1. Introduction
The coronavirus disease 2019 (COVID-19) pandemic that suddenly expanded in the spring of 2020 put decision-makers and law drafters in an unexpected and unprecedented situation. In order to prevent the spread of the virus, regulations were prepared on a fast schedule, which meant, at the same time, interfering with fundamental and human rights, for example, in the form of restrictive measures for freedom of trade and movement. Law-drafting during the pandemic was characterized by exceptional urgency and uncertainty, as well as making decisions based on incomplete information.
Extensive interference with fundamental and human rights means that the restriction regulations have had an exceptionally large social significance. This, furthermore, underlines the essentiality of both regulatory ex ante and ex post impact assessments. Ex ante impact assessment can be used to assess the societal costs and benefits of various regulatory alternatives and thereby obtain information on the net benefits of COVID-19 restrictions. Correspondingly, ex post impact assessments have an important role in obtaining information for future crises. However, in crises (like COVID-19), there is some contradiction in terms of ex ante assessment. On the other hand, the social significance of the restrictions would favor a very diligent impact assessment, but at the same time, haste and incomplete information create obstacles to a careful assessment.
The impact assessments in so-called “normal conditions” have been examined in various studies during the last decades (see e.g. Hahn et al., Reference Hahn, Burnett, Chan, Mader and Moyle2000; Renda, Reference Renda2006; Hahn & Dudley, Reference Hahn and Dudley2007; Belcore & Ellig, Reference Belcore and Ellig2008; Cecot et al., Reference Cecot, Hahn, Renda and Schrefler2008; Ellig et al., Reference Ellig, McLaughlin and Morrall2013; Fritsch et al., Reference Fritsch, Radaelli, Schrefler and Renda2013; Ellig & Fike, Reference Ellig and Fike2016; Uusikylä et al., Reference Uusikylä, Keinänen, Vartiainen, Ervasti, Salminen, Kettinen, Lintinen, Köppä and Lindström2023). Impact assessments and law-drafting have been studied in recent years, also in the context of the COVID-19 pandemic. For example, Staronova et al., (Reference Staronova, Lacková and Sloboda2023) found that the quality of law-drafting decreased in the post-pandemic period from the perspective of stakeholder participation in Slovakia. Using Polish data, Jonski and Rogowski (Reference Jonski and Rogowski2023) found that COVID-19 impact assessments and thereby evidence-based decision-making were widely lacking.
In this study, the law-drafting and impact assessments during the COVID-19 pandemic are examined using Finnish data. The Finnish COVID-19 impact assessments have been studied earlier by Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024), where the analysis was conducted using quantitative methods and regression estimations. According to the findings of Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024), the rush and uncertainty brought by the pandemic were connected to a lesser utilization of an evidence-based approach in law-drafting. In the various areas of an evidence base, the rush was particularly evident in the skipping of consultations.
In Vartiainen and Keinänen’s (Reference Vartiainen and Keinänen2024) study, impact assessments were examined with a relatively large amount of data (311 government bills), which made it possible to produce generalizable information about the utilization of an evidence base. However, the evaluation metrics used by Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024) were quite rough “yes/no” level indicators. In other words, the Finnish government bills were examined to see if a certain quality criterion, such as the assessment of regulatory alternatives, was included in the government bill documents or not. In this study, the approach is different, and the COVID-19-related government bills are analyzed using qualitative methods. The purpose is to produce more in-depth information about the practices and problem areas of the impact assessments by focusing on socially significant government bills related to fundamental rights restrictions under the Finnish Constitution.
The analysis of the government bills is divided into two parts. The government bill documents are used to analyze both the carrying out of ex ante impact assessments and the preliminary consideration of implementation (enforcement) and ex post impact assessments. The research questions are set as follows:
-
1. How have ex ante impact assessments been carried out in COVID-19-related law-drafting?
-
2. How have the forthcoming implementation and ex post impact assessments been considered during the law-drafting phase?
The aim of this study is to provide information on law-drafting during the COVID-19 pandemic to support preparedness and resilience for future crises. The results can help identify successes and failures from an impact assessment perspective and thus help in learning to pay attention to quality issues in law-drafting in urgent and unexpected circumstances.
The article has the following structure. The next section introduces the theoretical background. When analyzing impact assessments, it is important to identify the criteria to which the COVID-19 government bills are reflected. In this study, the criteria are based on rationality, the key principles of which are to assess the benefits and costs of different regulatory alternatives (in this case, COVID-19 restrictions) to enhance the overall societal benefits. The third section displays the data and methods, and the fourth and fifth sections present the results. The sixth section concludes the most central findings and shows the recommendations for preparedness for future crises.
2. Rationality in law-drafting
When examining the restrictions planned to prevent the spread of the coronavirus, it is essential to assess, as thoroughly as possible, their societal benefits and costs. The comparison of benefits and costs is emphasized in the rational choice theory, which has served for decades as a mainstream economic theory in explaining the behavior of various actors, such as people and businesses (Cooter & Ulen, Reference Cooter and Ulen2012, 50–51). The rational choice theory is also applied in the context of law-drafting. In contrast to the rational choice theory, the purpose of rationality in law-drafting is not to explain or predict how the actors (in this case, law drafters and decision-makers) make choices or how law-drafting is conducted in practice. Instead, the purpose is to define the criteria for a high-quality law-drafting and impact assessment process. The phases of the rational (law-drafting) model have been reviewed in several different sources (see, e.g. Patton & Sawicki, Reference Patton and Sawicki1993, 46–52; Bohne, Reference Bohne and Verschuure2009, 63–64, 67; Popelier & Verlinden, Reference Popelier, Verlinden and Verschuure2009; Carroll, Reference Carroll2010, 114; Tala, Reference Tala2010, 208; Schwartz, Reference Schwartz, Dunlop and Radaelli2016, 35–41; Vartiainen, Reference Vartiainen2022, 290; Vartiainen & Keinänen, Reference Vartiainen and Keinänen2024, 3–4).
In accordance with the rational model, at first, the societal problem must be defined and a goal setting created to solve the problem. For COVID-19 restrictions, the societal problem was related to the spread of the virus and the adequacy of healthcare’s capacity. After goal setting, the impacts of regulatory alternatives should be assessed. It is essential to assess the costs and benefits of different alternatives and choose the alternative with the highest societal net benefits. Whenever possible, the costs and benefits should be quantified and monetized in a comparable manner. Various impact assessment methods, such as benefit–cost analysis, can be used in the evaluation. The impact assessments of COVID-19 restrictions would, therefore, have to evaluate the benefits and costs between a completely unrestricted society, partial restrictions, and complete closure of society. In line with the principles of the rational model, the assessment should utilize a comprehensive evidence base and scientific knowledge.
The final stages of the rational model consist of the implementation of the chosen regulatory alternative and its ex post impact assessment. The success of implementation is a key factor determining the effects of regulations, and implementation should be considered already in the law-drafting process. Important information can be obtained, for example, through consultation processes, which can be used to identify potential problems during the implementation phase (Naundorf & Radaelli, Reference Naundorf, Radaelli, Karpen and Xanthaki2017, 202–203).
Ex post impact assessment should also be considered in the law-drafting process. This can include planning the timing, resources and the choice of the organization or operator that will conduct the evaluation. Furthermore, law drafters can pre-identify evaluation subjects and data on which the evaluation will be based (Harrington & Morgenstern Reference Harrington and Morgenstern2004, 20; Keinänen et al., Reference Keinänen, Heikkinen and Rinne2019; Bennear & Wiener Reference Bennear and Wiener2021, 489). If the upcoming evaluation is planned as precisely as possible already in the law-drafting, the planning will help the actual execution of the assessment: it is more probable that the evaluation will actually take place if it is clear who will be responsible for performing it and how it will be actually performed. The planning of ex post evaluation can practically mean, for example, data collection from the starting point, in which case social development can be compared before and after the regulatory change (Vartiainen Reference Vartiainen2021, 138). Harrington and Morgenstern (Reference Harrington and Morgenstern2004, 19) point out that a proper impact assessment has data on the pre-regulatory environment. According to them, the drafters of the ex ante impact assessments (regulatory impact assessments; RIA) are in the best position to define the data, the models and the time to complete the ex post evaluations. As Harrington and Morgenstern (Reference Harrington and Morgenstern2004, 20) conclude: “the best time to begin an ex post evaluation of a regulation is before the regulation becomes effective.”
Ex post impact assessment is an integral part of the rational model, as the information obtained through ex post evaluation can be used in the ex ante impact assessments of future regulatory projects (Mergaert & Minto Reference Mergaert and Minto2015, 48; Vartiainen Reference Vartiainen2022). Hence, the rational process does not end at the ex post evaluation, but the ex post phase connects back to the first stages concerning problem definition, goal setting and ex ante impact assessment. Therefore, it can be concluded that ex post evaluation should be considered, as far as possible, already in the law-drafting phase, even in an urgent COVID-19-related drafting process. In this way, the best possible conditions would exist for conducting future ex post impact assessments.
The rational model’s principles as factors determining high-quality law-drafting emerge not only through the literature but also through official law-drafting and impact assessment guidelines. The practical requirements for the evidence-based comparison of benefits and costs of different regulatory alternatives originate especially from the 1980s in the United States and the administration of President Reagan. In recent decades, rational model principles have been prominently featured in the European Union’s operations for better regulation, in the OECD’s (Organisation for Economic Co-operation and Development) recommendations, as well as in the law-drafting guidelines and quality improvement measures of several other countries in addition to the United States (Harrington & Morgenstern, Reference Harrington and Morgenstern2004, 2; Hahn & Tetlock, Reference Hahn and Tetlock2008, 68; Bohne, Reference Bohne and Verschuure2009, 64; Baldwin, Reference Baldwin, Baldwin, Cave and Lodge2010, 5–6; Tala, Reference Tala2010; Ellig et al., Reference Ellig, McLaughlin and Morrall2013, 153; Vartiainen, Reference Vartiainen2022, 289–290).
The principles of the rational model can thus be recognized also in the Finnish law-drafting and impact assessment guidelines that embody, on a practical level, the role of the rational model as a central standard for high-quality law-drafting. The guidelines (Finnish Government, 2022; Ministry of Justice Finland) emphasize, among other things, evidence baseness, assessment of alternatives, and the advantages of ex post impact assessment. According to the guidelines, the government bill must set out the relevant research evidence and present a comparison of the positive and negative impacts of different regulatory alternatives. When applicable, alternatives can be compared, for example, using a cost–benefit analysis. Furthermore, the exhortation to analyze impacts quantitatively and in monetary form is highlighted especially for economic impacts. The government bill must also disclose how the proposed regulation will be monitored (or evaluated ex post). The guidelines specify that the need for monitoring, as well as the data, schedule, and resources of monitoring, are necessary to consider already in the law-drafting phase.
The comparison between benefits and costs also appears in Finnish legislation in the principle of proportionality and necessity. Proportionality and necessity mean that the benefits of the COVID-19 restrictions must be weighed against the disadvantages they cause, for example, in the context of freedom of trade and movement. During the COVID-19 pandemic, the Finnish Constitutional Law Committee repeatedly stressed compliance with the proportionality and necessity principle. Furthermore, the Committee emphasized how compliance with these requirements requires, on the one hand, an assessment of the regulatory alternatives of restrictions of different intensities and, on the other hand, a sufficient justification of the necessity of the proposed restriction.
Even though law-drafting is guided, even on a practical level, to be performed in line with the rational model, actual law-drafting by far does not always operate in accordance with those principles. In practical law-drafting and decision-making, different values and interests exist between politicians and stakeholders, and regulation is often shaped as a compromise between these interests (Baldwin, Reference Baldwin, Baldwin, Cave and Lodge2010, 11–12; Carroll, Reference Carroll2010, 120–121; Tala, Reference Tala2010, 210–211; Vartiainen & Keinänen, Reference Vartiainen and Keinänen2024, 3–4). In addition, the reasons behind the inadequate compliance are usually related to the lack of resources, expertise, and proper research and data, and furthermore, to tight law-drafting schedules and weak regulatory management (Carroll, Reference Carroll2010, 121; Vartiainen & Keinänen, Reference Vartiainen and Keinänen2024, 19).
Lack of compliance with the rational model can be observed in many studies, including those on the United States and the European Union (see e.g. Hahn et al., Reference Hahn, Burnett, Chan, Mader and Moyle2000; Renda, Reference Renda2006; Hahn & Dudley, Reference Hahn and Dudley2007; Belcore & Ellig, Reference Belcore and Ellig2008; Cecot et al., Reference Cecot, Hahn, Renda and Schrefler2008; Ellig et al., Reference Ellig, McLaughlin and Morrall2013; Fritsch et al., Reference Fritsch, Radaelli, Schrefler and Renda2013; Bennear & Wiener, Reference Bennear and Wiener2021). According to the previous research, the assessment of alternatives, costs, and net benefits and the quantification of the estimated impacts, as well as conducting ex post evaluations and taking them into account in the law-drafting phase have been incomplete. Findings of Finnish law-drafting are quite similar. Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024) examined the compliance of rationality in Finnish government bills in 2020. According to the results, the average bill (that was not related to the COVID-19 pandemic) only considered four of the seven areas that were recognized as the basic seven “subareas” of rationality. The results of Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024) show that even non-COVID bills (so-called normal law-drafting) do not necessarily contain the basic information of impact assessments. In other words, several government bills have not, for example, assessed alternatives, identified negative impacts, presented quantitative assessments, or considered ex post evaluation. Finally, in addition to the shortcomings related to the law-drafting phase, the executed post-legislative measures have also been found to be inadequate. In Finland, ex post impact assessments have been carried out sporadically and unsystematically due to, among others, the lack of resources and tight schedules (Uusikylä et al., Reference Uusikylä, Keinänen, Vartiainen, Ervasti, Salminen, Kettinen, Lintinen, Köppä and Lindström2023). This has led to too few ex post evaluations being carried out.
When interpreted at the theoretical level, the absolute compliance of the evidence-based policy-making emphasizing the rational model limits the space for political interference and protects the public interest in decision-making (Carvalho et al., Reference Carvalho, Rondon and Cunha Marques2020, 2–3). However, it is worth remembering that in a democratic system, politicians are and should be the ones who make decisions. As Baldwin (Reference Baldwin, Baldwin, Cave and Lodge2010, 6) has stated, “RIAs are intended to inform decision-making, not to determine decisions or to substitute for political accountability.” According to Harrington and Morgenstern (Reference Harrington and Morgenstern2004, 3), impact assessments can, however, bring new discipline and rigor to the rule-making process and force decision-makers to think critically about the regulations they propose. Belcore and Ellig (Reference Belcore and Ellig2008, 17) aptly point out that two different decision-makers can, based on their values, reasonably disagree regarding what to do, even if they have the same information about the benefit–cost comparisons. Belcore and Ellig continue by stating that this should not mean that the information should be thrown away but rather make decisions based on supposition and hope. Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024, 4) emphasize the role of the rational model by arguing that the assessment of alternatives, identification of costs, operating ex post evaluations, and utilization of research findings should always be a part of high-quality law-drafting.
Therefore, it can be stated that despite the obvious political nature of decision-making, law-drafting must be required to follow the central principles of rationality. As COVID-19 restrictions are socially highly significant, the evidence-based assessment of alternatives should play a prominent role. Albeit the law-drafting environment has been different from so-called “normal law-making” due to time pressures and incomplete information, even in critical situations regulations should be justified by stating clearly why the proposed regulation would achieve more benefits than costs. The social significance of the COVID-19 restrictions also emphasizes the success of the implementation phase and the need for ex post evaluations, the success of which can be influenced already during the law-drafting phase.
3. Data and methods
This study examines the ex ante impact assessments and the planning of forthcoming implementation and ex post impact assessments in Finnish government bills related to COVID-19 restrictions. In Finland, the vast majority of legislative changes are based on government bills that are presented to the Finnish Parliament for discussion and decision-making. Government bills are documents that are prepared in ministries and include a law proposal together with a description of the impact assessments that have been conducted and planned during the law-drafting process. Therefore, government bills serve as the most important source of data when assessing the quality of regulatory impact assessments.
The data used in this study consist of all the COVID-19-related government bills from the years 2020 and 2021, in which restrictions to the fundamental rights laid down in the Finnish constitution were proposed. A total of 17 such government bills were identified. The government bills related to COVID-19 restrictions were chosen because of their societal significance. The restrictive measures were related to constitutional rights, such as equality, freedom of movement, electoral and participatory rights, and cultural rights, as well as the right to work and the freedom to engage in commercial activity.
The reason for choosing government bills specifically related to fundamental rights restrictions was to select those bills that are most socially significant from among all the bills related to COVID-19. The restrictions on fundamental rights thus serve as a criterion for identifying the most socially significant legislative projects. The fact that the selected government bills happen to be related to fundamental rights does little to change the criteria for a high-quality impact assessment from the perspective of rational law-drafting. For example, the freedom of trade involves restrictions for businesses, which, from an impact assessment perspective, means assessing business impacts. However, when fundamental and human rights are restricted in the Finnish legal system, the legislation must meet certain requirements (Viljanen, Reference Viljanen2001, 37–38), out of which the requirements of necessity and proportionality are the most central in the context of impact assessments. Necessity and proportionality are directly linked to the benefit and cost evaluation of the proposed laws. Thus, they do not actually bring additional requirements to rational law-drafting since the rational model itself includes the idea that only necessary and proportionate measures are rational (as explained in Section 2).
The central goal of the law proposals was to limit contact between people and thereby reduce coronavirus infections and slow down the spread of the disease. In the government bills, restrictions were proposed, among other things, on the opening hours and the number of customers in restaurants. Furthermore, legislative changes were proposed to enable directing restaurants and other businesses to close their premises temporarily in case the COVID-19 situation worsened. It was also suggested that schools could arrange their teaching, either in part or completely, as remote teaching. In addition, restrictions on moving outside of one’s residence, restrictions on prisoners’ rights (e.g. prisoner meetings and exit permits), and postponing the election date of municipal elections were proposed. One of the government bills was about health tests or COVID-19 certificates required from people arriving in Finland. In the same context, it was proposed that private businesses and event organizers could require their customers to have a health certificate (a so-called corona passport). Some of the government bills examined in this study are about continuing or reintroducing restrictions that had already been used previously during the pandemic.
The bills examined are presented in more detail in Table A1. In Table A1, there is a short description of the content of each government bill, and information about whether or not the bill was accepted in the Parliament. The identification code of each bill carries information about when the bill was submitted to Parliament: for example, the government bill “HE 25/2020 vp” is the 25th law proposal submitted in the parliamentary year 2020.
As can be seen from Table A1, all except one of the bills (HE 39/2021 vp) were accepted – and the one not accepted was actually canceled by the Government itself after receiving very critical feedback from the Constitutional Law Committee, stating that the proposed restrictions were not proportionate and necessary. Thus, all the bills that were voted on were accepted regardless of the quality of the impact assessments. This, however, is not surprising. It is worth noting that in the Finnish system, laws proposed by majority governments are accepted in the Parliament almost without exception, which was also the case during the COVID-19 pandemic. It is also typical for the Finnish system that amendments are made to government bills based on statements and reports of different committees in which government bills are discussed. For example, in 2020 and 2021, a total of 349 government bills unrelated to COVID-19 (so-called normal bills) were discussed in Parliament. Only two of those bills were rejected by the Parliament, and two were canceled by the government. A total of 149 of the bills were accepted unchanged and 196 accepted with amendments.
In the case of the COVID-19 restrictions, the Constitutional Law Committee had a major role in requiring amendments to government bills. Based on committee discussions, the vast majority of the bills examined in this study were amended before being accepted, and the one abovementioned bill was canceled altogether (see Table A1). When government bills are amended based on committee discussions in normal times, the changes are typically minor and technical. During the pandemic, however, the Constitutional Law Committee repeatedly stressed that the restrictions need to be necessary and proportionate, which led to the role of impact assessments growing during the crisis (see e.g. PeVL 14/2020 vp, 5; PeVL 44/2020 vp, 5–6; PeVL 10/2021 vp, 3–4; PeVL 13/2021 vp, 4). This emphasizes the significance of high-quality impact assessments themselves, and the importance of doing research about impact assessments in times of crises.
In this study, sections of the government bills are examined, in which the ex ante impact assessments and the planned implementation, as well as the ex post impact assessments, are described. The data, consisting of only 17 government bills, does not offer enough data for a meaningful quantitative analysis. Hence, the perspective in this study is qualitative, and the approach can be characterized as deductive qualitative content analysis (see e.g. Mayring, Reference Mayring2000; Graneheim et al., Reference Graneheim, Lindgren and Lundman2017, 30–31). In this context, “deductive” means that the data have been systematically reviewed using perspectives based on the theory of rationality in law-drafting (see Section 2). “Qualitative content analysis” refers here to an approach in which the content of the data has been observed, analyzed and interpreted from the perspective of auxiliary questions derived from the theoretical background (see the following paragraphs).
Although the approach is qualitative, numerical data are also presented to create an overview of the extent to which certain perspectives have been considered in the government bills. However, the aim of this study has not been to draw broadly generalizable quantitative conclusions from the data but instead to recognize both the best practices and the problem areas of legislative processes in times of crisis. In other words, the purpose has been to analyze in depth the most socially significant government bills. Within the framework of this analysis, it is, therefore, neither possible nor intended to attempt a ceteris paribus type setup of the impact of uncertainty brought by the pandemic in which COVID bills are compared to so-called normal government bills (as in Vartiainen & Keinänen, Reference Vartiainen and Keinänen2024), holding other factors affecting the quality of the impact assessments constant. Notwithstanding, the basic assumption is that many shortcomings exist in the impact assessments related to COVID-19 restrictions, considering that many shortcomings have been observed even in the impact assessments in so-called normal law-drafting both in Finland and internationally (see Section 2).
Based on the rational model, it is essential in the ex ante impact assessments to examine the benefits and costs of regulatory alternatives, to evaluate the effects of the different alternatives in a comparable manner (preferably quantitatively) and to base the assessments on a reliable and sufficient evidence base (see Section 2). In order to reach the objectives of the legislation, the best regulatory alternative in terms of the net benefits to society should then be chosen. The Finnish guidelines for law drafters have a lot in common with the rational model. Although the guidelines do not explicitly mention the net benefits to society, they do advise law drafters to write the government bills in a way that justifies – based on the impact assessments – why the objectives of the regulation are best achieved with the regulatory alternative that is suggested in the bill (see Ministry of Justice Finland). Thus, regarding the ex ante impact assessments, this study examines (in Section 4) the following: (i) whether the benefits and costs of different regulatory alternatives have been taken into consideration in the government bills, the nature of the relationship between the chosen regulatory option and the objectives of the government bill; (ii) whether the impacts have been estimated quantitatively as far as possible; and (iii) whether it is clear, based on impact assessments, why the chosen regulatory option is the best one for achieving the objectives of the legislation.
The role of ex post impact assessments in a rational process is, on the one hand, to ensure the functionality of the regulation and, on the other hand, to produce information for the future. As stated in Section 2, the ex post assessments are more likely to be conducted and their results to be utilized when they are planned comprehensively before the law comes into force. Keinänen et al., (Reference Keinänen, Heikkinen and Rinne2019) have, in their study on the quality of ex post impact assessments, examined the following questions: Have the ex post impact assessments been considered in the government bills at all? Is it clear who is responsible for conducting the ex post impact assessments? Has what will be assessed and when been defined? Has the data on which the ex post impact assessments will be based been defined? In this study, the same aspects have been examined in the government bills concerning COVID-19 restrictions (Section 5). In addition, how the implementation of the law has been described in the government bills has been examined. It is worth mentioning that the aim of this study is not to analyze how the implementations of the laws and ex post impact assessments were actually and eventually carried out. Instead, the aim is to analyze how these topics were considered in the law-drafting phase.
Some key findings are summarized in Table A2 in a rough yes/no form, visualizing whether certain attributes of high-quality impact assessments are present in the government bills or not. However, this rough summary does not tell the whole truth about the quality of the impact assessments. Many government bills mention, for example, different regulatory alternatives and their impacts. This finding appears on the yes/no scale as a sign of high-quality law-drafting. However, a more qualitative approach reveals that the alternatives are not necessarily compared in a way that would give the decision-makers enough information about why the chosen option is better than the others. Thus, the key findings in Table A2 should be read together with the qualitative analysis (Sections 4 and 5) in which more in-depth conclusions about the findings have been drawn.
Evaluating the quality of impact assessments using government bills as a source includes specific uncertainties. By examining government bills, it is only possible to get direct information about what kind of information is written in the bills. Thus, it is not necessarily possible to get a complete picture of how the impact assessments have been conducted in practice. The information presented in the government bills can, for example, be based on high-quality assessments even if the impact assessment process was not clearly explained or the sources of information were not mentioned in the text. In such cases, the impact assessments appear worse than they actually were. On the other hand, a government bill text that meets the ideals of rationality is not necessarily a guarantee of good impact assessment practices. On the contrary, sometimes certain issues in government bills can be mentioned only to mimic good practices, even though the issues mentioned have not affected the actual impact assessment process.
However, on the basis of the Finnish guidelines for law-drafting, government bills should always – alongside the results of the impact assessment – include information about the methods and the data that were used (Ministry of Justice Finland). As Vartiainen and Keinänen (Reference Vartiainen and Keinänen2024, 7) have stated, there are two possibilities for lacking information, both of which indicate poor law-drafting quality: either a certain aspect has not been considered in the law-drafting process, or it has been considered but not been noted in the government bill. In addition, government bills are the most important base for parliamentary discussion and decision-making, which is why the benefits and costs of different regulatory alternatives should be evident within government bills. The Finnish Constitutional Law Committee, for instance, has emphasized the role of government bills as a means to realize Parliament’s constitutional right to access information. The Committee has stated that Parliament and its committees must be able to trust that all the relevant information for decision-making is presented to them in government bills (Constitutional Law Committee, 2020, 3). Therefore, it is purposeful to evaluate impact assessments based on government bill documents.
4. The ex ante impact assessments
4.1. The different regulatory alternatives and their impacts
In all 17 government bills, the ex ante impact assessments regarding the proposed regulatory alternative are presented to varying extents. In some of the bills, however, the impacts of the proposed legislation or the impacts of the regulatory alternatives are only referred to very briefly. In these cases, it is mentioned that the impact assessments were described more thoroughly in previous government bills that laid out similar restrictions, and the reader is instructed to take a look at other government bills in which the impact assessments were described in more detail (HE 31/2021 vp, 5; HE 73/2021 vp, 9, 10; HE 6/2021 vp, 12; HE 139/2020 vp, 11).
Referring to previous government bills is, in these cases, justified because it was proposed that restrictions that were used earlier should be continued, reactivated or clarified. On the one hand, if it is assessed that continuing, reactivating or clarifying regulation does not have any independent impacts in relation to previous regulation, it can be well-grounded not to repeat what was already stated. This may help to keep the government bill concise and easy to approach – especially in a situation in which the earlier parliamentary discussions about similar restrictions are still fresh in memory. On the other hand, the pandemic situation sometimes changed very rapidly. Hence, it can be questioned whether the impacts of the same restrictions are actually similar if some months have passed between formulating the original restrictions and proposing their continuation. Therefore, even if the impacts were assessed earlier, it would be reasonable to assess the relevance of the previous impact assessments in that particular situation in which the restrictions are continued or reactivated.
In the majority of the government bills, some alternatives for the chosen regulation are presented (see Table A2). However, the option that nothing would be done is only assessed in five government bills (HE 86/2020 vp; HE 218/2020 vp; HE 39/2021 vp; HE 73/2021 vp; HE 93/2021 vp). Furthermore, in some of the bills, the reader is not able to find the impact assessments of the regulatory alternatives in one place. Instead, small pieces of impact assessments can, in these cases, be found in different parts of the text, which makes it difficult to get a full picture of the impacts of different legislative options (see e.g. HE 32/2021 vp and HE 33/2021 vp).
In most of the bills, several reasons are given to justify why the chosen regulatory alternative is considered to be better than other possible alternatives. Various positive effects accompany the chosen regulatory measures, whereas the other alternatives are presented with effects that are more negative. This kind of approach can be seen as a form of weighing benefits and costs. The credibility of these assessments, however, is often undermined by the fact that the analyses are not presented equally between different kinds of impacts (e.g. health effects and economic effects). Instead, each regulatory alternative is often presented with a couple of more or less random advantages and disadvantages that are very difficult to compare. This leads to the impression that the law-drafting process has not even attempted to justify the choice of regulatory options by weighing their benefits and cost (see e.g. HE 25/2020 vp, 6).
The government bill HE 72/2020 vp can also be observed as an example of the abovementioned phenomenon. In this bill, the government proposed that restaurants’ opening hours, alcohol serving times and the number of customers would be limited by law in order to reduce contact between people. The bill states that the restrictions would cause costs to businesses, and that the regulations would treat different kinds of businesses unequally. In the government bill, recommendations for restaurants to improve their safety and self-regulation of the restaurant industry are recognized as alternatives to the restrictions prescribed by law. Recommendations and self-regulation are seen as potentially effective measures: in the bill, it is stated that earlier COVID-19-related recommendations were followed comprehensively, and that it is also in the interests of the restaurants themselves to make customers feel safe. These alternatives are nonetheless rejected because – as stated in the bill – giving up law-based restrictions would pose too high a risk concerning the spread of the disease.
Simply stating that a risk is too high remains at an overly generalized level. In urgent circumstances, a full-scale benefit–cost analysis might be too much to ask. As Manski (Reference Manski2020, 342) has pointed out, data and modeling uncertainties limit the ability to predict the impacts of alternatives. Still, in order to make the benefits and costs comparable, the magnitude of the risk should be assessed at least to some degree. The bill also does not include assessments of the financial effects of the non-law-based regulatory alternative or the effects on the equality of different kinds of businesses. Furthermore, no quantitative estimates about the differences between different degrees of restrictions are given. Based on the government bill, it is thus difficult or impossible to gain any kind of understanding of whether the overall benefits of the chosen regulatory measures are greater than those of the presented alternatives.
On the other hand, COVID-19 restrictions have been justified with a precautionary principle, according to which precautionary measures can be implemented when scientific evidence on human health risks is uncertain, but the risks are high (see e.g. Aldred Reference Aldred2022). Based on the precautionary principle, restrictions could be prescribed already in a situation in which it was known that COVID-19 posed a severe human health risk, but in which the knowledge about the impacts of the planned restrictions was insufficient. This was the case especially in the early stages of the pandemic. Obviously, in a situation of urgency and incomplete information, it is not realistic to expect a comprehensive benefit–cost analysis. However, the Chancellor of Justice (the supreme guardian of the law in Finland, who, among other things, oversees the legality of the activities of the Finnish Government) has stated that regulation that enables restrictions of people’s rights should always be based on an appropriate and up-to-date assessment of the effects of the restrictions (Chancellor of Justice, 2023). Hence, the precautionary principle does not diminish the responsibility to carry out impact assessments in the most comprehensive way possible, comparing benefits and costs based on the information available at a certain moment.
4.2. The objectives of the regulation and the assessed impacts of the proposed laws
The main objective of all the government bills has been to reduce contact between people and to slow down the pandemic and protect people’s lives and health. Therefore, one would expect to find assessments, especially regarding the regulation’s impacts on the spread of the virus and people’s health and well-being. Surprisingly, the effects of the regulation on health and well-being are mentioned only in 10 out of 17 government bills (see Table A2).
Even when the impacts on people’s health are mentioned, the actual assessment of the impacts is not always described. Instead, it is often stated that assessing the health effects of the restrictions is difficult. In a government bill proposing restrictions on restaurants (HE 139/2020 vp), for instance, it is stated that although it is difficult to predict how any individual measures effect the epidemic, the overall health impacts of the restrictions are estimated to be positive. According to the rational model, however, the benefits and the costs of the restrictions should first be compared, and then a decision should be reached that not only has positive effects, but that the positive effects outweigh the costs or disadvantages. Of course, it is also a political question of values when the benefits are weighty enough to justify the costs. This, however, does not overrule the fact that in order to be able to make informed political choices, politicians need appropriate and accurate information about the effects. Especially when the politicians need to decide about restricting fundamental rights, it is extremely important that the impact assessments are executed as elaborately as possible. When restricting fundamental rights, the lawmakers themselves are also restricted by certain principles, such as the principles of necessity and proportionality. Evaluating the necessity and proportionality of the restrictions in Parliament is difficult if there is no information about the magnitude of the assessed effects in the government bills.
Another example of a government bill that does not clearly describe the relationship between the objectives of the regulation and the assessed impacts of the proposed regulatory measures is HE 73/2021 vp, regarding the continuation of special hygiene obligations in restaurants. In the bill, it is stated that the proposed regulation alone cannot affect the spreading of the coronavirus, but that such restrictions have clearly been necessary as a part of the means used, formed by recommendations, restrictions, and obligations (HE 73/2021 vp, 9). It is not clear from the text what this assessment is based on, and it is not mentioned in the bill why such restrictions have been necessary. In addition, the statement that the proposed regulation alone cannot affect the spread of the virus is peculiar in the sense that each restriction proposed should have an independent effect on the disease situation. The most likely meaning is that one type of restriction is not enough to slow down the spread of the virus on its own. However, it would be important to be able to get information about the effects of different restrictions separately to make rational decisions about the costs and benefits, and thereby whether or not the different measures are necessary and proportionate.
In the government bills regarding closing schools and offering remote teaching during the pandemic (HE 218/2020 vp and HE 93/2021 vp), comprehensive quantitative information is presented about how many infections there have been in the school environment before implementing restrictions on schools. However, as in the previously mentioned examples, the justification of the restrictions remains unclear since the amount of the infections in schools had been low, and in the latter case, there are references to studies showing that the risk of becoming infected after exposure to the virus in a school environment is significantly lower than in many other circumstances (HE 93/2021 vp, 13). Furthermore, both of the bills underline that closing schools would have severe adverse effects on the students’ well-being, learning results, and equality. On the one hand, it is positive that such studies have been utilized in impact assessments: openly presenting insecurities and opposing views gives credibility to the assessments. On the other hand, the bill should still clearly justify why such restrictions on schools are proportionate.
4.3. The accuracy level of the impact assessments
In 13 out of 17 government bills, some quantitative information about the impacts is presented (see Table A2). In most of these cases, quantitative information is offered about the assessed economic impacts and other kinds of impacts have mostly been assessed qualitatively. This, of course, is to be expected, as it is not easy or even possible to quantitatively assess all the effects, especially in the urgent circumstances caused by the pandemic. It is more convenient to assess quantitatively the financial effects on companies than, for example, the effects on people’s mental well-being. It is also emphasized in the guidelines for Finnish law drafters that quantitative information should be given, especially in the case of economic impacts (see Ministry of Justice Finland). However, even when it is not possible to quantify or monetize all of the impacts themselves, at least the size of the target group (the number of people, groups, businesses, etc.) that is affected by implementing or not implementing certain restrictions should be evaluated quantitatively. Only then will the decision-makers be able to assess the proportionality of the restrictions.
In the cases in which no quantitative impact assessments have been presented, it is stated in the bill that the restrictions proposed either do not have significant economic impacts (e.g. HE 44/2021 vp) or there is a reference to other government bills in which quantitative assessments about similar restrictions are presented (HE 31/2021 vp; HE 32/2021 vp; HE 73/2021 vp). Thus, it seems the law drafters have tried to give information about the quantitative effects of the regulation, at least when it was assessed that the proposed law would have economic impacts.
Only the government bills suggesting closing schools (HE 86/2020 vp; HE 218/2020 vp; HE 93/2021 vp) discussed the economic impacts of the legislation without any quantitative information. In the first of the abovementioned bills, remote teaching was proposed for the first time. In the latter two, the possibility for schools to switch to remote teaching was proposed to be continued. In the first bill, it is only stated that the proposed restrictions may have minor economic impacts, but that closing schools and adopting remote teaching might even cost less than classroom teaching. Based on that short statement, neither the direction (positive or negative) nor the magnitude of the impacts is presented. In the latter two government bills, the assessments of the economic impacts have been broadened by explaining how the costs of the regulation might be formed. Similar broadening of impact assessments can also be seen in some other government bills in which certain restrictions were reactivated (see e.g. HE 72/2020 vp and HE 6/2021 vp that propose restrictions for restaurants and later continue them). This indicates that some ministries might have been aware of the quality problems of law-drafting during the COVID-19 pandemic, and that the ministries later wanted to correct the shortcomings of the early-stage law-drafting.
Broadening the impact assessments when more information was available can be seen as a positive development. This kind of procedure can also be seen as necessary when the amount of information increases, and the precautionary principle can no longer be applied. Since the data consist of only 17 government bills, widely generalizable conclusions about temporal changes in the quality of law-drafting cannot be drawn. However, the qualitative analysis suggests that the impact assessments were not systematically elaborated, and the requirements of rationality were not always met, even when more information for impact assessments became available. This is clearly visible, for instance, in the cases in which the earlier impact assessments of similar restrictions were seen to be sufficient (see Section 4.1), although the rapidly changing pandemic situation might have required revising the impact assessments after time passed and the amount of information increased.
Within individual government bills, all the assessed impacts are not described with the same level of accuracy. This alone is not problematic: detailed and thorough impact assessment is a time-consuming and possibly also an expensive process, which is why it makes sense to use more resources for assessing those impacts that are thought to be the most significant. However, in the COVID-19-related government bills, it is not always clear why certain impacts have been assessed more thoroughly and in a more detailed manner than others.
The government bill HE 245/2020 vp, in which the responsibilities of social and healthcare services were redefined and in which new hygiene requirements were prescribed on businesses offering services for people, serves as an example of impact assessments that are not conducted considering the significance of the impacts. The bill states that the economic impacts of the law on social and healthcare services would be significant, and that the costs would be related to personnel, equipment acquisitions and protective gear. The economic impacts are described as “significant,” which is a relatively vague expression of quantity. When the impacts are expected to be significant – and not “minor,” for instance – it would have been appropriate to assess the impacts in a more precise manner. In addition, a list of only a few examples is given about the ways in which costs may be formed for social and healthcare services. It seems a thorough analysis of different impact mechanisms has not been conducted. On the one hand, the imprecision of the impact assessments can be explained by the fact that the information available during the pandemic was limited, and the measures to slow down the pandemic needed to be taken in a hurry. On the other hand, impact assessments should be produced even in times of crises – when possible – at a level of precision that makes it possible to compare the benefits and costs of regulations. This is especially important when assessing the most significant impacts of the regulation.
In the same government bill, the costs subjected to the cultural sector are presented by giving precise quantitative information about the current financial situation of the sector and the possible losses of (ticket) income. In addition, various other mechanisms through which the proposed legislation can affect the cultural sector are recognized and analyzed in the bill (HE 245/2020 vp, 38–42). Based on the bill, it is not clear why it would make sense to assess the costs for the cultural sector in a more precise way than those for social and healthcare services. It seems that resources for assessing the impacts were not allocated based on the significance of the impacts. Instead, the impacts seem to be described more thoroughly when the information for the assessments was easily accessible and did not require much research. This, of course, can also be seen in a positive light: it is positive that easily accessible information was utilized.
In certain cases, quantitative information has been included in government bills, but the relevance of this information from the perspective of assessing the impacts is unclear. In a government bill in which the temporary closure of restaurants was proposed (HE 25/2020 vp, 6), for instance, the numbers of the affected companies, their sales and employees are presented to the reader. This information, however, is not connected to the impact assessments. It would have been essential to describe, at least broadly, what kind of impacts the proposed regulation would have on these numbers: How many of the companies would suffer financial losses, and how much would these losses be? How many of the employees would be furloughed or left without a job? It seems, in some cases, quantitative information might have been included in the impact assessments to mimic fulfilling the guidelines for law-drafting, without the information having any actual role in the impact assessments.
4.4. The justifiability of the proposed restrictions in the light of the impact assessments
As has been presented in the previous sections, the relationship between the impact assessments and the key objectives of the proposed laws is often superficial and contains plenty of uncertainty. This is understandable in the sense that the time to prepare the bills was very limited, and, especially in the early stages of the pandemic, there was not much information on the effects of similar kinds of restrictions. Due to the urgency, the schedules and methods for consulting interest groups on the proposed legislation differed from the normal procedures in all of the examined regulatory projects, which also contributes to the lack of knowledge about the impacts on different groups of people (see also Vartiainen & Keinänen Reference Vartiainen and Keinänen2024). The uncertainty of the impact assessments is often brought out directly by stating, for instance, that the combined effects of different restrictions are difficult to predict. This kind of openness is desirable because it helps to promote transparency in decision-making.
Based on the government bills, the situational picture of the pandemic seems to have been relatively good, as the numbers of infections were monitored extensively during the pandemic. Collecting data from the initial situation is essential for both ex ante and ex post impact assessments. However, integrating this information into the impact assessment processes has not always been successful. Thus, based on the government bills, it is not necessarily clear what kind of effect the proposed restrictions would have on the initial situation described in the government bills.
Sometimes, however, even in very uncertain circumstances, the law drafters succeeded in offering information about the estimated impacts of the regulation on the situation in which the restrictions are proposed. The government bill HE 39/2021 vp can be mentioned as an example of a thorough presentation of the impacts of the proposed restrictions on people’s health and well-being. The bill depicts different scenarios in which either the existing restrictions are continued, some of the restrictions are discontinued, or new restrictions are introduced. The scenarios are based on the modelings of the Finnish Institute for Health and Welfare, and they include quantitative information on both the number of infections and the probabilities of the different scenarios.
The bill HE 39/2021 vp was presented to Parliament at the end of March 2021. The scenarios suggest, for instance, that discontinuing certain restrictions (e.g. the temporary closure of restaurants) would, by mid-May, lead to a peak of ∼1,200 new confirmed infections per day in the healthcare service area of Helsinki and Uusimaa. Continuing with current restrictions would have led to a peak of ∼800 confirmed infections per day, and regulating certain new restrictions would lower the number of confirmed infections temporarily from 600 to 400 new infections per day. Describing the impact assessments in a detailed manner gives decision-makers essential information concerning the potential impacts of the restrictions – information that could have enabled them to make their decision about whether or not to accept the proposed law.
Interestingly, the above-described Government bill – presenting very detailed information about the estimated impacts of the proposed legislation – was canceled because of the Constitutional Law Committee’s critical remarks. According to the Committee, the bill did not comply with the requirements of proportionality and necessity because it would have probably been possible to reach the desired effects with milder measures (PeVL 12/2021 vp, 9–13). Similar kinds of statements about necessity and proportionality were issued about several Government bills that needed to be modified before adopting the legislation (see Table A1). The requirements of proportionality and necessity, together with the remarks issued by the Constitutional Law Committee, underline the importance of comparing the impacts of the proposed restrictive measures with other measures with which the objectives of the legislation could possibly be reached. That is to say, without assessing rationally the impacts of different kinds of legislative alternatives, it is not even possible to conclude that the proposed legislation offers the least restrictive means for achieving the objectives of the legislation. Thus, it is not sufficient that one part of the impact assessments (e.g. the impacts of the proposed restrictions) is carefully conducted; other areas of the impact assessments (e.g. the impacts of legislative alternatives) need to be taken into consideration as well.
Surprisingly, the COVID-19-related government bills often overemphasize the negative side effects and the costs of proposed restrictions compared to the intended positive effects. For example, a government bill proposing a temporary closure of restaurants (HE 32/2021 vp) emphasizes that the negative effects of the restrictions would be very significant. Moreover, in a government bill about organizing children’s education as remote teaching (HE 93/2021 vp), the negative effects of remote teaching on the equality and well-being of the students are described in a detailed way based on various reports. In addition, it is stated that the benefits of classroom teaching are generally greater than the actual risk of coronavirus infection in schools. As shown by these examples, it sometimes remains unclear why certain restrictions were proposed, when the positive (health) effects were presented very vaguely, and the negative effects were explained extensively.
Being transparent about the side effects and costs of a regulation can be considered positive. Because of political pressure, impact assessments are sometimes conducted and described in a way that the benefits of the regulation are overemphasized, and the costs are downplayed (Bohne Reference Bohne and Verschuure2009, 65–66). Being open about negative effects can, on the one hand, be considered as a sign of good law-drafting, as it increases the credibility of the impact assessments and gives the politicians better grounds for forming their opinions. On the other hand, one can reasonably question the net benefits of the proposed restrictions if, for instance, there are no stronger justifications given for issuing new regulations than for not taking any regulatory measures. In these kinds of cases, one may get the impression that reacting to the problem – doing anything at all – has been considered more important than doing something effective and efficient. Based on the impact assessments, it should always be clear why the proposed regulation is better than other regulatory alternatives or the option that no government measures would be taken.
5. Planning the implementation of the laws and the ex post impact assessments
5.1. Describing implementation in the government bills
The implementation of the proposed legislation is described in the government bills at varying levels. In some of the bills, only the authorities responsible for implementing the restrictions are mentioned, whereas other bills describe what these authorities do in practice and with what resources. An example of a very abstract planning of implementation can be seen in the government bill HE 39/2021 vp, in which limitations on people’s right to move outside of their own residence were planned. The bill only stated that the Government is responsible for implementing the law and that the law would be implemented by a Government Decree introduced by the Government office.
The nature of the law proposal in question is such that the implementation of the restrictions requires a separate decree to be laid out if implementing the restrictions would, at some point, be necessary. This may partly explain why it was not considered essential to describe the practical implementation of the regulation while drafting the bill. On the other hand, it would be reasonable to plan the implementation already at the stage of preparing for such restrictions. If these restrictions had to be put into action, it is likely that the situation would be even more urgent than when drafting a government bill enabling such restrictions. In addition, planning the implementation already in the government bill would give Parliament the opportunity to assess the functionality and thus the necessity of the proposed regulation.
The government bills HE 131/2021 vp and HE 44/2021 vp can be mentioned as positive examples of more concrete planning of implementation. The former bill was (among other things) related to health certificates required from people arriving in Finland. The bill explains in detail which authorities are responsible for implementing different parts of the regulation. Furthermore, it even describes how the necessary changes to IT systems due to the regulatory changes would be made. The latter was related to restricting prisoners’ rights to have visitors or to have holidays outside of prison. The bill states that the Prison and Probation Service of Finland and the healthcare services for prisoners were involved in the preparation of the legislation, which is why they could already take measures in advance for the implementation of the restrictions. Based on the bill, the executive authorities had been involved in the preparation of the law proposal from the beginning: helping to assess the needs for such restrictions in the first place, giving statements while preparing the law and commenting on the law proposal in its final stages.
5.2. The planning of ex post impact assessments
In 12 out of 17 government bills, some information about the planned ex post impact assessments was presented (see Table A2). In all of these bills, the authority responsible for the assessments is named. In most of the cases (in eight law proposals: HE 72/2020 vp; HE 86/2020 vp; HE 139/2020 vp; HE 218/2020 vp; HE 6/2021 vp; HE 44/2021 vp; HE 73/2021 vp; HE 93/2021 vp), the assessments were planned to be carried out by the ministry that had drafted the government bill. In seven of these bills, another organization – a state office, city or municipality or a research organization – was mentioned to be cooperating with the ministry (HE 72/2020 vp; HE 86/2020 vp; HE 139/2020 vp; HE 218/2020 vp; HE 6/2021 vp; HE 73/2021 vp; HE 93/2021 vp).
In four of the bills, however, the authorities assessing the law are referred to in a very vague way (HE 245/2020 vp; HE 31/2021 vp; HE 39/2021 vp; HE 131/2021 vp). A bill suggesting that the Communicable Diseases Act should be temporarily changed for the special needs of the COVID-19 pandemic (HE 245/2020 vp) serves as an example. It is described in the bill that the implementation of the law, additional costs to authorities and effects on different areas of society are intended to be monitored on the Finnish Government level as a part of the ongoing work to prevent the spreading of the COVID-19 epidemic. The way of expressing the intention of monitoring (“are intended to be monitored”) does not create an image of concrete plans. In comparison, in another bill suggesting changes to the Communicable Diseases Act (HE 31/2021 vp), it is mentioned that the effects “will be monitored” on a regional level and on the Government level. In neither of the bills, however, are the organizations or institutions responsible for the monitoring identified: the expressions “on a regional level” or “on the Government level” do not clearly state who is actually responsible for the monitoring.
What will be monitored and when the monitoring will take place are specified in eight government bills (see Table A2). It is noteworthy that the focus seems to be on monitoring the pandemic situation and the combined effects of all the restrictions – not on evaluating the effects of the restrictions proposed in the individual government bills. It is very often mentioned, for example, that the authorities will produce a situational assessment of the epidemic both nationally and regionally at certain intervals, for example, once a week.
On the other hand, the bills concerning educational arrangements during the pandemic (HE 86/2020 vp, HE 218/2020 vp and HE 93/2021 vp) can be mentioned as examples of concentrating specifically on the effects of the regulation in question. In these cases, the ex post activities are connected to projects with different universities and research organizations. However, the projects do not assess the effects of the proposed remote learning arrangements on the pandemic situation. Instead, the purpose is to produce information about the effects on the well-being of children and families, educational equality and the teaching practices of schools. In this way, information is produced about the side effects of the regulation – which is also one of the interests of ex post assessment – but not about the intended effects of the restrictions. Furthermore, the ex post activities seem to have been planned to be more observing and monitoring rather than assessing or evaluating the actual impacts of single legislative actions.
Sources from which data for ex post assessments are collected are presented in six government bills. Out of the six bills, three are about remote teaching arrangements (HE 86/2020 vp; HE 218/2020 vp; HE 93/2021 vp) and another three are about restrictions on restaurants (HE 139/2020 vp; HE 6/2021 vp; HE 72/2020 vp). In the three bills concerning remote teaching arrangements, impact assessments and on what kind of information they will be based are described in detail. The bills describe that the Ministry of Education and Culture will regularly gather experiences and feedback from larger cities, and that an inquiry about the impacts will be addressed to all organizers of teaching. In addition, a committee was assigned to monitor equality and inequality in education during the pandemic. Furthermore, more precise assessments of different kinds of impacts (e.g. impacts on the well-being of children and families and on teaching and management practices in schools) are connected to several projects carried out by different universities and other research organizations.
In the bills proposing restrictions on restaurants, on the other hand, the sources of data are described with varying degrees of accuracy. Bills HE 139/2020 vp and HE 6/2021 vp stated in a very general way that monitoring will be based on epidemiological, medical, and functional indicators, which could refer to almost any kind of data on the spread of the virus. The same statement is also included in the bill HE 72/2020 vp, but it is followed by a concrete list of what is included in these indicators. The advantage of identifying the sources of information is that it makes it possible to evaluate the credibility of the data. However, it is worth noting that the more detailed bill (HE 72/2020 vp) has been given before the other ones. In such a situation, it may have been thought that there is no need to repeat what was mentioned in the earlier bill.
In summary, assessing the effectiveness of the proposed restrictions has not been discussed in the government bills at all. The ex post impact assessments have been mentioned in a large number of the government bills (12 out of 17, see Table A2). However, in all of these cases, the assessments focus on observing and monitoring the pandemic situation, the combined effects of different restrictive measures, or the side effects of the regulation. Thus, the intended effects of the proposed restrictions remain unassessed. Monitoring the pandemic situation and the combined effects of the restrictions is, of course, necessary: if the pandemic situation eases up, the conditions for limiting people’s fundamental rights may not be met anymore, in which case the restrictions must be removed. However, that kind of assessment does not convey whether a certain individual restrictive measure alone influences the disease situation (and if so, how much).
In order to gain information about the effectiveness of different regulatory alternatives, ex post impact assessments should focus, for example, on how much interpersonal contacts have been reduced by certain restrictions, and what kind of role specific restrictions have played in preventing the virus from spreading. Only when the ex post impact assessments are targeted at the causal effects of individual restrictive measures can the results of these impact assessments be fully utilized in the future if society faced a similar crisis. Evaluating the causal effects of individual restrictive measures is, of course, difficult in a situation in which several interventions are taking place at the same time. However, it can still be argued that legislators should have a more ambitious approach to ex post impact assessments instead of just monitoring the disease situation and the combined effects of different regulatory measures.
6. Conclusions
In this study, regulatory impact assessments related to COVID-19 restrictions were examined to gain information about successes and failures of law-drafting in times of crisis. The study aimed to learn from the past and to produce guidelines for future crises characterized by emergency conditions and a high level of uncertainty. The examination was based on the principles of rational law-drafting that created the criteria for the analysis. The qualitative content analysis of 17 Finnish Government bills showed that the rational model was rarely applied adequately, which led to problems with the justifiability of the restrictions.
It was not always clear why the proposed restrictions are more beneficial than harmful – let alone whether they would be optimal in terms of the overall societal benefits. This was due to the vagueness of ex ante impact assessments regarding the intended health effects. Furthermore, ex post impact assessments planned in the bills seem to produce information only about the general development of the pandemic and not about the effectiveness of individual restrictions. However, a positive aspect of the law-drafting processes was that the uncertainties and negative effects of the restrictions were brought out openly. Furthermore, it was positive that the impact assessments were sometimes elaborated when similar restrictions were proposed to be continued or reintroduced in later stages of the pandemic. Also, the situational image of the pandemic seemed to have been accurate, which was a necessary basis for conducting impact assessments.
The findings of the study are mostly consistent with observations made about the lack of proper regulatory impact assessments in Finland in the so-called normal times (e.g. Uusikylä et al., Reference Uusikylä, Keinänen, Vartiainen, Ervasti, Salminen, Kettinen, Lintinen, Köppä and Lindström2023; Vartiainen & Keinänen, Reference Vartiainen and Keinänen2024). Hence, the findings of this study can, to some extent, be applied to law-drafting in general, since similar aspects about assessing costs and benefits of different regulatory alternatives, as well as planning implementation and ex post impact assessments should be considered in “everyday law-drafting,” not just in times of crisis. However, there are some key lessons to be learned about impact assessments, especially in times of crisis:
-
1. The precautionary principle justifies actions in high-risk situations even when adequate information is not available. Impact assessments should still take a stand on why the proposed regulation is expected to bring more benefits than costs and is the preferred option compared to other alternatives. Justification of costs and benefits helps to ensure that decision-making is not arbitrary, even in urgent situations.
-
2. When more evidence is available and acting based on the precautionary principle is no longer necessary, the impact assessments should be conducted in more detail.
-
3. Attention should be paid to assessing the benefits and costs of different degrees of restrictions (comparing not implementing any restrictions to different scenarios with a more or less restricted society). This is essential for evaluating the necessity and proportionality of the restrictions.
-
4. When previously used restrictions are continued or reintroduced, the impact assessments should be reconducted to a relevant extent to evaluate the benefits and costs of the restrictions in the possibly changed situation.
-
5. In times of crisis, it is important to prepare for the implementation of the proposed regulation already in the law-drafting phase to ensure fast and efficient implementation. This can be facilitated by involving the authorities responsible for implementing the legislation in the law-drafting process.
-
6. Evidence base from ex post impact assessments is crucial in urgent situations in which there is not much time for adequate ex ante impact assessments. Ex post impact assessments should be planned as concretely as possible to ensure their implementation. If ex post evaluations are not eventually carried out, the same situation of an inadequate knowledge base will arise again during the next crisis.
-
7. Ex post impact assessments should, as much as is feasible, focus on evaluating the effects of individual restrictions to gain information about which restrictive measures are effective and which are not.
Finally, it could be beneficial if there were an institutional body responsible for conducting comprehensive ex post impact assessments of both individual restrictions and the cumulative effects of different regulatory measures (see also Bennear & Wiener, Reference Bennear and Wiener2021, 486). Establishing such an institution would help to ensure that ex post impact assessments would actually be performed, because assessing the impacts of crisis regulation would then be clearly designated to the responsibility of a certain entity. Ex post evaluations should also be facilitated by making the government responsible for overseeing the evaluation procedures.
Funding statement
The research is part of the Information Resilience in a Wicked Environment (IRWIN) project funded by the Academy of Finland. IRWIN is a 3-year project (2021–2023), where researchers from the University of Vaasa, the University of Eastern Finland, the Laurea University of Applied Sciences and the National Defence University seek to study Information Resilience in Complex Environments. IRWIN seeks to develop a participatory model of national preparedness in which decision-makers, civil society and the business sector work together to promote crisis preparedness.
Competing interests
The authors declare none.
A. Appendix
Finnish government bills related to COVID-19 restrictions from the years 2020–2021
Key findings summarized
Open access
