• A Introduction

  After the May 2015 Better Regulation (BR) package was published by the European Commission, it was met with a flurry of attention from politics,¹xIn the Netherlands both the House of Representatives and the Senate have been asking critical questions. For the debate in The Netherlands, see Communication from the Commission, ‘E150011 – Communication: Better Regulation for Better Results – An EU Agenda’, 2015, available at: <https://www.eerstekamer.nl/eu/edossier/e150011_commissiemededeling_betere>. the policy world²xA. Renda, ‘Too Good to be True? A Quick Assessment of the European Commission’s New Better Regulation Package’, CEPS Special Report No. 108, April 2015, available at <https://www.ceps.eu/system/files/SR108AR_BetterRegulation.pdf>. and academia.³xA. Alemanno, ‘How Much Better is Better Regulation’, European Journal of Risk Regulation, Vol. 6, No. 3, 2015, pp. 344-356. Apart from the ex ante impact assessments that are the focus of this article, the package also proposed to reinforce ‘fitness checks’ in key legislative areas, general policy evaluations and stronger ex post evaluation in order to turn the EU lawmaking process into a ‘regulatory cycle’.⁴xEuropean Commission, ‘Better Regulation for Better Results, An EU Agenda’, COM(2015) 215 final. See also ‘Special Issue on the Better Regulation Package’, European Journal of Risk Regulation, 2015, p. 3. In some respects the package promised a real step forward in turning BR into an issue for the European legislature in its entirety. The draft Inter-Institutional Agreement on Better Regulation (IIA-BR), in particular, led commentators to wonder whether impact assessment (IA) was set to become a more central tool in EU lawmaking.⁵xAlemanno, 2015, pp. 344-356. The draft even went so far as to mention a tripartite ‘Joint Panel’ that would be tasked with assessing the quality of IA carried out on ‘substantial amendments’. The final version of the IIA-BR as it was adopted by all three institutions, however, was heavily watered down on crucial elements such as the aforementioned. The IIA-BR reiterates that European legislative and regulatory acts should be comprehensible and clear; include reporting, monitoring and evaluation requirements; avoid overregulation and unnecessary administrative burdens; and be practical to implement. The commitment by the European Parliament and Council to carry out their own IAs on ‘substantial amendments’ was reinforced somewhat, but without the institutional backup that was part of the draft version.⁶xThe European Parliament, The Council of the European Union & The European Commission. ‘Inter-Institutional Agreement on Better Lawmaking’, 2015, available at: <http://ec.europa.eu/smart-regulation/better_regulation/documents/20151215_iia_on_better_law_making_en.pdf>. However, the sentiment that prompted the European Commission to propose more radical reforms remained. As the Commission noted in its communication, in the period 2007-2014 it published over 700 IAs, while the European Parliament carried out approximately 20 and the European Council none.⁷xEuropean Commission, 2015. And for all the attention adjacent measures such as REFIT have received,⁸x See ‘REFIT Platform’, available at: <https://ec.europa.eu/info/law/law-making-process/overview-law-making-process/evaluating-and-improving-existing-laws/reducing-0_en> (last accessed 3 March 2017). impact assessment remains the key tool in the BR programme because of the breadth of its scope and the way in which it worked itself into the lawmaking routine.
  So, as the issue of how to turn IA into a proper inter-institutional tool is still pending in the background, the question of the appropriate litmus test for the impact of IAs remains relevant. Owing to methodological constraints, the aggregated effects of individual IAs on the economy is not a workable measure of success. Therefore, any test that is geared towards qualitative changes in the process of lawmaking is a better candidate. As Lenaerts proposed, BR instruments should be incentives for legislators to “genuinely investigate alternative instruments and policies rather than be focused narrowly on justifying their chosen policy option”.⁹xK. Lenaerts, ‘The European Court of Justice and Process-Oriented Review’, Yearbook of European Law, Vol. 31, No. 1, 2012, pp. 3-16, at 7. Also, IA was for long, and in the current version of the European Commission’s BR policy still is, the main tool to try and make the ‘Better Regulation’ agenda live up to its five core principles: effectiveness, coherence, participation, openness and accountability.¹⁰xOne of the textual proposals for the TTIP regulatory coherence chapter as published by the European Commission also mentions a “shared commitment to good regulatory principles and practices, such as those laid down in the OECD (Organization for Economic Cooperation and Development) Recommendation of 22 March 2012 on Regulatory Policy and Governance”. See <http://trade.ec.europa.eu/doclib/press/index.cfm?id=1230#regulatory-cooperation>. However, the jury is still out as to what the overarching aim of BR, and of IA specifically, is. Therefore one way to assess the direction in which IA as a tool is developing is to consider the evolution of IA from three distinct perspectives, each of which is present in the policy discourse: a ‘deregulation perspective’, in which curtailing regulatory intervention is the main goal, a ‘technocratic perspective’, which is mainly interested in improving the rationality of regulatory decision-making, and a ‘participatory ‘perspective’, which wants to open up the lawmaking process with the ultimate aim of defying regulatory capture by ‘levelling the playing field’.
  Relating these perspectives to the overarching aims of BR, the first does not really feature any of the values of BR, not even effectiveness, which ultimately is about achieving policy goals, and not about cutting red tape; the second perspective emphasizes effectiveness and coherence, whereas the third champions participation, openness and accountability. One thing that these perspectives do have in common is the centrality of ‘analysis’, whether it be an instrumental kind of analysis, to achieve cutting of red tape, analysis that is as comprehensive as possible or analysis with an added element of procedural justice. One mechanism through which the challenge of ensuring that the heart of IA is ‘analysis’, as opposed to a box ticking exercise, has been taken up is review of IAs as a more routinized or even institutionalized exercise. What that looks like in more concrete terms, and across an array of potential reviewing bodies, is the focus of this article.
  Over the years a consensus has arisen that some form of review of individual IAs is needed for IA to make a difference in the legislative process. Therefore, this contribution asks, as a matter of its central research question, what forms of review of impact assessment exist and what they tell us about the direction in which IA as a tool of European lawmaking is developing. The most explicit initiative taken in this regard concerns the establishment of an ‘Impact Assessment Board’, recently renamed ‘Regulatory Scrutiny Board’. However, to complete the inventory of review mechanisms, this article also looks at external reviewers such as the European Court of Auditors, the European Ombudsman and the European Court of Justice as well. The article begins with a brief introduction of IA and the main concerns regarding its functioning that come out of the literature. It then covers ‘internal’ review and continues with ‘external’ types of review of IA, before briefly presenting potential scenarios for IA’s further development as a tool of EU lawmaking and drawing conclusions regarding the ways in which the various reviewers tackle their tasks.

• B BR and IA: An Overview

  I The Aims of IA

  Going far beyond the context of risk management,¹¹xFor this type of analysis, see, e.g., N. de Sadeleer, ‘The precautionary principle in EC health and environmental law’, European Law Journal, Vol. 12, No. 2, 2006, pp. 139-172. the European Commission introduced ‘full-blown impact assessment’ around 2003 as legitimacy concerns started to dawn on ‘Brussels’.¹²xA.C.M. Meuwese & S. van Voorst, ‘Impact Assessment in Legal Studies’, in C. Dunlop & C.M. Radaelli (Eds.), Handbook on Impact Assessment, Cheltenham, Edward Elgar, 2016, pp. 21-32. IA as a tool is supposed to instil a culture of critical self-reflection and, more specifically, an internal check on common regulatory biases such as confirmation bias in the working routines of policymakers.¹³xC. Dunlop & C.M. Radaelli, ‘Impact Assessment in the European Union: Lessons from a Research Project’, European Journal of Risk Regulation, Vol. 6, No. 1, 2015, pp. 27-34. Today, IA is compulsory for all major regulatory actions (including important delegated and implementing acts, legislation and international treaty negotiations), although it is not a legal requirement as such.¹⁴xMeuwese & Van Voorst, 2016, p. 24. Rather than a strictly enforceable norm, it can be characterized as a heavily institutionalized practice.¹⁵xA.C.M. Meuwese, ‘Inter-Institutionalising EU Impact Assessment’, in S. Weatherill (Ed.), Better Regulation, Oxford, Hart Publishing, 2007, pp. 287-309. The IA template asks for the identification of objectives, alternatives and impacts across three ‘pillars’ (economic, social and environmental) and a comparison based on trade-offs across a fairly flexible set of decision criteria. The aim of imposing a certain economic discipline on lawmakers exists in several varieties that roughly correspond to the ‘perspectives’ laid out in the introduction. ‘Multi-criteria analysis’, rather than prescribing that the policy option with the greatest net benefits to European society always be picked, is best suited to accommodate all perspectives simultaneously. As to the ‘flavour’ of the analysis, in the literature we find assumptions that the information in IAs is, in essence, economic information (e.g. cost-benefit analysis), but also assumptions that they are about ‘scientific advice’ in a wider sense, albeit “considered through a legal, economic and political prism”.¹⁶xP. Hines, ‘Enlightening EU Policy-Making: Evolving Scientific Advice’, EPC Policy Brief, 27 September 2016. The latter view immediately broadens the range of methods and types of evidence, making them “extremely hard to reconcile into balanced, concise advice”.¹⁷x Ibid. For a few short years the European Commission had a Chief Scientific Adviser, who tried to reform the IA process so as to prioritize the use of the scientific method in the assessment of regulatory impacts.¹⁸xW. Douwma, ‘The Role of Science in Better Regulation’, speech at the symposium Better Regulation in the EU Revisited Benefiting Business and Citizens, The Hague, T.M.C. Asser Institute, 23 April 2015. Today, a New Scientific Advice Mechanism (SAM), drawing from the vast array of existing advisory organs, has been put in place.¹⁹xHines, 2016. Its mandate includes advising on future legislation, but until now ‘scientific advice’ appears to exist mainly as a pillar institutionally separate from the IA process. For this reason, this mechanism is not including in the inventory of review bodies below. What could, of course, happen is that this new position for the ‘scientific view’ will lead to renewed calls for independent review of IAs.²⁰x Ibid.

  II Overview of Quality Concerns

  Before we look at what various mechanisms for review of IAs add to the process of lawmaking in the next section, let us list a series of findings regarding what we may loosely label as ‘quality concerns’. A recent study on what actually goes on in IA processes within the European Commission was carried out by Smismans and Minto. Specifically, they looked at the extent to which the Commission in its IAs took into account what they call ‘mainstreaming objectives’.²¹xS. Smismans & R. Minto, ‘Are Integrated Impact Assessments the Way Forward for Mainstreaming in the European Union?’, Regulation & Governance, Vol. 10, 2016: “IAs aim to define the objectives of a particular policy initiative, to set out the different policy options, and to provide an impact assessment of these different options. By definition, such ex ante screening of both objectives and impacts of new policy initiatives requires that there is assessment of the impacts of the initiative beyond the sector in which it originated. Therefore, this systematic, horizontal screening of all main policy initiatives is a potentially useful tool for mainstreaming, as it enables the assessment of whether the policy initiative is likely to have a positive, negative, or negligible impact on the mainstreaming objectives.” The latter term refers to one particular objective of IA, which is to ensure that certain horizontal policy objectives and points of attention are screened for in every policy development initiative. These overarching policy objectives and points are: ‘fundamental rights’, ‘gender’, ‘horizontal social clauses’, ‘discrimination’, ‘environment’ and ‘consumers’. Analysing 35 IAs in the period 2011-2014, Smismans and Minto found that the authors of IA reports currently do not sufficiently take into account these six considerations. In fact, in many of the IA reports analysed several of the objectives and points mentioned were completely absent. ‘Gender’ and ‘non-discrimination’, in particular, were overlooked a lot. Only three out of the 35 IAs referred to all mainstreaming objectives.
  Although there is consensus that Commission IAs have improved significantly in recent years, it is also likely that the following critical findings from earlier studies still apply to some extent. For instance, IAs contain so-called ‘non-falsifiable statements’, seen by Luchetta and Hoepner as a way to avoid accountability towards the Council and the European Parliament.²²xG. Luchetta & S. Hoepner, ‘Praising Their Own Wine? EU Legislators and Non-falsifiable Statements in Impact Assessments’, 2012, p. 12, available at <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2149424>. Another recurring observation relates to selective assessment of benefits. Souto-Otero finds that “[i]n a number of IAs, the social, economic and environmental impacts are only assessed for the preferred option, only listing the negative impacts of the other options”.²³xM. Souto-Otero, ‘Is Better Regulation Possible? Formal and Substantive Quality in the Impact Assessments in Education and Culture of the European Commission’, Evidence & Policy, Vol. 9, 2013, pp. 513-529, at 523. Finally, it has been established in the past that the quality of IAs increases with the expected cost of an initiative.²⁴xC. Cecot et al., ‘An Evaluation of the Quality of Impact Assessment in the European Union with Lessons for the US and the EU’, Regulation and Governance, Vol. 2, No. 4, 2008, pp. 405-424.
  Another aspect that is not captured adequately by in-built quality mechanisms of IA is the quality of the problem definition. Souto-Otero analysed a number of Commission IAs and concluded that problems tended not to be expressed in quantitative terms and that justifications for public intervention, a crucial part of the IA framework, were substandard.²⁵xThe IAs concerned legislative proposals affecting education. Pre-review quality controls of IA could not counter a tendency to reinforce the fallacy “that there is a right way to produce policy through a set of predefined procedures to be universally applied, even though the types of problems identified in the IAs were certainly heterogeneous”.²⁶xSouto-Otero, 2013, p. 521. Finally, the literature reports a “residual lack of transparency in selecting evidence: [the BR Guidelines document] does not force the authors to give the methodology used to acquire and screen evidence”.²⁷xHines, 2016, p. 3. Better Regulation Toolbox, which complements the BR Guidelines, does contain a lighter requirement, namely for an annex to the IA report to “[e]xplain which evidence has been used in the impact assessment together with sources and any issues regarding its robustness (i.e. has the information been quality assured?)”.²⁸xEuropean Commission, ‘Better Regulation Toolbox’, 2015, p. 49, available at: <http://ec.europa.eu/smart-regulation/guidelines/docs/br_toolbox_en.pdf>.

• C Inventory of Regulatory Review Mechanisms

  This section offers an inventory of four bodies with a formal role or a role that fits their institutional position in the EU policy process in reviewing Commission IAs. An important difference between the three non-judicial reviewers (the Regulatory Scrutiny Board, the European Court of Auditors and the European Ombudsman) and the judicial one is the object of their review efforts. The following table illustrates this, while simultaneously capturing where the particular reviewers may be placed on the procedure–substance spectrum:

  Table 1. Overview of reviewers and the focus of their review activities

  Reviewer	Review of the IA report as such	Review of the legislation/regulation (with the help of IA)
  Non-judicial
  Regulatory Scrutiny Board	Procedural (Substantive)	X
  European Court of Auditors	(Substantive)	X
  European Ombudsman	Procedural	X
  Judicial
  Court of Justice	X	Procedural-Substantive

  I Internal: Review by the Regulatory Scrutiny Board

  The initial review forum for Commission IAs consists of the inter-service steering group, in which all the Directorates-General (DGs) who are affected by the regulation are represented and to which the unit carrying out the IA is supposed to report. However, for the sake of comparability, this article takes the first moment that an explicit review takes place (i.e. of which there is a paper trail) as the main internal review mechanism. Perhaps best described as operating at arm’s length of the primary IA process, the Impact Assessment Board was renamed and reconceptualised in the wake of the publication of the IIA-BR. The current Regulatory Scrutiny Board (RSB) includes three members from outside the Commission, selected “on the basis of their proven academic expertise in impact assessment, ex-post evaluation and regulatory policy generally”.²⁹xCommunication from the Commission, 2015, p. 3. All members, so also those recruited from within the Commission, are expected to carry out their assessments independently, work full time exclusively for the board and be selected transparently on the basis of their expertise.³⁰xFor an overview of the current members of the Regulatory Scrutiny Board, available at: <http://ec.europa.eu/smart-regulation/impact/iab/members_en.htm>. Its role is still very similar to that of the former Impact Assessment Board (IAB): scrutinizing draft IA reports and issuing recommendations to the Commission service responsible for the IA report.³¹xA. Alemanno, ‘The Better Regulation Initiative at the Judicial Gate: a Trojan Horse within the Commission’s Walls or the Way Forward?’, European Law Journal, Vol. 15, 2009, pp. 382-400, at 389. As a matter of political commitment on the part of the College of Commissioners,³²xThe Commissioners have committed to not deciding on any legislative proposals that do not carry a positive opinion from the RSB. a positive RSB opinion is needed to go ahead with a proposal, although the RSB does not have a veto right in any strict legal sense.³³xAlemanno, 2009, p. 390. An RSB opinion is considered ‘negative’ when it contains the verdict that “substantial improvements are needed on a number of significant issues”. In that case, a revised IA report needs to be submitted. It happens occasionally that the second opinion is negative again, but only in very rare cases is the third opinion still negative. The RSB also reviews ex post evaluations, a novelty since the new set-up was announced.³⁴xA.C.M. Meuwese & S. Gomtsian, ‘Regulatory Scrutiny of Subsidiarity and Proportionality’, Maastricht Journal of European and Comparative Law, Vol. 22, 2015, pp. 483-505, at 487.
  As the result of an analysis of opinions from the Board from 2010 and 2011, Meuwese and Gomtsian point to the former IAB’s knack for procedural interpretation, at least when it comes to subsidiarity and proportionality review, over economic and legal interpretations.³⁵xMeuwese & Gomtsian, 2015. More than 75% of the Board’s comments on subsidiarity were based on procedural interpretation. The number is even higher for proportionality – about 97%. The institutional changes that took place when the IAB became the RSB, which mean that it is conceived of as a ‘college of experts’ rather than as a ‘body of peers’, imply that more attention for substantive instead of procedural review may be expected.³⁶xA.C.M. Meuwese, ‘Regulatory Scrutiny in Transition’, European Journal of Risk Regulation, Vol. 3, 2015, p. 360. Yet anecdotal evidence suggests that the ‘procedural approach’, also beyond the principles of subsidiarity and proportionality, remains an attractive option for a review body. As, for example, an analysis of the review of the IA on the proposal for a ‘Regulation to ensure the crossborder portability of online content services in the internal market’ shows, the tendency is still to ask for ‘better, more objective reporting’ (on the evidence base for all policy options, not just the preferred one), and ‘better explaining’ (the need to act, in this case).³⁷xRegulatory Scrutiny Board, SEC(2015)484 (draft version of 5 October 2015), available at: <http://ec.europa.eu/smart-regulation/impact/ia_carried_out/docs/ia_2015/sec_2015_0484_en.pdf>. This example is quite representative of the pattern the approaches and methodologies of the IAB/RSB have followed so far.³⁸xMeuwese & Gomtsian 2015; F. Kartner & A.C.M. Meuwese, ‘Responsiveness towards Fundamental Rights Impacts in the Preparation of EU Legislation’, in E. Brems & J. Gerards (Eds.), Procedural Review in European Fundamental Rights Cases, Cambridge, Cambridge University Press (forthcoming). In particular, it is common for the RSB to tell the drafters of an IA report to better explain the need for the preferred option and “offer a real choice for decision makers rather than promote the preferred solution”.³⁹xThis exact phrase comes from SEC(2015) 484.

  II External: Review by the European Court of Auditors

  In the past, IAs occasionally featured in regular reports and opinions by the European Court of Auditors (ECA).⁴⁰xA.C.M. Meuwese, Impact Assessment in EU Lawmaking, The Hague, Kluwer Law International, 2008, p. 176. However, after the ECA’s decision in 2008 to audit the entire Commission’s IA system, this body placed itself more firmly on the BR map.
  Falling into the category of ‘general review’, this particular audit was carried out at a macro level: it posed the rather ‘soft’ question of whether IAs support decision-making in the EU institutions. The ECA examined over 100 Commission IA produced from 2003 to 2008 and carried out around 190 interviews. The ensuing report ‘Impact assessments in the EU institutions: do they support decision-making?’⁴¹xEuropean Court of Auditors, Impact Assessments in the EU Institutions: Do They Support Decision Making?, Special report 3/2010, Luxembourg, Publications Office of the European Union, available at <http://ec.europa.eu/smart-regulation/impact/docs/coa_report_3_2010_en.pdf>. called IA one of the cornerstones of the Commission’s BR policy for the improvement and simplification of new and existing legislation.⁴²x Ibid, p. 6. It has been suggested that the rigour of the audit method in fact stood in the way of the ECA carrying out a meaningful study for which social scientific methodology would have been more appropriate.⁴³xA.C.M. Meuwese, ‘The European Court of Auditors Steps Out Of Its Comfort Zone With An Impact Assessment Audit’, European Journal of Risk Regulation, No. 1, 2011, pp. 104-107. However, the report still generated certain insights that confirmed a more widely shared impression regarding the state of development of EU IA, highlighting several more critical points of feedback: the need for increased transparency regarding the reasons for carrying out an IA, the need for simpler language in the IA reports, the issue of superficial treatment of alternative options, caused by the fact that a decision to go ahead with a proposal is often taken before the IA process is started.⁴⁴xEuropean Court of Auditors, 2010. Currently, the ECA is undertaking an audit on a Better Regulation tool at the other side of the ‘regulatory cycle’,⁴⁵xE. Mastenbroek, S. van Voorst & A.C.M. Meuwese, ‘Closing the Regulatory Cycle? A Meta-Evaluation of Ex-post Legislative Evaluations by the European Commission’, Journal of European Public Policy, 2016. ex post evaluation of legislation.
  Under the heading of concrete reviews of IA by the ECA we do encounter some concrete references to IA in ECA reports. For instance, in 2012, the ECA published an opinion regarding a financial interest protection,⁴⁶xEuropean Court of Auditors, ‘Opinion No 3/2012 on a Proposal for a Regulation of the European Parliament and of the Council on the Hercule III Programme to Promote Activities in the Field of the Protection of the European Union’s Financial Interests (2012/C 201/01)’, 2012, available at: <www.eca.europa.eu/Lists/ECADocuments/OP12_03/OP12_03_EN.PDF>. which states:

  In the Court’s opinion, the ‘Intermediate review of the achievement of the objectives of the Hercule II programme’ (4) and the impact assessment are limited in terms of measuring the achievement of the objectives. They merely report the inputs and outputs of the programme over the period 2007-2010. This is partly due to the lengthy procurement procedures and projects whose impact cannot be immediately measured.

  In its 2014 Activity Report, the ECA states:

  The audit found that while the Commission has increased the quality of its impact assessments over time, it still does not sufficiently analyse the economic impact of preferential trade agreements. The EU loses revenue because of weak Member State customs controls that fail to prevent some imports from wrongly benefiting from preferential tariffs.⁴⁷xEuropean Court of Auditors, Activity Report, 2014, available at: <www.eca.europa.eu/Lists/ECADocuments/AAR_14/AAR_14_EN.pdf>.

  Furthermore, in 2014, the ECA published a Special Report on preferential trade arrangements.⁴⁸xEuropean Court of Auditors, ‘Special Report Are Preferential Trade Arrangements Appropriately Managed?’, 2014, available at <www.eca.europa.eu/Lists/ECADocuments/SR14_02/QJAB14002ENC.pdf>. In the report, the ECA mentions IA several times, explicitly stating the differences with sustainability impact assessments (SIAs). The ECA analysed six different IAs concerning preferential trade agreements (PTAs),⁴⁹xIAs concerning central America, the Andean Community, the Republic of Korea, India, Council Regulation (EC) No 732/2008 of 22 July 2008 applying a scheme of generalised tariff preferences for the period from 1 January 2009 to 31 December 2011 and amending Regulations (EC) No 552/97, (EC) No 1933/2006 and Commission Regulations (EC) No 1100/2006 and (EC) No 964/2007 (OJ L 211, 6.8.2008, p. 1) and Regulation (EU) No 978/2012 of the European Parliament and of the Council of 25 October 2012 applying a scheme of generalised tariff preferences and repealing Council Regulation (EC) No 732/2008 (OJ L 303, 31.10.2012, p. 1). stating that IA should have been carried out in 13 cases.⁵⁰x See European Court of Auditors, 2014, p. 18. Its main conclusions were that the Commission “has not appropriately assessed all the economic effects of PTAs and that the completeness of revenue collection is not ensured. However, the use of the impact assessment tool has increased and there has been progress in the quality of the analysis conducted”.⁵¹x Ibid, p. 33. Finally, there are some references to IA in the Rolling Check-list of 2015, which presents a comprehensive overview of the European Court of Auditors’ (ECA) Special Reports.⁵²x See Special Reports of the European Court of Auditors, ‘A Rolling Check-List of Recent Findings’, available at: <www.europarl.europa.eu/RegData/etudes/STUD/2015/536342/EPRS_STU(2015)536342_EN.pdf>.

  III External: Review by the European Ombudsman

  The European Ombudsman acts mainly on the basis of complaints about maladministration in EU institutions, bodies, offices and agencies. Anyone with an EU address and any organization with an office registered in the EU, regardless of whether they have been individually affected by the maladministration, can file a complaint with the European Ombudsman.⁵³xEuropean Ombudsman, ‘Problems with the EU? Who Can Help You?’, 2015, available at: <www.ombudsman.europa.eu/showResource?resourceId=1457445736660_Whocanhelpyou_2015_EN.pdf&type=pdf&download=true&lang=en>. These wide ‘standing rights’ in combination with the wide scope of the European Ombudsman’s mandate made it likely that, sooner or later, IA would appear on the radar of this ‘generic accountability watchdog’.
  First, the Ombudsman has been active in the field by responding to the Commission’s Public Consultation on Stakeholder Consultation Guidelines of 2014, endorsing the need for providing IAs in the preparation of legislative initiatives by the European Commission. On this occasion the European Ombudsman stated that it would be in the interests of citizens, and of the EU institutions, for the Commission to respond positively to the Berlinguer Report that has been adopted by the Parliament on 14 January 2013.⁵⁴x See K. Giannoulis, ‘Berlinguer’s Report on Administrative Procedure Law Welcomed by the Commission – New Europe’, New Europe, 2013, available at: <https://www.neweurope.eu/article/berlinguer-s-report-administrative-procedure-law-welcomed-commission/>. See also European Parliament, ‘Motion for a European Parliament Resolution with Recommendations to the Commission on a Law of Administrative Procedure of the European Union’, 2012/2024(INI), available at: <www.europarl.europa.eu/sides/getDoc.do?type=REPORT&reference=A7-2012-0369&language=EN>.
  With respect to the good administration of EU procedural law, the following was stated as a general starting point:

  The preparation by the Commission of each major policy initiative and of each proposal for a significant legal act of general application (whether legislative or non-legislative) shall normally include (a) an impact assessment and (b) wide consultation with affected interests.

  (a) The Commission shall organise and conduct impact assessments according to the following principles:

  (…)

  (b) The Commission shall organise and conduct consultations (including consultations that form part of impact assessments) according to the following principles:
  (i) Timeliness. Consultation aims to involve interested parties in the development of a policy at a stage where they can still have an impact on the formulation of the main aims, methods of delivery, performance indicators and, where appropriate, the initial outlines of that policy.
  (ii) Clarity. All communications relating to consultations should be clear and concise, and should include all necessary information to facilitate responses. Consultations which are addressed to the general public shall be published in all the official languages.
  (iii) Transparency. Notice of all consultations with interested parties shall be published in the Official Journal. The notice shall include a concise statement of the issues that will be developed, who will be invited to respond and the mechanisms that will be used. Consultation documents shall be made available to the public, at least on request. Responses to consultations shall be accessible to the public in accordance with Regulation 1049/2001. Where possible, such responses shall be made available on-line.
  (iv) Public participation. Any person may respond to a consultation, whether invited to do so or not.

  Second, in the absence of further elaboration of the principles that should accompany the provision of impact assessments, the Ombudsman stated that it “may consider an own-initiative inquiry in the future on the Commission’s procedures for impact assessment”.⁵⁵xEuropean Commission, ‘Public Consultation on Stakeholder Consultation Guidelines 2014’, (Response of the European Ombudsman). Available at: <www.ombudsman.europa.eu/showResource?resourceId=1433773928901_Commission%20-%20stakeholder%20consultation%20guidelines%20-%20EO%20input-final.pdf&type=pdf&download=true&lang=en>. So far, such an inquiry does not appear to have been initiated.
  Third, the European Ombudsman has ventured into several reviews of actual IA, although this practice is very far from amounting to a systematic review procedure. Probably the most well-known example of ‘Ombudsman-intervention’ in the field of IAs concerns its ‘Decision on the failure of European Commission to conduct a prior human rights impact assessment of the EU–Vietnam free trade agreement’ of February 2016.⁵⁶xEuropean Ombudsman, ‘European Ombudsman’s Decision on the Failure of European Commission to Conduct a Prior Human Rights Impact Assessment of the EU Vietnam Free Trade Agreement’, 2016, available at: <www.ombudsman.europa.eu/nl/activities/speech.faces/nl/64453/html.bookmark> & <www.ombudsman.europa.eu/en/cases/caseopened.faces/en/54682/html.bookmark>. Trade negotiations being exempted from the ‘integrated impact assessment’ system, the case concerned the question whether the European Commission should have carried out a specific ‘human rights impact assessment’ in the context of its negotiations to conclude a free trade agreement with Vietnam. The complainants believed that such an assessment was necessary, whereas the Commission’s position was that it was not necessary since a sustainability IA had already been carried out in 2009 on a proposed EU/ASEAN free trade agreement, which included Vietnam. The Ombudsman’s conclusion was that the Commission’s failure to carry out a specific human rights IA, in relation to Vietnam, constituted maladministration. In March 2015 the Ombudsman provided the Commission with the following recommendation:

  Taking into account the above findings, the Commission should carry out, without further delay, a human rights impact assessment in the matter”.⁵⁷x See ‘Draft Recommendation of the European Ombudsman in the Inquiry into Complaint 1409/2014/JN Against the European Commission’, available at: <www.ombudsman.europa.eu/en/cases/recommendation.faces/en/59398/html.bookmark>. The case ended without a win for the European Ombudsman. The Commission refused to conduct such a human rights impact assessment, arguing that its “non-trade policy instruments” and the human rights clauses in the partnership and cooperation agreement achieved that same purpose. As the Agreement had been concluded in the meantime, the Ombudsman had no choice but to close the case with a critical remark.⁵⁸x See ‘Decision in Case 1409/2014/MHZ on the European Commission’s Failure to Carry Out a Prior Human Rights Impact Assessment of the EU-Vietnam Free Trade Agreement’, available at: <www.ombudsman.europa.eu/en/cases/decision.faces/en/64308/html.bookmark>.

  Prior to the aforementioned decision, the Ombudsman acted on a complaint concerning “an alleged failure by the European Commission to carry out an adequate public consultation in advance of drawing up its Proposal for a Regulation concerning the European single market for electronic communications and a Connected Continent”.⁵⁹xEuropean Ombudsman, Decision in Case 904/2014/OV on the European Commission’s Public Consultation Prior To Its Legislative Proposal For A Regulation Concerning The European Single Market For Electronic Communications, 2015, available at: <www.ombudsman.europa.eu/nl/cases/decision.faces/nl/60965/html.bookmark>. See also ‘Failure to carry Out a Public Consultation Prior to a Commission Legislative Proposal on a European Single Marker for Electronic Communications’, available at: <www.ombudsman.europa.eu/cases/caseopened.faces/en/54560/html.bookmark>. Below is a summary of the points that concern the alleged maladministration with regard to the IA that has been carried out.
  The complaint was submitted by the European Competitive Telecommunications Association (ECTA) and concerns the Commission’s alleged failure to carry out an adequate public consultation and IA before submitting its Proposal for a Regulation of the European Parliament and Council laying down measures concerning the European single market for electronic communications and to achieve a Connected Continent on 11 September 2013, which is best known for gradually phasing out roaming surcharges.⁶⁰xEuropean Commission, ‘Regulation of the European Parliament and of the Council’, COM(2013) final 627, 2013. One of the points alleged by ECTA concerned failure on the part of the Commission services to address the points raised by the Impact Assessment Board, as well as a deliberate attempt to conceal the lack of a public consultation. The complainant also alleged a failure to comply with the Commission’s Impact Assessment Guidelines of 15 January 2009. More specifically, not only did the Commission twice fail to obtain the approval of the IA Board; also, even if, in its third opinion of 6 September 2013 (one of those rare cases, see above), the IAB acknowledged some improvements in the IA, it still highlighted several points that needed to be addressed, none of which in the complainant’s view were addressed in the final IA. The Commission defended itself by stating that the final version of the IA does contain a complete overview of the way in which the IAB’s feedback was taken into account. This defence convinced the Ombudsman, since “[t]he fact that the Commission’s draft Impact Assessment report twice received a negative opinion from the IA Board cannot be considered as maladministration” and since the rule in the Guidelines is that “[t]he final version of the IA report should briefly explain how the Board’s recommendations have led to changes compared to the earlier draft”. The implication of this very procedural approach to IA review is that a heavy burden of proof is put on any complainant who seeks to establish that the Commission failed to fulfil its IA obligations to the point of maladministration.

  IV External: Review by the European Court of Justice (CJEU)

  Review of IA by the European Court of Justice (CJEU) – in any of the forms discussed below – occurs less frequently than one might imagine. A major legal argument against a role by the CJEU in any kind of direct IA review is that an IA report is not a reviewable act for the purposes of bringing an action for annulment. Be that as it may, there are plenty of possibilities to indirectly involve an IA report in review activities of legislative and regulatory acts in front of the Court, so much so that it has spawned a specialized strand of literature.⁶¹xA. Alemanno, ‘Courts and Regulatory Impact Assessment’, in C.A. Dunlop & C.M. Radaelli (Eds.), Handbook of Regulatory Impact Assessment, Cheltenham, Edward Elgar, 2016, pp. 127-141; Alemanno, 2009, pp. 382-401; D. Keyaerts, ‘Ex ante Evaluation of EU Legislation Intertwined with Judicial Review? Comment on Vodafone Ltd v Secretary of State for Business, Enterprise and Regulatory Reform (C58/08)’, European Law Review, Vol. 35, No. 6, 2010, pp. 869-884; A.C.M. Meuwese & P. Popelier, ‘The Legal Implications of Better Regulation: An Introduction’, European Public Law, Vol. 17, No. 3, 2011, pp. 455-466. Of course, the formal status of IA reports as Commission staff working papers may still be an obstacle here: quite often an IA report is not ‘up to date’ because amendments have been introduced by the Council and European Parliament (EP) that the IA does not reflect. The main issue regarding some form of indirect review of IA by the CJEU, however, is whether it should be procedural or substantive in nature.⁶²xE. Brems & J. Gerards (Eds.), Procedural Review in European Fundamental Rights Cases, Cambridge, Cambridge University Press (forthcoming). Should the Court mainly check whether the Commission (and co-legislators) ‘did their homework’ or should it also go into the matter of whether that homework was correct? It is important to note that even in the latter approach, which amounts to a procedural kind of review, significant obligations on the part of the European Commission may result. General principles such as equal treatment and legitimate expectations may further strengthen Court-imposed procedural requirements,⁶³xLenaerts, 2012, p. 3. so as to incentivize the Commission to take its IA obligations seriously.⁶⁴xG. Bermann, ‘Taking Subsidiarity Seriously: Federalism in the European Community and the United States’, Columbia Law Review, Vol. 94, 1994, p. 392; Alemanno, 2009, p. 400.
  The Court’s involvement with IA started out with the case Spain v Council,⁶⁵xCase C-310/04, Kingdom of Spain v. Council of the European Union, [2006] ECR I-07285. in which Advocate-General Sharpston suggested that “the lack of impact study makes choices by Commission and Council appear arbitrary”.⁶⁶xOpinion of Advocate General Sharpston in Case C-310/04, Spain v. Council. In that particular case, however, which at the time ruffled some feathers in the legal services of the Institutions,⁶⁷xMeuwese, 2008, p. 163. the Court in its judgment did not refer to ‘impact assessment’ as such explicitly. In what could be called the ‘next generation’ cases referring to IA, Vodafone ⁶⁸xCase C-58/08 Vodafone Ltd and Others v. Secretary of State for Business, Enterprise and Regulatory Reform, [2010] ECR I-04999. and Luxembourg/Slovak Republic vs EP/Council ⁶⁹xCase C-176/09, Grand Duchy of Luxembourg v. European Parliament and Council of the European Union, [2011] ECR I-03727. the Court went further. In Vodafone, the CJEU explicitly referred to the IA report as part of its proportionality review of the Roaming Regulation, and problems with the IA were a contributing factor to the conclusion that there had been a violation.⁷⁰xCase C-58/08, Vodafone Ltd and Others v. Secretary of State for Business, Enterprise and Regulatory Reform, [2010] ECR I-4999, para. 55. In Luxembourg/Slovak Republic vs EP/Council, too, the Court cautiously referred to the analysis conducted in the IA as one argument among several when reviewing for the appropriateness of EU action. In Volker and Markus Schecke,⁷¹xJoined cases C-92/09 and C-93-09, Volker and Markus Schecke GbR and Hartmut Eifert v. Land Hessen, [2010] ECR I-11063. the Court established that the Council and Commission should have looked at a less intrusive policy option, leading it to declare the relevant provisions invalid.⁷²xLenaerts, 2012, pp. 10-12. In Afton Chemical, the CJEU held that amendments should be based on scientific data.⁷³xC-343/09, Afton Chemical Limited v. Secretary of State for Transport, [2010] ECR I-07027. These cases, however, should still be classified as ‘procedural review’, albeit with clear substantive implications. Finally, it is important to note that they concern review “not of but through IA”.⁷⁴xAlemanno, 2016.
  In this limited body of case law, one could read an evolutionary development towards the view, here represented through the exact wording used in Afton Chemical that “even though […] judicial review is of limited scope, it requires that the Community institutions which have adopted the act in question must be able to show before the Court that in adopting the act they actually exercised their discretion, which presupposes the taking into consideration of all the relevant factors and circumstances of the situation the act was intended to regulate”.⁷⁵xC-343/09, Afton Chemical Limited v. Secretary of State for Transport, [2010] ECR I-07027, para. 122. What is clear is that the CJEU does require a certain degree of evidence-based reasoning in order to uphold legislation.⁷⁶xKartner & Meuwese (forthcoming). Less clear is how much weight the Court is prepared to attach to IA as a specific instrument, with specific procedures, and to what extent the activities of the reviewers mentioned above are part of the judicial considerations in the background.

  V Summary Table

  The aforementioned findings may be summarized in a table as follows:

  Table 2. Overview of IA reviewers, the type of review and their core principles

  Reviewer	Type of review	Core principles
  Non-judicial
  Regulatory Scrutiny Board	From ‘peer’ to ‘expert’?	‘Soundness’? Proportionality Subsidiarity Compliance with IA Guidelines Procedural justification EU action
  European Court of Auditors	Expert (limited so far)	(Cost-)effectiveness (in theory)
  European Ombudsman	Watchdog	Administrative justice (transparency, rights) Compliance with IA Guidelines
  Judicial
  Court of Justice	Legal	Proportionality Procedural justification EU action

• D Scenarios for the Development of IA

  To assess the effects of increased and more diverse review on the nature of IA as a BR tool, we must have an idea of the various scenarios for the future of IA.
  A first scenario is for EU IA, currently mainly a tool for internal use by the European Commission, to develop into a ‘tripartite’ instrument. This expression kept returning in the context of the negotiations on a new Inter-Institutional Agreement on Better Regulation. The concept of IA as a ‘tripartite’ instrument would involve a greater degree of ‘ownership’ of the IA process on the part of the European Parliament and the Council. As the House of Lords had already stated in 2010, “it does seem to be the case that the Council of Ministers and the European Parliament are not making as full use of impact assessments as they might”.⁷⁷xHouse of Lords, European Union Committee, 4th Report of Session 2009-2010, ‘Impact Assessments in the EU: Room for Improvement?’, Report with Evidence, 9 March 2010. Given the watering down of the IIA-BR, an evolution into a tripartite tool is not a likely scenario for the near future, and therefore it will not play a large role in the analysis below. However, in the longer run, the scenario is likely to reappear.
  A second scenario is that IA as a tool will develop more and more as an analytical corollary to the consultation process as it takes place during the Commission phase of the legislative process, helping to ‘structure the discourse’. This scenario champions a pluralist vision on the EU lawmaking process. One of the main elements here is increased transparency of the legislative process overall, catalyzed by IA. Of course, the price to pay for more openness – and for increased legitimacy – is acknowledged in the literature. More transparency means less capability for reaching compromise solutions owing to reputational concerns⁷⁸xA. Alemanno, ‘A Meeting of Minds on Impact Assessment – When Ex Ante Evaluation Meets Ex Post Judicial Control’, European Public Law, Vol. 17, 2011, pp. 485-505, at 486. and may cause lawmakers to take more extreme positions.⁷⁹xJ.P. Cross, ‘Striking a Pose: Transparency and Position Taking in the Council of the European Union’, European Journal of Political Research, Vol. 52, 2013, pp. 291-315, at 293-294. This direction for the IA process as a whole would likely be accompanied by a procedural focus in the various IA review mechanisms, which would likely be set up as ‘peer review’ or assume a ‘watchdog style’ of review.
  A third scenario involves a return to the ‘policy coordination tool’ IA was originally envisaged to be.⁸⁰xMeuwese, 2008. Rather than focusing on participation and transparency as such, the emphasis here is on ‘highlighting trade-offs’ across regulatory policy options in a sophisticated and objective manner. This direction is not necessarily at odds with the idea of IA as a tripartite tool, as both scenarios may call for capacity for updates during the legislative process. The scenario in which IA is mainly a technocratic policy coordination tool would likely be accompanied by a more substantive focus in IA review, which would preferably be carried out by ‘independent experts’.

• E Conclusion

  As the inventory demonstrates, it is still fairly early days yet for institutionalized review of IA. All four bodies presented in the inventory appear to still be ‘finding their feet’ as reviewers of Commission IA. Although the importance of (external) review is widely recognized as important to the quality of the process and/or the content (depending on the perspective one takes), none of the external reviewers takes a comprehensive view on this task.
  Of the four ‘reviewers reviewed’, the RSB is probably the one with the highest expectations to live up to. From research done on this body so far, it appears that in terms of the overarching direction in which this particular form of review is pushing, ‘structuring the discourse’ most adequately captures the primary concern of this reviewer. For a focus on ‘highlighting trade-offs’ and a pure ‘technocratic perspective’, the overall approach is too broad and too procedurally oriented. Although it would go a bit too far to say that the RSB fully embraces a ‘participatory perspective’, it is attempting to contribute to an informational level-playing field. One risk related to this type of review is that its role in the Commission part of the IA process becomes so routinized that Commission services primarily responsible for the IA start relying on the RSB as an additional source of expertise, rather than trying to meet its standards as a reviewer.
  Although the ECA has taken some interesting initiatives reviewing BR procedures in more general and abstract terms, the concrete review of IA undertaken by this body – although great in potential, especially when it comes to ‘expert review’ of substantive aspects such as effectiveness in the future – remains limited in scope.
  The same could be said for the European Ombudsman, who has focused mainly on the specific context of trade negotiations in which regular IA requirements do not apply as such. In the most prominent ‘regular’ IA case brought to its attention to date, the one on roaming surcharges, the Ombudsman assessed not the substance of the IA but merely the procedural aspects of IA, and in a surprisingly formalistic manner. However, there is also a clear potential on the part of the Ombudsman to help shape the normative and procedural framework for IA, as demonstrated by the more general observations on IA that recently were expressed by this institution.
  A survey of the European Court of Justice case law shows an incremental increase of review activities that involve IA. Here there is a slow but clear evolution towards integrating IA into the pre-legislative space for which the onus is on the Commission to prove it actually exercised its discretion. There is some support for the thesis that the case law signals support for a more technocratic perspective on IA, and the corollary scenario of IA as ‘policy coordination tool’, since an actual exercise of discretion requires high-quality insights into regulatory trade-offs.
  Observing a dominant position for the ‘technocratic perspective’ on BR, interestingly, none of the reviewers appear to embrace a ‘deregulation perspective’. The RSB can at times be strict when it comes to demanding that ‘confirmation bias’ (in favour of EU regulatory intervention) is effectively countered in the IA report, but its dominant procedural approach towards achieving that means that the European Commission gets a lot of leeway for such intervention after all. Finally, one may have expected a championing of the ‘participatory perspective’ and the related ‘structuring of the discourse scenario’ on the part of the European Ombudsman, and we may very well still see this develop in the future, but so far the concrete ‘case law’ does not appear to be there.
  The way in which these types of review are developing are interlinked. For instance, in the future, the Court of Justice could resort to a ‘meta-review’ of subsidiarity by scrutinizing whether “the justification for EU action contained in the impact assessment appears merely formal, scant, or exiguous”.⁸¹xP. Craig and G. de Búrca, EU Law: Text, Cases and Materials, 5th edn, Oxford, Oxford University Press, 2011, p. 99. See also P. Craig, ‘Subsidiarity: A Political and Legal Analysis’, Journal of Common Market Studies, Vol. 50, 2012, p. 78. Meta-review here means that the Court is ‘reviewing the review’ by the Commission, including the Regulatory Scrutiny Board (which, after all, is much better described as ‘internal’ than as ‘independent’). This could also involve looking at the soundness of the quality control procedure in general and using a favourable judgment on the overall quality of the process as an argument for relaxing the standard of review. In a similar vein, if review by the RSB falls below a certain standard, the European Ombudsman may become more prepared to take on its watchdog role and take a less formalistic approach to guarding IA procedures, and/or the ECA might be more likely to offer itself as the alternative ‘expert body’ in the pluralist space of IA reviewers increasingly operating in each other’s shadows.

Noten

• * The author would like to thank Edwin Alblas for his research assistance and Thomas van Golen for his feedback on a draft version of this paper.

• 1 In the Netherlands both the House of Representatives and the Senate have been asking critical questions. For the debate in The Netherlands, see Communication from the Commission, ‘E150011 – Communication: Better Regulation for Better Results – An EU Agenda’, 2015, available at: <https://www.eerstekamer.nl/eu/edossier/e150011_commissiemededeling_betere>.

• 2 A. Renda, ‘Too Good to be True? A Quick Assessment of the European Commission’s New Better Regulation Package’, CEPS Special Report No. 108, April 2015, available at <https://www.ceps.eu/system/files/SR108AR_BetterRegulation.pdf>.

• 3 A. Alemanno, ‘How Much Better is Better Regulation’, European Journal of Risk Regulation, Vol. 6, No. 3, 2015, pp. 344-356.

• 4 European Commission, ‘Better Regulation for Better Results, An EU Agenda’, COM(2015) 215 final. See also ‘Special Issue on the Better Regulation Package’, European Journal of Risk Regulation, 2015, p. 3.

• 5 Alemanno, 2015, pp. 344-356.

• 6 The European Parliament, The Council of the European Union & The European Commission. ‘Inter-Institutional Agreement on Better Lawmaking’, 2015, available at: <http://ec.europa.eu/smart-regulation/better_regulation/documents/20151215_iia_on_better_law_making_en.pdf>.

• 7 European Commission, 2015.

• 8 See ‘REFIT Platform’, available at: <https://ec.europa.eu/info/law/law-making-process/overview-law-making-process/evaluating-and-improving-existing-laws/reducing-0_en> (last accessed 3 March 2017).

• 9 K. Lenaerts, ‘The European Court of Justice and Process-Oriented Review’, Yearbook of European Law, Vol. 31, No. 1, 2012, pp. 3-16, at 7.

• 10 One of the textual proposals for the TTIP regulatory coherence chapter as published by the European Commission also mentions a “shared commitment to good regulatory principles and practices, such as those laid down in the OECD (Organization for Economic Cooperation and Development) Recommendation of 22 March 2012 on Regulatory Policy and Governance”. See <http://trade.ec.europa.eu/doclib/press/index.cfm?id=1230#regulatory-cooperation>.

• 11 For this type of analysis, see, e.g., N. de Sadeleer, ‘The precautionary principle in EC health and environmental law’, European Law Journal, Vol. 12, No. 2, 2006, pp. 139-172.

• 12 A.C.M. Meuwese & S. van Voorst, ‘Impact Assessment in Legal Studies’, in C. Dunlop & C.M. Radaelli (Eds.), Handbook on Impact Assessment, Cheltenham, Edward Elgar, 2016, pp. 21-32.

• 13 C. Dunlop & C.M. Radaelli, ‘Impact Assessment in the European Union: Lessons from a Research Project’, European Journal of Risk Regulation, Vol. 6, No. 1, 2015, pp. 27-34.

• 14 Meuwese & Van Voorst, 2016, p. 24.

• 15 A.C.M. Meuwese, ‘Inter-Institutionalising EU Impact Assessment’, in S. Weatherill (Ed.), Better Regulation, Oxford, Hart Publishing, 2007, pp. 287-309.

• 16 P. Hines, ‘Enlightening EU Policy-Making: Evolving Scientific Advice’, EPC Policy Brief, 27 September 2016.

• 17 Ibid.

• 18 W. Douwma, ‘The Role of Science in Better Regulation’, speech at the symposium Better Regulation in the EU Revisited Benefiting Business and Citizens, The Hague, T.M.C. Asser Institute, 23 April 2015.

• 19 Hines, 2016.

• 20 Ibid.

• 21 S. Smismans & R. Minto, ‘Are Integrated Impact Assessments the Way Forward for Mainstreaming in the European Union?’, Regulation & Governance, Vol. 10, 2016: “IAs aim to define the objectives of a particular policy initiative, to set out the different policy options, and to provide an impact assessment of these different options. By definition, such ex ante screening of both objectives and impacts of new policy initiatives requires that there is assessment of the impacts of the initiative beyond the sector in which it originated. Therefore, this systematic, horizontal screening of all main policy initiatives is a potentially useful tool for mainstreaming, as it enables the assessment of whether the policy initiative is likely to have a positive, negative, or negligible impact on the mainstreaming objectives.”

• 22 G. Luchetta & S. Hoepner, ‘Praising Their Own Wine? EU Legislators and Non-falsifiable Statements in Impact Assessments’, 2012, p. 12, available at <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2149424>.

• 23 M. Souto-Otero, ‘Is Better Regulation Possible? Formal and Substantive Quality in the Impact Assessments in Education and Culture of the European Commission’, Evidence & Policy, Vol. 9, 2013, pp. 513-529, at 523.

• 24 C. Cecot et al., ‘An Evaluation of the Quality of Impact Assessment in the European Union with Lessons for the US and the EU’, Regulation and Governance, Vol. 2, No. 4, 2008, pp. 405-424.

• 25 The IAs concerned legislative proposals affecting education.

• 26 Souto-Otero, 2013, p. 521.

• 27 Hines, 2016, p. 3.

• 28 European Commission, ‘Better Regulation Toolbox’, 2015, p. 49, available at: <http://ec.europa.eu/smart-regulation/guidelines/docs/br_toolbox_en.pdf>.

• 29 Communication from the Commission, 2015, p. 3.

• 30 For an overview of the current members of the Regulatory Scrutiny Board, available at: <http://ec.europa.eu/smart-regulation/impact/iab/members_en.htm>.

• 31 A. Alemanno, ‘The Better Regulation Initiative at the Judicial Gate: a Trojan Horse within the Commission’s Walls or the Way Forward?’, European Law Journal, Vol. 15, 2009, pp. 382-400, at 389.

• 32 The Commissioners have committed to not deciding on any legislative proposals that do not carry a positive opinion from the RSB.

• 33 Alemanno, 2009, p. 390.

• 34 A.C.M. Meuwese & S. Gomtsian, ‘Regulatory Scrutiny of Subsidiarity and Proportionality’, Maastricht Journal of European and Comparative Law, Vol. 22, 2015, pp. 483-505, at 487.

• 35 Meuwese & Gomtsian, 2015. More than 75% of the Board’s comments on subsidiarity were based on procedural interpretation. The number is even higher for proportionality – about 97%.

• 36 A.C.M. Meuwese, ‘Regulatory Scrutiny in Transition’, European Journal of Risk Regulation, Vol. 3, 2015, p. 360.

• 37 Regulatory Scrutiny Board, SEC(2015)484 (draft version of 5 October 2015), available at: <http://ec.europa.eu/smart-regulation/impact/ia_carried_out/docs/ia_2015/sec_2015_0484_en.pdf>.

• 38 Meuwese & Gomtsian 2015; F. Kartner & A.C.M. Meuwese, ‘Responsiveness towards Fundamental Rights Impacts in the Preparation of EU Legislation’, in E. Brems & J. Gerards (Eds.), Procedural Review in European Fundamental Rights Cases, Cambridge, Cambridge University Press (forthcoming).

• 39 This exact phrase comes from SEC(2015) 484.

• 40 A.C.M. Meuwese, Impact Assessment in EU Lawmaking, The Hague, Kluwer Law International, 2008, p. 176.

• 41 European Court of Auditors, Impact Assessments in the EU Institutions: Do They Support Decision Making?, Special report 3/2010, Luxembourg, Publications Office of the European Union, available at <http://ec.europa.eu/smart-regulation/impact/docs/coa_report_3_2010_en.pdf>.

• 42 Ibid, p. 6.

• 43 A.C.M. Meuwese, ‘The European Court of Auditors Steps Out Of Its Comfort Zone With An Impact Assessment Audit’, European Journal of Risk Regulation, No. 1, 2011, pp. 104-107.

• 44 European Court of Auditors, 2010.

• 45 E. Mastenbroek, S. van Voorst & A.C.M. Meuwese, ‘Closing the Regulatory Cycle? A Meta-Evaluation of Ex-post Legislative Evaluations by the European Commission’, Journal of European Public Policy, 2016.

• 46 European Court of Auditors, ‘Opinion No 3/2012 on a Proposal for a Regulation of the European Parliament and of the Council on the Hercule III Programme to Promote Activities in the Field of the Protection of the European Union’s Financial Interests (2012/C 201/01)’, 2012, available at: <www.eca.europa.eu/Lists/ECADocuments/OP12_03/OP12_03_EN.PDF>.

• 47 European Court of Auditors, Activity Report, 2014, available at: <www.eca.europa.eu/Lists/ECADocuments/AAR_14/AAR_14_EN.pdf>.

• 48 European Court of Auditors, ‘Special Report Are Preferential Trade Arrangements Appropriately Managed?’, 2014, available at <www.eca.europa.eu/Lists/ECADocuments/SR14_02/QJAB14002ENC.pdf>.

• 49 IAs concerning central America, the Andean Community, the Republic of Korea, India, Council Regulation (EC) No 732/2008 of 22 July 2008 applying a scheme of generalised tariff preferences for the period from 1 January 2009 to 31 December 2011 and amending Regulations (EC) No 552/97, (EC) No 1933/2006 and Commission Regulations (EC) No 1100/2006 and (EC) No 964/2007 (OJ L 211, 6.8.2008, p. 1) and Regulation (EU) No 978/2012 of the European Parliament and of the Council of 25 October 2012 applying a scheme of generalised tariff preferences and repealing Council Regulation (EC) No 732/2008 (OJ L 303, 31.10.2012, p. 1).

• 50 See European Court of Auditors, 2014, p. 18.

• 51 Ibid, p. 33.

• 52 See Special Reports of the European Court of Auditors, ‘A Rolling Check-List of Recent Findings’, available at: <www.europarl.europa.eu/RegData/etudes/STUD/2015/536342/EPRS_STU(2015)536342_EN.pdf>.

• 53 European Ombudsman, ‘Problems with the EU? Who Can Help You?’, 2015, available at: <www.ombudsman.europa.eu/showResource?resourceId=1457445736660_Whocanhelpyou_2015_EN.pdf&type=pdf&download=true&lang=en>.

• 54 See K. Giannoulis, ‘Berlinguer’s Report on Administrative Procedure Law Welcomed by the Commission – New Europe’, New Europe, 2013, available at: <https://www.neweurope.eu/article/berlinguer-s-report-administrative-procedure-law-welcomed-commission/>. See also European Parliament, ‘Motion for a European Parliament Resolution with Recommendations to the Commission on a Law of Administrative Procedure of the European Union’, 2012/2024(INI), available at: <www.europarl.europa.eu/sides/getDoc.do?type=REPORT&reference=A7-2012-0369&language=EN>.

• 55 European Commission, ‘Public Consultation on Stakeholder Consultation Guidelines 2014’, (Response of the European Ombudsman). Available at: <www.ombudsman.europa.eu/showResource?resourceId=1433773928901_Commission%20-%20stakeholder%20consultation%20guidelines%20-%20EO%20input-final.pdf&type=pdf&download=true&lang=en>.

• 56 European Ombudsman, ‘European Ombudsman’s Decision on the Failure of European Commission to Conduct a Prior Human Rights Impact Assessment of the EU Vietnam Free Trade Agreement’, 2016, available at: <www.ombudsman.europa.eu/nl/activities/speech.faces/nl/64453/html.bookmark> & <www.ombudsman.europa.eu/en/cases/caseopened.faces/en/54682/html.bookmark>.

• 57 See ‘Draft Recommendation of the European Ombudsman in the Inquiry into Complaint 1409/2014/JN Against the European Commission’, available at: <www.ombudsman.europa.eu/en/cases/recommendation.faces/en/59398/html.bookmark>.

• 58 See ‘Decision in Case 1409/2014/MHZ on the European Commission’s Failure to Carry Out a Prior Human Rights Impact Assessment of the EU-Vietnam Free Trade Agreement’, available at: <www.ombudsman.europa.eu/en/cases/decision.faces/en/64308/html.bookmark>.

• 59 European Ombudsman, Decision in Case 904/2014/OV on the European Commission’s Public Consultation Prior To Its Legislative Proposal For A Regulation Concerning The European Single Market For Electronic Communications, 2015, available at: <www.ombudsman.europa.eu/nl/cases/decision.faces/nl/60965/html.bookmark>. See also ‘Failure to carry Out a Public Consultation Prior to a Commission Legislative Proposal on a European Single Marker for Electronic Communications’, available at: <www.ombudsman.europa.eu/cases/caseopened.faces/en/54560/html.bookmark>.

• 60 European Commission, ‘Regulation of the European Parliament and of the Council’, COM(2013) final 627, 2013.

• 61 A. Alemanno, ‘Courts and Regulatory Impact Assessment’, in C.A. Dunlop & C.M. Radaelli (Eds.), Handbook of Regulatory Impact Assessment, Cheltenham, Edward Elgar, 2016, pp. 127-141; Alemanno, 2009, pp. 382-401; D. Keyaerts, ‘Ex ante Evaluation of EU Legislation Intertwined with Judicial Review? Comment on Vodafone Ltd v Secretary of State for Business, Enterprise and Regulatory Reform (C58/08)’, European Law Review, Vol. 35, No. 6, 2010, pp. 869-884; A.C.M. Meuwese & P. Popelier, ‘The Legal Implications of Better Regulation: An Introduction’, European Public Law, Vol. 17, No. 3, 2011, pp. 455-466.

• 62 E. Brems & J. Gerards (Eds.), Procedural Review in European Fundamental Rights Cases, Cambridge, Cambridge University Press (forthcoming).

• 63 Lenaerts, 2012, p. 3.

• 64 G. Bermann, ‘Taking Subsidiarity Seriously: Federalism in the European Community and the United States’, Columbia Law Review, Vol. 94, 1994, p. 392; Alemanno, 2009, p. 400.

• 65 Case C-310/04, Kingdom of Spain v. Council of the European Union, [2006] ECR I-07285.

• 66 Opinion of Advocate General Sharpston in Case C-310/04, Spain v. Council.

• 67 Meuwese, 2008, p. 163.

• 68 Case C-58/08 Vodafone Ltd and Others v. Secretary of State for Business, Enterprise and Regulatory Reform, [2010] ECR I-04999.

• 69 Case C-176/09, Grand Duchy of Luxembourg v. European Parliament and Council of the European Union, [2011] ECR I-03727.

• 70 Case C-58/08, Vodafone Ltd and Others v. Secretary of State for Business, Enterprise and Regulatory Reform, [2010] ECR I-4999, para. 55.

• 71 Joined cases C-92/09 and C-93-09, Volker and Markus Schecke GbR and Hartmut Eifert v. Land Hessen, [2010] ECR I-11063.

• 72 Lenaerts, 2012, pp. 10-12.

• 73 C-343/09, Afton Chemical Limited v. Secretary of State for Transport, [2010] ECR I-07027.

• 74 Alemanno, 2016.

• 75 C-343/09, Afton Chemical Limited v. Secretary of State for Transport, [2010] ECR I-07027, para. 122.

• 76 Kartner & Meuwese (forthcoming).

• 77 House of Lords, European Union Committee, 4th Report of Session 2009-2010, ‘Impact Assessments in the EU: Room for Improvement?’, Report with Evidence, 9 March 2010.

• 78 A. Alemanno, ‘A Meeting of Minds on Impact Assessment – When Ex Ante Evaluation Meets Ex Post Judicial Control’, European Public Law, Vol. 17, 2011, pp. 485-505, at 486.

• 79 J.P. Cross, ‘Striking a Pose: Transparency and Position Taking in the Council of the European Union’, European Journal of Political Research, Vol. 52, 2013, pp. 291-315, at 293-294.

• 80 Meuwese, 2008.

• 81 P. Craig and G. de Búrca, EU Law: Text, Cases and Materials, 5th edn, Oxford, Oxford University Press, 2011, p. 99. See also P. Craig, ‘Subsidiarity: A Political and Legal Analysis’, Journal of Common Market Studies, Vol. 50, 2012, p. 78.