4
95
4.21
Transportation
73
39
112
34.82
Law & Crime
98
35
133
26.32
Social welfare
43
21
64
32.81
Housing
96
8
104
7.69
Banking & Finance
55
76
131
58.02
Defence
21
2
23
8.70
Technology
13
13
26
50.00
Foreign trade
8
3
11
27.27
International affairs
106
109
215
50.70
Government operations
196
24
220
10.91
Public lands/Water management
36
1
37
2.70
1188
485
1673
28.99
Total
Source: Legitech dataset, Chamber archive, own coding and calculations
Table 4: Europeanized laws by initiator 1986-2006 (laws still in force)
201
Nr of Europeanized
Nr of NonEuropeanized Initiator
Laws
Laws
Total
%
Government
1160
479
1639
29.23
Parliament
20
3
23
13.04
Both
8
3
11
27.27
Total
1188
485
1673
28.99
Source: Legitech dataset, Chamber archive, own coding and calculations
202
Table 5: Laws transposing a directive 1986-2006 (laws still in force)
Nr of Europeanized
Nr of NonEuropeanized N° of directives transposed
laws
Laws
Total
%
0
1169
363
1532
23.69
1
16
68
84
80.95
2
2
28
30
93.33
3 and more
1
26
27
96.30
Total
1188
485
1673
28.99
Source: Legitech dataset, Chamber archive, own coding and calculations Table 6: Adoption rate of amendments by initiator (sample of laws still in force)
Initiated
Adopted
%
Government amendments
61
59 96.7%
Committee amendments
85
83 97.6%
Amendments in plenary
42
Total
131
3
7.1%
121 92.4%
Source: Legitech dataset, Chamber archive, own coding and calculations Table 7: Amendment rate of laws per year (sample of laws still in force)
Year
Not amended
Initiated
Amended
Total
%
1986
46
20
16
62 25.8%
1991
41
21
21
62 33.9%
1996
44
22
18
62 29.0%
2001
50
33
32
82 39.0%
2006
81
35
34
115 29.6%
203
Total
262
131
121
383 31.6%
Source: Legitech dataset, Chamber archive, own coding and calculations Table 8: Amendment rate among Europeanized laws per year 1986-2006 (sample of laws still in force) Non-Europeanized
Not amended
Year
Europeanized
Amended
%
Not amended
Amended
%
1986
42
13
23.6%
4
3
42.9%
1991
37
15
28.8%
4
6
60.0%
1996
34
13
27.7%
10
5
33.3%
2001
31
13
29.5%
19
19
50.0%
2006
45
15
25.0%
36
19
34.5%
Total
189
69
26.7%
73
52
41.6%
Source: Legitech dataset, Chamber archive, own coding and calculations
Table 9: Amendment rate among Europeanized laws per policy area1986-2006 (sample of laws still in force) Non-Europeanized
Not amended
Year
Amended
Europeanized
%
Not amended
Amended
%
Macroeconomics
17
6
26.1%
3
7
70.0%
Civil rights
3
1
25.0%
2
0
0%
Health
5
2
28.6%
0
2
100.0%
Agriculture
0
2
100.0%
0
2
100.0%
Labour
6
3
33.3%
2
6
75.0%
Education
13
7
35.0%
3
0
0%
204
Environment
5
3
37.5%
9
2
18.2%
Energy
1
0
0%
0
0
-
Immigration
25
2
7.4%
0
0
-
Transportation
12
4
25.0%
9
1
10.0%
Law & Crime
11
9
45.0%
5
7
58.3%
Social welfare
4
5
55.6%
3
0
0%
Housing
15
3
16.7%
0
1
100.0%
Banking & Finance
4
4
50.0%
8
10
55.6%
Defence
3
0
0%
0
0
-
Technology
2
1
33.3%
1
4
80.0%
Foreign trade
2
0
0%
1
0
0.0%
International affairs
21
0
0%
26
6
18.8%
Government operations
26
16
38.1%
1
4
80.0%
Public lands/Water managem.
14
1
6.7%
0
0
-
Total
189
69
26.7%
73
52
41.6%
Source: Legitech dataset, Chamber archive, own coding and calculations
205
9. Myths and Milestones: The Europeanization of the Legislative Agenda in The Netherlands Gerard Breeman (Wageningen University Research Centre) & Arco Timmermans (Montesquieu Institute and Leiden University)
Introduction: Dutch Sensitivity to European Mythology In July 2010, the Chamber of Representatives in The Netherlands, the Tweede Kamer, used a new parliamentary tool to constrain the government’s mandate on European policy proposals. This tool enables the Chamber to list issues for discussion with ministers in the national parliament before they represent the country in European policy making.1 This first case of early parliamentary detection and involvement in EU policy making followed after several years of stalemate and silence over European issues on the national political agenda. The political legacy of Europeanization in the Netherlands is a particularly strong case of what Majone (2005) calls ‘integration by stealth’. The Netherlands are a founding member and have a long standing reputation as a loyal European partner in matters of high and low politics. Commissioners from the Netherlands were found to be among the most dedicated to the European case2. At home, critique on Europeanization was rare and debate was mostly absent. The economic benefits from European integration spilled over to what seemed to be a broad political and public acceptance. Yet, the national referendum in June 2005 in which nearly two thirds of the voters rejected the European Constitutional Treaty brought euroskepticism and even outspoken antiEuropean sentiment at the surface. This referendum was a shock event to most of the established parties, as they did not know how to respond except by avoidance. The 1
Such a list was made in April 2010 (Tweede Kamer 2009-2010, 22112, no. 1012). For example, in the 1960s Sicco Mansholt designed the important Common Agricultural Policy, in the 1990s Frits Bolkestein further opened the internal market, and recently Neelie Kroes as commissioner for competition policy vigorously applied sanctions against companies violating the European rules. 2
206
referendum campaign organized by the government had demonstrated a lack of willingness to engage in political debate. This fuelled strong reactions from parties that did mobilize on the theme of Europeanization – but mostly with a negative tone. While parties in government issues averted attention to the EU and even dropped their support for national referendums as a mechanism of citizen participation, opposition parties became more profiled in challenging the Dutch position in the European Union. The success of the Freedom Party (PVV) led by Geert Wilders in the parliamentary elections of June 2010 came with populist claims to resist ‘multicultural Eurabia’, reshape the EU back into a European Economic Community, and abolish the European Parliament. These developments widened the gap between two quite distinct worlds of attention to Europeanization in the Netherlands. One is the sphere of mostly technically defined issues addressed in closed communities of experts and policy makers, operating at a low tide of public interest: the domain of ‘policy without politics’ (Schmidt 2005). The other world is in Schmidt’s characterization the opposite, ‘politics without policy’, in which mostly critics of Europeanization exploit venues of attention for diffusing dramatic images. It is the sphere of what Van Middelaar (2009) calls ‘the street versus the state’. The existence of these two different spheres of attention to Europe made the Dutch particularly sensitive to the Myth of Delors on the extent of Europeanization of national policy making In recent years, scholars made first attempts at assessing the actual degree of European regulatory influence. The results are a beginning of demystification, although the relevant studies are all limited in scope. In a quick scan of a few policy fields, De Jong and Herweijer (2004) found considerable variation. This finding of variation also appeared in a more systematic comparison by Douma et al (2007) of regulation in the fields of education and environment. These authors concluded that on 1 January 2005, 6 percent of all national regulation on education and 66 percent of environmental regulation stemmed directly from European legislation and jurisdiction. Bovens and Yesilkagit (2005; 2010) considered the effects of European directives on formal legislation and delegated regulation in force on 31 July 2003. Their main finding is that on average, only about 16 percent of national regulation (laws, orders in council and ministerial decrees) can be traced back to European directives. The authors also mention that 12.6 percent of formal Dutch legislation in force on 1 July 2003 contained the transposition of European directives (2010: 61). Most of this transposition
207
happens via delegated regulation in which the national parliament is not involved. Finally, Hogenbirk and Princen (2010) report first findings on the relative importance of European regulation compared to national regulation, and conclude that European regulation is not more (nor less) important and complex than national regulation. Finally, the official website of the chamber of representatives, the Tweede Kamer, mentions that ‘more than half of all new laws is the consequence of European policy’.3 In this contribution we systematically analyze the presence or absence of EU links in national legislation during the period from 1981 to 2009. For this analysis we constructed a dataset containing nearly 5,000 bills submitted over the thirty year period. We begin this chapter with a brief description of the legislative process in The Netherlands and what institutional arrangements exist for the Europeanization of this process. Most visibly, such arrangements relate to the transposition of European directives and the institutional opportunities for parliamentary scrutiny. Next, we present our data and analyze legislative productivity with and without EU links, and consider different main topics that are more or less Europeanized. A following section deals more specifically with the extent to which parliamentary parties used their formal tool to amend bills. Finally, we discuss our findings in the context of the existence of the two worlds of Europeanization in Dutch politics and society.
Law Making in The Netherlands The Dutch parliamentary system has a bicameral parliament, in which the Chamber of Representatives, the Tweede Kamer, is directly elected. The presence of multiple political parties from left to right in this lower house is facilitated by the electoral system of proportional representation. Size differences between traditional parties and new, increasingly radical parties have diminished, thus making it harder to make and maintain government coalitions. Producing legislation in the Netherlands is based on a coalition agenda set in government formation. In considering bills, the Tweede Kamer has more powers than the Eerste Kamer, the Senate. Only the Tweede Kamer can introduce amendments to bills, the 3
See: http://www.tweedekamer.nl/hoe_werkt_het/tweede_kamer_en_europa/index.jsp.
208
Senate must vote on bills after they have (or have not) been amended in the lower house. The use of amendments has increased over time, but the number varies from one year to the next, with a tendency for amendments to appear more often towards the end of a governmental term in office. The Dutch government uses its informational advantage and administrative machinery to absorb almost all preparatory activity of legislation; it initiates some 95 percent of bills. As governments are always coalititons, institutional mechanisms evolved for managing interparty conflict and streamlining policy making. These mechanisms include coalition agreements, arenas for managing dispute, and informal rules of discipline. Matters on which members of the coalition parties in parliament are not expected to be disciplined and are allowed more leeway are explicitly mentioned in the coalition agreement. But such matters are few. Parliamentary parties taking stock against government bills are most frequently members of the opposition, and if they manage to form ad hoc legislative coalitions it is on issues of lower importance to the the government (Timmermans and Andeweg 2000: 373; Timmermans 2011). Influence via amendments is more frequent, on average about one third finds its way into the final bill text. Most of these amendments affect some points, not the main thrust of the initial bill. While bills usually take about a year to reach the stage of approval in the Tweede Kamer, and some three months in the Senate, controversial bills can take up to four years before they are adopted. Bills that fail are sometimes stuck in the machinery for many years before they are rejected or withdrawn. The European dimension of national law making has become institutionalized, but in political terms it is still mostly low key. Government departments have their EU detection and policy preparation units, and European matters reach the formal agenda of the Council of Ministers if these are considered key points of government policy. This usually does not happen unless some event triggers special attention. In parliament, matters with a link to the European Union used to be considered primarily in the European Affairs Committee, but recently all permanent committees of the legislative chambers have started to consider the European dimension of national legislation. European proposals also are reviewed increasingly on subsidiarity criteria, and a new element of parliamentary agenda setting on European matters is that European policy issues can be listed for discussion with responsible ministers before they represent the Dutch government in European policy making. This parliamentary scrutiny reservation (behandelvoorbehoud) has a signaling role. Plenary
209
debates on such matters are quite rare, however. Further, the standard rule of tacit ratification of European and international treaties may be exempted if at least one fifth of the members of either of the legislative chambers expresses the intention to scrutinize it.
Research Focus and Dataset
There are different types of bills, varying for institutional and other reasons, and the inclusion or exclusion of these types of bills in the present analysis has consequences for findings on the extent of Europeanization. Following the common approach to the empirical investigation of European influence in national law production presented in the introductory chapter, we acknowledge the possible differences in kind, size, importance etc. between pieces of legislation. Such variation exists in the Netherlands as in other countries. Thus our dataset includes not only politically debated bills, but also bills with modest or just technical changes, bills that impacted social behavior as well as those that turned out to be merely symbolic. But it is reasonable to assume that such variation existed throughout the entire period, so that analysis of developments in Europeanization of legislation is left relatively unaffected by it. Some types of bills are more likely to contain references to European institutions or policy than others. We exclude those types of bills that are part of legislative packages for formal reasons, and these types have a low institutional likelihood of EU links. Excluded are bills containing naturalizations of individuals and bills formally presenting annual departmental budgets. This last type of bill comprises about one third of the overall volume of legislation in the Netherlands. Budget bills do not vary much in number from year to year, they are a mandatory and institutionally fixed element in the policy process. Bills included in the analysis are ordinary bills, Constitutional revision bills, and bills ratifying international and European treaties. As said, in The Netherlands, treaty ratification bills appear only when parliament explicitly demands a scrutiny procedure. We included, for these types, both government bills and legislative proposals made by members of parliament (private member’s bills).
210
In the following empirical approach and analysis of European links in national legislation, we distinguish between two types of EU influence. One is the category of bills mandated directly by EU policy, often directives and regulations, European treaty obligations commitments, and European Court rulings. In this sense these bills follow from European ‘constraints’; they are instances of European policy implementation by law. Many of these are instances of transposition. In this contribution, we do not consider negative agenda setting, that is, cases in which the government or parliament refrains from initiating bills because of European constraints or more indirect sources of European influence. Power and influence, of course, do include such an element of negative agenda setting. The second category of EUlinked bills is those containing links to institutions or policies of the EU but not based directly on obligations of implementation. Instead they contain phrases such as the ‘desirability’ of new or changed legislation given European policy. These EU-linked bills, thus, are more voluntary, and they can be anticipatory. As we will show in our findings, the vast majority of EU-linked bills in the Netherlands is of the mandatory type; only a small part involve bills produced voluntarily. The third category of bills includes those for which no links to EU institutions or policy were found. The data collected for this analysis are part of the Netherlands Policy Agendas Project, mapping attention to problems across all policy fields in different institutions over a long period. For this we use a Dutch version of an international coding scheme (Breeman and Timmermans 2008). As in the policy agendas project we study the legislative agenda, our units of analysis are bills submitted in parliament. The vast majority of such bills is formally approved. The reason why we focus on bills is that, as policy items framed for presentation on the legislative agenda in parliament, they attract more (or less) attention and do (or do not) provoke formal amendments in the Tweede Kamer. The government almost always initiates of the framing and presentation of these bills, and these are the best proxy to the legislative agenda in the institutional interplay between the government and parliament when studying influences of Europeanization. Single pieces of legislation and bills can vary enormously in size. The legislative agenda may be limited institutionally by the number of cabinet ministers, the size of legislative chambers, the number of parliamentary committees, time tables for legislative scrutiny and voting, etc. But individual bill texts have fewer length constraints. We agree with
211
analysts who argue that bill text length can be an indicator of importance, and we also acknowledge that longer texts as data sources increase the likelihood of hits in EU keywords searches. In this project, however, we limit our unit of analysis to titles and preambles of bills, and thus do not consider full texts or underlying policy rationales of bills. Such an extended content analysis project is beyond the scope of the present study. We hold our unit of analysis constant for the full period on which empirical data are collected, so there is no variation in types of data sources over time to avoid biased results. We also kept track of the number of key word hits per bill, for an indication of the magnitude of EU-links in bills over time and across different policy fields. We use the set of key words mentioned in the introductory chapter with some additional key words; these are synonyms or extensions to include references to specific European treaties. The extended list of key words for this analysis is given in Appendix 1. The dataset contains 4,677 content coded bills for the period 1981-2009.4 Specific dates of bill submission could be extracted from the data source only for the period after 1995, for earlier years only the parliamentary session years could be extracted. A parliamentary year does not follow the normal calendar from January to December, and in order to be able to analyze for normal calendar years, we standardized our data and set all bills proposed during a parliamentary session on 1 January of the year of submission.
Patterns of Legislative Activity We begin the empirical analysis with some general patterns of legislative production, considering the annual volume of bills, the proportion of types of bills, allocation of legislative attention to policy topics, and the limited subset of private member’s bills. This analysis is made for the set of 4,677 bills produced between 1981 and 2009. Figure 1 shows the annual numbers of bills included in the analysis. Figure 1 here 4
The total volume of bills in this period was 7,257 bills (of which 2,580 budget bills are excluded from the analysis).
212
Figure 1 Bill production 1981-2009 (excluding budget bills) 250
200
150
100
50
Center-Right Governments: CDA-VVD
Governments with Social Democrats (PvdA)
CenterRight Govt
Gvt w. P vdA
0 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
The number of bills submitted in parliament varies between some 100 and over 200 per year, with an average increase over time, particularly in the 1990s. Earlier years in the studied period reveal shorter intervals of activity and more frequently dropped bills (such drops were not due to shorter government duration, which influences bill production cycles). Typically, legislative productivity increases towards the end of the legal term in office of governments, and years of transition from one government to the next show low numbers. While these findings may provide some evidence for the idea of political business cycles5, , it must be realized that this analysis is based on bills which may be delayed during the process of parliamentary scrutiny even after the elections are held. The most straightforward interpretation of the cyclic pattern is that governments in office need some preparation and perhaps political incubation time for getting all their legislative projects on track. This finding corresponds to comparative work on a shorter time period, in which governments in The Netherlands were found to submit bills later during their term in office than governments in some other European countries ( Martin 2004). Different institutional types of bills were produced in quite different numbers; the vast majority were ordinary bills (4,428 cases), followed at large distance by bills for treaty 5
The theory of political business cycles states that parties in office strategically aim at producing most legislative output towards the end of the term, in order to strenghten their electoral position.
213
ratification (206 cases) and attempts at Constitutional revision (33 cases). Private member’s bills can be of either of these three types (they are hardly of any of the nonincluded types), but their number is quite modest: in the thirty year period, 208 were submitted. These numbers of bills per type vary with the institutional thresholds for submission and parliamentary approval. The rate of parliamentary approval of bills is, in most years, around 90 percent. Private member’s bills are less successful, they reach the stage of adoption in only 50 percent of the cases. Often, such initiatives are made by opposition parties (Timmermans et al 2008). In all legislative activity, the party composition of government coalitions varies only to some extent with the volume of bills produced. Governments including left wing parties with a programmatic and state steering orientation could be expected to set a larger legislative agenda than governments with a center-right profile. This is visible to some extent in 1990s compared to the 1980s (see figure 1), but when in 2002 a center-right coalition was formed, it did not reduce the volume of bills produced. This suggests that, apart from the presence of cycles within governmental terms in office, longer trends in legislative production are only in part attributable to the party profiles of governments. One reason for the (modest) incline in the volume of bills in the analyzed period may be legislative ‘downloading’ of European policy. This will be a central point of analysis further below. Finally, we mapped the substantive topics of all content coded bills. Figure 2 presents the ranking of bill topics, for the entire period. Figure 3 shows the relative distribution of attention to topics in bills over time.
Figure 2 here
214
Figure 2: Bills per topic, absolute numbers 1981-2009 12. Justice & Crime 20. Public Administration 6. Education & Culture 15. Commerce 13. Social Security 1: Macroeconomy 5. labor 10. Transport 2. Rights, Immigration, Integration 7. Environment 3. Health 19. Foreign Policy 14. Housing 4. Agriculture & Fisheries 16. Defense 17. Science & Technology 21. Planning 8. Energy 18. Foreign Trade 0
100
200
300
400
500
600
As figure 2 shows, justice and crime issues were the most salient on the legislative agenda between 1981 and 2009. Foreign trade attracted least attention. The organization of government and relations to subnational political institutions also took a considerable share of legislative interest. Domestic regulatory issues of education, commerce and business, and social security also lead the legislative agenda. The policy topics resulting in a smaller volume of bills may be less typically matters of regulatory policy, but they also may involve delegated regulation such as ministerial decrees which are not included in the present analysis.6
Figure 3 here
6
Private member’s bills involve a somewhat different allocation of attention. Labor issues are most salient, followed by social security, and public administration and the organization of government.
215
Figure 3: Proportional attention to topics on legislative agenda, 1981-2009 100%
21. Planning 20. Public Administration 19. Foreign Policy
90%
18. Foreign Trade 17. Science & Technology
80%
16. Defense 70%
15. Commerce 14. Housing
60%
13. Social Security 12. Justice & Crime
50%
10. Transport 40%
8. Energy 7. Environment
30%
6. Education & Culture
20%
5. labor
10%
3. Health
4. Agriculture & Fisheries 2. Rights, Immigration, Integration
0%
1: Macroeconomy
81 82 83 84 85 86
877 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
Figure 3 gives a more dynamic picture of the relative attention to policy problems in bills. It shows the ebb and flow in relative salience of issues on the legislative agenda. A number of main topics such as health, rights and immigration and spatial planning and water management have increased over time, and some have become less visible in the production of bills, such as macro-economics, defense issues, and education and culture. Topics also come and go on the legislative agenda, but this happened only with matters that receive, on average, quite small portions of attention (foreign trade, energy, agriculture). The overall pattern is that changes in the allocation of attention are modest in size. The legislative agenda, within capacity bounds, appears relatively stable over the thirty years period. This may indicate that both capacity and other institutional thresholds for change play a part in legislative agenda setting. Such thresholds may also hamper the impact of party composition of governments on legislative agenda setting. We see that the volume of bills shows more of a general increase than variation with government types, and intra-governmental production cycles exist. Shifts in attention to topics occur more often during governmental terms in office than at their beginning. This may indicate that if legislative attention shifts are programmed, they first need to overcome friction in the policy making machinery. We also found that bill production increases during terms in office, and this may indicate that there is a connection to topic change and agenda capacity.
216
European Influence on the Legislative Agenda
How many of these bills in the different domains of public policy carry European impulses as measured by our indicators? Is the Europeanization of legislation in the Netherlands as strong as Jacques Delors predicted in 1988, is it anywhere near the 50 percent mark mentioned at the public information website of the Tweede Kamer, or rather at mostly lower levels found in recent work on this question? We emphasize that our measurement does not include lower and delegated regulation, while European directives, for example, may be implemented by such lower regulation and not only by formal laws. Moreover, in the Netherlands, the ratification of European treaties happens by legislation only if parliament demands such an explicit procedure. Analyzing the set of 4,667 bills, we found 572 bills with one or more EU links, this is 12.3 percent. Table 1 shows some variation between bill types analyzed.
Table 1 here Type of legislation
No EU link
EU link
Total
Ordinary
3901
527 (11.9 %)
4428
Treaty ratification
162
44 (21.4 %)
206
Constitutional
32
1 (3.0 %)
33
Total
4095
572 (12.3 %)
4667
As we saw, in the entire package of bills produced over 30 years, ordinary legislation takes most legislative agenda space. On average, 11.9 percent of the ordinary bills carries one or more references to European institutions or policy. Contrary to the findings for some other countries such as France, constitutional amendments in The Netherlands are rarely linked to the EU; just 3.0 percent (one out of 33) is. While bills ratifying international treaties are not
217
numerous, they are more often EU-linked than any other type of bills: in 21.4 percent of the cases. As noted, in many cases, international treaties are ratified tacitly by parliament, not by formal legislation.
The institutional milestone effect Figure 4 presents the development over time, from 1981 to 2009, for all bill types taken together.
Figure 4 here Figure 4: Bills with and without EU links, 1981-2009 250
25%
200
20%
150
15%
100
10%
50
5%
0
0%
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 EU-linked bills
TOTAL amount of bills
Percentage of EU linked bills
Linear (Percentage of EU linked bills)
National bills became more influenced by European institutions and policy. The linear trend line shows the cumulative average, from under 5 percent in the early 1980s to above 20 percent in recent years. But the proportion of EU-linked bills did not increase gradually; there are clear ups and downs that sometimes happen from one year to the next. Year-to-year spikes in EUreferences in bills are more frequent than sudden large drops, but it seems that sudden
218
rises in the number and perhaps the magnitude of EU links in national bills also experience decline in following years. This may indicate the effect of European institutional milestones, in particular the successive treaties on European integration. We see some effect of the Single European Act in the year following 1987; then a next and steep rise in 1992 when the Maastricht Treaty was signed; the enforcement of the Amsterdam Treaty in 1999 (signed in 1997) is also visible, as is Nice which was accorded in 2001, took effect in 2003 and lifted the proportion of EU-referenced bills close to 25 percent. Finally, the Lisbon Treaty resulted in a modest peak in 2007. The later decline may be related to international and certainly also domestic confusion and controversy about the status of this Treaty until it was ratified in December 2009. This pattern suggests not only an ‘institutional milestone’ effect, but the declines in years after the event may also indicate that Europeanization of legislation was selective: it involved a series of bills to adapt to changes in European jurisdictions and policy. After these waves of attention the proportion of EU-linked bills went down again. Figure 4 also reveals that in years of lower legislative productivity, the proportion of EU-linked bills did not decline accordingly. In fact, years of low bill production are often associated with a higher proportion of Europeanized legislation. This may indicate that legislators are limited in their ability to include or ignore EU signals – the EU has become more of a constant factor in bill production, and is less sensitive to ups and downs in national legislative activity. Later we consider the mandatory or voluntary nature of references to the EU in national legislation.
Policy topic variation in Europeanization From a Europeanization perspective, not all policy topics are equal. Figure 5 presents the numbers of bills with one or more EU links and their proportion relative to the total volume of bills on that topic. The typical domains of Europeanization are visible, but more clearly in terms of absolute numbers than as proportions: domestic commerce and business is the most prominent topic of EU-linked bill production, before transport, foreign affairs and environment, with justice and crime and civil rights and immigration following at a lower production level. Agriculture is, in terms of numbers of bills, less a matter of formal legislative activity than its share in European spending may suggest. National policy on
219
macro-economic issues also involves just a modest number of bills with EU references. Domains of very low national activity in producing EU-linked legislation are housing, social security and defense.
Figure 5 here
Figure 5: Number and percentage of EU-linked bills 19. Foreign Policy 18. Foreign Trade 15. Commerce 17. Science & Technology 8. Energy 7. Environment 10. Transport 4. Agriculture & Fisheries 2. Rights, Immigration, Integration 3. Health 1: Macroeconomy 12. Justice & Crime 5. labor 20. Public Administration 21. Planning 6. Education & Culture 16. Defense 13. Social Security 14. Housing 0
20
40
60
Number EU linked Bills
80
100
120
140
percentage of EU linked bills
Although the area of commerce and business forms a considerable part of all EUlinked legislation (over 130 bills out of the 572 with EU links), the amount of legislation influenced by the European Union institutions and policies is still less than one third of the
220
total volume of bills devoted to this topic. Foreign policy and foreign trade bills are similarly influenced by the EU in relative terms; each with about one third of the total number of bills submitted between 1981 and 2009. Compared to the average of 12.3 percent of all bills with one or more EU links, science and technology, energy, environment, transport, agriculture as well as rights and immigration are topics on which legislation contains explicit reference to the EU. The other of the 19 main topic categories are domains of low European salience, sometimes despite volumes of bills that exceed those on more Europeanized domains. Justice and crime for example is ranked fifth in the number of Europeanized bills, but only comes on twelfth place when taking the proportion of EU-linked bills out of all bills on this topic. These findings speak to the institutional development of policy making jurisdictions in the European Union. Topics vary in salience at the European stage, and they also vary in their visible effect on national legislation. This topic diversity is related to what we call the ‘institutional milestone effect’: the scope of the Europeanized legislative agenda broadened in the early 1990s after the Maastricht Treaty, and has since expanded and contracted to some extent. This broadening and limiting of the scope of the agenda varied between 8 and 13 topics, out of a maximum agenda scope of 19 topics distinguished in the policy agendas coding scheme.7 Note that this variation in scope only relates to the agenda of Europeanized bills. Variation in the scope of the entire legislative agenda is much more limited. To conclude, on the legislative agenda with EU-links, one set of topics is always present, one set is almost never present despite expansion of European jurisdictions, but a third set of topics shows some rise and decline on the Europeanized legislative agenda as the process of European integration went from one treaty to the next.
Compexity and compulsion Two final aspects of the Europeanization of national bills are the complexity of EU-linked bills and the degree to which attention to the EU in legislation is compulsory, mandated by European institutional decisions and policies. By complexity we mean the extent to which 7
We also used a Herfindahl index to measure the degree of topic fragmentation over time for the legislative agenda with EU-linked bills. See appendix 2. The institutional milestone effect is visible, but the index also shows that over time, large shifts in attention between Europeanized topics on the legislative agenda became more rare (between 0.10 and 0.20 on a scale from 0.0 to 1.0). This indicates a cumulative effect of Europeanization.
221
Europeanized bills contain few or many key words as distinguished in our data approach. We did not qualify these individual key words, we consider simple counts. Figure 6 shows the results. The increase in Europeanization clearly involved a trend towards more complexity. The density of references to European institutions and decisions went up considerably over time, particularly after the Treaties of Nice and Lisbon. This trend may show the cumulative effect of the Acquis Communautaire on national legislation in The Netherlands, within the limits of the extent of Europeanization of bills that we found in the previous parts of the analysis.
Figure 6 here
Figure 6: Average number of references per EU-linked bill 1981-2009 10 9 8 7 6 5 4 3 2 1 0 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
Figure 7 shows the average level of complexity of EU-linked bills per policy topic. While the most Europeanized domains of legislation are high on the list (compare figure 7 to figure 5), also bills in areas of low key Europeanization can become quite complex on EU matters whenever they are mentioned. Planning and water management legislation does not, in many instances, contain EU references, but when this happens the density is higher than in any other domain. This also applies to education and culture bills with explicit links to the EU. Thus,
222
the degree of EU complexity and the degree of Europeanization in national bills do not always go together.
Figure 7 here
223
Figure 7: Average EU words per EU-linked bill 1981-2009
21. Planning 17. Science & Technology 7. Environment 15. Commerce 6. Education & Culture 3. Health 8. Energy 10. Transport 5. labor 12. Justice & Crime 1: Macroeconomy 2. Rights, Immigration, Integration 13. Social Security 20. Public Administration 18. Foreign Trade 14. Housing 4. Agriculture & Fisheries 19. Foreign Policy 16. Defense 0
1
2
3
4
5
6
7
8
9
224
Complexity is related to compulsion. The mandatory or voluntary nature of EU attention in national bills also is measured. How many of the 572 EU-linked bills contain a direct mandatory element in bill initiation? As mandatory EU attention in national bills we consider all cases in which the bill contains explicit mention of transposition or implementation of directives, treaties, court rulings etc. that require such follow up in member states. ‘Voluntary’ EU attention are instances of anticipation of EU policy or references to soft law less pressing on the national legislative agenda and involving more choice in tuning national legislation to EU policy. We also include the bills ratifying European treaties in this voluntary category, as in the Netherlands the decision to ratify via formal legislation is at the discretion of the national parliament. Figure 8 presents the development over time.
Figure 8 here Figure 8: proportion of EU mandated bills
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 81
83
85
87
89
91
93
95
97
99
01
03
05
07
09
After a period in which mandated and voluntary legislation with EU-links alternated, the trend is towards mandated EU links in bills. Peaks of compulsory EU attention in bills occur, and they correspond to the institutional milestone effect. The Single European Act
225
involved a clear spike in mandatory reference to the EU and in following years the proportion of mandated EU-links is at or above 80 percent mark most of the time. While bill production varies from year to year, this trend is another indication of a bottom line of attention to the EU in legislative activity in The Netherlands. Such EU-mandated bills also contain more references to the EU than voluntary ones, instances of symbolic lip service to the EU are rare in national legislation. Our findings correspond to the results of analysis by Bovens and Yesilkagit, who found that 12.6 percent of all legislation in place on 31 July 2003 was transposition of EU directives (2010: 61). We have shown the figures vary from year to year, but within a limited range.
Parliamentary Attention to Europeanized Bills
The vast majority of bills is initiated by the government. All kinds of mechanisms for coalition governance must secure their political endorsement. Coalition agreements set the initial legislative agenda, arenas for managing dispute must prevent that debate over issues gets out of political control, and informal rules of multiparty government constrain both cabinet ministers and the parliamentary groups supporting them. These mechanisms and rules also determine the discretion for parliamentary parties when scrutinizing bills and introducing amendments. Does this mean that parliament is a rubber stamp on matters of legislation? This question is particularly interesting for EU-influenced issues, as one widespread diagnosis of the democratic deficit in the European Union is that it exists at home – in the national parliaments of member states. The contention of Moravcsik (1994) that Europeanization has increased the information and power position of executives and further weakened the role of parliaments speaks to this domestic deficit. Other authors however argue that the risk of ministerial drift on policy issues induces legislative chambers to engage in legislative review (Martin and Vanberg 2005). In light of this theoretical expectation, bills with European content may lead the legislative chambers to intensify scrutiny. Thus rather than weakened and low key, parliaments are assumed to be more active in amending EU-linked bills.
226
We consider these two rival expectations and analyze how the Tweede Kamer as the chamber of representatives had used its competence to amend bills with and without references to the EU. The Dutch bicameral system is asymmetric; as said the Senate, Eerste Kamer, does not have the Constitutional right to amend bills and can only accept or reject them. On matters with EU content the Senate may have become more alert in recent years, but it uses its legislative veto power infrequently. More generally, the general adoption rate of bills is very high in the Netherlands. In cases of legislative ratification of treaties, such political endorsement happens on explicit parliamentary request. The bulk of legislation is ordinary bills, and also the vast majority of amendments was on this type of legislation. Data on these amendments are available for the period 1995-2009. In this period that includes 2,660 bills, the total volume of amendments was 7,045. On average this was 3.1 amendments per bill. Table 2 presents the proportions for both categories of bills.
Table 2 here Amendments to bills, 1995-2009 Bills with
Bills without
amendment
amendment
Total
29.7%
70.3%
100% N=411
EU linked bills 19952009 non EU linked bills 1995-2009
Total
100% 41.0%
59.0%
39.2%
60.8%
N=1043
N=1617
N=2249 100% N=2660
These findings show that attempts to amend bills were not distributed evenly. Most bills submitted between 1995-2009 were not amended (over 60 percent). While this was true also for bills without EU links, EU-linked bills were amended in less than 30 percent of the cases.
227
This is counterevidence to the assumption that parliaments are more alert and active when cabinet ministers address EU-related matters. Considering developments over time, figure 9 shows the pattern of parliamentary activity from year to year, with average numbers of amendments to bills with and without EU links.
Figure 9 here
5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Av. Amendments per bill without EU link
Av. Amendments per bill with EU link
Amendments are not only introduced in fewer cases of legislation when bills had EU-content, our findings also show a trend in the number of amendments per bill. This trend is that with some rise and decline, the average number of amendments to EU-links bills drops compared to bills without EU references. In the 1990s representatives in the Tweede Kamer tabled as many amendments to bills with explicit indications of EU influence as to other bills, on average more than three per bill.8 But compared to the two to three amendments for non EUlinked bills in more recent years, Europeanization seems to have made the Tweede Kamer much less active – and perhaps reluctant to act. This lower level and even decline in 8
But as said, this was done in fewer cases of legislation; thus amendment activity was more concentrated on fewer bills.
228
amendment activity on bills submitted for parliamentary scrutiny provides no evidence for the assumption that parliamentary parties use amendments to bills as a tool to constrain cabinet ministers on matters relating to European policy. In the Dutch case, there seems to be little in the way of cabinet ministers producing laws with EU references, a trend that speaks more to the argument of Moravcsik than the expectation of Martin and Vanberg on parliamentary control of ministerial drift. Only in some years has the parliament targeted its amending activity more to EU-linked bills. In the most recent six years this has not happened. Given the indications of selective attention, we also analyze possible variation between policy topics in parliamentary scrutiny through amendments. Figure 10 presents averages for EU-linked and non EU-linked bills for 19 topic categories. Agriculture, planning and water management, and in particular energy bills containing references to the EU appear quite sensitive to amendment. Bills regulating energy issues involve even more than twelve amendments on average when these bills contain links to European institutions or policy, but this is just a small category in absolute numbers of bills, so amendment activity is quite concentrated. Further, bills on planning issues attracted nearly eight amendments on average. This topic category also involved the highest complexity in case references to the EU. Environment, foreign policy and foreign trade are other topics on which legislative attention in the Tweede Kamer involved more amendment activity in case of EU links, but this happened at a rather lower level of intensity (barely more than one amendment per bill, on average). For most topics, however, the Tweede Kamer was more inclined to amend bills if these contained no clear references to the European Union.
Figure 10 here
229
Figure 10: Amendment rate per topic for EU linked and non EU linked bills 8. Energy 21. Planning 4. Agriculture & Fisheries 17. Science & Technology 5. labor 2. Rights, Immigration, Integration 10. Transport 15. Commerce 6. Education & Culture 20. Public Administration 3. Health 7. Environment 19. Foreign Policy 18. Foreign Trade 12. Justice & Crime 1: Macroeconomy 16. Defense 14. Housing 13. Social Security 0
2
Amendment rate EU-linked bills
4
6
8
10
12
14
Amendment rate non-EU linked Bills
In short, policy topics with the highest proportion of Europeanized bills involve just limited amendment activity. Even if some topics with EU-links attract more attention from parliamentary parties introducing amendments (such as energy, planning and agriculture), this activity is concentrated on a small number of bills. This pattern can be seen in connection with the increased complexity and compulsion of attention to EU matters in national legislation. The mandated nature of some part of national legislation following European decisions and the observed rise in complexity of references to these European decisions seem to have an adverse effect on political parties in the Tweede Kamer with regard to their use of legislative tools. Matters of domestic origin on
230
the legislative agenda are given far more attention than matters with EU content. Contrary to the findings for other countries in this volume, such as France, Germany and Italy, the chamber of representatives in the Netherlands seems either less willing or less equipped (or both) to amend bills with explicit links to European institutions and policy.
Conclusion
The Myth of Delors acquired political significance in the Netherlands as the process of European integration became controversial after the turn of the century. After many years without much explicit attention and debate, Europeanization became the impetus of intense debate. Costs and benefits of European membership no longer were defined just technically but came to be dramatized. This change fuelled mythology on the scope and degree of Europeanization of the domestic policy agenda. Analysis of the legislative agenda over the past 30 years, however, shows that even in the most Europeanized policy domains, the visibility of European institutions and policy via explicit references in bills is nowhere near 80 percent. It was never higher than about 30 percent. In numbers, bills on domestic commerce and business carry considerable European content, and in proportion, other topics such as foreign affairs and foreign trade also are close to this 30 percent level, followed by science and technology, energy and environment. But bills on other policy topics refer to the EU far less frequently. On average, 12.3 percent of the bills produced between 1981 and 2009 contained explicit reference to EU institutions or EU policy. Bills ratifying international treaties included such references (21.4 percent) more frequently, which does not come as a surprise as part of these treaties were on European integration. Constitutional revisions were hardly influenced by EU matters. This may indicate that in the Netherlands, constitutional engineering is framed very much in domestic terms and does not include much reference to institutional design and principles of policy at the European level.
231
The modest Europeanization of the national legislative agenda came in waves. While the trend was upward, the emerging pattern shows ups and downs following major events of European integration via ratification and enactment of European treaties. These treaties expanded the scope and magnitude of European competencies on main policy domains. The pattern we found however shows that rising attention to European policy in national bills mostly was temporary; attention did not remain at the level set at these major European events. Moreover, these rises in attention involved an expansion in scope of the EU-linked legislative agenda, followed by some contraction in next years. These episodic and modest effects of European integration on the legislative agenda came with a rather low key involvement of parliament. Since 1995, the Tweede Kamer used its constitutional power to amend legislation with references to the EU less and less. Compared to other countries in this volume, the Dutch chamber of representatives did not appear to show much alertness when European policies influenced the national legislative agenda. We acknowledge that this analysis of bills does not capture types and forms of delegated regulation. It also does not include transposition or implementation of European policy via other means than formal legislation. None the less, the analysis of explicit European impulses on the legislative agenda in the Netherlands over a thirty year period shows that the extent of this influence is far below what either enthusiasts or critics of European integration have claimed. It also shows that the nature of this influence mostly was mandatory. Instances of voluntary reference to EU policies were infrequent compared to indications of transposition and implementation of European policy decisions. The compulsory nature of attention to the EU on the legislative agenda included an element of high complexity; bills with EU links often contained several references to European decisions, particularly on topics that were defined technically. On such matters, information costs or political costs for parliamentary parties to become engaged are weighted against expected electoral benefits. For example, as environmental issues became more prominent on the European agenda, national attention in recent years declined. That is, they were addressed in mandatory legislative activity, but received much less political attention otherwise (Breeman and Timmermans 2008).
232
Political avoidance as a strategy accounts for the low and declining level of attention in the Dutch parliament to EU-linked bills. This avoidance of European matters may or may not happen in other arenas where policy agendas are set, such as coalition agreements of new governments, annual policy speeches of executives, and parliamentary questions and motions. These other venues are much more exposed to the audience, and for this reason they may be used by political parties to create dramatic images of problems. Recent work takes such other venues into account (Raunio and Wiberg 2009; Breeman and Timmermans 2010). It may involve both positive and negative agenda setting – issues rising and falling, dramatized or suspended, exploited politically or delegated to communities of specialists and experts. In the Netherlands, the dramatization of European integration in venues attracting public attention has involved populism and, as Schmidt (2005) put it, ‘politics without policy’. It also led parties with government responsibility to suppress debates, particularly since the national referendum on the European Constitutional Treaty in June 2005, in which the government lost much of its credibility on European affairs. This has reinforced the decline in parliamentary attention to bills with European links, given the high costs of legislative scrutiny and the risk of populist attacks by other means involving lower costs. Thus the trend in recent years was an increasing democratic deficit of the European Union in policy arenas and other venues of agenda setting at home in the Netherlands (Van Noije 2007). New European and national rules of the game were designed for empowering the national parliament in addressing issues of European integration. Political parties gained more opportunities for legislative scrutiny, for monitoring ministerial behavior in European policy making arenas, and for bringing EU related matters on the domestic agenda. But institutional tools depend in their effect on the expected benefits that parties see in increased attention to EU matters. The level of electoral uncertainty reached an unprecedented level in the parliamentary elections of June 2010, and this situation is likely to persist in coming years. Under these conditions, political parties still face large risks when becoming more active on issues with European content.
Acknowledgements
233
Thanks to Rosalie Belder, Josta de Hoog and Erlijn Wenink for their assistance in data collection. We are also grateful to the Parliamentary Documentation Center for its support in dataset construction, and Hessel de Zoeten for his comments on a draft version of this chapter. This research was sponsored by the Ministry of Internal Affairs and the Montesquieu Institute.
Appendix 1: Keywords European Union/EU European Community/EC European Economic Community/EEC European Coal and Steel Community/ECSC(-Treaty) European Monetary Union/EMU European Monetary System/EMS Common Market Single Market European Market Single European Act Merger Treaty Treaty of Nice/Amsterdam/Lisbon/European Constitution/Constitutional Treaty European Atomic Energy Community/EAEC/Euratom(-Treaty) European law European Union/EU/Community/EC law European/EC/EEC Directive/Framework Directive European (Commission) Regulation
234
European Parliament
Appendix 2: Herfindahl index of topic fragmentation on the EU-linked legislative agenda
Topic fragmentation, EU-linked legislative agenda
1.00 0.80 0.60 0.40 0.20 0.00 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009
References
Asser Institute (2007), Pilot-Monitor EU invloed. Een onderzoek naar de realiseerbaarheid vna een Permanente Monitor voor het meten van de invloed van Europese regelgeving op in Nederland geldende wet- en regelgeving, Den Haag (incl English summary) Breeman, Gerard and Arco Timmermans (2008), Politiek van de aandacht voor milieubeleid. Wageningen, Wot- rapport 77 (incl English summary). Breeman, Gerard and Arco Timmermans (2010), ‘The Europeanization of Laws, Queen’s Speeches and Coalition Agreements in The Netherlands’, Paper presented at the Council for European Studies Annual Meeting, Montreal 15-18 April 2010.
235
Bovens, Mark and Kutsal Yesilkagit (2005), ‘De invloed van Europese richtlijnen op de Nederlandse wetgever’, Nederlands Juristenblad 11 maart 2005: 520-528. Bovens, Mark and Kutsal Yesilkagit (2010), ‘The EU as Lawmaker: The Impact of EU Directives on National Regulation in The Netherlands’, Public Administration 88 (1): 57-74. Majone, Giandomenico (2005), Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth. Oxford, Oxford University Press. Martin, Lanny (2004), ‘The Government Agenda in Parliamentary Democracies’, American Journal of Political Science, 48 (3): 445-461. Martin, Lanny and Georg Vanberg (2005), ‘Coalition Policymaking and Legislative Review’, American Political Science Review 99 (1): 93-106. Noije, Lonneke van (2007), The Democratic Deficit Closer to Home. Agenda-building relations between parliament and the press, and the impact of European integration, in the United Kingdom, the Netherlands, and France, Doctoral dissertation, Free University Amsterdam. De Jong, Peter and Michiel Herweijer (2004), ‘Alle regels tellen’, Beleidswetenschap 18 (3): 26-40. Hogenbirk, Sjoerd and Sebastiaan Princen (2010), ‘Het soortelijk gewicht van Europese wetgeving. De invloed van de EU nader bekeken’, Bestuurskunde 2010/1: 71-79. Middelaar, Luuk van (2009), De passage naar Europa. Geschiedenis van een begin. Groningen: Historische uitgeverij. Moravcsik, Andrew (1994), ‘Why the European Union Strengthens the State: Domestic Politics and International Cooperation’, Center for European Studies Working Paper 52, Harvard University, Cambridge MA. Raunio, Tapio and Matti Wiberg (2009), ‘How to Measure the Europeanisation of a National Legislature?’, Scandinavian Political Studies 33 (1): 74-89. Schmidt, Vivien A. (2005), Democracy in Europe. The EU and National Polities, Oxford, Oxford University Press.
236
Timmermans, Arco (2011), ‘The Netherlands: Legislative Agenda Setting and the Politics of Strategic Lock-Ins’, in Bjorn Erik Rasch and Gerorge Tsebelis (eds.), The Role of Governments in Legislative Agenda Setting, London, Routledge. Timmermans, Arco and Rudy B. Andeweg (2000), ’The Netherlands: Still the Politics of Accommodation?’, in Wolfgang C. Müller and Kaare Strøm (eds.), Coalition Governments in Western Europe, Oxford, Oxford University Press. Timmermans, Arco, Peter W.A. Scholten and Steven Oostlander (2008), ‘Gesetzgebung im Politischen System der Niederlande’, in Wolfgang Ismayr (ed.) Gesetzgebung in Westeuropa, Wiesbaden, VS Verlag.
237
10. The Europeanization of law-making activities in Spain 1 Anna Palau & Laura Chaqués, University de Barcelona
Introduction About 50% of laws passed in Spain since 2001 have been directly related to the European Union (EU). The Europeanization of law-making activities has increased steadily since Spain’s entry to the European Economic Community (EEC, later EU) in 1986, especially since the implementation of the Economic and Monetary Union. For the last three decades, Spanish law-making activities have come under ever-increasing pressure to adapt to common European standards with regard to the what and how of regulation (Jenny and Müller 2009). Nonetheless, only 35% of all the laws passed over the last three decades have been linked to Europe; and as expected, the impact of the EU has been concentrated in specific policy areas, especially those related to the single market and the environment. Overall, 63% of environmental and agricultural legislation and 56% of economic and fiscal legislation is related to the EU, compared with only 25% of laws on welfare issues. The first goal of this chapter is to explain how Europeanization has affected Spanish legislative activity over time, from 1986 to 2007, and across policy areas. The analysis illustrates that the impact and scope of European integration on Spanish law-making is still far from reaching Jacques Delors’ prediction that by 1998 the European Community would be the source of 80% of national legislation (economic legislation with a possible extension to fiscal and social legislation). In this way, the Spanish case corroborates previous research about the Europeanization of law-making activities in other member states (e.g. Wiberg 1997; Töller 2006; Müller et al. 2010). As in the cases of France and Germany, the influence of the EU in setting the Spanish legislative agenda is increasing, but overall it only accounts for the greater
1
We would like to thank the rest of members of the Spanish Policy Agendas Project, Luz Muñoz, Ferran Davesa and Lluis Medir, for their help in the preparation of this article. The research for this paper is carried out as part of the project “The Politics of attention: West European politics in times of change: the case of Spain”, funded by the Ministerio de Ciencia e Innovación, (SEJ-2007-30760-E/SOCI), under the Eurocores ECPR 2007 of the European Science Foundation (ESF), and the AGAUR (SGR 536).
238
part of legislation in specific areas, and there are major variations across time and policy subsystems. The study of the Europeanization of laws is relevant not only for understanding the power of the European Union in setting the national legislative agenda – that is, the extent to which the legislation passed in Spain is defined by European imperatives – but also for the analysis of important normative concerns related to the “de-parliamentarization” process of national political systems (Chryssochoou et al. 1998). In this regard, the second goal of this chapter is to contribute to the debate on the democratic deficit by analyzing how far Europeanization has affected relations between the executive and Parliament in Spain. We argue that Europeanization has contributed to reinforcing the already predominant position of the executive with regard to the Parliament. The executive plays a key role as the legitimate national representative in the EU policymaking process, controlling the political agenda on EU affairs and most of the transposition of EU directives – 90% of the directives are transposed by secondary legislation – which finally results in the reduction of the already limited legislative capacity of the Spanish Parliament. Here we go beyond the analysis of formal transposition of directives and present the first systematic, comprehensive analysis of the Europeanization of all the laws and bills passed in Spain. Several authors have investigated the transposition patterns of EU directives across member states and policy areas (e.g. Mastenbroek 2003, Falkner et al. 2004, Kaeding 2005, Treutlein 2007, König and Luetgert 2008), or the variations of compliance with EU directives inside a single policy sector (e.g., Börzel 2000). From a quantitative and comparative perspective, these analyses explain how the case of Spain differs from other countries, why there are variations in the degree of compliance with EU directives across sectors, and to what extent these differences can be explained by the mobilization of social groups or the characteristics of the Spanish institutional context. In this chapter we depart from these analyses, and introduce a new concept of the Europeanization of laws that is not only limited to the transposition of directives. To do so, we analyzed the European content of all the bills and laws passed from 1986 to 2007 and classified them in 19 different policy areas, following the methodology of the Comparative Policy Agendas Project (www.comparativeagendas. com). We consider a law as Europeanized if it is totally or partially defined by a European binding regulatory act: a
239
Directive, Regulation, Decision, European Court of Justice decision, or Treaty provision. Other indirect forms of Europeanization, for example laws referring to EU “soft law” – for example, Commission White or Green Papers – or generic references to Europe are excluded from the analysis. The construction of a large database of law-making activities in Spain also permits us to go beyond previous investigations of the Europeanization of the Spanish political system which mainly focus on the formal institutional reforms since the country’s entry into the EEC (e.g., Morata, 1987, 1998, 2004, Closa 2001, Molina 2001, Cienfuegos 2001, Morata and Mateo 2007, Arregui 2007), or the patterns of Europeanization across different policy subsystems (e.g., Torreblanca, 2001, Lombardo, 2003, Subirats et al. 2001, Chaqués 2002, Palau, 2009). Most of these analyses centered exclusively on the Spanish case, providing detailed explanations of how Europeanization has affected a particular policy sector or political institution like the Parliament or the executive in recent decades. The results we present here provide a comprehensive explanation of how the Europeanization of law-making activities occurs, why there are differences across policy subsystems and to what extent European integration is contributing to a process of de-parliamentarization in Spain. The chapter is organized as follows. First, we contextualize the analysis of legal Europeanization from a theoretical perspective. Second, we explain the methodology and coding procedure. Third, we describe the Europeanization of national law-making activity in Spain, using Delors’ prediction as a benchmark. Fourth, we discuss these results in relation to the question of who benefits from Europeanization and in relation to the transformations generated by Europeanization in executive-parliament relations. The final part of the chapter summarizes and presents the conclusions.
Shrinking the role of the Spanish Parliament A central debate in the study of Europeanization is the impact of the EU on executiveparliament relations, that is, the extent to which Europeanization weakens national parliaments in favor of the executive (de-parliamentarization process), or, on the contrary, whether national parliaments maintain or even enhance their positions at the level of member states and within the European governance system (re-parliamentarization process) (Goetz and Meyer-Sahling 2008). From this perspective, national parliaments have redesigned themselves in order to enhance their access to information and play a more significant role in
240
formulating and monitoring EU policies with regard to the executive’s activity (Melsaether and Sverdrup 2004, Raunio 2005, Saalfeld 2005). In contrast, authors who regard national parliaments as “losers” argue that European integration has changed the balance of powers at the domestic level in favor of the executive (e.g. Chryssochoou et al. 1998, Hix 1999, Maurer and Wessels 2001, Jachtenfuchs 2001). Europeanization erodes the role of national parliaments mainly because the executive has a predominant role in EU policies –national constitutions generally designate the executive as the sole representative in EU negotiations— and monopolizes control over the agenda on European affairs (Moravsick 1994, Molina 2001, Olsen 2007, König 2009). The Spanish executive is recognized as the legitimate national representative in the EU policymaking process. This confers significant “gate keeping” power to it as well as an increasing capacity to veto policies, strategically manipulating the timing, sequencing and presentation of policies and issues (Moravsick 1994). However, as Capo (2003) illustrates, the Spanish executive already has a predominant position in the national legislative process, for several reasons: First, the Spanish Constitution empowers the executive to enact decree-laws and legislative decrees, which represented about 25% and 5% respectively of the laws passed from 1978 to 2008. Second, parliamentary rules give the executive a predominant position in the legislative process – for example, executive bills have priority over any other bills – and third, most of the laws that are actually passed by the Spanish Parliament are initiated by the executive; that is, the adoption rate for executive bills is 88%, compared with only 10% of parliamentary bills. The capacity of the Spanish Parliament to participate in European affairs and to oversee the executive’s conduct is even more limited than in the domestic policy process, for several reasons. First, the transaction costs of renegotiating EU agreements at domestic level are high: once a decision is taken in the European arena, renegotiation at the national level by the Spanish Parliament is costly and risky, since it delays the implementation of policies, creates uncertainty and heightens the risk that negotiation will collapse altogether (Moravsick 1994). By presenting Parliament with a “take or leave it” decision between the European agreement and the status quo, the executive has the power to choose from among the policies that lie within the domestic “win set”. According to this argument, we expect that the laws and executive bills with European content discussed by the Spanish Parliament will have a higher adoption rate, and a lower amendment rate, than that without European content.
241
Second, the de-parliamentarization process is also explained by the increasing role of secondary legislation2 in the transposition of EU directives. The executive acquires increasing legislative competences by using secondary legislation to adapt the policy making process to European standards, thus reducing the political autonomy of the Spanish Parliament in EU matters. Some authors explain the decrease in primary legislation as a rational solution aiming to respond to the increasing demands for legislative production on the national Parliaments. The nature of EU law and the increasing volume of European legislation have led to a shift towards secondary legislation in order to avoid the length, complexity, and lack of flexibility of parliamentary procedures (Ordoñez 1994). Other authors such as Treutlein (2007), Döring (2004) and Döring and Hallerbert (2004) explain the increasing use of secondary legislation to transpose EU directives as a contingent process that depends on institutional factors like the number of veto players: the higher the number of veto players, the greater the use of secondary legislation. Testing this hypothesis is of particular interest in the case of Spain. The country’s majoritarian parliamentary system allows for the creation of outright majorities. The first Socialist governments of Felipe González (1982 to 1993), and the second government of the conservative party (PP) leader José María Aznar (2000-2004) obtained absolute majorities, while for the rest of the period neither the PSOE (first with González from 1993 to 1996, and later with José Luis Rodríguez Zapatero, in power since 2004), nor the PP (the first Aznar government from 1996-2000) won an outright majority. When governing as a minority, both the PSOE and the PP depend on the support of other political parties, mainly regional parties, which are willing to support the national government in exchange for an increase in political autonomy for their territories (Chaqués and Palau 2011). So, when the executive is governing as a minority, the number of veto points increases; accordingly, we would expect a reduction in the use of primary legislation for the transposition of EU directives. In addition, the timing of the transposition within the policy cycle may also play a crucial role in explaining the use of secondary legislation to transpose EU directives (Treutlein 2007). As Döring (2004) argues, the marginal costs of passing additional bills tend to be higher towards the end of an election period. The policy cycle plays a crucial role for governmental preferences regarding transpositions, as the closer the transposition decision is 2
Secondary legislation refers to legislative actions directly passed by the executive, mainly decrees and ministerial orders, while primary legislation refers to national legal action passed by Parliament, mainly laws.
242
to the end of an election period, the lower the ratio of primary to total transposed directives, particularly with regard to highly controversial ones. According to this argument, we would expect the rate of primary versus secondary legislation to decrease before elections. Finally, the way in which formal parliamentary powers are affected by Europeanization also depends on public opinion and party stances on European integration. Several authors, like Raunio (2005) and Goetz and Meyer-Sahling (2008) argue that parliamentary control over executive activity on EU issues increases according to the degree of opposition to European integration among the general population and the political parties: the greater the opposition, the stronger the control over EU matters. One of the main characteristics of the Spanish case is that both political parties and public opinion have a positive attitude towards European integration. Spain is the only country in Europe where EEC accession was supported by all political parties (Álvarez 1996, Montabes and Szmolka 2000). This crossparty support for the EU has characterized the Spanish political party system, with the result that all EU treaties, including the European Constitution, have been ratified with the broadranging support of all political parties and the Spanish population. This positive attitude is attributed to the benefits that the country is believed to have
gained from European
integration at the political and economic level. After Franco’s death, Europeanization was seen as a unique opportunity for the consolidation of democracy and for the structural reform of the economic system, and, more generally, for the country’s modernization. Identification with the EU became an important element in the political discourse of democratization, which blurred ideological differences within political parties over EU issues, and made the emergence of Euroscepticism across parties unlikely (Closa, 1995, Closa and Heywood 2004). This strong support for the EU has prevailed throughout the democratic period, even when Spain lost its position as one of the main receivers of EU economic funds. The only exception occurred with the ratification of the European Constitution in 2005, when regional parties such as ERC in Catalonia, EA in the Basque Country, and BNG in Galicia, and left parties such as IU and IC-V urged their electorate to vote “no” because of the Constitution’s perceived minimization of national identities and neglect of social issues (Powell 2002, Font and Rodríguez 2006, Szmolka 2007). Nonetheless, the European Constitution had the support of more than 90% of the members of the Congreso de los Diputados3. 3
Between 1982 and the ratification of the European Constitution, there were no significant disagreements between political parties with regard to European integration. With the ratification of the Maastricht Treaty,
243
Spanish public opinion has always been more supportive and enthusiastic with regard to Europeanization than other EU countries (Szmolka 1999, Sánchez 1999, Díez 2003, 2007). Most Spanish citizens have a positive attitude towards the EU, even though 70% of them consider their knowledge of the EU in general, and of its functioning in particular, to be low or very low4. This lack of information on EU issues reinforces the predominant role of political parties which, as the main link between politicians and the general public, channel these positive ideas regarding the integration process. In general, individuals who support a particular political party tend to support that party’s position on European affairs; and this is particularly the case in countries like Spain where citizens’ familiarity with the workings of the EU is low (Anderson 1998, Steenberg and Jones 2002, Hooghe and Marks 2005). In short, institutions and party preferences are important factors in explaining the Europeanization of law-making activities in Spain. The institutional framework of the Spanish political system (for instance, the prominent role that the Constitution confers on the executive in the EU policy making process and in the development of legislative activities), and the positive attitudes of the Spanish population and political parties towards the EU has favored the Europeanization of the legislation introduced in the Spanish Parliament since 1986. The following sections in this chapter explore the degree of Europeanization of legislative activities and the extent to which the process of increasing Europeanization has been accompanied by a process of de-parliamentarization.
Methodology and coding procedure For our study we use the databases created by the Spanish Policy Agendas Project (www.ub.edu/spanishpolicyagendas) following the methodology of the Comparative Agendas Project (www.comparativeagendas.org). Each bill and law is coded according to a universal coding system based on 19 major topic categories (table 1), and 247 subcategories. First developed in the US (Baumgartner and Jones 1993, Jones and Baumgartner 2005), this coding system was adapted to the characteristics of the Spanish political system introducing some revisions such as questions dealing with the national health system or ETA terrorism. The Izquierda Unida (IU) criticized the Treaty for its lack of a strong social vision. However, the Treaty was ratified with almost complete political support (314 votes in favor, three against and nine abstentions). 4 This percentage is higher than in other EU countries (see Szmolka 2007). It is also interesting that when Spanish citizens define the country’s political situation as either bad or very bad, the proportion of citizens who identify the EU as a positive institutional framework increases (Sánchez Cuenca 1999, Kritzinger 2003).
244
analysis of Europeanization is based on 1,287 national laws (table 2) – including ordinary and organic laws, legislative decrees and decrees-laws – 949 executive bills (Proyectos de Ley) and 1,502 parliamentary group bills (Proposiciones de Ley). The time period covered is 19862007, with the exception of the calculations of amendment rates of bills which begin in 1987. The Congreso de los Diputados has only given information about whether a bill or a law has been passed with and without amendments since 1987, and this information on amendments only applies to organic and ordinary laws. Our analysis is based on direct forms of Europeanization. That is to say, we consider that a law (or bill) is Europeanized if it is totally or partially defined by a binding European regulatory act, such as a Directive, Decision, Regulation, European Court of Justice decision, or Treaty provision. Because the goal is to analyze to what extent the Spanish law-making activity is constrained by the European Union’s agenda setting power, indirect forms of Europeanization, such as references to EU “soft law” or generic references to the EU, are not considered. EU “soft law” works as an alternative form of regulation to promote policy change in those policy areas where it has traditionally been difficult to develop “hard law”, such as social policy, education or internal affairs (López-Santana 2006). In these sensitive areas we expect to find a low level of direct Europeanization, since any attempts to impose it are unlikely to gain the political support of member states (Jacobsson 2004). The coding procedure is based on a system of manual codification and a key word search. For the analysis of national legislation we use the 21 key word search defined by the Delors’ Myth Research Project (table 3). The search is made on the full text of the law using the information provided by the Iberlex database (www.iberlex.es). Two coders performed the search and decided whether a law was directly Europeanized, or whether it referred to a generic reference to the EU or to “soft law”. Our database also includes information on transposition of directives provided by the Spanish Boletín Oficial del Estado (BOE)5. The databases produced by this public institution include all legislation passed since 1960, including a legal analysis that gives information on formal transposition. For the analysis of bills, as it is not possible to do an automatic keyword search using the Iberlex database, the
5
The BOE is published by the Ministry of the Presidency, which is in charge of printing, distributing and selling the State Official Journal. It is also responsible for publishing reports, and compilations of legal texts and carrying out the official printing of assignments requested by Ministries, organisms and other public bodies.
245
coding procedure is limited to a manual codification procedure on the basis of the full text of the bill.
The Europeanization of legislative activity From 1986 to 2007, 35% of the laws passed by the Spanish Parliament were directly related to the EU (table 4). This is a long way short of Delors’ Myth of 80%, but a detailed analysis of the data suggests that an increasing number of laws are linked to Europe (figure 1), with significant variations across time, policy subsystems, and types of law. We have defined three main time periods for our study: the first from Spanish accession to the EEC and the Single European Act to the ratification of the Maastricht Treaty (1986-1993); the second from the implementation of the Maastricht and the Amsterdam treaties (1994-2000); and the third the period between the Nice and Lisbon Treaties (2001-2007). During the first period (19861993) only 26% of laws passed in Spain were linked to Europe; this percentage increased to 30% in the 1990s (1994-2000), and reached 49% after the implementation of EMU (20012007). In the new millennium, most of the laws passed by the Spanish Parliament are Europeanized, but we find some striking annual variation, particularly with regard to the transposition of directives. The comparison of the three periods shows that more directives were transposed from 1986 to 1993 and from 1994 to 2000 – 34% of the laws passed with EU content transposed directives – compared with 30% from 2001 to 2007. This is mainly due to the imperatives of the Economic and Monetary Union (EMU) and to the necessity to adapt previous legislation to the new political context after accession. With the ratification of the Single European Act and Spain’s entry into the EEC in 1986, the percentage of laws with EU content that transposed directives was very high (in 1986, 61% of laws with EU content transpose directives), as it was after the ratification of the Maastricht Treaty (the figure reaching 61% again in 1995). In order to capture the impact of Europe on legislative activities, we also considered Europeanization by type of law (organic, ordinary, decree law, and legislative decree). Table 4 suggests the following conclusions. First, ordinary laws are the most Europeanized type of law – 38% of the total passed throughout the whole period. The percentage of ordinary laws which have been Europeanized has increased since Spanish accession to the EEC – in the new
246
millennium more than 50% of the total ordinary laws passed have EU content – reflecting the EU’s increasing influence on Spanish legislative activity. In contrast, only 15% of the organic laws passed since 1986 have been related to the EU. This low percentage is explained by the fact that organic laws are limited to issues (mainly fundamental rights and public liberties, the general electoral system, and the modification of basic institutions like the Constitutional Court) which are regulated almost entirely by the Spanish Parliament, and thus, the level of delegation of political power is very low. Most organic laws having European content (42% of the total passed) refer to the ratification of EU treaties and the regulation of electoral procedures with regard to European elections. Second, in contrast to the case of organic laws, most legislative decrees passed since 1986 have been related to Europe; 35% of all legislative decrees are related to the EU, and in most years this percentage reaches 100%. In fact, most of the directives that had to be transposed in 1986 with the ratification of the Single European Act and Spain’s entry into the EEC were transposed using legislative decrees. More precisely, by the Ley 47/19856, the Spanish Parliament delegated the capacity to transpose EU directives to the executive through the use of legislative decrees. This decision was taken for rational purposes, and it aimed to favor the adoption of the “acquis communautaire” and the accession process; but at the same time it highlights the capacity of the Spanish executive to control the political agenda and the legislative process in matters related to the EU. Third, the percentage of decree-laws directly related to the EU was 32% of the total passed from 1986 to 2007. Decree-laws are rarely used for the transposition of directives (in fact, in only 8% of cases) but they have been increasingly linked to the EU, especially since the mid-1990s after the ratification of the Maastricht Treaty. This increase is related to the type of issues that are normally regulated by decree-law. In formal terms, decree laws can only be used in case of urgent necessity, and can only deal with issues that do not affect the regulation of the basic institutions of the State, the rights, duties, and liberties of the citizens regulated in Title I of the Constitution, Spain’s Autonomous Regions, or electoral law7. 6
Ley 47/1985, de 27 de diciembre, de Bases de Delegación al Gobierno para la Aplicación del derecho de las Comunidades Europeas 7 Section 86 of the Constitution states that “In case of extraordinary and urgent need, the Government may issue temporary legislative provisions which shall take the form of decree-laws and which may not affect the legal system of the basic State institutions, the rights, duties and freedoms of the citizens contained in Part 1, the system of Self-governing Communities, or the general electoral law. Decree-laws must be immediately submitted for debate and voting by the entire Congress, which must be summoned for this purpose if not already
247
However, in practice, decree-laws are one of the main legal instruments used by the executive to pass new legislation, especially in the areas of agriculture and industrial policy (38% of the total), which are among the areas with the highest degree of Europeanization. In short, the analysis of Europeanization according to type of law illustrates the predominant role of the executive in the legislative process. Despite their exceptional nature, decree-laws and legislative decrees account for 25% of the laws with EU content; the rest of the laws with EU content correspond to organic laws (6%) and ordinary laws (69%), which are mainly introduced as executive bills. The Europeanization of laws varies not only across time and types of legislation, but also across policy areas. As table 6 and figure 2 show, Agriculture and Environmental issues are the most Europeanized areas (63% of the laws passed in these areas are related to the EU), followed by issues related to Science and Technology (59%), Macroeconomics and Foreign Trade (56%), Energy (52%) and Commerce and Industrial policy (50%). That is, agriculture, the environment and the economy (from macroeconomics to industrial affairs) are the most Europeanized policy areas, and areas related to basic functions of the State like defense (2%), justice and internal affairs (12%) or civil rights (20%) are the least Europeanized. Europeanization is also rare in welfare issues: around 25% of the laws related to social policy and 33% related to health issues are linked to the EU, and 22% of the laws on education. As expected, the level of Europeanization in these policy areas is low, reflecting the difficulty of passing “hard law” at the EU level in these sensitive areas. Finally, most of the laws linked to the EU are concentrated in three policy areas – macroeconomics, commerce and industrial policy and agriculture – which represent 53% of all the laws that are Europeanized (figure 3). More precisely, 26% of the laws with EU content concern macroeconomic issues such as the regulation of prices, interest rates, or taxes. This is the policy area that has the highest percentage of Europeanized legislation, followed by the regulation of commerce and industrial policy (19%) and agriculture (8%). Environment and science and technology represent 5% of all Europeanized laws passed from 1986 to 2007, while energy and foreign trade represent only 3% and 1% respectively.
in session, within thirty days of their promulgation. The Congress shall adopt a specific decision on their ratification or repeal in the said period, for which purpose the Standing Orders shall contemplate a special summary procedure”.
248
Who benefits from Europeanization? The percentage of Europeanized laws has increased over time, though with significant differences across types of law and policy subsystems. The question we pose now is to what extent this process of Europeanization of the legislative process has transformed executivelegislative relations. To do so we first analyze the institutional reforms made by the Spanish Parliament after European integration and then consider the extent to which parliamentary groups effectively make use of their formal powers to control the executive activities in different parliamentary venues. Like most national parliaments in Europe, the Spanish Parliament was reformed in order to have access to EU-related information and to control the executive’s activity in EU affairs (Matia 1999, Storini 2005)8. In 1985, a Committee on EC affairs (Comisión Mixta para las Comunidades Europeas) was created, although its functions were limited to the control of the legislative decrees and to accessing EU-related information (Ley 47/19859). One of the main limitations of this committee is that the executive may inform Parliament “a posteriori”, so most of the debates concern issues on which decisions have already been taken by EU institutions (Storini 2005). As a consequence, the Committee was reformed in 1988 (Ley 18/198810) and in 1994 (Ley 8/199411) increasing its formal powers and number of members. In spite of this, the activity of this committee has been uneven: between 1987 and 1993, 83 oral questions were raised by different parliamentary groups; the figure rose to 120 during the period 1994-2000, and then fell again to 71 after the implementation of the EMU in 200012. The Committee on EU affairs does not have a monopoly on the discussion of issues related to European integration. In fact, depending on the type of issue, the debate about the EU in the Spanish Parliament has become increasingly fragmented. For example, a question about the CAP can be discussed simultaneously by the Committee on EU affairs, in a plenary meeting in the Parliament, and also by the Parliamentary Committee on Agriculture. As figure 4 illustrates, the number of oral questions with EU content introduced in plenary meetings and 8
Since the mid-1980s, and following Declaration 13 of the annexed protocol of the Maastricht Treaty, most national parliaments have introduced reforms to increase their role in EU issues (Storini 2005). 9 Ley 47/1985, de 27 de diciembre, de Bases de Delegación al Gobierno para la Aplicación del Derecho de las Comunidades Europeas 10 Ley 18/1988 de 1 de julio de modificación de la Ley 47/1985 de Bases de Delegación al Gobierno para la Aplicación del Derecho de las Comunidades Europeas). 11 Ley 8/1994, de 19 de mayo, por la que se regula la Comisión Mixta para la Unión Europea 12 In addition to oral questions, this Committee performs other control procedures such as the preparation of reports or hearings of officials and specialists (see Matia 1999, Cienfuegos 2001, Besabe and González 2001, Magone 2007).
249
Parliamentary Committees represents a small percentage of the total (between 5 and 10% between 1986 and 2007). Parliamentary control on EU issues mainly focuses on agriculture (25% of the oral questions about agriculture are linked to the EU) and on EU institutional reforms and treaty negotiations (coded as international affairs, and reflecting 16% of the total) (figure 5). In contrast, issues associated with a low level of Europeanization, such as health, social policy or education, are hardly ever discussed in Parliamentary plenary meetings making reference to the EU. This is also the case of highly Europeanized policy areas, like environment and science and technology, which are rarely associated with the EU when being debated in the Spanish Parliament. Only 6% of the oral questions concerned with the environment (48 out of 861 oral questions) have European content – a very low percentage, considering that 63% of the laws passed on environmental issues are Europeanized. Another way to test whether Europeanization has contributed to reinforcing the role of the Spanish executive vis-à-vis the Parliament is to assess whether laws and executive bills with EU content have higher adoption or amendment rates than those without. As discussed above, we should expect lower amendment rates and higher adoption rates for bills and laws connected to the EU, mainly because the cost of renegotiating policy decisions already taken by the executive at the supranational level is high (Moravsick 1994). Table 5 summarizes the differences between amendments rates of laws with and without EU content. Overall, the data suggest that the percentage of Europeanized laws passed with amendments is higher (87%) than the average (77%), and higher than the percentage of laws without EU content (71%). The amendment rate for Europeanized laws increased gradually from 79% (1987-1993) to 92% (2001 to 2007), but this rise is even higher for those laws without EU content – the percentage of laws amended from 1987 to 1993 is 66%, and 85% from 2001 to 2007. In short, amendment rates are always higher for laws related to European affairs, but the differences have been declining for the last decade, mainly because the percentage of laws passed with amendments has been increasing, especially for laws without EU content. As a result, since 2001 only 8% of the Europeanized laws have been passed without amendments, versus 15% of the laws not linked to the EU. The analysis of executive bills confirms our hypothesis regarding the adoption rate: 92% of the bills with EU content are passed versus 89% of those bills without EU content (table 7). However, in the case of amendment rates we find the opposite trend: governmental
250
bills that are Europeanized have a higher amendment rate (81%) than those without EU content (65%). As in the case of laws, the amendment rate of Europeanized bills also increased gradually, from 61% (1987-1993) to 93% (2001-2007); but the amendment rate of bills without EU content also increased over time (from 55% between 1987 and 1993, to 80% from 2001 to 2007). From 2001 to 2007 only 5% of Europeanized bills were passed without amendments, compared to 15% of bills not related to the EU. The percentage of amended legislation according to policy area is summarized in table 8. We observe that most of the amended legislation in Spain concerns areas connected to the EU: this is especially true of laws related to the environment, in which 88% of the amended legislation has EU content, agriculture (77%), macroeconomics (71%), energy (70%), commerce and industrial policy (67%), and science and technology (61%). In contrast, in areas such as rights, labor, education or social policy, the amended legislation is mostly nonEuropeanized. If we consider only Europeanized legislation (figure 6) we observe that in defense, rights, social policy, and public land, 100 % of the Europeanized legislation is amended. Only international affairs – a policy area traditionally dominated by the executive– has more legislation without amendments (69%) than with amendments (40%). Overall, the analysis of amendment rates does not confirm the hypothesis of the increasing empowerment of the executive due to Europeanization. On the contrary, it seems that there is an increase in the percentage of laws with EU content submitted to control by parliamentary groups. However, it is difficult to arrive at a definitive conclusion because (1) the amendment rate is high and has increased for all types of laws and bills, with and without EU content, suggesting that it is probably related to other political factors in addition to Europeanization; and (2) the origin and motivation of amendments can be related to factors other than Europeanization, from technical factors to ideological issues. In this connection, further analyses should test whether amendments are motivated by EU-related aspects of the law, or whether they are related to other factors. Another question is to what extent the executive avoids passing legislation through Parliament to transpose EU directives, that is, to what extent the executive uses secondary legislation to regulate European issues, bypassing the Parliament. In Spain, almost 90% of the legislation used to transpose EU directives is secondary legislation (mainly decrees) passed by the executive; this is a general trend, with no important variations since 1986 (figure 7). The
251
only exception is the year of Spain’s entry into the EEC (1986), when 68% of the directives were transposed with primary legislation. This increase in the use of primary legislation cannot be understood as reflecting the predominance of the Spanish Parliament in this first period of accession to the EEC. In fact, the transposition of most of these directives was made by legislative decrees, that is, by delegation of the legislative tasks to the executive. Finally, figure 7 shows that, as predicted, the use of primary legislation decreases when election periods are approaching (the rate of primary versus secondary legislation declines significantly in the year of elections in 1990, 1993, 1996, 2000 and 2004). However, these results do not support the argument that the stronger the government and the higher the number of veto players, the higher the rate of secondary versus primary legislation. In Spain, from 1993 to 2000 and after 2004 there was no absolute parliamentary majority and governability depended on the agreement of central and regional parties. During these periods we do not see an increasing use of secondary legislation to transpose EU directives. In 1986, when the PSOE governed with a majority, primary legislation to transpose EU directives was used more frequently than secondary legislation (68% vs. 32%), but this percentage fell drastically in 1987, after accession, when 100% of legislation transposing directives was secondary legislation. By the same token, in 1995 Felipe González’s minority government recorded one of the highest rates of use of primary legislation (19%). The analysis of policy areas (figure 8) illustrates that laws about health, agriculture, education, energy, transport, housing and international affairs are mostly transposed with secondary rather than with primary legislation. In contrast, the transposition of directives in policy areas like macroeconomics, rights, labor, environment, justice and internal affairs, social policy, commerce and industrial policy, science and technology, foreign trade, government affairs and public lands, are more often carried out with primary legislation. Some authors (Olsen 1980, Treutlein 2007, 2009) consider that the use of primary legislation increases in policy areas related to the economy and issues that the government considers important and symbolic. In contrast, the use of primary legislation is considered costly for more technical issues such as transport, energy or environmental policies (Treutlein 2007:34). However, this theory is not completely supported by our data: we find a high use of secondary legislation for the case of transport and energy, but not for the environment. Besides, the existence of formal and material reservations of law in some policy areas in Spain explains why some policy areas are more regulated by primary legislation. This is the case, for
252
example, of the regulation of rights, a policy area that has necessarily to be legislated by organic laws and cannot be delegated to the executive.
Conclusions The Europeanization of law-making activities in Spain still lags far behind Delors’ figure of 80%, but it is taking on increasing relevance. Overall, 35% of the laws passed in Spain from 1986 to 2007 were related to the EU; the percentage has increased gradually from the entry into the EEC in 1986, especially after the implementation of EMU, to reach 50% in the new millennium. This process of Europeanization reflects the increasing power of the EU to set the national agenda, particularly in areas related to agriculture, the environment, science and technology, and economics. In other areas, like defense, rights issues, crime and justice or welfare the national government continues to enjoy a great deal of autonomy. The high percentage of Europeanized legislation in Spain comes as no surprise if we consider the wide support of the political class and public opinion for the EU. References to Europe in the national legislation are frequent because political parties and citizens have long identified the EU as an institutional framework that makes a positive contribution to the country’s economic and political modernization. Since 2006, coinciding with the failure of the European Constitution project, there has been a slight increase in negative attitudes towards the EU, but there is still wide support for European integration in Spain. The Europeanization of legislative activities in Spain has been accompanied by a process of a moderate reduction in the capacity of the Spanish Parliament to carry out its basic functions. This is explained by institutional factors such as the leading role that the Spanish Constitution assigns to the executive in the EU policy making process and in the development of legislative activities at the national level, but also because (1) parliamentary groups have delegated their legislative power to the government (legislative decrees) for the transposition of EU directives, especially since the entry into the EEC and the ratification of the Single European Act; (2) the executive has also taken the initiative to press forward most of the laws linked to the EU by introducing a decree-law or an executive bill; and (3) most EU directives are transposed with secondary legislation, leaving only a marginal scope for the Spanish Parliament to carry out this task. The analysis of legal Europeanization in Spain also shows that, in contrast to the hypothesis developed in the literature (Döring 2004, Döring and
253
Hallerbert 2004, Treutlein 2007) variations in the use of secondary legislation for the transposition of EU directives are not explained by the existence of minority governments. Overall, the EU contributes to reducing the Spanish Parliament’s already limited capacity to control the legislative agenda, confirming to a certain extent the arguments of the authors who support the idea that Europeanization involves a “de-parliamentarization” process of national political systems (e.g. Moravsick 1994, Hix 1999, Maurer and Wessels 2001, Chryssochoou et al. 1998, Jachtenfuchs 2001). Our results provide evidence that parliamentary control over EU affairs is low compared to the percentage of Europeanized legislation. Even in highly Europeanized policy areas such as the environment, the percentage of oral questions with EU content in the various parliamentary venues is low. In contrast, the analysis of amendment rates suggests that a high percentage of laws with EU content are submitted to control by parliamentary groups, since the amendment rate is higher for laws related to the EU than for the rest. Further research into the origins and motivations of these amendments is necessary to confirm whether the Spanish Parliament exercises strong control over the implementation of EU law.
254
References Álvarez Miranda, B. (1996): El Sur de Europa y la Adhesión a la Comunidad (Madrid:Centro de Investigaciones Sociológicas). Anderson, Ch. J. (1998) “When in Doubt, Use Proxies: Attitudes toward Domestic Politics and Support for European Integration”, Comparative Political Studies, 31:569-601. Arregui, J. (2007): “Europeización y transformación institucional del sistema político español”, in F. Morata and G. Mateo (eds.) España en Europa-Europa en España (1986-2006), Fundació CIDOB: Barcelona, pp. 123-148. Baumgartner, F. and B. D. Jones (1993): Agendas and Instability in American Politics (Chicago: University of Chicago Press). Besabe, Ll. F. and M.T. González (2001): “The Parliament of Spain: slowly moving onto the European direction?” in A. Maurer and W. Wessels (eds.) National Parliaments on their Ways to Europe: Losers or Latecomers?, Baden-Baden: Nomos Verlag, pp. 199-222. Börzel, Tanja A. (2000): “Why there is no Southern Problem. On Environmental Leaders and Laggards in the EU”, Journal of European Public Policy 7 (1): 141–162. Capo (2003): “The Spanish Parliament in a triangular relationship”, Journal of Legislative Studies, 9 (2): 107-129. Cienfuegos, M. (2001): “Las Cortes Españolas ante la Integración Europea”, en C. Closa (ed.) La Europeización del sistema politico español (Madrid:Istmo). Closa, C. (1995): “Spain: the Cortes and the EU. A growing together”, Frank Cass Journals, pp. 136-150. Closa, C. (ed.) (2001): La Europeización del sistema politico español (Madrid:Istmo). Closa, C. and P.M. Heywood (2004): Spain and the European Union (New York: PalgraveMacmillan). Chaqués, L. (2002): Estructura y Política farmacéutica (Madrid: Centro de Investgaciones Sociológicas).
255
Chaqués, L. and A. M. Palau (2011) “Explaining issue competition in Spain: a comparison of law production at sub-national level from 1982 to 2007”, Comparative Political Studies, (under review) Chryssochoou, D., S. Stavridis, and K. Infantis (1998): Theory and reform in the European Union (Manchester: Manchester University Press). Díez Medrano, J. (2003): Framing Europe: Attitudes to European Integration in Germany, Spain, and the United Kingdom (Princeton, NJ: Princeton University Press). Díez Medrano, J. (2007): “La opinion pública española y la integración europea”, in F. Morata and G. Mateo (eds.) España en Europa-Europa en España (1986-2006), Fundació CIDOB: Barcelona, pp. 205-233. Döring, H. (2004): “Controversy, Time Constraint and Restrictive Rules”, in H. Döring and M. Hallergerg (eds.) Patterns of Parliamentary Behaviour (Aldershot: Ashgate Publishing Limited). Falkner, G., O. Treib, M. Hartlapp and S. Leiber (2004) Complying with Europe. EU Minimum Harmonisation and Soft Law in the Member States (Cambridge: Cambridge University Press). Flynn, B. (1997) ‘Subsidiarity and the rise of “soft law” in EU environmental policy: beyond who does what, to what it is they actually do?’, The European Policy Process, The Human Capital and Mobility Network Final Workshops, University College Dublin, Occasional Paper No. 40. Font, J. and E. Rodriguez (2006): “The Spanish Referendum on the EU Constitution: Issues, party cues and second order effects”. Paper presented at the 64th Annual Meeting of the Midwest Political Science Association, Chicago, 20-23 april. Goetz, K.H. and J. Meyer-Sahling (2008): “The Europeanisation of domestic political systems”, Living Reviews in European Governance, June. Hix, S. (1999): The political system of the European Union (Basingstoke:Macmillan). Hooghe, L. and Marks, G. (2005): “Calculation, Community and Cues: Public Opinion on European Integration”, European Union Politics, 6: 419.
256
Jacobsson, K. (2004): “Soft regulation and the subtle transformation of states: the case of EU employment policy”, Journal of European Social Policy, 14: 355-70. Jachtenfuchs, M. (2001): “The Governance Approach to European Integration”, Journal of Common Market Studies, 39:2. Jenny, M. and Müller, W. C. (2009): “From the Europeanization of law-making to the Europeanization of national legal orders: the case of Austria”, Paper presented for the WZB conference on “Measuring the Europeanization of Public Policies Beyond the 80% Myth”, Berlin, 27 – 28 February 2009. Jones, B. D. and F. R. Baumgartner (2005): The Politics of Attention (Chicago: University of Chicago Press). Kaeding, M. (2005): “Determinants of transposition delay in the European Union. The member state perspective”. University of Leiden working paper. König, T. (2009): “Measuring the Europeanization of lawmaking in Institutional context: from executive dominance to parliamentary scrutiny?”, Paper presented for the WZB conference on “Measuring the Europeanization of Public Policies Beyond the 80% Myth”, Berlin, 27 – 28 February 2009. König, T., B. and B. Luetgert (2008): “Troubles with transposition? Explaining trends in member state notification and the delay transposition of EU Directives”, British Journal of Political Science, 39: 163-194. Kritzinger, S. (2003): “The influence of the Nation-State on Individual Support for the European Union”, European Union Politics, 4 (4): 219-241. Lombardo, E. (2003): “La Europeización de la política española de igualdad de género”, Revista Española de Ciencia Política, 9, 65-82. López-Santana, M. (2006) “The domestic implications of European soft law: framing and transmitting change in employment policy”, Journal of European Public Policy 13(4): 481–99.
257
Magone, J. M. (2007): “South European national parliaments and the European Union. An inconsistent reactive revival, in J. O. Brennan and T. Raunio (eds.), National Parliaments within the enlarged European Union. From victims of integration to competitive actors? (London: Routledge), pp. 116-31. Mastenbroek, E. (2003) ‘Surviving the Deadline: The Transposition of EU Directives in the Netherlands’, European Union Politics 4:(4), 371–95. Matia, F. J. (1999): Parlamentos Nacionales y Derecho Comunitario (Madrid: Centro de Estudios Políticos y Constitucionales). Maurer, A. and W. Wessels (2001): National Parliaments on their ways to Europe: losers or latecomers? (Nomos, Baden-Baden). Melsaether, J., and U. Sverdrup (2004) “The Parliamentary Challenge in the EU and the EEA: An Increasing Gap”, ARENA Working Papers, No. 17, ARENA Centre for European Studies, Oslo. Molina, I. (2001): “La adaptación a la Unión Europea del poder ejecutivo español”, en Closa, C. Ed. La Europeización del sistema politico español (Madrid:Istmo). Montabes, J. and I. Szmolka (2000): “Los programas electorales”, in A. Martínez and M. Méndez (eds.), Las elecciones al Parlamento Europeo, 1999 (Valencia: Tirant lo Blanch). Morata, F. (1987): Autonomía regional i integració europea (Generalitat de Catalunya: Institut d’Estudis Autonòmics). Morata, F. (1998): “Spain: modernization through integration”, in K. Hanf and B. Soetendorp (eds), Adapting to European integration: small states and the European Union (London: Longman), pp. 100-15. Morata, F. (2004): Gobernanza Multinivel en la Unión Europea (Valencia: Tirant lo Blanch). Morata, F. and Mateo, G. (2007): España en Europa-Europa en España (1986-2006), Fundació CIDOB: Barcelona.
258
Moravsick, A. (1994): “Why the European Union Strengthens the State: Domestic Politics and International Cooperation”, Center for European Studies, Working Paper Series 52. Müller, W.C.; M. Bovens; J.G. Christensen; M. Jenny; and K. Yesilkagit (2010): “Legal Europeanization: comparative perspectives”, Public Administration, 88: 75-87. Olsen, J. P. (2007): Europe in Search of Political Order (Oxford: Oxford University Press). Olsen, M. (1980): The legislative Process: A Comparative Approach (New York: Harper & Row). Ordoñez, David (1994): La ejecución del Derecho Comunitario Europeo en España, Cuadernos de Estudios Europeos (Madrid: Civitas). Palau, A. (2009): “The Europeanization of the Spanish Food Safety Policy: framing and institutional change and its relation to European soft law”, Paper presented at the ECPR General Conference in Postdam (September 2009). Powell, Ch. (2002): “Spanish membership of the European Union revisited”, Working Paper 2002/02, Real Instituto Elcano de Estudios Internacionales y Estratégicos, 17 de junio. Raunio, T. (2005): “Holding governments accountable in European affairs: explaining crossnational variation”, Journal of Legislative Studies, 11 (3/4),319-42. Saalfeld, T. (2005): “Deliberate delegation of abdication? Government backbenchers, ministers and European Union Legislation”, Journal of Legislative Studies, 11 (3-4), 343-71. Sánchez, I. (1999): “El déficit democrático en la Unión Europea”, en I: Llamazares y F. Reinares (eds.), Aspectos políticos y sociales de la integración europea (Valencia: Tirant lo Blanch), pp. 94-116. Steenbergen, M. and J. Bradford (2002): ‘Modeling Multilevel Data structures’ American Journal of Political Science 46: 218–37. Storini, C. (2005): Parlamentos Nacionales y Unión Europea (Valencia: Tirant lo Blanch)
259
Subirats, J.; Q. Brugue and R. Goma (2001): “Multilevel governance and Europeanization” in K. Featherstone-G. Kazamias (eds.) Europeanization and Souther Periphery (Frank Cass, London), pp. 95-118. Szmolka, I. (1999): Opiniones y actitudes de los españoles ante el proceso de integración europea, Madrid: CIS. Szmolka, I. (2007): Veinte años de pertenencia de España a la Unión Europea: actitudes de los españoles ante el proceso de integración comunitaria (Madrid: CIS). Töller, A. E. (2006): “How European Integration Impacts on National Legislatures: The Europeanization of the German Bundestag”, Center for European Studies Program for the Study of Germany and Europe, Working Paper Series 06.2. Torreblanca, J. I. (2001): La Europeización de la política exterior española (Madrid: Istmo). Treutlein, D. (2007): “What actually happens to EU Directives in the member states? A cross.country cross-sector view on national transposition instruments”, CESifo Working Paper, 2098. Treutlein, D. (2009): “Zooming in on Transposition: National Execution Measures for EU Directives 1986-2002”, CESifo Economic Studies, 55 (1): 94-109. Wiberg, M. (ed.) (1997): Trying to make democracy to work: the Nordic parliaments and the European Union (Stockholm: Gidlunds).
260
Tables and Figures Table 1. Issue Topics of the Policy Agendas Project
1.Macroeconomics 2. Civil rights 3. Health 4. Agriculture 5. Labor 6. Education 7. Environment 8. Energy 10.Transportation 12.Internal affairs and justice 13. Social Policy 14. Housing 15. Commerce and industrial policy 16. Defense 17. Science and Technology 18. Foreign Trade 19. International Affairs 20. Government and Public Administration 21. Public Lands and water management
261
Table 2. Number of laws and proportion of the different legal status of laws per year (1986-2007) Ordinary Laws Organic Laws Decree-Laws Legislative-Decrees Total Year total % total % total % total % 1986 25 52 4 8 3 6 16 33 48 1987 34 69 7 14 7 14 1 2 49 1988 44 75 7 12 7 12 1 2 59 1989 20 65 3 10 7 23 1 3 31 1990 32 78 0 0 6 15 3 7 41 1991 31 74 5 12 5 12 1 2 42 1992 44 77 6 11 6 11 1 2 57 1993 23 50 0 0 22 48 1 2 46 1994 42 63 10 15 13 19 2 3 67 1995 45 61 16 22 11 15 2 3 74 1996 15 44 1 3 17 50 1 3 34 1997 65 65 6 6 29 29 0 0 100 1998 50 65 7 9 20 26 0 0 77 1999 56 63 10 11 22 25 1 1 89 2000 16 42 7 18 10 26 5 13 38 2001 26 53 6 12 16 33 1 2 49 2002 53 71 11 15 10 13 1 1 75 2003 62 70 20 22 7 8 0 0 89 2004 3 12 3 12 11 44 8 32 25 2005 32 60 5 9 16 30 0 0 53 2006 43 69 6 10 13 21 0 0 62 2007 60 73 9 11 11 13 2 2 82 Total 821 149 269 48 1287
Table 3. Key words used for the analysis of Europeanization 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
European Union European Community European economic Community Common Market Single Market European Market European Coal and Steel Community European Atomic Energy Community European monetary Union European monetary System
11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
EU EEC EC ECSC EAEC Euratom EMU EMS Directive Community law European law
262
Table 4. Proportion of Europeanized laws per year in Spain (1986-2007)
Year 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Total/mean
% Ordinary laws Europeanized
% Organic laws Europeanized
% Decree-Laws Europeanized
24 12 18 25 25 32 27 39 36 40 33 20 46 38 31 42 49 61 67 53 53 58 38
25 14 0 0 0 0 17 0 10 6 0 17 29 0 0 33 27 30 33 40 33 11 15
0 29 29 14 17 40 17 14 15 36 35 21 35 27 60 38 30 71 45 31 23 73 32
% LegislativeTotal % Decrees Laws Total Europeanized Europ. Passed 100 100 100 100 0 0 0 0 0 0 100 0 0 0 40 0 100 0 87.5 0 0 50 35
23 8 11 7 9 12 14 12 18 23 12 20 32 27 13 19 33 49 15 24 28 45 454
48 16 19 23 22 29 25 26 27 31 35 20 42 30 34 39 44 55 60 45 45 55 35
263
Table 5. Laws Europeanized with and without amendments*
1987 1988 1989 1990 1991 1992 1993 period 1987-1993 1994 1995 1996 1997 1998 1999 2000 period 1994-2000 2001 2002 2003 2004 2005 2006 2007 period 2001-2007 Average
Non Europeanized % without % with amendments amendments 33 67 35 65 28 72 50 50 15 85 32 68 50 50
Europeanized % without % with amendments amendments 40 60 13 88 0 100 13 88 30 70 15 85 33 67
34
66
21
79
17 31 9 51 53 24 33
83 69 91 49 47 76 67
6 5 0 21 40 14 0
94 95 100 79 60 86 100
34
66
17
83
47 23 13 0 6 8 3
53 77 87 100 94 92 97
15 0 9 0 16 8 6
85 100 91 100 84 92 94
15
85
8
92
29
71
13
87
* Includes organic and ordinary laws
264
Table 6. Legislation passed and % Europeanized by policy area (1986-2007) Policy Area
Total laws passed
Macroeconomics Rights Health Agriculture Labor Education Environment Energy Transport Internal Affairs Social Policy Housing Commerce Defense S&T Foreign Trade International Affairs Government Affairs Public Lands Total
206 40 40 60 82 54 38 23 79 142 16 18 169 43 41 9 61 120 46 1287
%/total passed 16 3 3 5 6 4 3 2 6 11 1 1 13 3 3 1 5 9 4 100
% Europeanized 56 20 33 63 26 22 63 52 22 12 25 11 50 2 59 56 25 26 20
%/total Europeanized 26 2 3 8 5 3 5 3 4 4 1 0 19 0 5 1 3 7 2 100
1987 1988 1989 1990 1991 1992 1993 period 1987-1993 1994 1995 1996 1997 1998 1999 2000 period 1994-2000 2001 2002 2003 2004 2005 2006 2007 period 2001-2007 Average
Table 7. Adoption and amendment rate of governmental bills Non Europeanized Europeanized % without % with % % % without % with amendment amendment no passed no passed amendments amendments s s 33 67 0 100 0 0 38 63 0 25 75 0 14 40 46 0 67 33 46 50 4 0 100 0 21 79 0 33 67 0 19 57 24 15 54 31 36 29 36 0 57 43 29
55
16
17
61
22
9 20 6 48 49 20 26
84 55 76 43 41 65 74
6 24 18 9 11 15 0
10 7 0 14 36 9 0
90 86 100 86 64 64 100
0 7 0 0 0 27 0
28
59
13
16
78
7
38 17 13 0 11 9 7
59 83 88 83 89 67 89
3 0 0 17 0 24 5
0 0 5 0 20 6 0
100 100 90 100 80 94 96
0 0 5 0 0 0 4
15
80
5
5
93
2
24
65
11
11
81
8
266
Table 8. Laws: amendment rate by policy area (1987-2007)*
% amended % legislation % legislation legislation/total amended amended non Policy area passed Europeanized Europeanized Macroeconomics 70 71 29 Rights 84 25 75 Health 80 35 65 Agriculture 79 77 23 Labor 88 27 73 Education 72 32 68 Environment 89 88 13 Energy 67 70 30 Transport 51 30 70 Internal Affairs 90 12 88 Social Policy 93 23 77 Housing 86 8 92 Commerce 81 67 33 Defense 85 4 96 S&T 93 61 39 Foreign Trade 86 33 67 International Affairs 35 21 79 Government Affairs 82 26 74 Public Lands 77 30 70 Total/Average 78 39 61 * Includes organic and ordinary laws
267
Figure 1. Laws passed and Europeanized (1986-2007) 120
100
Number of laws
80
60
40
20
0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Total laws passed
Europeanized
% Europeanized
Figure 2. Legislation Europeanized by policy area (1986-2007) 250
Number of laws
200
150
100
50
nd s
Pu b
lic
La
irs
irs nm
en
tA ffa
de
Af fa
Tr a
na l G
te rn tio
In
Total laws passed
ov er
S& T
ig n Fo
re
fe ns e
De
in g
m er ce
us
Co
m
ol icy
Ho
irs So cia
lP
rt
lA ffa
sp o In
te rn a
rg y
Tr an
en t
En e
io n En vir o
nm
r
ca t
La bo
Ed u
th
ts
re ltu
al
Ag ric u
He
Ri gh
M
ac ro e
co n
om
ics
0
Total Europeanized
268
Figure 3. Contribution to Europeanization by policy area Macroeconomics
26 19
Commerce Agriculture
8
Government Affairs
7
S&T
5 5
Environment Labor
5
Internal Affairs
4
Transport
4
Interntional Affairs
3
Health
3
Energy
3
Education
3
Public Lands
2
Rights
2
Foreign Trade
1
Social Policy
1
Housing
0
Defense
0 0
5
10
15
20
25
30
%
Figure 4. Oral Questions with EU content raised in Parliamentary Comittees and Plenary Meetings (%/total passed) 25
20
%
15
10
5
0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Parliamentary Committees
Plenary meetings
269
9
0
lA ff a irs So ci al Po lic y H ou si ng C om m er ce D ef en se
14
te rn a
100
sp or t
86
Tr an
91
In
en t En er gy
17
io n
13
nm
15
En vi ro
20 83
ca t
88
r
85
La bo
80
Ed u
Ag ric ul tu re
0
ea lth
ts
ig h
%
100
H
R
ic s
co no m
ac ro e
Total raised
100 100
88 89
13
11
0
0
Without amendments
S& Fo T re ig n In Tr te ad rn ti o e na G l ov Af er fa nm irs en tA ffa irs Pu bl ic La nd s
M
95
5
13
6
G
67
33
94
87
lic
Pu b
La
nd s
irs
irs
de Af fa
Tr a
S& T
tA ffa
en
nm
ov er
na l
io
ig n
re
Fo te rn t
In
Tr an sp or te t rn al Af fa irs So cia lP ol icy Ho us in g Co m m er ce De fe ns e In
io n
en t En er gy
nm
En vir o
r
La bo ca t
Ed u
th
ts
re
al
ltu
Ag ric u
He
Ri gh
om ics
M ac ro ec on
total passed
Figure 5. Oral Questions with EU content raised on Parliamentary Committees and Plenary Meetings
4000
3500
3000
2500
2000
1500
1000
500
0
With EU content
Figure 6. Amendment rate of Europeanized legislation by policy area (1987-2007)
120
100 91 100
60 69
50
40
31
0
9
0
With amendments
270
Figure 7. Percentage of primary and secondary legislation passed to transpose EU Directives (1986-2007)
120,0
100,0
100
99 93
99 92
97
96 91
96 89
87
90
98
98
91 87
86
89
89 83
81
80,0
%
68 60,0
40,0 32
20,0
19 13 9
0,0
17 11
10
13 9
14
11
11
8 7 4 4 3 2 2 1,4 1 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Primary Legislation
Secondary Legislation
271
Primary legislation
G
ts
Fo
re ig n
lic
Pu b
La
nd s
irs
irs
de Af fa
Tr a
tA ffa
en
nm
ov er
na l
S& T
sp In or te t rn al Af fa irs So ci al Po lic y H ou si ng C om m er ce D ef en se
rg y
En e Tr an
io n
r
en t
ca t
La bo
nm
Ed u En vi ro
te rn tio
In
ea
lth Ag ric ul tu re
H
ic s
ig h
om
R
co n
ac ro e
M
%
Figure 8. Transposition of Directives by policy area and type of legislation (primary vs secondary)
120
100
80
60
40
20
0
Secondary legislation
272
11. So close, yet so far? The EU footprint in Swiss legislative production Roy Gava & Frédéric Varone University of Geneva
1. Introduction1
What proportion of Swiss legislation is influenced by the European Union? This question represented the essence of a parliamentary intervention in 2006 that set out to establish an official and systematic identification of Europeanized domestic legislative acts2. The Government dismissed this proposal3 and admitted that such information was not available4. This chapter embarks on the investigation of this question by undertaking an exploratory quantitative assessment of the Europeanization of Swiss legislative production during the period 1987-2007. Although it is neither a member of the European Union (EU) nor of the European Economic Area (EEA), Switzerland is undeniably a case worth considering for students of Europeanization. Firstly, besides occupying a major and recurrent place on the foreign policy agenda, the process of European integration has increasingly affected domestic politics and policies. Swiss citizens have recurrently been summoned to vote on issues related to European integration (Schwok, 2006). In a country characterized by the importance and frequent use of its direct democratic instruments, the European question is not left unnoticed at the ballot box: the referendum on the EEA membership
1
This chapter is an output of the project Agenda Setting in Switzerland funded by the Swiss National Science Foundation (ref. 105511-119245/1), and part of the EUROCORES project The Politics of Attention: West European Politics in Times of Change, funded by the European Science Foundation. We are grateful to Nino Landerer and Lisa Marx for their help in the data collection process. A previous version of this chapter was presented at a Staff meeting of the Department of Political Science at the University of Geneva. We are grateful to the participants, in particular René Schwok, for their valuable comments. 2 Nordmann, Roger. Adaptation sur une base volontaire du droit suisse aux normes européennes. Rapport et mise en évidence dans le RS. Postulate 06.3839, December 20th, 2006. 3 Federal Council. Réponse du Conseil Fédéral, Postulate 06.3839, February 14h, 2007. 4 Federal Council. Rapport du Conseil fédéral sur les effets de divers instruments de politique européenne sur le fédéralisme de la Suisse, p. 12, June 15th 2007.
of 1992 still holds the sixty year record for voters' participation in a popular vote and was preceded by what constituted "probably the most intensive campaign in the history of Swiss democratic voting" (Kriesi, 2005: 116). In the case of Switzerland, the question of European integration represents a salient issue, which is repeatedly mobilised in partisan competition (Kriesi et al., 2008). Secondly, by providing variation on a key variable, such as the institutional link with the EU, Switzerland emerges as a potentially valuable “control case” for cross-national inquiries aiming to trace the net impact of the EU on national political systems and outcomes (Bartle, 2006; Haverland, 2007; Fontana et al., 2008). Several empirical studies have analyzed the impact of EU and international factors on Switzerland's decision-making processes and public policies (for an overview see Fischer, 2007). For instance, studies of internationalization have already offered insights regarding the transnational influence on decision-making processes (Sciarini et al., 2002). Legal scholars have discussed the ways in which EU norms influence the Swiss legal order (Epiney, 2009; Kohler, 2009; Mallepell, 1999). Attention has also been devoted to the question of whether EU membership would have limited direct democratic institutions and the celebration of popular votes (Kellerhals, 1999). Nevertheless, the net traces of the EU in the overall legislative production across time and policy domains remains unexplored. By means of a content analysis of official documents, this chapter performs a quantitative assessment of explicit textual references to the process of European integration in Swiss legislative proposals and acts over the past two decades. In this contribution, manifest references to the EU which are observable in official texts are referred to as an "EU footprint". The sample of documents retained for analysis includes both the legislative inputs and outputs of the Swiss Parliament during the past five legislatures. This chapter deals, on one hand, with reports presenting legislative proposals, whether they are put forward by the Government, the Parliament or by citizens and, on the other hand, with constitutional amendments, federal laws and federal decrees enacted from 1987 to 2007. The extent to which legislative inputs and outputs carrying an EU footprint compare to the ones which are devoid of it, is therefore explored in documents of political and juridical nature. Finally, a closer look is devoted to the fate of legislative outputs in the referendum phase. 274
In contrast to former contributions dealing with the Europeanization of Swiss politics, the present chapter benefits from the adoption of a relatively extended time-frame. The comparison across legislative periods becomes particularly appealing when taking into account the evolution of the process of European integration and the changing nature of Swiss-EU relations across time. Additionally, in comparison to earlier studies focusing on the impact of supranational and international factors on domestic legislative activity, the EU is in this case isolated from other transnational actors or arenas. Although previous studies have dealt with the influence of the EU on domestic legislative processes, this chapter offers a glance at both the input and output of legislative production, while simultaneously benefiting from a larger number of observations. Last but not least, the application of a common coding scheme of policy sectors of legislative activity and EU references, offers a rare opportunity to effectively put the Swiss situation in comparative perspective with EU member states. This chapter is structured as follows. Section 2 provides an overview of the relations between Switzerland and the EU. Section 3 introduces the reader to the different types of legislative proposals and legislative acts under study in this contribution. Section 4 briefly reviews previous studies on the extent to which the Swiss legislative production has been influenced by EU law. Section 5 presents the data and discusses methodological issues. Section 6 presents the descriptive statistics of the Swiss case. Finally, the conclusion puts forward a broad discussion based on the empirical findings.
2. Switzerland and the EU
The title of a recently published book (Church, 2007) depicts, in a nutshell, the relationship between Switzerland and the process of European integration since the early days of this supranational construction: it is “close, contradictory and generally misunderstood”. As a small country landlocked in Western Europe, the process of European integration has remained a longstanding and central issue in Swiss politics. On the one hand, given its profound links with EU member states, the prospect of economic and political isolation does not represent a sustainable alternative for the 275
Swiss European policy. As a matter of fact, the EU is Switzerland's most important commercial partner: the EU is the destination of 60% of the Swiss exports and the origin of 80% of its imports5. On the other hand, the country has proved recurrently reluctant to the option of EU membership. Although over time there has been some variation in its position vis-à-vis the EU, it could be stated that Switzerland has been following a particular “third way” towards European integration: while dismissing full participation, it has tried to avoid marginalization (Church, 2007; Dupont and Sciarini, 2001; Schwok, 2006).
2.1 The Swiss "Third Way" Switzerland is a founding member of the European Free Trade Association (EFTA). Under its framework, Switzerland concluded a Free Trade Agreement (FTA) for manufactured goods with the European Community (EC) in 1972. This treaty became the cornerstone of the Swiss-EC relationship for the following fifteen years, and was domestically considered as an advantageous deal for Switzerland, given that the FTA allowed the country to benefit from high economical gains at low political costs. However, during the second half of the 1980s developments at the external front dramatically changed this situation (Dupont and Sciarini, 2001). With a new impulse given to the European integration process, particularly by the 1985 Single European Act and the prospect of the new Common Market, a series of considerable economic discrimination threats against EFTA countries arose. Jacques Delors' proposal of an EEA in 1989, based on the "four EC freedoms" (goods, capital, services and people), between EC and EFTA countries eventually gave the latter the possibility to participate in the incipient Single Market. However, the framework of the EEA did not grant EFTA members rights of co-decision over the future development of the internal market legislation. Faced with this situation, the Swiss Government (Federal Council) signed the EEA treaty and, along with other EFTA countries, applied in May 1992 for full accession to the EC.
5
Figures from the Swiss Federal Department http://www.eda.admin.ch. Data accessed on June 10th 2009.
of
Foreign
Affairs,
available
at:
276
In accordance with the Swiss Constitution, entry to the EEA required a mandatory referendum. On December 6th 1992, a day which became known as the "Black Sunday", the Swiss citizens rejected in a popular vote, and in contradiction to their Government’s position, joining the EEA. Through this refusal, Switzerland found itself with a disadvantageous access to the single market, when compared to its EFTA partners. Since taking up membership of either the EEA or the EU has proved unviable due to domestic opposition, the Government has pursued a strategy consisting of the completion of “sectoral bilateral agreements” with the aim of countering economic isolation since early 1993 (Dupont and Sciarini, 2001). In parallel, this policy of bilateral international treaties has since been complemented with cases of "autonomous adaptations" (i.e. without any formal constraint from the EU) of the Swiss legal order to European standards. Negotiations over a first set of bilateral treaties between Switzerland and the EU began shortly after the Swiss EEA rejection and lasted four years. Seven different policy sectors and issues (agriculture, public procurement markets, technical barriers to trade, research, civil aviation, road transportation and the free movement of persons) were included in the negotiations, and were bound by a "guillotine clause"6. Two negotiations proved particularly salient and difficult both at the external and domestic levels: road transportation and the free movement of persons. Nevertheless, the seven treaties were finally jointly signed in 1999 and ratified in 2000 by the Swiss citizens (Dupont and Sciarini, 2007). In 2002, talks over a second round of bilateral negotiations started and comprised nine issues (taxation of savings, fight against fraud, processed agricultural products, media, education, pensions of EU officials, statistical data collection and the Swiss adhesion to the Schengen/Dublin Area and the European Environment Agency). This time three issues proved difficult to agree upon: taxation of savings, the fight against fraud and Schengen/Dublin. The agreements were ratified separately by the Parliament in 2004. A referendum was launched against the Schengen/Dublin agreement, but was finally accepted by the citizens in 2005 (Afonso and Maggetti, 2007). 6
While the first set of bilateral agreements was formed by seven individual treaties, the EU imposed a "parallelism of the talks" over the negotiation and ratification of processes (Dupont and Sciarini, 2007). This set of treaties "could only enter into force collectively and the non-prolongation or cancellation of each single one would result in a concomitant cancellation of all other agreements" (Epiney, 2009: 185).
277
Switzerland and the EU resumed bilateral talks in 2008. This time the four fields retained for negotiation included free-trade in the agricultural and food sector, food safety, consumer products safety, and cooperation in the area of public health7. At present these different negotiations are still open and ongoing. The two packages of bilateral agreements concluded between Switzerland and the EU in 1999 and 2004, in conjunction with the 1972 FTA and a 1989 bilateral agreement on insurances, represent not only the basis of the relationship between the two: they are additionally clear examples of the Europeanization of different policy domains of the Swiss domestic legislation.
2.2 Direct and Indirect Europeanization Europeanization is considered to exist in two forms in Switzerland (Sciarini et al., 2004): A direct Europeanization mechanism is at work when Switzerland negotiates bilateral or international agreements with the EU entailing a domestic change. But Europeanization may also occur in a more indirect way, whenever Switzerland (as a non EU-member state) unilaterally adapts its domestic rules to existing EU norms. This second mechanism is generally termed "autonomous adaptation" (autonomer Nachvollzug) by the Swiss political elites. An example of the unilateral adaptation of Swiss legislation in the absence of bilateral agreements with the EU is the adoption of the so-called "Swisslex" legislative package. In order to comply with the requirements of the EEA membership in 1992, the Government prepared a package including the modification of 70 federal laws and the creation of 12 new legislative acts in different domains which became known as "Eurolex"8. With the popular rejection of the Switzerland's membership to the EEA in 1992, Eurolex was ultimately not enacted. Nevertheless, the Government decided to enhance compatibility with European legislation and proposed a less ambitious package, this time re-named “Swisslex”. Swisslex retained 27 of the original Eurolex
7
Integration Office, Federal Department of Foreign Affairs. http://www.europa.admin.ch/, last accessed on October 22nd 2009. 8 Federal Council. Message I sur l'adoption du droit fédéral au droit de l'EEE (Message complémentaire I au message relatif à l'Accord EEE), p. 8. May 27th, 1992.
278
modifications and was adopted by the Parliament in 19939. No referendum has been launched against these modifications. Some other important domestic policy processes and designs that have been identified by the literature as being indirectly Europeanized include, for example, the liberalization of the telecommunications (Fischer et al., 2003; Sciarini et al., 2004) and electricity markets (Fontana et al., 2008) as well as regulatory policies in the economic sector (Mach et al., 2003; Mach, 2006). Furthermore, as early as 1987, the Federal Council introduced a "euro-compatibility" assessment in its reports accompanying and explaining Government sponsored bills to the Parliament. From that point onwards, these reports have included a section that explicitly assesses the adjustment of the new piece of legislation with the existing European legal order. This originally voluntary practice by the Government was institutionalized and eventually incorporated in 2002 within the federal law regulating the relations between the Executive and the Legislative10. Consequently, regardless if the outcome is "euro-compatible" or not, most of the legislative processes already incorporate a "European reflex"11. This "European reflex" of the legislator is supposed to exert its influence to the extent that "the eurocompatibility of Swiss legislation is the rule and the adoption of a legislative act which is not in conformity with European Community Law is the exception for which good reasons must be given" (Epiney, 2009: 181).
3. Legislative production in Switzerland
3.1 Legislative proposals In terms of legislative impetus, and in a comparative cross-national perspective, the Swiss political system can be depicted as being open. A wide array of political actors can initiate the decision-making process at the Federal level (Sciarini, 2006). When 9
Federal Council. Message sur le programme consécutif au rejet de l'Accord EEE. February 24th,
1993. 10
See the art. 141, al 2, let a. of the Law on the Federal Assembly. See also the Federal Council answer to the Parliamentary Question Swisslex II (98.1032), May 13th 1998. 11
279
strictly speaking of the presentation of legislative proposals for the Parliament's consideration, three key actors can draft and prepare new bills: the Government, the Parliament and the citizens12. In Switzerland, where pre-parliamentary negotiations and consultation procedures are considered particularly important, most of the bills are drafted and prepared by the Government. Besides its own legislative proposals, the Federal Council is in charge of the drafting of bills triggered by the means of a parliamentary motion, an instrument by which the Parliament constrains the Government to propose legislation related to a given issue. Legislative proposals prepared by the Government are submitted to the Parliament and generally accompanied by an explanatory report, known as message du Conseil Fédéral, which supports and presents bills before the Parliamentary chambers. Although in quantitative terms the lion's share of legislative proposals are prepared by the Government, the Parliament remains responsible for bill drafting when legislative processes are initiated by a parliamentary initiative (Lüthi, 2006). In these cases, a parliamentary committee manages the equivalent pre-parliamentary stage. The committee in charge proceeds to draft bills and prepares a report in which the legislation is presented to the Parliament. Finally, though they are restricted to constitutional amendments, legislative proposals can be drafted and put forward by the citizens through popular initiatives.
3.2 Legislative acts In addition to legislative proposals, this chapter explores legislative production outputs. More precisely, this contribution aims to quantitatively assess references to EU rules in the content of new legislative acts or amendments to these, that are subject to either mandatory or optional referendum. According to Linder (Linder et al., 1985: 23-24; Linder, 2005: 245), governmental and administrative decisions included, this represents approximately 15% of the entire Swiss decision-making output. Nevertheless, it could 12
It should be noted that by the means of a cantonal initiative, cantons can propose legislation at the federal level. If a cantonal initiative is accepted by both parliamentary chambers it follows the procedure reserved for parliamentary initiatives (Law on the Federal Assembly, Art. 117). The four cantonal initiatives that led to a legislative proposal prepared by a parliamentary committee in our sample are then treated as parliamentary proposals in this chapter..
280
be argued that the legislative acts that are subject to one of the two types of referendum cover the most important political decisions (Linder, 2007: 108-109). Our sample of legislative acts consequently includes amendments to the federal Constitution, federal laws, federal decrees and ratifications of international treaties. Table 1 provides an overview of the Swiss legislative production for the period 1987-2007.
Table 1 around here
Amendments to the federal Constitution require a popular vote (mandatory referendum) with a double majority of citizen's votes (more of 50% of the people) and of cantons (12 or more cantons). Besides emanating from the Parliament or Government, proposals to amend the Constitution can also be put forward by citizens, by the means of a popular initiative (100,000 signatures required). If the popular initiative meets all the formal requirements, it is deliberated by the Federal Council and the Parliament, and then presented to the voters with a (generally negative) recommendation. Simultaneously, federal authorities may also submit a counterproposal to the voters. In the presence of such a case, an additional question is put on the ballot, asking the voters which of the two options (i.e. original popular initiative versus counterproposal) they would prefer in the event that both are accepted. At the federal level, legislative acts containing general and abstract rules (federal laws or lois fédérales) are voted by the bicameral Parliament. All federal laws are subject to an optional referendum. As a consequence, in the case of laws, the decision-making process in Switzerland does not end with a positive Parliamentary vote. An optional referendum can be launched once a bill has been adopted by the Parliament, if a minimum of 50,000 citizens or eight cantons ask for a popular vote within 100 days. If the bill submitted to the popular vote fails to obtain a simple majority (more than 50% of the citizens), then it is not enacted. Legislative acts containing other types of parliamentary decisions are referred to as federal decrees (arrêtés fédéraux). Depending on their content, the Constitution stipulates which federal decrees are subject (or not) to one of the two types of referenda 281
(art. 140 and 141). From 1848 until 1999 a specific type of federal decree could also contain time-bounded rules (arrêtés fédéraux de portée générale). However, with the constitutional reform of 1999 this latter category disappeared. These type of rules have ever since been adopted by the Parliament as federal laws (Auer et al., 2000: 500-502). International treaties are ratified by the Parliament by means of a federal decree. In relation to their degree of commitment and their impact on the Swiss legal order, international treaties can be subject (or not) to one of the two types of referenda. The underlying principle can be summarized as follows: the more constraining the treaty, the more demanding the ratification procedure13.
4. EU law and domestic legislative production: previous assessments Different contributions have already investigated the extent to which legislative activity is permeable to supranational developments in Switzerland (Linder et al., 1985; Linder et al., 2009). Poitry (1988: 187) found that 13% of the legislative acts adopted during the years 1971-1975 could be linked to the signature of international treaties, while Sciarini et al. (2002) reported 17% for the period 1995-1999. Furthermore, for this last legislative term, 31% of the total legislative activity is considered to have been "highly" indirectly internationalized (Sciarini et al., 2002: 12-14). A series of studies striving to determine the proportion of EU-influenced domestic legislative production have been feeding the political debate in Switzerland. Although guided by similar starting questions these empirical analysis were carried out with different operationalization and sampling procedures. In consequence, even though the three contributions discussed below relied largely on the textual analysis of eurocompatibility tests included in Federal Council messages, their results are not directly comparable. To the best of our knowledge, the first of these attempts to evaluate the proportion of the domestic legislative activity influenced by EU law was performed by Mallepell (1999). Through the analysis of the tests included in the messages and brief case studies, the 13
However, it should be noted that empirical studies have pointed out that federal authorities have a discretionary room of manoeuvre when interpreting these constitutional rules (Klöti et al., 2005; Hirschi et al., 1999).
282
euro-compatibility of domestic legislative production was assessed with a sample of 47 legislative acts enacted between 1993 and 1995. In the retained sample, 46.8% of the legislative acts were identified as being euro-compatible, while 12.8% were not compatible with EU law. In the remaining 40.4% of the cases, the performed eurocompatibility tests showed no direct links with existing EU norms. In a similar vein, Kohler's study (2009) considers 107 bills enacted between 2004 and 2007. Results are presented in accordance with a typology of different Swiss adaptations to EU law. According to this paper, the official euro-compatibility tests enable the assessment of the link of domestic bills with EU law in 55% of the observations. Partial adaptations to EU law (35.5%) appeared more frequent than complete adaptations (15.9%). Furthermore, cases in which adaptation to EU rules is the main objective of the bill (5.6%) are less frequent than cases of bills that present adaptation to EU law as a marginal objective (15%). Arbia's (2008) contribution has the added value of providing a comparison of the Swiss and Austrian cases, the consideration of parliamentary debates to further grasp evidence of EU influence and the inclusion of a measure of parliamentary conflict by taking into account the result of final votes. The analysis comprehends a sample of 186 legislative acts for Switzerland, enacted between 1996 and 2005. The reported results show that half of the analyzed Swiss laws appear to be either medium or highly Europeanized, with Austria displaying similar figures. Furthermore, in the case of this group of medium and highly Europeanized laws a lower level of parliamentary conflict is reported. The quantitative exploration offered by this chapter does not attempt to evaluate the extent to which the Swiss domestic legislative activity is euro-compatible. Nevertheless, as a distinctive feature from previous studies, it provides an exploration of the proportion in which textual references to the EU can be found in the input and output instances of the overall domestic legislative production.
5. Data and methods
283
In order to assess the presence of an EU footprint in the domestic legislative production, the time-frame retained for analysis encompasses five legislative periods, from the 43rd (1987-1983) to the 47th (2003-2007) Legislature14. With regard to the input of legislative production, all the legislative proposals that have been published in the weekly official Federal journal (Feuille fédérale) and which legislative output has been subject to referendum, are taken into consideration. Units of observation consequently consider: first, reports presenting bills drafted by the Government and presented by the means of Federal Council's messages; second, bills proposed by parliamentary committees in their respective reports; and third, reports regarding popular initiatives proposed by the citizens15. Regarding legislative production outputs, all the legislative acts, as well as their modifications, that have been adopted by the Federal Assembly by the successive Legislatures from 1987 to 2007, and that have been subject to either mandatory or optional referendum, are included as units of observation in our analysis16. In this case data was collected on the basis of the weekly editions of the Official collection of Federal Law (Recueil officiel du droit fédéral). Data on referenda comes from the Swissvotes databank17. Observations on legislative production inputs and outputs have been systematically coded according to their policy domain and the type of EU footprint. The attribution of policy domains has been performed manually in keeping with the Swiss version of the
14
The Agenda-Setting in Switzerland project collected data on legislative activities for the period 19782008. While this chapter focuses on 1987-2007, some references have been made for illustrative purposes to the 41st and 42nd legislative terms. 15 It should be noted that messages and parliamentary reports can propose more than one bill. This helps to understand the lower number of legislative proposals when contrasted to that of legislative acts present in our sample. 16 Given that we do not possess data on the legislative stock at the beginning of the period under study, the analysis is limited to the flow of Swiss legislative production per legislature. On the one hand, the legislative stock makes reference to the total number of legislative acts existing at a given point in time. On the other hand, legislative flows make reference to the frequency of legislative activity (creation, abrogation, modification) during a given time-frame (Linder et al., 1985). 17 Année Politique Suisse / www.swissvotes.ch (2009). Datensatz der eidgenössischen Volksabstimmungen ab 1848. Bern: Institut für Politikwissenschaft, version of November 24th 2009, URL: http://www.swissvotes.ch
284
"Comparative Policy Agendas" topics codebook18. Regarding the type of EU reference, automatic and manual content analysis coding techniques were applied. With regard to automatic coding, through the application of a pre-established keywords dictionary common to the "Delors' Myth"19 project, a dummy variable was automatically attributed to the observations that present one of the pre-defined EUrelated keywords in their text. However, as we have previously pointed out, influence of the EU on the Swiss political system can be either of direct or indirect nature. In order to capture this distinction, a manual coding strategy is most suitable. Concerning its operationalization, the following types of references are considered as being of a "direct" nature: negotiations or results of international sectoral treaties between Switzerland and the EU, situations in which Switzerland depends on the EU's approval to deploy its effects (such as in cases of Swiss participation in EU funded programmes) and instances in which sanctions can be deployed against Switzerland in circumstances of the non respect of EU norms. In contrast, the EU footprint of an "indirect type" includes all other instances of textual references to the EU in official documents. This latter group accounts for cases of explicit "autonomous adaptation" and other allusions to EU rules and standards. The literature has developed a more nuanced distinction of the modes in which the EU has an indirect impact on Swiss legislative activity (Epiney, 2009; Sciarini et al., 2002). For instance, Epiney notes that the question of whether "there is a real autonomous adaptation (of Swiss law) or only a punctual inspiration from European Community Law has to be answered by interpreting the text and the history of the legislative act in question" (2009: 182). Given the extended time-frame retained for analysis and the number of legislative acts under scrutiny, such an in-depth inquiry is however not included in the scope of this chapter. Our discussion is therefore limited to the manifest content of official documents.
18
See the Comparative Policy Agendas project website: http://www.comparativeagendas.org. This project's codebook distinguishes among more than 200 political issues within 20 main categories of public action. 19 In the case of Switzerland, we relied on official documents in French and adapted the keywords dictionary of the "Delors' Myth" project in consequence.
285
The automatic coding applying the "Delors' Myth" keywords and the manual coding turned out to offer largely similar results for our sample20. Given that the distinction between direct and indirect references appears particularly useful for the assessment of the Swiss case, the discussion of the following sections is based on the results of the manual coding. For distinct reasons, the coding of legislative proposals (messages of the Government and reports of parliamentary committees) has been restrained to the evaluation of their summary (condensé). Firstly, we assume that if the EU is considerably linked to the proposed legislation an explicit reference is foreseeable in the report's summary. Secondly, the extensive length of these documents implies that a manual assessment of their full text represents a resource-intense exercise. Thirdly, given that the Federal Council has been including a special section assessing the euro-compatibility of the bills (c.f. section 2 above), the application of a dictionary-based automatic coding on the full text of the reports would offer misleading results. In the case of legislative acts, their entire adopted text has been taken into consideration.
6. The EU footprint and the legislative process
This section presents and discusses the extent to which an EU footprint is identifiable with regard to three successive stages of the Swiss legislative process: the introduction of legislative proposals, the enactment of legislative acts and their fate during the referendum phase.
6.1 Legislative proposals Data displayed in table 2 provide an overview of the legislative proposals containing bills that were subject to referendum during a 20-year period. As is widely 20
For the sample retained for analysis, additional legislative acts carrying an EU footprint have been identified through manual coding in contrast to automatic coding. This is mainly explained by references to the EU terms by the means of EU managed programmes (e.g. INTERREG) or areas (e.g. EEA, Schengen/Dublin). In order to facilitate cross-national comparisons, the results obtained by means of the automatic coding are reported in the annex of this chapter and in the comparative chapter.
286
acknowledged, most of the legislative proposals are introduced by the Government. The Parliament increased its participation in the preparation of bills by the means of parliamentary initiatives across legislatures, a development that corresponded with institutional reforms that aimed at strengthening its position vis-à-vis the Federal Council (for a discussion see Lüthi 2006). Popular initiatives put forward by citizens seem to remain relative stable in terms of overall legislative impetus.
Table 2 around here
Interestingly enough, the Government is not only the main actor responsible for the input of legislative proposals; it also represents the political actor whose figures appear most linked to carrying an EU footprint. The evolution of the proportion of messages with an EU footprint does not seem linear. The peak of EU references in messages is observable during the 45th legislature with 21.43%, a period that was characterized by governmental efforts to counter the negative effects of the citizen's refusal to join the EEA. Finally, during the last decade the EU footprint is observable in roughly 15% of messages from the Federal Council. While the proportion of legislative proposals drafted by parliamentary committees slightly increased across time, the EU footprint is absent in this group of documents. Not a single legislative proposal triggered by a parliamentary initiative and subject to referendum, has been identified as carrying an EU footprint during 1987-2007. As has been previously stated, citizens can only propose constitutional amendments by relying on popular initiatives. Figures show that the overall number of popular initiatives that can be linked to the EU is relatively low. Citizens voted on only four popular initiatives that carried such a manifest reference. These popular initiatives concerned the transport sector and negotiations regarding Swiss membership to the European Union. Interestingly enough, the two popular initiatives that regarded the opening of EU membership negotiations, which would both eventually have been rejected at the polls, were conceived with diametrically opposed political intentions. As
287
it will be shown below, the two remaining popular initiatives in transport policy illustrate the permeability to EU related events of domestic debates in this domain. As previously mentioned, references to the EU in official Swiss documents are considered to be either of a direct or indirect nature. While the cumulative number of bilateral agreements to which documents can be linked increases considerably with the signature of the first (1999) and second (2004) packages of international treaties between Switzerland and the EU, indirect references to the EU are still the most frequent. This outnumbering of indirect EU references is notable throughout the 1990s, and particularly during the 45th legislative term. Consequently, these figures support insights from the literature regarding the strategy of the Swiss Government strategy of "autonomous adaptation" after the 1992 refusal of the EEA membership in the ballot box. Faced with the impossibility of taking part in the Single Market, the Government pursued a series of unilateral legislative modifications that aimed to reduce regulatory discrepancies with EU norms and standards, as an attempt to minimize the costs of non participation for the Swiss economy. With the exception of a reference to the 1972 FTA included in a 1981 message regarding commercial competition regulation, the permeability of Swiss legislative proposals to the process of EU integration can be traced from 1989 onwards. The occurrence of this particular turning point is not surprising since it coincides with the regulatory construction of the Single Market at the EU level and with Swiss expectations to join the EEA.
Graph 1 around here
The presence of the EU footprint on legislative proposals across policies domains is shown in Graph 1. With an average value of 12% of EU references for the period 19872007 as a reference point, several groups of sectors are discernible. At the top of all references linked to the EU appear legislative proposals concerning transports (33.3%) and domestic commerce (29%). The high figures for these two policy domains are largely attributable to indirect references to the process of EU integration. Given the 288
country's geographical situation, lying at the heart of a traffic intense EU north-south axis, the development of transport policy in Switzerland is intimately connected to that of the EU. Legislation with an EU footprint in this policy domain can be linked to measures related to the transportation of goods across the country, the resulting development of infrastructure projects, as well as to the harmonization with EU regulatory standards regarding road, rail and air traffic. The comparable high ratio reported for domestic commerce is attributable to the efforts deployed in promoting a domestic regulatory framework compatible with the EU Single Market. A second group of relatively high permeable sectors consists of foreign trade, energy and science and communications.
The case of foreign trade is predictable as
Switzerland and the EU are linked by international agreements in this domain. Furthermore, the EU is by far the largest trading partner of Switzerland. In the energy sector, the liberalization of the Swiss electricity market has been intimately linked to that of the EU, to the extent of being pointed out as an example of the indirect Europeanization of Swiss policies. Finally, Swiss participation in EU research programs, in hand with the liberalization and regulation of the telecommunications sector, contributes to foster the ratio of EU references in the science and communications category. Legislative proposals with the lower proportion of an EU footprint include policy domains such as defense and health. The rather surprising presence of labor and employment among the least permeable sectors could be related to the use of legislative proposals as the unit of analysis. As it will be seen in the next section, domestic legislative acts in this domain have indeed been reached by the EU footprint, a consequence of the signature of the free movement of person's agreement. This quantitative exploration of legislative proposals provides some fresh empirical support for discussions regarding how the EU is related to Swiss legislative production. First, since the 43rd legislature references to the EU start to be identifiable in this type of documents, in line with the strengthening of the EU integration process and the willingness of Switzerland to participate in the Single Market. Second, actors with control over the drafting process contribute unevenly to the inclusion of EU references in legislative proposals. Government legislative proposals are systematically more inclined than their Parliamentary counterparts to include textually identifiable references 289
to the EU. Third, the considerable contribution of the so-called indirect EU footprint enables to better grasp the alternative ways in which the EU is present beyond the bilateral Swiss-EU lines of co-operation. Fourth, EU influences in Switzerland are deeply unequal across policy domains, with transport and domestic commerce appearing as the most permeable sectors.
6.2 Legislative acts There are remarkable differences across the types of legislative acts carrying an EU footprint (see Table 3). During the period under study the Swiss Constitution incorporated numerous amendments and was subject to major revision in 1999. Not one of these constitutional amendments, independent of whether they emanate from the federal authorities or from the people, had a textual reference to the EU. Two popular initiatives concerning the opening of EU membership negotiations, the 1997 "EC Negotiations Before the People!" and the 2001 "Yes to Europe!", that would have incorporated such an explicit reference, were eventually rejected. Consequently, until 2009 the Swiss Constitution has no explicit references to the EU. In order to trace the EU footprint in the Swiss legislative activity, it seems appropriate to focus on legislative acts of a lower legal hierarchy but which are still subject to referendum: federal laws, federal decrees and ratifications of international treaties.
Table 3 around here
In our sample, federal decrees appear as the type of legislative act that are the least permeable to an EU footprint. Additionally, it should be noted that from 1999 onwards their absolute number decreased dramatically, with the suppression of the arrêtés fédéraux de portée générale. As far as ratifications of international treaties are concerned, their number considerably increased during the last legislature under study. Simultaneously, in this last legislative period the proportion of EU related observations identified was highly influenced by the ratification of seven of the nine different treaties 290
that formed the second package of bilateral agreements between Switzerland and the EU21. Focusing the attention on federal laws provides more fertile soil for identifying textual EU references. The EU footprint in Swiss federal laws increases across time. 1992 represents a turning point from which the text of federal laws became permeable to the EU footprint. The proportion of federal laws carrying an EU footprint has continually progressed, reaching 11.7% for the last legislature under study. The 45th and 47th legislative represented a considerable leap forward (+5.17% and +4% respectively). The tendency of a growing presence of an EU footprint in federal laws is confirmed, regardless of whether we concentrate solely on direct references or whether we extend our focus to indirect EU references. In our sample, after the signature of the different packages of bilateral agreements, several federal laws have been enacted or modified to account for this enhanced and binding co-operation between Switzerland and the EU. The latter logically influences the frequency of the direct EU references found in the text of laws, along with the rising participation of Switzerland in EU programs in different domains.
Graph 2 around here
How does the EU footprint in legislative acts vary across policy domains in the 20-year period between 1987-2007? According to the figures shown in Graph 2, 6.4% of the enacted legislative acts carried some kind of EU footprint in their text. Immigration stands out from other domains, with an impressive 34.78% of its legislative activity carrying an EU footprint. References to an EU footprint are mostly of a direct 21
The 1989 insurance agreement between Switzerland and the EU has not been subject to either mandatory or optional referenda. Therefore, its ratification in the form of a federal decree is not present in the sample. Although it consists of seven different treaties, the first package of bilateral agreements signed in 1999 had the particularity of being ratified by a single federal decree. This was related to the socalled clause guillotine. The second set of bilateral agreements signed in 2004 accounts for 7 international treaties in the sample. The agreements on education and processed agricultural products are not included. From a legal standpoint, the agreement on education has been endorsed through an official letter exchange. Concerning the agreement on processed agricultural products, it was ratified by a federal decree which was not subject to an optional referendum.
291
nature in this sector, as a result of the legal modifications prompted by the bilateral agreements on the free movement of people and the inclusion of Switzerland in the Schengen and Dublin areas. Besides the outlier position for immigration, legislative acts concerning commercial regulation and education appear as relatively permeable to an EU footprint. Furthermore, these two sectors account for most of the indirect EU references identified in the sample. Among bilateral treaties, the free movement of people and Schengen/Dublin agreements account for most of the references in the text of legislative acts. They additionally share the characteristic of deploying effects across different policy domains. For instance, textual references to the agreement on the free movement of people are identifiable in a wide array of policy domains, including employment, education, social affairs, domestic commerce and immigration. In the case of Schengen/Dublin, textual references to this treaty are identifiable in legislative acts related to justice and police issues, in addition to what has already been discussed with regard to the immigration domain.
At the same time it can be noted that legislative acts within some policy sectors remain particularly immune to explicit EU references. The group of least EU permeable areas includes defense, rights and health related issues, among others. The contrasting low figures displayed in the case of energy, when compared to the ones presented in the previous assessment of legislative proposals, provide a clear example of the variation of results obtained depending on whether we focus on the text of legislative proposals or that of legislative acts. This point is further discussed in the conclusion.
6.3 Referenda The previous section concentrates on enacted legislative acts, which became an integral part of the Swiss legal order. By doing this, the discussion is circumscribed to objects that successfully passed the particular and last stage of the Swiss decision-making process: the referendum phase. 292
Table 4 around here
How are referenda on federal laws and international treaties related to the presence of an EU footprint? Between 1987 and 2007 the Federal Assembly adopted 687 federal laws, of which 7% were challenged by an optional referendum and 2% were eventually rejected at the polls. Data displayed in Table 4 suggest that federal laws carrying an EU footprint are more frequently subject to optional referenda. While offering a first overview, interpretations regarding these figures should be regarded with caution given the absence of control variables (such as the weight of the EU reference for the legislative act in question as well as the relative importance of the act itself). Of the 50 optional referenda votes held, 12% were identified as carrying an EU footprint in their text. The Swiss-EU treaty on the free movement of people accounts for 3 of the 4 federal laws on which Swiss citizens voted that carried a direct EU footprint; the other being a modification of the Law on Asylum referring to the Schengen/Dublin Agreement. The remaining two objects concerned indirect references to the EU: the 1998 modification of the Law on Asylum, which included a broad reference to the European policy in this domain; and the 2006 Law on the Cooperation with Eastern European States, related to the Swiss financial contribution to EU enlargement. All the while not carrying an explicit reference to the EU in its text but in its corresponding message, the 2000 Law on the Electricity Market, was rejected at the polls. Regarding the ratification of international treaties, the EU can be linked to the 4 votes held during the period 1987-2007. The 3 optional referenda launched and accepted at the polls were the Schengen/Dublin agreement in 2005, the first set of bilateral agreements in 1999 and the extension of the free movement of people to the ten countries that joined the EU in 2004. The 1992 rejection of the mandatory referendum on the EEA membership offers a supplementary illustration of the limitations of any quantitative assessment of the EU footprint on legislative acts that does include an indicator of their relative importance. If this single legislative act had not been defeated at the polls, the Swiss legal order would have incorporated a considerable share of the acquis communautaire in a wide variety of policy domains through its ratification. 293
In sum, it is difficult to assess with this type of data whether the optional referendum is distinctively activated in the case of federal laws carrying an EU footprint. Nevertheless, it is still worth noting that most of the referenda launched against legislative acts with EU references are related to the cooperation of Switzerland and the EU in terms of borders control and immigration policy22.
7. Conclusion
In both academic and political circles, the figure of 80% of EU members' national lawproduction originating at the European level continues to be echoed without empirical support (König and Mäder, 2008). In the case of Switzerland, which is known to be permeable to Europeanization processes despite its non-member status, the question of the EU influence on the national legal order acquires a particular democratic relevance given the country's lack of participation in EU decision-making institutions. Through content analysis this chapter provides a first quantitative evaluation of the direct and indirect EU footprint in Swiss legislative production during the period 19872007. By relying on messages and reports supporting bills, it has been shown that the Government is the actor most likely to include references to the EU in its legislative proposals. While the EU footprint is identifiable in governmental proposals from the 43rd legislature onwards and across sectors, the few EU related popular initiatives put forward by citizens are grouped in terms of time (1990s) and policy domains (Swiss membership to the EU and transport policy). No legislative proposal whose bill drafting was managed by the Parliament was identified as carrying an EU footprint. Regarding the extent to which EU references are observable in the full text of legislative acts, there is clearly a variation across types of legal objects. On the one hand, among the types of legislative acts retained for study, federal laws and ratifications of international treaties appear as the most permeable to EU references. Between 1987 and 2007, 6.84% of the 673 new or amendments to federal laws that became part of the 22
In 2009 two additional referenda with an EU relation were held. A first one related to the renewal and extension of the free movement of persons agreement to Bulgaria and Romania and a second one related to the development of the Schengen Area (biometric passports). Both legislative acts were accepted by the citizens.
294
Swiss legal order carried an EU footprint. On the other hand, the Swiss Constitution still lacks an explicit reference to the EU. There are no great surprises regarding the different policy sectors in which the EU footprint is observable. Figures largely correspond to the domains for which bilateral agreements currently exist and to different EU programmes in which Switzerland participates. At the same time, the frequency of references to the EU of an "indirect" nature reminds the reader that the EU influence in Swiss legislative production is identifiable even when the EU does not participate as an active actor in "domestic" decision-making processes. Two bilateral agreements, the free movement of people and Schengen/Dublin deserve a special note. First, they account for an important share of the textual references encountered in legislative acts. Second, they both concern border and immigration control, a salient issue in Swiss-EU relations whether we focus on international negotiations or referendum phases. In order to capture the EU footprint in the domestic legislative process we relied on the content analysis of documents representing legislative inputs (legislative proposals in the form of messages and reports) and legislative outputs (legislative acts). It should be noted that both types of documents vary fundamentally in terms of their legal status. In contrast to the juridical nature of legislative acts, messages and reports represent a statement of grounds in which political rhetoric is put forward to support a bill before the Parliament. Given that Switzerland is not an EU member, it is without surprise to observe higher levels of references to the EU in political texts rather than in their juridical counterparts. The Swisslex package provides an illustration in this regard. While the legislative proposal explicitly aimed to unilaterally adapt Swiss norms to that of the EEA, not a single of its 27 corresponding enacted legislative acts carried an explicit EU textual reference. While representing a first assessment of EU influence on Swiss legislative production across legislatures and policy domains, this exploration suffers from a number of limitations. A first group of limitations can be related to the sample of observations retained for analysis. Firstly, we concentrate on the legislative acts that were adopted by the Federal Assembly from 1987 to 2007. This restricts our analysis to legislative flows, 295
therefore ignoring relative impacts of the EU on the overall legislative stock (Linder et al., 1985; Linder et al., 2009). More noteworthy, by focusing uniquely on legislative acts that succeed in the final parliamentary vote, it is impossible to assess the extent of self-imposed limitations and non-decisions (e.g. deliberate decisions of the Parliament in order not to adopt a legislative act that would be incompatible with EU norms). Secondly, while legislative acts subject to either optional or mandatory referendum arguably represent the most important set of legal texts, second order legislation and regulations (Ordinances by the Parliament, the Federal Council or a Federal Department) covering a substantial part of decision-making outputs are out of the scope of this chapter. Some scholars may even highlight the relevance of court rulings (Kälin and Rothmayr, 2006: 188-190) and decisions by independent regulatory agencies (e.g. in liberalized network industries such as telecommunications, energy or railways) for an adequate assessment of the Europeanization of the Swiss legal order (Epiney, 2009). A second set of limitations of this exploration can be linked to other inquiry choices. First, by enabling the consideration of an extended time-frame and number of documents, the EU footprint captured by an exclusive focus on the manifest content of legislative acts is restricted to its explicit form. As a consequence, power games beneath decision-making processes in which EU factors may have played a role either directly or indirectly remained veiled (e.g. differential impacts on political actors, blame-shifting strategies). Second, even if the EU footprint can be identified, what remains missing is an evaluation of the degree of relative importance of this reference within the observations under scrutiny. Third, a measure of the importance of legislative proposals and acts (and of their induced policy change) would represent a valuable indicator for weighting the frequency of EU references. Fourth, while two broad categories of EU influence (direct and indirect) are distinguished here, mixed research designs may contribute to better grasp the ways and the extent to which these influences work (Epiney, 2009; Arbia, 2008; Kohler, 2009). Questions regarding the proportion and mechanisms of the Europeanization of domestic legislative production over time and policy sectors pave the way for a demanding but stimulating research agenda. First, quantitative assessments of supranational impacts on Swiss legislative acts may benefit from disaggregation into identifiable international institutions. Europeanization could then be put in perspective with the impacts of other 296
international actors and arenas over time, such as the Council of Europe, the United Nations and the World Trade Organization. Second, such large-scale assessments may shed
new
light
on
vibrant
academic
debates
on
the
consequences
of
Europeanization/Internationalization processes for domestic actors and institutions. This exploration provides a first step in the assessment of the relative importance of the EU in Swiss legislative production across time and sectors, a question that has proved to be of political and scientific relevance. While methodological issues may arguably continue to fuel debates, what is certain is that a comprehensive and all-encompassing assessment of the Europeanization of Swiss legislative production, as called for by the parliamentary intervention in 2006, remains an open and stimulating challenge.
297
References
Afonso, A. & Maggetti, M. (2007) Bilaterals II: Reaching the limits of the Swiss third way? IN Church, C. (Ed.) Switzerland and the European Union: a close, contradictory and misunderstood relationship. London and New York, Routledge. Arbia, A. (2008) The Road not Taken: Europeanisation of Laws in Austria and Switzerland 1996-2005. Etudes & Travaux. Geneva, Graduate Institute of International and Development Studies. Auer, A., Malinverni, G. & Hottelier, M. (2000) Droit constitutionnel suisse, Bern, Staempfli. Bartle, I. (2006) Europeans outside the EU: Telecommunications and Electricity Reform in Norway and Switzerland. Governance, 19, 407-436. Church, C. (Ed.) (2007) Switzerland and the European Union: A close, contradictory and misunderstood relationship, London and New York, Routledge. Dupont, C. & Sciarini, P. (2001) Switzerland and the European Integration process: Engagement without marriage. West European Politics, 24, 211-232. Dupont, C. & Sciarini, P. (2007) Back to the future: The first round of bilateral negociations with the EU. IN Church, C. (Ed.) Switzerland and the European Union: A close, contradictory and misunderstood relationship. London and New York, Routledge. Epiney, A. (2009) How does European Union law influences Swiss laws and policies. IN Nahrath, S. & Varone, F. (Eds.) Rediscovering public law and public administration in comparative policy analysis. Lausanne and Bern, Haupt and Presses poliytechniques et universitaires romandes. Fischer, A. (2007) Internationalization of Swiss Decision-Making Processes. IN Klöti, U., Knoepfel, P., Kriesi, H., Linder, W., Papadopoulos, Y. & Sciarini, P. (Eds.) Handbook of Swiss Politics. Zurich, NZZ Libro.
298
Fischer,
A.,
Sciarini,
P.
&
Nicolet,
S. (2003)
La
politique
suisse
des
télécommunications: entre pression internationale et résistance nationale. Politiques et management public, 21. Fontana, M.-C., Afonso, A. & Papadopoulos, Y. (2008) Putting the Special Case in Its Place: Switzerland and Small-N Comparison in Policy Research. Swiss Political Science Review, 14, 521-50. Haverland, M. (2007) Methodology. Europeanization: New Research Agendas. New York, Palgrave Macmillan. Hirschi, C., Serdült, U. & Widmer, T. (1999) Schweizerische Aussenpolitik im Wandel: Internationalisierung, Globalisierung und Multilateralisierung. Swiss Political Science Review, 5, 31-56. Kälin, W. & Rothmayr, C. (2006) The Judicial System. IN Klöti, U., Knoepfel, P., Kriesi, H., Linder, W., Papadopoulos, Y. & Sciarini, P. (Eds.) Handbook of Swiss Politics. Zurich, NZZ Libro. Kellerhals, A. (1999) Auswirkungen einer Mitgliedschaft der Schweiz in der Europäischen Union bzw. im Europäischen Wirtschaftraum auf die Ausübung des Referendumsrechts auf Bundesebene. Staatsrechtliche Auswirkungen einer Mitgliedschaft in der Europäischen Union. Conséquences institutionnelles d'une adhésion à l'Union européenne. Zurich, Schulthess Polygraphischer Verlag. Klöti, U., Hirschi, C., Serdült, U. & Widmer, T. (2005) Verkannte Aussenpolitik im Wandel: Internationalisierung, Globalisierung un Multilateralisierung. Swiss Political Science Review. Kohler, E. (2009) Influences du droit européen sur la législation suisse: analyse des années 2004 à 2007. Jusletter. König, T. & Mäder, L. (2008) Das Regieren jenseits des Nationalstaates un des Mythos einer 80-Prozent-Europäisierung in Deutschland. Politische Vierteljahresschrift, 49, 438-463. Kriesi, H. (2005) Direct Democratic Choice, Lanham, Lexington. 299
Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S. & Frey, T. (2008) West European Politics in the Age of Globalization, Cambridge, Cambridge University Press. Linder, W. (2005) Schweizerische Demokratie. Institutionen, Prozesse, Perspektiven, Bern, Haupt. Linder, W. (2007) Direct Democracy. IN Klöti, U., Knoepfel, P., Kriesi, H., Papadopoulos, Y. & Sciarini, P. (Eds.) Handbook of Swiss Politics. Zurich, NZZ Libro. Linder, W., Hümbelin, O. & Sutter, M. (2009) Die Entwicklung der eidgenössischen Gesetzgebungstätigkeit 1983-2007: eine quantitative Analyse. Bern, Institut für Politikwissenschaften. Linder, W., Schwager, S. & Comandini, F. (1985) Inflation législative? Une recherche sur l'évolution quantitative du droit suisse 1948-1982, Lausanne, IDHEAP. Lüthi, R. (2006) The Parliament. IN Klöti, U., Knoepfel, P., Kriesi, H., Linder, W., Papadopoulos, Y. & Sciarini, P. (Eds.) Handbook of Swiss Politics. Zurich, NZZ Libro. Mach, A. (2006) La Suisse entre internationalisation et changements politiques internes: législation sur les cartels et relations industrielles dans les années 1990, Zurich and Chur, Verlag Rüegger. Mach, A., Häusermann, S. & Papadopoulos, Y. (2003) Economic regulatory reforms in Switzerland: adjustment without European integration, or how rigidities become flexible. Journal of European Public Policy, 10, 301-318. Mallepell, R. (1999) Der Einfluss des Gemeinschaftsrechts auf die schweizerische Gesetzgebung. Swiss Papers on European Integration, 21. Poitry, A.-V. (1988) La fonction d'ordre de l'Etat: Analyse des mécanismes et des déterminants sélectifs dans le processus législatif suisse, Bern, Peter Lang. Schwok, R. (2006) Suisse-Union Européenne: L'adhésion impossible?, Lausanne, Presses polytechniques et universitaires romandes. 300
Sciarini, P. (2006) The Decision-Making Process. IN Klöti, U., Knoepfel, P., Kriesi, H., Linder, W., Papadopoulos, Y. & Sciarini, P. (Eds.) Handbook of Swiss Politics. Zurich, NZZ Libro. Sciarini, P., Fischer, A. & Nicolet, S. (2004) How Europe hits home: evidence from the Swiss case. Journal of European Public Policy, 11. Sciarini, P., Nicolet, S. & Fischer, A. (2002) L'impact de l'internationalisation sur les processus de décision en Suisse: Une analyse quantitative des actes législatifs 1995-1999. Swiss Political Science Review, 8, 1-34.
301
9. Tables and Graphs
Table 1: Legislative production
Table 2: EU footprint in legislative proposals
302
303
Graph 1: EU footprint in Legislative proposals per policy domain 1987-2007 (N=763)
33.3%
Transports 29.0%
Domestic Commerce 20.7%
Foreign Trade Energy
19%
Science & Communications
19% 17%
Foreign Policy 13.6%
Immigration & Integration
13%
Urban & Regional Development Agriculture
12%
Education and Culture
11.1%
EU Direct EU Indirect
9.5%
Macroeconomics Environment
8.3%
Social Policy
6.7%
Law & Crime
6.5%
Government & Administration
4.1%
Rights
3.6%
Health
1.8%
Labor & Employment 0% Defense 0% Public Lands 0% Total
12.1% 0%
5%
10%
15%
20%
25%
30%
35%
304
Table 3: EU footprint in legislative acts
305
Graph 2: EU footprint in legislative acts per policy domain 1987-2007 (N= 916)
34.8%
Immigration and Integration 12%
Domestic commerce Education and culture
11.1%
Foreign Trade
11.1%
Transports
8.1%
Labor & Employment
7.4%
Social Policy
6.5%
Foreign Policy
5.9%
Science & Communications
5.7%
Urban & Regional Development
5.6%
EU Direct EU Indirect
4.2%
Law & Crime Environment
3.6%
Government & Administration
3.3%
Health
2.3%
Agriculture
2.1% 2%
Macroeconomics Rights 0% Energy 0% Defense 0% Public Lands 0% Total
6.4% 0%
5%
10%
15%
20%
25%
30%
35%
306
Table 4: EU footprint and optional referenda on federal laws (1987-2007)
307
Annex: Delors' Myth common tables and figures
1. Number of laws and proportion of the different legal status of laws per year (1986-2007)
308
2. Number and proportion of Europeanized laws per year (1986-2007) N=705
Figure 2. Europeanization of federal laws (1986-2007) 70 60 50 40 30 20 10 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Total Laws
Laws Europeanized
% Europeanized
309
3. Proportion of Europeanized laws per policy area (1986-2007)
310
12. Going beyond: causes of Europeanization T. Koenig and L. Maeder
The aim of this book is the empirical evaluation of Brussels’ influence on domestic legislatures. The previous chapters have empirically determined the amount of legislative activities at the EU level, which potentially impact the activities in the domestic legislatures. Formally, we saw that the EU has deepened the relationships among the member states during the last 30 years by extending the policy competencies to additional policy areas and expanding qualified majority voting in the Council with the participation of the European Parliament. These changes were performed by treaty revision and intended to facilitate and legitimate the adoption of Commission proposals. At the same time, the EU has increased the number of member states from nine to 27 countries, starting with Southern enlargement in the beginning of the 1980s, followed by the Northern round in the mid-1990s, and most recently the Eastern enlargements in the beginning of this century. Given the increasing number of countries and policy competencies, it is hardly surprising that the absolute number of decisions, which are directed to individual members, has slightly increased over time. Similar to the type of instrument, the type of legislation has also changed over time: secondary legislation, in which the Council and the European Parliament are directly involved, significantly decreased in this period. Tertiary legislation, which is delegated to the Commission and usually of a more bureaucratic nature, is the major growth factor of EU legislative activities, and in particular the share of delegated regulations significantly expanded over time. Compared to regulations and decisions, the total number of directives, which are expected to affect domestic legislatures more directly due to their transposition requirement into domestic law, remained about the same. On closer inspection of directives, we also find that the share of delegated directives under tertiary legislation almost doubled in size. This suggests that even though the number of EU legislative activities has only slightly increased over time, the type of activities significantly shifted from secondary to tertiary legislation of the Commission, which is considered to be more bureaucratic in nature.
311
In a further effort, the country chapters investigated the impact of EU legislative activities for eight domestic legislatures and one country outside EU. Compared to the mythical 80% narrative on Brussels impact, they found that the average impact of EU activities on domestic legislation is significantly lower but has increased over time, ranging from 14 per cent in France to 26 per cent in Austria, with a very low influence of only 2 per cent in the nonmember state of Switzerland. Figure 1 displays the number of adopted domestic laws, the number of Europeanized domestic laws and their percentage for eight of the countries included in this volume: Austria, France, Germany, Italy, Luxembourg, Netherlands, Spain, and Switzerland across 19 policy areas over the period between 1987 and 2005.
Figure 1 about here
According to Figure 1, most of the domestic laws were adopted in the policy areas of Law, Macroeconomics, and Government operations. In addition to the variation among countries, the share of Europeanized domestic laws varies considerably across the different policy areas. Compared to the policy areas of Agriculture, Banking & Finance, and Environment indicating a relative high level of Europeanization with about 32 per cent of Europeanized domestic laws, the level of Europeanization remains relatively low in the policy areas of Defence, Housing, and Social Welfare with less than eight per cent of Europeanized domestic laws. In addition to this area-specific development, Figure 2 illustrates a common trend towards a higher level of Europeanization of domestic laws over time.
Figure 2 about here
The number of Europeanized domestic laws fluctuated around six per cent in 1987, but this share increased significantly until 1993, reaching almost 21 per cent. In the following years, the share of Europeanized domestic laws slightly decreased to 15 per cent until 1995 – the date of Northern enlargement. Since 1996 the share of Europeanized domestic laws has almost constantly grown every year, reaching its highest value in 2005 with almost 29 per cent. Before Eastern enlargement, there was a significant increase of almost six per cent in 312
2003. This trend towards more Europeanized domestic law is almost identical in all eight countries, but more pronounced in Spain, Luxembourg and Italy.
The relative impact of EU legislative activities and legislative review On closer inspection of these developments along country- and area-specific patterns, we are interested in evaluating the empirical implications of two kinds of factors, which contribute to this development. Taking a “top down”-perspective, the central question is whether and to what extent does the increase of EU (tertiary) legislation explain this area-specific domestic development. In other words, if tertiary legislation is responsible for the trend towards the Europeanization of domestic laws, we have good reason to believe that this type also accounts for a bureaucratization of domestic legislatures. A more “bottom up”-related perspective on domestic-related factors draws attention to the “means” of Europeanization, where the share of Europeanized domestic laws indicates the activities of domestic parliaments in terms of using legislative review. In examining the means of Europeanization, we follow the recent literature on legislative review, according to which coalition conflict increases the likelihood for ministerial drift which coalition partners can overcome by using legislative review (Epstein and O’Halloran 1999, Martin and Vanberg 2004, 2005, König and Mäder 2008). Our focus on bottom up and top down points to the two most extreme perspectives on Europeanization, which refer to arguments on down- and uploading. We will examine whether and to what extent these two factors are particularly important for the evaluation of the country- and area-specific Europeanization trend, which we found in the investigated countries. For this purpose, we generate a dataset on the annual share of Europeanized domestic laws which varies across countries and policy areas. We operationalize the arguments of both perspectives on Europeanization using our data on EU legislative activities and construct new area-specific estimates for measuring coalition conflict in the eight countries. Finally, we evaluate the empirical implications of both perspectives and discuss their impact for the current debate on Europeanization.
Policy Growth by EU legislative activities For considering the empirical implications of Europeanization, EU legislative activities are an essential precondition for Brussels’ impact on domestic parliamentary legislatures. Apart from 313
treaty conditions and court decisions, a certain number of adopted Commission proposals, i.e. decisions, regulations and directives, is a necessary condition for the influence of the EU on the number of domestic laws. We label this expectation a policy growth model and expect that the share of Europeanized domestic laws increase with the number of adopted Commission proposals. In particular directives, which require transposition into domestic law, should have a marked effect on the share of Europeanized domestic laws. Compared to decision and regulations, directives imply the “structural necessity” and obligation of the member states to enact transposition measures in order to give it full legal effect (Capotorti 1988:156). Some scholars argue that this obligation creates a ‘two-stage legislation’ or ‘indirect legislation’ (Prechal 1995:18). In the first stage, the EU defines the goals of directives, which have to be transposed into the national legal order in the second stage of this process. Some scholars conceive policy growth of EU legislative activities as a sufficient condition for Brussels impact on domestic parliamentary legislatures, which they operationalize by the ratio of the number of secondary Commission proposals to the number of domestic laws in force on a given day (Höscheidt and Hoppe 2010, 234). We believe that this approach has too many drawbacks: first, it assumes that member states transpose one piece of secondary legislation by one domestic law, which is far from empirical evidence. By contrast, comparative transposition studies have shown that member states vary in their transposition record and mainly transpose directives by executive instruments (König and Mäder 2007). Furthermore, transposition frequently involves multiple instruments, while one piece of national law sometimes bundles the transposition of multiple directives. This does not mean that we do not expect a positive relationship between EU legislative activities and the number of Europeanized domestic laws. However, since executive instruments are frequently used for transposition, we only see the number of adopted Commission proposals as necessary but not sufficient condition for explaining the share of Europeanized domestic laws. This share may not only vary across countries, policy areas and over time, but also caused by other factors. Regarding EU legislative activities, we expect that the share of Europeanized domestic laws will increase with the number of adopted Commission proposals. According to this policy growth model, the likelihood for a reference to the EU should correspond to the total number of adopted decisions, regulations and directives. Hereby, directives are considered to have a more pronounced effect for domestic legislatures which we measure by a two-year lagged variable because directives usually grant a two-year transposition period for implementation into domestic law. In addition to this distinction between directives and decisions/regulations, 314
we believe that the type of secondary or tertiary legislation is also important for the evaluation of Brussels’ impact on domestic legislatures. Secondary legislation is decided by the Council and – to an increasing amount – in collaboration with the European Parliament, while tertiary legislation is delegated to the Commission, in particularly for regulating technical and bureaucratic matters.1
Domestic factors by coalition politics Europeanization of domestic legislatures may also follow domestic concerns. The “deparliamentarization” literature is concerned about the (informational) advantages, which Europeanization provides governmental actors vis-à-vis domestic parliaments. On closer inspection, many governments consist of two or more political parties, which differ in their policy interests but share a common interest in forming coalition government. When forming a governmental coalition, they must agree on a cabinet and delegate ministers to it, which have an incentive and the resources to pursue their own party interests at the expense of the coalition partner (Martin and Vanberg 2004; 2005).2 In the event of ministerial drift, coalition partners can however make use of parliamentary oversight by legislative review to overcome their delegation problems. Because ministers have a particular informational advantage in European affairs, the number of Europeanized laws may also indicate how often coalition partners make use of legislative review in the parliamentary legislature to control for ministerial drift. According to the parliamentary scrutinizing approach developed by Martin and Vanberg (2004, 2005), coalition partners are inherently plagued by delegation problems when the distance between coalition partners is large. In combination with an informational advantage of bureaucratic expertise, this distance indicates the risk of ministerial drift, which coalition partners can overcome by legislative review. Compared to the minister, coalition partners
1
Legally, only those rules should be considered as tertiary legislation whose legality directly depends on rules of secondary legislation, but this is not the case when secondary legislation empowers the Commission to enact further legislation. Such delegated legislation, made at the time when the Commission has been authorized to do so, remains in force when secondary legislation granting the power is withdrawn. 2 Huber and Shipan (2002:183-185) suggest that ministers dominate the parliamentary and the executive process. In parliamentary legislation, the minister is responsible for preparing the initial proposal. However, at this stage, a minister is often constrained by other members of the cabinet because the initial proposal is usually discussed in cabinet and requires approval before sending it to the floor. Compared to the parliamentary process, ministers are less closely monitored in executive matters. They are the head of department and define the guidelines of how policy should be implemented by an agency.
315
need however to invest time and resources for parliamentary oversight by legislative review. This makes their decision to scrutinize ministers conditional on the costs of legislative review. When these costs are low, coalition partners will more often scrutinize ministerial policies when the risk of ministerial drift increases. Consequently, the share of Europeanized laws should be a function of ministerial drift and relative costs for legislative review. By contrast, Epstein and O’Halloran (1999) assume that policy implementation is always associated with a certain level of uncertainty and that ministers, due to the expertise of their executive agencies, have an informational advantage over parliaments. Accordingly, legislative review follows the interaction between the risk of ministerial drift and uncertainty associated with EU policy. For any fixed level of risk of ministerial drift, the share of Europeanized domestic laws will shrink when the level of uncertainty increases. In contrast to Martin and Vanberg (2004; 2005), this perspective predicts that low uncertainty produces a higher share of legislative review. To measure ministerial drift both approaches use the size of coalition conflict, which we have to measure for eight countries across 12 policy areas. Instead of using standard estimates for political parties, which provide little variation over time and across policy areas, we apply our smart-tagging procedure which produces variation by relating these estimates to area-specific context (for more detail about smart-tagging, see König et al. 2010). Briefly summarized, we consider context by running a keyword search with the categories of the Party Manifesto Research Group (MRG) in the descriptions of the Policy agenda project, according to which our policy areas are defined. This search with 3 to 4 keywords of the MRG categories produces an area-specific hit list, which we use to calculate a context-related left-right position of political parties by the following formula: !"#$%&'()*+,( = log 12 3% 456$%&'(,% 8 − log 12 3% 456*+,(,% 8 %
%
where 3% represents the number of hits from our keywords in the description of the policy areas of the Policy agenda project, 456$%&'(,% the number of a party’s right-wing and 456*+,(,% the number of left-wing quasi-sentences.3 This context-related estimate expresses the size of
3
For the allocation of the different MRG categories to the left-right dimension we follow Lowe et al. William Lowe et al., "Scaling Policy Preferences from Coded Political Texts," Legislative Studies Quarterly 26, no. 1 (2011)..
316
an area-specific coalition conflict according to the absolute distance between the two most extreme governmental parties.4 To identify the costs for legislative review we use an indicator which measures the average involvement of domestic parliaments in the transposition of directives (König and Mäder 2007). Hence, coalition partners are faced with different costs for legislative review in the event of ministerial drift, which will determine the share of Europeanized domestic laws. Put differently, when costs for legislative review are extremely high because implementation has been delegated to the executive or because domestic parliaments have weak standing committees etc., ministers can more easily pursue own interests and implement by executive measures in the event of coalition conflict. Thus, we expect not only a significant positive effect of regular parliamentary involvement on the share of Europeanized domestic laws, but also - conditional on costs – that the share of Europeanized domestic laws changes with coalition conflict.
The area-specific Europeanization of domestic legislation over time For the analysis of the explanatory power of EU- and domestic-related factors, we start by creating a pooled-time-series-cross-section dataset with 2.888 observations for the period from 1987 to 2005 (8 countries * 12 policy areas * 19 years). Next, we connect the 19 domestic policy areas to the corresponding 11 EU policy areas plus a residual area that contains Health, Education, Law, Social Welfare, Housing, Defence, Technology, and Public Lands, where we do not have a corresponding area or only a few activities at the EU level. From the 1.824 cases of this dataset, we drop 171 Austrian cases for the period prior to accession in 1995. For this sample, the dependent variable is the percentage of Europeanized domestic laws per policy area and year (;%
