3. A closer look at regulatory frameworks and practices
This chapter provides an in-depth analysis of Lithuanian regulatory frameworks and practices. It observes that poor implementation of better regulation practices may be a cause of legislative inflation. The chapter discusses the government’s efforts to strengthen regulatory impact assessments (RIAs), such as establishing higher impact legislation lists. It suggests deepening these efforts by creating a Regulatory Oversight Board to strengthen quality assurance, as well as clarifying and homogenising RIA processes and the legal framework. The chapter also provides an assessment of the nascent system for ex post evaluations of regulations, pointing to potential issues with evaluation clauses that are not associated with necessary funding or data-gathering requirements; and with the insufficient current capacities for co-ordination. Finally, it discusses the possibilities for regulatory rationalisation and codification as means to alleviate the burden of the growing legislative stock in light of best international practices.
This chapter examines the use of regulatory management tools and practices in Lithuania and how these contribute to the achievements of economic and social goals. It discusses the challenges posed by legislative inflation to Lithuania, due to a legalistic culture and lack of forward planning in policy making, and how this is impeding the development of a good regulatory environment. It then highlights the difficulties in integrating the key regulatory management tools - ex ante Regulatory impact assessment (RIA), stakeholder engagement and the nascent ex post review system - into the policy-making process, and ensuring that they contribute to evidence-informed approaches. A lack of robust quality assurance mechanisms is also highlighted. It sets out a series of concrete measures to tackle legal inflation in the short and longer term, including creating a forward planning system, reviewing the processes of the transposition of EU directives as well as some forward-looking initiatives. It also sets out measures to enhance the effectiveness of the regulatory management tools, including establishing an independent “Regulatory Oversight Board” for strengthening quality assurance processes.
Legislative inflation overall represents a key challenge in Lithuania, which the government has only been able to partly tackle to date by the existing regulatory management tools.
More legislative acts are passed in Lithuania than neighbouring countries
Legislative inflation is an important challenge facing Lithuania’s system of government. More legislative projects are discussed in the Lithuanian parliament than in neighbouring countries, with over 1 000 draft laws registered in the Parliament of 2016-2020 annually (5077 in total over 4 years) (Parliament of Lithuania, 2020[1]). By contrast, over the period of 2014-2016, 700 laws were discussed in Lithuania, while only 200 laws were discussed in Estonia and 417 in Latvia (Supreme Audit Institution, 2018[2]).
A number of factors have been identified as driving this phenomenon, including a legalistic culture, in which the performance of parliamentarians is perceived as being indicated by the quantity of legislation they are involved in producing (Lukošaitis, 2020[3]). Members of Parliament are particularly active in initiating legislation and legislative amendments, with nearly half (2 877 out of 5 077) of the legislative projects registered for the discussion in parliament initiated by parliamentarians (Parliament of Lithuania, 2020[1]). However, only 1 481 (51% of its total) of the legislative proposals initiated by the parliamentarians were adopted during this tenure, whereas the corresponding figure for government initiatives was 1534 (78% of its total) (Parliament of Lithuania, 2020[1]). An example of another smaller EU nation that has experienced the challenge of legislative inflation in the past is Belgium, which developed a number of specific approaches to address it (Box 3.1 below).
Another factor behind legislative inflation historically has been the excessive use of the legislative urgency procedures by parliamentarians. Under these urgency procedures, the legislative process is shortened, and the opportunity to thoroughly scrutinise legislation (including through RIA and stakeholder engagement) is significantly reduced (Lukošaitis, 2020[3]). However, Lithuania has recently faced a significant reduction in the use of the legislative urgency procedure, following several Constitutional Court rulings on the 16th of April 2019,1 complemented by additional rulings on 13 May and 7 June 2021.2 These decisions have influenced the reduction in the use of such practice; 52.5% and 36.5% of the laws were discussed using this procedure by the 2008-2012 and 2012-2016 Parliaments respectively whereas only 9.1% of laws were discussed using this procedure by the Parliament of 2016-2020 (Parliament of Lithuania, 2020[1]) (5% being a European average (Supreme Audit Institution, 2018[2])). The reduction is likely to continue following the most recent rulings.
The pursuit of Better Regulation in Belgium has been motivated by the need to address the challenge of legislative inflation. The federalisation of the Belgium state, from 1970 onwards, has resulted in a complicated regulatory environment, in need of simplification.
Additional causes of legislative inflation included a tendency for administrations to respond to any crisis or pressing issue with a new regulation, as well as preparing regulations at short notice under “urgency” procedures, which were often of poor quality. Many draft regulations were exempted from any form of ex ante RIA. RIA was often done too late and became an ex post justification for decisions which had already been reached. Political nominees themselves rather than civil servants often were involved in drafting laws, often causing implementation problems downstream and requiring revisions to the law in the worst cases. Institutional frameworks were weak and generally unable to challenge poorly implemented assessments. Transparency was also weak with often limited efforts to consult with stakeholders and little effort at publication. Finally, Belgium had a weak framework of ex post policy evaluation and stocktaking.
Political commitment to simplification has been expressed in successive government policy statements. Therefore, in 2004 the move towards legislative simplification and digital government started in both of the main federal constituents as well as by the central government in Brussels. In 2003, the government-wide initiative called “Kafka” was started which provided a platform for citizens, businesses and public servants across Belgium to propose ideas to cut the red-tape regulations. Moreover, the government recognised the importance of RIA in curbing excess legislation and the need to move beyond the traditional evaluations of administrative burden to more holistic considerations. A number of codification projects have been undertaken to create new codes or update existing codes. The biannual surveys of the Belgian Federal Planning Bureau have indicated that administrative burdens on businesses decreased from an estimated 3.5% of GDP in 2000 to 1.72 % of GDP in 2008.
Source: OECD (2010[4]), Better Regulation in Europe: Executive Summaries, pp. 56-72, www.oecd.org/gov/regulatory-policy/45079126.pdf.
Despite this useful legal remedy, legal inflation remains an important issue in Lithuania as each parliament keeps discussing and ratifying more legal acts than the preceding one (Parliament of Lithuania, 2020[1]).
Legislative inflation is impeding the development of a good regulatory environment
Such a situation clearly impairs the whole functioning of government in the country, leading to the uncontrolled proliferation of legal texts, and creating many opportunities for lack of compliance with laws and regulations, as well as excessive regulatory burdens. It is worth noting that Lithuania is also one of the European economies with the highest share of informal economy3. One of the determinants of the level of informality in an economy is the quality of the regulatory environment. For example, an OECD study of regulatory policy in Chile found there to be a number of regulatory barriers preventing formalisation of SMEs including taxation, business registration/licensing requirements, and compliance with labour laws. Evidence shows that reducing tax rates and simplifying the paperwork and tax compliance helps decrease the share of informal activities in the economy (OECD, 2016[5]).
The Supreme Audit Institution has highlighted the potential negative consequences of such a legalistic culture, stating that
“unclear, over-regulation, inconsistent legislation complicates law application and causes legal disputes, increases the administrative burden on the activities of public administrations, create preconditions for corruption, inefficient use of state budget funds and other adverse events occur” (Supreme Audit Institution, 2018[2]).
In this context, ex ante RIAs and ex post evaluations, together with administrative simplification and codification, can provide important mechanisms to address the challenges and consequences of legislative inflation. Codification in systems of civil law means consolidating all the amendments made over time to a set of related laws. It may also mean assembling an original legal act and all subsequent modifying acts into one new legal text (OECD, 2010[6]). It may also imply, at the same time, simplifying the existing stock of laws, as was attempted in France in the past.
In addition, adequate timeframes and sufficient importance bestowed upon conducting public consultations early in the policy process, can both slow down legislative processes and assure against a lack of attention to potential negative impacts. These tools provide assurance mechanisms against the negative impacts on the economy, businesses and society stemming from poorly designed regulations, as well as fostering public trust in the legislative system (OECD, 2020[7]).
A clear understanding of the problem of legal inflation amongst the political leadership offers a window for action
There is a clear understanding of the issue of legal proliferation among the Lithuanian political leadership as illustrated by the Programme of the 18th Government. Point 190 of this programme states that the new government will focus on “quality rather than the quantity” of new regulations, while improving existing regulations (Government of Lithuania, 2020[8]). It also recognises the importance of the principles of proportionality, legislative professionalism, inter-institutional co-operation and public consultations in assessing the anticipated impacts of the planned legislation. The government is also committed to pursuing a programme of legislative simplification and “debureaucratisation”. One of the concrete planned measures is the creation of the electoral code, through the consolidation of different pieces of legislation related to electoral procedures (Government of Lithuania, 2020[8]). This presents an opportunity to address this recurring challenge. The current report will provide a set of recommendations and good practices that could be mobilised to this effect, through fully maximising the potential of regulatory management tools.
Use of these tools is critical in supporting evidence-informed policy making
Regulatory management tools, including ex ante RIA and ex post evaluation of regulations, among others, play an important role in supporting evidence-informed policy making. RIA is a process of policy making whereby the likely impacts and consequences of various regulatory and non-regulatory options are considered and evaluated in order to choose a suitable one (OECD, 2018[9]). RIAs help to substantiate decisions not to intervene in the markets when the costs are too high relative to benefits or to communicate the arguments when such intervention is found necessary. It ensures that the impact on a particular societal or interest group that might be marginalised or absent from the mainstream public debate is assessed (OECD, 2020[10]).
Ex post evaluations are systematic reviews of the stock of significant regulation against clearly defined policy goals, including consideration of costs and benefits, to ensure that regulations remain up to date, cost-justified, effective and consistent, and deliver the intended policy objectives. Ex post evaluations complete the ‘regulatory cycle’ that begins with ex ante assessment of proposals and proceeds to implementation and administration (OECD, 2018[9]). Well-established practices and processes of ex post regulatory assessment can ensure the effectiveness, value-for-money, accountability and transparency of policy making. Such reviews enable the government to identify the effects of existing legislation and can help to improve the design and administration of new regulations (OECD, 2020[7]).
Gaps in the implementation of these tools to date have limited their effectiveness
Previous OECD studies have identified a number of challenges to successfully implementing the key regulatory management tools within the Lithuanian rule-making process, including an OECD review of Lithuania’s regulatory policy processes in 2015,4 as well as the OECD Regulatory Policy Outlook 2018 which map the efforts of members and accession counties to improve regulatory quality - see Box 3.2 for more information on these challenges (OECD, 2018[9]; OECD, 2015[11]).
Since 2015, Lithuania has undertaken several reforms to improve the quality of the legislative process and in particular, the modernisation of these tools to contribute to evidence-informed policy making (OECD, 2015[11]). The Office of the Government has started developing lists of Higher-Impact Legislation, which have to undergo thorough detailed RIAs. STRATA was given the role of scrutinising the quality of these assessments. In addition, the new policy framework for implementing ex post regulatory review into the legislative process was established with the Ministry of Justice as a co-ordinating institution.
Against the general trends in regulatory policy observed in OECD countries (see Box 3.2), there is no single formal government regulatory policy in Lithuania, though some elements are embedded in several strategic documents. Consultations in the development of regulations are anchored in the administration and interaction between stakeholders and the government sometimes takes place before a decision to regulate is made. Yet the implementation of consultation practices remains inconsistent across ministries and RIA documents are not well utilised in the consultation process. While the impacts are required to be assessed for any legislative acts, RIA remains a largely formal exercise to justify choices already made, rarely based on data or analysis of alternative options. There are some general requirements to conduct monitoring and ex post reviews of existing regulations, and the government has run pilots of more in-depth ‘fitness checks’. Quality control of regulatory management tools tends to occur late in the rule-making cycle, and mainly focuses on the procedural quality of RIA, rather than broader policy goals.
The OECD 2015 and 2018 Regulatory Policy Outlooks map countries’ efforts to improve regulatory quality based upon the IREG (Indicators for Regulatory Policy and Governance). The key findings are:
The vast majority of OECD and accession countries have a whole-of-government regulatory policy and have entrusted a body with promoting and co-ordinating regulatory quality across government.
OECD countries increasingly undertake stakeholder engagement with citizens and businesses on forthcoming laws and regulations. However, consultation outcomes could be better taken into account in regulatory design.
RIA has become an important step in the rule-making process of most countries. However, in many jurisdictions, RIA has become over-procedural and is not targeted to the most significant laws and regulations, either because there is no triage system or because regulatory proposals with significant impacts are exempted.
Most strikingly, the “lifecycle” of regulations remains largely incomplete. Countries are more adept at designing laws and regulations, than they are at the later stages of enforcing and reviewing them.
Some OECD countries have established regulatory oversight capacities and functions. However, in many countries, responsibility for regulatory oversight is split among several institutions.
Every institution initiating a legislative proposal is mandated to conduct ex ante RIA
According to OECD data, RIA is now required in almost all OECD countries for the development of both primary laws and subordinate regulations (OECD, 2018[9]). In 2003, Lithuania introduced Government Resolution No.276, which contained a requirement to conduct ex ante RIA when preparing new legislation. Later, the 2012 Law on the Legislative Framework set out the main principles of legislation and stages of a legislative process. Article 15 of the law mandates that institutions drafting legal acts are obliged to conduct ex ante RIA when new regulations are introduced or existing regulations are changed substantively. Moreover the law states that “a decision on the assessment of the effect of envisaged legal regulation shall be adopted by the drafter”, meaning that the scope of RIA and the processes whereby it is conducted are decided by the drafting institution itself (Parliament of Lithuania, 2012[12]). Additional measures have been taken to improve the methodology for conducting RIA and ensure its implementation. A list of the legislative milestones in the establishment of better regulation in Lithuania are set out in Box 3.3.
The 2012 law sets the following principles to guide the use of ex ante RIA:
The decision to conduct the impact assessment will be taken by the sponsoring institution, based on a principle of proportionality (i.e. depending on the extent of the expected impacts).
Positive and negative impacts on the relevant sector, as well as on individuals or groups of individuals, affected by the draft regulation should be assessed.
In developing the RIA, the ministry should consider the potential impacts of the draft regulation on: the economy, public finance; social environment; public administration; legal system; crime and corruption; environment; administrative burden; regional development; competition; other relevant areas. The results of the assessment have to be included in the explanatory note to a law or as a separate document (Parliament of Lithuania, 2012[12]). The note also lists the institutions and stakeholders to be consulted during the preparation of the draft regulation. However, in practice, only a couple of sentences on the estimated impacts have tended to be included in the explanatory note and the notes remain mostly descriptive – this was also a finding in the previous OECD review in (2015[11]).
It is usually the ministry’s department or unit in charge of the relevant policy area that is primarily responsible for the quality of the RIA. The ministry’s legal department conducts a legal quality check of the draft legislation, including the explanatory note and the conclusions on RIA results, before it is posted on the online legislative platform and other ministries and external stakeholders can provide comments. The Rules of Procedure of the Government determine when a certain act must be sent to a specific ministry. Otherwise, the drafting ministry decides which ministry to consult.
A proportionate ex ante RIA process has been in place since 2013, whereby a list of priority initiatives was developed, containing laws that were to be submitted to an in-depth assessment of their impacts. The list drew on proposals submitted by line ministries and proposals selected by the Office of the Government drawing to a large extent upon the Annual Legislative Programme (OECD, 2015[11]). As of 2020, this proportionality process has been further developed, as line ministries are mandated to develop and submit preliminary information on the expected impacts of their legislative projects to the Office of the Government, who then decides, together with advice from STRATA, which pieces of legislation shall be included in a (since renamed) list of higher-impact legislation. A set of criteria has been developed by the Office of the Government for deciding whether a proposed piece of legislation will be included in this list - see Table 3.2. Criteria for a law to be included in the list of higher impact legislation later in this chapter for more information about this.
The key steps in the development of Lithuania Better regulation policy are presented below:
26 February 2003: Government Resolution No. 276, introducing the requirement to make an impact assessment when preparing new legislation (recast in 2012 by Government Resolution No 1276).
12 February 2008: Govnment Resolution No 185 introduced the Better Regulation Progamme to act as a framework document to implement a broad range of better regulation initiatives.
23 February 2011: Government Resolution No. 213 “Regarding Adoption of Methodology for Identifying and Evaluating Administrative Burden for Citizens and Other Persons”.
11 January 2012: Government Resolution No. 4 “Regarding Adoption of Methodology for Identifying Administrative Burden for Businesses”.
18 September 2012: Law on the Legislative Framework (XI-2220), which set out the main principles of legislation, stages of a legislative process, as well as rights and obligations of state and municipality institutions and entities, and other persons, that take part in the legislative process.
8 November 2012: the Lithuanian Republic Law on Administrative Burden Reduction (XI-2386, defines the concept of administrative burden and creates the Better Regulation Supervisory Commission.
17 April 2013: Government Resolution No. 337 “Regarding Adoption of the Rules of Procedure of the Government of the Republic of Lithuania” sets out the main principles of carrying out legal impact assessment.
8 May 2013: Order of the Minister of Justice No 1R-142 “Regarding Adoption of Procedure for Conducting Monitoring of Legal Regulation”, which sets out the monitoring framework (recast on the 15th of May 2021).
23 December 2013: Order of the Minister of Justice No. 1R-298 “Regarding Adoption of Recommendations for Drafting Legal Acts”, which sets out technical requirements for legal drafting.
22 July 2014: The amendment to the Methodology for Determining the Administrative Burden of Business for establishing a new goal of the reduction of administrative burden applicable to each authority individually.
28 May 2019: Amendment of the Law on Legislative Framework Nr. XIII-2134 includes the notion of the ex post evaluation of regulations into the Law of Legislative Framework, and details under what conditions a legal act needs to undergo an ex post evaluation and how.
15 January 2020: Decision of the Government Meeting on the planned legislation quality improvement delegates to STRATA the duty to verify the quality of the RIAs of the legal acts included in the list of the higher impact legislation.
29 January 2020: Government Resolution 69 on the implementation of Law on Legislative Framework” gives the Ministry of Justice the role to co-ordinate of ex post regulatory assessment.
5 May 2021: Government Resolution No. 308 adopts the Methodology for ex post evaluation of existing legislation
Source: OECD (2015[11]), Regulatory Policy in Lithuania: Focusing on the Delivery Side, http://dx.doi.org/10.1787/9789264239340-en; and author’s elaborations; and Ministry of Justice (2021[13]).
Several other actors are involved in the co-ordination of RIAs
Several institutions are in charge of co-ordinating the implementation of RIAs:
The Office of the Government is the main institution in charge of co-ordinating the RIA process by:
formulating the overall policy, requirements and methodologies for RIA;
Some responsibilities have recently been transferred to STRATA:5
The Ministry of Economy and Innovation (MEI) co-ordinates the impact assessment on the administrative burden on business (Government of Lithuania, 2012[14]). In particular, it provides methodological support to other ministries in conducting these assessments, by providing feedback and helping draft conclusions. Since 2014, the ministry also ensures that the administrative burden approach of “One-In, One-Out” is followed. Moreover, the MEI is responsible for the better regulation policies for economic activity (administrative and compliance burden reduction, regulatory simplification and regulatory quality assurance)6 (Government of Lithuania, 1998[15]).
The Ministry of Interior is responsible for co-ordinating the assessment of regulations that affect administrative burdens on citizens, by providing methodological support to ministries (Government of Lithuania, 2011[16]).
The Competition Council provides methodological support and guidelines to the ministries for the impact assessments on competition. According to the Government Resolution on the ratification of the methodology for regulatory impact assessment (Chapter 2, point 10), the impact on competition shall be evaluated based on the guidelines on the “impact assessment of decisions on competition” prepared by the Competition Council (Government of Lithuania, 2003[17]). If a new legislation concerns competition, it is still the responsibility of the drafting institution to conduct an impact assessment, but Competition Council may be consulted. The Council has a right to intervene unilaterally and conduct the RIA themselves.
The Ministry of Justice reviews all draft laws and regulations, and provides a legal opinion (OECD, 2015[11]). It is also in charge of co-ordinating ex post assessment of regulations. The Ministry also co-ordinates and monitors the process of transposition, whereby EU directives are implemented by the Lithuanian administration (Government of Lithuania, 2004[18]). The Ministry is also responsible for the issue of “gold-plating”7 in the transposition process, and provides advice to ministries and has produced methodological guidance for preventing creation of unnecessary administrative burdens in the process of EU law implementation. This ministry used to have a department specialising in EU regulatory matters, however, it was recently consolidated with another department in the Ministry, thus reducing the scope for specific quality assessment related to the transposition of EU directives.
The Ministry of Foreign Affairs has the leading role with regard to negotiations of EU legislation, co-ordinating the inputs into the Impact Assessments at European level ex ante (Government of Lithuania, 2004[18]). It co-ordinates the positions of line ministries in their own field of policy responsibility for the negotiations on the production of EU regulation, but plays no role in the downstream aspects of creating domestic regulations as a result of EU processes.
However, despite Lithuania putting in place a comprehensive set of ex ante RIA requirements, it is apparent that most RIAs appear to be often conducted as a formality, with limited or no impact on the legislative process. Current quality assurance mechanisms are also lacking, with responsibilities dispersed across the government and proving ineffective at driving up RIA standards. Recently, there has been an attempt to strengthen and clarify the proportionality process, by developing a set of criteria to determine which proposed laws will form part of the List of Higher Impact Legislation and therefore undergo more in-depth RIA. The proposed laws on this list undergo a thorough quality control check by the Office of the Government, with advice from STRATA, but this only applies to a very small subset of legislations. This proportionality process does not currently have any legal status (e.g. through a Government Resolution), as have other parts of the RIA framework.
The nascent framework for ex post evaluation also presents significant shortcomings with regard to capacities in ministries to undertake this function. However, the overall context of legislative proliferation impedes the possibility of conducting quality RIAs and maximising the effectiveness of consultation. Until recently, the problem has also been compounded by an overall lack of planning of higher impact legislation, with insufficient time to conduct effective RIA or consultation for proposed laws. The recent establishment of a high-level Regulatory Impact Assessment schedule for high-impact laws represents a welcome improvement. A more proactive government-wide approach will be necessary precondition to create effective conditions for the update of regulatory management tools in Lithuania. The rest of this chapter will explain these challenges and proposed recommendations in detail.
RIA often remains a formal exercise
RIA appears to be viewed by officials as an administrative burden rather than a valuable tool for decision making, which may be a result of legal inflation (as well as making it more difficult to stop legal inflation). RIA does not appear to be undertaken at an early stage of the policy process, when there is a genuine interest in identifying the best available solution and there is an opportunity to consider alternatives to regulation (STRATA, 2020[19]) (STRATA, 2019[20]).
Assessments are rarely based on hard data or comparative analysis of alternative options. In most cases, they are prepared in parallel with the drafting of the legislative proposal and justify the choice already made, rather than offering a basis for evaluating alternative solutions to a public policy problem (STRATA, 2020[19]) (STRATA, 2019[20]). In addition, ex ante RIAs do not appear to be systematically carried out, even when ministries are mandated to do so. According to the Supreme Audit Institution, 10% of laws adopted by Parliament between 2014 and 2018 included an explanatory note in which it was indicated that no negative impact is foreseen with no further justification (Supreme Audit Institution, 2018[2]). In practice, there is a tendency to prepare RIA and a draft legal act at the same time, and to choose the regulatory measure, instead of a non-regulatory alternative.
Members of Parliament, have the right to initiate new legislation and, as mentioned previously, are particularly active in initiating legislation and legislative amendments, with nearly half of the legislative projects registered for the discussion in parliament initiated by parliamentarians. The adoption of amendments during the last stage of voting can reduce the effectiveness of the process for making high-quality legislation, initiated by either the executive or the parliament. Whilst draft laws initiated by Members of Parliament are subject to similar procedures as those initiated by the executive, with each legislative proposal needing to be accompanied by the explanatory note containing a discussion on impacts, there is no requirement for a RIA to be produced.
A number of short and longer-term reforms involving the executive and legislative branch should be considered to increase the use of RIAs in policy making
In order to increase the quality and number of RIAs conducted, Lithuania should consider a number of reforms, both within the executive branch, and also reaching out to the legislative branch.
In terms of the legislative branch, the Office of the Government and STRATA could seek to raise awareness with members of parliament in order to increase their understanding of the RIA process and its value for evidence-informed decision making, as well as the need for better planning of legal activity. Specifically, the Office of the Government could invite the parliament to express a commitment, through a memorandum of understanding (MoU), with the desired outcome of increasing the number of RIAs prepared for laws initiated by Members of Parliament. This MoU could also be utilised to generate a joint understanding between the executive and the parliament of the planned calendar for the work of parliament.
Beyond such short-term fixes, effectively tackling the problem of legislative inflation may require Lithuania to take a broader, forward-looking perspective to its processes of law making. The Lithuanian Government should consider setting up a Strategic Task Force on the future of the Lithuanian legal framework, aimed at reflecting on the long term costs to the economy of current legal inflation and the implications of the current legal culture in Lithuania (e.g. with laws that are very detail-oriented). This Task Force could be comprised of prominent members of the government, parliament and the legal profession (including legal academics) but should also be open to foreign expertise from countries that have faced similar challenges. The goal would be to raise awareness, which may not always exist within the legal profession in Lithuania, and to ensure that any reforms receive a large degree of consensus and are likely to remain durable over time. The Task Force should discuss what are the implications of such a detail-oriented approach in the drafting of laws, and the current imbalance in the law-making processes, which in the longer term could have adverse effects for the country.
Ensuring the quality of regulation is important as there may be unintended consequences and, ultimately, negative impacts for citizens, businesses and society as a whole that result from badly designed interventions (OECD, 2020[10]). Hence, it is crucial to ensure that the legislative processes are carried out correctly and using reliable evidence. In addition, OECD best practice suggests that effective regulatory oversight is a crucial precondition for a successful RIA process, which includes putting in place mechanisms for ensuring the quality control of RIAs.
As of 2020, some quality control mechanisms exist in Lithuania. The Office of the Government, advised by STRATA, oversees the quality of RIAs accomplished for the draft laws that are included in the Higher-Impact Legislature list, and all legislative proposals undergo a thorough legal quality check by the Ministry of Justice. However, internal guidelines and structures, setting out in detail how policy teams should integrate RIA into their processes, are missing inside ministries and RIA is often perceived as a formal exercise intended to justify choices already made (STRATA, 2020[19]). In addition, insufficient attention is focused downstream on the domestic regulatory implications of implementing EU directives.
Certain ministries are equipped with a unit that is responsible for making sure that an explanatory note, containing the RIA results, accompanies each legislative proposal. However, these units offer little support with regards to quality assurance or capacity building for RIA as they do not have such responsibility. Indeed, these units are generally composed of staff members with legal backgrounds, who do not possess the necessary analytical capacities to offer analytical insights into the quality of RIAs.
Fragmented responsibilities for quality assurance
Several institutions are in charge of developing government-wide guidelines for RIA. The methodology on impact assessment of draft legislative acts, approved by the government in 2012, serves as the main guide for preparing RIAs (Government of Lithuania, 2003[17]). This methodological guide recommends the use of Cost-Benefit Analysis (CBA) or Cost-Effectiveness Analysis (CEA), when possible. The Office of the Government has developed specific guidance to help ministries apply social CBA and CEA (Office of Government, 2011[21]); it can be found on the website of the Ministry of Finance. The Ministry of Economy and Innovation and the Ministry of Interior have produced guidelines on how to calculate administrative burden respectively on businesses and citizens, based on the Standard-Cost Model. STRATA has also recently published its own guidelines for conducting ex ante RIAs (STRATA, 2020[19]).
The existence of multiple methodological guidelines may create confusion for ministries over their legal status, and when it is appropriate to utilise them in policy making. These documents can be found on different institutional webpages and the extent of their use is at the discretion of public servants drafting RIA. Lithuania would benefit from consolidating these documents into a set of methodological guidance with a clear status as part of an integrated strategy, which would serve as a main reference for ministries while developing RIAs.
To improve the quality of RIAs, there is a need to invest upstream on skills development and downstream through establishing a Regulatory Oversight Board
There is therefore a need to work on two fronts to improve the quality of RIAs:
The first front is an upstream investment in skills and capacities in the ministries, so that they are effectively able to deliver the analysis. As detailed in Chapter 1, the Lithuanian government could benefit from the creation of an analytical stream within the civil service to attract highly qualified analytical personnel. This should be accompanied by the government-wide strategy on analytical skills. Such a strategy could include the provision of scholarship programmes as a short-term measure and the development of academic programmes in Lithuania for quantitative policy analysis and economics for the long-term supply of skills.
The second front is an investment downstream to strengthen the quality assurance process, through the establishment of a proper “Regulatory Oversight Board” which could be envisaged in the medium term after an experimental phase, and a strengthened oversight of processes by the Office of the Government. The oversight board could be tasked with carrying out quality control of ex ante RIA documents for higher impact RIAs, as well as providing publically available advice on the quality of the RIAs that would have to be shared with Parliament. This advice would confirm whether due process and the core steps of the RIA process have been followed. This advice could be placed on a website for the Regulatory Oversight Board as is common for similar oversight bodies in a number of OECD countries (OECD, 2018[22]).
The Office of the Government should send the oversight board’s official advice to the government in a public manner, before a law is transmitted to Parliament, as is the case in other jurisdictions e.g. with the advice of the Dutch Advisory Board on Regulatory Burden. The oversight body should also provide regular assessments of the overall quality of the RIA process across government and publish their findings in publically available reports, as suggested by the 2012 Recommendation of the OECD Council (OECD, 2012[23])8. The oversight body does not necessarily imply creating a new or large agency, but it does require establishing an independent voice, through setting up a board of external, prominent personalities to provide the opinions on the RIAs.
STRATA would remain available to provide the secretariat for the oversight body, and to provide methodological assistance to ministries on how to conduct ex ante RIA analysis. Box 3.4 provides detailed information on three European regulatory oversight bodies, each of which operates with a degree of operational independence from the executive.
Germany - National Regulatory Control Council - NKR
The Nationaler Normenkontrollrat (NKR) established by law in 2006 scrutinises draft impact assessments accompanying primary and secondary law proposals, with regard to underlying assumptions on compliance costs. It normally provides initial advice at pre-consultation stage during the inter-service consultation with all ministries. The final opinion is released after the post-consultation (final) stage, when the proposal is ready to be tabled at and adopted by Cabinet (council of ministers). In between these stages, there is a working level exchange between NKR Secretariat and lead ministries. The parliament may request the NKR to provide its opinion on the quality of an impact assessment; for a parliamentary proposal, even if the initiation of laws by the MPs is rare. The NKR has received other functions including the “one in one out” approach, Cutting Red Tape efforts, and the scrutiny of evaluation clauses in the legal acts (obligatory for all regulations that induce compliance costs of over 1 million euros). The NKR also publishes an annual report that provides a comprehensive view of Better Regulation policy. The members of the NKR are nominated by the Federal Government and appointed by the Federal President. The Chancellor of the Federal Government designates the Chair of the NKR. The NKR committee is supported by a 15 civil servants’ secretariat and a 1.5 million euros budget.
Czech Republic - Regulatory Impact Assessment Board
The Regulatory Impact Assessment Board (RIAB) was established in 2011 to review the quality of submitted RIAs accompanying draft primary and secondary legislation. The RIAB is one of the working groups of the Government Legislative Council (GLC) and its activities are supported by the Office of the Government. The RIAB may also undertake consultation with ministries concerning aspects of RIA during the preparatory and drafting phase; and issue non-binding opinions. It is sometimes contacted by external stakeholders to receive more detailed information on the impacts of the proposed legislation. Based on the evaluation of “Overviews of Impacts” (provided to the drafting authorities in a template), the RIAB issues opinions as to which planned pieces of draft legislation should undergo a regular RIA process. The suggestions of the RIAB are to be approved later as a binding obligation in the Plan of Legislative Work of the Government, respectively in the Plan of Preparation of Decrees and officially published. The opinions are made publicly available (on the website http://ria.vlada.cz). The RIAB is composed of 15 external experts (economists, lawyers, business representatives, etc) who meet once every 3 weeks. The RIAB The RIAB members report potential conflict of interests at the RIAB meetings.
Netherlands - Dutch Advisory Board on Regulatory Burden – ATR
The ATR is an independent advisory body established by law with budgetary autonomy. The board consists of three members, who are supported by a director, 8 senior policy advisors and 2 supporting staff. The body decides what it advises on (within its mandate) as well as on the content of these opinions. Both government and Parliament can approach the body for advice. The ATR advises ministries mainly in the early stages of the policy process i.e. problem identification. It does not operate according to a proportionality principle and assesses the regulatory consequences on business of all primary legislative proposals and administrative measures, and also scrutinises ministerial decrees with substantial compliance costs. Recently, the ATR has taken on a new role of scrutinising the new SME Test.
Source: OECD (2018[22]), Case Studies of RegWatchEurope Regulatory Oversight Bodies and European Union Regulatory Scrutiny Board, http://www.oecd.org/gov/regulatory-policy/regulatory-oversight-bodies-2018.htm.
A small number of RIA for “higher impact legislation” undergo thorough quality control
Since 2018, the Office of the Government has drawn up a semi-annual list of draft legislation considered of higher impact (the list replaced the Priority Legislation List which was in existence between 2013 and 2016). A set of criteria has been developed by the Office of the Government for deciding whether a proposed piece of legislation will be included in this List of High Impact Legislation - see Table 3.2 below. This list is discussed and approved during an inter-ministerial meeting.
These selection criteria however have not been formalised in any binding legal act and remain of an indicative nature. The first steps in systemising the triage process have been taken by the previous 17th Government and the implementation plan of the new government programme mandates the Office of the Government to develop a new set of criteria for the selection of higher-impact legislation as part of the improvements for the RIA methodology and processes (measure 8.1.2.) (Government of Lithuania, 2021[24]). In addition, the previous government has started to publish these bi-annual lists on its website. The current 18th government has amended this practice by preparing one list of higher impact legislation for the period until the end of its term in office (2021-2023). This list includes 35 legislative projects and sets the timeline for impact assessment, public consultation, submission to the government and parliamentary deliberation.
As of 2020, line ministries must develop and submit preliminary information on the expected impacts of their legislative projects to the Office of the Government who then decides, together with advice from STRATA, which pieces of legislation shall be included in the list of higher-impact legislation. 13 legislative projects were included in the list for the spring session of 2019, 12 for the autumn session of 2019, 6 for the spring session of 2020 and 7 for the autumn session of 2020, for 2021 – 2023 there are 35 planned legislative projects of higher-impact.
Ministries can choose to consult STRATA on the quality of RIAs for higher-impact laws earlier on in the legislative process. Once the RIA is completed, it has to be sent to STRATA, which then scrutinises the quality of the RIA based on a list of predetermined questions, and provides advice to the Government, on whether an evidence-informed decision can be made based on the RIA. Between March and December 2020, STRATA has provided its opinion on 12 such RIAs for 8 draft laws (STRATA, 2020[19]). Based on the decision of the government, the drafting institution prepares either a draft law or a non-regulatory measure. Box 3.5 below provides examples from OECD members of how they have applied different forms of triage processes to introduce proportionality into their RIA systems.
The introduction of a set of criteria for determining whether a proposed piece of legislation will be included in the high impact list, and the strengthening of quality control for those laws has enhanced the proportionality process for RIA. This appears to follow the best practice of a number of OECD countries. However, this list only presently applies to a very small subset of legislations. Given the number of bills in the Government Legislative Plan, it will be important to make sure that proposed legislation with potentially high impacts does not avoid scrutiny. Also, this proportionality process does not currently have any formalised legal status (e.g. through a Government Resolution), as have other parts of the RIA framework. Therefore, it may be beneficial to formalise the proportionality process and the role of STRATA in quality control, potentially as part of the proposed Government Resolution on Better Resolution (see the sub section “In order to facilitate structural uptake of these findings, Lithuania could consider developing a coherent framework for better regulation”). As this is a recent process, it is not possible yet to draw a conclusion as to how effectively it has been implemented.
In the United Kingdom, a de minimis rule has been introduced to ensure proportionality and give departments greater flexibility to determine the appropriate level of analysis to demonstrate the rationale for a regulation. The Better Regulation Executive and the regulatory oversight body, the Regulatory Policy Committee (RPC), have jointly produced guidance stipulating that only measures with significant regulatory impacts (greater than GBP -/+ 5million threshold) are expected to have full RIAs and be submitted to the RPC for scrutiny. However, measures that fall below the de minimis threshold may still be expected to produce a full RIA if they are estimated to have significant distributional impacts; disproportionate burdens on small businesses; significant wider social, environmental, financial or economic impacts; or significant novel or contentious elements. All other regulatory measures are still expected to produce a proportionate level of analysis to support stakeholder and parliamentary scrutiny of the proposal. In addition, departmental Chief Analysts are responsible for ensuring that the analysis used for measures which are under the GBP -/+ 5 million threshold is sufficiently robust.
The South Korea test requires quantitative RIA to be undertaken if it affects more than 1 million people and/or 10 million Won, there is a clear restriction on market competition or a clear departure from international standards.
Belgium uses the approach whereby some topics are evaluated only based on the values of some indicators whereas the other impacts have to be evaluated in depth. In total, there are 21 impacts to be assessed but only 4 impacts have to be evaluated using the detailed approach (administrative burden, gender and policy coherence for development).
In Australia, each submission to the parliament has to be accompanied by a RIA. For the subordinate legislation, a preliminary assessment is conducted to determine if a RIA is required for both primary and subordinate legislation. A RIA is also mandatory for any non-Cabinet decision made by any Australian Government entity if that decision is likely to have a measurable impact on businesses, community organisations, individuals or any combination of them.
Source: OECD (2015[25]), OECD Regulatory Policy Outlook 2015 , http://dx.doi.org/10.1787/9789264238770-en; OECD (2020[26]), Review of International Regulatory Co-operation of the United Kingdom, https://doi.org/10.1787/09be52f0-en.
There are few internal established processes and analytical support inside ministries for the staff that are drafting the RIA
At the level of line ministries, however, there are few internal established processes to support the implementation of the RIA methodologies inside the ministries. As previously mentioned, the strategic planning units within most ministries typically offer little support with regards to providing analytical support or capacity building for RIA. They are generally composed of staff members with legal backgrounds, who do not possess the necessary analytical capacities to offer insights into the quality of RIAs.
The responsibility to conduct RIA is highly dispersed among the staff of ministries, many of whom may have general legal or institutional skills, but not the analytical skills that would be necessary for conducting RIAs. Indeed, the Supreme Audit Institution has estimated that 2000 ministry employees conduct RIAs, but that many of them lack adequate training. An audit from the Supreme Audit Institution reported that 10% of the staff responsible for drafting legislative proposals during the audited period (2014-2017) had had training on RIA. The data was collected through the ministries’ surveys (Supreme Audit Institution, 2018[2]). STRATA is running a course of seminars on RIA, and is aiming at training 150 civil servants with a plan of extending these numbers. In addition, ministries can outsource advice and analysis for RIA, to make up for a lack of analytical skills or timing constraints. These services are usually acquired through public procurement procedures (OECD, 2015[11]).
On the other hand, ministerial dependencies/ agencies, which often serve as the analytical arms of ministries (e.g. Enterprise Lithuania), do not have a formal role in the RIA process. Not only do the ministries drafting the legislation work on RIAs in silos, but also the staff that is conducting the analysis do not have much support inside their own ministry.
The process for preparing RIAs and new regulatory proposals would benefit from some clarification within ministries
Lithuania could consider consolidating methodological guidelines as mentioned above, complemented by trainings and tools, in order to support the RIA process at the line ministry level. Firstly, the process for preparing RIAs and new regulatory proposals within line ministries would benefit from some clarification. This advice also echoes that of consolidating capacities for analysis in each ministry, so that policy units in charge of preparing RIAs can easily identify internal resources for quality assurance of RIAs. Furthermore, ministries could ensure that there are staff members who can provide methodological support for RIAs in-house. In doing so, ministries could look to the example of the process elaborated by the Ministry of Social Security and Labour in recent years, by which line policy units can require support from the “Strategic Decisions Support and International Cooperation Group” to conduct an assessment of the monetary impacts of legislative projects. Box 3.6 below provides some examples of OECD governments that have established networks of officials within line ministries to provide advice on RIA and other better regulation processes.
Regular programmes of RIA training for new and existing civil servants should be established together with digital tools supporting the process
Other than methodological support, developing staff capacities for analysis is another way that countries can promote the quality of RIAs, which goes back to the issue of analytical capacity addressed above. In Lithuania, there have been good examples of capacity-building activities e.g. STRATA has run workshops to train civil servants on how to conduct RIAs. However, this training should become more structured and systematic to ensure that RIA capacity is firmly embedded within ministries over time. Therefore, Lithuania could consider establishing a regular programme of RIA training for new and existing civil servants. In addition, more specialised RIA training could be focused upon the units responsible for internal RIA co-ordination, to ensure that each ministry has a hub of expertise that can provide advice to other officials.
In addition to training, Lithuania could also rely on digital tools to promote the quality of RIA. More recently, a number of countries have developed software-based tools that can be used to assist in RIA development to estimate various kinds of costs.
United Kingdom
In the United Kingdom, government departments with a responsibility for producing regulations have a Better Regulation Unit (BRU) consisting of a team of civil servants that oversee the department’s regulatory management processes and advises on the compliance with the Better Regulation requirements. It is at the discretion of each department to determine the scope of the BRU’s role, its resourcing (i.e. staff numbers, composition of policy officials and analysts, and allocation of time on this agenda versus others) and position within the departmental structure. However, their functions generally include promoting the use and application of better regulation principles in policy making, advising policy teams on how to develop a RIA (or post-implementation review) including queries on methodology and analysis, and advising policy teams on the appropriate schedule to submit a RIA to the oversight body (the Regulatory Policy Committee) for scrutiny.
Mexico
In Mexico, all regulatory impact assessments are reviewed by the National Regulatory Improvement Commission (CONAMER). Its opinion on RIA has to be received before any regulation is adopted. If RIA is deemed non-satisfactory, CONAMER may request the drafting ministry to develop or amend it or it can hire an external consultant to conduct an independent RIA.
CONAMER interacts with an important network of officials that co-ordinate RIA processes within the line ministries. Each head of public administration entity appoints an official that acts as a liaison between the institution and CONAMER. They are charged with submitting to CONAMER the better regulation programme related to the regulations and procedures of their institutions. They are also responsible for the co-ordination of better regulation processes within their institutions. In order to enhance the political weight of better regulation, these officials are usually selected from vice-ministers or chief administrative officers.
Source: OECD (2019[27]), Implementing Regulatory Impact Analysis in the Central Government of Peru: Case Studies, https://doi.org/10.1787/9789264305786-en; OECD (2020[26]), Review of International Regulatory Co-operation of the United Kingdom, https://dx.doi.org/10.1787/09be52f0-en.
These tools are presented in Box 3.7 below. These calculators are, in some countries, accessible also to stakeholders, which can calculate the costs of current, drafted or potential regulations or their changes. One such calculator has also been developed by STRATA in Lithuania.9 This calculator should be widely publicised to ministries to spread awareness, as well as firmly ‘signposted’ it within RIA training programmes and within accompanying methodological guidance.
To increase the quality of RIA and lessen the burden of preparing RIA statements in South Korea, e-RIA was launched in 2015. It is linked to the national statistical database and provides the public officials who prepare RIAs the possibility to automatically obtain the necessary data for cost-benefit analysis, and a sufficient amount of descriptions and examples for all fields. As all fields are mandatory, e-RIA also prevents users (regulators) from omitting important data and information. RIAs are produced automatically upon completion of all fields.
In 2010, Mexico has introduced a software calculator, the Regulatory Impact Assessment Calculator (RIAC) that helps to determine what type of RIA should be conducted for a specific planned regulation. An evaluator needs to answer 10 questions and insert the relevant data, which then determines if the associated potential costs of the planned regulation fall below the threshold, which allows regulations to be exempted from RIA. In 2012, the analysis of competition and risk was integrated into the RIAC.
Australian government has developed the Regulatory Burden Measure (RBM), the software tool that allows the public servant and the public to calculate the potential compliance costs of planned regulations (accessible on https://rbm.obpr.gov.au/home.aspx). The calculator gives the evaluators a template and guidelines to logically evaluate the potential compliance costs. It also safe time of report generation and repetitive research/calculations. The user of this calculator needs to:
Source: OECD (2017[28]), Regulatory Policy in Korea: Towards Better Regulation, http://dx.doi.org/10.1787/9789264274600-en, OECD (2015[25]), OECD Regulatory Policy Outlook 2015, http://dx.doi.org/10.1787/9789264238770-en.
STRATA should nurture a RIA “community of practice” to facilitate exchanges of best practices across ministries
In addition, STRATA should nurture a RIA “community of practice” through the facilitation of exchanges of best practices across ministries. This should be co-ordinated by the Office of the Government to facilitate knowledge sharing, a shared understanding of the processes, and greater access to analytical content with the support of STRATA through seminars and other networking opportunities. It should build upon existing networking opportunities that line ministries and agencies may already have. International examples of such communities of practice are provided in Box 3.8.
Canada’s Community of Federal Regulators (CFR) is a partnership of Canadian regulatory organisations at the federal level that aims to facilitate professional development, collaboration and advancement of the regulatory field. The CFR focuses its events, activities and resources to meet three strategic objectives:
Talent Management – initiatives to strengthen the regulatory profession across the system;
Collaboration – events to connect organisations to foster collaboration and sharing of regulatory expertise;
Experimentation – pursuing prototypes, projects and other activities to increase community understanding of innovative regulatory concepts and enabling their application.
Key activities include an annual two-day Regulatory Conference, annual one-day Law Enforcement Symposium, Regulatory Professional Development Program, Regulatory Speaker Series, Regulatory Excellence Awards, Prototyping Workshops and Communities of Practice/Working Groups on specific regulatory issues.
The community serves approximately 40 000 regulatory professionals who support Canada’s regulatory lifecycle including policy analysts, programme officers, compliance and enforcement officers, performance evaluators, risk assessors, legal counsellors and cost-benefit analysts. The community is governed by a Deputy Minister Champion, two Assistant Deputy Minister Co-Champions and representatives from each of the departments and agencies providing financial support to the community, responsible for setting direction and areas of focus for the community in conjunction with the CFR Office.
New Zealand’s Government Regulatory Practice Initiative (G-REG) is a network of central and local government regulatory agencies established to lead and contribute to regulatory practice initiatives. G-REG focuses on developing people capability, organisational capability, and building a professional community of regulators.
G-REG’s primary activity to date has been the development and delivery of a qualifications framework. Having a common qualification in the public sector is intended to make it easier for regulatory agencies to work together, when their people have common ways of operating and transferable skills and qualifications. G-REG’s primary activity to date has been the development and delivery of a qualifications framework. Having a common qualification in the public sector is intended to make it easier for regulatory agencies to work together, when their people have common ways of operating and transferable skills and qualifications.
G-REG is working to unify and professionalise the regulators of New Zealand, by bringing it together through a series of workshops, annual conferences, articles in industry journals. It also has worked to increase its intellectual credibility by establishing a Chair in Regulatory Practice at Victoria University of Wellington, whose research programme incorporates advances in regulatory practice outside New Zealand, focusing on innovative regulators, regulatory instruments and processes. This enables international regulatory best practice and knowledge to be disseminated to G-REG and the wider regulatory community (through blogs, seminars and guest lectures).
Source: OECD (2020[26]), Review of International Regulatory Co-operation of the United Kingdom, https://doi.org/10.1787/09be52f0-en.
Consultation procedures are well established and have been strengthened but remain rather formal
OECD best practices on RIA recommend that stakeholder engagement be incorporated systematically in the RIA process to give an opportunity for all stakeholders to participate in the regulatory process (OECD, 2020[10]). Indeed, consultation and user engagement can give important information on the feasibility of proposals, on the alternatives considered, and on the degree to which affected parties are likely to comply with the proposed regulation. Furthermore, the assumptions underlying RIAs can be improved if they are tested after the carrying out of the RIA through public disclosure and consultation.
In Lithuania, according to the 2012 Law on the Legislative Framework, the ministry responsible for drafting the legislation decides whether public consultation is necessary given the scope of the proposal, as well as its extent and any methods employed. Moreover, the methodology on impact assessment stipulates that public consultation can be arranged at different stages of the assessment: when defining the problem, formulating the goals or regulation alternatives, assessing the impact of alternatives, etc. Thus, it can be said that the consultation process for RIA is formally established within the policy-making processes in Lithuania. (Government of Lithuania, 2003[17]; OECD, 2015[11]).
In practice, ministries tend to consult with stakeholders through formal and informal working groups and networks, sometimes in the early stages of preparation of the draft legislation. For example, a working group consisting of the representatives of different universities was gathered by the Ministry of Education, Science and Sports to draft and evaluate the reform of the restructuring of universities. However, stakeholders do not appear to be systemically involved in or alerted to upcoming consultations. Lithuania is not unusual in this respect as OECD Regulatory Policy Outlook data indicates that most consultation efforts in OECD countries continue to focus on later stages of the rule-making process, i.e. when a preferred solution has been identified and/or a draft regulation has been prepared (see Figure 3.4 below).
Consultation can also take place online through the “E.Citizen” website (accessible on https://epilietis.lrv.lt). This is considered to be good practice according to the OECD Best Practice Principles for Regulatory Impact Assessment (OECD, 2020[10]). In this case, citizens as well as stakeholders and academics can then submit their comments and proposals for the legislation and the RIA in question. However, the use of the website by ministries is voluntary and stakeholders do not automatically receive notifications about the ongoing consultations. In addition, the draft legislation and the explanatory notes, that include RIA results, have to be published on the Legal Acts Information System (TAIS) (Parliament of Lithuania, 2012[12]). However, according to the audit results conducted by the SAI, the share of legislative proposals that receive comments from the public is low (Supreme Audit Institution, 2018[2]). The number was 8.4% of all legislative proposals in 2014, 9.7% in 2015 and 5% in 2016.
It appears that feedback from stakeholders is often sought more to confirm a preferred option, rather than to identify and evaluate different options. This is due, in part, to the explanatory note accompanying draft legislation providing too little information to engage stakeholders in a constructive discussion on the impacts of the proposed legislation. Accordingly, it can be said that public or stakeholder feedback rarely alter the substantive content of the RIA.
The information relative to these consultations needs to be presented to the legislative body and published on the TAIS. However, some of the accompanying documents are often missing and only the RIA conclusion, rather than the whole analysis, is presented (Supreme Audit Institution, 2018[2]). This represents an important gap with the international best practice, which is to publish the RIA also with the results of the consultation exercise. Ministries have an obligation to assess the proposals collected during the consultation exercise and incorporate them into the legislative project or provide arguments why the submitted proposals were not taken into account. The Office of the Government has also developed methodological guidance for consultation, although its use by ministries is voluntary (Office of the Government, 2018[29]).
Strengthening public consultation through effective engagement with stakeholders should be a priority
Lithuania should consider making consultations when drafting legislative proposals more systematic, effective and less formal, with a view to engaging in constructive dialogues, using substantive RIA documents, about policy options with stakeholders (see Box 3.9).
In Canada, a variety of methods to engage with stakeholders is used that combines the direct targeting of identified stakeholders as well as the open-access consultations. Some examples of these methods are emails, phone calls, third-party-facilitated sessions, roundtable meetings and online consultations. Each RIA is published together with the legislative proposal on the Canadian public consultations’ Gazette. Stakeholders can then express any issues they have with the conducted evaluation of impacts or submit their-own alternative evaluation. The drafting ministry is then obliged to respond to these comments and explain the rationale behind their decision of the amendments to their initial RIA or the non-incorporation of suggestions.
European Commission has developed its two-stage public consultation on RIA. Since 2015, before the final draft of RIA, the Inception Impact Assessment is prepared which contains different options to solve the policy issue as well as the impacts associated with each alternative. This initial document is then discussed and consulted on for 4 weeks. At the end of this process, an official impact assessment is proposed. Using this impact assessment, the commission conducts a 12-week consultation process to draft the finalised legislative proposal. This proposal together with RIA is published for the public comments. Moreover, the draft subordinate legislation undergoes a 4-week consultation process. Once the legislation is ratified, all the impact assessment and the opinions of Regulatory Scrutiny Board (RSB) are published online.
Source: OECD (2016[30]), OECD Investment Policy Reviews: Philippines 2016, https://dx.doi.org/10.1787/9789264254510-en; OECD (2017[28]), Regulatory Policy in Korea: Towards Better Regulation, https://doi.org/10.1787/9789264274600-en; European Commission (2019[31]), Regulatory Scrutiny board, https://ec.europa.eu/info/law/law-making-process/regulatory-scrutiny-board_en (accessed on 11 May 2021); OECD (2017[32]), OECD Best Practice Principles on Stakeholder Engagement in Regulatory Policy: Draft for Public Consultation, https://www.oecd.org/gov/regulatory-policy/public-consultation-best-practice-principles-on-stakeholder-engagement.htm (accessed on 11 May 2021).
In Lithuania consultation could be made mandatory for all legal acts and RIAs, and the use of online consultation favoured as a default option (all the while maintaining a principle of proportionality). It could become mandatory to make consultations open for a minimum period of 3 to 6 weeks for all legal acts and RIA – indeed a majority of OECD countries systematically make use of such minimum periods with a view to ensuring stakeholders have sufficient time to provide meaningful input in the rule-making process (see Figure 3.5 below). The Office of the Government should also be charged with validating the consultation process, as a requirement for law proposals to be tabled in the Council of Ministers, to examine whether the minimum consultation standards have been correctly followed, and whether results have been incorporated into the RIA starting with high impact RIAs.
There is also a need to continuously provide training on the consultation methodological guidance and promote the sharing of best practices. Some of these capacity-building and knowledge-sharing initiatives are in fact implemented on the government level. Trainings on consultations have been organised for ministries; there is also a network of consultation co-ordinators and the Office of the Government organises some best-practice sharing events.
More generally, there seems to be an understanding of the need to strengthen consultation practices in Lithuania. Decision making based on consultations and evidence is one of the 7 expressed principles of the new 18th Government of Lithuania in the Programme of the Government (Government of Lithuania, 2020[8]). The Government Programme Implementation Plan contains commitments to make discussions on RIA an integral part of public consultation, and to enhance the use of the E.Citizen platform as well as updating related regulations and methodologies and employing innovative communication tools (Government of Lithuania, 2021[24]).
Moreover, public consultations could take the form of a continuous engagement with stakeholders allowing them not only to provide opinions upon the request, but to also have a permanent channel to raise concerns and give suggestions on improvements to the current regulatory environment. One good example of such a “bottom-up approach” to regulatory change is provided by the Danish Business Forum (see Box 3.10 below).
In Denmark, the Ministry for Business and Growth launched the Danish Business Forum in 2012 to identify and discuss the compliance and administrative burden that businesses face. The members of the forum include industry and labour organisations, businesses, as well as experts with expertise in simplification. The forum gathers 3 times a year and sends common proposals to the government on the possible avenues for regulatory simplification. These proposals are subject to a “comply or explain” approach whereby the government is obliged to either pursue the proposed initiatives or to explain why these are not pursued. As of 2016, 603 proposals have been made by the forum of which 191 were fully and 189 partially implemented. The total savings to businesses from the implementation of these simplification measures were estimated to amount to 790 million Danish crowns.
The forum helps to bring up the issues that can often seem too minor on a political agenda but represent a significant unnecessary burden for businesses. Examples of regulatory changes introduced by the forum have included amending regulations relating to warehouse fire safety in Denmark. These changes were inspired by the Swedish regulatory environment, which had leaner but “smarter” regulations in this area. The change resulted in a 10% savings on new buildings and operational costs. Another example involved amending regulations to enable car licence plates to be sent directly to garage mechanics, as supposed to them having to collect the plates directly from the issuer.
Source: OECD (2017[32]), OECD Best Practice Principles on Stakeholder Engagement in Regulatory Policy: Draft for Public Consultation, https://www.oecd.org/gov/regulatory-policy/public-consultation-best-practice-principles-on-stakeholder-engagement.htm (accessed on 11May 2021); Danish Business Authority (2021[33]), Presentation: Ex-post Evaluation: Denmark.
Most regulatory proposals are only subject to legal quality control while leaving the rest of the content unchecked
The current process provides several vetting instances from a legal standpoint, while an analysis of the content in substantive terms and the supporting analysis seems to be missing, except for the higher impact RIAs, where some elements are in place. Given the fragmentation of tasks highlighted previously, there appears to be scope for streamlining the legal control functions between the Office of the Government and the Ministry of Justice, as there are some overlaps in the current mandates of the two institutions.
A two-level quality control check for RIAs should be implemented
Lithuania could consider implementing a two-level quality control check of the analytical content of RIAs. The first level of scrutiny, covering the majority of legislative proposals, could be carried out by the in-house teams within ministries, responsible for RIA co-ordination, who can provide feedback on RIA quality to the policy team responsible for drafting the legislative proposal. The Ministry of Justice would continue to deal with the legal aspects of quality control, while the Office of the Government would deal with the quality of RIAs. However, for higher impact legislative proposals, the independent Regulatory Oversight Board would provide publicly available advice on the quality of the corresponding RIAs.
The legislative process lacks sufficient forward planning
The Lithuanian legislative process lacks sufficient forward planning, with insufficient time built into the process for ministries to carry out ex ante RIA and comprehensive stakeholder engagement at a sufficiently early point in the legislative process. This limits the opportunity for these regulatory management tools to inform the development of legislative proposals and may lead to a bias towards legislative solutions for policy problems.
A forward-planning system should be implemented with a clear 18-month rolling calendar of upcoming legislation
In order to address this issue, the Lithuanian Government could consider setting up a forward-planning system. This forward plan would involve a clear 18-month rolling calendar, published online and updated annually, for the development of new legal initiatives, which would be coherent with the Programme for the Government for the political term. The plans would identify from the start any upcoming primary and secondary legislation, and the time implications for the government to undertake necessary stakeholder consultation and ex ante RIA, early in the policy-making cycle. These plans should be developed in co-ordination with the Government’s Legislative Programme. The current Lithuanian government has recently published a similar plan for the whole tenure (2021-2024) for the first time (see the box below) (Government of Lithuania, 2021[34]).
In May 2021, the Lithuanian government published the first medium-term legislative plan covering the period of 2021-2024. The plan includes 260 legislative proposals as well as 90 proposals required for the transposition of EU directives. These legislative proposals include both primary and some secondary legislation. Each proposal is associated with the responsible sponsoring ministry and the timeline. To supplement the legislative plan, the Higher Impact Legislation list for 2021-2023 was published including 35 legislative projects. The plan also foresees the timeline for regulatory impact assessments and public consultations for these legal acts.
The life span of the legislative plan covers the whole tenure of the government and the Office of the Government is charged to update it.
This initiative aims at reducing the annual flow of legislative amendments, by enabling policy makers to plan and co-ordinate their legislative work in advance. It is expected to reduce the frequency of different institutions amending the same legal act, as well as consolidating technical changes of legal acts.
The legislative plan will be published and updated on the E.Citizen website, together with the list of the higher impact legislation, so as to enable greater outreach for the eventual public consultations.
Source: Government of Lithuania (2021[34]), 2021-2024 Legislative Plan of the Government, https://lrv.lt/uploads/main/documents/files/VPA20210331.pdf (accessed on 27 May 2021).
This could be inspired by the practice of a number of OECD jurisdictions that have established forward regulatory planning systems (see Figure 3.6 below) that can inform stakeholders of upcoming programmes of new regulation, so that they can engage in regulatory development and plan for future regulatory changes at the earliest opportunity. A majority provides forward planning by publishing such lists on primary laws and around one third of countries do so for subordinate regulations. A selection of international examples of such forward planning systems are set out in Box 3.12.
In Canada, departments and agencies must post their plans to develop or amend regulations over the next 24 months publicly on their websites. Plans must be updated on an annual basis. Stakeholders should be given the opportunity to provide comments to departments and agencies on their Forward Regulatory Plans, once published, to inform the government on whether the regulatory priorities reflect the issues they face. Initial consultation on instrument choice should occur before a proposed regulatory change is formally incorporated into a Forward Regulatory Plan.
In Germany, the Coalition agreement is adopted at the beginning of each legislative term and it sets the main elements of the government’s policy and planned projects. The exact implementation of those legislative plans are then the competence of the responsible ministers. However, the centre of the government still closely monitors the implementation as each minister has to inform the Chancellor on the progress of implementation and participates in the closed-doors conferences to discuss key objectives. Moreover, Germany has a dedicated unit set in the federal Chancellery that is responsible for the legislative forward planning. This unit maintains a database and tracks how line ministries follow the plans set by the political coalition agreements. This unit helps the inter-ministerial co-ordination and legislative planning enactment.
In South Korea, the Regulatory Reform Committee (RRC) has launched a programme, entitled the Futuristic Regulatory Map, whereby each public administration agency will map its regulatory plans for new emerging industries. Under this programme, the RRC selects emerging industries whose trends are analysed to predict future technology and industry development. Based on this analysis, the regulatory reform plans are devised to plan for the accommodation of future needs. Hence, the RRC uses top-down approach for long-term regulatory planning. The challenges that this programme tries to address are resistance of existing regulations that hinder the entry of new industries and technology, and the lack of necessary regulations in time for the introduction of emerging industry. The pilot project of the programme was conducted on the self-driving cars industry.
Source: OECD (2017[28]), Regulatory Policy in Korea: Towards Better Regulation, http://dx.doi.org/10.1787/9789264274600-en; OECD (2018[22]), Case Studies of RegWatchEurope Regulatory Oversight Bodies and European Union Regulatory Scrutiny Board, https://www.oecd.org/gov/regulatory-policy/Oversight-bodies-web.pdf (accessed on 11 May 2021); https://www.canada.ca/en/government/system/laws/developing-improving-federal-regulations/regulatory-planning/government-wide-forward-regulatory-plans.html; OECD (2010[35]), The Development of New Regulations (accessed on 24 June 2021), https://www.oecd.org/gov/regulatory-policy/45049504.pdf.
Impact assessment upstream during negotiation
The process for co-ordinating the position of Lithuania in the negotiation process for EU legislation is similar to many other OECD countries which are EU members - there are currently 13 OECD member states that require RIA to be conducted in order to determine the country’s position during the negotiation phase (OECD, 2019[36]). This is important since the original impact assessment of the Commission does not necessarily include an identification of the impacts on individual countries. It means the expected individual member state impacts may not have been identified or assessed through a domestic impact assessment process.
In Lithuania, individual ministries are responsible for drafting policy positions in their areas of competence (Government of Lithuania, 2004[18]). The Ministry of Foreign Affairs is in charge of co-ordinating the preparation of Lithuanian positions with regard to draft EU legislative proposals (Government of Lithuania, 2004[18]). Draft positions for the European Council and the Council of the EU meetings are adopted by the Governmental Commission on the EU Affairs and endorsed by the government. A special information system called LINESIS enables online co-operation among state institutions involved in the negotiation process in real time. NGOs and social partners can get access to this system.
A typical position should contain an impact assessment of the legislative draft, as stipulated in the Government Resolution No. 21 of 9 January 2004 “Regarding Coordination of European Affairs”. This Resolution also introduces an obligation to conduct impact assessment according to the methodology adopted by the Government Resolution No. 276. As part of an EU-funded project “Increasing the Efficiency of Lithuania’s EU Policy”, the Ministry has also produced a set of qualitative methodological guidance to assist officials in assessing the impact of draft new EU legislation and their potential impact on Lithuania. (Ministry of Foreign Affairs, 2019[37]). As part of the guidance, officials are advised to undertake a baseline impact assessment on proposed EU legislation and examine the potential impact of a new EU initiative project on 4 key issues:
However, as with domestic legislation, the assessments do not appear to provide a detailed description of the estimated impacts of the legislative proposal. There is no effective quality control of these impact assessments, as the Ministry of Foreign Affairs does not appear to have the analytical capacity to scrutinise the quality of the underlying evidence. The objective could be to assess whether some of the “high impact” EU legislation may require some form of an impact assessment ex ante, and whether this should receive some form of scrutiny or control. If scrutiny or control is warranted, it would have to be co-ordinated by the Office of the Government.
When countries have sophisticated and well-established apparatus for RIA and quality control, they can mobilise those for the purpose of assessing the impact of European proposals. One example from another EU member state, is the National Regulatory Control Council in Germany, which is the national oversight body that co-operates with some ministries but also the European Commission in elaborating “national” impact assessments on the Commission’s legislative proposals. However, for Lithuania, there are many intermediate steps that would need to be envisaged to reach such a situation. The aspiration for Lithuania should be to ensure that proposed European regulations, estimated to contain significant regulatory impacts, should produce RIAs should with a sufficient level of analysis of key issues to properly inform Lithuania’s negotiating position, and these RIAs would undergo quality control from the Office for the Government, with advice from the future Regulatory Oversight Board.
Impact assessment downstream and process for transposition
RIA is conducted more often during the transposition process than the negotiation phase of EU legislation in most of the EU countries. According to OECD data, 26 countries obliged the transposition project to be accompanied by the RIA and 24 countries had the same rules for the RIA of transposition as for the domestic regulations (OECD, 2019[36]).
The process of transposition and implementation of the EU law in Lithuania is highly centralised and well established. A detailed system of allocation of institutional responsibilities and procedural aspects of co-ordination and monitoring of the implementation process is set by the Law on the Government and the Government Resolution No. 21. The European Union Legal Group of the Ministry of Justice is the main institution co-ordinating and monitoring the transposition using the electronic system LINESIS. The Group supervises how the institutions are following the plan and the deadlines. It is required to complete and attach a correlation table to every single draft legal act which is implementing the EU legal act. The European Union Legal Group then reviews the table with its EU law legal experts. All the correlation tables must be uploaded to LINESIS.
However, the fact that the civil servants that were participating in the negotiating process on a particular piece of legislation at the EU level are not involved in its transposition represents a drawback. Therefore, the institutional memory is not fully utilised within ministries. This is, however, a general problem in many other EU member states. Furthermore, as EU directives are transposed into national law, they will be subject to the same obligations to undergo ex ante RIA as mandated by the Government Resolution No.276 and the subsequent analysis is generally subject to the same shortcomings as detailed earlier in this section (Government of Lithuania, 2003[17]).
The European Union Legal Group is also responsible for reducing administrative burdens during the process of EU law transposition, which implies reducing the scope for “gold-plating”, that is to create more administrative requirements and burdens than would be otherwise necessary to just implement the directive. To better formalise the process of preventing gold-plating, the institutional predecessor to the Group (the European Law Department) carried out an EU-funded project in 2015 called “Implementation of European Union Legislation in National Law and Recommendations for Assessing the Justification of Administrative Burdens”. The project’s aim was to create a set of recommendations for public institutions and a concrete methodology for preventing creation of unnecessary administrative burdens in the process of EU law implementation. (Ministry of Justice, 2015[38])
Despite this worthy project, gold-plating remains a problem within Lithuania and maybe is contributing to legal inflation, with ministries over-implementing EU Directives by imposing various national requirements or failing to take advantage of exemptions and other flexibilities written into the directives. As previously mentioned, this may be due in part to a legalistic culture, with a suspicion towards administrative discretion and a lack of overall planning for new legislation. In addition, the transposition process was previously controlled within European Law Department under the Ministry of Justice, which was better resourced than the current EU Legal group and therefore better able to oversee and scrutinise the transposition process, including controlling the issue of gold-plating. Currently, 16 EU countries have specific provisions that require a separate assessment of the impact of the gold-plating of the EU legislation, according to OECD data (OECD, 2019[36]). For example, the government of the Czech Republic has introduced a requirement for ministries to identify whether the transposition of an EU directive goes beyond the requirements of the directive. In cases where no gold plating has been identified, or the expected impacts do not pass a predetermined threshold, the transposed legislative drafts are exempted from RIA (OECD, 2019[36]). A description of Denmark’s approach to ensuring that transposition of EU law does not result in extra burdens to business can be viewed in Box 3.13.
The responsibilities for co-ordinating the transposition process and the capacity of the EU Legal Group should be reviewed
Therefore, in order to strengthen the co-ordination of the transposition process, the Lithuanian Government should consider consolidating the responsibilities for transposition with the Office of the Government, who will have a better strategic view and will better be able to exercise control over the legislative agendas of other ministries. The capacity of the European Union Legal Group at the Ministry of Justice should be reviewed, to examine whether they require additional resourcing for undertaking quality control of transposed legislation, with a view to limiting the possibility for gold plating. It might be helpful for the Office of the Government to require a “gold plating impact review”, as a requirement to be produced by the Ministry of Justice, before agreeing to proceed with transposition. In general, laws issued from the transposition of EU directives should be subject to the same process as other laws, including for the “higher impact” regulations. In this context, the new “Regulatory Oversight Board”, as part of its mandate to scrutinise the quality of RIAs for higher impact laws, should provide opinions on to the evidence supporting the “gold plating” of EU directives.
In 2015, Denmark established a system of two-step quality oversight of the transposition of EU legislation. This new system aimed at ensuring a systematic and uniform approach towards the implementation of EU legislation across government and at avoiding additional burdens for businesses through this process.
The Inter-Ministerial EU Implementation Committee comprises of eight government ministers, situated in the Ministry of Employment, which examine all national legislative proposals deriving from business-oriented EU legislation to ensure that they follow five principles for implementation. These principles include, inter alia, provisions to avoid burdens for businesses stemming from the transposition of EU directives and an implementation going beyond the minimum requirements set in EU legislation. As part of the development of legislation implementing business-oriented EU legislation, all ministries need to submit an implementation schedule to the secretariat of the Committee, explaining whether the five principles have been followed. If a draft law is not in compliance with the five principles, the matter is put before the Inter-Ministerial Committee, which can approve or reject measures going beyond what is required as part of implementing EU legislation.
The external EU Implementation Council comprises 11 members from business, consumer, employers and labour organisations that advises the Committee in its efforts to prevent unnecessary costs for business in implementing new EU legislation. This advisory body has a secretariat situated at the Danish Agency for Labour Market and Recruitment under the Ministry of Employment. If the Council identifies burdensome future EU legislation, it can advise the government through the Inter-Ministerial EU Implementation Committee to lobby proactively already at the development stage of EU legislation. In addition, the Council advises ministries on the transposition of new EU legislation, and all ministries are required to submit an implementation plan to the Council within 4 weeks of the adoption of the directive in Brussels, indicating the planned process and method of implementation. The Council sends recommendations to the ministries on this basis, which are subsequently discussed in the Implementation Committee. The Council can also request a ministry conduct a study of how a certain EU directive is implemented in another EU country and to benchmark it against Danish practice.
Source: OECD (2019[36]), Better Regulation Practices across the European Union, https://doi.org/10.1787/9789264311732-en.
Ex post regulatory assessments are currently co-ordinated by the Ministry of Justice
Regulations should be periodically reviewed to ensure that they remain fit for purpose. Ex post reviews complete the “regulatory cycle” that begins with ex ante assessment of proposals and proceeds to implementation and administration. The broadly accepted notion of a ‘’regulatory cycle’’ recognises that regulations are akin potentially to depreciating assets that require ongoing management and renewal (see Box 3.14 on New Zealand’s novel approach to “regulatory stewardship”). Even if they start out well, many regulations may no longer be fit for purpose some years after. The accumulated costs of this in economic or social terms can be high.
The government of New Zealand has developed a unique policy to managing its regulatory stock, entitled “regulatory stewardship”. Under regulatory stewardship, government agencies (including line ministries) are obliged to adopt a whole-of-system view, and a proactive, collaborative approach to the monitoring and care of any regulatory system in which they hold a policy or operational role.
The government agreed to a set of “Government Expectations for Regulatory Stewardship” in 2013, updated in 2017, in order to give departments more direction as to how they should discharge their regulatory stewardship obligations. These expectations are set out in Part B of the “Government Expectations for Good Regulatory Practice” (where Part A sets out expectations for the design of regulatory systems). In addition, the regulatory stewardship principle is enshrined in the Public Service Act of 2020, which stipulates 5 public service principles.
Regulatory stewardship involves:
Monitoring, reviewing and reporting on existing regulatory systems - this includes working collaboratively:
to monitor the ongoing performance and condition of a regulatory system and the regulatory environment in which it operates, recognising that this environment is continually changing
reviewing the system regularly to determine whether it’s still fit for purpose and likely to remain so in the medium to longer-term
Robust analysis and implementation support for changes to regulatory systems - this includes a systematic impact and risk analysis process before changes are made, and ensuring affected and other interested parties can directly contribute to the design of regulations.
Good regulator practice - this includes providing accessible and timely information and support to help regulated parties understand their obligations, and equipping regulatory workforces with the necessary knowledge and skills.
The Regulatory Quality Team within the Treasury, a central agency, exercises stewardship over the regulatory management system to maintain and enhance the quality of government-initiated regulation. This team is the lead agency on good regulatory practice for New Zealand.
Source: Treasury of New Zealand (2020[39]), Regulatory Stewardship, https://www.treasury.govt.nz/information-and-services/regulation/regulatory-stewardship (accessed on 12 May 2021).
However, according to the OECD iREG data, ex post evaluation systems are still rudimentary in most member countries, and it is still not mandatory to conduct an ex post review in one third of OECD countries. (OECD, 2018[9]). In most countries where a requirement exists, it does not apply systematically to all or major regulations. OECD countries have put in place different types of requirements to trigger ex post evaluations, including “thresholds”, “sunsetting” clauses or automatic evaluation requirements. A growing number of countries conduct evaluations of regulations on similar issues as a “package” (see Figure 3.8 for more information).
In Lithuania, the system of ex post regulatory assessments is being institutionalised. Article 241 of the Law on Legislative Framework mandates the “co-ordinator of ex post regulatory assessments” to (Parliament of Lithuania, 2012[12]):
draft, update and publish the list of legal acts which are to be evaluated ex post
analyse the practices and the tendencies of ex post evaluations and report the findings to the government
recommend to institutions to conduct ex post evaluations or to include an evaluation clause into draft legal acts.
A further Government Resolution 69 of the January 29th, 2020 on the implementation of Law on Legislative Framework mandated the Ministry of Justice to be the “co-ordinator of ex post regulatory assessment” (Government of Lithuania, 2020[40]).
The co-ordination of ex post regulatory assessments should be entrusted to another institution with a broader mandate, such as the Office of the Government
Whilst the Ministry of Justice has a high level of legal expertise, it is not appropriately equipped in terms of analytical capacities for providing a leading function for ex post regulatory assessments. Therefore, the co-ordination of ex post regulatory assessments should be entrusted to another institution with a broader mandate. The task for ex post regulatory assessment should in fact be consolidated with that of ex post evaluation of public policies and programmes. The government of Lithuania could therefore consider attributing the co-ordination mandate for ex post assessment to the Office of the Government, whilst mandating STRATA to provide methodological and analytical support for ex post evaluation, with a broad perspective on ex post evaluation of laws, as well as policies and programmes.
Furthermore, in the medium to long term, to ensure adequate quality control of the content of ex post evaluations, Lithuania could consider implementing a two-level quality control check process, similar to that proposed for ex ante RIAs. The first level of scrutiny, covering the majority of legislative proposals could be carried out by the in-house teams within ministries, responsible for RIA and ex post evaluation co-ordination, who can provide feedback on analytical quality to the policy team responsible for drafting the evaluation. The Office of the Government supported by STRATA for that task in substantive terms would deal with the quality of ex post evaluations. For higher impact legislative proposals, the proposed independent regulatory oversight body would provide publically available advice on the quality of the corresponding ex post evaluations.
While new laws can include an ex post evaluation clause as of 2020, this has not been associated with funding requirements and does not specify a data-gathering process
The OECD Best Practise Principles for Reviewing the Stock of Regulations state that at the stage when regulations are being developed, there would generally be a clearer appreciation of the sort of ex post review that would be most appropriate, given the nature of the regulation, its context and any potential uncertainties about its effects. There should be explicit provision in agency budgets to cover the costs of reviewing the regulations for which they have responsibility. It also states that if the funding for the evaluation is not determined at the inception of a regulation it is more likely that they would be avoided, deferred or undertaken in a form inadequate for the purpose (OECD, 2020[41]).
According to Article 24 of the 2012 Law on the Legislative Framework, if a law regulates a previously unregulated field or amends it significantly, or has a high impact to a specific policy area, individuals or groups of them, an ex post evaluation clause should be included (Parliament of Lithuania, 2012[12]). In practice, however, the law does not detail how these evaluations can be funded, nor makes provisions for data collection. To date, no ex post evaluation has been conducted, as the evaluation has to be conducted at least two years after the law comes into effect. So, Article 24 is currently de facto ineffective as it stands.
Ex post evaluation clauses should be associated with clear financial resources and plans for gathering data
Lithuania could consider associating ex post evaluation clauses with clear financial resources, as well as making sure that institutions anticipate data gathering for the evaluation during the policy implementation phase.
There is also potential for greater coherence between ex ante RIA and ex post evaluation requirements. Indeed, OECD Best Practice states that ex ante RIAs should establish monitoring indicators and data gathering to enable ex post evaluation to take place (OECD, 2020[10]). A responsive administration performs an ex ante RIA, but also provides for monitoring, data collection and evaluation indicators and an ex post evaluation, which itself leads to the identification of the need for further action and a new ex ante assessment phase (STRATA, 2020[19]).
The European Commission is arguably one of the most vocal advocates of the new course given to ex post evaluation of regulation. It has introduced the so-called “evaluating first principle”, according to which the Commission commits “(…) [not to] examine proposals in areas of existing legislation until the regulatory mapping and appropriate subsequent evaluation work has been conducted.” (European Commission, 2012[42]).
The commitment was announced in the political guidelines that President Barroso publicly issued in 2009, at the outset of his second term in office, as well as in various public speeches (Barroso, 2009[43]). The commitment to the principle was renewed as part of the 2015 Better Regulation Package and readdressed in its recent 2017 Better Regulation Communication (European Commission, 2017[44]).
The principle is expected to help the Commission, in the short to mid-term, to re-allocate the services’ resources according to priority axes, raising at the same time the relative importance of ex post evaluation within the policy cycle. The evaluating first principle, if systematically applied, has clear repercussions on the re-organisation of the planning phase of evaluations. This principle bestows the greater importance upon ex post evaluation changing its place in the policy cycle. It has a big impact on the organisation and planning of ex post evaluations of the stock of regulations.
Source: OECD (2019[36]), Better Regulation Practices across the European Union, https://doi.org/10.1787/9789264311732-en; European Commission (2012[42]), EU Regulatory Fitness; Barroso (2009[43]), Political Guidelines for the Next Commission, http://ec.europa.eu/commission_2010-2014/president/pdf/press_20090903_en.pdf (accessed on 12 May 2021); European Commission (2017[44]), Completing the Better Regulation Agenda: Better solutions for better results, COM(2017) 651 final, https://ec.europa.eu/info/sites/info/files/completing-the-better-regulation-agenda-better-solutions-for-better-results_en.pdf (accessed on 12 May 2021).
While ex post evaluations are to be conducted by ministries in their areas of competence, they could solicit help from STRATA or private consultancies as required.
As for ex ante RIAs, ex post evaluations have to be conducted by the staff of the ministries that drafted the regulation. Other state and municipal institutions can help in providing the necessary data for the analysis. The methodology approved on the 5th of May 2021 sets to establish that if more specific skills are required for the evaluation (such as data analysis or conducting surveys) external consultants can be contracted or the help of STRATA can be solicited. For assessments that concern horizontal issues, an inter-ministerial or expert working group can be set up (Ministry of Justice, 2021[13]).
Lithuania could further implement strategies for the management and rationalisation of the stock of existing regulations
The OECD Best Practice Principles on Reviewing the Stock of Regulation state that a “portfolio” of approaches to managing the stock of regulation will generally be needed. In broad terms, such approaches range from programmed reviews, to reviews initiated on an ad hoc basis, or as part of ongoing “management” processes. Most countries have adopted more than one of these approaches utilising forms of review within each category listed below (Table 3.3). These draw on a taxonomy developed by the Australian Productivity Commission.
Reviews often need to be initiated on an ad hoc basis in response to an emergent issue or crisis, such as a natural disaster or major public health problem. Principle-based reviews, which are the most frequently used, can also be established to address a more general theme or concern, such as impediments to competition, or to focus on a particular economic activity or segment of society, such as regional development. “In-depth” reviews are most effective when applied to evaluating major areas of regulation with wide-ranging effects. They seek to assess the appropriateness, effectiveness and efficiency of regulation – and to do so within a wider policy context, in which other forms of intervention may also be in the mix. Public “stocktakes” of regulation are useful for soliciting public views about current problems and priorities. They can also be an effective means of identifying cumulative regulatory burdens or detecting adverse interactions across different regulations. Benchmarking can potentially provide useful information on comparative performance, leading practices and models for reform across jurisdictions and levels of government.
The Lithuanian government has undertaken a number of initiatives over the years with the attempt of rationalising the stock of regulations and reducing burdens on businesses. There have also been projects on reviewing licences, improving enforcement and implementation. A Better Regulation Programme was adopted in 2008, which was intended to introduce a broad range of better regulation initiatives, including those aimed at administrative burden reduction on businesses and prevention, as well as simplification of licences and permits. A Law on Administrative Burden Reduction came into force on 1st of July 2013 (with some amendments in January 2014). The new law also sets an obligation to prepare and submit to the government (and subsequently the parliament) two-year Administrative Burden Reduction Plans.
The Ministry of the Economy and Innovation is responsible for co-ordinating major developments, related to administrative burden measurement and reduction for business, continues its work in prevention of new administrative burdens on business, calculate administrative burden reduction target of every governmental institution and overall countrywide. However, as another line ministry, it is likely to be difficult for it to put pressure on other line ministries or agencies not to increase administrative burdens stemming from regulations in their area of competence. Furthermore, as the measurement of administrative burdens has a separate legal basis to the RIA framework, this may increase the perception that is a separate from the overall RIA process and may increase the difficulty in integrating these considerations into the development of new laws. Therefore the requirements regarding administrative burden measurement could form part of the proposed Government Resolution on Better Regulation (see the sub section “In order to facilitate structural uptake of these findings, Lithuania could consider developing a coherent framework for better regulation”).
The government should initiate a process of codification to simplify the statute book and increase capacity for compliance
Despite previous efforts at legislative simplification and administrative burden reduction, the issue of legislative inflation over time has contributed to an overly complex statute book, that is difficult for external stakeholders (e.g. Small and Medium Sized Enterprises) to understand and navigate. The Government could therefore look to undertaking a process of codification, with a clear plan, co-ordinated by the Ministry of Justice, for the existing stock of regulations, with a view to achieving clearer language and increased capacity for compliance.
An international example of codification has included France, which has made substantial and sustained efforts over time to codify the law (see Box 3.16 for more information). Codification – the work of rationalising and producing a systematic inventory of the law – has resumed over the past 20 years, partly with the aim of addressing the consequences of legislative inflation. Today, more than 40% of the laws in force are grouped into almost 70 codes (OECD, 2010[4]). Another example of codification is Greece, which has been carrying out several reforms of its regulatory framework, including the establishment of a long-term codification plan of the main regulations in 2016 and creation of an electronic portal for access to regulations as well as simplification of law in selected areas (labour law, VAT) in 2015 (OECD, 2018[9]).
Experiences of legislative inflation and drift over the period 1946-1958, and lack of parliamentary stability inspired a series of provisions aimed at rationalising legislative work, as part of the “rationalised parliamentarism”, adopted under the 1958 constitution. Under this constitution of the Fifth Republic, the scope for Parliament to address matters through legislation was defined in a restrictive way, through the introduction of the definition of the “domain of law” whereby the decision not pertaining to this definition are not considered a law and have to be regulated by secondary legislation. This implies that any legal text, which would not be part of the “domain of the law”, could be changed or revised by decree by the government.
This system sets strict limits on the legislative and management prerogatives of parliament to the benefit of the government. Thus Members of Parliament may introduce draft legislation, but article 40 of the Constitution prevents this if its adoption would decrease public financial resources or increase public expenditure. The constitutional reform of July 2008 has lessened the oversight of the executive (notably through the introduction of a shared agenda), without completely writing off rationalised parliamentarism.
Nevertheless, despite this constitutional reform, legislative inflation continued to be a problem in France throughout the 20th century, resulting in an ever-growing stock and complexity of regulations. Seeking alternatives to regulation has not been a key feature of the French legal system, due to both the centralised practice of a country with written Roman law, and the need to accommodate the European framework. In addition, law in France has no automatic sunsetting clauses. Therefore, during the first 30 years of the Fifth Republic the average flow of new laws increased by 35% each year. In addition, the government ratified 670 degrees a year on average and the official gazette increased by the factor of 2.4 between 1976 and 1990.
To respond to the challenges of such a legal proliferation, France has engaged in multiple rounds of codification and administrative simplification. The two main rounds of administrative simplification started in 1983 and 1996 during which the number of forms and declarations were reduced upstream. Moreover, the codification was resumed in 1989. Between 1989 and 1996, the Codification Council presided by the Prime Minister produced 5 codes that were adopted by decree of the Council of State. An important simplification for codification was a Constitutional Council decision of 1999 that confirmed that codification has a constitutional value. Following this decision, the government has ratified 9 more legal codes from 1999 to 2004. As of 2010, 40% of laws were arranged in 70 codes. However, unlike in Lithuania, the government is in a position to proceed with codification through “law decrees”, i.e. through a “Habilitation Law”, which then allows the government to proceed and cancel a block of existing legislation and to replace it with a code. Such a code may not cover 100 % of the stock of existing laws, but it may offer an opportunity for simplification, which is allowed through codification (i.e. known as “codification at non constant legal scope”).
Source: OECD (2004[45]), Regulatory Reform in France: Government Capacity to Assure High Quality Regulation, https://doi.org/10.1787/9789264015487-en.
To date, there is no single formal government framework supporting regulatory policy in Lithuania. This would be particularly needed in order to clarify how ex ante and ex post assessment contribute to bettering regulation. It could also provide a means for the government to link better regulation tools to its overall legislative goals e.g. reducing legislative inflation. This point was underlined by the OECD in (2015[11]), and it still stands today:
While a number of pieces of legislation and government resolutions have been adopted, this significant effort still falls short of an overall policy for better regulation. Lithuania could consider bringing the different elements of regulatory policy together in an integrated strategic plan for Better Regulation, with identified objectives and a clear communication strategy.
This strategy or document could take the form of a Government Resolution on Better Regulation, that would bring together the different existing provisions on better regulation, as well as identify the lead institution(s) in charge of co-ordinating its implementation and clarify the objectives of such a policy. This resolution could also be used to place a number of different parts of the regulatory policy process, on to a more formal legal basis e.g. it could formalise the existing proportionality process, including the developing of the Lists for Higher Impact Legislation, as well as formalising the role of STRATA in advising on the quality control of these proposed laws. It should also be used to ensure that the measurement of administrative burdens for business, overseen by the Ministry of the Economy & Innovation and that of administrative burden on citizens, overseen by the Ministry of Interior, form a key part of the RIA process instead of being separate to it. Having such a framework would underline high-level political support for a better regulation agenda. An example of the Canadian Government’s whole-of-the-government regulatory policy can be viewed in Box 3.17.
Canada has a long history of instituting a whole-of-the-government policy for better regulation. The first such policy was first put in place in 1999 with the Government of Canada Regulatory Policy. Its last edition came in the form of Cabinet Directive on Regulation in 2018. The directive clearly states that “the regulatory life cycle approach requires departments and agencies to examine and analyse regulations through all stages of their life cycle”.
It establishes the requirements for developing subordinate regulations, with ex ante RIA being mandatory and made public on a central registry, along with the draft legal text. Open consultation is conducted for all subordinate regulations and regulators must indicate how comments from the public were addressed, unless the proposal is exempted from the standard process. Departments and agencies are required to undertake periodic reviews of their regulatory stock to ensure that regulations achieve intended objectives. The directive sets four guiding principles that the departments and agencies need to follow at each step of the policy cycle.
Regulations protect and advance the public interest and support good government: Regulations are justified by a clear rationale in terms of protecting the health, safety, security, social and economic well-being of Canadians, and the environment.
The regulatory process is modern, open, and transparent: Regulations, and their related activities, are accessible and understandable, and are created, maintained, and reviewed in an open, transparent, and inclusive way that meaningfully engages the public and stakeholders, including Indigenous peoples, early on.
Regulatory decision making is evidence-informed: Proposals and decisions are based on evidence, robust analysis of costs and benefits, and the assessment of risk, while being open to public scrutiny.
Regulations support a fair and competitive economy: Regulations should aim to support and promote inclusive economic growth, entrepreneurship, and innovation for the benefit of Canadians and businesses. Opportunities for [international] regulatory co-operation and the development of aligned regulations should be considered and implemented wherever possible.
Source: Government of Canada (2020[46]), Cabinet Directive on Regulation, https://www.canada.ca/en/government/system/laws/developing-improving-federal-regulations/requirements-developing-managing-reviewing-regulations/guidelines-tools/cabinet-directive-regulation.html#toc4 (accessed on 12 May 2021); OECD (2019[36]), Better Regulation Practices across the European Union, https://doi.org/10.1787/9789264311732-en.
Implement a series of concrete measures to tackle the consequences of legal inflation, by:
Launching a process of codification for the existing stock of regulations, with a view to achieving clearer language, increased capacity for compliance.
Setting up a forward planning system, by putting in place a clear 18-month rolling calendar for the development of new legal initiatives, in co-ordination with the key government programme initiatives, identifying from the start any upcoming primary and secondary legislation, what are the time implications of conducting stakeholder consultation and RIA.
Reviewing the process for implementing EU directives; consolidate responsibilities for co-ordinating transposition processes with the Office of the Government, while strengthening the quality control capacity at the Ministry of Justice to limit the possibility for gold plating. The new “Regulatory Oversight Board”, as part of its mandate to scrutinise the quality of RIAs for higher impact laws, should provide opinions on the evidence supporting the “gold plating” of EU directives.
Set up a strategic framework to address the longer-term challenges of legal inflation, by:
Establishing a memorandum of understanding between the executive branch with the Parliament, which would help to create political momentum for increasing the number and enhancing the quality of RIAs for legislative proposals initiated by the Members of Parliament, as well as leading to a joint understanding of the planned calendar for the work of Parliament.
Setting up a strategic task force on the future of the Lithuanian legal framework, aimed at reflecting on the long-term costs to the economy of current legal inflation and the implications of the current legal culture in Lithuania (with laws that are very detail-oriented).
Develop a formal policy for better regulation from a whole of government perspective.
This policy could take the form of a Government Resolution on Better Regulation. It should:
Bring together different elements of better regulation policy currently regulated by separate government resolutions (e.g. RIA, ex post review, administrative burden on business, administrative burden on citizens) under a single better regulation framework.
Clarify mandates, responsibilities and processes across the government for ex ante RIA and ex post evaluation and rationalise co-ordination functions with a network co-ordinated from the Office of the Government.
Formally establish a two-tier RIA system to strengthen the effectiveness, with quality control conducted by STRATA for high impact laws and the Office of the Government from a general perspective.
Enhance overall transparency of the quality of public consultation for RIAs. It could become mandatory to make consultations open for a minimum period of 3 to 6 weeks for all legal acts and RIAs, with online consultations being the default option.
Strengthen the management of RIA, through the following steps:
The Office of the Government could clarify further RIA procedures and practices. It could develop, jointly with STRATA, tools for government analysis to improve and facilitate RIA.
Ministries could clarify the processes and responsibilities as regards the RIA process and legislature drafting. They could mobilise their internal analytical capacities so that the available analytical resources can be easily identified by the policy units drafting a legislative project.
All impact assessments and the outcome of the consultation process should be made public on a dedicated government website.
STRATA could offer targeted support on demand from ministries, and with due compensation, for higher impact RIAs.
A RIA community of practice should be established, co-ordinated by the Office of the Government to facilitate knowledge sharing, a shared understanding of the processes, and greater access to analytical content with the support of STRATA through seminars and other networking opportunities.
Consider establishing an independent “Regulatory Oversight Board” that would:
Provide public advice on the quality of the evidence supporting ex ante RIAs for high impact laws and provide regular assessments of the overall quality of the RIA process across government, and communicate the results with the parliament and the public.
Ensure the credibility and independence of this board to guarantee the impact and legitimacy of its advice.
Be supported by STRATA, which should serve as the secretariat of this board.
Develop systematic and permanent training programmes for civil servants on RIA. In particular, STRATA should further:
Streamline the quality control and co-ordination functions between the Office of Government and the Ministry of Justice:
While the responsibility for ex ante legal conformity control would be entrusted to the Ministry of Justice, the Office of Government would focus exclusively on the overall quality of higher impact legislation.
The co-ordination function of the Ministry of justice vis-a-vis the implementation of EU directives would be transferred back to the Office of the Government.
Improve the effectiveness of the ex post regulatory evaluation framework by:
Mandate the Office of Government to co-ordinate ex post policy and regulatory evaluations across government.
STRATA should act as the main institution in charge of promoting ex post evaluation, guidance with possibility of establishing a working group involving the OG, CPMA, the Supreme Audit Institution and the Ministry of Finance.
Consider how ex post evaluation clauses could be associated with the subsequent data gathering strategy and clear financial resources to increase the link between RIA and ex post evaluation.
References
[43] Barroso, J. (2009), Political Guidelines for the Next Commissio, https://ec.europa.eu/archives/commission_2010-2014/president/pdf/press_20090903_en.pdf (accessed on 12 May 2021).
[47] Constitutional Court (2019), Lietuvos konstitucinio teismo nutarimas dėl Lietuvos Respublikos miškų įstatymo, Lietuvos Seimo statuto 164 straipsnio 1 dalies atitikties Lietuvos Respublikos Konstitucijai.
[33] Danish Business Authority (2021), Presentation: Ex-post Evaluation: Denmark.
[31] European Commission (2019), Regulatory Scrutiny Board, https://ec.europa.eu/info/law/law-making-process/regulatory-scrutiny-board_en (accessed on 11 May 2021).
[44] European Commission (2017), Completing the Better Regulation Agenda: Better solutions for better results, https://ec.europa.eu/info/sites/info/files/completing-the-better-regulation-agenda-better-solutions-for-better-results_en.pdf (accessed on 12 May 2021).
[42] European Commission (2012), EU Regulatory Fitness.
[46] Government of Canada (2020), Cabinet Directive on Regulation, https://www.canada.ca/en/government/system/laws/developing-improving-federal-regulations/requirements-developing-managing-reviewing-regulations/guidelines-tools/cabinet-directive-regulation.html#toc4 (accessed on 12 May 2021).
[24] Government of Lithuania (2021), Aštuonioliktosios Lietuvos Respublikos Vyriausybės programos nuostatū įgyvendinimo planas Nr. 21-20649 (The Implementation Plan of the 18th Government of the Republic of Lithuania), https://lrv.lt/uploads/main/documents/files/VPN%C4%AEP%20projektas.pdf (accessed on 27 May 2021).
[34] Government of Lithuania (2021), Vyriausybės 2021-2024 metų teisėkūros planas (2021-2024 Government Legislative Plan), https://lrv.lt/uploads/main/documents/files/VPA20210331.pdf (accessed on 27 May 2021).
[40] Government of Lithuania (2020), Nutarimas dėl Lietuvos Respublikos teisėkūros pagrindų įstatymo įgyvendinimo (Government Resolution on the implementation of the Law of the Legislative Framework ).
[8] Government of Lithuania (2020), “Ruling on the Program of the Government of the Republic of Lithuania”, https://www.e-tar.lt/portal/lt/legalAct/ed22bb703bc311eb8d9fe110e148c770.
[14] Government of Lithuania (2012), Lietuvos Respublikos vyriausybės rezoliucija dėl administracinės naštos ūkio subjektams nustatymo metodikos patvirtinimo (Government Resolution Regarding the Methodology for Evaluating Administrative Burden on Business).
[16] Government of Lithuania (2011), Lietuvos Respublikos vyriausybės rezoliucija dėl administracinės naštos piliečiams ir kitiems asmenims nustatymo ir įvertinimo metodikos patvirtinimo (Government Resolution on the Methodology of Administrative Burden on Citizens and Other Persons Evaluation).
[18] Government of Lithuania (2004), Europos Sąjungos reikalų koordinavimo taisyklės (Rules of the Coordination of EU Affairs), Nr. 21, https://e-seimas.lrs.lt/portal/legalAct/lt/TAD/TAIS.224896/cnXUUgmheP.
[17] Government of Lithuania (2003), Lietuvos respublikos vyriausybės nutarimas dėl numatomo teisinio reguliavimo poveikio vertinimo metodikos patvirtinimo (Government Resolution on Draft Regulation Impact Assessment Methodology), https://e-seimas.lrs.lt/portal/legalAct/lt/TAD/TAIS.205970/JVbZEZPFRn.
[15] Government of Lithuania (1998), Nutarimas dėl Lietuvos Respublikos Ekonomikos ir Inovacijų Ministerijos nuostatų patvirtinimo “Resolution on the Statute of the Ministry of Economy and Innovation”.
[53] Lithuania, G. (1998), Nutarimas dėl Lietuvos Respublikos Ekonomiko ir Inovacijų Ministerijos nuostatų patvirtinimo “Resolution on the Statute of the Ministry of Economy”.
[50] Lithuanian Free Market Institute (2019), Šešėlinės ekonomikos tendencijos Lietuvoje (Tendencies of Shadow Economy in Lithuania).
[3] Lukošaitis, D. (2020), “Legislation in Lithuania: Institutional and Cultural Fragments of Problem Perception”, Parliamentary Studies, http://www.parlamentostudijos.lt/Nr28/files/24-46.pdf.
[37] Ministry of Foreign Affairs (2019), Recommendations on new EU initiatives: project impact assessment, Government of Lithuania, Eugrimas Publishing House, Vilnius.
[13] Ministry of Justice (2021), Ex-post Vertinimo metodologija (Ex-post Evaluation Methodology).
[38] Ministry of Justice (2015), Implementation of European Union Legislation in National Law and Recommendations for Assessing the Justification of Administrative Burdens.
[7] OECD (2020), Building Capacity for Evidence-Informed Policy-Making: Lessons from Country Experiences, OECD public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/86331250-en.
[10] OECD (2020), OECD Best Practice Principles for Regulatory Policy: Regulatory Impact Assessment, OECD, Paris, https://doi.org/10.1787/7a9638cb-en.
[26] OECD (2020), Review of International Regulatory Co-operation of the United Kingdom, OECD Publishing, Paris, https://dx.doi.org/10.1787/09be52f0-en.
[41] OECD (2020), Reviewing the Stock of Regulation, OECD Best Practice Principles for Regulatory Policy, OECD Publishing, Paris, https://doi.org/10.1787/1a8f33bc-en.
[36] OECD (2019), Better Regulation Practices across the European Union, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264311732-en.
[27] OECD (2019), Implementing Regulatory Impact Analysis in the Central Government of Peru: Case Studies 2014-16, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264305786-en.
[22] OECD (2018), Case Studies of RegWatchEurope Regulatory Oversight Bodies and European Union Regulatory Scrutiny Board, OECD, Paris, http://www.oecd.org/gov/regulatory-policy/regulatory-oversight-bodies-2018.htm.
[9] OECD (2018), OECD Regulatory Policy Outlook 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264303072-en.
[32] OECD (2017), OECD Best Practice Principles on Stakeholder Engagement in Regulatory Policy: Draft for Public Consultation, OECD, Paris, https://www.oecd.org/gov/regulatory-policy/public-consultation-best-practice-principles-on-stakeholder-engagement.htm.
[28] OECD (2017), Regulatory Policy in Korea: Towards Better Regulation, OECD Reviews of Regulatory Reform, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264274600-en.
[30] OECD (2016), OECD Investment Policy Reviews: Philippines 2016, OECD Investment Policy Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264254510-en.
[5] OECD (2016), Regulatory Policy in Chile: Government Capacity to Ensure High-Quality Regulation, OECD Reviews of Regulatory Reform, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264254596-en.
[25] OECD (2015), OECD Regulatory Policy Outlook 2015, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264238770-en.
[11] OECD (2015), Regulatory Policy in Lithuania: Focusing on the Delivery Side, OECD Reviews of Regulatory Reform, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264239340-en.
[23] OECD (2012), Recommendation of the Council on Regulatory Policy and Governance, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264209022-en.
[52] OECD (2010), Better Regulation in Europe: An OECD Assessment of Regulatory Capacity in the 15 Original EU Member States.
[4] OECD (2010), Better Regulation in Europe: Belgium, OECD, Paris, https://www.oecd.org/gov/regulatory-policy/betterregulationineuropebelgium.htm.
[6] OECD (2010), Regulatory Policy and the Road to Sustainable Growth, OECD, Paris, http://www.oecd.org/regreform/policyconference/46270065.pdf.
[35] OECD (2010), The Development of New Regulations, OECD, Paris, https://www.oecd.org/gov/regulatory-policy/45049504.pdf (accessed on 24 June 2021).
[45] OECD (2004), OECD Reviews of Regulatory Reform: France, OECD Publishing, Paris, https://doi.org/10.1787/9789264015487-en.
[21] Office of Government (2011), Soccialinių sąnaudų-naudos analizės metodinės gairės (CBA and CEA Methodological Guidelines), https://finmin.lrv.lt/uploads/finmin/documents/files/sanaudu_gaires.pdf (accessed on 24 November 2020).
[29] Office of the Government (2018), Methodology and Guidelines of Public Consultation (Viešūjų Konsultacijų Metodika ir Taikymo Gairės), Office of the Government, Vilnius.
[1] Parliament of Lithuania (2020), Teisės aktų leidyba 2016-2020 m. kadencijos Seimo I-IX sesijoje (2016-11-14 - 2020-11-12) (Legislative Process in the I-IX sessions of the 2016-2020 Parliament (14-11-2016 - 12-11-2020)).
[51] Parliament of Lithuania (2012), Lietuvos Respublikos administracinės naštos mažinimo įstatymas (Law on the Reduction of Administrative Burden).
[54] Parliament of Lithuania (2012), Teisėkūros pagrindų įstatymas XIII-3243, Law on the Legislative Framework of the Republic of Lithuania (last amended 26 June 2020).
[12] Parliament of Lithuania (2012), Teisėkūros pagrindų įstatymas XIII-3243, Law on the Legislative Framework of the Republic of Lithuania, (last amended 26 June 2020).
[48] Prime Minister’s Office (2020), Ministro pirmininko nota dėl numatomo teisinio reguliavimo poveikio vertinimo kokybės gerinimo (Note of the Prime Minister on the RIA Quality Improvement).
[49] Schneider, F. and D. Bouman (2019), “IFO World Economic Survey 2019”, Vol. 18/04, pp. 1-23, http://hdl.handle.net/10419/219172 (accessed on 27 November 2020).
[19] STRATA (2020), Numatomo Teisinio Reguliavimo Poveikio Vertinimas (Ex ante Regulatory Impact Assessment).
[20] STRATA (2019), Valstybės tarnautojų apklausa apie poveikio vertinimą ir tyrimo rezultatai (Civil Service Survey on Impact Assessment and study results).
[2] Supreme Audit Institution (2018), Teisėkūros procesas (Legislative Process) Nr. VA-2018-P-40-6-2, Supreme Audit Institutoin, Vilnius, Lithuania.
[39] Treasury of New Zealand (2020), Regulatory Stewardship.
Notes
← 1. The Constitutional Court ruled that the use of the urgency procedure must be limited to the circumstances listed in the constitution, namely, under the state of emergency, mobilisation, when taking the decision to use armed forces, when there is an urgent need to fulfil international obligations, when there is a threat to the safety of the state or the public that necessitates a legislative response (Constitutional Court, 2019[47]).
← 2. https://www.lrkt.lt/lt/apie-teisma/naujienos/1331/konstitucinis-teismas-vien-poreikis-patvirtinti-valstybes-biudzeta-nera-konstituciskai-pateisinama-aplinkybe-leidzianti-mokesciu-istatymams-skubiai-isigalioti:332
← 3. Dr. Friedrich Schneider (Schneider and Bouman, 2019[49]) using Multiple Indicators and Multiple Causes method estimates Lithuanian shadow economy to be 21.9% of GDP (EU average being 15%). Using the same estimation method the figure was 30% for 2010, 26% for 2015 and 24% for 2017 (Lithuanian Free Market Institute, 2019[50]).
← 4. OECD review conducted in 2015 that focuses on the delivery of Lithuanian regulatory policy and licencing. It can be retrieved from https://www.oecd.org/gov/regulatory-policy-in-lithuania-9789264239340-en.htm.
← 5. This transfer was documented in the minutes of the meeting of the 16th January 2020 of the Cabinet of Ministers (2020[48]). However, these processes will be further formalised in the appropriate legal acts.
← 6. Apart from the general RIA framework as detailed by the Law of 2012 on Legislative Framework (Parliament of Lithuania, 2012[54]), the administrative burden assessment is mandated by the Law of 2012 on the Administrative Burden Reduction (Parliament of Lithuania, 2012[51]). Hence, the special focus on the administrative burden assessment as supposed to other areas of competence of the ministry. The ministry is also responsible for the policy formation in other areas of competence (Lithuania, 1998[53]).
← 7. Gold-plating can be defined as the over-implementation of an EC Directive through the imposition of national requirements going beyond the actual requirements of the Directive, resulting in extra costs and burdens. (OECD, 2010[52])
← 8. The 2012 Recommendation of the Council of the OECD on Regulatory Policy and Governance recommends that governments regularly publish reports on the performance of regulatory policy and reform programmes and the public authorities applying the regulations. Such reports should also include information on how regulatory tools such as Regulatory Impact Assessment (RIA), public consultation practices and reviews of existing regulations are functioning in practice. (OECD, 2012[23])
← 9. The calculator recently developed by STRATA helps institutions to calculate the ex ante impact of regulations but still needs to be tested in practise. It is based on the CBA calculator developed by the CPMA to evaluate investments. It can be retrieved from https://strata.gov.lt/lt/poveikio-vertinimas/sna-skaiciuokle.