In Search for ‘Good’ Evidence and Its Use for Better EU Legislation

• A Context: A Growing Emphasis on Evidence-Based Lawmaking

  The European Union (EU), as a regulatory authority,¹x G. Majone, ‘The Rise of the Regulatory State in Europe’ West European Politics 17 (1994), pp. 77-101. has a long history of evidence-based policymaking, dating back to 2002.²x European Commission, Communication, ‘Action Plan “Simplifying and Improving the Regulatory Environment”’, COM(2002) 278 final, 5 June 2002. With its 2015 Better Regulation Agenda,³x European Commission, Communication to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, ‘Better Regulation for Better Results – An EU Agenda’, COM(2015) 215 final, 19 May 2015. The documents included in the ‘Better Regulation’ package are available at: https://ec.europa.eu/info/law/law-making-process/better-regulation-why-and-how_en#need. the European Commission stressed the importance of evidence-based legislation⁴x ‘Policy’ is a wider term than legislation and encompasses alternative measures, e.g. the EU Guidance or the promotion of self-regulation, as well as measures to strengthen law and policy implementation or enforcement. as a means to ensure that a policy is developed on the basis of a “systematical analysis of the potential economic, social and environmental impacts of proposed policies”. Evidence-based legislation should thereby help policymakers to draft ‘high-quality’ legislation that is effective and efficient. The Commission has reiterated its commitment to ‘evidence-based lawmaking’ in its 2021 Communication on Better Regulation.⁵x European Commission, Communication to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, ‘Better Regulation: Joining Forces to Make Better Laws’ COM (2021) 219, 29 April 2021, p. 4. Over the past few decades, the EU machinery for evidence-based lawmaking has progressively become more sophisticated. This has resulted in new procedures devised to ensure that legislative proposals are based on an assessment of the potential impact of the legislative measure.⁶x For more references see, inter alia, M. Dawson, ‘Better Regulation and the Future of EU Regulatory Law and Politics’ Common Market Law Review 53 (2006), pp. 1209-1235. This emphasis on evidence-based lawmaking requires a thorough reflection on what relevant evidence or good evidence is and how this is used and should be used to achieve the objective of better regulation.
  In both the 2021 Communication and the Better Regulation Guidelines, ‘evidence’ is defined as

  multiple sources of data, information, and knowledge, including quantitative data such as statistics and measurements, qualitative data such as opinions, stakeholder input, conclusions of evaluation, as well as scientific and expert advice.⁷x COM (2021) 219, p. 4, Commission Staff Working Document, ‘Better Regulation Guidelines’ SWD (2021) 305, 3 November 2021, p. 5. In addition, it is worth recalling that the Organisation for Economic Co-operation and Development (OECD), Regulatory Impact Assessment, OECD Best Practice Principles for Regulatory Policy, Paris, 2020, p. 8, defines ‘evidence-based policymaking’ as follows: “Policies and/or regulations should always be based on the best available information … and take into account different solutions to a problem.” For references see A. Alemanno, ‘How Much Better is Better Regulation?’, European Journal of Risk Regulation 3 (2015), pp. 344-356 notes that methods for gathering evidence, monitoring and consultations have not undergone drastic changes with the 2015 Better Regulation Agenda.

  This EU-wide notion of evidence is necessarily a wide definition, as the answer to the question of what data, information and knowledge are available, reliable and convincing may very well differ depending on the policy area. Accordingly, in an organization, like the EU, characterized by a broad competence in a plethora of policy domains, a univocal understanding of evidence would not be possible, also because of the different nature of these policy domains, including even politically sensitive areas, such as the policy fields of the Area of Freedom, Security and Justice (AFSJ). In this connection, although the Commission links evidence-informed lawmaking to high-quality legislation that achieves ‘better’ results, that does not mean that that evidence subsequently substitutes political choices in the EU legislative process.⁸x SWD (2021) 305, p. 10. On the contrary, evidence is to be used to support the adoption of duly informed choices. In other words, discretion is exercised while taking into account relevant evidence and scientific data available during the legislative process.⁹x CJEU 21 June 2018, Poland v. Parliament and Council, C-5/16, EU:C:2018:483, paras. 160-163.
  Even though the quality of legislation, and its – presumably optimized – effects are frequently referred to in the context of evidence-informed lawmaking, the question of what ‘good’ evidence entails is much less present in Commission communications, despite the call for horizontal standards for scientific evidence.¹⁰x G. Listorti et al., ‘The Debate on the EU Better Regulation Agenda: A Literature Review A Contribution to the Stocktaking of the Commission’s Better Regulation Approach’ JRC Science for Policy Report 2019, p. 18. The lack of a clear view on what constitutes ‘good’ evidence may partially be due to the wide range of policy domains and the many forms of evidence used in those domains, as well as, possibly, the goal-oriented nature of evidence-gathering in a political context.
  The literature has extensively addressed issues linked to better regulation at the European and national levels¹¹x Cf e.g. C. Radaelli, ‘Whither Better Regulation for the Lisbon Agenda’, Journal of European Public Policy 14 (2007), pp. 190-207; M. Kaeding, ‘In Search of Better Quality of EU Regulations for Prompt Transposition: The Brussels Perspective’ European Law Journal 14 (2008), pp. 583-603; C. Radaelli & A. Meuwese, ‘Better Regulation in Europe: Between Public Management and Regulatory Reform’, Public Administration 87 (2009), pp. 639-654; C. Dunlop et al., ‘The Many Uses of Regulatory Impact Assessment: A Meta-Analysis of EU and UK Cases’ Regulation and Governance 6 (2012), pp. 23-45; S. Tombs, ‘Making Better Regulation, Making Regulation Better?’ Policy Studies 37 (2016), pp. 332-349. and has raised questions that are still pertinent, due to the ongoing and future-oriented approach to better regulation pursued by the European Commission. This Special Issue aims to provide an interdisciplinary overview on this question, exploring what evidence has been considered by whom and for what purpose in the EU legislative process, in the areas of consumer law, competition law, asylum law, climate law, regional policy, artificial intelligence and multilateral rules on e-commerce.

• B Insights: Generating ‘Good’ Evidence in Different Policy Domains

  The European Commission recognized that generating evidence, especially scientific evidence ‘produced by the research community’, can be a complex and lengthy process.¹²x COM (2021) 219, 29 April 2021, p. 4. In order to streamline this process, the Commission has established the ‘Have your say’ portal, where stakeholders and citizens have the opportunity to provide feedback for a certain amount of time on the Commission’s proposals for EU legislative acts, draft-delegated acts and draft-implementing acts.¹³x The portal is available at: See https://ec.europa.eu/info/law/better-regulation/have-your-say_en. Moreover, gathering evidence is not, in itself, neutral, especially not if political choices must be made.¹⁴x COM (2021) 219, p. 1. It is, for example, apparent from the Better Regulation Toolbox that gathering evidence is a targeted exercise, aimed at informing policy options or responding to evaluation questions, with the help of ‘appropriate’ methods “to collect, interpret, process and transform evidence and data”.¹⁵x Better Regulation Toolbox, Tool #4, available at: https://ec.europa.eu/info/sites/default/files/br_toolbox_-_nov_2021_-_chapter_1.pdf, p. 20. In its most recent annual report, the Regulatory Scrutiny Board – an independent body within the Commission that provides central quality control and support for Commission impact assessments and evaluations at early stages of the legislative process – even noted that impact assessments frequently reasoned towards a politically preferred option.¹⁶x Regulatory Scrutiny Board, Annual Report 2021, p. 18.
  The contributions will analyse the collection and use of evidence in the light of the indications, provided by the non-binding Better Regulation Guidelines and the Toolbox on evidence. The contributions touch upon the question what ‘good evidence’ is. The Commission’s definition of evidence, its recommendations in the Better Regulations and Toolbox, and the consequences for non-compliance with these recommendations leave room for doubt.
  Firstly, evidence should consist of ‘multiple’ sources,¹⁷x See on this question S. Ranchordas, ‘Consultations, Citizen Narratives and Evidence-Based Regulation: The Strange Case of the Consultation on the Collaborative Economy’ European Journal of Law Reform 19 (2017), pp. 1-33. which, if planned well, will enable drafters to triangulate data by comparing, for example, responses to consultations to Eurobarometer studies.¹⁸x Better Regulation Toolbox, Tool #67, available at: https://ec.europa.eu/info/sites/default/files/br_toolbox_-_nov_2021_-_chapter_8.pdf, p. 574. In itself, however, even the use of an impressive number of sources need not equal evidence-informed lawmaking, depending on the use of those sources, their reliability and the existence of other relevant sources. How to choose between the many possible sources of data, and what if the evidence is inconclusive,¹⁹x This may also be reflected in impact assessments, see A. De Feo, ‘Scrutiny of EU Policies’ in A. De Feo and B. Laffan (eds.), Scrutiny of EU Policies. Contributions to the Workshop Organised by the RSCAS, 27 February 2017 (European University Institute, Robert Schuman Centre for Advanced Studies, 2017), pp. 2-13. contradictory or absent? The discretion of the Commission should not be underestimated here. In areas where the EU legislator must analyse complex technical and scientific facts, especially when policy initiatives address evolving, complex technology, this discretion may even include the “finding of basic facts”.²⁰x CJEU (n 9), paras. 150-151.
  Secondly, the Better Regulation Guidelines provide some answers to the questions of how to select between possible data or information, albeit in a scattered manner.²¹x See critically on the different standards for collecting evidence for respectively ex post and ex ante evaluations S. Smismans, ‘Policy Evaluation in the EU: The Challenges of Linking Ex ante and Ex post Appraisal’, European Journal of Risk Regulation 6 (2015), pp. 6-26. They indicate that, for one, impact assessments should consider the results of ex post evaluations.²²x SWD (2021) 305, p. 10. In addition, the positions of stakeholders on various policy options should be made clear in both ex post and ex ante evaluations,²³x Ibid., pp. 29, 34. and, if policy options proposed by stakeholders were rejected. Public consultation is also underscored as essential, especially for impact assessments.²⁴x Ibid., p. 9. Does that mean these sources are of special relevance and should that not mean they are also of at least a minimum standard of quality? The 2021 annual report of the RSB indicates that, regardless of the quality, the subsequent use of consultations may be questionable. The findings of consultations may for example be used “as if they were the result of a representative survey”, rather than a means to gain opinions on future initiatives.²⁵x Annual report 2021, p. 18. In this respect, the selective collection and interpretation of evidence may even serve as an instrument for political lawmaking, which contradicts the notion that evidence-informed lawmaking should contribute to rather than hamper objective lawmaking.
  Thirdly, even if these internal, non-binding Guidelines, as well as the opinions of the Regulatory Scrutiny Board, provide some starting points, not following recommendations in the Guidelines does not necessarily render measures vulnerable to challenges on the basis of, most prominently, principles of proportionality and subsidiarity. For instance, the Court has upheld initiatives even if they have not been accompanied by impact assessments, as the obligation to conduct an impact assessment for all initiatives does not follow from the Interinstitutional Agreement on Better Lawmaking. Thus, no conflict with proportionality arises “where the EU legislature is in a particular situation requiring it to be dispensed with and has sufficient information enabling it to assess the proportionality of an adopted measure”.²⁶x CJEU 3 December 2019 (Czech Republic v. European Parliament and Council of the European Union), C-482/17, ECLI:EU:C:2019:1035, paras. 83-86.
  This raises the question: how should it be determined what ‘good evidence’ entails, in what context and by whom?

• C Prospects: Which Good Evidence can Guide the Future of Better Regulation?

  In an attempt to contribute to the ongoing debate on better regulation, this Special Issue collects multidisciplinary articles from an array of EU policy domains with the ambition to illustrate how to choose ‘good evidence’ to improve the quality of the EU lawmaking. In this connection, despite the difficulties to identify a univocal understanding of the evidence that can be considered equally relevant in all policy areas, some recommendations can be, nonetheless, drawn to at least reflect on how evidence can guide the future of Better Regulation. The rich contributions collected in this Special Issue have provided a critical reflection on the potential and shortcomings of the use of evidence in EU law and policies. Depending on the area of law, various sources are or should be identified as relevant. Similarly, while mentioning the limits of the use of the evidence in specific policy domains, the articles in this Special Issue help provide an answer to how to use good evidence.
  The articles in this Special Issue are categorized into three broad themes. The first theme considers legislative endeavours in EU environmental law, and features articles by Jennings, and Khadim and Van Eijken. We then turn to a second theme on legislation in territorial matters, with articles by Unfried et al. and Nicolosi. The third theme considers legislation in the broad field of market regulation, with contributions by Sluijs, Van Schagen et al., Grozdanovski and De Cooman, and Munu. We close this Special Issue with a synthesizing article by Princen.
  The first article, written by Rhoda Jennings, investigates the value of scientific evidence in environmental legislation as attributed by the CJEU. By analysing case law relating to the Habitats Directive and the Birds Directive, she argues how the CJEU has set a scientific evidentiary requirement for the application of key elements of both directives. In the absence of impact assessments of both Directives, the Court has taken on a clarifying and purposive role in determining an evidentiary standard. As Jennings argues, this will be consequential for the application of the precautionary principle for EU environmental legislation going forward.
  Continuing the theme of environmental legislation, Asmaa Khadim and Hanneke van Eijken critically assess the evidentiary basis of the European Green Deal (EGD). They argue that while the EGD is thoroughly grounded in environmental and economic factors, the purported ‘just transition’ of the EGD for EU citizens is not as strongly supported by evidence. The authors draw attention to the participatory nature of social and human rights impact assessment as important tools in the creation of a ‘just’ transition for citizens. This leads Khadim and Van Eijken to recommend that these tools should be included in all of the key phases of decision-making of the EGD, at the European, national and local levels.
  The third article by Martin Unfried, Pim Mertens, Nina Büttgen and Hildegard Schneider considers EU legislation as it affects cross-border regions in the EU. They draw attention to the many cross-border regions in the EU, and posit the cross-border impact assessment as a special variation of the territorial impact assessment for legislative procedures in the EU. The authors argue that including the cross-border dimension into the standard impact assessment methodology at the national and EU level will enrich the evidence-based approach to the legislative process.
  In the fourth article by Salvatore Nicolosi, the tension is explored between the gathering and processing of evidence and increasing politicization of legislation in EU migration and asylum law. The author analyses the reform of the Common European Asylum Procedure (CEAS) and demonstrates how instrumentalization and simplification of available evidence for political purposes has impeded compromise between Member States to the detriment of a truly migrant-oriented focus of EU migration and asylum law.
  The fifth article by Jasper Sluijs kicks off the section on market regulation. By means of a case study of the 2018-2022 revision of the Vertical Block Exemption Regulation (VBER), the author scrutinizes the tension between the broad evidentiary standard of the EU Better Regulation Agenda and the narrower focus on economic evidence in EU competition law under the Commission’s ‘More Economic Approach’ (MEA). In the VBER revision process, economic evidence was one of many sources of evidence. Moreover, the Commission has been selective and strategic in its use of economic evidence during the evaluation and impact assessment of the VBER.
  The sixth article of Jaap Baaij, Isabelle Rueda and Esther van Schagen bridges legislation in energy markets and EU consumer law. The authors demonstrate how the present revision of the EU Energy Efficiency Directive illustrates that the inclusion of both behavioural and legal design insights can lead to information obligations that are shorter, more accessible and thus more effective for consumers. These behavioural and legal design insights should therefore be considered as ‘relevant’ and ‘best available’ evidence in the sense of the Better Regulation Guidelines and Toolbox.
  Next, the seventh article by Ljupcho Grozdanovski and Jérôme De Cooman critically assesses the process of risk regulation during the legislative drafting of the EU’s Artificial Intelligence (AI) act of 2021. Against the backdrop of the evidentiary requirements of EU Better Regulation Agenda, the authors conclude that the regulatory approach of the AI Act has not been premised on the gathering and processing of evidence. Rather, a pre-existing strategy on AI seems to have formed the foundation on which the Commission designed the regulatory framework, which took shape in the AI Act.
  The eighth article by Martin Munu considers the role of evidence-based legislation in the EU can have an impact on e-commerce negotiations within the World Trade Organization (WTO). The author points out inadequacies in the scattered e-commerce regulation at the WTO, and argues how experiences in evidence-based legislative procedures of mainly the EU Digital Services Tax and the New Deal for Consumers can inform further legislative endeavours for e-commerce at the WTO level.
  The ninth and final article of this Special Issue is by Sebastiaan Princen. His contribution synthesizes the previous seven articles, by reflecting on the nature of ‘evidence’ and its role in the crafting and application of legislation. This way, the article aims to reach a more balanced and nuanced view on the role of evidence in evidence-based legislation and the ways in which legal research can facilitate the fruitful use of evidence going forward. Specifically, the author draws attention to the burden of evidence in legislation, the criteria for determining what counts as ‘good’ evidence and the various purposes for which evidence is used in the legislative processes.

Noten

• 1 G. Majone, ‘The Rise of the Regulatory State in Europe’ West European Politics 17 (1994), pp. 77-101.

• 2 European Commission, Communication, ‘Action Plan “Simplifying and Improving the Regulatory Environment”’, COM(2002) 278 final, 5 June 2002.

• 3 European Commission, Communication to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, ‘Better Regulation for Better Results – An EU Agenda’, COM(2015) 215 final, 19 May 2015. The documents included in the ‘Better Regulation’ package are available at: https://ec.europa.eu/info/law/law-making-process/better-regulation-why-and-how_en#need.

• 4 ‘Policy’ is a wider term than legislation and encompasses alternative measures, e.g. the EU Guidance or the promotion of self-regulation, as well as measures to strengthen law and policy implementation or enforcement.

• 5 European Commission, Communication to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, ‘Better Regulation: Joining Forces to Make Better Laws’ COM (2021) 219, 29 April 2021, p. 4.

• 6 For more references see, inter alia, M. Dawson, ‘Better Regulation and the Future of EU Regulatory Law and Politics’ Common Market Law Review 53 (2006), pp. 1209-1235.

• 7 COM (2021) 219, p. 4, Commission Staff Working Document, ‘Better Regulation Guidelines’ SWD (2021) 305, 3 November 2021, p. 5. In addition, it is worth recalling that the Organisation for Economic Co-operation and Development (OECD), Regulatory Impact Assessment, OECD Best Practice Principles for Regulatory Policy, Paris, 2020, p. 8, defines ‘evidence-based policymaking’ as follows: “Policies and/or regulations should always be based on the best available information … and take into account different solutions to a problem.” For references see A. Alemanno, ‘How Much Better is Better Regulation?’, European Journal of Risk Regulation 3 (2015), pp. 344-356 notes that methods for gathering evidence, monitoring and consultations have not undergone drastic changes with the 2015 Better Regulation Agenda.

• 8 SWD (2021) 305, p. 10.

• 9 CJEU 21 June 2018, Poland v. Parliament and Council, C-5/16, EU:C:2018:483, paras. 160-163.

• 10 G. Listorti et al., ‘The Debate on the EU Better Regulation Agenda: A Literature Review A Contribution to the Stocktaking of the Commission’s Better Regulation Approach’ JRC Science for Policy Report 2019, p. 18.

• 11 Cf e.g. C. Radaelli, ‘Whither Better Regulation for the Lisbon Agenda’, Journal of European Public Policy 14 (2007), pp. 190-207; M. Kaeding, ‘In Search of Better Quality of EU Regulations for Prompt Transposition: The Brussels Perspective’ European Law Journal 14 (2008), pp. 583-603; C. Radaelli & A. Meuwese, ‘Better Regulation in Europe: Between Public Management and Regulatory Reform’, Public Administration 87 (2009), pp. 639-654; C. Dunlop et al., ‘The Many Uses of Regulatory Impact Assessment: A Meta-Analysis of EU and UK Cases’ Regulation and Governance 6 (2012), pp. 23-45; S. Tombs, ‘Making Better Regulation, Making Regulation Better?’ Policy Studies 37 (2016), pp. 332-349.

• 12 COM (2021) 219, 29 April 2021, p. 4.

• 13 The portal is available at: See https://ec.europa.eu/info/law/better-regulation/have-your-say_en.

• 14 COM (2021) 219, p. 1.

• 15 Better Regulation Toolbox, Tool #4, available at: https://ec.europa.eu/info/sites/default/files/br_toolbox_-_nov_2021_-_chapter_1.pdf, p. 20.

• 16 Regulatory Scrutiny Board, Annual Report 2021, p. 18.

• 17 See on this question S. Ranchordas, ‘Consultations, Citizen Narratives and Evidence-Based Regulation: The Strange Case of the Consultation on the Collaborative Economy’ European Journal of Law Reform 19 (2017), pp. 1-33.

• 18 Better Regulation Toolbox, Tool #67, available at: https://ec.europa.eu/info/sites/default/files/br_toolbox_-_nov_2021_-_chapter_8.pdf, p. 574.

• 19 This may also be reflected in impact assessments, see A. De Feo, ‘Scrutiny of EU Policies’ in A. De Feo and B. Laffan (eds.), Scrutiny of EU Policies. Contributions to the Workshop Organised by the RSCAS, 27 February 2017 (European University Institute, Robert Schuman Centre for Advanced Studies, 2017), pp. 2-13.

• 20 CJEU (n 9), paras. 150-151.

• 21 See critically on the different standards for collecting evidence for respectively ex post and ex ante evaluations S. Smismans, ‘Policy Evaluation in the EU: The Challenges of Linking Ex ante and Ex post Appraisal’, European Journal of Risk Regulation 6 (2015), pp. 6-26.

• 22 SWD (2021) 305, p. 10.

• 23 Ibid., pp. 29, 34.

• 24 Ibid., p. 9.

• 25 Annual report 2021, p. 18.

• 26 CJEU 3 December 2019 (Czech Republic v. European Parliament and Council of the European Union), C-482/17, ECLI:EU:C:2019:1035, paras. 83-86.