This is the accepted version of the journal article: Fernández Pasarín, Ana Mar; Dehousse, R.; Plaza, J.P. «Comitology : the strength of dissent». Journal of European Integration, Vol. 43 Núm. 3 (2021), p. 311-330. DOI 10.1080/07036337.2020.1740697 This version is available at https://ddd.uab.cat/record/259720 under the terms of the license Comitology: The Strength of Dissent, Journal of European Integration Fernández-Pasarín, Ana Mar; Dehousse, Renaud and Plaza, Joan Pere Abstract This article sheds new light on deliberative dynamics at work in comitology. Starting from the findings of previous research on the frequency of consensus within the comitology committees of the EU in 2008, this article seeks to provide a measure of the intensity of opposition to the Commission’s draft implementing proposals before and after the 2011 comitology reform. Applying this approach to data from the comitology register (2008-2013), the article analyses the structure of contestation, proposes an index of opposition and tests the factors that may explain variations. Attention is paid to procedural choices and sociological arguments before and after the last reform of the comitology system. Our findings are fourfold. First, conflict with the Commission’s positions, although weak, is not that rare during deliberations. Second, patterns of support and opposition are stable over time. Third, sharp cross-sectorial variations exist. Four, procedural choices matter for contestation. When member states have strong national preferences, they opt for more constraining procedures and will be less inclined to engage in consensual dynamics. Keywords: comitology; contestation; council; delegation of power; European commission; executive decision-making 1 Introduction Delegation of power without contestation? This article sheds new light on deliberative dynamics at work in comitology. Starting from the findings of a previous study on the frequency of consensus within the comitology committees of the EU in 2008, which unveiled the existence of contested votes in over a quarter of member-state representatives’ deliberations (Dehousse et al. 2014), this article seeks to provide a measure of the intensity of opposition to the Commission’s draft implementing proposals before and after the 2011 comitology reform. Using data from the comitology register (2008-2013), the article analyses the structure of contestation, proposes an index of opposition and briefly explores the factors that may explain variations. Attention is paid to procedural choices and sociological arguments. Our findings are threefold. First, they reveal that conflict with the Commission’s positions, although weak, is not that rare during deliberations. Second, they demonstrate that patterns of support and opposition are stable over time, i.e., the 2011 comitology reform has not fundamentally altered either the frequency, the intensity or the direction of dissent. Third, sharp cross-sectorial variations exist. Four, procedural choices matter for contestation. When member states have strong national preferences, they opt for more constraining procedures at the legislative stage and will be less inclined to engage in consensual dynamics and to accept Commission’s proposals without contestation at the comitology level. The article is organised as follows. Section I addresses the theoretical debate on the nature of interactions in EU comitology committees and discuss, in particular, the question of the practical denaturation of comitology as a vector of intergovernmentalism. 2 Section II present the data and method of the research with particular emphasis on the construction of an index of opposition in the delegation phase of the EU decision-making process. Section III lays out the patterns of contestation to the Commission's implementing proposals as revealed by the empirical study. Section IV attempts to explain these findings by testing two hypotheses derived from the literature. The paper ends with some remarks on the findings of this study, its scope and limits as well as further research questions. Comitology: Delegation without contestation? Since the 60s, the main principle of governance underlying the configuration of the comitology system has been the control of the delegated competences entrusted by the Council to the Commission as regards the implementation of EU common policies. The origin and development of comitology reflect nothing other than the need to articulate interinstitutional relationships at the regulatory level of EU decision-making and to crystallise, in particular, member states’ control over the Commission’s executive powers. As such, comitology was configured as a classic instrument of intergovernmental control of supranational institutions. By imitating the Council’s territorial and sectorial composition as well as its modes of decision, comitology committees look like ‘Councils in miniature’ (Blom-Hansen 2013: 427). Yet, they differ from their master’s institution in at least three aspects. First, their highly specialised areas of expertise and their backstage position on the EU decision- making scene invest them with less political aura. With the exception of the Appeal Committee, which is composed, as a general rule, of representatives with “a high and 3 horizontal profile” not below the level of members of the Committee of Permanent Representatives1, comitology committees appear to be eminently technical and administrative, even though the regulatory tasks they perform are actually far from being anecdotic, as reflected, for instance, by the important lobbying activity deployed in this domain (cf. Wetendorff Nørgaard et al. 2014; interview Comitology lobbyist, Brussels, 24 July 2015). Second, despite significant efforts in terms of transparency since 2008 with the launching of the new Comitology Register, the activities of these meso-level institutions still have a certain air of opacity. Illustratively, contrary to the rule in force at the Council level, there is no disaggregated information made public about the votes of national representatives in committees. Third, and more crucially from an interinstitutional point of view, comitology committees are intergovernmental units embedded in the Commission and thus placed under its supranational leadership. This last peculiarity has generated a recurrent debate in the literature about the nature of day-to-day interactions within comitology committees beyond their formal definition as guardians of national interests (Brandsma and Blom-Hansen 2010; Egeberg et al. 2003; Joerges and Neyer, 1997a and 1997b; Trondal and Veggeland 2003). Do member states keep strict control over the Commission’s room for manoeuvre in the executive domain, as could be expected given the essence of comitology and suggested by the successive 1 Appeal Committee, ‘Rules of procedure’, Official Journal of the EU, 2011/C183/05, 24 June 2011. In practice, the summary records of this committee reveal that national representatives are usually members of the Permanent Representation, occasionally accompanied by representatives from the ministries or national agencies. 4 ‘games of control positions’ surrounding the reforms of the comitology system (Brandsma and Blom-Hansen 2012) or is the Commission ‘escaping’ from this monitoring and actually making use of comitology procedures to strengthen its own institutional position? In brief, to what extent is the institutional design of comitology as a vector of intergovernmentalism challenged on the ground by its daily functioning and, in particular, by the Commission’s tendency to regard this system as a mode of self- affirmation? These questions about the virtual denaturalisation of comitology as a mechanism of member states’ control have led to different interpretations. Based on the classical dichotomy between intergovernmental and supranational theories, most of the contributions have framed their reasoning in terms of principal-agent tension. In this vein, whereas part of the literature sustains that comitology committees maintain their original control function (Blom-Hansen 2013; Pollack 2003), other scholars argue that this control is undermined by the Commission’s strategy of devising operative mechanisms to free itself from these constraints and forge support for its implementing proposals, anticipating, for instance, adverse reactions in both the Council and Committees (Ballmann et al. 2002; Craig 2016). This discussion in terms of inter-institutional battles and actors’ relative power, which was extended to the European Parliament following the introduction of the right of scrutiny in the 1999 Comitology Decision and its elevation to the rank of ordinary co-legislator in the Treaty of Lisbon, has naturally been complemented by the question of the representation of interests. Thus, whereas from an intergovernmental perspective, national representatives in comitology committees are considered mere transmitters of member states’ preferences, from a deliberative supranationalist view, meetings between national representatives, who are high-profile 5 experts, are privileged arenas for the creation of a European epistemic community that enhances actors' autonomy from command-and-control governmental approaches (Böhling 2014; Dehousse 2003; Joerges and Neyer 1997a and 1997b). Between these two polarised views, an intermediate interpretation considers that committee members combine and deal with both identities in a kind of schizophrenic exercise (Brandsma 2010: 491; Brandsma 2013; Egeberg et al. 2003). In short, the main question addressed by the literature is that of the match between the original institutional purpose of comitology and its actual functioning in day-to-day activities. To what extent is comitology a real national counterweight to the Commission’s delegated powers as initially established? What kind of interest do national representatives actually represent when they are working in comitology committees, i.e. under the umbrella of the Commission? This debate about the theoretical conceptualisation and practical evolution of comitology has gained special relevance with the entry into force of the Lisbon Treaty and the 2011 comitology reform. The Lisbon Treaty divided the old regime of comitology into two distinctive executive parts –the delegated acts (quasi-legislatives; Art. 290 of the Treaty on the Functioning of the EU (TFEU)) that have substantially empowered the operational range of the Commission and the implementing acts proper (Art. 291 of the TFEU), which remain under the control of comitology. In February 2011, a new Regulation of the European Parliament and of the Council revised both the number and typology of comitology procedures (Regulation (EU) Nº182/2011). Since the reform, the control by member states of the Commission’s implementing powers is channelled through two main procedures: the advisory (Art. 4) and the examination (Art. 5) procedures. The 6 selection of the procedure to be used is determined by the legislator in the basic legal act. As a general rule, the examination procedure should apply to implementing acts of general scope; the common agricultural and fisheries policies; the environment; health protection; safety of humans, animals or plants; trade or taxation policies. The advisory procedure should apply, for its part, to the other implementing acts. Furthermore, whereas the advisory procedure is merely consultative, the examination procedure has a more constraining effect on the Commission’s room for manoeuver in that it “[that procedure] should ensure that implementing acts cannot be adopted by the Commission if they are not in accordance with the opinion of the committee […]” (Regulation Nº182/2011). In addition, the regulatory procedure with scrutiny (RPS), set out in the Art. 5a of the 1999 Comitology Decision which gives a right of veto to the Council and the European Parliament even if the committee has delivered a positive opinion, is still applied to existing legal acts which make reference to it (European Commission 2013)2. The 2011 comitology reform stimulated new reflections about the institutional scope and practice of comitology (Bergström and Ritleng 2016; Brandsma 2013; Brandsma and Blom-Hansen 2012 and 2016; Siderius and Brandsma 2016).The objective of this article is to contribute to this debate by bringing new empirical data on the member states’ behaviour regarding the Commission’s action in the executive domain before and after the most recent comitology reform. A previous study on the frequency of member states’ dissensus with the Commission’s regulatory measures in comitology committees in 2008 showed that conflicts, i.e. contested votes, existed in over a quarter of member 2 Since 2013 this procedure can no longer be used in new legislation. See European Commission (2013) Report on the working of committees during 2012, COM (2013) 701 final, Brussels, 10 October. 7 states’ deliberations (Dehousse et al. 2014). This finding revealed, thus, that national representatives’ opposition to the Commission’s draft implementing proposals, although not frequent, did exist. The present research seeks to refine these initial results by 1) looking at the structure of contestation over a longer period, encompassing the pre- and the post comitology reform in 2011, 2) attempting to map and quantify the frequency and intensity of this conflict and 3) test explanatory variables. Data and method. Measuring opposition in comitology procedures The empirical contribution to be found in this article relies on a novel dataset encompassing six years of comitology committees' workings and voting records which are available on the new Comitology Register for the period 2008-2013 (N=8021)3. In those cases where on-line information from the Comitology Register's website was incomplete or incorrect voting records were discarded. In addition, instances of contested votes under the advisory procedure whose structure of votes was expressed through the number of member states and not through a weighted vote were also eliminated as the calculation of the index opposition was here impossible (N=684). The complete information was downloaded from the Register, entered into the database and continuously updated in order to keep up with the many changes that the new Comitology register website suffered throughout this period. All in all, the database holds 91.4% of the voting procedures initiated in the framework of the comitology system for the period 2008-2013 (N=7.337). These quantitative data have been complemented with attendance lists to the committee’s meetings which are also available in the Comitology Register. As 3 The new Comitology register was created in April 2008. 8 regards quantitative data processing, we proceed in two steps. The first is of a descriptive nature. It consists of identifying patterns of support and opposition to the Commission’s draft implementing proposals on the basis of member states’ votes within comitology committees. All the voting sheets available in the new Comitology Register for the years 2008-2013 were coded so as to identify the structure of votes and, in particular, the scope of contestation over the period under scrutiny. Concretely, for any instance of voting, we have counted the number of votes ‘FOR’, votes ‘AGAINST’ and ‘ABSTENTIONS’ as well as the number of votes ‘NOT REPRESENTED’4. This information was classified in terms of committees (222), procedures (advisory, management, regulatory, regulatory with scrutiny or safeguard procedure before the entry into force of the 2011 comitology reform; advisory, examination or regulatory with scrutiny procedures after the reform) and policy domains according to the Comitology Register’s classification5. 4 By way of example, the voting procedure V024762/01, issued in 2012, was approved with 345 votes “For”, 0 votes “Against”, 0 votes “Abstaining” and 0 votes of “Not represented” MS. The structure of vote of this particular voting procedure will thus be noted as follows: SVV024762/01 = 345−0−0−0. By contrast, voting procedure V021969/01 was approved with 333 votes “For”, 0 votes “Against”, 0 votes “Abstaining” and it is also said that 1 MS was “Not Represented”. In this case, we can assume that, in fact, 12 votes were “not represented”. Consequently, its structure of vote will be noted as follows: SVV021969/01= 333−0−0−12. 5 The classification system of the Comitology Register includes all the Commission’s DGs as well as administrative bodies such as Eurostat or the General Secretariat. Our data reflect the current categories set in the Register (March 2019). Thus, DG’s acronyms have been retroactively updated following changes introduced in the DG’s nomenclature. 9 Preliminary observations at this stage include the following. 1) In contrast with the Council’s minutes, the Comitology Register does not provide information about the votes cast by each delegation. Member states’ preferences appear, thus, as aggregated. 2) Member states ‘not represented’ who did not make any explicit mention of opposing or abstaining the Commission’s draft implementing act before the expiry of the time limit are regarded as having tacitly agreed to the Commission’s implementing proposal (Art.3. parag. 5 of the Regulation (EU) No 182/2011). It is worth noting that according to our dataset, 25.6 % of the cases under scrutiny show a varying intensity of “tacit agreement”. 3) In the case of the advisory procedure, decisions can be adopted through a simple majority of votes. A QMV similar to that of the Council is the rule for the other procedures (since the accession of Croatia, 260 favourable votes out of 352). The second step is of an analytical nature. We propose the construction of an index of opposition (O) with the aim of capturing and measuring the strength of dissent in comitology deliberations for the period 2008-2013 (N=7337), in addition to its frequency. As mentioned before, we decided to discard from the sample any instances of contested votes under the advisory procedure whose structure was expressed through the number of member states. Nevertheless, cases where the committee expressed full support or opposition (28 Member states in favour or against) or those where the opinion was delivered through a general allusion to “consensus” or “unanimous favourable” (N=195), were included in the final sample. We consider the index of opposition as a proxy to measure the intensity of conflict with the Commission’s executive proposals. For the purpose of this study, the notion of dissent or opposition refers to any possible situation where member states’ preferences 10 on any particular policy issue at hand differ in varying intensities from the original draft implementing proposal from the Commission. Here, we assume that 1) non-support (tacit or explicit) is regarded as opposition to the Commission’s draft implementing proposal; 2) abstention is considered a lesser form of opposition than votes against (cfr. Hayes- Renshaw et al. 2006; Mattila 2004). Conventionally, we decide to consider abstention as half-opposition. The index of opposition is, thus, conceived to measure to what extent the voting results display evidence of opposition to the Commission’s draft implementing act (beyond a final ‘favourable opinion’ to the Commission’s proposal).6 The index we propose is as follows7: ‘O’ is equal to the number of votes against plus the abstentions (considered as half-opposition) divided by the total number of votes at stake8. 6 It is worth noting that this index of opposition does not allow the ordering of committees’ opinions that, although presenting very different structures of vote, have the same value of opposition. Illustratively, SVV005863/01=167-109-65-4 (year 2009); SVV021131/01=149-87-109-0 (2012) and SVV027002/01=187-125- 33-0 (2013) all present the same value of opposition: O =0.41014. Which of these three voting procedures should be considered the most contested is, however, a question that remains open. 7 This index is a variation of the Rice Index of Cohesion developed by Stuart Rice in 1924. 8 In all policy sectors with the exception of Justice, in which the opt-out clause applies for Denmark, Ireland and the UK, the total number of votes at stake was 345 before the accession of Croatia in 2013 and 352 since then. When the opt-out clause was activated before the accession of Croatia, the total number of votes could be either 302 (all three countries) or 338 (opt-out of Denmark). Since the entry of Croatia, the total number of votes in case of opt-out could be respectively 309 or 345. 11 Against+0.5∙Abstention O= For+Tacit Agreement+Against+Abstention We additionally identify six categories of opposition on the basis of the intensity of opposition to the Commission’s proposal. The construction of these categories is based on the majority required for the adoption of decisions at the Council level. We assume that: (1) Absence of opposition (O =0) is a situation where there is unanimous support for the Commission’s proposal. In this event, O equals 0. Voting sheets referring to “consensus” or “unanimous favourable” are also placed in this category. (2) Weak opposition is a situation where there are votes against and/or abstentions in a number below the blocking minority (i.e. at most 90 votes against -92 since the accession of Croatia- or any equivalent structure of votes with the same value for the index of opposition). (3) A medium opposition is a situation where there are votes against and/or abstentions in a number equal to or higher than the blocking minority but lower than the absolute majority of votes against (i.e. from 91/93 to 172/176 votes) against or any equivalent structure of votes with the same value for the index of opposition). (4) A majoritarian opposition is a situation where there are votes against and/or abstentions in a number equal to or higher than the absolute majority but lower than the qualified majority of votes against (i.e. from 173/177 to 254/259 votes against or any equivalent structure of votes with the same value for the index of opposition). 12 (5) A qualified opposition is a situation where there are votes against and/or abstentions in a number equal to or higher than the qualified majority of votes against but lower than full opposition (i.e. from 255/260 to 342/349) or any equivalent structure of votes with the same value for the index of opposition. (6) Full opposition (O =1) is a situation where all the votes are against the Commission’s draft proposal, thus there are neither votes in favour nor abstentions Figure 1. Scale of opposition in voting procedures [Figure 1 near here] Findings Globally speaking, data on the structure of contestation for the period 2008-2013 confirm previous findings for the year 2008 (Dehousse et al. 2014). In the vast majority of cases, deliberations end with a favourable result for the Commission (96.98%). Noticeable cases of opposition, i.e. cases that ended with a ‘non-favourable opinion’ on the Commission’s draft implementing proposals are extremely rare (table 1). Actually, during the period 2008-2013 only 9 cases of this sort were recorded (0.12% of the total volume of comitology opinions under scrutiny). For the rest, unanimity is the prevalent but not the only mode of decision in comitology: on average, 77.35% of the opinions on the Commission’s draft proposals delivered by comitology committees through a vote between 2008 and 2013 were uncontested, i.e. all member states voted in favour (either explicitly or through tacit agreement) or a general allusion to consensus or unanimous 13 favourable appeared on the voting sheet as a result of the committee’s deliberations. Yet, in a non-negligible 22.65% of the total sample, the decision was contested to some extent through the expression of at least one vote against, an abstention or both, without this preventing the committees’ members from delivering a final positive opinion (table 2). In a nutshell, these data confirm that the final adoption of the Commission’s draft implementing proposal does not preclude the existence of dissensus during previous deliberations. Member-state opposition is not typical behaviour in comitology committees but does nevertheless exist in over one in five voting procedures. In 13.40% of the cases at least one member state expressed a negative opinion while in 16.34% of the votes at least one member state abstained. When member states choose to manifest dissensus, they therefore opt more frequently for abstention than for explicit opposition. Last but not least, no noticeable variations are appreciated between the pre- and the post 2011 comitology reform. [Table 1 near here] [Table 2 near here] If we look at this opposition in terms of intensity, our findings reveal that, in general, opposition levels are low and stable across time (table 3 and figure 2). Between 2008 and 2013, only 4 instances of qualified opposition against the Commission’s proposals were registered (of which 3 in 2011). Cases of majoritarian opposition are not exactly rare but there are nevertheless very few: 19 cases were recorded during the period, 14 mainly concentrated in years 2011 and 2012 (8 and 6 cases respectively). By contrast, instances of medium opposition (153 cases concentrated in years 2011, 2013 and 2012) and, above all, of weak opposition (1486 cases) are markedly more frequent and actually represent the dominant feature of the contestation landscape in comitology. Finally, opposition descriptives have slightly increased with the coming into force of the 2011 comitology reform (table 4). [Table 3 near here] [Figure 2 near here] [Table 4 near here] On another note, our data reveal that contestation is not equally distributed across policy sectors (tables 5 and 6). In addition to the Secretariat General which deals with the instances of votes taking place at the level of the Appeal Committee (i.e. a special committee’s configuration that replaces the Council in its function of second instance since the 2011comitology reform9), the most contested policy domains for the period in terms of frequency are Budget (100%), followed by Taxation and Customs Union (issues 9 The Appeal Committee can block the Commission’s draft implementing proposals through a qualified majority against. Between 2011 and 2013, the period under scrutiny in this article, this special committee intervened on 23 occasions (Between 2011 and 2018, the total number of interventions ascends to 86). Source: Register of Comitology (Last accessed: 18 October 2019). For an analysis of this committee, see Christiansen and Dobbels (2016). 15 related to the negotiation of the Customs Union Code) (67.83%), Climate (51.43%), Environment (49.10%), Eurostat (46.79%), Energy (46.27%) and Agriculture (38.29%). Compared to these data, the Health and Food Safety sector (SANTE) seems to be much less conflictual (17.16% of contested votes and a mean of opposition that is below the highest ten policy domains). These aggregated data must however not be misinterpreted. Indeed, this domain presents a particularly high volume of activity, namely 39.4% of the total number of comitology opinions for the period under scrutiny. In addition, one of its most active committees, the C2500-Standing Committee on Medicinal Products for Human Use (1257 voting records between 2008 and 2013, which stand for 17.13% of the total sample, and 43.40% of the opinions issued in SANTE) tends to be very consensual due to its internal rules of procedure which facilitate pre- agreements at the level of national medicines agencies (Interview European Commission’s Official, DG SANCO, Brussels, 13 June 2014). The combination of these two variables serves to blur the existence of very few but nevertheless strong and noticeable contestation cases. As a matter of fact, three out of the four most conflictual instances of votes for the period 2008-2013 in our dataset correspond to this policy domain and two of them to this latter committee: typically cases related to regulations on NGOs and medicines for human use (table 9). Concretely, in 2008, the Standing Committee on Plants, Animals, Food and Feed - Section: "Biological safety of the food chain" (C20404) delivered an unfavourable opinion on a Commission draft proposal to implement Regulation (EC) No 853/2004 of the European Parliament and of the Council concerning the use of antimicrobial substances to remove surface contamination from poultry carcasses. The structure of votes was then 0 votes in favour-316 votes against-29 abstentions and 0 votes not represented (SVV000724). Besides, in 2011, the aforementioned 16 Standing Committee on Medicinal Products for Human Use (C02500) issued an unfavourable opinion on the market authorisation of the ‘orphacol-cholic acid’ for human use with 79 votes in favour-266 votes against-0 abstentions and 0 votes not represented (SVV017628/01). Last but not least, the third one refers to a decision of the Appeal Committee ratifying the negative opinion delivered in first instance by the C02500 regular committee, with the following voting result: 64 votes in favour-281 votes against-0 abstentions and 0 votes not represented (SVV017668/01). The fourth case recorded in 2011 took place, for its part, in the field of Education and Culture. The Committee of the Integrated Action Programme in the field of lifelong learning (C30800) issued a negative opinion on the ‘Evaluation of the current application system for Study Visits and proposals for 2012’. In this latter case, the formal result of voting was 39 votes in favour- 256 votes against-36 abstentions and 14 votes not represented (SVV014513/01). [Table 5 near here] [Table 6 near here ] Explaining contestation How can we explain these results? This section briefly explores the explanatory power for the period 2008-2013 of two elements derived from the intergovernmental and the deliberative supranationalism theses. These are respectively: 1) the selection of 17 comitology procedure and 2) the frequency of meetings and the composition of comitology’ committees. A first interpretation inferred from intergovernmentalism would be that contestation is modulated by governments’ preferences. At the Council level, research on voting records reveals that explicit opposition is infrequent (Hagemann et al. 2017), though increasing over time (Hagemann and Franchino 2016), and that dissent can be shaped by economic preferences (Bailer et al. 2014), government ideology (Hagemann 2008) or public opinion (Hagemann et al. 2017), among other factors. At the level of comitology, these arguments are difficult to test because governments’ positions are blurred by the highly specialised and technical nature of committees as well as by the fact that voting results are not disclosed by member states. By contrast, the selection of procedure appears as a useful indicator to assess member states’ preferences. Procedural choices do matter because they are ‘political weapons’ (Blom-Hansen 2011) that reflect member states’ strategy of variable control over the Commission’s autonomy. Depending on the nature of the issues at stake in the basic legal act, member states would be inclined to opt for more or less constraining procedures at the comitology stage. If this is the case, one would expect to find more frequent and acute controversy in management, regulatory and regulatory with scrutiny committees (before the 2011 reform) and in examination and regulatory with scrutiny committees (after the 2011 reform) than in advisory committees since governments agreed on stricter control in the first two categories (Brandsma and Blom-Hansen 2010; Dogan 2000; Pollack 2003). A second idea derived from deliberative supranationalism would be that low levels of contestation are the product of a dynamic of socialisation: delegates who are experts, meet 18 regularly and frequently make decisions together might be more inclined to make efforts to understand each other’s concerns and viewpoints, and therefore more willing to transcend national interest and engage in some form of supranational ‘logic of appropriateness’ (Egeberg et al. 2003; March and Olsen, 1984; Quaglia et al. 2008). The following tables contrasts these hypotheses with empirical evidences for the period 2008-2013. To start, tables 7 and 8 provide a breakdown of voting results according to the type of comitology procedure. The data seem to broadly confirm the intergovernmental hypothesis. Although contestation can be found in all categories, dissent appears to be more frequent in committees where the procedures are more constraining for the Commission. Before the 2011 comitology reform, contestation was markedly more frequent in the regulatory with scrutiny (40.71%) and regulatory (36.32%) procedures than in the merely advisory ones (9.84%). The 2011 reform has not fundamentally altered this path: the regulatory with scrutiny (43.03%) and the examination procedures (21.02%) still produce a much higher frequency of dissent than the advisory procedure (0.75%). In summary, our data seem to ratify the idea that member states’ institutional choice for more or less constraining procedures and, consequently, control over the Commission’s implementing powers depends on the importance they attribute at the legislative stage – during the selection of the comitology procedure– to the issues at stake and their anticipation of potential inter-institutional conflict at the comitology committee level. [Table 7 near here] 19 [Table 8 near here] As regards the intensity of opposition, weak opposition is the rule for all procedures. Nevertheless, it is stronger when comitology procedures foresee tighter control. Tellingly, the four cases of qualified opposition included in our dataset correspond to one regulatory procedure prior to the 2011 comitology reform in the Health and Consumer Protection domain (V000724/01, O=0.95, year 2008) and to three examination procedures after the reform, all three in 2011: one in the Education and Culture domain (V0014513/01, O=0.79, EAC) and another in the Health and Consumer Protection domain (V017628/0, O=0.77). The third refers to the intervention of the Appeal Committee on the same SANTE issue (V017668/01, O=0.81, 2011). Cases of majoritarian opposition are distributed across more policy domains: Agriculture (AGRI), Climate (CLIMA), Mobility and Transport (MOVE) and Research and Innovation (RTD) and do also coincide with the stricter procedures. On the other hand, the sociological institutionalist hypothesis would lead us to expect that frequent meetings among experts and the habit of making decisions together should be conducive to greater levels of mutual understanding, and therefore to a higher percentage of decisions made unanimously. Our data do not lend support to this. Beyond the fact that hierarchical control and follow-up exerted by governments on their national representatives cannot be underestimated (Interview, European Commission official DG AGRI, 12 June 2014; Interview, General Secretariat of the Council official, Brussels, 1 20 March 2018; Interview, Spain’s National Representative, Ministry of Education, Madrid, 12 April 2016), there are elements of committees' functioning in practice that question the idea of a general and strong socialisation effect. First, as revealed by our dataset, there is no significant relationship between the frequency of meetings and unanimity in comitology. The Pearson’s correlation coefficient between these two variables for the period 2008-2013 has the value r= -0.06. Second, as revealed by the summary records and the attendance lists available on the comitology register, the composition of comitology committees is a complex and flexible mosaic that can vary from one meeting to another. The number, institutional affiliation and position of national representatives are not uniform and even rather eclectic (cfr. Egeberg et al. 2003). Depending on the issues at stake, a government can send a varying number of representatives and these can come from very different backgrounds. By way of example, during the 242nd meeting of the ‘Management Committee for the Common Organisation of the Agricultural Markets-All Sectors’ (CMTD(2010)1375) that took place on 16 December 2010, Belgium was represented by four members of the Flemish government and three members of the Wallonia Region Ministry, France was represented by two members of the Ministry of Agriculture and two members of France Agrimer10, Spain was represented by one member of the Ministry of Agriculture, Fishery and Food, one member of the Ministry of Industry, Tourism and Trade and one representative of the Autonomous Communities and Denmark by one member of the Danish Food Industry Agency. During the same meeting, Greece, Luxembourg and Romania were represented by members of their Permanent Representations (European Commission, S012646/01). 10 Etablissement National des Produits de l’Agriculture et de la Mer. 21 During the next meeting of the same committee, which took place one month later, on 19 January 2011, Belgium was represented by only two participants (one from the Wallonia Public Service and one from the Flemish Ministry of Agriculture), France by one member of the Ministry of Agriculture and one member of Agrimer, Spain was represented by three members of the Ministry of Environment and Denmark by one member of the Ministry of Agriculture and one member of the Danish Food Administration (European Commission, S01285701). This example illustrates the absence of a common pattern as regards both the number and institutional affiliation of the national representatives. Depending on the issue, a government can send several representatives from one Ministry, several from different ministries, a combination of ministerial representatives and agency members, only agency members or members of the Permanent Representation to the committee meeting. In other cases, external actors can also attend the meetings. For instance, representatives from the German company BiPRO GmbH participated in the meeting of the Committee for the ‘Adaptation to Scientific and Technical Progress and Implementation of the Directives on Waste Established under Article 39 of Directive 2008/98/EC (C37000) on 14 March 2011 as invited experts (European Commission, S013807/01). Second, turnout is not unimodal and constant: it can take place through physical attendance but also through a written procedure or the delegation of vote to another state. Illustratively, Cyprus was represented by Greece, Italy by France, Luxembourg by Belgium, Slovenia by Bulgaria, and Malta and Romania were not represented during the meeting held by the C34200 REACH committee (Registration, Evaluation, Authorisation and Restriction of Chemicals) in September 2012 (European Commission, S023462/01). 22 Moreover, a certain absenteeism has been observed. As revealed by our dataset, in 2008, in 43.72% of the votes, all member states were not represented despite the possibility of delivering an opinion through the written procedure or the delegation of representation. Between 2009 and 2013, the rate of non-participation decreased, but still remain significant: 33.77 % in 2009; 27.68 % in 2010 and 19.91% in 2013). At first sight, these data seem to unveil certain member states’ disinterest towards the secondary legislation of the EU. With the exception of sensitive dossiers such as those related to the biological security of the food chain or GMO, which have actually sometimes involved the intervention of the Appeal Committee, government’s attention seems to decline once the legislative moment has passed. However, an in-depth qualitative analysis of the rationale behind these data would be necessary to corroborate this idea. Last but not least, the combination of technical and political profiles with the concomitant participation of ministerial experts, agencies’ experts and members from the Permanent Representation, which is, moreover, usually, the body in charge of representing the member states in appeal in the event of conflict with the Commission’s proposals, poses questions about both the consistency and the equal composition of comitology committees. In sum, member states seem at first glance to decide the composition of their delegation on a case by case basis. The heterogeneity observed in the few examples we randomly selected to illustrate committees’ attendance patterns prevents us from drawing significant conclusions regarding the impact on national representatives’ behaviour. Yet, it suggests that a systematic scrutiny of the attendance lists and summary records of these meetings would contribute to better capturing the prevailing style of interactions in comitology committees. Recent studies on member states’ behaviour have underlined that 23 the level of individual negotiation participants has an influence on interactions and negotiation results in EU institutions (Panke 2016). Other specific qualitative studies on comitology have demonstrated that the way comitology committees internally operate is a combination of ‘integrative bargaining and technical debate’ (Siderius and Brandsma 2016). Arguably, an exhaustive analysis of the composition of committees, unveiling its multiple variations would help to refine these findings. Conclusion Delegation does not involve the absence of contestation. In this article, the objective was to refine the study of the contestation landscape in the comitology committees of the EU by systematically looking at the frequency and strength of dissent over a six-year period, encompassing the pre- and the post-comitology reform in 2011. In order to do so, we have built a novel dataset including all the voting records made available by the register of comitology for the period 2008-2013. First, these data were coded so as to identify the structure of vote and, unveil, in particular, the frequency of contestation. Then, the objective was to measure the strength of opposition and, to this effect, we proposed an index of opposition. The combination of these two exercises confirmed that 1) opposition to Commission draft implementing proposals is rare and weak and 2) the 2011 comitology reform, which is generally assumed to be a turning- point in the path of the comitology system, involving a requalification of the balance of power between the Commission and the member states, does not appear to have had a decisive impact at this level. 24 In addition to map and measure contestation in comitology committees between 2008 and 2013, the purpose of this article was to briefly explore the rationale behind dissensus. In order to do so, two hypotheses derived respectively from the intergovernmental and deliberative supranationalism theories were tested. On one hand, the selection of comitology procedure as a proxy to capture member states’ preferences. On the other hand, the frequency and composition of comitology meetings. The findings suggest that comitology procedural choice matters for contestation. When member states have strong preferences, they opt for more constraining procedures at the legislative stage and will less be inclined to engage in consensual dynamics at the comitology level. Having said this, and bearing in mind that our data only consider votes in comitology and thus do not trace the inter-institutional negotiations that precede themcases that ended with a final rejection of the draft implementing proposal presented by the Commission are extremely minoritarian. This low level of conflict and, thus apparent harmony between the Commission and the member states, seems to lend support to the idea of a certain denaturation of comitology in practice. From a theoretical point of view, the practical deviation of comitology as a vector of intergovernmentalism illustrates the contrast between original institutional design and the unexpected consequences of institutional choices over time, while, arguably, exemplifying at the same time the ‘integration paradox’ identified by Bickerton et al. (2015). Beyond the question of comitology’s role conception and of its evolution over time, these findings have important implications in terms of democratic accountability as they engage with the significant question in literature on how decisions are made at the meso-level of EU government. Further analyses would benefit from complementing these findings with qualitative data and from focusing on specific policy areas. Further research 25 would also benefit from extending the study of dissensus. First, at the legislative stage of EU policy-making, by building a bridge between the scope of contestation in the Council and its impact on comitology. Second, at the European executive level, by comparing the patterns of contestation in comitology with the situation in the case of the delegated acts channelled through Art. 290 of the TFEU. Both analyses would contribute to better capture the nature of inter-institutional relationships at the heart of the EU political system. References Bailer, S., Mattila, M. and Schneider, G. (2014) ‘Money makes the EU go round. The objective foundations of conflict in the Council of Ministers’, Journal of Common Market Studies 9:1-20. Ballmann, A., Epstein, D. and O’Halloran, S. (2002) ‘Delegation, comitology, and the separation of powers in the European Union’, International Organization 56(3): 551-574. Bergström, C.F. and Ritleng, D. (eds.), Rulemaking by the European Commission: The New System for Delegation of Powers, Oxford: Oxford University Press Bickerton, Ch, Hodson, D. and Puetter, U. (2015) ‘The New Intergovernmentalism: European Integration in the Post-Maastricht Era’, Journal of Common Market Studies 53(4): 703-22. 26 Blom-Hansen, J. (2011) ‘The EU comitology system: Taking stock before the new Lisbon regime’, Journal of European Public Policy 18(4): 607-17. Blom-Hansen, J. (2013) ‘Legislative Control of Powers Delegated to the Executive: The Case of the EU’, Governance 26(3): 425-48. Böhling, K. (2014) ‘Sidelined Member States: Commission-learning from Experts in the Face of Comitology’, Journal of European Integration 36(2): 117-34. Brandsma, G. (2010) ‘Accounting for input in comitology committees: an uncomfortable silence’, Journal of European Public Policy 17(4): 487-505. Brandsma, G., Curtin, D. and Meijer, A. (2008) ‘How Transparent are the EU ‘Comitology’ Committees in Practice?’, European Law Journal 14 (6): 819-38. Brandsma, G. and Blom-Hansen, J. (2010) ‘The EU Comitology System: What Role for the Commission?’, Public Administration 88(2): 496-512. Brandsma, G. and Blom-Hansen, J. (2012) ‘Negotiating the Post-Lisbon Comitology System: Institutional Battles over Delegated Decision-Making’, Journal of Common Market Studies 50(6): 939–57 Brandsma, G. (2013) Controlling comitology. Accountability in a Multi-Level System, London: Palgrave. 27 Brandsma, G. and Blom-Hansen, J. (2016) ‘Controlling delegated powers in the post- Lisbon European Union’, Journal of Common Market Studies 23(4): 531-49. Christiansen, T. and Dobbels, M. (2016). ‘Interinstitutional tensions in the New System for Delegation of Powers’ in Bergström, C.F and Ritleng, D. (eds.) Rulemaking by the European Commission: The New System for Delegation of Powers, Oxford: Oxford University Press, pp.87-103. Craig, P. (2016) ‘Comitology, Rulemaking, and the Lisbon Settlement. Tensions and Strains’, in C.F. Bergström and Ritleng, D. (eds.), Rulemaking by the European Commission: The New System for Delegation of Powers, Oxford: Oxford University Press, chapter 9, doi:10.1093/acprof:oso/9780198703235.003.0009 Dehousse, R. (2003) ‘Comitology: who watches the watchmen?’, Journal of European Public Policy 10(5): 798–813. Dehousse, R., Fernández Pasarín, A. and Plaza, J. (2014) ‘How Consensual is Comitology’, Journal of European Public Policy 21(6): 842-59. Dogan, R. (2000) ‘A cross-sectoral view of comitology: incidence, issues and implications’, in T. Christiansen and E. Kirchner (eds.) Committee Governance in the European Union. Manchester: Manchester University Press, pp.45-61. Egeberg, M., Schaefer, G. and Trondal, J. (2003) ‘The Many Faces of EU Committees Governance’, West European Politics 26(3): 10-40. 28 Council of Ministers and European Parliament (2011) ‘Regulation (EU) Nº 182/2011 of the European Parliament and of the Council of 16 February 2011 laying down the rules and general principles concerning mechanisms for control by Member States of the Commission’s exercise of implementing powers’, Official Journal of the European Union, L55/13, 22 February 2011, Brussels, available at https://eur-lex.europa.eu/legal- content/en/TXT/?uri=celex:32011R0182 (Last accessed: October 2019) European Commission, (2010) ‘Summary Report of the 242nd Management Committee for the Common Organisation of the Agricultural Markets-All Sectors’, Summary record Nº S01264601, 16 December 2010, Brussels, DG Agriculture and Rural Development, available at https://ec.europa.eu/transparency/regcomitology/index.cfm?do=search.documentdetail& Dos_ID=3716&DS_ID=12646&Version=1 (Last accessed: October 2019) European commission (2011) ‘Summary record of the meeting of the Committee for the adaptation to scientific and technical progress and implementation of the directives on waste established under article 39 of directive 2008/98/EC- WEEE’, Summary record Nº S013807/01, 25 March 2011, Brussels, DG Environment, available at http://ec.europa.eu/transparency/regcomitology/index.cfm?do=search.documentdetail& Dos_ID=11460&ds_id=41880&version=1&page=1 (Last accessed: October 2019) European Commission (2012) ‘Summary record of the meeting of the Committee established under the Regulation concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH)’, Summary record Nº S023462/01, 27 29 September 2012, Brussels, DG Internal Market, Industry, Entrepreneurship and SMEs, available at http://ec.europa.eu/transparency/regcomitology/index.cfm?do=search.documentdetail& Dos_ID=11663&ds_id=41836&version=1&page=1 (Last accessed: October 2019) European Commission (2013) Report on the working of committees during 2012, COM (2013) 701 final, Brussels, 10 October Fernández Pasarín, A. (2016) ‘The state back in. La comitologie ou la delegation contrôlée et coordonnée du pouvoir’, Revue Française d’Administration Publique 158(2):463-75. Hagemann, S. (2008) ‘Voting, statements and coalition-building in the Council from 1999-2006’. In D. Naurin and H. Wallace (Eds.) Unveiling the Council: Games governments play in Brussels (London:Palgrave Macmillan), pp.36-63. Hagemann, S. and Franchino, F. (2016) ‘Transparency vs efficiency? A study of negotiations in the Council of the European Union’, European Union Politics 17(3): 408-428. Hagemann, S., Hobolt, S. and Wratil, C. (2017) ‘Government responsiveness in the European Union: Evidence from Council voting’, Comparative Political Studies 50(6): 850-76. Hayes-Renshaw, F., Van Aken, W. and Wallace, H. (2006), ‘When and why the EU 30 Council of Ministers votes explicitly’, Journal of Common Market Studies 41(1): 161- 94. Joerges, Ch. and Neyer, J. (1997a) ‘From intergovernmental bargaining to deliberative political processes: the constitutionalisation of comitology’, European Law Journal 3(3): 273–99. Joerges, Ch. and Neyer, J. (1997b) ‘Transforming strategic interaction into deliberative problem-solving: European comitology in the foodstuffs sector’, Journal of European Public Policy 4(4): 609–25. March, J.G. and Johan P. Olsen. 1984. “The New Institutionalism: Organizational Factors in Political Life.” American Political Science Review 78 (September): 734–49. Mattila, M. (2004). “Contested decisions: Empirical analysis of voting in the European Union Council of Ministers”. European Journal of Political Research 43(1): 29-50. Panke, D. (2016) ‘Quantitative Analysis: Negotiation Success’ in Small States in the European Union: Coping with Structural Disadvantages (London: Routledge), chapter 8, pp.127-42. Pollack, M. (2003) ‘Control mechanism or deliberative democracy? Two images of comitology’, Comparative Political Studies 36(1–2): 125–55. 31 Quaglia, L., De Francesco, F and Radaelli, C. (2008) ‘Committee governance and socialization in the European Union’, Journal of European Public Policy 15(1): 155-66. Siderius, K. and Brandsma, G. (2016) ‘The Effect of Removing Voting Rules: Consultation Practices in the Commission’s Delegated Act Experts Groups and Comitology Committees’, Journal of Common Market Studies 54(6): 1265-1279. Spendzharova, A. and Versluis, E. (2013) ‘Issue salience in the European policy process: what impact on transposition?’, Journal of European Public Policy 20(109:1499-1516. Trondal, J. and Veggeland, F. (2003) ‘Access, voice and loyalty: the representation of domestic civil servants in EU committees’, Journal of European Public Policy 10(1): 59- 77. Wetendorff Nørgaard, R., Nedergaard, P. and Blom-Hansen, J. (2014) ‘Lobbying in the EU Comitology System’, Journal of European Integration 36(5): 491-507. Interviews Interview European comitology lobbyist, Brussels, 24/07/2015. Interview European Commission’s Official, DG SANCO, Brussels, 13/06/2014. Interview General Secretariat of the Council’s Official, DG Agriculture, Fisheries, Social Affairs and Health, Brussels, 01/03/2018. Interview European Commission’s Official, DG AGRI, Brussels, 12/06/2014. 32 Interview Spanish Diplomat, Ministry of the Presidency, Madrid, 24/09/2015. Interview Spain’s National Representative, Ministry of Education, Culture and Sport, Madrid, 12/04/2016. Tables Table 1. Final opinions delivered by comitology committees (2008-2013) Table 2. Frequency of contestation (2008-2013) Table 3. Strength of opposition (2008-2013) Table 4. Strength of opposition. The 2011 comitology reform effect Table 5. Frequency of contestation. Breakdown by policy domains (2008-2013) Table 6. Strength of opposition. Breakdown by policy domains (2008-2013) Table 7. Frequency of contestation. Breakdown by type of procedure (2008-2013) Table 8. Strength of opposition. Breakdown by type of procedure (2008-2013) Figures Figure 1. Scale of opposition in voting procedures Figure 2. Distribution of opposition in voting procedures (2008-2013) 33