Analyzing argumentation patterns in political discourse for better policy design

ABSTRACT

A new policy is needed to manage Switzerland’s increasing urbanization and growing population. The second stage of the revision of the Swiss Spatial Planning Law, which has been ongoing since 2014, aims to create the legal framework to reduce or prevent the negative consequences of soil sealing, such as loss of biodiversity or urban sprawl. The revision process is characterized by substantial opposition among the actors involved; an acceptable draft revision is not conceivable. Using a structuring qualitative discourse analysis, we coded the consultation responses from the three consultation processes to date. We based our code system on acceptance factors existing in the literature and analyzed the content of the consultation responses over time and by actor. The results show that while a consensus on the instruments of the policy is emerging, there is great disagreement about the exact design and the resulting effects on the actors involved. The relative advantage of the policy and its compatibility with existing regulations are not sufficiently elaborated and presented in a comprehensible way. Ultimately, we identify several patterns of argumentation that should be considered by the policy-makers involved in the further revision, especially to address the critical arguments of the cantons and municipalities.

Highlights

• Points of conflict and consensus during policy design were identified using QDA.

• Structuring the content of the policy discourse can support consensus-building.

• Typical argumentation patterns for opposing or accepting a policy were identified.

• Growing consensus entails shifts of actor conflicts from mid- to low-level topics.

• Relative advantage of a policy solution is key to increasing its acceptability.

KEYWORDS:

• Policy design
• qualitative discourse analysis
• structuring content analysis
• acceptance factors
• argumentation patterns

1. Introduction

Worldwide, urbanization processes are putting soils under pressure (FAO, 2015). As a consequence, the continuing sealing of valuable soils due to growing settlements has irreversible impacts on the provision of life-sustaining services such as food production, clean drinking water or habitats for plants and animals (Blum, 2005). New policies are therefore needed to guide the sustainable use of soil resources (cf. Peake & Robb, 2021).

However, designing such policies is a major challenge because of complex conflicts of interest among the stakeholders involved (Ostrom, 2015; Peake & Robb, 2021; Haelg et al., 2020). The moderation of strongly opposing opinions (e.g. between two groups of actors) and the development of compromise solutions to which even disfavored actors may agree are therefore of great importance for successfully designing a policy. If such conflicts cannot be resolved, the policy design process may be further delayed, leading to unnecessary additional costs, legal uncertainties, and a lack of evidence for decision-making in individual contexts (Devine-Wright, 2011; Solly, 2021). Often though, the roles of the various actors in the policy design process remain vague, even though they play a crucial role in shaping the final policy outcome (Haelg et al., 2020). Therefore, gaining a better understanding of how consensus and dissent are formed among the involved actors in designing a policy poses a significant challenge.

Examining the characteristics of opposition, identifying its origins, and understanding the reasoning behind opposition can help resolve potential conflicts already during designing a policy (cf. Schmidt et al., 2019). To be able to better understand the specific arguments of the different stakeholders for the rejection or acceptance of a policy, policy-makers can analyze a political discourse early in the design process (cf. Sudau et al., 2022; Ellis et al., 2007). For example, consultation responses submitted by stakeholders during a public consultation phase can provide a robust, unbiased, and meaningful dataset suitable for systematically analyzing and understanding aspects that actors support or criticize (cf. Schmidt et al., 2019; Le Gouais & Wach, 2013) and to map the development of the discourse over time as well as shifts in conflict and consensus-building (cf. Bowen, 2009; Morgan, 2022). Fisher et al. (2013) have, for example, analyzed congressional testimonials on national climate change politics in the United States to identify points of consensus and conflict among the policy-makers involved. Cowell and Devine-Wright (2018) used public documents, such as transcripts of speeches or consultation responses, as a data source and analyzed them to examine how UK governments engaged the public in developing the national environmental policy. Both examples show how public documents in which stakeholders articulate their political perspectives and arguments can improve the understanding of political concerns.

Applying such methodological approaches, can help generating important knowledge to improve consensus-building among relevant actors during the design of a policy, increase the traceability of design decisions, and enhance the acceptability of the policy (cf. Levi & Spears, 1994). Simultaneously, the challenge to better understand the considerations of policy-makers that ultimately lead to the rejection or acceptance of a policy can be addressed. These considerations are made under the influence of abstract paradigms (Schmidt et al., 2019), which are difficult to measure. Such paradigms sharpen policy-makers’ understanding of problems and, in turn, form the basis for the development of explicit policy goals as well as the design of policy instruments (ibid.). Paradigms can be understood here as individual thought patterns of policy-makers, which are abstract because their genesis is unknown and influencing variables, such as the political interests of their own party, personal values and experiences, or moral aspirations of policy-makers are almost impossible to measure (Schmidt et al., 2019). As a result, the decision-making process of policy-makers tends to appear non-transparent, as it does not necessarily coincide with the arguments prevailing in public or political discourse, and policy decisions can therefore deviate from the interests of the relevant actors, such as municipalities or business associations (cf. Stadelmann-Steffen & Eder, 2021).

To address the highlighted challenges during the design of a policy, our study aims to delve into the following research questions: (RQ1) What arguments do actors involved in the process of designing a policy contribute, and how do these arguments influence consensus-building and the acceptability of a policy? (RQ2) What specific patterns or priorities in policy discourse can be identified that can help policy-makers to overcome political inertia (Schmidt et al., 2019)? (RQ3) How can the identified patterns in policy discourse be utilized by policy-makers to design more acceptable policies by addressing specific concerns, and thereby improve the effectiveness of policy elements?

In this contribution, the goal is to examine the complexity of designing a policy, particularly the role of stakeholders’ arguments and the influence of abstract paradigms on decision-making. Using a case study in which, during several unsuccessful revisions, the substantive debates between the actors involved have constantly shifted (cf. Schmidt et al., 2019; Jenkins-Smith et al., 2014), we demonstrate how consultation responses¹ can be used as a novel type of data to analyze the patterns of argumentation on key issues and controversial topics of the case study (cf. Morgan, 2022). Using qualitative discourse analysis (QDA), we also analyze the temporal and thematic shifts in the political discourse to improve the design process of a policy. This methodological approach, with a focus on factors that influence the acceptability of a policy, can help to better understand the dynamics of stakeholders’ arguments and those aspects of the revision process that have a negative influence on acceptance and thus are particularly obstructive for consensus-building.

2. Theoretical background

In order to address the challenges described, it is first necessary to briefly explain the process of designing a policy and, based on other studies, to show how the conflicting interests of the various actors involved in the design of a policy can be methodologically and conceptually analyzed. Subsequently, the selected case study and its suitability to the research will be introduced.

2.1. Policy design process

A policy is developed in an iterative process of a broad network in which policy-makers as well as other actors are actively involved, for instance because they implement the policy or provide the necessary resources (e.g. financial, human, informal) (Kammermann & Ingold, 2019). The development and detailed elaboration of a policy thus often becomes time-consuming and can take several years, culminating in a legislative proposal that can achieve the acceptance of a political majority. Consequently, policy-makers are generally concerned with gaining public support for their governing actions during the policy design process. They strive to anticipate the interests of the people or groups they represent politically (cf. Heyen, 2022) and take public opinion, preferences and attitudes into account. They prioritize the political feasibility of a policy over maximizing its problem-solving effectiveness to reduce the risk of lacking public support (Wicki et al., 2020). In addition, other factors are also crucial to a successful policy design process and should be appropriately considered in consensus-building (Head, 2008; Cairney, 2016; Horton & Brown, 2018). These include, for example, the ideological and political values of the actors involved or individual disadvantages they may face as a result of introducing a new policy (cf. Huijts et al., 2007).

In order to avoid rejection of a policy and to achieve the best possible trade-off between effectiveness and acceptance of a policy by the relevant actors, a policy is composed of different elements whose effects have different levels of impact on the actors, but also on the degree of goal achievement (policy packaging, Wicki et al., 2020). For the design of an acceptable policy, it is therefore necessary to configure the elements of a policy in such a way that the overarching policy goals can be achieved and at the same time the interests of the stakeholders involved can be adequately addressed.

A policy design process can be roughly divided into five phases: agenda setting, formulation, adoption, implementation, and evaluation of the policy (cf. Jenkins-Smith et al., 2014). In particular, in the formulation and adoption phases, the various actors (e.g. policy-maker, administration, economic and environmental associations or the population) can engage actively and represent their interests, for example, through comments, contributions to political debates, or as participants in relevant commissions or working groups (cf. Haelg et al., 2020). The analysis of consultation responses and written comments on political debates is a promising approach to study the process of designing a policy, especially in Western societies where such documents are often publicly available (cf. Morgan, 2022; Le Gouais & Wach, 2013).

2.2. Swiss case study – revision of the spatial planning act

Switzerland is a suitable context for analyzing a policy design process because of the formalized, participatory, and structured procedure of the legislative process (Haelg et al., 2020). It usually has 16 stages and is initiated when a new or revised legislation is proposed by the Federal Council², the Parliament or the cantons³. A preliminary draft is then prepared, which is commented on through office consultations and public consultation procedures before being revised by the responsible department and submitted to the Federal Council. During the iterative revision process, there are several opportunities for parliament, political parties and stakeholders to comment on, discuss and propose changes to the draft. The public consultation processes are one important way to contribute to the resolution of disagreements between stakeholders and formulating an acceptable law. After a successful final vote, there is the possibility of a facultative referendum⁴, in which the population has the final say. Finally, after a successful national vote, the new law is enacted by the Federal Council (Linder & Mueller, 2020, p. 173ff; FCh, 2023; cf. Swiss policy cycle, Appendix Figure C1).

We focus on the design of the Swiss Spatial Planning Law in Switzerland, which has been unsuccessfully revised several times in recent years. Spatial planning in Switzerland comprises the planning of space and the environment at the level of the federal state, the 26 cantons, and the over 2,000 municipalities. The necessary principles and objectives are regulated by the Spatial Planning Law (RPG). Since the introduction of the RPG in 1980, the law requires the sustainable management of soil, but this goal has not been achieved to the desired degree, particularly due to the rapid growth of the population, as well as changes in demography and lifestyle (EspaceSuisse, 2022; Mann, 2009). Furthermore, agriculture is undergoing a structural change and agricultural use is increasingly being abandoned and replaced by other uses. Today, of the approximately 590,000 buildings (ca. 20% of all buildings) located outside of building zones in Switzerland, about 190,000 buildings include residential units used as primary or secondary residences or for tourism purposes (cf. Mann, 2009). The overall goal of the revision of the RPG, which is ongoing since 2008, is to shape spatial development in Switzerland economically, ecologically and socially with enhanced sustainability (high-level abstraction goal; cf. Haelg et al., 2020).

The revision process was divided into two stages to account for the variety of the different conflicting topics to handle, as well as their prioritization (EspaceSuisse, 2022; APS, 2021; FOSD, 2021). The first part (RPG1) was approved by parliament in 2013 and has been in force by law since 2014. The main objective of RPG1 was to mobilize the potential for inward settlement development (inner development) by densifying settlement areas and, e.g. by converting brownfield sites. In this way, the excessive consumption of valuable cultivated land for settlement development is to be limited, and the high costs of developing new building zones with utility infrastructure are to be avoided (EspaceSuisse, 2022; cf. van Vliet et al., 2015). In the current second stage of the revision (RPG2), the complex set of regulations managing building outside the building zones is to be revised (cf. Solly, 2021). This goal is also supposed to reduce the loss of valuable cultivated land in Switzerland in order to meet society’s increasing need for large-scale natural landscapes, attractive local recreation areas and the protection of biodiversity (UREK-S, 2022; cf. Peake & Robb, 2021). In RPG2, eleven of the 39 articles of the RPG are planned to be modified, but several drafts of RPG2 were already rejected in the political process (APS, 2021; UREK-S, 2022). A chronological and thematic overview of the development of the topic of building outside the building zones in Switzerland’s spatial planning law can be found in the Appendix in Table C1. Figure 1 illustrates the aspects that will be regulated under RPG2 (FC, 2018; UREK-S, 2022). The main objectives and measures of the revision are structured by the level of abstraction following Haelg et al. (2020, p. 11).

Figure 1. Policy design elements of the RPG2 revision (FC, 2018; UREK-S, 2022) structured by policy focus and level of abstraction (cf. Haelg et al., 2020, p. 11).

The political process of the RPG2 revision has already been ongoing since 2014 and the substantive debates between the actors involved repeatedly shifted and new agendas or controversies emerged, while other aspects remained an enduring and entrenched substantive conflict between the actors (FOSD, 2018; UREK-S, 2022; cf. Schmidt et al., 2019; Jenkins-Smith et al., 2014; APS, 2021; cf. Table C1). The aggregated feedback from the actors’ consultation responses is documented in a results report (UREK-S, 2022). There, it is summarized that a majority of the actors are in favor of the general objectives of the draft law (necessity of the revision). However, it is rejected overall by the majority of actors because, for example, certain regulations are missing, concerns over the effectiveness of some of the newly proposed instruments are raised, or because of objections about the legal complexity of the draft that might shift the political responsibilities of the cantons (APS, 2021). Despite increasing acceptance of the draft law, continuing resistance and political inertia (Schmidt et al., 2019, p. 772) are hindering the revision of RPG2, requiring conflict resolution, also to avoid a possible referendum on the initiative of the Swiss population (cf. Solly, 2021).

3. Method and study design

Our workflow comprised three steps. (i) First, we compiled the consultation responses submitted in the three consultation phases of RPG2 and processed them for further analysis. (ii) In a next step, we coded and analyzed the consultation responses using QDA. (iii) In a final step, we analyzed the coded consultation responses according to different actor categories as well as in the course across the three consultation phases.

3.1. Qualitative discourse analysis – data and method

To examine the changes in the political discourse around the RPG2 revision over time, we focus on the analysis of the consultation responses submitted by the various actors during the consultation phases of the policy design process. During the three consultation processes in 2015 (N = 227), 2017 (N = 245) and 2021 (N = 188), a total of 660 consultation responses from a wide range of actors (e.g. cantons, political parties, interest groups) were submitted indicating the broad interest of the Swiss population (APS, 2021; FC, 2022; UREK-S, 2022). In principle, these are publicly and digitally available documents. They were made available to us in form of a pooled data set by the Federal Office for Spatial Development (FOSD). All duplicates, poorly scanned documents, and non-German language responses were excluded (N = 25) from the analysis beforehand for reasons of comparability of coding and due to deficiencies in the 2017 French-language revision documents (FOSD, 2018). The latter resulted in omitted or unstructured consultation responses from the French part of Switzerland, not suited for analysis. The adjusted dataset finally comprised 405 usable consultation responses. Hereof, we selected all consultation responses from actors who submitted a consultation response in each of the three consultation phases. Thus, a total of 156 consultation responses were coded and analyzed across three time steps, 2015, 2017, and 2021 (Table 1), which corresponds to approximately one quarter of all 660 consultation responses submitted (Bowen, 2009, p. 33ff.; cf. Schmidt et al., 2019). In order to prepare the consultation responses for coding and analysis, they were first assigned to the corresponding actor categories and (publication) year as individual PDF files and then imported into the maxQDA 22 software environment.

Table 1. Consultation responses used in the qualitative discourse analysis (absolute numbers; share in %).

Consultation responses	2015		2017		2021		Total
Received during consultation phases	227	(100%)	245	(100%)	188	(100%)	660	(100%)
Provided by FOSD for analysis	115	(51%)	179	(73%)	111	(59%)	405	(61%)
Analyzed	52	(23%)	52	(21%)	52	(28%)	156	(24%)

In the dataset provided by FOSD, the consultation responses were classified by the actor categories municipalities, cantons, parties, economy, non-invited and others. To ensure a better comparability of our analysis of actors with these categories, we adopted them as far as possible but reclassified the consultation responses in the categories non-invited and others into the categories spatial planning (e.g. universities, research companies, spatial planning actors) and environment (e.g. environmental protection associations). Only actors that could not be assigned to one of these two newly introduced categories remained under others (cf. Appendix, Table B1). We made this distinction, on the one hand, to be able to contrast the category economy (e.g. tourism, agriculture) with two further influential and relevant actors in spatial planning (cf. Perić et al., 2022). On the other hand, we expected that actors from spatial planning and environment would bring more critical perspectives or interests, or even opposing arguments into the revision process due to their professional expertise, exposed role in the initiation of the RPG2 revision, or strongly opposing interests, e.g. compared to agriculture (e.g. no building on green fields vs. need to expand residential infrastructure for touristic purposes).

For our analysis, we applied QDA as recommended by Morgan (2022), Mayring (2014), and Bowen (2009). The consultation responses analyzed were particularly suitable for such type of analysis, because they were elaborated by several people, represented a concerted and well-considered formulation of representative interests, and were also structured similarly in terms of content (cf. Morgan, 2022, p. 66; Arnautu & Dagenais, 2021; Le Gouais & Wach, 2013). The used data thus met the basic requirements of authenticity, credibility, representativeness, and meaning for the applicability of a QDA (cf. Morgan, 2022; Mayring, 2014; Bowen, 2009). After coding the documents, 20 of them (13%) were randomly selected and the codes were reviewed by an independent second researcher to ensure the validity and reliability of the coding (Baek et al., 2021, p. 4; Wang, 2011). In this iterative process, in addition to assessing the intercoder reliability, the codebook was also refined and improved (cf. Rädiker & Kuckartz, 2019, p. 287ff.; Wang, 2011).

3.2. Acceptance factors as categories for a code system

The forming of categories of arguments through qualitative analyzes of a policy discourse, allows for a comprehensive and contextual understanding of stakeholders’ decision-making processes in designing a policy. The studies by Haelg et al. (2020) and Sudau et al. (2022) are good examples and were both conducted in the Swiss context. While Haelg et al. (2020) based their analysis on the design process of a national instrument in the context of the Swiss energy strategy, Sudau et al. (2022) investigated the acceptance of different instruments of a spatial planning policy. This provided valuable insights into the reasoning of policy-makers (Haelg et al., 2020, p. 3) as well as the population (cf. Sudau et al., 2022), which are transferable to other contexts.

For example, the acceptance factors identified by Sudau et al. (2022) allow us to link our findings to existing knowledge about factors that influence the acceptance of a policy and can thus contribute to a better understanding of the paradigms under which policy-makers make decisions (cf. governance capacity; Dang et al., 2016, p. 1158). The framework of Haelg et al. (2020) could have been similarly useful, but it was based on comments from policy-makers (expert knowledge, different data type). We discuss these methodological choices critically later.

The framework of Sudau et al. (2022) describes the factors that influence the acceptance of policy instruments. In order for the statements and acceptance factors to be applicable to our analysis as categories of the code system, it was first necessary to slightly modify the naming of the acceptance factors and statements for our context (policy rather than instruments). The descriptions of the subcategories were discussed iteratively with a second, independent researcher in the context of the intercoder reliability check and revised according to the coded contents. An overview of the modifications made can be seen in Table A2.

Consequently, we applied the acceptance framework by Sudau et al. (2022) as a structured category system for coding (cf. Rädiker & Kuckartz, 2019; Mayring, 2014; Bowen, 2009), as it represents argumentation patterns used by a variety of stakeholders (e.g. public, interest groups). We used the five acceptance factors as the main code category and the statements each as a subcategory. We additionally distinguished whether an argument was positive or negative and respectively identified and coded the arguments expressed by the actors in the analyzed consultation responses. Depending on how the argument is contextually applied, it was assigned to the corresponding main category, subcategory, and direction (negative or positive) of our code system.

The code category design of the policy describes all arguments that refer to the design variables of a policy, such as the type of intervention, its configuration, anticipated costs, or the basis of assessment of the measures. We use the code category effect on actors for arguments with which an actor expresses how appropriate it perceives the effects of a policy on its own interests, the congruence of the goals and means of the policy with its political attitudes, or its sense of proportionality and fairness of the effects of the policy. We use the code category institutional embeddedness to code arguments about the necessity of the policy and its compatibility with existing policy regulations, and about potential shifts in political responsibility and authority caused by the policy. The code category effect of the policy describes arguments about the perceived efficiency and effectiveness and the relative advantage of the policy in terms of its purpose and compared to existing regulations and procedures. Finally, the code category understanding of the policy captures arguments about the overall credibility of the targeted effect as well as the basic understanding of the policy in terms of how it works and how it is presented. The five main categories of our code system are composed of a total of 19 subcategories, whose description is summarized in Table A1. A detailed description of our code system with corresponding examples of coding can be found in our codebook in Table A4 in the Appendix.

4. Results

In total, the 156 coded consultation responses spanned 1,206 pages (avg. = 7.73 pages; med. = 6 pages) and ranged from one to 31 pages in length. Across all consultation responses, we coded 4,215 arguments, for an average of 3.5 arguments per page and 27.02 arguments per consultation response. This includes some consultation responses with no arguments, compared to 137 arguments in the most argumentative consultation response (med. = 23 arguments). The intercoder reliability was tested to be > 75% (Appendix, Table A3).

Overall, significantly more negative than positive arguments were used in the consultation responses. In 2015, there were 188 positive (14%) versus 1,173 negative arguments, compared to 207 (19%) versus 904 in 2017 and 383 (22%) versus 1,360 in 2021, and overall 778 (18%) positive versus 3,437 negative arguments. The cantons (N = 18), as well as the actors categorized into spatial planning (N = 10) and economy (N = 12) accounted for the largest share of arguments. It is also noticeable that fewer arguments were used overall in 2017 than in 2015 and 2021 (Table B1).

4.1. Argumentation patterns over time

Figure 2 shows which statements and acceptance factors were particularly relevant in the respective consultation phase of the RPG2 revision and whether they were used by an actor as a negative or positive argument. For example, it is noticeable that almost exclusively arguments of the acceptance factors design of the policy and effect on actors were used for acceptance. In particular, arguments of the subcodes type of the intervention (Y-01) and actor’s political attitude (Y-18) were used. Over time, these arguments were mentioned with increasing frequency, which indicates that the acceptability of the RPG2 revision has increased overall from 2015 to 2021. Analogous to the higher acceptance of the type of intervention (Y-01), negative arguments of this subcode (N-01) were used decreasingly from 2015 to 2021. However, for some negative arguments an increase over time was also observed; for example, the subcodes configuration (N-02), basis for assessment (N-03) and allocation of funds (N-04). The fairness of the policy (N-06) also did not seem to improve from the perspective of the majority of actors, and neither did the compatibility of RPG2 with existing regulations (N-16). These findings indicate that until 2021 rather fundamental conflicts from the first draft revision in 2015 could be reduced or resolved and the discussion shifted more towards details and individual interests.

Figure 2. Change of importance of statements during the three consultation processes of the revision of the second stage of the Swiss Spatial Planning Law; shown as frequencies in all consultation responses of the respective year; left column = positive arguments, right column = negative arguments; acceptance factors indicated as colors: red = design of the policy, blue = effect on actors, orange = institutional embeddedness, purple = effect of the policy, green = understanding of the policy.

4.2. Argumentation patterns across actors

Figure 3 shows the differences of the patterns of argumentation of the seven actor categories over time. The RPG2 policy increasingly gained acceptance among parties and environmentalists, while the cantons still have objections, especially on issues related to the specific design of the policy. The strongest opposition, however, seems to come from municipalities, representatives of spatial planning as well as from the economy.

Figure 3. Change of importance of acceptance factors during the three consultation processes of RPG2 and differentiated by actor categories; shown as average frequencies per consultation response of the respective year; colors indicate the respective acceptance factor.

On the side of the municipalities, the resistance was mainly argued with the unclear and possibly high costs of the RPG2 implementation (N-05), as well as with the rejection of the type of intervention (N-01), the allocation of funds (N-04), and the thereby expected negative effects on the municipalities in Switzerland (N-19). Although the municipalities acknowledged the improved institutional embeddedness of the draft law over time, there were also doubts about the effectiveness of the policy (N-12) in terms of achieving its objectives. Over time, the cantons increasingly agreed with the type of intervention (Y-01), but their criticism of individual design features of the policy (N-02) and the basis for assessment (N-03) of individual interventions increased. This reinforces their doubts about the compatibility with existing regulations (N-16), the effectiveness of the policy (N-12) and the fairness of the policy (N-06). There was a growing consensus among the parties, and the overall improvement in the understanding of the policy was positively acknowledged. What needs to be resolved in this regard was the continuing skepticism regarding the effectiveness of the policy (N-12) and the shift in the power of institution (N-13) that would occur as a result of the RPG2 policy in the direction of the cantons, which would receive more power compared to the municipalities; this also coincides with the general opposition of the municipalities. The expansion of the cantons’ jurisdiction was also criticized by the representatives of spatial planning. This reinforces the lack of understanding of the functionality (N-22), calls into question the effectiveness of the policy (N-12) and would have to be solved by substantially revising the configuration (N-02) of the policy. As a result, this category of actors also had strong concerns about the fairness of the policy (N-06) and was increasingly opposed to the draft law. Actors from the environment category saw their interests positively taken into account over the course of the policy design process (actor’s political attitude, Y-18) and acknowledged the improved fairness of the policy (Y-06). However, the compatibility with existing regulations (N-16) and the complexity due to the increased number of regulations (N-20) would have to be improved. The latter should be coupled with improved explanations of the functionality (N-22) and representation (N-21) of the policy. The representatives of the economy showed an increasing acceptance of the types of intervention (Y-01) of the individual components of the RPG2 policy. In contrast, a particular need for improvement with regard to the configuration (N-02) and the basis for assessment (N-03) was expressed. Although agreement with their actor’s political attitude (Y-18) was increasing overall, they still saw themselves as being affected too negatively (effect on self-interest, N-19) and argued that the fairness of the policy (N-06) was not given.

4.3. Typical patterns of argumentation

Figure 4 illustrates the code map⁵ for the positive and negative arguments across all actors and consultation phases and is based on the distance matrix (Appendix Figure B2). Nine clusters (A-I) were formed, each containing arguments that are commonly used together and reflect typical patterns of argumentation by an actor in a consultation response. Thus, negative arguments on allocation of funds (N-04), credibility of development (N-17), or necessity of the policy (N-07) were made as stand-alone arguments against the RPG2 revision (Clusters G-I). In addition, if the effect on self-interest (Y-19) of an actor was considered positive, acceptance of the configuration (Y-02) of an intervention was often also expressed (Cluster F). Cluster D represents a collection of a total of eleven positive arguments. Here, it should be emphasized that positive arguments of the acceptance factor understanding of the policy were used in combination with acceptance of design characteristics of a policy (e.g. costs, effectiveness, basis for assessment). The clusters A-C represent three patterns of argumentation, which, on the one hand, occurred frequently and, on the other hand, were also used in combination with each other (grey links), which indicates a mutual reinforcement.

Figure 4. Code map of the subcodes (arguments from the consultation responses); size of the subcodes indicates frequency of usage; bold arguments are most frequent; colors indicate the respective main code category (acceptance factors, Sudau et al., 2022); grey links indicate >50 relations between the clusters; clusters A-I calculated using unweighted average linkage clustering method based on occurrence in document, N = 4,215 coded arguments.

Cluster A represents a pattern of argumentation that expresses particular concerns about the configuration (N-02), but approves the type of intervention (Y-01) and the agreement with the actor’s political attitude (Y-18), i.e. is not strongly opposed to the policy. Improvements in the explanations of the exact mode of operation (N-23) and adjustments to the basis for assessment (N-03) could well contribute to a consensus here (cf. Appendix Figure B1). Cluster B shows a pattern of argumentation that criticizes in particular the poor or missing compatibility with existing regulations (N-16), from which the fairness of the policy (N-06) and the effect (N-09) are questioned. The resistance of this pattern of argumentation, which is characterized by the acceptance factor institutional embeddedness, can be weakened by a better alignment of the policy draft with existing laws and a more precise explanation of the functionality (N-22) of the policy. Cluster C, by contrast, represents a pattern of argumentation that is strongly opposed to the policy. More fundamental aspects such as the type of intervention (N-01), undesirable shifts in the power of institution (N-13) and the resulting strong negative effects on self-interest (N-19) are the main arguments, often used in combination. In order to reach a consensus with the actors who argue with this pattern of argumentation, extensive revisions of the policy seem to be necessary.

5. Discussion

The analysis of different arguments in consultation responses allows us to identify relations and typical patterns of argumentation. Thereby, we offer a way to map the shift in content of the political discourse (differentiated by actor perspectives) during the design of a policy in a structured way and to identify the causal relationship of different arguments. With this approach, the participation of relevant actors in the design process of a policy can be additionally supported by addressing their concerns directly, which enhances the acceptance of design decisions (cf. collaborative governance, Kammermann & Ingold, 2019; Ellis et al., 2007). The approach provides an opportunity for policy-makers in other countries or contexts to determine the progress of consensus-building during the design of a policy and to better manage and monitor it, as well as to identify substantive conflict points between relevant actors that fundamentally stand in the way of the policy’s acceptability (RQ3; cf. Gärling & Loukopoulos, 2007). The direction (positive or negative) of an argument, as well as its importance in increasing the acceptability of the policy as a whole, can thus be taken into account during the design process (cf. Schmidt et al., 2019; Jenkins-Smith et al., 2014; Fisher et al., 2013).

Related to our case study, our findings confirm that the newest revision of the Spatial Planning Act can still not rely on the broad stakeholder support it needs (cf. UREK-S, 2022) and is an example of a conflicted and unsuccessful policy design process (cf. political inertia, Schmidt et al., 2019, p. 772). Compared to the first draft in 2015, some substantial resistance among actors could be resolved (e.g. introduction of a deconstruction payment, exemptions for structures outside the building zones such as cell towers, introduction of the stabilization target), but the negative arguments outweigh the positive and the discussion has developed around low-level goals and details of the design and compatibility with existing regulations (e.g. financing and payment modalities of the deconstruction payment). This illustrates that the challenge for legislators now is to reconcile the very diverse and argumentatively different criticisms and to resolve the conflicts (cf. Levi & Spears, 1994, p. 348f.; Stadelmann-Steffen & Eder, 2021).

Over time across the three consultation phases, it can be seen that the revision of the policy draft gained acceptance for the types of intervention (Y-01), and that concerns about the associated costs of the policy (N-05) were decreasing as a result (RQ1). There was also increasing recognition of the necessity of the revision (Y-07), but this may also be related to the long period of time over which the political discourse on RPG2 has spanned. These more fundamental improvements to the policy draft suggest that a consensus is emerging. However, the shift in the substantive discourse toward aspects of design, effectiveness, or compatibility with existing regulations also indicates where the need for improvement currently lies (RQ2). The argument relative advantage of the policy could play an important role for consensus-building. As a negative argument (N-08), it was most frequently mentioned with the three main arguments of Cluster C (N-01, N-13, N-19), and as a positive argument (Y-08) in Cluster D, it was frequently associated with good representation (Y-21) and explanations of functionality (Y-22) and the mode of operation (Y-23). A good understanding of the relative advantage of a policy seems to have a positive effect on its acceptance. Actors also aligned criteria such as fairness, concerns about self-interest, or power of institution, which is known from other contexts (cf. Peters et al., 2018). When revising the draft, the revised articles should thus be described in the explanatory report in a clear and understandable manner so that relevant actors can better assess the effectiveness, efficiency, and thus the advantages of the policy over previous regulations and its compatibility with other existing laws or procedures (cf. Wicki et al., 2020).

Our results show that an actor almost exclusively argues positively only if it sees its own interests or beliefs reflected in the law. Analogously, a poor understanding of the policy reinforces a negative attitude toward the draft law. This confirms previous research findings (cf. Huijts et al., 2007; Peters et al., 2018; Sudau et al., 2022) and shows that this acceptance factor is highly relevant in policy discourse. Criticism of insufficient revision documents (esp. 2017) or unclear formulations raise questions about the effect or implementability of the policy (Cluster B, Figure 4). The quality of the presentation and explanation of the policy draft also has an important influence on the perception of an actor’s own concerns as well as on the complexity of the policy draft (Cluster C; cf. RQ3).

Surprisingly, when we take a closer look at the arguments raised by the different actors, we find a strong resistance from the municipalities, which tends to increase over time. A solution to the conflict seems achievable only if their main points of criticism (financing, costs, negative effects on the self-interests of the municipalities) are addressed. Since they implement the laws on the ground, municipalities are one of the three constitutionally defined administrative levels of spatial planning in Switzerland and their opposing perspective should be given priority for the next revision steps (cf. Jenkins-Smith et al., 2014). At the same time, comprehensive revisions of the details of the design of the various elements would have to take place in order to increase acceptance of cantons, spatial planning and the economy. The strong opposition of spatial planning, in particular to shifts in competencies and concerns about fairness, must be seriously addressed. In addition, the institutional concerns of the environment actor group should be taken into account as a further aspect in the revision of the policy, particularly with regard to its integration into existing laws and established procedures (cf. Jenkins-Smith et al., 2014; RQ2).

Furthermore, the results of this study confirm the important role of the understanding of the policy and show how the patterns of argumentation in political discourse differ from those of a population (cf. Ellis et al., 2007; Stadelmann-Steffen & Eder, 2021). Institutional and design aspects have a higher importance for the acceptance of a policy in the design process. In contrast, the population evaluates a policy more on the basis of personal disadvantages or with the effects of an intervention in relation to its goals (cf. Sudau et al., 2022). This also suggests that even if a political consensus over RPG2 is reached, the draft law must also meet these requirements of the population, so that a referendum, for example, does not ultimately lead to a rejection of the law at the ballot box (cf. Solly, 2021; Levi & Spears, 1994). Compared to the patterns of argumentation of a broad population, arguments of the acceptance factor effect of the policy were rather disregarded in the context of the three consultation phases. This may be because the actual effects of a policy are difficult to assess by the actors at the higher level of abstraction, which leads to a subjective perception on the part of the actors. Here, an underused potential for the further design process of the policy could be to explicitly address, where possible, the increased effectiveness through the proposed changes and the impact on the different actors in the communication and explanation on the draft, e.g. impact assessments or test planning (cf. Peake & Robb, 2021; Devine-Wright, 2011; Peters et al., 2018; Stadelmann-Steffen & Eder, 2021; Sudau et al., 2022).

With regard to the methodology we used, it should be noted as a possible limitation that we only focused on the responses of stakeholders who participated in each of the three consultation phases (156 out of 660), particularly in order to understand the shift over time in the identified patterns of argumentation. However, this may omit insights from stakeholders who did not participate throughout or from those who expressed brief and concise acceptance of the draft, which could lead to a loss of information. Another limitation of our study is the exclusion of consultation responses in Italian and French. However, it should be noted that these would not have been included in our analysis anyway, as no French or Italian actors submitted consultation responses in all three phases. More detailed material such as meeting protocols, interviews or comments, e.g. from press releases or transcripts of political debates, could further help to identify party affiliations or strategic (mis)communication of individual actors (e.g. tactical arguments to enforce self-interests). This is not possible on the basis of written consultation responses. Moreover, the use of an existing code system can, on the one hand, lead to oversimplifying individual details of the discourse and, on the other hand, the positions represented within an actor’s category can vary considerably, which becomes more difficult to comprehend through categorization. In this context, it can be stated that we have decided in favor of the content structuring analysis based on an existing code system, since the context (management of the resource soil, building outside of building zones) of the acceptance framework by Sudau et al. (2022) fits very well to the RPG2 revision and thus the link to acceptance research can be established (cf. Mayring, 2014; Bowen, 2009).

While other concepts such as ACF (Jenkins-Smith et al., 2014) or design coalitions (Haelg et al., 2020) could have also been used as a code system, they are rather focusing on policy-makers beyond stakeholders and were developed on specific design characteristics of an instrument or used more complex data such as meeting protocols and interviews. The acceptance framework by Sudau et al. (2022) was less abstract and aimed to anticipate public opinion – an important consideration for policy-makers seeking to avoid a referendum. Finally, the demonstrated analysis of consultation responses provides a cost-effective and efficient approach to monitor consensus-building during the process of designing a policy.

However, further research is needed, e.g. to validate the suitability of our methodological approach in other research contexts (e.g. for the analysis of a different policy or in other countries). The approach could also be tested for its applicability to policy-makers themselves, in order to improve the information quality of summaries that are confusing or lack transparency in terms of content, such as the UREK-S (2022) results report. In this way, it would also be possible, for example, to comprehensibly examine specific aspects in detail, such as a particular conflict regarding a revised article or intervention, in order to derive technical proposals for solutions that take into account the different interests of the actors. In addition, individual actors and their development of argumentation and position could also be analyzed to avoid too strong generalizations of actors of one category.

6. Conclusion

Qualitative discourse analyses are suitable for structuring and analyzing the content of a political discourse and deriving recommendations for policy-makers. Our findings are particularly useful for prioritizing the content of the next steps of the RPG2 revision for successful consensus-building (Horton & Brown, 2018; cf. Levi & Spears, 1994). By experimentally applying Sudau et al.’s (2022) acceptance factors and statements as a code system to consultation responses around the revision of the spatial planning act in Switzerland, we found that institutional embeddedness, moderate and fair effects on the relevant actors, and the effectiveness of a policy are key factors for its acceptability. If a policy is insufficiently explained or incomprehensibly formulated, this has negative effects on its acceptance, the political discourse shifts to insignificant aspects and consensus-building is delayed. If impact assessments (e.g. through expert opinions or model calculations) are prepared for the main points of conflict of a policy early in the design process and attached to the explanatory documents, this can strengthen the problem awareness and the general understanding of a policy among the involved actors (cf. Heyen, 2022). In particular, for the RPG2 revision to be successful and to develop an acceptable draft law, the interests of municipalities should be sufficiently taken into account and compatibility with existing regulations and established procedures of policy-makers should be ensured to support the development of new policies for the effective soil resource management (cf. Peake & Robb, 2021; EspaceSuisse, 2022).

Acknowledgements

We would like to thank the Federal Office for Spatial Development for providing the consultation responses, Andreas Moser for his great support in checking the intercoder reliability, and Dr. Nicolas Schmid and Prof. Dr. David Kaufmann for their valuable input on policy change and the QDA methodology. Furthermore, we thank the reviewers for their constructive comments that significantly helped to improve the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Notes

1 For the remainder of the paper we use the term consultation response for the German word Stellungnahme, whichcaptures the essence of actors providing an opinion or position on a specific topic during a consultation phase.

2 The Federal Council of Switzerland is the highest executive authority of the country, constituted by seven members from (currently four) different political parties (cf. FCh, 2023).

3 Cantons are defined as federated states with varying degrees of autonomy and distinct political, economic, and cultural characteristics (e.g., cantonal parliament, main language German, French, or Italian), whereas municipalities are smaller administrative units within the cantons (cf. Linder & Mueller, 2020).

4 Switzerland’s direct-democratic system gives citizens the opportunity to actively express their political opinions. If a law is not in line with public opinion, a referendum can be initiated (if 50,000 valid signatures of Swiss voters against the enactment are collected within 100 days of its publication), which may result in the policy being overturned by a majority in a national vote (cf. FCh, 2023).

5 In QDA, a code relation measures the similarity between two coded arguments based on their frequency and co-occurrence in the examined dataset, calculated in form of a distance matrix. A distance of zero means that two arguments only occur together; the maximum distance means that two arguments never occur together. The distance matrix can then be visualized in a two-dimensional code map to form a conceptual network (Schmidt et al., 2019, p. 774f.) and to identify clusters and patterns of argumentation.

Appendix

A1 – Technical documentation of qualitative discourse analysis.

Table A1. Code system used in our study; adapted from Sudau et al. (2022)

Main category: Acceptance factors	ID	Subcategory: Statements	Description
Design of the Policy	01	Type of intervention	The type of the policy intervention is a useful measure to achieve the intended goal.
	02	Configuration	The configurations of the policy intervention is appropriate to achieve the intended goal.
	03	Basis for assessment	The basis for the assessment of the policy (e.g. for configuration or impact assessment) is useful and applicable.
	04	Allocation of funds	The funds allocated for the financing of the policy measure and their use are appropriate.
	05	Costs of implementation	The costs/resources needed for implementing the policy are appropriate.
Effect on Actors	06	Fairness of the policy	The allocation of policy impacts among the respective actors is fair.
	18	Actor’s political attitude	The policy intervention is necessary and useful to shape our space and strengthen settlement development within and outside of settlement areas.
	19	Effect on self-interest	The effect of the policy on the interests of the respective actor (stakeholder affection) is appropriate.
Institutional Embeddedness	07	Necessity of the policy	The policy intervention is necessary to shape our space and strengthen settlement development within and outside of settlement areas.
	13	Power of institution	The policy empowers the right actors in spatial planning to decide on how to use and manage space.
	16	Existing regulations	The policy is compatible with existing regulations and the interference with established processes is appropriate.
	20	Number of regulations	Introducing the policy increases the quantity of regulations in an appropriate amount.
Effect of the Policy	08	Relative advantage of the policy	Introducing the policy provides an advantage towards existing spatial planning instruments and policies.
	09	Effect	The intended effect can be achieved with this policy intervention and no undesirable other effects are expected.
	12	Effectiveness of the policy	This policy intervention can achieve the intended goal effectively and in a goal-oriented manner.
Understanding of the Policy	17	Credibility of development	The intended developments (goal achievement) are explained in a transparent and credible manner and the procedural implementation of the policy will be successful as presented.
	21	Representation	The manner in which the policy was presented was understandable and the documents provided contribute to a comprehensive understanding of the proposed measures.
	22	Functionality	The functionality of the policy intervention is understandable and clear.
	23	Mode of operation	The mode of operation of the policy intervention and its procedural sequences are understandable.

Table A2. Development of the code system based on the acceptance factors and statements of Sudau et al. (2022); shown are the names of the categories and subcategories as well as the changes made (bold) or deletions (-).

Acceptance Factor	ID	Old category name (cf. Sudau et al., 2022)	Revised formulation (Code system)	Old statement description (cf. Sudau et al., 2022)	Revised formulation (Subcode system)
Design of the Policy	01	Type of intervention	Type of intervention	Type of intervention: ⇉ type ⇇ is a useful measure to ensure the protection of good quality soil	The type of the policy intervention is a useful measure to achieve the intended goal.
	02	Configuration	Configuration	Configuration: The ⇉ instrument ⇇ is ⇉ type ⇇ which is the right strategy for spatial planning	The configurations of the policy intervention is appropriate to achieve the intended goal.
	03	Basis for assessment	Basis for assessment	Basis for assessment: soil quality as basis for the assessment where to build is useful	The basis for the assessment of the policy (e.g. for configuration or impact assessment) is useful and applicable.
	04	Allocation of funds	Allocation of funds	Allocation of funds: the generated funds are usefully applied	The funds allocated for the financing of the policy measure and their use are appropriate.
	05	Costs of implementation	Costs of implementation	Costs of implementation: the costs for implementing the ‘⇉ instrument ⇇’ are appropriate	The costs/resources needed for implementing the policy are appropriate.
Effect on Actors	06	Fairness of the instrument	Fairness of the policy	Fairness of the instrument: it is fair that landowners have to pay for the costs of the ⇉ instrument ⇇	The allocation of policy impacts among the respective actors is fair.
	18	Personal political attitude	Actor‘s political attitude	Personal political attitude: the ⇉ instrument ⇇ matches my personal political attitude	The policy intervention is necessary and useful to shape our space and strengthen settlement development within and outside of settlement areas.
	19	Effect on myself	Effect on self-interest	Effect on myself: the ⇉ instrument ⇇ is not affecting me myself as I am not owning land	The effect of the policy on the interests of the respective actor (stakeholder affection) is appropriate.
Institutional Embeddedness	07	Necessity of the instrument	Necessity of the policy	Necessity of the instrument: The ⇉ instrument ⇇ is necessary to shape our space and strengthen settlement development within current urbanized area as well as the protection of good quality soil	The policy intervention is necessary to shape our space and strengthen settlement development within and outside of settlement areas.
	13	Power of municipality	Power of institution	Power of municipality: the instrument empowers the municipality to decide on how to use and manage the soil	The policy empowers the right actors in spatial planning to decide on how to use and manage space.
	16	Existing regulations	Existing regulations	Existing regulations: the ⇉ instrument ⇇ is compatible with existing regulations	The policy is compatible with existing regulations and the interference with established processes is appropriate.
	20	Number of regulations	Number of regulations	Number of regulations: introducing the ‘⇉ instrument ⇇’ increases the quantity of regulations	Introducing the policy increases the quantity of regulations in an appropriate amount.
Effect of the Policy	08	Relative advantage of the instrument	Relative advantage of the policy	Relative advantage of the instrument: introducing the ‘⇉ instrument ⇇’ provides an advantage towards existing spatial planning instruments	Introducing the policy provides an advantage towards existing spatial planning instruments and policies.
	09	Effect	Effect	Effect: the intended effect of the instrument (reduction of urban sprawl and protection of good soil) is achieved	The type of the policy intervention is a useful measure to achieve the intended goal.
	10	Difference of sealed surface	–	Difference of sealed surface: the difference of sealed surface with and without the ⇉ instrument ⇇ is appropriate	–
	11	Difference of consumed soil quality	–	Difference of consumed soil quality: the difference of consumed soil quality with and without the ⇉ instrument ⇇ is appropriate	–
	12	Effectiveness of the instrument	Effectiveness of the policy	Effectiveness of the instrument: the ⇉ instrument ⇇ helps to ensure good quality settlement growth	This policy intervention can achieve the intended goal effectively and in a goal-oriented manner.
	14	Quantity of sealed surface	–	Quantity of sealed surface: implementing the ⇉ instrument ⇇, the quantity of sealed surface is appropriate	–
	15	Consumption of good quality soil	–	Consumption of good quality soil: the consumption of good quality soil is reduced applying the ⇉ instrument ⇇	–
Understanding of the Policy	17	Credibility of development	Credibility of development	Credibility of development: the shown developments are likely to happen given the implementation of the ⇉ instrument ⇇	The intended developments (goal achievement) are explained in a transparent and credible manner and the procedural implementation of the policy will be successful as presented.
	21	Representation	Representation	Representation: the way of representing the ⇉ instrument ⇇ was understandable	The manner in which the policy was presented was understandable and the documents provided contribute to a comprehensive understanding of the proposed measures.
	22	Functionality	Functionality	Functionality: the functionality of the ‘⇉ instrument ⇇’ was clear	The functionality of the policy intervention is understandable and clear.
	23	Mode of operation	Mode of operation	Mode of operation: the mode of operation of the ‘⇉ instrument ⇇’ was clear	The mode of operation of the policy intervention and its procedural sequences are understandable.

Table A3. Intercoder reliability by actor category; occurrence of same subcode in document (N = 20; one independent reliability coder).

Actor	N	Argument			Percentage
		Agreement	Non-Agreement	Sum
Municipalities	1	18	1	19	94.74
Cantons	6	63	25	88	71.59
Parties	2	20	0	20	100.00
Spatial Planning	3	39	14	53	73.58
Environment	2	33	5	38	86.84
Economy	4	49	23	72	68.06
Others	2	32	8	40	80.00
Total	20	254	76	330	82.12

Table A4. Codebook used; including examples for coded arguments for each subcode; Y = positive argument, N = negative argument.

Main category: Acceptance factors	ID	Subcategory: Statements	Description	Examples as coded in the consultation responses
Design of the Policy	01	Type of intervention	The type of the policy intervention is a useful measure to achieve the intended goal.	Y: Most cities welcome the fact that the stabilization goal is to be achieved through an incentive (deconstruction payment).
				Y: The planning approach in the non-building zone with uses to be compensated with the possibility to specifically address cantonal peculiarities (Art. 8c and 18bis) is welcomed in principle.
				N: The proposed new instrument of deconstruction payments as an incentive system to reduce the number of buildings outside the building zones should be questioned.
				N: The ‘Spatial Concept Switzerland’ is not a binding instrument, but merely an orientation aid. A change of this understanding, as defined at the beginning, is rejected.
	02	Configuration	The configurations of the policy intervention is appropriate to achieve the intended goal.	Y: We also find it appropriate that ancillary businesses without a close factual connection to agriculture are not included.
				Y: In particular, we also welcome the fact that the proof of long-term viability can be dispensed with, or that this requirement only has to be roughly examined according to Article 23b paragraph 5RPG.
				N: Agriculture is in conformity with the zoning in the agricultural zone, which is why it must be correctly excluded from the limitation of the number of buildings.
				N: However, the GREENS reject that such a reward is also paid out if the demolition takes place in favor of a replacement agricultural building.
	03	Basis for assessment	The basis for the assessment of the policy (e.g. for configuration or impact assessment) is useful and applicable.	Y: It is particularly welcomed that the restriction refers not only to the number of buildings outside the building zones, but also to the sealed areas.
				Y: Furthermore, the RZU supports the criteria according to which soil can now be allocated to a special agricultural zone (cf. Art. 16a para. 3).
				N: However, the chosen indicators must be measurable, based on existing and available data or at least on data that can be collected with reasonable effort.
				N: As already stated with regard to the stabilization goal, in Art. 24 g para. 1 let. a, the sum of the floor areas of the buildings should be decisive as a criterion for soil consumption.
	04	Allocation of funds	The funds allocated for the financing of the policy measure and their use are appropriate.	Y: -
				N: Financing via the added value levy weakens spatial planning within the building zones and is not possible due to a lack of revenue in individual cantons.
				N: According to the proposal of the UREK-S, the funds for the financing of demolition costs should primarily be provided by the cantons, namely from the revenues from the added value levy. EspaceSuisse is critical of this.
	05	Costs of implementation	The costs/resources needed for implementing the policy are appropriate.	Y: Even if the newly postulated acceptance requirement for the waiver of the restoration of the legally permissible state by a cantonal authority (para. 4) is associated with increased personnel requirements in most cantons, we can agree to the amendment of the law with a focus on the reduction of illegal constructions.
				N: The required comprehensive reporting ties up resources for the preparation and ongoing monitoring. Whether the resources and appropriate methods for monitoring are available is questionable.
				N: In addition, a considerable effort is to be expected for the spatial observation as well as the reporting, which cannot easily be afforded by smaller cantons.
Effect on Actors	06	Fairness of the policy	The allocation of policy impacts among the respective stakeholders is fair.	Y: We appreciate the work done by the various committees and note that the UREK-S has taken equal account of the interests of agriculture, landscape protection and the cantons.
				Y: Agriculture would not incur any additional costs as a result of the compensation, since according to the draft this is to be paid by the general public.
				N: In contrast, urban and agglomeration communities, which generate an above-average share of the total revenue from added value compensation, especially from rezoning and upgrades, will benefit only marginally from the effects of the deconstruction payment.
				N: However, agriculture as an economic sector should have the same opportunities everywhere in a unified economic area of Switzerland.
	18	Actor’s political attitude	The policy intervention is necessary and useful to shape our space and strengthen settlement development within and outside of settlement areas.	Y: We share the federal government’s direction as set out in the explanatory notes.
				Y: We take note of the addition and find the shift appropriate in terms of subsidiarity.
				N: The proposed regulation does not yet meet our requirements for a stable legal regulation.
				N: The initiators are very disappointed that the important components of the initiative (proposed Art. 24b 1 bis, Art. 24c para.2, 24d para.2b) are not addressed at all in this legislative amendment.
	19	Effect on self-interest	The effect of the policy on the interests of the respective stakeholder (stakeholder affection) is appropriate.	Y: HotellerieSuisse welcomes Article 8c, which reflects the concerns of the tourism industry.
				Y: The amendment is welcomed, as the handling of non-agricultural ancillary businesses without a close connection to agriculture has often been fraught with problems up to now.
				N: In addition, the canton of Nidwalden hardly has sufficient financial means available for financing the deconstruction payments via the added value levy due to the low revenues.
				N: The proposed procedure favors the infrastructure offices of the federal government and unduly restricts the competences of the cantons and ignores the interests of the landowners.
Institutional Embeddedness	07	Necessity of the policy	The policy intervention is necessary to shape our space and strengthen settlement development within and outside of settlement areas.	Y: In our opinion, it is urgently necessary to provide for effective penalties at the level of federal law for violations of the provisions of the RPG, especially as far as non-building areas are concerned.
				Y: The Canton of St. Gallen recognizes the necessity of the partial revision of the Federal Spatial Planning Law (SR 700; abbreviated as RPG) in the area of regulations on building outside building zones.
				N: The proposed procedure favors the federal infrastructure offices and unduly restricts the competencies of the cantons and overrides the interests of landowners.
				N: To date, it has remained unclear why it is mandatory for these residuals to be included in the Spatial Planning Law.
	13	Power of institution	The policy empowers the right actors in spatial planning to decide on how to use and manage space.	Y: The Government Council welcomes paragraph 4, according to which only the responsible cantonal authority can decide on the exceptional waiver of the restoration of the lawful state.
				Y: The Green Liberals welcome the fact that in the future the cantons can also be stricter than the RPG in the case of zone-compliant buildings in the agricultural zone and site-specific buildings.
				N: Unfortunately, we have to note that the draft, with all its reporting and monitoring mechanisms (Art. 24 g, Art. 38b, etc.), its provisions on demolition (e.g. Art. 25 Para. 3 and 4), the threatened consequences if the stabilization targets are not met (cf. Art. 38c), shows strong centralist patterns.
				N: That the cantons should be allowed to make autonomous restrictions here is extremely questionable and is categorically rejected by us. As a consequence, the cantons could pursue their own agricultural structural policy via spatial planning, although this is within the sovereignty of federal special legislation.
	16	Existing regulations	The policy is compatible with existing regulations and the interference with established processes is appropriate.	Y: We welcome in particular that the principle of separating building and non-building areas is maintained.
				Y: The addition to Art. 37 of the fact that temporary zoning can also be established for ‘infrastructures of national importance’, on the other hand, seems sensible (and in combination with the existing instruments also sufficient).
				N: The present wording would give general priority to agriculture over other uses that conform to zoning regulations and would thus contradict Art. 16 Para. 1. We assume that this is a legislative oversight in this regard.
				N: It must not be made easier for utility infrastructure to be relocated to sites outside the building zones, especially since there is already a licensing and court practice in this regard.
	20	Number of regulations	Introducing the policy increases the quantity of regulations in an appropriate amount.	Y: The high level of detail is justified, as the provisions regulate building outside building zones conclusively, i.e. without further legal specifications by cantons and municipalities.
				Y: The complexity has been reduced compared to the version of the Federal Council.
				N: With the numerous amendments, the impression is created that all the wishes of the federal administration have been incorporated into the law ‘textbook-like’, comprehensively and unfiltered, which is detrimental to the general law character of the RPG, but rather leads to unnecessary overregulation.
				N: A regulation of pre-concepts is to be dispensed with without substitution. The clear structure of ‘concept/property plan - structure plan – land use plan’ (Art. 13 RPG) should not be clarified unnecessarily and without corresponding gain.
Effect of the Policy	08	Relative advantage of the policy	Introducing the policy provides an advantage towards existing spatial planning instruments and policies.	Y: Thus, the planning approach and the consideration of interests offer the opportunity, through the holistic planning of a perimeter, to take better account of regional needs on the one hand and to achieve qualitatively better solutions than with the existing legislation on the other.
				Y: From the point of view of nature and landscape protection as well as biodiversity and its promotion, this paragraph in conjunction with Art. 10’ of the Environmental Protection Act represents an improvement.
				N: The new system forces the users to work with a concordance table. For the government council it is therefore questionable whether the effort of a revision is worthwhile. In any case, the added value of the simplification is rather low.
				N: The newly inserted paragraph 3 regarding sustainability assessment does not bring any added value compared to paragraph 1 (planning and voting obligation) in combination with RPV Art.47 (planning report).
	09	Effect	The intended effect can be achieved with this policy intervention and no undesirable other effects are expected.	Y: The VSLG only welcomes the fact that valuable cultivated land (crop rotation areas) will be better protected.
				Y: The deconstruction payment creates an incentive for farmers to replace old buildings with more efficient ones and to modernize their farms. This contributes to various agricultural, environmental and animal welfare concerns.
				N: Spatial planning stifles structural change instead of allowing it, as agricultural policy demands. Article 23f para. 2iand 3 RPG are to be deleted.
				N: It competes with agriculture, drives up land prices and leads in the long term to a compulsory zoning.
	12	Effectiveness of the policy	This policy intervention can achieve the intended goal effectively and in a goal-oriented manner.	Y: The principle of separation is better complied with by deleting this option.
				Y: This will create the legal basis for setting the right course in good time and accelerating the planning processes.
				N: It is questionable whether this incentive system alone will be sufficient to achieve the desired stabilization.
				N: If the topic of biodiversity is nevertheless addressed in the Spatial Planning Law, however, RZU believes that it is not enough to merely preserve biodiversity. Rather, it must be strengthened and promoted in a targeted manner.
Understanding of the Policy	17	Credibility of development	The intended developments (goal achievement) are explained in a transparent and credible manner and the procedural implementation of the policy will be successful as presented.	Y: -
				N: It should also be pointed out at this point that the French documents must be described as deficient.
				N: A purely declaratory addition to the provision is simply unnecessary and tends to create confusion and legal uncertainty.
				N: However, the draft still appears immature and its effects cannot be assessed. In its present form, its practicability is doubtful and would first have to be demonstrated.
	21	Representation	The manner in which the policy was presented was understandable and the documents provided contribute to a comprehensive understanding of the proposed measures.	Y: The consolidation of the various articles in the three sections creates clarity and supports enforcement.
				Y: The systematic structure of the provisions is improved overall, which should facilitate access for persons who have not previously dealt with the provisions.
				N: The draft submitted for consultation has fundamental shortcomings that cannot be remedied by a simple revision. Before another revision of the RPG can be started, a clear legislative concept must be developed
				N: In the context of RPG2, the amendments to the RPV were not submitted either to the present draft or to the consultation of 2014/2015. This is regrettable, because numerous regulations in the law are strongly dependent on the implementation at ordinance level for their effect.
	22	Functionality	The functionality of the policy intervention is understandable and clear.	Y: We welcome the better description of the zoning conformity of the agricultural zone and its delimitation.
				Y: The clear division of the content into chapters (general provisions / provisions for buildings and facilities conforming to the zone / exceptions) makes the law clearer, which we welcome because it facilitates enforcement.
				N: With regard to the provisions of the ordinance, which are not available, we expect that the demarcation between land-dependent and land-independent activities, or between core agriculture and the parts of the business that complement core agriculture, will be based on the current solution.
				N: Thus, the stabilization objective cannot be absolute. At least at the level of the ordinance, the handling of this area of tension as well as the consequences in case of failure to achieve the goal must be shown.
	23	Mode of operation	The mode of operation of the policy intervention and its procedural sequences are understandable.	Y: In particular, SBS supports the open conception of the article and the possibility mentioned in the explanatory report to combine the concerns of tourism promotion with those of landscape development.
				Y: The corresponding provision in the Spatial Planning Ordinance (Art. 3 RPV) would make sense to specify the methodology of the consideration in more detail.
				N: A specification of the type and quality of the structural changes is appropriate here, in order to preserve the building cultural values in the non-building area or to create them in the case of a replacement new building.
				N: In the case of the deconstruction payment, it should also be clarified whether such a payment is also made if there is already a legal, decreed or contractual removal obligation in individual cantons.

B1 – Further results from the qualitative discourse analysis.

Figure B1. Pearson correlation matrix all coded arguments (N = 4,215); −1 = negative correlation/never used in combination; 1 = positive correlation/always used in combination; bolded/boxed >0.5 and p < .0000.

Figure B2. Distance matrix all coded arguments (N = 4,215); Y-04 Allocation of funds and Y-17 Credibility of development haven’t been coded and are thus excluded from this matrix. Distances are without unit of measurements and calculated in maxQDA 22; 2D space is without unit of measurements and has a height of X = 628.96 and a width of Y = 843.83.

Table B1. Coded arguments (frequency), differentiated by year and actor categories.

Category	N	Arguments total			Arguments average (%)
		2015	2017	2021	2015	2017	2021
Municipalities	2	58	50	65	29.00	25.00	32.50
Cantons	18	348	411	627	19.33	22.83	34.83
Parties	3	64	82	66	21.33	27.33	22.00
Spatial Planning	10	376	206	458	37.60	20.60	45.80
Environment	3	104	118	116	34.67	39.33	38.67
Economy	12	321	170	266	26.75	14.17	22.17
Others	4	90	74	145	22.50	18.50	36.25
Sum	52	1,361	1,111	1,743
Average	7.43	194.43	158.71	249.00	27.31	23.97	33.17

C1 – Case study RPG2 – further information.

Figure C1. Legislative process (policy cycle) in Switzerland (Linder & Mueller, 2020, p. 174).

Table C1. Chronological and thematic summary of the most important political processes on the subject of building outside the building zones (based on APS, 2021).

Year	Type	Content
12/1991	Motion Zimmerli for a more flexible order of the in the agricultural zone (90.780)	Representative Zimmerli calls for a more flexible regulation of the use regulations applicable in the agricultural zone.
05/1992	Partial revision of the Spatial Planning Law (BRG 96.040)	In response to Zimmerli’s motion (90.780), the Federal Council forms a commission of experts from agriculture, nature and cultural heritage protection, spatial planning and the economy.
09/1994	Development of building sites outside the building zone (Mo. 93.3311)	A motion of Mr. Bisig demanded a revision of the RPG (Art. 24) in order to extend the exceptions for infrastructure facilities for the development of building zones also outside the building zone.
01/1996	Basic decision of the Federal Supreme Court concerning rural residential buildings outside the building zone	In a landmark decision, the Federal Supreme Court clarified the decisive aspects for the approval of agricultural residential buildings outside the building zone in accordance with the Spatial Planning Act (RPG) of 1980.
02/1999	Partial revision of the Spatial Planning Law	Following a referendum proposal, the Swiss population approved the revision of the Spatial Planning Law in a popular vote. The partial revision of the RPG was primarily aimed at relaxing the building and use regulations in the agricultural zones.
09/2000	Partially revised RPG and Spatial Planning Ordinance with regard to the conversion of buildings in agricultural zones	The Federal Council put the revised Spatial Planning Law and the application ordinance into force. The degree of deregulation by the ordinance was controversial between farmers and environmentalists until the very end. Finally, the Federal Council did not relax the legal provisions for the agricultural zone much.
05/2005	Partial revision of the Spatial Planning Law (BRG 05.084)	With the planned partial revision of the Spatial Planning Law, the Federal Council aims to improve secondary income opportunities for agriculture, facilitate buildings for livestock farming, and allow the addition of more storeys to and conversion of existing agricultural residential buildings for non-agricultural housing (e.g. tourism).
12/2008	General revision of the Spatial Planning Law failed in the consultation process	The main objective of the revision was to prevent uncontrolled urban sprawl. The majority of the 275 statements received were in favor of a partial revision of the 30-year-old RPG, but explicitly rejected a general revision. The majority of stakeholders criticized in particular the insufficient involvement of important stakeholders in the process as well as the simplification of the zone types. The back-zoning of oversized building zones without compensation (reserve zones) or the building obligation for owners of building plots also encountered strong resistance.
09/2009	Decision on partial revision	Due to the negative reactions, the government decided to abandon a general revision of the Spatial Planning Law and proposed a partial revision instead.
03/2013	Conversion of unused agricultural buildings (Mo. 11.3285)	A motion submitted in 2011 called for the relaxation of the legal provisions on the conversion of unused agricultural buildings. Since such buildings are already accessible, it would make sense to make it easier to convert them to residential purposes in order to avoid sealing additional cultivated land.
03/2013	First stage of the revised Spatial Planning Law (RPG1) passed in a popular vote	The result of the popular vote on the RPG1 revision was clearly positive (62.9% in favor with a participation rate of 46.5%). The aim of the first stage of the revision was, due to the limited space available in Switzerland, to use soil resources more economically. The principle of inner development was introduced accordingly with this revision. The partially revised RPG finally came into force in 2014.
12/2014	2nd stage of the revision of the Spatial Planning Law (BRG 18.077)	The Federal Council opened the consultation on the RPG2 revision, which aims in particular to improve the protection of cultivated land. Until the end, various stakeholders had called in vain for a postponement of the RPG2 revision. Concerns were directed especially at the resources available at the cantons and municipalities, as they were strongly engaged in the implementation of the RPG1 revision and its ordinance, which came into force in spring 2014.
12/2015	RPG2 failed in the consultation process	A large majority of the participants in the consultation rejected the proposed revision in this form. On the one hand, it was again criticized that there were too few resources available for the actors, as the implementation of RPG1 was time-consuming and in progress. On the other hand, the draft was found to be immature in terms of content, in particular because there was no overarching concept and no need for such a far-reaching revision. However, it was further criticized that adjustments were still required in the areas of soil protection and the protection of crop rotation areas as well as in building outside the building zones. As a consequence of the negative feedback, the Federal Council decided to work out a revised version together with cantons and municipalities and to limit itself to the issues of building outside the building zones as well as to spatial planning in the underground and in functional areas.
06/2017	Maximum use of the existing building volume outside the building zones (Mo. 16.3697)	With his motion, National Councilor Page aimed to expand the use of the existing building volume outside the building zones. This would be achieved by amending Article 24c of the RPG and existing restrictions in the law on the expansion of the building volume should be deleted.
06/2017	RPG2 failed in the consultation process	The second draft of the RPG2 revision, which the Federal Council sent out for consultation in June 2016, was aimed specifically at regulating building outside the building zones. More emphasis was placed on preserving the distinction between building and non-building zones and allowing the cantons more flexibility to take regional conditions into account. For this purpose, the so-called planning and compensation approach is now proposed. How exactly this compensation should work, however, was not further specified and should be addressed by the cantons. In particular, the planning and compensation approach was critically commented by most stakeholders. Many different arguments were raised against it, often revolving around time concerns as well as legal uncertainties regarding the practical implementation of the mechanism, and concerns about the enforceability and sophistication of the provisions.
10/2018	Formation of a working group to concretize the planning and compensation approach	The task of the working group consisting of representatives of various cantons and the BPUK was to concretize the mechanism and to examine the feasibility of possible designs of the approach. The results of this process were subsequently discussed with various other stakeholders from the fields of economy, environment and landscape planning as well as with representatives of the association of cities and municipalities and the Federal Supreme Court. According to the Federal Council’s report, which was presented at the end of October 2018, the stakeholders attested a ‘high degree of maturity’ to the results.
03/2019	Popular initiative to restrict building outside the building zones (landscape initiative)	The consortium behind the popular initiative aimed to ensure a clear distinction between building and non-building zones and to prevent the expansion of buildings outside the building zones in terms of quantity and surface area. The initiative wants to establish a strict separation of building zones and non-building zones in the Federal Constitution and demands that building outside the building zones be strictly regulated.
09/2019	Supplementary consultation on the planning and compensation approach	Once again, there was a strong majority of critical voices. The necessity of this proposal and the grip of the effectiveness of the mechanisms were criticized equally. The latter ultimately also led to the launch of a popular initiative at the beginning of 2019. In June 2019, the UREK-N finally rejected the RPG2 revision, and in December 2019, the National Council did so as well. Although the majority of the delegates again agreed on the need for action, the main point of contention was the introduction of the planning and compensation approach. Points of criticism were the high effort, the necessity as well as the feasibility of the approach.
05/2021	Re-revised draft of the RPG2 revision	Once again, the UREK-S submitted a revised draft of the controversial revision draft for consultation in spring 2021. The draft adopted the largely undisputed parts of the previous process, but fundamentally retained the planning and compensation approach and added provisions aimed at stabilizing the number of buildings located outside the building zones with the help of financial incentives for the removal of buildings. In particular, the object approach was removed as the basis of assessment of the planning and compensation approach in order to reduce the complexity of the draft. The Commission thus considered the central demands of the Landscape Initiative to have been incorporated and consequently presented its preliminary draft as an indirect counter-proposal to the popular initiative.
01/2022	RPG2 failed in the consultation process	The UREK-S submitted the preliminary draft for consultation for the third time by September 2021. In its results report, the commission summarized the feedback by stating that the direction of the draft was generally supported. Thirteen cantons, the center, FDP, SVP and GPS parties, as well as numerous organizations and associations indicated that they support the major goals of the draft. Three cantons as well as the SP and the GLP rejected the draft because they were of the opinion that it did not fulfill the objectives set. The responsible association of the landscape initiative also criticized that important concerns of the initiative had not been taken into account. The SP and the GLP as well as various conservation organizations also did not consider the draft to be a valid indirect counterproposal. The Greens, on the other hand, saw the draft as an admissible counter-proposal to the landscape initiative. The commission explained that the individual proposals for improvement would now be carefully examined and the draft revision adapted if necessary.