Making Socio-Technical Transition Pathways: The establishment of the Swedish Climate Policy Council, an Argumentative Policy Analytical case study

Making socio-technical transition pathways

The establishment of the Swedish Climate Policy Council, an Argumentative Policy Analytical case study

Eskil Engström

Department of Political Science Degree 30 HE credits

International master's programme in Environmental Social Science (120 HE credits)

Spring term 2021

Abstract

In recent years, several nations have adopted institutional framework laws, so-called Climate Change Acts (CCAs), as means to enforce Paris-compliant mitigation pathways. A key institutional feature to ensure policy stability and compliance with CCAs has been the establishment of

independent advisory bodies, tasked with advising on mitigation targets and policy instruments, as well as the, monitoring and evaluation of target attainment. These advisory bodies are endowed with a crucial role in the long-term evaluation and planning process: examining how the low-carbon transition pathways might be achieved. Calling attention to the question of how transition pathways should be conceived and approached, whether it is in 'bio-physical' (climate science), 'techno- economic' (technology assessment/economics) or 'socio-technical' (socio-technical transition field) terms. Recent studies have indicated that a socio-technical transitions is increasingly framed as a question of removing carbon energy from various practices and infrastructures, challenging the dominant techno-economic approach of emissions reductions using carbon-pricing instruments.

This thesis explores this challenge, drawing upon a case study of the establishment of the Swedish Climate Policy Council, by means of analyzing the process of institutionalization and how

transition pathways are (re)produced discursively through the practices of climate policy evaluation and planning. The main findings of this thesis is that a cross-party consensus behind the Swedish CCA was formed around institutionalizing a 'bio-physical’ mitigation pathway, monitored and safeguarded by the Council which could assign 'political embarrassment' to governments failing to comply with the interim and long-term GHG mitigation targets. Beyond this consensus, the

institutional design of the Council is the result of discursive struggles between actor-coalitions supporting techno-economic versus socio-technical transition pathways. However, the recently formed Council has come to challenge previously dominant techno-economic practices of

forecasting cost-efficient emissions reductions. This has been accomplished by introducing a novel socio-technical approach to climate policy evaluation: the backcasting of interrelated technological and institutional shifts believed to be necessary in bringing about a low-carbon transition or

transformation. Nevertheless, as this socio-technical practice primarily backcasts upon a number of key technological innovations, with limited changes to current industrial patterns of production and consumption, doubts are raised if this approach is to be considered as constitutive of transformative transition pathways.

Keywords

Transitions discourse, climate change acts, socio-technical transitions, transition pathways, argumentative discourse analysis, climate policy evaluation

Tables and Figures 4

Abbreviations and acronyms 4

1. Introduction 5

1.1 Long-term climate policy planning and evaluation 6

1.2 Research aims and questions 8

2. Literature review 9

2.1 Swedish climate policy discourse 9

2.2 Climate Change Acts 12

3. Theoretical framework 13

3.1 Evaluating transition pathways 13

3.2 Politics in low-carbon transitions 15

3.3 Processes of path-making 17

3.4 Argumentative Policy Analysis 18

3.5 Strategic practices 19

4. Method 20

4.1 Materials and delimitations 20

4.2 Operationalization and analysis 22

5. Results and discussion 22

5.1 The climate law storyline 23

5.2 Stabilizing a bio-physical pathway 25

5.3 Reproducing techno-economic pathways 28

5.4 Transforming socio-technical pathways 30

5.5 Producing the role of the Council 33

5.6 Climate policy evaluation practices 35

6. Conclusion 38

7. Appendix 41

7.1 List of interviews 41

7.2 Interview questionnaire 41

8. Bibliography 43

Tables and Figures

Table 1. Main Swedish climate policy actors

Table 2. Contrasting conceptions of transition pathways Table 3. Empirical material

Table 4. Analytical questions

Table 5. Summarizing pathway conceptions

Figure 1. Transition model

Figure 2. Argumentative Analytical Framework Figure 3. Operationalization

Abbreviations and acronyms

UNFCCC - United Nations Framework Convention on Climate Change COP21 - 2015 United Nations Climate Change Conference

CCA - Climate Change Act

Framework - Swedish Climate Policy Framework Council - Swedish Climate Policy Council

EOC - Environmental Objectives Committee CCC - UK Committee on Climate Change

SEPA - Swedish Environmental Protection Agency SEA - Swedish Energy Agency

NIER - Swedish National Institute of Economic Research NAO - Swedish National Audit Office

SOU - Swedish Government Official Reports SSNC - Swedish Society for Nature Conservation CSE - Confederation of Swedish Enterprise

STUC - Swedish Trade Union Confederation FSF - Federation of Swedish Farmers

CCS - Carbon Capture Storage

1. Introduction

The main objective of the 2015 Paris agreement (COP21), holding the global mean temperature 'well below 2°C ... and to pursue …1.5°C', is increasingly addressed by the adoption of national framework

laws establishing long-term climate policy targets, as well as the institutions and planning processes seen as necessary to achieve those targets (Averchenkova et al., 2020). In the preface to COP21, Christiana Figueres, executive secretary of the UNFCCC at the time, went so far as to argue that such national climate laws would constitute the 'solid foundation' of the Paris agreement (Climate Law Study 2015). With this turn to national climate governance, the Kyoto understanding of climate change as a global 'emission reduction problem' (calling for policy responses in the form of international agreements and carbon markets) has been complemented and possibly challenged by the emerging understanding of climate change transitions as a question of 'decarbonization' (Shaw 2011, p.743) The decarbonization objective which foregrounds low- carbon transitions as an issue of removing the use of carbon energy from various practices and

infrastructures in complex societal systems (Stripple & Bulkeley, 2019).

Since COP21, more than 70 countries have operationalized a decarbonization objective in the form of mitigation targets to reach net zero GHG emissions by 2050 (UNFCCC, 2019). A number of countries have also launched comprehensive national climate policy frameworks to meet these targets, all of them modelled on the 2008 UK Climate Change Act (Muinzer, 2019). Climate Change Acts (CCAs) which share the commonality of establishing binding long-term mitigation targets, policy planning mechanisms, and

monitoring and evaluation functions in the form of advisory bodies (ibid.). In sum, this institutional design of CCAs have been argued to constitute the most comprehensive approaches to national climate policy

governance to date (Averchenkova et al., 2020), drawing attention to how CCAs functions.

Much of the initial attention to CCAs was centered on the design of legally 'self-binding' long-term mitigation targets (Muinzer 2019). From such rationalistic and legalistic perspectives 'climate laws' could bind governments to deliver on GHG mitigation targets (Knaggård & Phil, 2015; McHarg, 2011). However, the validity of such explanations is questionable as no 'climate laws’ to date have been enforceable via prosecution or been linked to any direct penalties for breaches (Nash & Steurer, 2019). Alternatively, it has been argued from an institutionalist perspective that the mechanism by which governments are held accountable to CCAs is the risk of political embarrassment when governments fall under the scrutiny of independent advisory bodies (Averchenkova et al., 2018).

This institutionalist literature give prominence to independent advisory bodies as the key innovation of CCAs, since the former are authoritative institutions designed to be supportive of a low-carbon transition (Rosenbloom et al., 2019). However, as the status of these bodies is purely advisory, with most lacking formal authority, it has been argued from a discursive perspective that the prospects for holding governments accountable by assigning 'political embarrassment' is dependent upon the general climate policy discourse in a given country (Lockwood, 2013; Gillard, 2016). In the absence of a strong climate policy discourse, in

support GHG mitigation efforts, governments can simply disregard the advice from their advisory bodies, or even move to repeal CCAs (ibid.). Moreover, the primary role of these advisory bodies has been far from clear cut, spanning over at least three separate functions, (1) advice on policy targets (for instance in the form of five-year carbon budgets), (2) monitor and evaluate target attainment, and (3) advice on policy

instruments (McGregor et al. 2012). In light of this discussion, it is important to recognize the role of climate policy discourse in relation to the differing functions independent advisory bodies can take upon in their work of monitoring, evaluating and advising on low-carbon transitions.

1.1 Long-term climate policy planning and evaluation

Despite the significance accredited to independent advisory bodies, to what Sluisveld and colleagues (2017, p.89) calls “the governance of long-term policy planning and evaluation processes”, less attention has been directed at how these bodies are instituted and function in specific national contexts. In the case of Sweden, no peer-reviewed articles have been published with an explicit focus on the Swedish CCA, the Climate Policy Framework (henceforth the Framework), nor the advisory body, the Climate Policy Council (henceforth the Council). This thesis seeks to attend to this gap, drawing on unique empirical material, including seventeen in-depth interviews with high profile policy official, experts and politicians who are and have been central to the establishment of the Council, as well as, Council members and secretariat, combined with an extensive text-corpus of policy documents.

The period studied in this thesis can be separated in three general phases, firstly from an agenda setting phase, starting in 2012, when the Swedish Green Party first proposed the adoption of a CCA modeled upon the UK counterpart (Romson, 2012). The second phase initiates in 2014 as the newly elected Social Democratic-Green government commissions the Environmental Objectives Council (EOC), to inquire and negotiate a proposal for a Framework (including Council). The EOC which gathered climate policy experts and representatives from all parliamentary parties (except the Sweden Democrats), business, trade unions, and environmental organizations. The third phase starts in 2017, with the adoption of a Framework including (1) a net-zero emissions target by 2045 with intermediary targets for 2030 and 2040, (2) a so-called 'climate law' obligating governments to include climate reports in each budget bill and present a climate action plan every fourth year, (3) and a Climate Policy Council (Proposition [governmental bill] 2016/17:146)). The third phase commences as the Council is formed during 2018 and begins its work of monitoring and evaluating whether the government’s policies in aggregate are consistent with the climate targets, published in the form of three yearly reports (2018/2019/2020).

The Swedish Framework has been adopted with the stated aim of fulfilling the objectives of the Paris

Agreement (Lövin 2016/17:24, address 187). However, while both the Swedish and the UK CCAs have been framed as Paris-compliant, Andersson and colleagues (2020) have argued that the yearly GHG mitigation rates demanded under the respective CCAs is only about half of what would be required to be Paris- compliant. This calls into question, to what extent there exists a substantive commitment to a

decarbonization objective in Sweden. In point of fact, this objective is not a novelty in the Swedish policy context, where a cross-party consensus has existed, forwarding a vision of a low-carbon economy, since 2009 (Hildingsson & Khan, 2015). However, this vision has not prevented several actors from maintaining a

'Kyoto mindset,’ that is, attaining short term 'emissions reductions' by market-based approaches at the lowest cost (ibid., p.157). In this sense, contestations in the Swedish climate policy context have not been 'if', but rather 'how' to pursue a low-carbon transition. While this consensus has been reflected in a general commitment to Sweden's supposed role as an environmental frontrunner, it has been accompanied by differing interpretations of the policy-making requirements under ecological modernization (Lidskog &

Elander, 2012).

A split in ecological modernization discourse and practice has been recognized by an number of scholars, though they have tended to use different labels to refer to the two main Swedish climate policy discourses (Hildingsson, 2014a; Anshelm & Hultman, 2014; Tobin, 2015; Zannakis, 2009, 2015). For the sake of simplicity, in this thesis, these two discourses are defined as 'techno-economic', and ’socio-technical'

respectively. While 'techno-economic' discourse and practice have been recognized as dominant (Anshelm &

Hultman, 2014, Zannakis, 2009, 2015), the notion of low-carbon transitions as a question of decarbonizing complex societal systems has been identified as emergent amongst socio-technical actors (Hildingsson, 2014a/b). More recent studies have also indicated, but not fully explored, a possible shift between 2011 and 2016, during which the decarbonization objective has become increasingly institutionalized in policy practice (Kronsell et al., 2019; Hildingsson et al., 2019). Given the recorded dominance of techno-economic discourse and policy practice in the Swedish climate policy-making context, this possible shift presents a puzzling development, which warrants further exploration.

In the Swedish climate-policy-context it has also been recognized how climate policy inquires and

evaluations have been central to the contestations on the low-carbon transition (Knaggård, 2014). Exhibiting a pattern where actors within a dominant techno-economic coalition have commissioned economic

assessments, empathizing various economic sacrifices associated with a transition, while socio-technical actors, reversely, have commissioned assessments showcasing the opportunities associated of a low-carbon transition (Nilsson, 2005; Neby & Zannakis, 2020). This has demonstrated that the methods employed to evaluate policy pathways exist in dialogue with political contestation on approaches to a low-carbon transition.

Recently, Rosenbloom (2017), reviewing the expanding literature on low-carbon transitions, has identified a similar pattern, namely how different scientific fields of inquiry conceive of transitions either in terms of bio- physical, techno-economic or socio-technical pathways. Bio-physical pathways are tethered to climate science communities seeking to develop emission scenarios and/or models through which GHG emissions can be controlled and stabilized, setting the context for mitigation targets and GHG emissions reductions trajectories. Meanwhile, techno-economic pathways have focused on the progression of technological rollout and carbon pricing approaches as means to an efficient transition, equivalent with the 'emissions reductions' framing of low-carbon transitions. The socio-technical pathways compares to the 'decarbonization' concept, that is, of a complex interlocking socio-technological system, which demands a series of technological and institutional reforms to effectively un-lock transition pathways.

1.2 Research aims and questions

As an increasing number of countries have expressed the intention to embark on pathways toward net zero GHG emissions Rosenbloom’s (2017) analysis shows that it is important to distinguish between the different types of pathways being imagined, originating in diverse scientific disciplines and involving separate governance approaches. However, it is also vital to recognize that this use of pathways in policy practice and the transition literature commonly signifies an 'entity', a specific route or sequence of actions to bring about a low-carbon economy (Stripple & Bulkeley 2019). To the degree that political or other social process are considered in relation to these pathways it is commonly within a specific route or sequence of technological rollout, i.e. the expansion of renewable energy sources. In line with Stripple and Bulkeley (2019), an

alternative understanding is proposed in this thesis, namely that of transition pathways as a process: ”the way in which the path is forged” (p.53). This means conceptualizing the construction of pathways as a discursive process of evaluating and planning the actions involved in creating low-carbon futures, as a practice of 'path- making'.

The overarching aim of this thesis is to explore how the 'decarbonization' objective presents a possible shift in Swedish climate policy-planning and evaluation of transition pathways. This aim is pursued through a two part research question; (1) how did the public policy process shape the institutional design of the Swedish Climate Policy Council; and (2) how were the practices of climate policy evaluation institutionalized by the subsequent Council?

Studying the process of institutionalization of the Council gives opportunity to explore the science-policy- interface, where transition pathways are envisioned and shaped in the intersection between political visions and scientific practices of climate policy evaluation. This in addition to explore how the possible conceptual shift from techno-economic 'emissions reductions', towards socio-technical 'decarbonization' of complex systems, manifest in policy discourse and practices of long-term climate policy planning and evaluation. In that capacity, this case is theoretically motivated and suited to answer the calls emanating from the transition literature for a stronger attention to politics in low-carbon transitions (Gillard et.al 2016). Empirically, this case certainly has relevance to the domestic academic discussion upon Swedish climate policy-making.

Single case studies of Swedish climate policy-making are also frequently claimed to have international relevance, as Sweden’s record as an environmental frontrunner makes it an example other states could follow (e.g OECD 2014). Considering the high Swedish per capita carbon footprint of ten tons/CO2 per person and year (SEPA 2020), this justification could certainly be questioned. Nevertheless, the 'perception' alone of Sweden as a climate frontrunner might lead other states to consider Swedish climate governance as a model, giving cause for critical examinations.

However, before commencing empirical analysis (in section 5), the literature review proceeds to undertake a 'discursive' reading of Swedish climate politics (in section 2.1), theorize CCAs (in section 2.2) and link these discussions to practices of climate policy evaluation (in section 3.1). This, while developing a theoretical framework to conceptualize politics in low-carbon transitions/transformations, relating to how transition pathways are made and come into existence (in section 3.2-3.5).

2. Literature review

2.1 Swedish climate policy discourse

From the late 1980s Sweden took on the role as one of the earliest and most enthusiastic advocates of ecological modernization theory and discourse (Lundqvist 2004). As theory, ecological modernization centered on the supposedly 'autonomous' sphere of environmental reform, commonly by localizing empirical examples of environmental frontrunners (Mol et.al 2012, p.17). As discourse, ecological modernization, outlined in Hajer’s (1995;1996) seminal works, challenged the 'limits to growth' thesis, suggesting that material and economical flows could be decoupled if institutional arrangements were reformed to integrate environmental concerns. Hajer’s (1995;1996) work on ecological modernization also introduced an influential approach to the analysis of the role of discourse in policy change (see section 3.4 for in-depth discussion on argumentative discourse analysis). Asserting that policy shifts may well take place as new storylines emerge, drawing together components of a broader discourse into condensed and compelling short stories that frame policy problems, solutions, and responsibilities. These policy storylines can also have a coordinating role, in that actors who adhere to storylines come to form discourse coalitions by acting upon and adding to the development of the storylines. When a particular set off storylines becomes structured, that is, the dominant way of making sense of a 'policy-problem' in a particular context, they can give rise to policy change. Discourse institutionalization taking place as discourse coalitions moves to replace policy practices, so that they are conducted according to the prescribed actions of the storyline.

In their study of two main Swedish climate policy discourses, which arguably is the most comprehensive in recent years, Anshelm and Hultman (2014) draw on Hajer’s typology of two contending understandings of ecological modernization as 'technocratic project' or 'institutional learning. Although these two discourse both acknowledge the same systems of IPCC science, market-economy and liberal values associated with ecological modernization, they differentiate in their prescribed policy response to climate change. Either arguing that international agreements, market approaches and large-scale technical solutions are sufficient (technocratic project), or that a deeper institutional reconfiguration by means of reformism associated with the welfare state is required (institutional learning). The policy actor coalitions articulating these two discourses are in line with widely recognized actor relations in the Swedish (climate) policy-making context (Hildingsson 2014a; Tobin 2015; Zannakis 2009, 2015).

A shortcoming with Anshelm and Hultman's (2014) division is that it succumbs to a conflation of 'how' to realize the transition with 'what' climate targets to purse, presuming that actors expressing a technocratic project discourse necessarily oppose, while institutional learning actors necessarily advocate, for stringent climate policy targets. This has not always been the case. On account of these definitional issues, in combination with the plethora of differing definitions of a similar division line in ecological modernization

Table 1. Main Swedish climate policy actors

Technocratic project (techno-economic) Institutional learning (socio-technical) Actors Liberal-Conservative coalition government 2006-2014,

the Moderate Party (Moderaterna), the Center Party (Centerpartiet), the Liberal Party (Liberalerna), Christian Democratic Party (Kristdemokraterna), the organizations of trade and industry, and parts of the labour movement

Social Democratic-Green coalition government ruling from 2014-present, the Social Democratic party (Socialdemokraterna), the Green Party (Miljöpartiet), together with the Left Party (Vänsterpartiet), as well as key climate policy experts, scientists and environmental organizations

discourse (for instance see Christoff, 1996; Audet 2016; Hajer, 1996), is henceforth approached by using Rosenbloom's (2017) typology. This gives us a labelling of the two discourses on 'how' to pursue a transition either as techno-economic (technocratic project) or socio-technical (institutional learning), while adding that bio-physical discourses pertains to the question on 'what' mitigation targets to pursue, i.e. the speed, scope and depth of GHG mitigation. This definition also resembles Audet’s (2016), who found two main

contending transition discourses, as 'economic' and 'institutionalist' (socio), both based rotted in ecological modernization.

While all central Swedish policy actors generally have acknowledged the IPCCs (bio-physical) account of climate change, Knaggård (2014) has demonstrated that climate scientific arguments have had marginal influence on the formulation of national mitigation targets. Rather, Sweden’s first comprehensive climate bill in 2002 (Proposition 2001/02:55) was the result of a series of negotiations and inquiries, ending up in a three party agreement setting targets for the levels of GHG reductions deemed ”politically feasible” at that time (Knaggård 2014, p.32). Moreover, the question of 'what' mitigation targets pursue has been inartistically linked to the question of 'how' to achieve GHG emissions reductions, as the process of setting national mitigation targets in the early 2000s institutionalized a model of parliamentary committees/inquiries, simultaneously assessing 'feasible' climate policy instruments and targets (Nilsson 2005). In the period leading up to the adoption of the next comprehensive climate bill in 2009, these parliamentary committees and inquires would play a crucial role in producing climate policy evaluations and in defining what were deemed feasible policy pathways for reducing GHG emissions (ibid.). These inquiries also exhibited a pattern whereby a (techno-economic) coalition, including the Ministry of Industry and its agencies, routinely commissioned economic modelling assessments, which predicted economic sacrifices associated with climate policies (ibid.). Conversely, though, a (socio-technical) coalition, including the Ministry of Environment, its agencies, environmental NGOs, and scientists put forwards assessments which arrived at the opposite conclusion, that a carbon-free energy system was feasible and an economic opportunity (ibid.).

Zannakis (2015) suggests that this relative (bio-physical) consensus on Sweden’s role as a climate

frontrunner, matched by (techno-economic versus socio-technical) conflicts of how to purse this objective, is related to differing interpretations of the same collective action problem. The problem is formulated as follows: while climate change is global, Sweden’s pursuit of GHG mitigation is only effective if it is

matched by high-emission countries. From the socio-technical perspective, the solution has been formulated in terms of an opportunity storyline, asserting that if Sweden can display the opportunities for innovation and exports of climate friendly solutions, then others will soon follow the same pathway (ibid.). However, those advocating the techno-economic perspective have countered by articulating a sacrifice storyline, arguing that Sweden should indeed be a frontrunner, but this must be accomplished without jeopardizing Sweden’s position as a rich welfare society, because in that case, no country would want to follow Sweden’s example.

In this thesis it is proposed that the common understanding at the root of these divergent perspectives could be construed as a 'frontrunner dilemma', which has been institutionalized into the very practices of climate policy evaluation and planning.

This frontrunner dilemma would also manifest in relation to Sweden’s second comprehensive climate bill (Proposition 2008/09:162). The Liberal-Conservative government (2006-2014) adopted a 40% GHG mitigation target for 2020, allowing for a third of the reductions to be meet by 'flexible mechanisms,' that is purchasing GHG emissions reductions on an international market (Tobin, 2015)). This target was opposed by the Left-Green oppositional parties, which demanded a purely domestic 40% GHG reduction, arguing that flexible mechanisms would slow down the necessary domestic transition of the economy (ibid.). In response, the Liberal-Conservative government argued that climate change is global, making it irrelevant where emission reductions take place, but also that Sweden had already accomplished substantial GHG reductions and that further large domestic reductions would be unnecessarily costly and pose a risk to the

competitiveness of Swedish industries (Anshelm & Hultman 2014). Aside from these contestations, the 2009 climate bill also introduced a vision statement of net zero emissions by 2050, accepted by all parties,

demonstrating the broad political consensus attached to the overall objective of a low-carbon transition (Hildingsson & Khan 2015). However, this long-term vision was not institutionalized into Swedish climate governance, seeing as operationalized targets and policy strategies were lacking (ibid.). The absence of such a framework, stretching beyond the 2020 targets, was brought up as a key shortcoming by climate-policy- circles in a 2011 interview study (Kronsell et al. 2012).

During 2011, policy officers within the Swedish Environmental Protection Agency (SEPA) would effectively establish support for a public inquiry into a Swedish Low-carbon Roadmap (roadmap being synonymous to pathway in Swedish) (Hildingsson 2014b)). The long-standing conflict of how to conduct climate policy evaluation and planning (Nilsson 2005; Neby & Zannakis 2020) would also develop in relation to the use of cost-efficiency analysis in this inquiry. Hildingsson (2014b:5) interprets the government’s directive to the SEPA to conduct the inquiry 'after consultation' with the National Institute of Economic Research (NIER) and Swedish Energy Agency (SEA) as an insertion of discursive watchdogs into the inquiry, ’’to secure support for and institutionalize market-liberal climate policy discourse” (p.5) (for a definition of discursive watchdogs see section 3.5)). Nevertheless, the final report from the Roadmap inquiry would present two scenarios, stating that all major economic sectors (excluding agriculture) could feasibly be transitioned towards 2050, a key statement which established and structured a decarbonization storyline (ibid.) Hildingsson and Khan (2015) defined the discursive struggles during this time as ”contestations in policy- making circles over the means for governing the low-carbon transition, if climate governance merely entails regulating emission reductions by means of carbon pricing or whether it should also foster decarbonization by transforming societal structures and systems” (p.169).

Returning to climate-policy-circles in 2016, Kronsell, Khan and Hildinsson (2019) identified a marked change from 2011, where ”the planning horizon for climate policy has extended to address long-term goals of decarbonization both on state, industry, and business agendas” (p. 406). In passage, Kronsell and colleagues refer to the Framework as a example of how the decarbonization objective has been

institutionalized into long-term climate policy planning, attributing this shift to corporatist actor relations in Swedish climate policy-making, which have allowed for compromises between established societal interests through a series of policy dialogues (ibid.). Although this study gives indication to that a decarbonization storyline has become increasingly institutionalized in Swedish climate-policy-making, it does not offer a

detailed analysis of the Framework and Council, and relate this discussion to CCAs and long-term climate policy planning and evaluation mechanisms. Such a discussion will be entered upon in the following section (2.2).

2.2 Climate Change Acts

Since the adoption of the UK Climate Change Act (2008), CCAs have proliferated as a means for states to institutionalize Paris-compliant mitigation pathways (Averchenkova & Matikainen 2016), with six more European countries, including Sweden, having passed similar legislative frameworks. These frameworks include (1) long-term targets, (2) planning mechanisms in the form of carbon budgets or mitigation plans, and (3) independent advisory bodies (Nash & Steurer 2019). In all but one case, European CCAs have operationalized the low-carbon frame into net zero (or near to zero) targets around 2050, passing through national parliaments with broad political support (ibid.). This raises the question whether or not

comprehensive institutional design and widespread consensus behind CCAs testify to a stable institutionalization of the decarbonization objective.

Rationalist explanations, that climate laws enforce 'self-binding' (bio-physical) mitigation pathways, while governments have the democratic control over determining how to pursue GHG reductions (Knaggård &

Phil 2015; McHarg 2011), have show to be invalid, as CCAs have not been enforceable via prosecution or linked to any direct penalties (Nash & Steurer 2019). Instead, it has been suggested at the outset, by those promoting institutionalist perspectives, that it is the larger institutional framework, instilling climate policy with institutional stability, that is the primary innovation of CCAs (Rosenbloom et al. 2019). Climate policy stability has emerged as a central concern both in policy discourse and socio-technical transition literature as a precondition for (1) enabling low-carbon investments and (2) a continuous, durable, and self-reinforcing institutional support for a decarbonization (ibid.). The first precondition, predominantly voiced by parts of the business community, was an influential argument for the establishment of the UK CCA (Lockwood 2013). However, the UK CCA has not produced major changes to fossil investments patterns (ibid.), and it has been highlighted that business interest may call for climate policy stability as means of resisting more stringent regulation and the disruptive changes entailed by a decarbonization process (Rosenbloom et al.

2019). The second precondition is that path-dependency theories predict that stable action over time can be self-reinforcing, as each step in a certain direction makes it increasingly hard to change course (Levin et al.

2012). From this perspective, the creation of independent advisory bodies is the key novelty of CCAs, as they form authoritative institutions that are purposely supportive of a transition, in contrast to other governmental institutions defending business as usual (Rosenbloom et al. 2019). However, Lockwood (2013), analyzing the institutional stability of the UK CCA claims that it remains at risk of reversal. Since the UKs Committee on Climate Change (CCC) authorities are not formal, but based on reputation, their

durability is sensitive to ideational hostility towards climate policy. Such hostility risks undermining or bringing about a reversal of the Act, as the UK CCA has not succeeded in producing a “discursive

transformation,” changing preferences and identities amongst groups that are antagonistic towards climate policy in general (ibid, p.1346).

It has been suggested that it is the risk of political embarrassment that will stabilize CCAs, as governments are under scrutiny of independent advisory bodies (Averchenkova et al. 2018). However, as Lockwood (2013) argues, the prospects for political embarrassment are dependent on the discourses that infuse the general climate policy debate. Gillard (2016) draws a similar conclusion, finding that during 2010-2015 the UK CCC continuously reemphasized the argument from the Stern Review, stressing cost-effectiveness of early mitigation efforts (ibid.) However, the message of the UK CCC was effectively suppressed towards 2015 by dominant discourses on austerity and neoliberal governance, resulting in a form of climate policy skepticism (Gillard 2016). In conclusion, while the UK CCC rose to the expectations of defending a decarbonization objective, successive UK governments were evidently not embarrassed for their

prioritization of short term savings and their failure to deliver on long-term targets. Upon this conclusion, Gillard (2016) argues that the consensual strategies of CCAs are used to signal a safely institutionalized decarbonization objective, while hiding discursive contestation amongst policymakers and inadequate policy-making practices.

Moreover, the framing of CCAs as institutionalizing Paris-compliant mitigation pathways (Averchenkova &

Matikainen 2016) is questionable. Andersson and colleagues (2020) suggest that a fair redistribution of the remaining global carbon budgets to achieve the demands of the Paris agreement, show that the UK and Sweden’s mitigation rates should be at minimum 10% and 12% per annum respectively. Despite these findings, the total carbon budgets implicit in the current CCAs under the (bio-physical) mitigation pathways imply GHG reductions of only about 5% per annum. The implication is that the Swedish CCA constitutes a mitigation pathway far less demanding in terms of climate policy than a Paris-compliant mitigation pathway ought to be (ibid.). This further underscores the key function of advisory bodies, in bringing forward policy targets, evaluations and advice on policy instruments, composing transition pathways which are Paris- compliant. However, the literature on CCAs has so far failed to recognize that transition pathways represents separate discourses, entities and processes, and so delineate different practices of evaluating and planning how low-carbon transitions can take place. This shortcoming is addressed in the following theoretical section (3.1).

3. Theoretical framework

3.1 Evaluating transition pathways

In one of the previous sections (2.1) Rosenbloom’s typology of pathways was employed to analyse Swedish climate policy discourses and practices. Pathways as discourse relate to the so-called policy-science-interface (Wesselink et al., 2013), where expert knowledge is interwoven with political consideration, into what Huitema and colleagues (2011, p.195) have called the 'politics of environmental evaluation'. The strength of the pathway concept, however, is that it also refers to 'entities', such as transition strategies or plans produced by researchers, key intergovernmental bodies and state agencies (Wiseman, et al. 2013). Moreover, below in section (3.3), a third use of pathways will be proposed, as the process of 'path-making' through which plans or strategies are made and come in to existence. This three-part use of the pathways concept, as (1)

discourse, (2) entity, and (3) process, certainly has been a source of confusion, blunting its analytical value.

Nevertheless, keeping these definitional differences in mind, pathways are useful concepts for the very same reason, since they cover a field of transition theories-in-use, where transition actors and transition scholars affect each other (Audet, 2014). Rosenbloom (2017) identifies three such fields where this interaction takes place, giving rise to separate approaches, for instance adopted by climate change advisory bodies, in the valuation and planning of low-carbon transitions.

'Bio-physical pathways' are anchored in a climate research strand concerned with mapping global trajectories of GHG emissions, in order for GHG stabilization and abatement (see table 1). This concept is most

prominently advanced by the IPCC’s summaries for policy-makers, presenting graphs with dotted mitigation pathways reaching net zero somewhere beyond 2050, keeping the global temperature rise below 1.5/2.0 degrees (within a margin of probability). Actors operating on the basis of this pathway concept commonly distances themselves from giving direct policy advice, instead emphasizing value-neutrality, establishing the environmental 'boundaries' which politics is to operate within, but leaving open the policy instruments with which to pursue mitigation (Rosenbloom, 2017). The bio-physical pathways conception can also be

identified in the Swedish climate policy context, in the form of arrangements created to monitor current levels of GHG emissions and effect-assessment of policy instruments, forming the basis for the process of setting national climate targets (Knaggård, 2014).

'Techno-economic' pathways have emerged from ideas of economics and technology assessments, aimed at pursuing the most cost-efficient means of transitioning currently existing economic sectors (see table 1). This conception is commonly based in neoclassic economics, where carbon pricing is understood as a means to internalize the socially externalized cost of pollution, thus directing mitigation efforts to the sectors where marginal abetment costs are lowest (Stern, Stiglitz, 2017). Further, carbon pricing approaches are seen as the first best policy alternative precisely because they provide flexibility for market actors to choose the means of reducing emissions, in contrast to 'inflexible' regulation (Baranzini et al., 2017). The notion of achieving emission reductions via the means of carbon pricing and market-based approaches has also been dominant in Swedish climate policy discourse and practice (Hildingsson, 2014b).

'Socio-technical pathways' emerge from the academic transition field (discussed in detail in the next section), which has its origin in technological innovation studies, but which is increasingly engaging with political dynamics in how socio-technical regime shifts take place (see table 1). In the institutionalist direction of this field it has become a premier question to instill climate policy with 'stability' as means to create new path- dependencies towards decarbonization (Rietig & Laing ,2017; Jordan & Matt, 2014). The apprehension of society as a complex socio-technical system, has also incurred the promotion of a wider set of policy instruments (in tandem with carbon pricing) to reconfigure production and consumption patterns (ibid.), so- called policy mixes (Hildingsson, 2014b). This could mean the introduction of a policy-mix calls for more pro-active state intervention (ibid.).

Methodologically, the three conceptions plot out possible transition pathways using differing scientific practices. Moreover, in relation to policy evaluations, these scientific practices can relate to 'ex-post',

retrospective assessments of policy outcomes (Schoenefeld & Jordan, 2017), or 'ex-ante', prospectively “(…)

seek to inform decision makers by predicting and evaluating the consequences of various activities according to certain convention’’ (Owens et al. 2004, p.1944). Bio-physical evaluations monitors ex-post GHG

emissions and forecasts ex-ante GHG emissions using research infrastructures and practices belonging to the natural sciences (Rosenbloom, 2017). Techno-economic evaluations are usually based upon a combination of backcasting (working backwards from a desirable target) and ex-ante forecasting, employing quantitative economic models to asses the potential of technology and optimal levels of carbon pricing to achieve the target (ibid.). Socio-technical pathways have advanced the use of ex-ante back-casting, lending attention to the social processes believed to take place within sustainability transitions (Smith & Kern, 2009). Yet, the socio-technical field has been criticized, for under-appreciating how power and knowledge is favoring certain pathways over others (Stirling, 2014), and for employing back-casting methods primarily focused on technological change, devoid of whom and how social agents and structures should change (Wangel 2011).

To reflect this critique, it has been suggested that 'transformational pathways' should be separated from socio-technical pathways (see table 1 ( Stirling, 2014)). The following section seeks to disentangle how these debates have unfolded, in relation to low-carbon transitions or transformations as political processes.

3.2 Politics in low-carbon transitions

Transition or transformation concepts have gained in scholarly popularity over the past decade, sharing concepts of 'fundamental shifts' of 'complex systems' as common features (Feola, 2014). Beyond these general characteristics, the socio-technical transition field have conceptualized transformations as one possible transition pathway amongst others, albeit being more forceful in pace, scope and depth as compared to other transition pathways (Geels, 2014; see Markard et al., 2012 for a review of the socio-technical transition field). In opposition, a critical social science research field has characterized transformations as qualitatively different from transitions, criticizing the socio-technical transition field for under-theorizing the interlinkage of power, politics and knowledge rendering ”which pathways get supported and legitimized, and which are ignored and so fail to gain traction” (Scoones et al., 2015 p.7; see Feola, 2014 for a review of the transformation field). In response to this criticism, there has been a growing engagement with politics within the socio-technical transition field, particularly around how 'niche' actors initiate, and 'incumbent' actors resists regime change (Geels, 2019).

Table 2. Contrasting conceptions of transition pathways

Bio-Physical Techno-Economic Socio-Technical Transformational

General conception of transition pathway

”Long-term trajectories of GHG emissions linked to particular stabilization targets and derived from macro-level parameters describing human-climate interactions over time”

(Rosenbloom 2017, p.40)

”Sequences of techno- economic adjustments linking current sector configurations to desirable low-carbon future states ” (ibid.)

”Unfolding socio-technical patterns of change within societal systems as they move to meet human needs in a low-carbon fashion” (ibid.)

”the way in which authoritatively asserted

‘scientific’ knowledges can have the effect of asymmetrically emphasising particular favoured pathways at the expense of others. It confirms that the obstacles to transformative change are manifestly more institutional and cultural (and epistemic and normative), than they are material or technical.” (Stirling 2014, p.87)

Research strands

Climate science Technology assessment/

Economics

Socio-technical transition field

Critical perspectives

Governance in the Swedish policy context

National climate targets (Knaggård, 2014)

Emissions reductions via market-based approaches (Hildingsson, 2014b)

Decarbonization via policy-mix (Hildingsson, 2014b)

However, the socio-technical transition field still has a tendency to regard power as the mobilization of resources (economic, institutional, discursive), by agents with specific intentions, rather than structural and internal to the political process (Stripple & Bulkeley, 2019). As a result, within the socio-technical transition field, politics have been perceived as struggles over which technological pathways to choose (as entity), rather than constitutive of what alternatives are even perceived as possible in the first place (ibid.).

Moreover, this conceptualization of the political process in the socio-technical transition literature is commonly informed by path-dependency theories, claiming that institutional legacies exerts a constraining effect on transitions, which possibly can be turned to a enabling effect through institutional (re)design (Fünfschilling & Truffer, 2014). In other words, enabling transitions is about designing 'sticky policies' that disincentivizes investments in carbon technologies and create self-reinforcing institutions which are supportive of a transition (Levin et al., 2012). In such a manner, 'progressive incrementalism' is believed to pick-up speed over time which could bring about a broader transformation (ibid.).

Even if incorporating possibilities for ideational change, institutionalist path-dependency theories are susceptible to the criticism of being overly deterministic, by virtue of viewing institutions as predetermining the progression of new institutions, in an evolving chain, with the occasional strategic intervention or

exogenous shock (Schmidt, 2010). Moreover, such notions within the transition field are linked to the idea of 'transition management,' that in all their complexity, transition processes can be controlled via strategic planning of an identifiable number of variables (Shove & Walker, 2007). In contrast, there have been attempts in recent times to develop a discourse-theoretical understanding of institutions as the result of discursive struggles and path-dependent properties as a result of hegemonic discourses temporary stabilizing paths-in-the-making (Buschmann & Oels, 2018). These sort of transition-discourse crossovers have explored the role of discourses in the creation of protective spaces for niche innovations (Raven et al., 2016; Smith et al., 2014), framing of technologies (Bosnan et al., 2014; Geels & Verhees, 2011), coalition building to undermine or reinforce institutional arrangements (Fünfschilling & Truffer, 2016; Markard et al., 2016), and storylines and discourse coalitions in framing technological pathways (Rosenbloom, 2018; Rosenbloom et al., 2016). The uniting orientation of this field, is that all these studies have inquired particular processes of technological rollout (for instance renewable energy infrastructures), instead of national climate politics in a broader sense. Secondly, on the scale of the agent-structure dualism, the tendency is to perceive agents as employing storylines (or discourses) strategically in order to initiate or resist technological regime changes.

Gillard and colleagues (2016) have interpreted this tendency as a result of over-theorizing complex system, and under-theorizing structural constraints and socially embedded power relations.

The socio-technical transition field has also been slow to recognize how transition theories-in-use now themselves are implicit in the construction of new transition discourses, effecting the political process of evaluating and planning long-term transitions (Audet, 2016). In this thesis it is suggested that a stronger conceptualization of the structural-agency dualism can be retained by returning to the field of argumentative discourse analysis (see section 3.3). This, while maintaining some of the transition field’s disposition towards agency-based explanations, by adopting concepts of how key discoursing agents in some instances can

function as 'policy entrepreneurs,’ exploiting opportunities to transform policy discourse and 'discursive watchdogs' to keep an eye on the (re)production of hegemonic climate policy discourses (see section 3.5).

3.3 Processes of path-making

The recognition that transition pathways does not unfold along pre-given routes, compile multiple intentions, give cause to unintended consequences and are interwoven in complex relations of power and agency, have given rise to new definitions of pathways as unfolding through processes of ”path-creation” (Buschmann &

Oels, 2018, p.3), or ”wayfinding” (Stripple & Bulkeley, 2019, p.53). While drawing upon these definitions, this thesis does not adopt the full theoretical approach of these two studies, instead advancing a separate concept of 'path-making.' Path-making refers to the political work of finding a way towards net zero emissions, in this case the establishment of an institution (the Council) intended to aid in this task by producing policy planning and evaluations. Hence, in this thesis, path-making does not mean the concrete policy measures taken or the material paths that envelops in an absolute sense, but the politicized work of evaluating and planning transition pathways.

The recent developments in conceptualizing transitions/transformations as discourse (Audet, 2016; Leipold et al., 2019; Feola & Jaworska, 2018) and differentiation between pathways imagined (Rosenbloom, 2017) are important. However, these accounts have not tended to the question of how transition discourses effect the transition process and are translated into policy practices. Bailey and Wilson (2009) have sought to develop a heuristic model of this process, that is of pathways as developing along 'decision-making corridors' (see the adapted figure 1). Decision-making corridors are constituted by societal paradigms, of ideologies and institutions, restricting which decisions making alternatives are perceived as possible. Pathways reaching outside this corridor are even ’unthinkable,' and for new pathways to develop, the current dominant decision- making corridor must be incrementally 'shifted' or radically 'ruptured' (ibid.).

In resemblance with path-dependency theories, this transition model emphasizes continuity over change, attributing incremental shifts in policy to endogenous explanations, while admitting that exogenous shocks

could cause radical ruptures.

However, ruptures substantially altering course of pathways are perceived as unlikely as they tend to

”occur within the bounds of what society deems to be the 'parameters of the possible' at any time” (ibid., p.2330). In summation, Bailey and Wilson’s (2009) transition model is a useful heuristic tool, but gives little or no insight into how transitional shifts (or ruptures) take place and bring about policy change. Insofar, in this thesis, it suggested that the mechanisms by which decision-making corridors remains stable or change, is 'black boxed'. In the

Figure 1. Transition model (Adaption of Bailey and Wilsons (2009) model))

following section (3.3), an Argumentative Policy Analysis (ADA) framework is developed, proposing that such an approach could unpack the 'black box', and for this purpose the original transition model by Bailey and Wilson (2009) has been adapted (see figure 1).

3.4 Argumentative Policy Analysis

Fischer and Forester (1993), famously labelled studies attentive to the constitutive role of language in public policy-making, as part of an ”argumentative turn in policy analysis and planning.” Scholars working within this field have called attention to what politicians and policy experts ”only know too well,” that is, ”policy analysis and planning are practical processes of argumentation” (ibid., p.1). Arguments arise as policy- making is a discursive struggle over how to frame problems, the proper policy response, and public perception, as well as criteria for evaluation (Stone 2002). This form of interpretative policy analysis is constructivist, challenging the two core assumptions of philosophical realism; (1) that the social world exists independent of the comprehension and construction of it, and (2) that the external world is accessible to us in 'pure form', that is, not contingent on the techniques that are used to capture and understand it (Behagel, 2012; Hajer & Wagenaar, 2003).

In Maarten Hajer’s (1995, 2003, 2006) seminal contributions to Argumentative Discourse Analysis, practice and discourse are seen as being in a dialogical relationship, both being affected and affecting. More

precisely, discourses are ”a specific ensemble of ideas, concepts, and categorizations that are produced, reproduced, and transformed in a particular set of practices and through which meaning is given to physical and social realities” (Hajer, 1995, p.44). For Hajer (1993, p.45), the 'real challenge' that faces policy analysis is to combine the analysis of discursive production with the analysis of extra-discursive practices that give rise to social constructs. To elucidate the relation between discourse, practice and institutionalization, Hajer (1993; 1995) develops a number of useful analytical concepts. These include instances of 'interpellation', which opens up spaces for counter-discourses to be raised and new 'storylines' to be articulated to challenge routinized proceedings and meanings. Hajer (1995) argues that storylines in this capacity “(…) play a key role in the positioning of subjects and structures. Political change may therefore well take place through the emergence of new storylines that re-order understandings. Finding the appropriate storyline becomes an important form of agency” (p.56). This confrontation comes from a group of actors that share a rival social construct (i.e., a divergent view of reality), a so-called 'discourse coalition.' However, discourse coalitions should not be understood as a set of actors strategically acting together. Rather, separate actors are

coordinated through the development of a shared social construct, represented in a collection of storylines.

Storylines are ”condensed statements summarizing complex narratives used by people as short hand in discussion” (ibid., 2006, p.69). By virtue of being condensed, storylines facilitate cooperation between actors who may not hold the same meaning to the storyline, such flexibility is important in bringing together coalitions. What attracts actors to a set of storylines can vary, but what is most important is that they 'sound right.' Qualifying what ‘sounds right’ is dependent upon the subjective predisposition of the listener, filtered through the discourses and practices they adhere too. The influence of a discourse can be measured

employing a two-step procedure, ”if many people use it to conceptualize the world (discourse structuration) and if it solidifies into institutions and organizational practices (discourse institutionalization)” (Hajer, 2006,

p.70). Hence, when both these conditions are fulfilled, and the policy process is conducted by institutional practices given by the ideas from a particular discourse, then it can be regarded as 'dominant' or

'hegemonic' (Hajer, 1993). However, Mert (2015) cautions against an understanding of institutionalization as a single discourse solidifying into a single organization. Instead, a single organization can be understood as the result of conflicting claims, often sedimenting into incongruent organizational designs.

In summation, ADA offers a number of useful concepts to analyze the role of discourse in the political process of 'path-making,' interpreted as argumentative struggles over the appropriate policy practices to realize a certain path. As such, path-dependent properties are not inherent to the institutions themselves, but rather the result of a discursive (re)production and stabilization process (Schmidt, 2010). However, Hajer’s (1995) treatment of the structure-agency dualism leaves room for interpretation. Agency is attributed to discourse coalitions in that they form new storylines of how a particular issue should be understood and dealt with, a capacity which arguably has been extended very far in discourse-transition crossovers, of actors using discourses strategically to certain ends. Nevertheless, in agreement with Hajer (2006), we also have to see to how actors are structurally bound to subject positions which are discursively constituted, considerably restricting their room for agency. Here it will be suggested, in line with Leipold and Winkel’s (2016) discursive agency approach, that agency has partly been attributed in the wrong place. These discourse- transition crossovers have so far overstated agency in structuring discourses, while also failing to recognize how practices of discursive (re)production also leave room for a bounded form of discursive agency. In other words, in tending to the question of shifts in transition pathways, we also have to see to the (methods) practices with which we understand where we are (evaluation) and where we are heading (planning). This question will be discussed further in the following section (3.5).

3.5 Strategic practices

Drawing upon Leipold and Winkel’s (2016) discursive agency approach, a concept of 'strategic practices' is adopted. While chiefly embracing Hajer’s understanding of discourse institutionalization (and the associated ontological/epistemological assumptions), these authors contends that storylines are not sufficient to account for strategic action, arguing that practices can also have strategic elements. This refinement is proposed with the explicit aim of finding a middle-range between discourse analysis (tending to be structure-focused) and agency-focused approaches. For this purpose, they define strategic practices as ”all practices that target the creation (and institutionalization) of a particular political truth about an issue and one’s position in relation to it” (ibid:15). This definition places agency in actors’ abilities of (strategic) positioning and self-identification with particular subject positions, while maintaining that actors (i.e., identities and political preferences) are dialectically constituted in relation to discourses. Among a number of strategic practices, including coalition building and production of storylines, Leipold and Winkel (2016) put forward 'governance strategies' as those strategies that target the restructuring of the policy-making process in favor of a certain position. For instance, Hildingsson (2014b) identified how 'cost-efficiency' is an integral part of a market liberal policy discourse, while simultaneously denoting a practice of conducting cost-efficiency analysis in climate policy evaluation and planning, thus (re)producing techno-economic descriptions of low-carbon transitions.

Moreover, Hildingsson (2014b) suggests that the discoursing agents, which in this case defended the practice of cost-efficiency analysis, could be regarded as 'discursive watchdogs.' In this sense, discursive watchdogs

could be conceptualized as agents actively working for the continuity of a dominant institution, the direct opposite to Kingdons (2003) 'policy entrepreneurs' seeking to exploit 'windows of opportunity' to produce policy shifts. However, while 'watchdogs' and 'entrepreneurs' could be used as an heuristic tool to introduce agency in the analysis of institutional continuity and change, it will be maintained that the boundary of the decision-making corridor and the subject position agents can adopt are co-constituted by discursive structure.

Relating the discussion above to Bailey and Wilson’s (2009) transition model gives our analysis four interacting levels in the process of 'path-making,' which defines the boundaries of the 'decision-making' corridor, and which account for 'path' stability (reproduction) and shifts (transformation) (see figure 2). The purpose of this framework steers away from the account of interested actors, in different power constellations, fighting over the formal design of institutions supporting a certain pathway, whereas the process of path-making in-itself is seen as constitutive of power relations and interest, and pathways as always indeterminate and provisional. It also leaves room for agency in articulating storylines and redefining practices, and so the means by which pathway concepts are being performed or imagined. As such, paths are not directly determined by a certain set of formal institutions, but they are shaped by the discourses and practices that have been employed in making them.

4. Method

In the previous section, a theoretical discourse-analytical approach was developed with its specific ontology and epistemology. Discourse analysis can, however, also denote a methodological approach (Jørgensen &

Phillips 2000). These discursive methodologies are rarely consolidated in specified techniques, but designed upon concepts and tools of various approaches to fit the specific research problem at hand (ibid.). In this thesis a three-step research strategy is employed drawing upon the research steps and methods presented by Hajer (2006, p.73), a model similar to that used by Leipold and Winkel (2016, p.19).

4.1 Materials and delimitations

In step (1) an overview of the chain of events was established, compiling three of Hajer’s (2006) steps for conducting ADA (desk research, helicopter interviews and document analysis). This first step also provided the basis for developing an interview questionnaire (appendix 7.2). A newspaper search establishes that the first Swedish mention of a Framework and Council was in a debate article (DN 2012) authored by the Green Party, with the first mention in an official policy document having been in an amending directive (2014:65) by the newly elected Social Democratic-Green coalition government. The amending directive (2014:65) changes the previous Liberal-Conservative government directive (2014:53) for inquiry into a new climate bill post-2020. The EOC inquired and negotiated (via party representatives) a proposal for a Framework, including a Council, submitted as a report in 2016 (Swedish Government Official Reports (SOU) 2016:21)).

The parliament debated and passed EOCs Framework proposal in 2017 and the Council was formed during 2018, having produced three yearly reports since then. While this text corpus offers an extensive material for

Figure 2. Argumentative Analytical Framework

document analysis, official documents or public statements does not necessarily give a representative or complete account of discursive patterns of conflict (Mert 2015). For this purpose, interviews with key actors can establish casual chains and interpretations of particular events (Hajer, 2006), just as they can be used to locate strategic practices (Leipold & Winkel, 2016). In addition, multiple sources of material can be utilized to triangulate results and strengthen the interpretative validity (ibid.).

In step (2) respondents from the EOC and Council were recruited on the basis of strategic sampling (Bryman 2008), both bodies offering 'administrative boundaries' with relatively limited groups of key actors. From these two groups, respondents were selected on the basis of two 'helicopter interviews', that is, interviews with respondents who have an overview of the process and who were able to point out key actors. While also seeking to establish a representative group by assembling respondents from differing political parties and layers of officialdom. The more elusive category of agenda setting, lacked the same administrative boundaries, relying more on the 'helicopter interviews' to point out some of the most central individuals during 2012-2014.

This sampling strategy entails two delimitations and weaknesses. Firstly, by selecting texts and respondents on basis of actors which are sanctioned to be a part of the policy process, it centers elite-discourses. This weakness can partly be justified on the basis that the Swedish climate policy sphere have been shown to be dominated by highly organized networks and groups in close collaboration with the state, effectively excluding radical green counter-discourses (Lidskog & Elander, 2012). Secondly, this weakness prevails as the general public debate over climate policy, as reflected in newspapers, has not been included in the text corpus. This weakness is partly remedied by including parliamentary records and reservations to the EOC (SOU 2016:21), a system giving societal actors a right to add their reservation in the appendixes of

governmental official reports. However, everything considered, the selection of material as-well-as analytical framework is purposely biased towards elite discourse and endogenous over exogenous explanations.

Following standard practices for qualitative interviewing (Bryman, 2008), seventeen semi-structured interviews was conducted during 2020, spanning between 30-60 minutes (8 telephone and 9 on-site

interviews), recorded and transcribed (using transcription software and manual correction). All interview and policy texts are in Swedish and excerpts included in the thesis have been translated by the author. This form of interpretative research approach puts a high demand on knowing the subject, terminology, and social situation of the respondents, even more so in the case for elite-interviews (Brinkmann & Kvale, 2014). In general, this could be facilitated by the authors decade-long engagement with Swedish climate politics, and by a meticulous familiarization with the policy documents prior to the interviews. Moreover, elite-interviews in a politicized context demands higher standards for confidentiality as the identification is easier in small

Table 3. Empirical material

Interviews Policy documents

Helicopter interviews ²

Agenda setting 2012-2014

3 Greens debate article 2012, Governmental directives for new climate targets post-2020 (Directive 2014:53/65)

EOC 2014-2017 ⁷ EOC SOU2016:21, Parliamentary records preceding the vote on the Framework 2016/2017:125 Council 2018-2020 5 Council reports, 2018,2019,2020, Directive 2017/1268

groups (ibid.). Anonymous interviewees were chosen to get access to accounts of conflicts that could have been sensitive to disclose with the respondent’s name published. To secure that the respondents were well informed and their data was handled according to their will, a record of consent was obtained from all participants.

4.2 Operationalization and analysis

In step (3) interview transcripts and policy documents were thematically coded according to recurring categorizations, concepts, and ideas (Hajer, 2006), using MAXQDA software for qualitative data analysis.

Ordering the codes according to the analytical questions (table 3), of actors, storylines (and subcategories), institutional practices and strategic practices. The resulting coding schemes from the analysis of the

interview transcript was in a later stage used to delimit the extensive text corpus in the policy documents and to triangulate interpretations/factual information.

The analytical framework is circled in a larger conceptual model (figure 3) which draws from Bailey and Wilson (2009, p.2329) model of pathways as decision-making corridors, but unpacking (the 'black box') how transitional shifts can take place using an ADA framework.

Analyzing how dominant pathways are (re)produced or transformed, while inserting the empirical case (pathway —> Agenda setting —>EOC—>Council—

pathway), as a circular process of path-making. Moreover, this model pre-assumes discourses upon pathways to be captured by either three of Rosenbloom’s (2017) core conceptions, while being uncommitted and open to the exact content of these discourses.

5. Results and discussion

In the first sub-section of the results and discussion (5.1), the aim is to explain how a 'climate law-storyline' facilitated a cross-party consensus backing up the Framework. The Framework was constituted by interim- and long-term mitigation targets, planning processes and monitoring functions, described as institutionalizing

Table 4. Analytical questions

Bio-physical Techno-economic Socio-technical

Main actors Who are the key actors forming coalitions around certain storylines?

Main storylines What are the main storylines and how do they stand in relation?

Institutional practices Which institutional arrangements are sought and why?

Strategic practices How are institutional practices used strategically to establish certain political truths?

Figure 3. Operationalization (Adaption of Bailey and Wilsons (2009) model))