Chapter 3. Regulatory impact assessment across the European Union

Regulatory impact assessment (RIA) is a key tool for policy makers to decide on whether and how to regulate to achieve public policy goals. RIA assists policy makers in identifying the most efficient and effective policy before making a decision. The use of RIA has expanded over the past 30 years across EU countries. This chapter analyses the EU’s and EU Member States’ use of RIA and explores how EU Member States consider various options and impacts when conducting RIA. It also includes a discussion of the application of individual Member State’s RIA requirements to regulatory proposals of the European Commission.

    

Note by Turkey: The information in this document with reference to “Cyprus” relates to the southern part of the Island. There is no single authority representing both Turkish and Greek Cypriot people on the Island. Turkey recognises the Turkish Republic of Northern Cyprus (TRNC). Until a lasting and equitable solution is found within the context of the United Nations, Turkey shall preserve its position concerning the “Cyprus issue”.

Note by all the European Union Member States of the OECD and the European Union: The Republic of Cyprus is recognised by all members of the United Nations with the exception of Turkey. The information in this document relates to the area under the effective control of the Government of the Republic of Cyprus.

Introduction

Regulatory impact assessment (RIA) is both a tool and a process designed to help inform political decision makers on whether and how to regulate to achieve public policy goals. Improving the evidence base through an ex ante impact assessment is one of the most important regulatory tools available to governments. The aim is to improve the design of regulations by assisting policy makers identify and consider the most efficient and effective options — including non-regulatory options — before making a decision. One method of doing so is by analysing the expected costs and benefits of regulation and of alternative means of achieving policy goals and to identify the approach that is likely to deliver the greatest net benefit to society.

The consideration of a range of alternative approaches to traditional “command and control” regulation — including complementary measures such as co-regulation — helps to ensure that the most efficient and effective approaches are used in attaining policy goals. Experience shows that governments must lead strongly to overcome inbuilt inertia, risk aversion and a “regulate first, ask questions later” culture. At the same time, care must be taken when deciding to use light-handed approaches such as self-regulation, to ensure that public policy objectives are achieved.

To that end, the OECD Recommendation of the Council on Regulatory Policy and Governance provides that member countries should “Integrate Regulatory Impact Assessment (RIA) into the early stages of the policy process for the formulation of new regulatory proposals. Clearly identify policy goals, and evaluate if regulation is necessary and how it can be most effective and efficient in achieving those goals” (OECD, 2012[1]).

For nearly 20 years the European Union has focused on improving regulatory outcomes through its better regulation initiatives (see Chapter 1). As part of that, RIA has been recognised as a centrally important regulatory management tool by the European Commission.

This chapter analyses EU Member States’ RIA requirements and practices as reported in the Indicators of Regulatory Policy and Governance (iREG) survey and its extension to all EU Member States. The reader’s guide explained the construction of the surveys. This chapter also presents results from the iREG survey as they relate to the European Union and other OECD member countries where appropriate. The section below presents the key findings of EU Member States’ RIA requirements and their implementation. The second section provides an overview of key results that emanate from the iREG survey. The third section discusses the scope of RIA in EU Member States, examining exceptions to conducting RIA, and requirements to adhere to the proportionality principle. The section on the application of Member States’ RIA to EU legislation covers EU Member States’ RIA requirements as they relate to EU-made laws. The section on options considered in RIA discusses the requirements and implementation of the use alternative options in RIA. The following section discusses the types of impacts assessed by EU Member States. The section on effectiveness of RIA presents information on the extent to which Member States have assessed the effectiveness of the RIA systems in practice.

Key findings

Nearly all EU Member States have embedded RIA as a core part of their regulatory management tools, and generally have explicit high-level political support for RIA. Malta does not currently have RIA requirements for primary laws, and Bulgaria and Croatia have no RIA requirement for subordinate regulations.

There is a gap between the level of outward commitment and what is witnessed in practice. In particular, there is a substantive gap between the implementation of RIA in practice to primary laws vis-à-vis subordinate regulations. This is a particularly acute problem given the lack of legislative oversight afforded to most delegated legislation. Only half of the EU Member States systematically publish RIA for the purposes of consultation on regulatory proposals relating to primary laws.

A central tenet of RIA is that it be proportionate to the level of expected impacts. Part of the proportionality principle refers to the need to triage regulatory proposals so as to ensure that impact assessment resources are appropriately targeted. Despite their importance, EU Member States do not use threshold tests to filter regulatory proposals; although the majority do have a requirement that the level of impact assessment undertaken be proportionate to the regulatory proposal’s expected impacts.

Only around half of EU Member States have a requirement to consider the baseline or ‘do nothing’ option. Perhaps more worryingly, in relation to subordinate regulations, in less than half of EU Member States are alternatives always identified and assessed. However, an integral part of a well-functioning RIA system is ensuring that all feasible options — including non-regulatory options — are systematically considered and assessed. Considering various options helps to inform decision makers about the expected impacts of regulatory proposals, and by doing so, helps to reduce the risk of regulatory failure.

EU Member States generally assess the following impacts: government, business (including small business), competition, and the environment. However, it should be noted that the impact assessment generally only relates to the preferred regulatory option and often does not involve a quantitative assessment of the regulation’s costs and benefits. Quantification of benefits are less common across EU Member States than quantifying costs, which can make assessing whether regulations justify their costs difficult.

The European Commission requires that RIA be conducted for regulatory proposals with significant economic, environmental or social impacts on the EU. The Commission publishes an inception impact assessment (IIA) that outlines the policy problem and a preliminary assessment of the anticipated impacts. Following public consultation on the IIA, the Commission undertakes a full RIA including data collection and evidence gathering, as well as public consultations. RIAs are subject to scrutiny by the Regulatory Scrutiny Board, where regulatory proposals may need to be revised.

In terms of EU-made laws, RIA is much more likely to be required when transposing EU directives than it is to form the basis of individual EU Member States’ negotiating position. Where Member States do not have formal requirements to conduct RIA at the negotiating phase, they do rely on the European Commission’s impact assessments, albeit not systematically. Generally when RIA is conducted for transposing EU directives, the same procedures apply as is the case with domestically made laws. Around half of the EU Member States systematically require an assessment of whether provisions have been added at the national level going beyond the requirements set out in the EU directives as part of their RIA when transposing EU directives.

Assessments of the effectiveness of RIA systems have been undertaken in a number of EU Member States, and include providing guidance on good regulatory practices, however these assessments would benefit from an increased focus on how systems can be improved over time.

General trends in RIA across the EU

There is quite a lot of variation across EU Member States’ RIA systems, reflected by the variance in composite indicators relating to RIA on primary laws (Figure ‎3.1). The reader’s guide at the beginning of this report covered the construction of the composite indicators, including the disclaimers concerning conclusions that may be drawn given the inherent limitations of cross-country comparable composite indicators.

The variance across Member States is quite pronounced and is around twice the variance that exists for stakeholder engagement in relation to primary laws. EU Member States have strong RIA requirements in place. Italy, Lithuania, and the United Kingdom have formal threshold tests for determining whether RIA should be undertaken at all; and Austria, Denmark, Estonia, Lithuania, Spain, and the United Kingdom have formal threshold tests to determine whether a simplified or full RIA needs to be undertaken. Croatia, Hungary, Ireland, Italy, and Slovenia include formal consequences in instances where regulatory proposals bypass the RIA system during regulatory design.

Figure 3.1. Composite indicators: regulatory impact assessment for developing primary laws, 2018
picture

Notes: Data for 2015 is based on the 34 countries that were OECD members in 2014 and the European Union, which included 21 of the current 28 EU Member States. The OECD average is based on the 34 member countries at the time of the survey. Data for 2018 includes the remaining EU Member States of Bulgaria, Croatia, Cyprus, Malta and Romania. The more regulatory practices as advocated in the 2012 Recommendation a country has implemented, the higher its iREG score. * In the majority of EU Member States, most primary laws are initiated by the executive, except for Bulgaria, where a higher share of primary laws are initiated by the legislature.

Source: Indicators of Regulatory Policy and Governance Surveys 2014 and 2017, http://oe.cd/ireg.

The survey results indicate that EU Member States have sound RIA methodologies in place for some areas, such as the specific impacts that are assessed. This is especially the case in the United Kingdom. However, it also suggests that there is room for improvement in RIA, such as in considering various options when designing regulatory proposals which relate to primary laws. Austria, Estonia, France, Italy, Poland, and the United Kingdom all have strong ex ante requirements in place to ensure that achievement of the regulation’s goals are assessed.

Transparency of RIA is relatively low across EU Member States and is just below the OECD average. RIAs are not made public for consultation at an early stage of policy development, and even after a regulatory decision has been made; only half of the EU Member States systematically publish RIA for consultation.

Oversight and quality control is the weakest area surveyed, and is lower than the OECD average. General quality control mechanisms such as assessing their own RIA systems and parliamentary oversight tend to be practises observed in only a few EU Member States. General oversight is reasonably strong in the Estonian, German, Lithuanian, and British RIA systems. The survey results indicate that Member States (and OECD countries) still have some way to go in order to reach the practices envisaged by both the European Union (European Commission, 2005[2]) and the OECD Recommendation.

Composite indicator scores for RIA relating to subordinate regulations are worse on average across each of the four dimensions than the respective scores for primary laws (Figure ‎3.2). Across the four areas, the results illustrate a similar pattern in that the strongest area is systematic adoption, followed by methodology, with transparency and oversight and quality control further behind.

Figure 3.2. Composite indicators: regulatory impact assessment for developing subordinate regulations, 2018
picture

Notes: Data for 2015 is based on the 34 countries that were OECD members in 2014 and the European Union, which included 21 of the current 28 EU Member States. The OECD average is based on the 34 member countries at the time of the survey. Data for 2018 includes the remaining EU Member States of Bulgaria, Croatia, Cyprus, Malta and Romania. The more regulatory practices as advocated in the 2012 Recommendation a country has implemented, the higher its iREG score.

Source: Indicators of Regulatory Policy and Governance Surveys 2014 and 2017, http://oe.cd/ireg.

Despite the strongest area being systematic adoption, there remains room for improvement. Compared with primary laws, RIA is conducted in practice less systematically across EU Member States. While some of this may be due to the fact that subordinate regulations can be of a more technical nature, it does not mean that the impacts are not significant.

In terms of RIA methodology, EU Member States tend to analyse a narrower range of economic, social, and environmental impacts for regulatory proposals relating to subordinate regulations, compared with those assessed for regulatory proposals relating to primary laws. In part, this can be explained by the fact that regulatory requirements are less comprehensive for subordinate regulations than they are for primary laws. Nevertheless, it is not immediately clear why RIA requirements should — by virtue of the regulatory instrument alone — be less stringent for subordinate regulations than for primary laws. Indeed, at the margin, this may create poor incentives for policy makers to design regulatory proposals as subordinate regulations where they should more appropriately be designed as primary laws, so as to subvert some RIA requirements, as well as potential parliamentary scrutiny.

The level of transparency associated with regulatory proposals relating to subordinate regulations is less systematic than it is for regulatory proposals relating to primary laws. In particular, the transparency covering instances where RIA is not conducted is more sporadically applied across the EU Member States than it is in relation to primary laws.

Similarly to regulatory proposals relating to primary laws, regulatory proposals relating to subordinate regulations fare poorly in terms of oversight and quality control. Where there are differences between the application of RIA to primary laws and subordinate regulations, they mainly relate to quality control aspects such as whether a method exists to ensure that an assessment of impacts is completed; as well as in relation to factors affecting oversight such as whether a government body outside of the proposing ministry is responsible for RIA quality control.

Scope of RIA

The application of RIA to regulatory instruments should be optimised to promote the best policy outcomes. Taking a proportionate approach is recommended so that the resources required by RIA are applied to those regulations likely to have the most significant impact on societal welfare (OECD, 2009[3]).

The Annex to the OECD Recommendation provides that member countries should “Adopt ex ante impact assessment practices that are proportional to the significance of the regulation, and include benefit cost analyses that consider the welfare impacts of regulation taking into account economic, social and environmental impacts including the distributional effects over time, identifying who is likely to benefit and who is likely to bear costs” (OECD, 2012[1]).

The vast majority of EU Member States have a universal requirement to conduct RIA to inform the development of primary laws (Figure ‎3.3). While this strong commitment to RIA is crucially important, it is worth noting that there is a gap between the commitment and the extent to which it occurs in practice.

The application of RIA to subordinate regulations is cause for concern: in only half of EU Member States is there a universal requirement to conduct RIA. That said, six Member States have a requirement to conduct RIA in relation to major subordinate regulations, perhaps indicating a more proportionate approach. It is important to recall that subordinate laws are generally subject to substantially less parliamentary oversight than primary laws, yet can have significant impacts on societal wellbeing. Moreover, subordinate regulations are often responsible for bringing primary laws ‘to life’, and as such represent a substantive proportion of the regulatory burden faced by citizens and business. Therefore, it is important to ensure that subordinate regulations are formed on an appropriately sound evidential basis.

Figure 3.3. Requirements and practice of conducting RIA
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Surveys 2017, http://oe.cd/ireg.

Exceptions to conducting RIA

Despite having broad requirements in place to conduct RIA, EU Member States currently have a number of exceptions (Figure ‎3.4). Around 40% of EU Member States provide for an exception to conducting RIA where regulation is introduced in response to an emergency. The European Commission currently also has exceptions to RIA relating to emergency measures, as well as in relation to regulatory proposals which have insignificant impacts.

Figure 3.4. Exceptions to conducting RIA
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

A number of Member States except RIA requirements where regulation relates to the implementation of international law, as well as in relation to a range of other circumstances. For instance in Belgium, laws that relate to the organisation of the State itself; implement a co-operation agreement between the federal state and regional entities; or relate to national security or public order are excepted from RIA. In Ireland, the following types of laws are excepted from RIA:

  • Consolidation of existing legislation where no regulatory changes are introduced

  • The law is as a direct consequence of a Court decision and leaves no discretion to consider alternative options or allow for meaningful consultation

  • Where the publication of RIA may not be appropriate, for example in the imposition of tax laws or the imposition of charges because of their sensitivity and the need to guard against possible invasion or avoidance.

Proportionality and RIA threshold tests

The proportionality principle is at the heart of impact assessment in both theory and practice. It governs the depth and breadth of analysis to be undertaken; the stakeholders to consult with; as well as considering matters such as data collection, compliance and enforcement, and eventual review.

Proportionality requirements exist in around 60% of EU Member States (Figure ‎3.5). The OECD has long endorsed the proportionality principle (OECD, 1995[4]), and it was formalised as part of the OECD Recommendation in 2012 (see above). Formal proportionality requirements are part of the RIA systems in nearly 70% of OECD member countries.

Figure 3.5. Proportionality requirements when conducting impact assessments
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg

RIA thresholds tests are important for a number of reasons. Firstly, they help to ensure that regulations with significant societal impacts are adequately assessed before being introduced. Secondly, they ensure that government resources are not unduly wasted in assessing regulatory proposals with only minor impacts, where the costs of conducting RIA would outweigh its benefits. In turn, this also helps to avoid consultation fatigue (and associated wasted resources) with the community — consultations should be conducted in circumstances where the regulatory impacts are expected to be significant. Thirdly, where the criteria are clear and straightforward to apply in practice, they help to improve the transparency of the overall RIA system. This helps to build trust in the community that regulatory proposals are appropriately subjected (and excepted) to RIA.

It is important that RIA threshold tests are based on the expected significance of impacts. The impacts include both positive and negative impacts to any area of society, and as such ought to be broader than assessing business impacts for instance. Despite their importance however, evidence from the iREG survey and its extension to all EU Member States indicates that EU Member States have been slow adopters. Only Italy, Lithuania, and the United Kingdom have a threshold test to determine whether RIA should be undertaken at all (Figure ‎3.6). None of the three Member States’ tests are explicitly based on the expected significance of the regulatory proposal’s economic, social and environmental impacts. That said, Italy’s threshold test does consider whether compliance costs are expected to be low, that there are a small number of affected entities, the regulatory proposal requires a small amount of public resources, and that there is a limited expected impact on the competitive structure of the affected market. It should also be noted that the European Commission has a comprehensive threshold test based on the regulatory proposal’s anticipated impacts to determine whether RIA should be undertaken at all.

Figure 3.6. RIA threshold tests
picture

Note: Data is based on 28 EU Member States

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

A greater number of EU Member States have a threshold to determine whether a full RIA (as opposed to a simplified RIA) is undertaken, yet it is still in the vast minority. The lack of a threshold at this stage of regulatory development is reflective more broadly of the dearth of two-stage RIA processes (preliminary/simplified and full) across EU Member States.

Overall, it appears that EU Member States have requirements to adhere to the proportionality principle, but often lack a formal threshold test with transparent criteria about when, and to what extent, RIA should be conducted.

The application of Member States’ RIA to EU legislation

Similarly to stakeholder engagement (Chapter 2), individual EU Member States have processes in place which require RIA to be conducted at the transposition stage in relation to EU-made laws, but much less at the negotiating phase. The process of creating EU-made laws was described in detail in Chapter 1, and the process that the European Commission follows for its RIA is described in Box ‎3.1.

Box 3.1. The European Commission’s impact assessment process

Impact assessments are required for all of the Commissions initiatives when the expected economic, environmental or social impacts of EU action are likely to be significant. The lead Commission service should establish as early as possible in the Commissions’ internal political validation process whether a RIA is required. Once a decision has been made that a proposal will be prepared, the Commission publishes an inception impact assessment (IIA) for all proposals subject to an impact assessment. IIAs present an outline of the policy problem, an initial mapping of the policy options as well as preliminary assessment of expected impacts. Where no impact assessment is prepared, the justification should be made public in a roadmap.

Following the public feedback period on the IIA (see Chapter 2), the Commission prepares and undertakes the full impact assessment process, including data collection, public and stakeholder consultations, expert hearings and/or seeking additional scientific evidence. The results of the impact analysis will be summarised in an impact assessment report and sent to the Regulatory Scrutiny Board (RSB) for a quality assurance review. Following potential revisions of the impact assessment based on the opinion from the RSB, the RIA is subject to internal consultation between the Commission’s departments together with the accompanying policy initiative. Once a regulatory proposal has been adopted by the College of Commissioners, the proposals and its accompanying impact assessment will be published online for feedback and sent the co-legislators for the negotiation of the Commission’s proposals.

Source: European Commission (2017), Better Regulation Guidelines, https://ec.europa.eu/info/law/law-making-process/planning-and-proposing-law/better-regulation-why-and-how/better-regulation-guidelines-and-toolbox_en (accessed September 2018); European Commission (2017), Better Regulation Toolbox, https://ec.europa.eu/info/better-regulation-toolbox_en (accessed September 2018).

Member States’ requirements to conduct RIA during the negotiation stage

RIA provides the largest net benefit to policy makers when it is commenced early in the regulation-making process; and the same applies to EU-made laws. Once the European Commission has reached a regulatory decision, individual Member States can begin to consider the individual impacts of it.

There are currently 13 EU Member States that have requirements to conduct RIA so as to define a negotiating position on European Commission regulatory proposals (Figure ‎3.7). The 13 Member States are: Austria, Bulgaria, Estonia, Finland, Germany, Hungary, Ireland, Italy, Lithuania, Malta, Poland, the Slovak Republic, and Slovenia.

Figure 3.7. Requirements to conduct RIA to define a negotiating position
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Germany has a specific threshold test during the negotiating stage. Where the Commission’s own impact assessment identifies expected compliance costs in excess of €35 million per year across the EU, German Ministries are required to carry out RIA to assess the compliance costs that are expected to arise in Germany. The RIA then forms the basis for the Federal Government in its negotiation of the legislative proposal at the European Union level.

In Ireland, departments are required in-principle to conduct RIAs on all Commission proposals leading to EU directives and significant EU regulations. The revised RIA guidelines indicate that the RIA process should be commenced as early as possible as and no later than four weeks from when the Commission publishes the proposed legislation and its own impact assessment. The RIA should contain a sufficient level of analysis of key issues to properly inform Ireland’s negotiating position, thereby minimising any potential negative implications for Ireland. The guidelines encourage officials to update the RIA periodically to take account of significant changes introduced at various stage of the proposal’s development (Department of the Taoiseach (Ireland), 2009[5]).

In Lithuania, policy makers are required to complete a ‘basic impact assessment’ in reaching its negotiation position. Basic impact assessments examine objectives and options for implementation of draft legislation, including quantitative data whenever possible. Extended impact assessments in Lithuania are more detailed and provide for more in-depth analysis when basic impact assessments are deemed insufficient or when substantive social, political or economic impacts are likely to materialise after implementation (Government of Lithuania, 2008[6]).

In addition to conducting RIA, Estonia, Hungary, Italy, Poland, the Slovak Republic, and Slovenia all indicated that they undertake stakeholder engagement during the negotiation stage (see Chapter 2).

For the 15 EU Member States that are not required to conduct RIA to inform a negotiating position, they are generally split on whether and to what extent they use the European Commission’s impact assessment (Table ‎3.1).

Table 3.1. Use of the European Commission’s impact assessment
To inform a negotiating position where no RIA requirement exists at an individual EU Member State level

Always

Frequently

Sometimes

Never

Member States

United Kingdom

France, Romania

Belgium, Croatia, Czech Republic, Luxembourg, Netherlands

Cyprus, Denmark, Greece, Latvia, Portugal, Spain, Sweden

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

In France for example, the 2017 Guidance on drafting legal norms provides that national impact assessments of transposition laws must include a reference to the impact assessment performed by the European Commission (Premier Ministre Secrétariat Général du Gouvernement, Conseil d'Etat (France), 2017[7]).

RIA at the transposition stage

Nearly all EU Member States are required to conduct RIA when transposing EU directives into national laws (Figure ‎3.8). The only Member State that does not require RIA is Romania. This is reflective of the fact that RIA for the transposition of EU directives is explicitly excepted under Article 6(2)(c) of Government Decision No. 561/2009.

Figure 3.8. Requirements to conduct RIA when transposing EU directives
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

In France, the 2017 Guide on the Drafting of Laws highlights the Prime Minister’s Administrative Regulation (circulaire) which provides that RIA must be undertaken at the earliest moment possible during the negotiation phase of draft EU laws. Once the EU directive is published, a legal impact assessment of transposition measures is undertaken, with a particular focus on precise and unconditional provisions. On EU regulations, the impact assessment is limited to the legal modifications that arise from the regulation’s direct effect (Premier Ministre Secrétariat Général du Gouvernement, Conseil d'Etat (France), 2017[7]).

Generally, the same RIA guidelines apply to the transposition of EU directives as they do to domestically-created laws. To some extent this is to be expected: transposing EU directives often includes amending existing national laws or developing new domestic regulations. The only countries where this is not the case are Romania, Estonia, and Sweden. Romania does not have a RIA requirement for EU directives, whereas domestic procedures do apply to domestically-created laws.

Estonia has produced bespoke guidance on domestic procedures to follow when dealing with EU-made laws. It is required that policy makers undertake a preliminary impact assessment which provides details on the expected Estonian budgetary impact of implementing the proposal, and the expected impacts to specific Estonian sectors and groups. In particular, it is required to indicate the data sources utilised; note whether a more detailed RIA is required; as well as to highlight significant effects on Estonia that the European Commission has not addressed in its impact assessment or where the assessments differ.

Requirements to assess gold plating

Just over half of the EU Member States include a specific assessment of provisions added at the national level which go beyond those established in the EU directive (Figure ‎3.9, left pane), so called “gold plating” (Box ‎3.2). This assessment is in effect an assessment of the total regulatory impacts, that is those imposed by the EU and then additionally via the individual Member State.

Figure 3.9. Requirements to assess gold plating and national additional implementation measures
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Some Member States have oversight institutions in charge of identifying gold plating of EU directives. For instance in the Slovak Republic, the Better Regulation Center is responsible for, among other things, identifying gold plating of EU directives. Having an entity directly responsible for assessing gold plating may provide feedback pressures on regulatory agencies when transposing EU directives. Similar efforts to provide oversight of the transposition of EU directives with the aim to avoid over-implementation in Denmark were discussed in Chapter 1.

Box 3.2. Gold plating of EU directives

Gold plating is a term specifically used in the EU context and describes over-implementation of an EC directive through the imposition of national requirements going beyond the actual requirements of the directive. Directives allow Member States to choose how to meet the objectives set out in the directive, adapting their approach to their own institutional and administrative cultures. Moreover, many directives are minimum directives allowing Member States to go beyond their requirements. It is therefore often at this stage that additional details and refinements, not directly prescribed by the directive, are introduced. These can go well beyond the requirements set out in the directive, resulting in extra costs and burdens.

Source: OECD (2009), “Better Regulation in Europe: An OECD Assessment of Regulatory Capacity in the 15 Original Member States of the EU, Project Glossary”, OECD, Paris, https://www.oecd.org/gov/regulatory-policy/44952782.pdf.

In its transposition guidance from February 2018, the UK Government included a guiding principle to ensure that UK businesses are not put at a competitive disadvantage with their European counterparts. Consequently, the United Kingdom should not exceed the minimum requirements of EU directives. Any gold plating must be explained in the RIA and is subject to oversight by the Reducing Regulation Committee, which was created in May 2010 to scrutinise transposition plans (Department of Business, Energy & Industrial Strategy (UK), 2018[8]). Similarly, the Czech Regulatory Impact Assessment Guidelines from February 2016 include an obligation to identify whether the transposition of EU directives goes beyond what is required by EU law. However, where the transposed directive provides for no discretion, or the draft regulation transposing the EU law does not exceed the minimum requirements, the draft regulation is exempted from Czech RIA requirements (Government of the Czech Republic, 2016[9]).

Whilst requirements to assess gold plating exists in just over half of the EU Member States, only eight are required to assess the marginal impact that the gold plating provisions have had (Figure 3.9, right pane). In Sweden for example, the consequences of regulations going beyond EU law are supposed to be separately reported in impact assessments. In particular, competition effects on companies can be reported under a specific heading. This does not apply exclusively to the transposition of EU directives, but also to the adjustments that might be required at national level to implement EU regulations. Examples of drafting templates are given to assist Swedish officials (Swedish Agency for Economic and Regional Growth, 2018[10]).

Options considered in RIA

In order to best assist decision makers in making better informed decisions, it is important that all feasible options are canvassed at an early stage of regulatory development. Consultation at an early stage is therefore important to ensure that all feasible options are genuinely considered and assessed (see Chapter 2). The OECD Recommendation provides that member countries should “Consider means other than regulation and identify the tradeoffs of the different approaches analysed to identify the best approach” (OECD, 2012[1]).

The consideration of alternative options is further elaborated on in the Annex to the OECD Recommendation: “Ex ante assessment policies should include a consideration of alternative ways of addressing the public policy objectives, including regulatory and non regulatory alternatives to identify and select the most appropriate instrument, or mix of instruments to achieve policy goals. The no action option or baseline scenario should always be considered. Ex ante assessment should in most cases identify approaches likely to deliver the greatest net benefit to society, including complementary approaches such as through a combination of regulation, education and voluntary standards” (OECD, 2012[1]). In a previous study of 15 EU Member States almost a decade ago, it was found that alternatives to regulation were not given sufficient attention, and that encouragement to consider alternatives was not always prevalent in impact assessment processes (OECD, 2009[11]).

The European Commission requires policy makers to identify and assess the impacts of: the preferred regulatory option; the baseline or ‘do nothing’ option; and regulatory and non-regulatory alternatives—in relation to all major regulatory proposals.

In 26 EU Member States, the preferred regulatory option needs to be identified and assessed for regulatory proposals relating to primary laws (Figure ‎3.10). The baseline or “do nothing” option is systematically considered as part of RIA in around two-thirds of EU Member States. Regulatory and non-regulatory alternatives are universally assessed in half of the EU Member States. Despite the recommendation above, OECD member states’ consideration of various policy options is in fact less prevalent than those relating to the EU Member States: around 80% of OECD members are systematically required to assess the preferred regulatory option; with broadly comparable results in terms of assessing the baseline options, and regulatory and non-regulatory alternatives.

Figure 3.10. Consideration of various options in RIA for primary laws
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Recalling that RIA is undertaken less systematically for subordinate regulations than it is for primary laws, it is perhaps to be expected that the consideration of various policy options is also less systematically undertaken for subordinate regulations compared with primary laws. Indeed, this is the case: for subordinate laws the preferred regulatory option is always identified and assessed in only half EU Member States (Figure ‎3.11). A similar story to that of primary laws is shown with the apparent lessening in considering the baseline options and regulatory and non-regulatory alternatives.

Figure 3.11. Consideration of various options in RIA for subordinate regulations
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Types of impacts assessed

A well-designed RIA can assist in promoting policy coherence by making transparent the inherent trade-offs in regulatory proposals, identifying likely beneficiaries from distributional effects of regulation as well as those that will bear the costs, and how risk reduction in one area may create risks for other areas of government policy. A comprehensive RIA incorporates an assessment of the economic, social and environmental impacts. RIA can improve the evidence base in policy making, identify an appropriate response to an identified problem, and reduce the incidence of regulatory failure arising from regulating when there is no case for doing so, or failing to regulate when there is a clear need.

The Annex to the OECD recommendation relevantly provides that “When regulatory proposals would have significant impacts, ex ante assessment of costs, benefits and risks should be quantitative whenever possible. Regulatory costs include direct costs (administrative, financial and capital costs) as well as indirect costs (opportunity costs) whether borne by businesses, citizens or government. Ex ante assessments should, where relevant, provide qualitative descriptions of those impacts that are difficult or impossible to quantify, such as equity, fairness, and distributional effects” (OECD, 2012[1]).

The European Commission requires RIAs to assess a broad range of economic, environmental, and social impacts. In addition though it also requires policy makers to consider issues relating to compliance and enforcement, distributional impacts, as well as identifying a process by which the law’s goals will be assessed.

EU Member States generally assess a wide range of impacts relating to regulatory proposals on primary laws, albeit with differing frequency (Figure ‎3.12). Impacts to government, competition, small business, and the environment are most prevalent across EU Member States. A number of social impacts such as income inequality and poverty are always assessed by half of the EU Member States. Specific regional impacts and impacts on foreign jurisdictions are least likely to be assessed in RIA.

Figure 3.12. Types of impacts assessed in relation to primary laws
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Requirements to assess government and environmental impacts are less systematic for regulatory proposals relating to subordinate regulations than they are for primary laws. RIAs are most likely to require an assessment small business and competition impacts (Figure ‎3.13), and a number of Member States have instituted explicit SME tests. For instance, in Spain, policy makers are required to carry out an SME test in accordance with the practices of the European Commission by virtue of the operation of Article 2 of Royal Decree 931/2017, which regulates the content of RIAs in Spain. In Finland, policy makers are required to identify and assess regulatory proposals’ impacts on small and medium-sized businesses. In addition however, they also need to identify and assess how the proposal affects entrepreneurship by increasing barriers to entry, as well as the impact on ‘growth entrepreneurship’ where such businesses play a major role in the creation of new jobs and in increasing productivity (Ministry of Justice (Finland), 2008[12]).

Figure 3.13. Types of impacts assessed in relation to subordinate regulations
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Assessment of costs

A central part of sound impact assessment is to identify and assess the expected costs and benefits that regulatory proposals will have on various economic actors. The Annex to the OECD recommendation provides that member countries should “Assess economic, social and environmental impacts (where possible in quantitative and monetised terms), taking into account possible long term and spatial effects” and “Evaluate the impact on small to medium-sized enterprises and demonstrate how administrative and compliance costs are minimised” (OECD, 2012[1]).

Across EU Member States, costs are more likely to be systematically identified rather than systematically quantified in relation to primary laws, and generally Member States are required to quantify costs wherever possible (Figure ‎3.14).

Figure 3.14. Requirements to identify and quantify costs of primary laws
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

As part of considering various alternative options, EU Member States were asked to provide information about whether those options are ever quantified in practice. Around half of the EU Member States have requirements to quantify the costs of more than one policy option (Figure 3.14).

With regards to subordinate regulations, EU Member States are less likely to systematically identify and quantify the costs of regulatory proposals compared with those for primary laws.

Where costs are quantified they would therefore generally relate to the preferred regulatory option and not include other regulatory or non-regulatory alternatives. The main costs that are quantified in RIA relate to the costs to government, followed by costs to business, for regulatory proposals relating to primary laws (Figure 3.15). EU Member States do not currently systematically quantify the costs of regulatory proposals in relation to individuals or non-government organisations (NGOs).

Figure 3.15. Quantification of costs for primary laws
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg

The quantification of costs to government, businesses, and individuals are undertaken sporadically across EU Member States in relation to regulatory proposals on subordinate regulations (Figure 3.16). Overall, costs imposed on NGOs are unlikely to be quantified by EU Member States.

Figure 3.16. Quantification of costs for subordinate regulations
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Assessment of benefits

Identifying the benefits of regulation is an important aspect of any sound impact assessment. It helps to explain to affected businesses and society more generally about why the regulation is required, as well as who stands to gain from it. It is more difficult for decision makers to provide informed decisions where benefits are not clearly identified as the expected impacts have not been completely assessed. Moreover, at a time of general distrust of governments, it is important that regulatory decisions are, and are seen to be, based on actual evidence (OECD, 2018[13]). When well-founded, regulations do provide benefits to society and these benefits ought to be an important aspect of RIA where policy makers can become better informed about both their magnitude and distribution. It also presents an opportunity for affected businesses and citizens to provide input into the expected benefits of regulation so as to better gain trust in the regulation-making process, as well as improve behaviour and compliance rates, potentially lowering costs to both government and businesses (OECD, 2018[14]).

Despite its importance, benefits are only required to be systematically identified in two-thirds of EU Member States (Figure ‎3.17). Where benefits are required to be identified, they are generally qualitatively assessed rather than quantified.

Figure 3.17. Identification and assessment of benefits for primary laws
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Half of the EU Member States have a requirement to identify the benefits of new subordinate regulations, and less than two-thirds have any sort of quantification requirement (Figure ‎3.18).

Figure 3.18. Identification and assessment of benefits for subordinate regulations
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Where benefits are quantified, the results of the iREG survey and its extension to all EU Member States indicates that there are currently 11 EU Member States where policy makers are required to systematically quantify the benefits of more than one option.

The most common benefits to be quantified by EU Member States are those affecting government, followed by businesses and individuals (Figure 3.19). That said, the systematic assessment of these benefits is still not commonplace.

Figure 3.19. Quantification of benefits for primary laws
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

For regulatory proposals in relation to subordinate regulations, quantification of benefits is the exception rather than the rule. Less than one-third of EU Member States always quantify benefits to government, and quantification in other areas is less frequent than that (Figure ‎3.20).

Figure 3.20. Quantification of benefits for subordinate regulations
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Consideration of costs and benefits in RIA

As the previous sections have illustrated, EU Member States tend to only thoroughly identify and assess the costs and benefits associated with the preferred regulatory option. It is much less commonplace that alternative regulatory and non-regulatory options are quantified, and even more so in relation to subordinate regulations.

A related issue is whether policy makers are required to show that the benefits of regulation outweigh their cost. This is an important aspect of any good RIA system as it helps to illustrate to the regulated community the expected benefits of the regulation, as well as providing information on who are likely to gain from it. Moreover, providing information that regulations are net beneficial helps to ensure that regulations are expected to enhance societal welfare. This helps to generate support for regulations from regulated entities and the broader community. Where the benefits of regulation are not or cannot feasibly be assessed during regulatory development, it is important that these regulations are subject to ex post evaluation after an appropriate period to ensure that they have in fact improved welfare, as well as providing an opportunity more generally to ensure that they remain fit-for-purpose (OECD, Forthcoming[15]).

The vast majority of EU Member States do not have a requirement to demonstrate that the benefits of regulation justify their costs, for either the development of primary laws or subordinate regulations (Figure ‎3.21). Hungary, Spain, and the United Kingdom have such a requirement for all regulatory proposals; Poland has a requirement in relation to major laws; and Lithuania has a requirement in relation to some laws. The European Union has such a requirement in relation to major laws.

Figure 3.21. Requirements to demonstrate that benefits justify the costs
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

Effectiveness of RIA

In order to build support with the community, countries’ RIA systems should be effective and efficient; focussing on regulatory proposals where the largest net benefits to society would come from subjecting them to impact assessment. Additionally, periodically assessing the effectiveness and efficiency of RIA systems helps to maintain political support for the process, as well as from within the civil service.

Previous OECD research found that there was a strong need to establish effective evaluations of 15 EU Member States’ processes (OECD, 2009[11]). Assessments of the influence that RIA has had in successfully modifying regulatory proposals have been undertaken in more than one-third of EU Member States (Figure ‎3.22). In order to help maintain support for RIA — particularly from within the civil service — a number of EU Member States have produced good practice examples which act as additional guidance for policy makers when developing regulations.

Figure 3.22. Assessments of RIA modifying regulatory proposals and the existence of good regulatory practices
picture

Note: Data is based on 28 EU Member States.

Source: Indicators of Regulatory Policy and Governance Survey 2017, http://oe.cd/ireg.

General regulatory oversight of EU Member States was discussed in Chapter 1. Some Member States have conducted evaluations of their RIA systems in practice. For instance, the members of RegWatchEurope (see Chapter 1) provide regular reports on the implementation of their RIA systems, as well as the RIA system overall (Box ‎3.3).

Box 3.3. Selected reports on the effectiveness of RIA by members of RegWatchEurope

The members of RegWatchEurope are the RIA oversight bodies from seven EU Member States: the Czech Republic, Finland, Germany, the Netherlands, Norway, Sweden, and the United Kingdom (see chapter 1). RegWatchEurope was formed to foster exchange and co-ordinate activities on regulatory oversight.

The members generally provide annual compliance reporting statistics on the quality of RIAs, as well as more qualitative information around matters such as impacts assessed. Some reports have also assessed the extent to which feedback to ministries has been taken into account in the final regulatory proposal.

Recent reporting by the Finnish oversight body noted that a number of shortcomings in RIAs could have been overcome via better adherence to the impact assessment guidelines. It did note however that some individual RIAs had undertaken sound impact assessment.

Source: Finnish Council of Regulatory Impact Analysis, Annual Review 2016, https://vnk.fi/documents/10616/1851227/Finnish+Council+of+Regulatory+Impact+Analysis+Annual+Review+2016/c367fbba-c0f3-4b6d-abe5-81ead1227046 (accessed September 2018).

Separately to those countries, Belgium has released a series of reports on the implementation of RIA. Its most recent report noted that RIA is not yet well integrated into Belgian policy making, and occurs too late in the policy development process. Although the review identified that the quality of RIAs was not satisfactory, it did recognise that there were some good examples that were explicitly highlighted for other ministries to follow (Comité d'analyse d'impact, 2016[16]).

A review of the Romanian RIA system found that there were a number of challenges in the full adoption of an effective RIA system including a bias towards regulatory options, and a lack of skills across the administration to apply analytical RIA steps in practice. It was identified however that RIA has been promoted throughout the administration and that the importance of evidence-based decision making was recognised (World Bank, 2014[17]).

References

[16] Comité d'analyse d'impact (2016), Comité d’analyse d’impact, Rapport 2015, http://www.simplification.be/sites/default/files/documents/Analyse%20d%27impact/AIR_EvaluationRapport2015_CAI.pdf (accessed on  October 2018).

[8] Department of Business, Energy & Industrial Strategy (UK) (2018), Transposition guidance: how to implement European directives effectively, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/682752/eu-transposition-guidance.pdf.

[5] Department of the Taoiseach (Ireland) (2009), Revised RIA Guidelines - How to Conduct a Regulatory Impact Analysis, http://www.oecd.org/gov/regulatory-policy/46272075.pdf (accessed on  October 2018).

[2] European Commission (2005), Better Regulation for Growth and Jobs in the European Union, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52005DC0097&from=EN (accessed on  September 2018).

[6] Government of Lithuania (2008), Resolution No21 of 9 January 2004 Official Translation 22 July 2008, http://www.oecd.org/gov/regulatory-policy/46272075.pdf (accessed on  October 2018).

[9] Government of the Czech Republic (2016), Regulatory Impact Assessment (RIA) Guidelines, https://www.vlada.cz/assets/ppov/lrv/ria/metodiky/OZ_RIA_-novela_2016_uplne-zneni-FINAL.pdf (accessed on  October 2018).

[12] Ministry of Justice (Finland) (2008), Impact assessment in legislative drafting guidelines, http://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/76118/omju_2008_4.pdf?sequence=1&isAllowed=y (accessed on  October 2018).

[14] OECD (2018), OECD Regulatory Enforcement and Inspections Toolkit, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264303959-en.

[13] OECD (2018), OECD Regulatory Policy Outlook 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264303072-en.

[1] OECD (2012), Recommendation of the Council on Regulatory Policy and Governance, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264209022-en.

[11] OECD (2009), Better Regulation in Europe: An OECD Assessment of Regulatory Capacity in the 15 Original Member States of the EU, General perspectives from the first reviews, http://www.oecd.org/gov/regulatory-policy/44983092.pdf (accessed on  September 2018).

[3] OECD (2009), Regulatory Impact Analysis: A Tool for Policy Coherence, OECD Reviews of Regulatory Reform, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264067110-en.

[4] OECD (1995), OECD Recommendation on Improving the Quality of Government Regulation, OECD Publishing, Paris, http://acts.oecd.org/instruments/showinstrumentview.aspx?instrumentid=128&instrumentpid=124&lang=en&book=false.

[15] OECD (Forthcoming), OECD Best Practice Principles for Regulatory Policy: Reviewing the Stock of Regulation, OECD Publishing, Paris.

[7] Premier Ministre Secrétariat Général du Gouvernement, Conseil d'Etat (France) (2017), Guide de Légistique, https://www.legifrance.gouv.fr/Droit-francais/Guide-de-legistique2 (accessed on October 2018).

[10] Swedish Agency for Economic and Regional Growth (2018), EU law and other international law, https://tillvaxtverket.se/amnesomraden/forenkling/handledning-for-konsekvensutredning/rattsliga-forutsattningar/eu-ratt-och-annan-internationell-ratt.html (accessed on  October 2018).

[17] World Bank (2014), Advisory Services Agreement on Strengthening the Regulatory Impact Assessment Framework in Romania, http://sgg.gov.ro/new/politici-publice-si-programe/documente/evaluarea-preliminara-a-impactului/.

End of the section – Back to iLibrary publication page