The early stages of the rule-making process – when officials first identify policy challenges and explore potential solutions – are fundamental in shaping the rules that ultimately affect citizens and businesses. Making use of sound evidence to compare and assess a variety of options and consulting with stakeholders and citizens is therefore essential to keep rules targeted and proportionate and avoid biased decision making. This chapter investigates how systematically and through which channels EU Member States use stakeholder engagement before developing a draft regulation to help define policy problems and identify potential solutions. It also considers how they use regulatory impact assessment from the outset to inform decisions on whether and how to regulate, including requirements to assess different options. Finally, it also considers practices to ensure proportionality of process in the use of evidence.
Better Regulation Practices across the European Union 2025
Abstract
Using sound evidence and taking into account a variety of views is key to accurately defining policy challenges and identifying solutions that are both effective and targeted. The considerations and choices at the outset of the policy-making process set the foundation and direction for the rules that will ultimately affect citizens and businesses. Basing them on a thorough understanding of the problem at hand and people’s priorities and lived realities leads to more effective, responsive and targeted rules. Moreover, policy approaches based on comprehensive evidence and a variety of options help avoid biases and narrow, ideologically-driven solutions. At the same time, efforts must be proportionate, as excessive procedural demands can delay beneficial changes and divert resources away from more critical policy areas.
Engagement and impact assessment often come too late – sometimes being used to justify, rather than inform, decisions that have already been made. Stakeholder engagement to help define problems and identify solutions significantly trails comparable procedures on draft rules. Analytical practices, such as options assessment, are relatively more advanced than engagement in these early stages of the rule-making process but there remain important gaps for EUMS to ensure that policy choices are well-founded.
Early engagement with citizens and other stakeholders remains a significant gap among EUMS and underutilised as a tool to inform policy choices. Almost all EUMS report undertaking some form of early engagement to define problems and identify solutions before developing a draft proposal, but only two of them do so systematically. With minimal improvements since 2021, this is an important area of focus for EUMS. When it is undertaken, policymakers tend to prioritise targeted engagement, especially with advisory groups with relevant expertise, or selected groups of stakeholders, foregoing the input of the wider public. However, on issues from climate to taxation, innovative deliberative models like citizen panels have emerged as encouraging practices to bring people meaningfully closer to policymaking from the outset.
Without sufficiently considering a variety of policy options, particularly non-regulatory ones, EUMS risk jumping to conclusions leading to presumptive or ineffective rules. Despite some improvements over the years, alternative options are still not consistently considered across all EUMS. Non‑regulatory approaches remain the least likely options to be assessed at all, and only five EUMS require identifying multiple non‑regulatory options. When it is required, options assessment is not sufficiently scrutinised, as a minority of EUMS indicate that oversight bodies can return impact assessments for revision if alternatives are not appropriately considered. With rising regulatory burdens across the EU and impact on competitiveness, robust options assessment can help policymakers minimise unnecessary rules without compromising fundamental policy objectives or protections.
EUMS have taken steps to ensure proportionality in how evidence is analysed in rulemaking but still have room to improve and maximise the use of limited government resources. Two-thirds of EUMS have a general requirement for proportionality in impact assessment. However, the use of objective criteria via threshold tests to guide proportionate analysis is comparatively less developed, with only one-third of EUMS using either a qualitative or quantitative trigger to determine whether a full or simplified regulatory impact assessment (RIA) should be undertaken. As a result, proportionality might be applied inconsistently and risking for impact and/or options assessment turning into a “check-box” exercise.
The decisions that influence the effectiveness of rules are made long before a legal draft is prepared. Before even deciding whether to regulate or not, policymakers need to be clear what challenges they are seeking to address through government intervention. Diligent agenda-setting lays the foundation for policies based on an accurate understanding of the problem, and a thorough instrument choice exercise helps to identify the most appropriate solution for the given circumstances. Conversely, taking shortcuts at this early stage and regulating “on a hunch” can have serious consequences – like jumping to ineffective and ideologically-driven solutions. However, overlooking critical impacts early, or missing public concerns cannot be effectively resolved later on in the process once expectations are set.
From the start of the policy-making process, people play a crucial role in informing effective and well-rounded policies. Listening to stakeholders early can help policymakers focus on the most pressing issues and gain a comprehensive understanding, including potential impacts, challenges, and opportunities. By engaging diverse perspectives early on, policymakers can uncover innovative ideas, identify potential unintended consequences, and build a more nuanced understanding of the problem they are trying to solve. Resulting policy choices based on the input shared may then be better aligned with public priorities and expectations. People, in turn, are more likely to trust a decision that is demonstrably based on their inputs than one made unilaterally by government behind closed doors (OECD, 2020[1]). Moreover, when citizens, businesses, and other affected parties feel their voices are heard from the outset, they may be more likely to support and comply with the resulting regulations (OECD, 2021[2]), ultimately enhancing the effectiveness of government action. Building on the discussion in this chapter, Chapter 3 further explores the role and importance of transparency and engagement later in the policymaking process when rules are being designed.
Informed by both diverse input and diligent analysis, building and then basing decisions on comprehensive evidence early in the policy development process is crucial for avoiding biased or narrow policy design. Confirmation bias occurs when policymakers favour information or options that confirm their pre-existing beliefs (Drummond, Shephard and Trnka, 2021[3]). For instance, an assumption that immigration targets are needed to protect local jobs may lead policymakers to overlook structural changes in the economy and demand for new skills that would require a more nuanced policy response. Relying on preconceptions can lead to rules that create unnecessary costs on businesses and citizens and that are fixated on (perceived) symptoms but do not address the root causes of the issue. Making use of evidence to consider and evaluate a range of options from the start, rather than using it to justify decisions that have already been made, helps to foster effective regulation. It can also have downstream effects on public trust, as people who believe that their government uses the best available evidence in decision-making are more likely to trust their government (OECD, 2024[4]).
At the same time, there is such a thing as too much preparation – and that in itself can have real consequences. The OECD Recommendation on Regulatory Policy and Governance emphasises that impact assessment should be proportionate to the anticipated impacts of a policy proposal (OECD, 2012[5]). Otherwise, overly extensive analytical requirements can hold up minor but beneficial changes. Instead of swiftly addressing simple regulatory updates, policymakers get bogged down in paperwork and analysis – which can undermine the buy-in for impact assessment as the crucial foundation for well‑informed policy rather than a procedural hurdle. Resulting delays can also increase burden and compliance costs for people and businesses that are forced to wait for regulatory changes that would reduce red tape or improve clarity. Diverting resources from more significant policy issues that require more thorough analysis can ultimately reduce the efficiency of governance.
This chapter delves into the specific requirements and practices that EU Member States (EUMS) employ when deciding whether and how to regulate. It begins by exploring how EUMS engage stakeholders at an early stage to help define policy problems and identify potential solutions. The chapter also examines how EUMS assess different policy options, emphasising the importance of considering a range of approaches before settling on a course of action. Finally, the chapter concludes by reviewing how EUMS set up targeted and proportionate processes for designing and assessing regulations.
The way that policymakers understand and define a given problem sets the course for the rest of the policy cycle, including how eventual regulations are designed and the outcomes they carry. A proper understanding of the situation should inform resulting policy objectives – not vice versa. Clear problem definition helps to ensure that the root causes of the issue are identified, allowing for targeted interventions that efficiently solve the problem. If a city defines its traffic congestion problem simply as “too many cars”, their response may be to restrict vehicle use or implement costly infrastructure changes. However, the real problem may lie in inefficient public transportation routes or poor urban planning. Resources may be wasted on interventions that are ineffective, costly, or meet public resistance. In this case, engaging with the public may have helped to identify the real challenge. Local businesses and commuters may have highlighted bottlenecks in specific bus routes or roads, which policymakers could have used to target changes.
Early-stage engagement with stakeholders is crucial to inform how problems are defined and how solutions are designed before rules are drafted. Compared to later stages of regulatory design, when engagement serves to refine how rules are drafted and inform how they are delivered (discussed in Chapter 3), early engagement allows stakeholders to be involved in a more foundational way when there is still substantial leeway to alter the direction of the policy. Early involvement fosters a sense of public ownership and commitment towards any solutions developed (OECD, 2021[2]). People also bring first-hand experiences, often with unique insights and practical knowledge, that can lead to a more accurate understanding of the problem (OECD, 2023[6]). Therefore, recognising its importance, this section explores how EUMS harness engagement to inform rulemaking early on.
Frequency and forms of early engagement
Fundamentally, early-stage stakeholder engagement remains a gap among EUMS – as it was in 2017 and in 2020. Almost all EUMS report conducting some form of engagement before drafting regulations to better understand policy problems and potential solutions (24 for primary laws, 22 for subordinate regulations). However, when asked how often this engagement is done, only 2 EUMS (Belgium and Italy) have reported doing so systematically (for all or major regulations) since 2017. Contrasting the lack of systematic early engagement, Chapter 3 explores how engagement becomes more systematic in the next stage of the policymaking process: regulatory design; by comparison, almost all EUMS engage systematically with stakeholders at a later stage, once rules have been drafted. This shift highlights the need for a more balanced approach, ensuring meaningful input from the start to create well-informed and effective regulations.
EUMS can also take steps to enhance the depth of early consultations that they do undertake by providing respondents with comprehensive supporting information that equips them to engage meaningfully. Twenty two EUMS provide at least some type of supporting document for their consultations, but almost none do so systematically, with minimal changes in their practices since 2017. It is worth noting that, compared to later-stage engagement, RIAs are not commonly provided at this stage; this could indicate that such documents in the form of an early-stage RIA comparing different options and setting out a rationale for intervention are not yet developed at this stage, suggesting that RIA takes place as a formal exercise downstream in the process. Consultation documents and green papers, however, can allow EUMS to share relevant research and analysis that is available at this stage, including background on the issue and why regulation is being considered. Consultees, in turn, may be more likely to participate if they are provided with detailed, complete information, rather than general descriptions (OECD, 2022[7]).
EUMS employ a range of methods, including public consultations, advisory committees, and targeted outreach, to gather input on potential regulations (general trends for primary laws, as shown in Figure 2.1, are consistent with those for subordinate regulations). However, there appears to be little consistency in how specific modes of engagement are being used, with most of them used ‘sometimes’ – rather than ‘frequently’ or ‘always’ – confirming a lack of established procedures at the early stage. Overall, EUMS show a preference for early targeted engagement, directly engaging with specific group of stakeholders relevant to the policy issue at hand. These could be selected experts, representative organisations or businesses in specific industries. However, it is important that EUMS not overlook broad and open engagement with the general public. Whilst public consultations will attract in particular those who feel affected by a particular issue or topic, their input can help shed light on an issue from a variety of angles and discuss the impacts of different potential solutions early on (OECD, 2020[8]).
Source: OECD Indicators of Regulatory Policy and Governance (iREG).
Figure 2.1 shows that digital platforms are a pillar for broad engagement. For instance, EUMS increasingly use virtual public meetings to support active participation and real-time feedback. This practice grew sharply between 2017 and 2020, which aligns with the Covid-19 pandemic driving governments to increasingly adopt virtual consultations as part of non-pharmaceutical interventions and social distancing. As virtual meeting technology continues to advance, governments have more dynamic tools than ever at their disposal to actively engage people in a more accessible and cost-effective manner. Moreover, 15 EUMS use broad online consultations at an early stage. For the other EUMS, adopting this practice may be low-hanging fruit to improve the reach and diversity of their early-stage engagement considering that most EUMS report already having the necessary infrastructure (i.e. a central government website for consultation). The European Commission, for instance, uses a single ‘Call for Evidence’ system on their online consultation portal to solicit input throughout the policy cycle, from the development of laws to their evaluation (Box 2.1). Recently, additional consultation mechanisms are being introduced: implementation dialogues at political level and reality checks at more technical levels to better understand issues encountered by stakeholders in implementing EU laws.
The Call for Evidence (CfE) system is the European Commission's primary means of explaining to the public and stakeholders the rationale behind a particular initiative, evaluation, or fitness check, outlining its objectives, and collecting feedback. Commission services are required to launch a CfE for new legislative proposals, evaluations, and fitness checks.
For new initiatives, the CfE bundles two previously separate consultation steps on the same initiative into a single engagement exercise. Stakeholders are asked to provide feedback on a CfE document (previously known as Roadmap/Inception Impact Assessment), which sets out the Commission’s understanding of the problem and possible solutions, including the objectives to be achieved, the need for EU action, its added value, and alternative policy options. Depending on the significance of the initiative, the document may also specify expected impacts for each policy option and further detail, such as on subsidiarity issues and impacts on Sustainable Development Goals. In addition, stakeholders are also asked to participate in a questionnaire-based consultation to share their views and any relevant information they may have. It is still possible however for Commission services to launch the public consultation at a later stage and only seek feedback on the CfE document first.
The CfEs are conducted via the Public Consultations and Feedback (“Have Your Say”) portal and are available in all official EU languages. When CfEs are accompanied by a questionnaire, stakeholders are given 12 weeks to respond to the consultations. A summary of the feedback received and how it has been addressed is published and complements the final IA (or evaluation) report.
The CfE is also used for evaluations of existing policy and fitness checks. Where this feeds back into changes to the existing regulations, the CfE for the evaluation and the impact assessment can be run as a single process. In this case, a public consultation with backward and forward-looking questions must be associated with this call unless a derogation is granted.
Targeted consultation with select stakeholders remains the most popular form of early engagement – even being undertaken by more EUMS at this stage than later in rulemaking. The different ways in which EUMS go about this are discussed further below.
Diverse methods for targeted engagement
Consulting with selected groups in a focused manner can be appropriate for situations where input from a specific group or a limited number of specific experts is needed. In fact, EUMS most commonly undertake formal targeted consultation with advisory groups that have particular expertise in the relevant policy area. In Germany, for example, the Federal Ministry of Justice and Consumer Protection established a working group on parentage law to assess the need for legislative reform. This group comprised experts from diverse fields, including family law, constitutional law, ethics, psychology, and medicine. The experts' recommendations informed the drafting of significant proposed parentage law reforms.
France’s Conseil National de la Refondation (CNR) exemplifies large-scale targeted consultation supported by broader input. The CNR is a national forum that brings together identified local, government, business, and social representatives for discussions on key challenges such as education, healthcare, digital transition, and climate change; the work of these representatives is informed by over 100 000 online contributions from the general public. At the national level, stakeholders have developed action plans like a roadmap to support digital inclusion and accessibility. At the local level, consultations on issues like health implicated local hospital administrators, medical professionals, and patients to discuss challenges and develop solutions tailored to their territory. For one town that was lacking a general medical practitioner, the territorial consultation on health led to nearby health centres committing their resources to provide the needed health services. The CNR seeks to reshape policymaking by promoting more inclusive and participatory approaches, ensuring that diverse perspectives are considered in the development of long-term reforms and public policies (Conseil National de la Refondation, 2023[12]).
Targeted consultations are not limited to experts or organised stakeholder associations. Governments can also consider using representative deliberative processes such as citizen panels to bring regular people closer to policymaking. Deliberative approaches typically refer to a randomly selected group of people broadly representative of a community spending significant time learning and collaborating through facilitated deliberation to form collective recommendations for policymakers (OECD, 2020[1]). OECD’s understanding of such approaches is based on three criteria: deliberation, representativeness, and impact – which is measured by a clear link between the process and decision-making (OECD, 2020[1]). Carefully choosing the number and issues covered is important given the resources involved in undertaking them (OECD, 2020[1]), coupled with the fact that they may result in delays to decision-making. The European Commission models the use of deliberative approaches through their Citizens’ Panels, which have tackled various policy challenges (Box 2.2).
The Citizens’ Panels are an essential element of the European Commission’s Citizens’ Engagement Platform, which provides a space for citizens to engage in European-level discussions regarding key policy areas. Overall, the citizen engagement process consists of several phases: participation, deliberation, follow-up and feedback.
Following the initial participation phase, where citizens can create, comment on, and vote for contributions from others on the Citizens’ Engagement Platform, the inputs feed into the Citizens' Panels to inform the deliberation. Built on the success of the citizens’ panels held at the Conference on the Future of Europe in 2021, the main objective of these panels is to allow European citizens to voice their expectations for the EU and have an active participation in shaping its future policies.
Each Panel brings together around 150 citizens from the 27 Member States (in which at least one third of the participants are between 16 and 25 years old), which are randomly selected based on a quota system to reflect various diversity criteria (educational background, geographical distribution, gender). Over three weeks, the selected individuals work together with the support of a facilitation team to develop recommendations for the European Commission. During the Panels, participants meet for three weeks during which they generate ideas. Through an iterative process, the Panels generate ideas, draft recommendations and, through an iterative peer review process, refine and expand on each other's work.
Since December 2022, these Citizens’ Panels have directly informed and impacted EU policies:
The Panel on Energy Efficiency’s recommendations are being used by the Commission to inform the development of a proposal for a Recommendation on the “Energy Efficiency First” principle
Recommendations of the Panel on Learning Mobility are informing the Commission’s work on a proposal for a Council recommendation promoting learning mobility for everyone in the European Education Area
Several recommendations of the European Citizens' Panel on Virtual Worlds were reflected in the Commission’s July 2023 communication “An EU initiative on Web 4.0 and virtual worlds: a head start in the next technological transition”.
Source: (European Commission, n.d.[13]), European Citizens’ Panels; (European Commission, 2024[14]), European Citizens' Energy Efficiency Panel - Final Report; (European Commission, 2024[15]), European Citizens’ Panel on Learning Mobility - Final Report; (European Commission, 2023[16]), European Citizens’ Panel on Virtual Worlds - Final Report.
At a national level, countries have employed citizens’ panels to bring ideas directly from people to decision makers. Denmark’s Youth Climate Council aimed to bring new climate policy ideas and provide input to the Minister for Climate on future climate solutions. The members of the Youth Climate Council are appointed for a one-year period and come from all parts of Denmark, from different educational backgrounds, so that they can bring to the table different views and approaches to address the climate challenge. In Finland, the Centre of Excellence in Public Opinion Research at Åbo Akademi University collaborated with the Finnish Parliament to organise a Citizens’ Parliament where randomly selected citizens participate in democratic deliberation and decision making. Guided by a moderator, participants discussed issues like fuel taxation and drug policy; a report summarising discussions will be submitted to relevant parliamentary committees. The aim of the Citizens’ Parliament is to increase diversity in public discourse and increase public acceptance of political decisions (Kansalaisparlamentti, 2023[17]).
After a policy problem is identified, policymakers must determine the best path forward to reach their desired objectives based on a thorough consideration of various policy options – or otherwise risk ineffective rules based on assumptions. This means considering evidence on how different options can help to achieve the desired outcomes, and weighing that against the different costs and risks that each option entails before settling on a specific course. This is important because it allows decision-makers to evaluate and contrast the potential direct and indirect costs and benefits – and their respective distribution (OECD, 2021[2]) – and potential side effects of each option. By considering a range of alternatives, governments can select a solution that is both effective and proportionate to the issue at hand. Critically, comparing different options based on objective evidence elicits inherent trade-offs, enabling decision makers to set out why they prioritised one approach over another. Early comparison of options also helps avoid rushed or poorly thought-out decisions, leading to better long-term policy outcomes. This section discusses how EUMS can better harness the value of thorough options analysis by addressing key gaps in impact assessment requirements and practices.
Identifying and assessing options
With both the Letta (2024[18]) and Draghi (2024[19]) reports calling for the modernisation and simplification of the regulatory framework, the systematic consideration of different and non-regulatory approaches, can play an especially important role in minimising unnecessary (or unnecessarily complex) new regulation without compromising policy objectives. This begins first with the careful assessment of what could happen if no policy action is taken. Importantly, this is not just an assessment of the current status quo, but an assessment of the assumed state of the world in the absence of any policy intervention, taking into account other factors that might influence the outcome over time (OECD, 2020[8]). This assessment provides a counterfactual for policy intervention options. Is a new rule likely to make the situation substantially better for people than the simple passage of time and the natural evolution of the situation? Or would a new rule unnecessarily add to the regulatory stock? Most EUMS already consider this to some degree, with most EUMS (78% for primary laws and 63% for subordinate regulations) systematically including impact analysis for a do-nothing scenario where no action is taken (Figure 2.2).
Source: OECD Indicators of Regulatory Policy and Governance 2024.
Most EUMS also systematically assess multiple alternative regulatory options to inform their decision‑making. However, non-regulatory alternatives remain the least likely to be assessed – and even when they are required, only five EUMS require identifying multiple non‑regulatory options. Countries would benefit from a more systematic consideration of both regulatory and non-regulatory approaches early in their policymaking to ensure that the approach ultimately identified is, in fact, best suited to achieving the desired impact within the given context (OECD, 2022[11]).
As a best practice, impact assessments should identify and assess all feasible alternative options for addressing the policy problem at hand – however, this analysis may either not be undertaken or conducted too late for the analysis to truly support decision‑making. There needs to be freedom to assess a range of genuine alternative solutions, rather than a more limited choice based on predetermined preferences from decision makers to regulate. A tendency remains to use evidence to justify a decision that has already been taken, instead of using evidence to inform the decision itself (OECD, 2020[8]). It risks overlooking alternative, potentially more effective ways of achieving the desired impact, or imposing unnecessary rules and burdens that can compromise the desired impact. To avoid this, decision makers need to be able to consider the implications of multiple options – including the option not to regulate – when selecting a path forward (OECD, 2012[5]) (see Box 2.3). “Doing nothing” also has costs and benefits for the population, like the inaction to address climate change or the unchecked deployment of new technologies, as discussed in Chapter 3.
Since remote gambling licenses were first issued in the Netherlands in October 2021, players lost a lot of money in a short time and new research showed that 31% of players classified as moderate or high‑risk gamblers, highlighting the greater addiction risks compared to gambling in physical locations.
Several key underlying causes were identified: 1) the choice architecture encourages high betting limits, 2) unclear processes around setting and adjusting limits, 3) no measures in the existing framework have restrictive or moderating effect on gaming behaviour, and 4) licensees do not intervene effectively when a player shows excessive or addictive behaviour.
Based on this, the government identified several potential approaches – regulatory and nonregulatory – to protect people against excessive gambling behaviours and prevent addiction.
The baseline “do nothing” option: Without intervention, the choice architecture would likely continue to place the growing number of online gamblers, especially young adults, at risk of addiction and significant monetary loss.
Self-regulation: The Government made some efforts to promote self-regulation through discussions with industry organisation. This led to some licensees adopting minor changes, like reducing pre-filled betting limits, but most remained reluctant to implement adequate measures to address the key underlying issues.
Provide better information: While this option would address the issue of licensees providing unclear or little to no information on playing limits – or making access to that information difficult – it would not address the other identified problems, making it an insufficient solution on its own.
Tighten supervision: The existing rules require licensees to provide clear information on playing limits and systematically monitor the gambling behaviour of players. Increasing supervision could prompt licensees to meet these existing obligations, but would be limited in scope (for instance, the rules don’t explicitly require real-time monitoring, which leads to delays in licensees intervening with problematic behaviour).
Change regulations: New and amended rules could address gaps and ambiguities in the existing framework that the other options could not sufficiently resolve (e.g. explicitly require a neutral setup for setting play limits that does not influence players to choose high limits; require players to confirm higher desired limits; require values to be displayed in euros; and require reminders to help players stay aware of their gambling behaviour, like how long they have been playing and when they are close to reaching their limit).
The proposed approach, aiming to address all the identified underlying problems, involved a combination of adjusting regulations and tightening supervision to ensure effective compliance with the amended rules.
Source: Dutch Ministry of Justice and Security (2024[20]), Regeling speellimieten en speelgedrag.
The timely assessment of all relevant policy options, including non-regulatory alternatives, is especially valuable for identifying the best approach to foster agile regulation that can keep pace with innovation (OECD, 2021[21]). Non-legally binding approaches, either as an alternative or as a complement to regulatory instruments, are a specific element of the OECD’s Recommendation on Agile Regulatory Governance to Harness Innovation (OECD, 2021[22]). Tools like high-level principles, guidelines, technical standards, and codes of conduct are increasingly employed to govern emerging technologies due to their flexibility and adaptability in the face of uncertain technological pathways. However, the suitability of such non-binding approaches must be carefully evaluated in each context, as they present challenges, particularly in terms of enforcement, given their non-binding nature (OECD, 2024[23]). Doing so is crucial to identify the best approach to address regulatory policy problems involving innovations.
Ultimately, each policy option will carry its own considerations regarding potential direct and indirect costs, benefits, and risks that will determine its suitability and feasibility. Policymakers may find themselves weighing immediate costs against long-term benefits or weighing the risk of different unintended consequences associated with different policy options. Chapter 3 includes further discussion on the assessment of these different factors for each option.
Ensuring guidance and oversight
Effectively defining and assessing relevant options can be a challenge for policymakers. Time and resource constraints, as well as political pressures can limit a thorough consideration of alternatives, pushing policymakers towards quicker, less nuanced solutions. Particularly if a preferred option already exists, the definition of alternatives can be an afterthought – a simple exercise in framing the preferred option as the “middle-ground” between an excessive approach and an insufficient approach. Effective guidance and oversight can help avoid the critical process of option defining turning into a checkbox exercise.
Since 2017, EUMS have increasingly provided written guidance on the definition and assessment of regulatory alternatives and baseline scenarios (Figure 2.3). All EUMS provide officials with some form of guidance on the preparation of a RIA and its scope. Guidance on identifying alternatives and baseline do‑nothing scenarios is comparatively trailing, but has seen consistent marginal improvement since 2017.
Source: 2024 OECD Indicators of Regulatory Policy and Governance.
Oversight, however, has been a gap since 2017 despite effective options analysis being a particular challenge for RIAs. Only a minority of EUMS indicate that regulatory oversight bodies can return the RIA for revision where an assessment of alternatives is not appropriately conducted (11 for primary laws and 7 for subordinate regulations). The European Commission’s Regulatory Scrutiny Board (RSB) can – and often does – issue negative opinions on impact assessments if they do not include a clear comparison of regulatory options, including cost-benefit analyses for each. In 2023, the definition of options was consistently the weakest element in first submissions reviewed by the RSB. Common issues included a narrow range of feasible options, options biased toward a preferred outcome, or a lack of consideration for alternative combinations that might arise during the decision-making process. The RSB highlighted that this limited the range of choices available to policymakers and hindered the making of balanced regulatory decisions. Year after year, options analysis remains one of the most challenging aspects of initial submissions (Regulatory Scrutiny Board, 2024[24]).
The Letta report recognises the challenges that European policymakers face in policy preparation, including time constraints and resource limitations, which can be exacerbated by excessive analytical demands (Letta, 2024[18]). Proportionality is an important principle to help limit the burden associated with the lawmaking process itself. This section explores how EUMS use threshold tests that trigger proportionality considerations and how they apply proportionality in determining the scope of RIAs, as well as mechanisms for oversight in both cases.
Threshold tests
Threshold tests are a means of implementing the proportionality principle through a first-decision judgement on the depth of impact analysis required. A threshold for triggering requirements for more extensive evidence can be based on qualitative or quantitative impacts, a mix of qualitative and quantitative criteria (e.g. the number of affected businesses or a subjective determination of the significance of identified impacts on key sectors), impacts on specific stakeholder groups, or the determination of a regulatory oversight body following initial analysis by the policymaker (OECD, 2020[25]). In Denmark, proposals expected to create more than 4 million DKK in administrative burden or more than 10 million DKK in direct compliance costs for regulated businesses would require a more in-depth assessment (including a detailed assessment and rationale for how relevant businesses will be impacted). The European Commission uses a qualitative determination of whether initiatives are expected to have “significant” social, economic, or environmental impacts and policy options, as a trigger for a full impact assessment. The different types of threshold tests and their usage across EUMS and OECD countries are expanded up in the previous edition of this report (OECD, 2022[11]) and in a dedicated annex to the OECD Best Practice Principles on RIA (OECD, 2020[25]).Without carrying out a threshold test, the costs of conducting an impact assessment of a policy with minor impacts may outweigh its benefits (OECD, 2020[25]).
Among EUMS, the use of threshold tests to guide impact analysis remains largely lacking since 2017. Only two EUMS (Italy and Lithuania) use threshold tests to determine whether a RIA is required at all, as established in the previous edition of this report (OECD, 2022[11]). This significantly trails the OECD average, where one-fifth of OECD members use a threshold test to make the same determination. As shown in Figure 2.4, just under one-third of EUMS use qualitative and/or quantitative threshold tests to determine the level of analysis in cases where a RIA is required, (i.e. whether a full RIA for primary laws and subordinate regulations should be undertaken, as opposed to a simplified RIA), compared to almost half of OECD members.
Source: OECD Indicators of Regulatory Policy and Governance (iREG) 2024.
Defining and applying a meaningful threshold can be challenging. It is important that thresholds are inclusive and based on the size of impacts across society rather than focusing on specific sectors or stakeholder groups (OECD, 2020[25]). There should also be consideration of how proposals that carry significant impacts not accounted for by the threshold test will be accommodated. On the other hand, overly complex or vague threshold tests may be more vulnerable to being employed inaccurately or inconsistently. With only 9 EUMS having written guidance on RIA that includes advice on threshold tests, the general lack of available guidance reflects overall gaps in the use of threshold tests among EUMS.
In addition to improving the availability of guidance, EUMS should look to bolster their transparency in the use of threshold tests. Among the minority of EUMS that employ threshold tests for their impact analysis, only four do so transparently. Estonia, Italy, Lithuania, and Croatia publish the results of a threshold test before the regulation is drafted, conveying to the public why a proposal is deemed to warrant a certain level of analysis. The European Commission, in its Call for Evidence for a proposed initiative, also states whether an impact assessment will be carried out for that initiative.
Proportionate impact assessment
The 2012 Recommendation advocates ensuring that the depth and scope of analysis underlying regulatory design are aligned with the significance of the proposed regulation’s potential impacts (OECD, 2012[5]). The majority of EUMS recognise that the level and depth of analysis should be aligned with the proposals’ expected impacts. Over two-thirds of EUMS require that impact assessments are proportionate to the significance of the proposal (22 for primary laws and 19 for subordinate regulations in 2024, up from 18 and 17 respectively in 2017). Contributing to this growth, and as noted in Chapter 1, the Netherlands and Romania are the latest to introduced proportionality requirements for conducting impact assessments since 2021 (Box 2.4). The previous edition of this report also breaks down in detail how proportionality is employed in other EUMS, like Germany, Denmark, Estonia, and Cyprus (OECD, 2022[11]). Previously‑identified membership and geographic trends remain relevant: 1) older and founding EUMS are more likely to require that RIAs are proportionate to the significance of the proposed, than EUMS that joined the EU since 2004; 2) southern EUMS are less likely to require applying proportionality principles to impact assessment (Arndt-Bascle and Davidson, 2022[26]).
The Netherlands
An online questionnaire helps policymakers determine the applicability of relevant impacts and tests for their proposal. The questionnaire requires identifying whether any of seventeen identified social, economic, or environmental impacts are anticipated for the proposal. For each impact that is foreseen, in-depth analysis should be undertaken to define the extent and management of that impact, as appropriate. For instance, answering positively to the question “do you expect regulatory burden effects?” will inform the policymaker that the following tests are mandatory for their proposal’s RIA: regulatory burden assessment, data protection impact assessment, and social cost benefit analysis.
The RIA template – referred to as a “Policy Compass” – is inherently modular. Policymakers complete relevant sections for their proposal, as advised based on their questionnaire.
Romania
Government Decision 443/2022 introduced, among other changes, proportionality for RIAs based on the level of expected impact. In-depth RIAs should only be conducted for proposals that are of “significant importance and complexity”, with the following criteria provided as guidance:
1. New regulations in a particular field or regulations that involve a substantial change to a current law;
2. Major reforms included in the Government Programme or other programmatic documents undertaken by the Government of Romania that generate systemic changes and that affect a large number of citizens, especially in areas such as education, social protection, health or business;
3. It directly affects vulnerable groups as defined by the Social Assistance Law;
4. It has an impact on at least 10% of the active companies or at least 20% of the employees in a certain sector of activity, according to the official data provided by the institutions with responsibilities in the field (National Institute of Statistics, National Bank of Romania, the National Commission for Strategy and Forecast, the Ministry of Labour and Social Solidarity etc.);
5. Its implementation involves a total net cost of more than 20 million lei over the general consolidated budget;
6. Generates an impact of over 100 million lei on public investments.
In-depth RIAs involve, for instance, more extensive risk assessment, as well as requirements to quantify more costs to businesses.
Source: OECD Indicators of Regulatory Policy and Governance Survey 2024.
The frequency and extent to which simplified and in-depth RIAs are employed varies across countries. For instance, in Austria, a simplified RIA is carried out for about two-thirds of all regulations and focuses on a simplified assessment of financial costs, while the full RIA evaluates a suite of economic, social, and environmental effects, and requires prescribing indicators to measure progress. In Estonia, a simpler preliminary RIA is developed for all primary laws and selected subordinate regulations. The level of analysis contained within their preliminary RIA has deepened over time, making in-depth RIAs increasingly irrelevant and rarely done (OECD, 2021[2])).
The practical difference in the level of analysis between a full and simplified RIA varies across countries. In Austria, a simplified RIA does not require policymakers to define indicators to measure progress, relying instead on a simplified assessment of financial costs and brief narrative descriptions. Cyprus similarly uses a more descriptive – rather than quantitative – analysis in their simplified RIAs. In Denmark, the simplified RIA must define Standard Cost Model variables to the greatest extent possible to assess whether the proposal meets the threshold for in-depth analysis; if so, a full RIA includes more thorough assessment of impacts on companies based on both costs and interviews.
In some cases, proportionality may deem that an impact assessment is not required. For example, minor legislative changes, administrative updates, or regulations that primarily involve existing frameworks may qualify for exemptions, allowing for a streamlined approach to rule-making. As well, the risk of not taking swift action may be so great as to warrant bypassing RIA requirements to expedite regulatory responses. There should be only limited exceptions to the general rule that RIA is required. Where exceptions are invoked (and to reduce the incentive for their misuse), it should be mandatory to conduct an ex post evaluation in such cases (OECD, 2020[8]) – though this is only the case for five of the EUMS that allow for RIA exemptions. In the European Commission, for example, if an impact assessment is waived due to urgency, an analysis must be published within three months of the initiative’s adoption.
Requirements, however, do not always translate into practice. The RSB observed that in their 2023 report that, across the RIAs they reviewed, “getting proportionality and the level of analysis right is probably the most difficult challenge” (Regulatory Scrutiny Board, 2024[24]). Croatia, meanwhile, adopted a new system moving away from proportionate RIAs in an effort to address compliance challenges with RIA requirements by making the process more straightforward. While the challenges associated with employing proportionality in regulatory design make effective oversight more important, that in itself is a gap for several EUMS with only 40% having a body responsible for reviewing the decision made by officials about whether a RIA is required. Chapter 3 further discusses RIA and the assessment of different social, economic, and environmental impacts.
Using open and evidence-based tools at the outset can pave the way for the following stages of the rule-making process. Once policymakers have engaged citizens and other stakeholders to identify a clear rationale for intervention and objectives, the consistent application of proportionality helps set the course for evidence-based and proportionate regulatory design process, as set out in Chapter 3.
References
[26] Arndt-Bascle, C. and P. Davidson (2022), European Union and Member States’ interface: Background paper, https://one.oecd.org/document/GOV/RPC(2022)9/ANN2/en/pdf.
[12] Conseil National de la Refondation (2023), La méthode CNR, https://conseil-refondation.fr/fonctionnement/fonctionnement/ (accessed on 23 October 2024).
[19] Draghi, M. (2024), The future of European competitiveness, https://commission.europa.eu/topics/strengthening-european-competitiveness/eu-competitiveness-looking-ahead_en.
[3] Drummond, J., D. Shephard and D. Trnka (2021), “Behavioural insight and regulatory governance: Opportunities and challenges”, OECD Regulatory Policy Working Papers, No. 16, OECD Publishing, Paris, https://doi.org/10.1787/ee46b4af-en.
[20] Dutch Ministry of Justice and Security (2024), Regeling speellimieten en speelgedrag, https://www.internetconsultatie.nl/speellimieten/b1.
[14] European Commission (2024), European Citizens’ Panel on Energy Efficiency - Final Report, https://citizens.ec.europa.eu/european-citizens-panels/energy-efficiency-panel_en.
[15] European Commission (2024), European Citizens’ Panel on Learning Mobility - Final Report, https://citizens.ec.europa.eu/european-citizens-panels/learning-mobility-panel_en.
[16] European Commission (2023), European Citizens’ Panel on Virtual Worlds - Final Report, https://citizens.ec.europa.eu/european-citizens-panels/virtual-worlds-panel_en.
[13] European Commission (n.d.), European Citizens’ Panels, https://citizens.ec.europa.eu/european-citizens-panels_en (accessed on 15 June 2025).
[9] European Commission (n.d.), Planning and proposing law, https://commission.europa.eu/law/law-making-process/planning-and-proposing-law_en#:~:text=quality%20is%20ensured-,Right%20of%20initiative,its%20citizens%20as%20a%20whole (accessed on 15 June 2025).
[10] European Commission (n.d.), Published initiatives, https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives_en?frontEndStage=PLANNING_WORKFLOW (accessed on 15 June 2025).
[17] Kansalaisparlamentti (2023), Hur arbetar medborgarparlamentet?, https://www.kansalaisparlamentti.net/sv/hur-jobbar-medborgarparlamentet/.
[18] Letta, E. (2024), Much more than a market: Empowering the Single Market to deliver a sustainable future and prosperity for all EU Citizens, European Union, https://www.consilium.europa.eu/media/ny3j24sm/much-more-than-a-market-report-by-enrico-letta.pdf.
[4] OECD (2024), Facts not Fakes: Tackling Disinformation, Strengthening Information Integrity, OECD Publishing, Paris, https://doi.org/10.1787/d909ff7a-en.
[23] OECD (2024), “Framework for Anticipatory Governance of Emerging Technologies”, OECD Science, Technology and Industry Policy Papers, No. 165, OECD Publishing, Paris, https://doi.org/10.1787/0248ead5-en.
[6] OECD (2023), Conference Proceedings – 14th OECD Conference on Measuring Regulatory Performance, https://one.oecd.org/document/GOV/RPC/RD(2023)6/REV1/en/pdf.
[11] OECD (2022), Better Regulation Practices across the European Union 2022, OECD Publishing, Paris, https://doi.org/10.1787/6e4b095d-en.
[7] OECD (2022), Establishing Regulatory Impact Assessment in Mauritius, OECD Publishing, Paris, https://doi.org/10.1787/2d072a27-en.
[2] OECD (2021), OECD Regulatory Policy Outlook 2021, OECD Publishing, Paris, https://doi.org/10.1787/38b0fdb1-en.
[21] OECD (2021), Practical Guidance on Agile Regulatory Governance to Harness Innovation, OECD, Paris.
[22] OECD (2021), Recommendation of the Council for Agile Regulatory Governance to Harness Innovation, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0464.
[25] OECD (2020), A closer look at proportionality and threshold tests for RIA: Annex to the OECD Best Practice Principles on Regulatory Impact Assessment, OECD Publishing, Paris.
[1] OECD (2020), Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave, OECD Publishing, Paris, https://doi.org/10.1787/339306da-en.
[8] OECD (2020), Regulatory Impact Assessment, OECD Best Practice Principles for Regulatory Policy, OECD Publishing, Paris, https://doi.org/10.1787/7a9638cb-en.
[5] OECD (2012), Recommendation of the Council on Regulatory Policy and Governance, OECD Publishing, Paris, https://doi.org/10.1787/9789264209022-en.
[24] Regulatory Scrutiny Board (2024), Regulatory Scrutiny Board Annual Report 2023, Publications Office of the European Union, https://doi.org/10.2792/93331.