1.Introduction

In a legal system characterised by intertwined levels of administrative authority, the implementation of EU law is necessarily a complex and multifaceted phenomenon, affecting the institutional actors involved both in the vertical and horizontal dimensions. At the same time, the legitimacy and accountability of the institutions and bodies entrusted with implementing powers is of crucial importance since, being an expression of the public power, the exercise of these powers towards individuals needs to comply with the tenets of democracy and the rule of law on which the European Union is based.Consolidated Version of the Treaty on European Union [2008] OJ C 115/13, art 2; Case 294/83, Les Verts v European Parliament EU:C:1986:166, para 23. However, holding these powers to account implies the need to have a clear overview of the mechanisms at stake and of the actors involved, to ensure that they do not remain hidden from view and, consequently, from the accountability mechanisms.With regard to the notion of accountability, we will adopt the notion proposed by Bovens, considering accountability as a “relationship between the actor and a forum, in which the actor has the obligation to explain and justify his or her conduct, the forum can pose questions and pass judgments, and the actor might face the consequence”. See Mark Bovens, ‘Analysing and Assessing Accountability: A Conceptual Framework’ [2007] European Law Journal 13(4), 447. See also Richard Mulgan, ‘Accountability: An Ever-expanding Concept?’ [2002] Public Administration 555-573, cited in Anthony Arnull & Daniel Wincott, Accountability and Legitimacy in the European Union (Oxford University Press 2002) 2; Madalina Busuioc, ‘Accountability, Control and Independence: The Case of European Agencies’ [2009] European Law Journal 607.

Therefore, this paper aims at bringing under the spotlight forms of implementation which have developed in the shadow of primary law and of the generally acknowledged hierarchy of EU norms. In particular, it will focus on the phenomena of “agencification”On the phenomenon of agencification, see, inter alia, Michelle Everson, Cosimo Monda & Ellen Vos, European Agencies. In between Institutions and Member States (Wolters Kluwer 2014); Merijn Chamon, EU Agencies. Legal and Political Limits to the Transformation of the EU Administration (Oxford University Press 2016); Edoardo Chiti, Le agenzie europee. Unità e decentramento nelle amministrazioni comunitarie (Cedam 2002); Carlo Tovo, Le agenzie decentrate dell’Unione europea (Editoriale Scientifica 2016); Jacopo Alberti, Le agenzie dell’Unione europea (Giuffrè 2018). and harmonised standardisation.On European standardisation, see, inter alia, H. Schepel, The Constitution of Private Governance (Hart Publishing 2006); Megi Medzmariashvili, ‘Delegation of Rulemaking Power to European Standards Organizations: Reconsidered’ [2017] Legal Issues of Economic Integration 353; Mariolina Eliantonio & Caroline Cauffman, The Legitimacy of Standardization as a Regulatory Technique in the EU (Edward Elgar, forthcoming). Thus, it will investigate to what extent the acts of EU agencies and European Standardisation Organisations (ESOs) may fall within the notion of implementation and what are the accountability mechanisms in place, as well as their problematic aspects. In particular, the scope of the analysis will be limited to the two most important aspects of accountability from the perspective of the rule of law. On the one hand, the political or democratic accountability of the relevant bodies vis-à-vis the European Parliament and, on the other, the judicial accountability before the Court of Justice of the European UnionFor the notions of political and judicial accountability, see Madalina Busuioc, European agencies: law and practices of accountability (Oxford University Press 2013) 46-64. will be considered in order to provide a limited, yet clear picture of the oversight of these institutions’ implementation of EU law.

2.The Plurality of Levels in Article 291 TFEU

The Lisbon Treaty has reformed the EU institutional framework profoundly and, especially, the regime for the adoption of non-legislative acts by the European Commission. Inspired by the need to establish a clearer separation of powers and hierarchy of norms in the EU, it has introduced a categorisation of legal acts, which was meant to simplify the legal framework for secondary rule-making.Final Report of Working Group IX on Simplification (Conv 424/02, WG IX 13), 9. See also, inter alia, Robert Schutze, ‘Sharpening the Separation of Powers through a Hierarchy of Norms?’ (2005) EIPA Working Paper; Deirde Curtin, ‘European Union Executives: out of the Shade, into the Sunshine?’ in J. De Zwaan, J. Jans & F. Nelissen (ed.), The European Union. An Ongoing process of integration (TMC Asser Press 2004) 99. With the splitting into two of what previously constituted a single regime for the implementation of EU law, the Lisbon Treaty now distinguishes between delegated acts, regulated in Article 290 TFEU, and implementing acts, in Article 291 TFEU.

The latter provision, in particular, confers the primary responsibility of implementing EU law on the Member States, giving express formulation in primary law to a principle already emerging in the case law.See Joined Cases 205-215/82, Deutsche Milchkontor GmbH and others v Federal Republic of Germany EU:C:1983:233, para 17. See also Joined Cases 89 and 91/86, L’Étoile commerciale and Comptoir national technique agricole (CNTA) v Commission of the European Communities EU:C:1987:337, para 11; Case C-476/93 P, Nutral v Commission EU:C:1995:401, para 14. Although it is expressed in terms of obligation for the Member States to adopt the necessary measures, the provision is considered to mean more than the general duty of “sincere cooperation” codified in Article 4 TEU.See Robert Schutze, ‘From Rome to Lisbon: “Executive Federalism” in the (New) European Union’ [2010] Common Market Law Review 1385, 1398. Indeed, Article 291(1) TFEU embraces the model of executive federalism (Vollzugsföderalismus) more clearly, according to which the competence to implement legislative acts is reserved for the Member States.See, inter alia, Schutze (n. 8) 1391; J.P. Jacqué, ‘Pouvoir législatif et pouvoirs exécutif dans l’Union européenne’ in J.B. Auby & J. Dutheil de la Rochère (ed.), Traité de droit administratif européen (Bruylant 2014) 50. Thus, it not only confirms and constitutionalises the rule that, in line with the needs of subsidiarity,Referring to “executive federalism” as a “specific application of the principle of subsidiarity”, Merijn Chamon, ‘The Influence of Regulatory Agencies on Pluralism in European Administrative Law’ [2012] Review of European Administrative Law 61, 66. On the fact that the subsidiarity principle also includes executive action, see Schutze (n. 8) 1411. implementation of EU law is carried out by indirect administration at the national level, but also, according to some authors, it recognises an “autonomous national competence” of the Member States”.Schutze (n. 8) 1397-1398. Therefore, Article 291 TFEU should “not be viewed from a horizontal separation of powers perspective, but […] from a vertical perspective that places Article 291 TFEU within the context of Europe’s executive federalism”.Robert Schutze, European Constitutional Law (Cambridge University Press 2012) 241. See also H. Hofmann, ‘Legislation, Delegation and Implementation under the Treaty of Lisbon: Typology Meets reality’ [2009] European Law Journal 482, 497-498.

Conversely, pursuant to Article 291(2) TFEU, where uniform conditions are needed, the EU is entitled to implement its own legally binding acts, allocating at the EU level the implementation which would be otherwise carried out by the Member States. In this case, the legislator shall confer implementing powers on the Commission or, “in duly justified specific cases”, on the Council, which thus retains its executive role. Therefore, Article 291(2) TFEU represents the legal basis for implementation at the EU level when justified by an “objective cause”,Opinion of Advocate General Cruz Villalón in Case C-427/12, European Commission v European Parliament and Council of the European Union EU:C:2013:871, para 50. i.e. the need to provide uniform conditions in the application of binding acts throughout the Union’s territory. As effectively expressed by Advocate General Cruz Villalón, “Article 291(2) TFEU is therefore primarily a rule that empowers the European Union, through the Commission, to use, in a subsidiary manner, a competence that belongs to the Member States.”Ibid [49].

Thus, Article 291 TFEU delineates a system clearly articulated in a plurality of levels, providing for the implementation to be carried out either at the national or at the EU level. However, it is noteworthy that, in the composite reality of EU administration, the distinction between direct and indirect administration is becoming increasingly blurred,See, inter alia, Sabino Cassese, ‘European Administrative Proceedings’ [2004] Law and Contemporary Problems 21; Giacinto Della Cananea, ‘The European Union’s Mixed Administrative Proceedings’ [2004] Law and Contemporary Problems 197. making the vertical division of competences more complex than the image enshrined in Article 291 TFEU.Hofmann (n. 12) 498. Not only are the Member States are entrusted with the control of the adoption of the implementing acts of the Commission through the comitology system, but their role is significantly relevant also in relation to other phenomena of implementation which have emerged in the reality of EU administration but which do not appear in the lines of Article 291 TFEU, such as EU agencies. From this perspective, the role of the Member States in the control of the exercise of implementing powers finds its logic in the multilevel composition of EU institutional framework and in the interplay between the different levels.

3.The De Facto Implementing Powers of EU Agencies

While the plurality of levels is clearly visible, a substantial part of the implementation activities carried out at the EU level remains “invisible” from the reading of Article 291 TFEU. In particular, primary law does not give account of the fundamental role which EU agencies have acquired in the implementation of EU policies.The “constitutional neglect” of EU agencies in Article 291 TFEU is even more remarkable considering that, during the Convention on the Future of Europe, the Parliament adopted a resolution on the hierarchy of norms in EU acts where it specifically recommended the addition of agencies in what is now Article 291 TFEU as bodies also exercising implementing powers. See European Parliament, Resolution on the typology of acts and the hierarchy of legislation in the European Union (2002/2140(INI)) OJ C 31/126. The establishment and the empowerment of permanent bodies with separate legal personality under EU public law, set up by the institutions through secondary legislation,According to the definition of EU agencies provided by Chamon (n. 3) 9. Compare this definition with the one proposed by Busuioc (n. 5) 21. undeniably represents a formidable development of EU administration. In the last few decades, agencification has progressively grown both in quantitative and qualitative terms. Indeed, from the first agencies in the 1970s, which were entrusted with merely informational tasks, the number of decentralised agencies has exponentially increased to more than 30 bodies, which are nowadays called to exercise important powers in complex and politically sensitive domains.For a recent overview of this phenomenon, see the study carried out by Ellen Vos on behalf of the European Parliamentary Research Service, ‘EU Agencies, Common Approach and Parliamentary Scrutiny’ http://www.europarl.europa.eu accessed 12 January 2019.

Although this empowerment without an express legal basis in the Treaties and without clear constitutional boundaries raised relevant concerns on the legitimacy of these bodies,Inter alia, Damien Geradin, ‘The Development of European Regulatory Agencies: Lessons from the American Experience’ in Damien Geradin, Rodolphe Munoz & Nicolas Petit (ed.), Regulation through agencies in the EU: a new paradigm of European governance (Edward Elgar 2005) 231; Ellen Vos, ‘Reforming the European Commission: What Role to Play for EU Agencies?’ [2000] Common Market Law Review 1113. the conferral of far-reaching powers to EU agencies was unequivocally sanctioned by the Court in the Short Selling judgment.See Case C-270/12, UK v Council of the European Union and European Parliament (Short Selling) EU:C:2014:18. In the light of the significant steps taken by the Lisbon Treaty in the institutionalisation of EU agencies,In particular, the insertion of EU agencies in Articles 263, 265, 267 and 277 TFEU. Inter alia, Curtin (n. 6) 146. The “constitutionalisation” of EU agencies remains, however, “unfinished” to an important extent, see Vos (n. 20) 43; M. Everson & E. Vos, ‘Unfinished Constitutionalisation: The Politicised Agency Administration and Its Consequences’, Paper presented at the TARN Conference (Florence, 10-11/11/2016), 14. the Court recognised the possibility to exercise decision-making powers under certain conditionsSee, in particular, Case C-270/12, United Kingdom of Great Britain and Northern Ireland v European Parliament and Council of the European Union EU:C:2013:562, para 45-47. and to adopt acts vis-à-vis third parties in the shadow of primary law and of the hierarchy of norms introduced by the Lisbon Treaty. Indeed, the Lisbon Treaty did not establish a “single legal framework under which certain delegated and executive powers may be attributed solely to the Commission”.Ibid [78]. See also Opinion of AG Jääskinen in Joined Cases C-404/12 P, Council of the European Union and European Commission v Stichting Natuur en Milieu and Pesticide Action Network Europe EU:C:2014:309, para 35. As held also in Spain v Council, “acts of secondary legislation may establish implementing powers outside the regime laid down in Article 291 TFEU”.Case C-521/15, Kingdom of Spain v Council of the European Union EU:C:2017:982, para 43.

In spite of the “constitutional neglect” of EU agencies in Article 291 TFEU,Michelle Everson & Ellen Vos, ‘European Agencies: What about the Institutional Balance?’ (2014) Maastricht Faculty of Law Working Paper 4/2014, 14. agencies can adopt acts of individual or general application which are amenable to judicial review before the Court of Justice. On one hand, agencies like EUIPO or CPVO have the power to adopt decisions of individual application.EUIPO and CPVO are examples of “genuine decision-making agencies” as defined by Stefan Griller and Andreas Orator, ‘Everything under Control? The “Way Forward” for European Agencies in the Footsteps of the Meroni Doctrine’ [2010] European Law Review 35(1), 13-15. Considering the nature of the powers, it appears that, since they apply the general rule to the particular case, they act within the realm of implementation.See Vos (n. 22) 44; M. Chamon, op. cit. (2014), 397; Tovo (n. 3) 73; D. Ritleng, ‘La nouvelle typologie des actes de l’Union. Un premier bilan critique de son application’ [2015] Revue trimestrielle de droit européen 28. See also Case C-16/88, Commission v Parliament EU:C:1989:397,

para 11; Case C-42/97, Commission v Parliament and Council EU:C:1999:81, para 37. In this sense, they are to be considered as atypicalSince Article 288 TFEU recognises as typical acts only the acts adopted by the EU institutions, from a formal perspective the agencies’ acts cannot fall within the category of implementing acts, but from a substantive perspective the powers delegated to EU agencies may be assimilated to the powers conferred on the Commission under Articles 291 TFEU. See Tovo (n. 3) 75. or de facto implementing acts.M. Chamon, ‘Institutional Balance and Community Method in the Implementation of EU Legislation Following the Lisbon Treaty’ [2016] Common Market Law Review 15501, 1536. See also Tovo (n. 3) 269. On the other hand, the same conclusion can be reached in relation to agencies’ acts of general application, although the powers conferred on these agencies cannot be defined a priori as implementing powers.Tovo (n. 3) 75. Indeed, acknowledging that one must be particularly careful to apply the terminology of Articles 290 and 291 TFEU to EU agencies,In this regard, it is important to underline that, although the Meroni judgment uses the notion of “executive powers”, they cannot be automatically assimilated to the concept of implementing acts, see C.F. Bergström, ‘Shaping the New System for Delegation of Powers to EU Agencies: United Kingdom v European Parliament and Council (Short Selling)’ [2015] Common Market Law Review 52, 238-239. the acts of these agencies may even supplement the relevant legislative acts, especially after the Short Selling judgment.Ibid [238]; J. Alberti, ‘Delegation of Powers to EU Agencies after the Short Selling Ruling’ [2015] Il Diritto dell’Unione European 480. However, it is arguable that the powers currently conferred to EU agencies, even the ones adopting acts of general application like ESMA or ECHA, are limited to implementing the relevant legislative acts since they provide further detail in relation to the content of the legislative acts especially in these highly technical domains.“To provide further detail in relation to the content of a legislative act” is the definition of implementing powers given by the Court in Case C-427/12, European Commission v European Parliament and Council (Biocides) EU:C:2014:170, para 39; Case C-88/14, Commission v Parliament and Council (Visa Reciprocity) EU:C:2015:499, para 30.

4.Harmonised Standardisation as a Form ofImplementation

The recent case law of the Court of Justice has unveiled the significance of another interesting phenomenon in the context of the implementation of EU law. Although the role of harmonized standardisation in support of the internal market policy dates back to the New Approach,See Commission of the European Communities, ‘Completing the Internal Market. White Paper from the Commission to the European Council’ (Milan, 28-29 June 1985) COM(85) 310 final; Council Resolution of 7 May 1985 on a New Approach to Technical Harmonization and Standards, OJ 1985 C 136/1; Council Resolution of 21 December 1989 on a Global Approach to Conformity Assessment, OJ 1990 C 10/1. recent developments in EU legislationRegulation EU No 1025/2012 on European Standardisation, OJ L 316/12. and in the case lawSee, inter alia, Case C-171/11, Fra.bo. v DVGW EU:C:2012:453. have led to a “juridification” of the harmonised standards,Inter alia, H. Schepel, ‘The New Approach to the New Approach: The Juridification of Harmo-nised Standards in EU Law’ [2013] Maastricht Journal of European and Comparative Law 521; R. Van Gestel & H.W. Micklitz, ‘European Integration through Standardisation: How Judicial Review Is Breaking Down the Club House of Private Standardisation Bodies’ [2013] Common Market Law Review 145. which has significant constitutional implications.

In this regard, it is important to recall that the New Approach consists of regulating through legislative acts only the essential requirements of general interests of a product, while referring the detailed definition of technical aspects to private organisations, composed of experts and representatives of the business sector, i.e. the ESOs. According to this method, after the adoption of a New Approach directive or regulation, the European Commission issues a request (also called “mandate”) to one of the ESOs to elaborate a document defining the technical rules applicable to the product. The subsequent publication of a reference in the Official Journal by the Commission endows these technical standards with the legal nature of “harmonised standards” and provides a presumption of conformity with the secondary law measures they are linked to.

In the James Elliott case, the Court for the first time established its jurisdiction to give a preliminary ruling on the interpretation of an harmonised standard adopted according to the Construction Products Directive.Case C-613/14, James Elliott Construction Limited v Irish Asphalt Limited EU:C:2016:821. While in his Opinion Advocate General Campos Sánchez-Bordona argued that the standards “should be regarded as ‘acts of the institutions, bodies, offices or agencies of the Union’”Opinion of Advocate General Campos Sánchez-Bordona in Case C-613/14, James Elliott Construction Limited v Irish Asphalt Limited EU:C:2016:63, para 40. and they constitute a “case of ‘controlled’ legislative delegation in favour of a private standardisation body”,Ibid [55]. the Court based its reasoning on the function of harmonised standards in relation to the relevant directive. In particular, the Court stressed that, although non-binding, the standard at issue is “a necessary implementing measure” of the Construction Products DirectiveIbid [43]. and, therefore, it can be considered as “part of EU law”.Ibid [40].

Building on the observations of the Court, it appears that, from a material perspective, harmonized standards bear strong commonalities with implementing acts.Carlo Tovo, ‘Judicial Review of Harmonised Standards: Changing the Paradigms of Legality and Legitimacy of Private Rulemaking under EU Law’ [2018] Common Market Law Review 1187, 1196. Adopted pursuant to a specific request of the Commission in accordance to the relevant New Approach legislative act, harmonized standards “support the implementation of Union legislation”European Commission, Vademecum on European Standardisation, SWD(2015) 205, 8-9. and “give concrete form on a technical level” to the relevant legislative act.Elliott (n. 39) [36]. Therefore, these specifications can be seen as acts de facto implementing EU legislation,In the same vein. Tovo (n. 44) 1195-1196; Linda Senden, ‘The Constitutional Fit of European Standardization Put to the Test’ [2017] Legal Issues of Economic Integration 337, 350. However, the neo-comitology system could apply to the Commission’s implementing decision containing the reference to the standard. adopted outside the regime laid down in Article 291 TFEU and in Regulation 182/2011 (the “Comitology Regulation”).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 [2011] OJ L 55/13.

In this respect, the possibility to consider the harmonized standards as acts of the European Commission has been debated. The Commission exercises significant control over the procedure for the drafting of harmonised technical standards, both ex ante (in the request issued pursuant to the relevant New Approach legislative act) and ex post (through the review prior to the publication in the Official Journal and through the procedure for lodging objections set forth by Article 11 of Regulation 1025/2012). Remarkably, the incorporation of harmonised standards in EU law and the production of their legal effects, namely the presumption of conformity, depend on the decision of the Commission to publish their reference in the Official Journal.The legal qualification of the publication is not clear in the Regulation, but it was considered a “decision” in Case T-474/15, GGP Italy v Italy, para 61. See also Tovo (n. 44) 1198. Also the recent shift in the practice of the Commission to publish the reference in the form of an Implementing Decision in the L seriesSee, for instance, Commission Implementing Decision (EU) 2018/2048 of 20 December 2018 on the harmonised standard for websites and mobile applications drafted in support of Directive (EU) 2016/2102 of the European Parliament and of the Council [2018] OJ L 327/84. (instead of in the form of a Communication in the C series) points to the direction of a more clear responsibility of the Commission.See also the Opinion of Advocate General Bobek in Case C-587/15, Dockevičius and Dockevičienė EU:C:2017:234, para 88: “once an EU institution decides to incorporate an originally external legal act into EU law and to draw legal consequences from it by effectively enforcing it internally […] the same institution cannot later turn a blind eye and suggest that since that act was originally drafted by a third party, it is therefore not an act of that institution”. However, the Court repeatedly maintained that harmonised standards are acts of private organisations.See Elliott (n. 39) [34] , and Case C-185/17, Mitnitsa Varna v SAKSA OOD EU:C:2018:108,

para 38. Neither the control exercised by the CommissionElliott (n. 39) [34]. nor the publication in the Official JournalDockevičius and Dockevičienė (n. 51) [39]. are sufficient for the standards to be attributed to the Commission, casting a shadow of doubt on the responsibility and, ultimately, on the accountability of harmonized standardisation.See, inter alia, Carlo Colombo and Mariolina Eliantonio, ‘Harmonized technical standards as part of EU law: Juridification with a number of unresolved legitimacy concerns’ [2017] Maastricht Journal of European and Comparative Law 323.

5.Controlling the Visible: The Democratic Account-ability of the Commission and the Council in the Adoption of Implementing Acts

In the light of these observations, recognising the functional similarity in the exercise of implementing powers by the Commission or the Council, by EU agencies and in harmonised standardisation, sheds light not only on the complexity of EU administration hidden behind the lines of Article 291 TFEU, but also calls for an assessment of the accountability of these forms of implementation. Indeed, the exercise of de facto implementing powers in the shadow of primary law entails the risk of circumventing the guarantees of democratic oversight which were progressively developed in relation to formal implementing acts, jeopardising the legitimacy of the EU legal system.

In this regard, it is important to recall that, in the 60 years of EU integration, the control of the Commission’s implementing acts has been the object of fierce interinstitutional battles which have seen, on one hand, the resistance of the Council to leave broad discretion to the Commission in this domain – which led to the creation of the highly idiosyncratic system of committees (i.e. the comitology system)For an analysis of the origin and the evolution of the comitology system (after Lisbon, neo-comitology), see D. Bianchi, De Comitatibus. L’origine et le rôle de la comitologie dans la politique agricole commune (L’Harmattan 2012); C.F. Bergström, Comitology. Delegation of Powers in the European Union and the Comitology System (Oxford University Press 2005). – and, on the other hand, the increasing role of the Parliament in its oversight due to a constant political pressure in this sense.Including the use of its budgetary powers and the conclusion of several inter-institutional agreements, see, inter alia, Kieran St Clair Bradley, ‘The European Parliament and Comitology: On the Road to Nowhere?’ [1997] European Law Journal 230; Jurgen Bast, ‘New Categories of Acts after the Lisbon Reform: Dynamics of Parliamentarization in EU Law’ [2012] Common Market Law Review 885. As a result of this evolution, the exercise of the Commission’s powers under Article 291 TFEU is currently embedded in a legislative framework, namely the Comitology Regulation, which provides for a direct control of Member States’ representatives through different committee procedures and an indirect control of the Parliament, formally at equal footing with the Council.Comitology Regulation, art 11. However, on the increasing substantive asymmetry between the Parliament and the Council, see P. Craig, ‘Delegated Acts, Implementing Acts and the New Comitology Regulation’ [2011] European Law Review 686.

In particular, pursuant to Article 11 of the Regulation, when the basic act is adopted under the ordinary legislative procedure, the Parliament and Council have a right of scrutiny (“droit de regard”) in relation to the exercise of implementing powers by the Commission. This consists of the possibility to indicate, at any time, that “a draft implementing act exceeds the implementing powers provided for in the basic act.”Comitology Regulation, art 11. In other words, the Council and the Parliament can react when they consider that the exercise of the powers is ultra vires: going beyond the scope of the empowerment.

The use of this mechanism is rather exceptionalBy the end of January 2016, the right of scrutiny was never used by the Council, while the Parliament used it only four times. Commission, ‘Report to the European Parliament and the Council on the Implementation of Regulation (EU) 182/2011’ COM(2016) 92 final, 8. and the consequence of the exercise of this right, in any case, is limited, especially in comparison to the objection and revocation powers under Article 290 TFEU. The Commission is bound to “review the draft implementing act, taking account of the positions expressed”.Comitology Regulation, art 11. Hence, it remains free to maintain, amend or withdraw the measure, duly informing the Parliament and the Council of its decision. Therefore, the right of scrutiny does not result in an actual right of veto. Should the Commission insist on the adoption of the implementing act allegedly considered to be ultra vires, the Council or the Parliament could react only by bringing an action before the Court of Justice to annul the measure.Bast (n. 57) 913; C. Blumann, ‘Un nouveau depart pour la comitologie. Le règlement n° 182/2011 du 16 février 2011’ [2011] Cahiers de droit européen 34.

Although rather “toothless”,Daniela Corona, ‘The Adoption of Secondary Legislation through Comitology in the EU: Some Reflections on the Regulation (EU) 182/2011 in Comparison with the Pre-Lisbon Reform’ [2014] The Theory and Practice of Legislation 100. the right of scrutiny guaranteed to the Parliament in this domain is still relevant in terms of oversight and, together with the Comitology Register,Comitology Regulation, art 10. it grants a certain degree of transparency and accountability to this form of implementationFor an analysis of the transparency of the comitology system, see G.J. Brandsma, D. Curtin & A. Meijer, ‘How Transparent Are EU “Comitology” Committees in Practice?’ [2008] European Law Journal 819. which, conversely, appears to be missing in the case of implementing acts of the Council. Indeed, in comparison to the highly formalised procedures for the adoption of implementing acts by the Commission, the exercise of implementing powers by the Council stands out due to the absence of a specific procedure for their adoption. Not specifically distinguished from the legislative procedure and the exercise of powers directly conferred by the Treaties, the procedural rules for the adoption of implementing acts do not differ from the ones applicable to the decision-making activities of the Council in general,For a description of the Council’s procedures, see Martin Westlake & David Galloway,

The Council of the European Union (John Harper Publishing 2004). The only relevant difference in this regard is the fact that draft legislative acts must be debated “in public” according to Art. 16(8) TEU. However, a form of “publicity” of the deliberations is ensured also for certain non-legislative acts by Articles 8 and 9 of the Council’s Rules of Procedure. which do not involve the Parliament. The absence of a role for the Parliament, which has neither a right of veto nor a right of scrutiny in the adoption of Council’s implementing acts, raises significant concerns from the accountability and institutional balance perspectives, especially in the light of the inherent risk of “sliding of powers” from the hands of the Parliament to the Council.On the issue, see the so-called Aglietta Report: Report of 3 August 1998 drawn up on behalf of the Committee on Institutional Affairs on the modification of the procedures for the exercise of implementing powers conferred on the Commission-comitology (Rapporteur: Marie Adelaide Aglietta), EP Doc A4-292/99, cited in C.F. Bergström, op. cit. (2005), 253.

6.Controlling the Invisible: The Democratic Accountability of EU Agencies and ESOs

In comparison with the accountability mechanism established in relation to the Commission’s implementing acts, the regimes applicable to EU agencies and ESOs arguably appear rather embryonic and also problematic to a certain extent.

6.1.The Accountability of EU Agencies to the Parliament

With regard to the implementing powers of EU agencies, it is important to underline that, often created to ensure credible policy commitments in controversial fields, and sometimes precisely as a reaction to credibility failures of EU institutions, EU agencies are meant to operate “at arm’s length from politics and political control”:Busuioc (n. 5) 114. “the independence of their technical and/or scientific assessments is […] their real raison d’être”.Commission, ‘The Operating Framework for the European Regulatory Agencies’ COM(2002) 718 final, 5. With the growth of agencification in quantitative and qualitative terms, however, the independence of these bodies was increasingly perceived as problematic, raising concerns that EU agencies might become “uncontrollable centres of arbitrary powers”.Michelle Everson, ‘Independent Agencies: Hierarchy Beaters?’ [1995] European Law Journal 180, 183. For this reason, proposals to strengthen the oversight of the Parliament, the Council or the Commission on the exercise of the delegated powers by EU agencies were put forward, both in the literatureSee, inter alia, Griller & Orator (n. 27) 27-29. and in official documents.See, inter alia, European Parliament, Resolution on the typology of acts and the hierarchy of legislation in the European Union, OJ C 31E/126 of 2002, point 17. The Parliament proposed, in particular, to introduce a scrutiny mechanism allowing the Commission, the Council and the Parliament to repeal agencies’ acts. On the relevance of control for the legality of the dele-gation of powers to these bodies, see Joined Cases 9/56 and 10/56, Meroni EU:C:1958:4, 152; Short Selling (n. 21) [50].

This tension between accountability and independence of EU agencies results in a peculiar mix of control mechanisms which stems from the institutional design and the obligations of these bodies.There is abundant literature regarding the accountability of EU agencies and its limits. See, inter alia, Busuioc (n. 5); Curtin (n. 6) 523-541; Anthony Arnull & Daniel Wincott, Accountability and Legitimacy in the European Union (Oxford University Press 2002); Busuioc & Groenleer, op. cit. (2014), 175-200; M. Buess, ‘European Union Agencies’ Vertical Relationships with the Member States: Domestic Sources of Accountability?’ [2014] Journal of European Integration 509; Busuioc (n. 5) 599-625; M. Scholten, ‘Independence v. Accountability: Proving the Negative Correlation’ [2014] Maastricht Journal of European and Comparative Law 197. Focusing on the control the Parliament exercises on the operation of EU agencies, it is noteworthy that this institution’s involvement with the activities of the agencies has progressively increased with the evolution of agencification,F. Jacobs, ‘EU Agencies and the European Parliament’ in M. Everson, C. Monda & E. Vos (ed.), European Agencies in between Institutions and Member States (Wolters Kluwer 2014) 202. culminating in the adoption of the Common Approach in 2012.Common Approach annexed to the Joint Statement of the European Parliament, the Council of the EU and the European Commission on decentralised agencies, signed on 19 July 2012.

In this regard, firstly, the Parliament appears to have gained an increased role in the appointment process of leadership positions within EU agencies.Jacobs (n. 74) 203. In particular, contrasted to the right to nominate one or two members of the Management Board, in some cases the Parliament was given a role in the appointment (and re-appointment)FRA Regulation, art 15. process of the agency’s director, who must appear before the Parliament before being formally appointed.See, for instance, in the case of EFSA and ENISA. However, the insertion of these provisions in the basic regulations of the different agencies is remarkably uneven,Compare, for instance, the founding regulations of ETF and EMA with the one of GSA or ESMA. For a comprehensive overview, see European Parliament’s study ‘EU Agencies, Common Approach and Parliamentary Scrutiny’, Annex 5, http://www.europarl.europa.eu, accessed

12 January 2019. and, although the requirement of the approval of the Parliament of the director of an agency was expressly debated in the negotiations,Jacobs (n. 74) 218. no provision was inserted in this respect in the Common Approach.

Secondly, the Common Approach codifies the practice of requiring the agencies to submit the annual report to the Parliament and to consult it on the multiannual work programmes of agencies.Common Approach annexed to the Joint Statement of the European Parliament, the Council of the EU and the European Commission on decentralised agencies, signed on 19 July 2012, paras 49 and 58. It is clarified, however, that “the purpose of the consultations is an exchange of views and the outcome is not binding on the agency”. Moreover, EU agencies’ repre-sentatives may be obliged to attend hearings before the Parliament, which may take place before the budgetary committees or the policy-specific committees of this institution.Busuioc (n. 5) 121. Empirical research, however, has shown that, due to information asymmetries and volatile political interests in this accountability exercise, there is a relevant underuse of these mechanisms, resulting in inadequacies in ensuring a comprehensive and systematic scrutiny of the agencies’ activities.Ibid [138].

Finally, the Common Approach innovatively introduced the so-called alert/warning system.Common Approach, para 59. This mechanism allows the Commission to react when it has “serious reasons of concern” related to certain Management Board’s decisions.The limitation of this power to the cases where the measures are adopted by the Management Board appears problematic. While the use of such a system would be pleonastic in relation to measures to be formally adopted by the Commission, in the variety of agencies’ procedures there might be cases of controversial measures adopted by other internal bodies. See, for instance, Council Regulation (EC) No 2100/94 of 27 July 1994 on Community plant variety rights, OJ L 227/1, art 35. In this case, the Commission “will raise formally the question in the Management Board”, requesting it to suspend the adoption or the implementation of the contested measure.Common Approach, para 59. Notably, this mechanism gives the Commission the power to react before the adoption of an unlawful measure by the agency, thus intervening during the exercise of the delegated powers and not just ex post. Should the request be ignored, the Commission will formally inform the Parliament and the Council. In particular, the system can be activated when the Board is about to adopt measures which are firstly, non-compliant with the mandate of the agency, secondly, in violation of EU law or thirdly, in manifest contradiction with EU policy objectives.Ibid. This mechanism, in other words, operates as a sort of “fire alarm” in the exceptional cases of ultra vires or contra legem exercise of the implementing powers.

However, the consequences of the alert/warning system are principally political, as it does not entail the power to withdraw a decision taken by the Board.This interpretation emerges from the Commission, ‘Progress report on the implementation of the Common Approach on EU decentralised agencies’ COM(2015) 179 final. In other words, the mechanism is intended to give the Commission the power to bring the attention of the Parliament and the Council to certain decisions of the Board which it considers problematic, preventing the adoption of such decisions behind the institutions’ back.See Vos (n. 22) 22. The trigger to introduce this system was the scandal which was caused by a report on anti-Semitism of the Management Board of the Fundamental Rights Agency (at the time, European Monitoring Centre on Racism and Xenophobia). Moreover, although interestingly involving the three institutions in an inter-institutional dialogue,Jacobs (n. 74) 219. the application of the alert/warning system appears to be problematic. The wording of the Common Approach seems to entail an obligation for the Commission to trigger the mechanism, posing the question as to who bears the political responsibility for those acts. Indeed, it is questionable whether this means that, if the Commission does not raise the alarm, it agrees with it and, thus, assumes a sort of “ministerial responsibility” for agencies acts.Vos (n. 22) 33. Clearly, in the absence of systematic information for the legislator on the agencies’ acts, and for its expertise in the policy field, the Commission is the institution best placed to track the agencies’ activities, but this mechanism puts it in an uncomfortable position, the institutional implications of which, especially in relation to the independence of EU agencies, need further consideration.Ibid.

6.2.The Accountability of the ESOs to the Parliament

With regard to the exercise of de facto implementing powers within the context of harmonised standardisation, it is noteworthy that concerns on its legitimacy and accountability were expressed in literature since the 1980s, raising strong doubts about the compatibility of the reference to harmonised technical standards in the New Approach directives with the Meroni doctrine.Herwig Hofmann, Gerard Rowe & Alexander Türk, Administrative Law and Policy of the European Union (Oxford University Press 2011) 248; Schepel (n. 4); E. Steindorff, ‘Quo vadis Europa? Freiheiten, Regulierung und soziale Grundrechte nach den erweiterten Zielen der EG – Verfassung’ in Forschungsinstitut für Wirtschaft (ed.), Weiterentwicklung der Europäischen Gemeinschaften und der Marktwirtschaft (Heymanns 1992) 11. From this perspective, steps towards the direction of enhancing these aspects were undertaken by the EU institutions which, already in the Council Resolution of 18 June 1992 on the role of European standardisation in the European economy,Council Resolution of 18 June 1992 on the role of European standardisation in the European economy, C 173 of 9.7.1992. emphasised the need for accountability of the ESOs in their activities under the New Approach.See also General Guidelines for Co-operation between CEN and CENELEC and the European Commission, adopted in 1984; and Commission, ‘Report of 13 May 1998 on efficiency and accountability in European standardisation under the New Approach’ COM(1998) 0291, 4.

However, it is with the adoption of Regulation 1025/2012 that a systematic accountability mechanism was introduced, thereby enhancing the institutional oversight of the Parliament on the activities of these private-law bodies. In particular, Article 11 allows the European Parliament and the Member States to raise formal objections to harmonised standards when one considers that “a harmonised standard does not entirely satisfy the requirements which it aims to cover and which are set out in the relevant Union harmonisation legislation”.Regulation 1025/2012, art 11. Accordingly, it informs the Commission, which shall examine the objection and may decide either, firstly, to publish, not to publish or to publish with restrictions the reference to the harmonised standard,In accordance with the advisory committee procedure, see Article 11(4). This provision applies to standards already adopted by the ESOs, but whose reference has not yet been published in the Official Journal of the European Union. or, secondly, to maintain, to withdraw or to maintain with restrictions the references to the harmonised standard.In accordance with the examination committee procedure, see Article 11(5).

In other words, similar to the right of scrutiny established in the Comitology Regulation, the Parliament is empowered to react to the publication of a harmonised standard,It may be interesting to note that the system established by Regulation 1025/2012 gives the Parliament a double possibility to contest the publication of the reference: firstly, in application of Article 11 of Regulation 1025/2012 and, secondly, in exercise of its right of scrutiny vis-à-vis the Implementing Decision adopted by the Commission on the maintenance of the reference under the Comitology Regulation. but without being provided with a veto right on the publication of a reference or its maintenance in the Official Journal. Although in line with the powers of the Parliament in other forms of EU implementation, the limited effect and the limited scope of the formal objection may appear particularly insufficient in the light of the deficiencies in the judicial account-ability of harmonised standards.

7.The Judicial Accountability of the Exercise ofImplementing Powers

In the light of the weak democratic control which characterises these forms of exercise of implementing powers, the issue of judicial accountability, i.e. the availability of judicial review of implementing acts, is crucial not only in its role as a tenet of accountability and throughput legitimacy, which could compensate the low performance in terms of input legitimacy,Fritz Scharpf, Governing in Europe: Effective and Democratic? (Oxford University Press 1999); Stijn Smismans, Law, Legitimacy and European Governance. Functional Representation in Social Regulation (Oxford University Press 2004). For a definition of throughput legitimacy, see, inter alia, V.A. Schmidt, ‘Democracy and Legitimacy in the European Union Revisited: Input, Output and “Throughput”’ [2013] Political Studies 2. but also as an essential element of the rule of law in the EU legal system.Case 294/83, Les Verts v European Parliament EU:C:1986:166, 23. In line with the relevance of the right to an effective remedy for the protection of rights and freedoms guaranteed by the law of the Union, the EU has developed a jurisdictional system which ensures application and effectiveness of the system of EU norms as a wholeG. Tesauro, Diritto dell’Unione europea (Cedam 2011) 229-231. On the limited jurisdiction of the Court in CFSP and PJCC matters, see Articles 275 and 276 TFEU. and which is articulated in a number of judicial remedies open not only to the institutions and Member States but also to individuals for the protection of their legal positions.On the peculiarity of the EU judicial system and its implications see, inter alia, B. Cortese, ‘A la recherche d’un parcours d’autoconstitution de l’ordre juridique interindividuel européen: essai d’une lecture pluraliste 50 ans après Van Gend en Loos et Costa’ [2015] Il diritto dell’Unione europea, 227-271. See also B. Cortese, L’ordinamento dell’Unione europea, tra autocostituzione, collaborazione e autonomia (Giappichelli 2018).

7.1.The Possibility of Judicial Review of the Acts of EUAgencies

With regard to the judicial review of implementing acts of the Commission and the Council, the possibility to challenge the validity of these acts is well established both directly through an action for annulment pursuant to Article 263 TFEU and indirectly through the preliminary reference to the Court of Justice pursuant to Article 267 TFEU.See also Articles 265, 268, 270 and 277 TFEU.

Conversely, the possibility to challenge the acts of EU agencies is the result of a tortuous evolution of the case law. Indeed, with the exception of the plea of illegality,See Case T-120/99, Kik v UAMI EU:T:2001:189, para 21. the jurisdiction on the acts of EU agencies was not immediately accepted by the Court, which for a long time adopted a formalistic interpretation of the treaty provisions on judicial review that did not mention them among the reviewable acts.Tovo (n. 3) 343. The turning point was the case Sogelma,Case T-411/06, Sogelma v EAR EU:T:2008:419. For a comment, see E. Bernard, ‘Recours contre les actes des agencies’ [2008] Europe 403, 14-16; E. Piselli, ‘Minimum Selection Criteria and their Application during the Evaluation Process: Sogelma Srl v European Agency for Reconstruction (EAR)’ [2009] Public Procurement Law Review 83; G. Vandersanden, ‘Arrêt “Sogelma”: l’annulation d’actes adoptés par des organes établis sur la base du droit dérivé’ [2008] Journal de droit européen 297. where the Court recognised that the impossibility to challenge these acts constituted an unacceptable “legal vacuum” in the EU system of judicial protection.EAR (n. 107) [40]. Therefore, since the EU is “a community based on the rule of law”,Ibid. Reference here was to Case 294/83, Les Verts v European Parliament, para 24. the Treaty provisions had to be interpreted as allowing the judicial review of the acts of EU agencies through a direct action.

The Lisbon Treaty expressly inserted the acts of EU agencies among the reviewable acts in Articles 263 and 267 TFEU and, thus, contributed to the constitutionalisation of this “specific institutional arrangement”.Edoardo Chiti, ‘European Agencies’ Rulemaking: Powers, Procedures and Assessment’ [2013] European Law Journal 93, 94. Yet, considering that the restrictive position of the Court on the standing of non-privileged applicants makes it particularly difficult for natural and legal persons to challenge these acts,See Case 25/62, Plaumann & Co. v Commission of the European Economic Community EU:C:1963:17, 107. See, inter alia, Bast (n. 57) 898-899; M. Eliantonio,‎ C.W. Backes,‎ C.H. Van Rhee,‎

T. Spronken & A. Berlee (ed.), Standing Up for Your Right in Europe: A Comparative Study on Legal Standing (Locus Standi) Before the EU and Member States’ Courts (Intersentia 2013) 45; P. Nihoul, ‘La recevabilité des recours en annulation introduits par un particulier à l’encontre d'un acte communautaire de portée générale’ [1994] Revue trimestrielle de droit européen, 171-194;

D.F. Waelbroeck & A.M. Verheyden, ‘Les conditions de recevabilité des recours en annulation des particuliers contre les actes normatifs communautaires: à la lumière du droit comparé et de la Convention des droits de l’homme’ [1995] Cahiers de droit européen 3-4, 399-441. On the qualification of agencies’ acts as “regulatory acts of general application”, see Case T-96/10, Rütgers Germany GmbH and Others v European Chemicals Agency (ECHA) EU:T:2013:109, para 58. putting into question the assumption that the Treaties have effectively established “a complete system of legal remedies and procedures”,Case C-583/11 P, Inuit Tapiriit Kanatami and Others v European Parliament and Council of the European Union EU:C:2013:625, para 92. See also Case C-50/00 P, Unión de Pequeños Agricultores v Council EU:C:2002:462, para 40; Case C-131/03 P, R J Reynolds Tobacco Holdings, Inc. and Others v Commission of the European Communities EU:C:2006:541, para 80; Case C-59/11, Association Kokopelli v Graines Baumaux SAS EU:C:2012:447, para 34. the possibility to hold EU institutions and agencies to account before the Court of Justice for their exercise of implementing powers is far from evident.

7.2.The Possibility of Judicial Review of Harmonised Standards

The availability of judicial review is even less evident for harmonised standardisation. In the past, the possibility to challenge harmonised standards before the Court of Justice was entirely excluded.P. Pecho & A. van Waeyenberge, ‘La normalisation technique européenne vue de Luxembourg’ [2010] RMC, 387, 393; J.L. Laffineur, M. Grunchard & C. Leroy, ‘Les possibilités de recours contre une norme technique dans l’Union européenne’ [2009] Revue européenne de droit de la consommation, 827. Following the James Elliott case, it is by now established that harmonised European standards form part of EU law “when that standard was conceived, managed and monitored by the Commission and when it produces binding legal effects following publication of its references in the Official Journal”.Case C-185/17, Mitnitsa Varna v SAKSA OOD EU:C:2018:108, para 39. Therefore, when these conditions are fulfilled,See, however, Case C-185/17, Mitnitsa Varna v SAKSA OOD EU:C:2018:108, where the Court considered that a harmonised standard, although fulfilling all the conditions set forth in James Elliott, was not to be considered part of EU law. See also Case C-630/16, Anstar Oy v Tukes EU:C:2017:971. they can be reviewed by the EU jurisdiction in preliminary questions of interpretationElliott (n. 39) [34]. and, perhaps, validityIn this sense, see Toolbox #18 ‘The Choice of Policy Instrument’, attached to the latest Better Regulation package. A number of doubts however remain, see Annalisa Volpato, ‘The harmo-nized standards before the ECJ: James Elliott Construction’ [2017] Common Market Law Review 591, 601. under Article 267 TFEU. The approach of the Court towards technical standards, however, is arguably inconsistent, especially when considering also the cases concerning standards elaborated by international standardisation organisations.See Case C-587/15, Lietuvos Respublikos transporto priemonių draudikų biuras v Gintaras Dockevičius and Jurgita Dockevičienė ECLI:EU:C:2017:463; Case C-399/12, Germany v Council ECLI:EU:C:2014:2258. For an analysis of the contradictory approach of the Court, see A. Volpato & M. Eliantonio, ‘The Contradictory Approach of the CJEU on the Judicial Review of Standards: A Love-Hate Relationship?’ in Eliantonio & Cauffman (n. 4).

Even less clear is whether these standards can be subject to direct action under Article 263 TFEU. Without considering the issues related to the locus standi of private individuals in such direct action,Mariolina Eliantonio, ‘Judicial Control of the EU Harmonized Standards: Entering a Black Hole?’ [2017] Legal Issues of Economic Integration 395, 399-404. the reviewability of an harmonised standard under Article 263 TFEU remains controversial since the Court refused to consider them as “acts of EU institutions, bodies, offices or agencies of the Union”.On this point, the reasoning of the Court differs from the one of the Advocate General Campos Sánchez-Bordona in James Elliott (n. 40). See, inter alia, Volpato (n. 117) 600-601. To reach the result of reviewing harmonised standards, however, possible alternative routes have been interestingly suggested.See, inter alia, Harm Schepel, ‘The New Approach to the New Approach: The Juridification of Harmonised Standards in EU Law’ [2013] Maastricht Journal of European and Comparative Law 521, 531. Especially after the recent shift in the practice of the Commission which now publishes the reference to harmonised standards in the L series of the Official Journal in the form of an Implementing Decision (instead of in the C series as a Communication),See the publication of Commission Implementing Decision (EU) 2018/2048 of 20 December 2018 on the harmonised standard for websites and mobile applications drafted in support of Directive (EU) 2016/2102 of the European Parliament and of the Council, OJ L 327/84. it seems to be possible to challenge the Commission’s decision to publish the reference to an harmonised standardSuch a possibility was presented as admissible by the Court already in Case T-474/15, Global Garden Products Italy SpA v European Commission ECLI:EU:T:2017:36, para 60. and, then, raise the issue of the legality of the adoption of the harmonised standard before the Court.See, by analogy, the case law on the preparatory measures (especially of EU agencies): Joined Cases T-74/00, T-76/00, T-83/00, T-84/00, T-85/00, T-132/00, T-137/00 and T-141/00, Artegodan GmbH v Commission EU:T:2002:283, para 201. See also Case C-120/97, Upjohn Ltd v The Licensing Authority EU:C:1999:14, paras 197-198; T-326/99, Olivieri v Commission and EMA EU:T:2003:351, para 55; Joined Cases 12/64 and 29/64, Ley v Commission EU:C:1965:28, para 118; Joined Cases T-10-12 and 15/92, Cimenteries and others v Commission EU:T:1992:123, para 31; Case T-123/03, Pfizer v Commission EU:T:2004:167, para 24; Case T-108/92, Calò v Commission EU:T:1994:22, para 13. See also Tovo (n. 44) 1207. Although the Court has not yet ruled in such a case, this possibility arguably paves the way for an indirect scrutiny of the activities of the ESOs.

7.3.The Intensity of the Review

The effectiveness and extent of judicial review depends not only on the conditions to have access to the Court, but also on the intensity of the judicial review exercised by the judges in the contested acts. In this regard, it is important to recall that judicial review generally involves a review on law, fact, and discretion.P. Craig, EU Administrative Law (Oxford University Press 2012), ch 13. While the Court fully substitutes judgements of the parties in relation to questions of law, the intensity of review of fact and discretion is different since it needs to respect the institutional prerogatives of the author in deciding on the merit.

Clearly, the exercise of implementing powers often involves the evaluation of complex economic and technical issues. When the challenged measure involves complex economic or technical appraisals, the Court has consistently recognised a broad discretion to EU institutions,Inter alia, Case 57/72, Westzucker GmbH v Einfuhr- und Vorratsstelle für Zucker EU:C:1973:30; Case C‑335/13, Robin John Feakins v The Scottish Ministers EU:C:2014:2343, paras 56-58. However, on the distinction between “discretion proper” and “technical discretion”, see, inter alia, M. Prek & S. Lefevre, ‘Administrative Discretion, Power of Appraisal and Margin of Appraisal in Judicial Review Proceedings before the General Court’ [2019] Common Market Law Review 339. limiting its review to examining whether the relevant act “is not vitiated by a manifest error or misuse of power or whether the institution in question has not manifestly exceeded the limits of its discretion.”Case C-369/95, Somalfruit and others v Ministero delle Finanze EU:C:1997:562, para 50; Case C-354/95 National Farmer’s Union and Others EU:C:1997:379, para 50.

While, in relation to the acts of the Commission, the approach of the Court has lately evolved into a more intensive review of the exercise of powers, applying the test for “manifest error” rigorously,Case T-13/99, Pfizer v Commission EU:T:2002:209. For a detailed analysis of the case, see Ellen Vos, ‘The European Court of Justice in the Face of Scientific Uncertainty and Complexity’ in B. De Witte, E. Muir & M. Dawson (ed.), Judicial Activism at the European Court of Justice (Cheltenham 2013) 142-166. See also Case C-12/03 P, Commission v Tetra Laval EU:C:2005:87, para 39. the judicial review of acts of EU agencies appears to be not particularly intensive. Indeed, it remains “of limited scope”,Case T-115/15, Deza v ECHA EU:T:2017:329, para 164; Case T-134/13, Polynt and Sitre v ECHA EU:T:2015:254, para 52. irrespective of the fact that the political control is rather weak.Tovo (n. 3) 364. For instance, in Rütgers, the Court recognised that the ECHA “has a broad discretion in a sphere which entails political, economic and social choices on its part, and in which it is called upon to undertake complex assessments”.Ibid [134]. As emerges from recent case law, such broad discretion concerns not only the nature and scope of the measures to be taken into account but also, to some extent, the finding of the basic facts.Case T-115/15, Deza v ECHA EU:T:2017:329, para 164; Case T-134/13, Polynt and Sitre v ECHA EU:T:2015:254, para 52. Accordingly, on these aspects the Court exercises only marginal scrutiny on the legality of the assessment, resulting in the annulment of the act only as far as it is proven that it is manifestly inappropriate.Case T-96/10, Rütgers Germany v ECHA EU:T:2013:109, para 134.

In this regard, it is arguable that the Court would adopt this deferential approach also in relation to the activities of the ESOs, characterised by a highly technical expertise.Tovo (n. 3) 1207. Considering that EU agencies and ESOs are generally delegated powers precisely to carry out technical and scientific work, the result is that most of their activities fall outside the scope of a full review by the Court. Although it is doubtful that the Court constitutes an appropriate forum for scientific assessments, a stricter standard of review, such as the one shown in relation to the Commission’s decisions in policy areas such as risk regulation and competition,See Prek & Lefevre (n. 126) 358, and the case law analysed therein. would appear more in line with the needs of accountability and, ultimately, with the limits of the delegation of powers to these bodies.Short Selling (n. 21) 53. On the point, see Hofmann, Rowe & Türk (n. 93) 244.

8.Conclusion

The analysis of the different forms of implementation of EU law which have emerged in the EU composite institutional reality has unveiled the complexity and the weakness of their accountability, shedding light on some blind spots in the democratic control and judicial review of these phenomena. While the plurality of levels involved are now visibly enshrined in primary law, what emerges from the lines of Article 291 TFEU does not give an exhaustive representation of the actual composite nature of EU implementation. In fact, it leaves in the shadow relevant phenomena which have become an inescapable reality in the practice of EU law. From a functional and material perspective, the qualification as implementing acts cannot be limited to the acts of the Commission and of the Council, but also the activities of EU agencies and harmonised standardisation arguably fall within the definition of an exercise of atypical, de facto implementing powers.

The recognition of this multilevel, composite nature of EU implementation highlights the need for more coherent and effective mechanisms to guarantee effective democratic and judicial accountability in this area of EU law. From the analysis of the oversight mechanisms in place - not only for the “visible” forms of implementation, but also for the “invisible” ones - a number of issues emerged. Firstly, while the rights of oversight of the Parliament bear strong similarities in the different cases, their effect is rather limited, if not absent altogether such as in the case of the implementing acts of the Council. In particular, despite the multiplicity of control mechanisms, the control over EU agencies still presents relevant shortcomings, raising critical doubts regarding the conclusion that the delegation to EU agencies is really “under control”.Echoing the phrase “nobody controls the independent agency, yet the agency is under control” used by G. Majone (ed.), Regulating Europe (Routledge 1996) 39. Secondly, in a system which guarantees the judicial review of implementing acts of the Commission, the Council and EU agencies before the Court, the same possibility for harmonised standards remains problematic, especially in relation to Article 263 TFEU. Moreover, the limited scrutiny of the Court in cases involving complex and technical assessments casts a shadow on the judicial accountability in this area of EU law and calls for further scholarly attention to the interplay between discretion, accountability and implementation of EU legislation.

In particular, in light of the existing hurdles to the judicial review of technical standards and the inconsistent approach of the Court, further evolution of the case law is needed to live up to the expectations of legitimacy and respect for the rule of law in this domain. The systematic rethinking of the position of harmonised standards in relation to EU law and to EU implementation should, thus, lead to an enhanced accountability of the ESOs before the Parliament and, especially, the Court. Bringing the judicial review of this exercise of public authority by private bodies in line with the one available for the other forms of exercise of EU implementing powers would, thus, ensure that the invisibility of this phenomenon in the Treaties does not result in its unaccountability.