Carbon Storylines : The discursive struggle over carbon offsets as a decarbonization pathway in the Swedish Climate Policy Framework

(1)

Department of Thematic Studies Environmental Change

MSc Thesis (30 ECTS credits) Science for Sustainable development

Sandra Ideskär

Carbon Storylines

The discursive struggle over carbon offsets as a

decarbonization pathway in the Swedish Climate

Policy Framework

(2)

ii Copyright

The publishers will keep this document online on the Internet – or its possible replacement – for a period of 25 years starting from the date of publication barring exceptional circumstances.

The online availability of the document implies permanent permission for anyone to read, to download, or to print out single copies for his/hers own use and to use it unchanged for non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility.

According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement.

For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its www home page: https://ep.liu.se/.

(3)

(4)

2

1. Abstract

This study addresses discourses and how they affect climate policy, through the example of carbon offsets as a tool to reach domestic emissions reductions in the Swedish Climate Policy Framework. An interpretation of Maarten Hajer’s argumentative discourse analysis is applied to understand the ideas and arguments that inform the policy debate on carbon offsets as a supplementary measure in this policy process. By mapping, comparing and finding dominating storylines, it presents how Swedish government actors, businesses- and non-governmental organizations legitimize, justify and contest carbon offsets. The findings suggest that the dominating storylines largely remain in the status quo on carbon offsets, connecting to the larger policy discourses of ecological modernization and green governmentality. They also show a potential attempt to divorce of international development from carbon offset mechanisms, as a way to increase efficiency. However, in a Paris Agreement and Article 6 landscape, room to reimage and critically evaluate the use of carbon offsets has also emerged. Established actors join civil society in raising uncertainty and doubt of the future of carbon offsets. These discursive shifts may impact how Sweden intends to exercise leadership in deep decarbonization going forward.

(6)

4

2. List of Abbreviations

BAU – Business as Ususal

BECCS - Bio-energy with Carbon Capture and Storage CDM – The Clean Development Mechanism

EU ETS - EU Emissions Trading System

LULUCF - Land Use, Land Use Change and Forestry NDCs - Nationally Determined Contributions

(7)

5

3. Introduction

Sweden has committed to reaching net-zero greenhouse gas emissions by 2045, with negative emission to follow. The goal is presented in the Climate Policy Framework, adopted by the Swedish parliament in 2017 to ensure policy coherence and long-term success for a climate transition in society (prop. 2019/20:65). The fossil-free transition has the explicit goal of being fair and with a broad consensus in society. The framework was created as a way to ramp up national ambitions in accordance with the spirit of the Paris Climate Agreement, where Sweden has pledged as part of the European Union (2020), with the goal of keeping temperatures well below 2C degrees from pre-industrial levels. In line with this increase in ambition, Sweden aims to become a world leader and the first fossil free welfare state, showcasing the possibility to combine fossil-free competitiveness and welfare. This includes creating policies ensuring emission are reduced within the national borders first, and later outside through technology transfer and by encouraging more stringent climate policies abroad (prop. 2019/20:65, p.9). In numerical terms, the goal of net-zero means that by 2045 Sweden’s territorial greenhouse gas emissions should be 85 percent lower than in year 1990. To eliminate the remaining 15 percent to reach net-zero, the Swedish climate policy framework introduces what it calls “supplementary measures” (prop. 2019/20:65, p.166; prop. 2016/17:146). These measures can include increased carbon uptake in forests and land, bio-energy with carbon capture and storage (BECCS) and verified emissions reductions in other countries. According to the framework, these supplementary measures can also account for reaching sub-goals before 2045 (eight per cent of emissions reductions before 2030, and two per cent of emissions reductions by 2040). The introduction of ‘supplementary measures’ in the Swedish climate policy framework marks a shift from previous Swedish climate policy commitments, where flexible mechanisms and similar were not counted towards national emission reduction goals (Zannakis, 2015; Friberg, 2008). Not all of these measures are, however, ready for immediate use (e.g. BECCS). Gaps in CCS technologies (SOU 2020:4) suggest that verified emission reductions could play a pronounced role in the near future.

Verified emissions reductions in other countries – also known as carbon offsets – have a long and contentious history in global climate politics. Born under the 1997 Kyoto Protocol, carbon offsetting is part of the Clean Development Mechanism (CDM), created as a market mechanism for decarbonization for the global north, while simultaneously promoting sustainable development in the global south (Bumpus & Liverman, 2011). Theoretically this made carbon offset markets a two-part win. One, they could reduce global emissions. Two, they could provide benefits to host countries by providing sustainable finance and green technology transfer. Co-benefits of carbon offsetting also included such things as improved livelihoods, health benefits and education in the project countries. These co-benefits have however been argued to be uneven, with issues of legitimacy, effectiveness and justice raised of carbon offsetting projects (Cavanagh & Benjaminsen, 2014; Lohmann, 2008; Okereke & Coventry, 2016). The effectiveness of carbon offsets in reducing emissions overall has also been questioned (Hulme, 2019; McAfee, 2016). In a Paris Agreement world these offset markets are moreover in an extended limbo, as Article 6 of the agreement is yet to be decided, with its structure determining their viability and form going forward.

The uncertainty surrounding the future of the carbon offset markets, the legitimacy issues surrounding the very function they are to provide raises question for Sweden. Question of how or if carbon offsets can be a responsible or fair tool in decarbonization. The decisions to

(8)

6

potentially include verified emissions reductions in the Swedish decarbonization pathway is not an apolitical process (Robbins, 2012). For Sweden to become a world leader, many have argued that historically, burdens of emissions reductions have been accounted nationally and not shifted elsewhere (Zannakis, 2015; Fiberg, 2008). The use of carbon offsets in reaching net-zero domestic emissions thus raises several questions of responsibility, justice and legitimacy. With the final function of the supplementary measures yet to be decided, as well as the pending Article 6 of the Paris Agreement, it is important to study how actors in the policy landscape are understanding and constructing the future of carbon offsetting. Carbon markets and other environmental policies do not become legitimized and institutionalized in a vacuum - but are part of the discursive practices and justifications that make them so (Bäckstrand & Lövbrand, 2006; Zannakis, 2015).

3.1 Aim and Research Questions

This study aims to understand the ideas and arguments that inform the policy debate on carbon offsets as a supplementary measure in the Swedish Climate Policy Framework. To that end, the thesis maps and compares different storylines articulated by Swedish government actors, businesses- and non-governmental organizations in response to the government inquiry SOU 2020:4, The Road to a Climate Positive Future (my translation). By analyzing the policy debate that followed the release of the government inquiry, it traces how actors justify, legitimize, and contest carbon offsets as a policy tool in reaching negative emissions and the potential implications of these perceptions. Studying this political space allows for an examination of the power of discourse in Swedish climate policy-making, and how ideas and meaning-making shape and form avenues for climate action.

The thesis is guided by the following research questions:

• What ideas, arguments and narratives are presented by Swedish government actors, businesses, and non-governmental organizations in relation to the use of carbon offsets in the Swedish Climate Policy Framework?

• Which storylines dominate in this policy debate and how do they work to justify and legitimize and contest carbon offsetting as a Swedish climate policy tool?

• What are the implications of the discursive struggle over carbon offsets in the Swedish climate policy framework? What does it tell us about the potential for how Sweden intends to exercise leadership in deep decarbonization?

3.2 Thesis structure

This study consists of six main parts. Following the above introduction and aim, chapter four presents the theoretical background of the study. Discourse as a theoretical framework and the functions of Maarten Hajer’s Argumentative Discourse Analysis will be discussed to anchor the study. Here, I also introduce policy discourses found by previous scholars examining carbon markets and environmental issues, to be used as a base in the study. Chapter five introduces the methodological consideration and process for analysis, as well as discussing the material used. A version of Hajer’s Argumentative Discourse Analysis will be introduced and operationalized for the study. In chapter six, I first offer an empirical background to the role of carbon markets and their future in the Paris climate regime. This forms the background for chapter seven, wherein I present the results of the discourse analysis. In chapter eight I discuss my results and draw conclusions.

(9)

7

3.3 Delimitations

Verified emission reductions in other countries, carbon offsets, international financing of carbon reductions projects and many more. The names for the practice most commonly known as carbon offsetting are many. In this study, this phenomena will primarily be referred to as carbon offsets. It may also be referred to as verified emissions reductions in other countries if discussed in the context of SOU 2020:4.

(10)

8

4. Theoretical Background

4.1 Discourse and the Making of Meaning

Swedish government actors, businesses- and non-governmental organizations carry different ideas and arguments that informs the debate on carbon offsets in the Swedish Climate Policy Framework. The presence of these ideas and arguments are however not always visible at first glance. Discourse analysis provides a vehicle to access these underlying discursive formations. Both a theory and a method, discourse analysis exists in a myriad of different versions and theoretical starting points, with the common thread of examining language and its meaning making (Boréus & Bergström, 2017). Language here is broader than linguistics, encompassing the “ideas, concepts and categories” that creates meaning to the world around us (Hajer & Versteeg, 2005, p.175). Language in this understanding does not mirror the world, but is what shapes the way actors view it (Hajer, 2006, p.66).

Discourse analysis cannot be summarized under one umbrella, with different epistemological and ontological traditions. The agency of actors, the rigidness of institutions and the connection to their social context differs from tradition to tradition. In this thesis, the main theoretical understanding of discourse has its basis in social constructivism. This assumes that reality and the world does not exist “out there” separate from its societal context, and is unfixed. Resting heavily on Foucauldian ideas of truth and knowledge, assumed truths, norms and practices and their underlying political processes become the object of study (Winther Jørgensen & Phillips, 2002). In turn, actors (in their broadest sense) do not become the main object of study for policy, but rather the ideas and norms they are steeped in (Stevenson and Dryzek, 2012). Discourses then represent a collection of ideas, norms and understanding and examines the way power is produced and tied to that of knowledge making (Bäckstrand & Lövbrand, 2007). Discourse analysis derived from this theoretical starting point is often critical, drawing a line between the historic and social context of an issue through to its knowledge making and finally the social action (Winther Jørgensen & Phillips, 2002).

Discourses and their functions become particularly interesting when viewing the functions of policy, as a nexus of the knowledge- and meaning making guiding society. This implies that dominating discourses wield power, and reject other knowledge production that may exist. Some discourse theorist see this as creating a hegemonic understanding, void of conflict while others see actors creating different understandings under a common discourse (Hajer, 1995). What discursive assumptions and actors that builds these hegemonic or dominating discourses enables some solutions and rejects others.

4.2 Argumentative Discourse Analysis

This study will utilize one of Hajer’s (1995) theories of discourse in environmental policy, Argumentative Discourse Analysis. Hajer builds argumentative discourse analysis on Foucaultian ideas – where historical contexts, power, institutional and social practices are all intertwined (Hajer & Versteeg, 2005). For this study, this means that what materially is happening to the environment may not be correspondent to the way discourse is presenting the problem and solutions. Environmental politics, due to its complexity, requires knowledge few apart from experts inhabit. Laypeople, politicians, civil society must therefore adapt these complex processes when entering the conversation. To understand political processes and their underpinnings - how discourses are constructed and differ then becomes the main point of environmental politics (Hajer, 1995).

(11)

9

In essence, argumentative discourse analysis traces and uncovers how language and its argumentative construction implicates political solutions. It does this through studying discourse in its institutional context and through its practices. Studying these argumentative constructions uncovers how actors communicate and in turn how discourses constrains and shapes their interactions (Zelli, Nielsen & Dubber, 2019). Discourses are approached with the assumption they will be fragmented (Hajer, 2006), with many actors co-producing and finding individual meaning through common narratives. Discourses are sustained by the actors reiterating them, informed by their historical understanding of an issue. As such, there is no unified understanding of environmental policy and concepts.

Actors in argumentative discourse analysis may prescribe to the same ideas and arguments, but they will still attach independent interpretations to them. They will not fully understand each other yet still create a common lane for political action (Hajer & Versteeg, 2005; Hajer, 2006). When talking about the destruction of a forest, every actor prescribe a new meaning to what the problem is. Is it the loss of habitats for animals? Loss of cultural value? Loss of revenue for the logging industry? In the context of this study, it means while actors may describe carbon markets as efficient or unjust, what they mean by these definitions may vary. These variations and clashing of definitions within and between these discourses then becomes an examination of something much bigger. How parallel narratives are formed and become dominant provides understanding of who forms and are affected by the political solutions in larger discourses. This makes discourse a means of accessing relations of political power – and the practices creating it (Svarstad et al., 2018).

While argumentative discourse analysis is mainly used to examine policy – and by extension politics and its actors – the actors affecting discourses can also be found outside the boundary area of politics. This as Hajer (1995), defines politics in a broad sense. Politics is a process where actors outside the traditional boundary areas of political parties are active. For this study, that means that NGOs, scientists, researchers and other “traditionally” non-political actors can be part in producing these discourses, as they are part of the political process.

4.2.1 Storylines and Discourse Coalitions

Hajer (1995) presents two main theoretical concept to analyze text in Argumentative Discourse Analysis to be used in this study. One is storylines, the other is discourse coalitions.

Storylines “is a generative sort of narrative that allow actors to draw upon various discursive categories to give meaning to specific physical or social phenomena” (Hajer, 1995, p. 56). Storylines are thus compressed and simplified articulations of ideas and arguments of larger discourses. These bite-sized, reoccurring formulations are more manageable for actors to call upon than overarching discourses that are not necessarily part of their everyday lives (Hajer, 1995, p.56). Storylines therefore exists for actors to redefine and produce their understandings of broader dominating discourses through short-cuts (Hajer, 2006). This reproduction of storylines continuously reimagines the problem definition of an issue – making actors operate an ever changing landscape of ideas.

For climate change, with issues incredibly complex and multifaceted in scope, no one actor can grip them fully. We do not experience increased invisible carbon emissions by ourselves, we experience it because experts presents these consequences to us (Hajer, 1995). In the case of this study, the focus is on the quantification of carbon through carbon offset markets. Actors

(12)

10

referring to carbon markets assume their audience “will know what he/she means or refers to” in conversation (Hajer, 2006, p.69). Instead these actors interpret issues and fill them with their own understanding, ideas and arguments. Together these actors may create a “false assumption” (Hajer, 2006, p.69) of mutual understanding of what carbon markets are - while still remaining fractured.

As such, storylines allows actors in political processes talk about issues they are not necessarily fully invested. Their simplification is key for Hajer (1995), as it enables the suggested common understanding and consensus, where there may be none. In turn, this assumption of common consensus and simplification is the very thing allowing for political action. As new storylines emerge, new coalitions follow, where actors re-arrange their understandings of a problem. Actors talk about the same subject, but carry different underlying assumptions that do not overlap. Or, they “talk at cross-purposes” (Hajer, 2006, p. 69).

The idea of framing, defining and simplification of problems for mutual understanding through policy is not unique for Hajer. However, the distinction in Hajer’s definition lies in how these storylines exists not because actors are actively positioning themselves or rejecting other view-points (though they may). Rather, they simply see the storylines as the way to talk about a subject at that moment in time (Hajer, 1995, p.56-57). It is in this space where different actors with varying interests can form coalition to create political action.

Hajer (1995, p.65) defines discourse coalitions as “the ensemble of (1) a set of story-lines; (2) the actors who utter these story-lines; and (3) the practices in which this discursive activity is based”. Discourse coalitions then represents the way actors create, reproduce and share a common language of a discourse and enable political action. While examining discourse coalitions the main focus should not be understood as the interest of the actors. Rather, the focus should be on the storylines. Actors with diverging interests may prescribe to a storyline, but this storyline may be located within several discourse coalitions. As the storylines change, the understanding of an actor’s interests may also change (Hajer, 1995). Studying discourse coalitions and their institutional implications fully is complex and requires extensive material not possible within the realm of this study. Discourse coalitions in this study will therefore primarily focus on the first two points presented by Hajer (1995), the storylines themselves and which actors are part of sustaining and re-producing them.

Storylines and discourse coalitions become useful in a project such as this, where the object of study is to understand, map and trace how actors justify, legitimize, and contest carbon offsets. Both as a policy tool in reaching emissions reductions and the potential consequences of these perceptions. How actors are positioned and struggle to define an issue by their underlying assumptions creates avenues for action. Swedish actors in the debate for carbon offsets will produce and re-produce what carbon offsets are based on certain assumptions, problem formulations and solutions. These constructions may provide different understandings of offsets as a policy measure. These actors can have different interests and goals, but through different storylines they communicate their simplified understanding of an issue.

It is the continued use of these storylines that culminates in different dominating discourses. Hajer (1995, p.60-61) describes dominating discourses through the terms structuration and institutionalization. Structuration ties to the credibility claim of the actor in sustaining discourses. In a structured discourse an actor cannot maintain their credibility of a given subject if they move outside the structured discursive frame. Institutionalization on the other hand is tied to action. It refers to how a discourse results in actual policies. Within the scope of this

(13)

11

study it will not be possible to claim that a storyline results in institutionalized or structured discourses. Instead, the prevalence of ideas and their storylines in the material will be measured, followed by a discussion on what these storylines indicate in terms of the institutionalization and structuration that Hajer describes.

4.3 Policy Discourses

Carbon offset markets offer an interesting case of an environmental mechanism within political economy – that operates within multiple levels of governance processes simultaneously (Newell & Bumpus, 2012). It is both abstract in conception, but material in execution. Discourse has the means to access how different actors create meaning to commodify, value and govern this environmental entity. Carbon markets are especially interesting in the mode of social constructivism, as they are managing an invisible environmental entity that must be constructed to be fully grasped in policy. An entity that cannot be seen with the bare eye, but where global constructions and power have local implications (Newell & Bumpus, 2012). A study on these discourses does not exist in a vacuum, and previous studies in the field can provide a starting point in how to situate different issues.

To make sense of storylines and coalitions of carbon offsets in the Swedish Climate Policy Framework, previous studies have identified environmental frames and discourses that can aid in this study (see Bäckstrand & Lövbrand, 2006; Teräväinen, 2010; Smith & Kern, 2009, etc.). As storylines are snippets of ideas and arguments in discourses, these discursive categories provide a way to understand storylines in a discursive and historical context.

This study will employ the three analytical categories as provided by Bäckstrand and Lövbrand (2006) in their study on carbon offset forestry in the CDM. The three main strands of discourse presented are: ecological modernization, green governmentality and civic environmentalism. These discourses are not to be seen as separate silos, but overlap, interact and challenge each other. They are not fixed and pre-determined, but rather act as analytical tools to understand in what context different storylines may function. With a social constructivist approach it is also assumed other interpretations of discourses will exists out there. These categories therefore provide an initial starting point in analyzing the storylines in a context, but are not seen as exclusive frames of understanding.

4.3.1 Ecological Modernization

Ecological modernization as a discourse presents the idea that technological solutions and the market can mitigate environmental degradation. Continuous growth and industrialization are not seen as incompatible for solving the environmental crisis faced, but rather as the solution in itself. Science and the resourcefulness of humans will be able mitigate continued consumption, growth of the population and industrialization within the three pillars of sustainable development. The assumption of the current economic system as the solution also means that capitalism itself becomes “green“ and environmentally friendly (Hamilton, 2016; Baker, 2015; Hajer 1995). Ecological modernization has become a popular way to approach environmental policy from the global north, as it does not require radical institutional change and does not necessarily prescribe to limits of economic growth (Baker, 2015). For Hajer (1995), this discourse may lessen the appetite for more radical action, similar to the discussion on sustainable development (Fischer & Hajer, 1999), as it dilutes the meaning of environmental degradation – making consensus policy making the only rational way forward rather than more radical solutions.

(14)

12

Because of the separation of environmental degradation and the economic system (and focus on technological solutions) ecological modernization prescribes to market driven and technocratic solutions (Bäckstrand & Lövbrand, 2006). This implies that ecological modernization represents a decentralized approached, with the role of the state as a curator of environmental process, and main focus and action falling on private actors. In essence, ventures proclaiming to increase efficiency while minimizing environmental negatives can be seen as belonging to ecological modernization. For states, this means policy that internalizes environmental externalizes, such as carbon taxes (Buttel, 2000).

Bäckstrand and Lövbrand (2006) however details that ecological modernization can exists in weaker or stronger variations of the concept. The first version, seen as weak, is understood as detailed above. It rejects stronger institutional reform, focuses on economic efficiency and technological solutions, informed by the hegemonic neo-liberal economic system. A “stronger” version of ecological modernization invites broadened participation, moving beyond a Western focus in policymaking to a further international understanding. It sees limits to top-down environmental governance, and accounts for perspectives of justice and democracy in the environmental process – making it a cousin to sustainable development. Of the two versions, the weak one has previously been most prevalent in discourse.

4.3.2 Green Governmentality

Green governmentality presents as a cousin to ecological modernization, where it is possible to manage global environmental change. Similar to ecological modernization, green governmentality is technocratic, with a top-down starting-point. However, green governmentality views the world from above as a single unity (or “one world”), with mega-science and expertise as the keys solving the environmental crisis (Rutherford, 2007). Climate change and environmental issues creates risks that transcends borders and physical dimensions. This creates a need for regulation and managerial solutions beyond what traditionally has been supplied (Soneryd & Uggla, 2015). Therefore experts become necessary to produce knowledge and legislative solutions that manage the earth globally.

In this discourse, the state still wields important power, but shares it with big-business and science. These actors prime and influence the responsibility of private actors to shape the future of the planet. The individual is given big importance through the priming of their actions – with consumer choice and its power used as a way to govern (Bäckstrand & Lövbrand, 2006; Soneryd & Uggla, 2015). NGOs, environmental experts, big business and the government are all experts in this global governing, and co-produce the knowledge and science used through it. The monitoring of environmental processes, the way in which it controls the action and responsibility of actors then stems from the knowledge constituted from these actors. Their problem formulations and structuring of problems found in the environment become “an expression of biopolitics that attends to the supporting environment necessary for life to survive” (Wang, 2015, p.324). Ideas such as sustainable development is part of this discourse, as it through expert knowledge tries to manage the perfect weights of the human-environmental relationship.

Similar to the discourse on ecological modernization, Bäckstand and Lövbrand (2006) highlight that green governmentality can be more or less reflexive. A rigid technocratic view can be viewed as elitist and detached. An administrative and managerial approach removed from its social context, where knowledge is produced in a detached way to right the bio-power of the

(15)

13

world in accordance to perceived eco-truths. A more reflexive discourse within green governmentality leaves the rigid global gaze and approaches environmental problems with consideration to their local and social contexts. It invites to participation and a co-production of knowledge that shifts power from elite productions only.

4.3.3 Civic Environmentalism

Civic environmentalism is a critical, bottom-up discourse. In contrast to the above discourses, it sees economic growth, the neo-liberal economic systems and its production as the main cause of environmental degradation. It is averse to technocratic solutions and wants delegation of environmental problem solving to be spread to local communities to strengthen their political participatory power. Social justice, marginalization, power-structures and levels of participation are part of the focus to counter-act these perceived problems (Orhan, 2008). Increased participation on all levels of environmental governing is seen as required to ensure those most affected by the current environmental crisis are given a voice in these processes. Marginalized groups (e.g. traditionally indigenous, women and economically disadvantaged) are prioritized in civic environmentalism with the assumption that no permanent solution to the ecological crisis can be reached without their engagement.

There is however no homogenous consensus on the function of this civic engagement, with many different sides housed under the same discursive umbrella (Agyeman & Angus, 2003; Bäckstrand & Lövbrand, 2006). Some see civic multilevel political participation as a necessary and important compliment to state-to-state and local governance, but relinquishes the ultimate decision-making to the official arena. Non-state actors act as watchdogs and ensures inclusion of local concerns and perspectives in policy discussions, speeding up the greening of policy processes through inclusion (John, 1994). As such, non-state actors are allowed a seat at table and become part of diverse coalitions of actors ranging from NGOs to businesses The above general aversion for market-based solutions and technocratic solutions is still valid. However, the key assumption is that inclusion of outside voices in the political process will act as a counter-weight, promoting greener results derived from a broad coalition of actors.

Other perspectives are further radical, rejecting that inclusion of actors into policy processes as a sufficient measure to solve the environmental crisis. Instead, this fraction of actors see the current and historical uneven power structures permeating multilateral and institutional decision making processes. Participation is still a key part of this discourse, but rather through challenging, resisting and rejecting status-quo. Solutions to environmental issues is instead seen through a radical transformative social process. It moves beyond and rejects neo-liberal institutions and the free market, aiming to expose the uneven power structures perpetuating climate change and environmental degradation (Agyeman & Angus, 2003). Movements such as climate- and environmental justice have grown into and embodied parts of the message of civic environmentalism. This coalition of actors are primarily grassroots non-state actors from civil society, but their messages permeates in other arenas as well.

4.4 Theoretical framework

Storylines and discourse coalitions are the tools to access how ideas and arguments about carbon offsets are justified, legitimized and contested by actors as part of the Swedish Climate Policy Framework. The three overarching discourses will act to inform the analysis and put the storylines in a historical context. Similar as in the analysis made by Zannakis (2015), these storylines will be found in the material first, before later being interpreted into the context of

(16)

14

the broader environmental discourses and how this affects action and the use of carbon offsets as a pathway for emissions.

(17)

15

5. Method and Material

5.1 Operationalizing Argumentative Discourse Analysis

How can you find which ideas and arguments discursively created by Swedish actors in the Swedish Policy Framework? And how do you find the dominating storylines and how they work to legitimize, justify and contest carbon offsets? As the previous chapter shows, argumentative discourse analysis provides theoretical tools to examine these discursive formations. While a fairly flexible method, it also requires operationalization to be workable. Hajer (1995) thoroughly presents his theoretic assumptions, but is less clear on how to operationalize these theoretical assumptions into a method. This provides both a challenge and opportunity for practitioners as argumentative discourse analysis. Certain steps in the process are outlines by Hajer (2006), but ample room is left for adaptation and interpretation by researchers.

This study has therefore taken inspiration from Boréus and Bergström (2017) and previous studies in structuring the analysis, with four main steps:

• First, an initial inductive read-through of the material was conducted. While time consuming, initially familiarizing with the material allows for a process of sorting and finding sections of text interesting for the study. If a section was seen as interesting it was highlighted and/or saved for further examination in later steps in the analysis. Notes were also taken during this process to save interesting sections and quotes.

• Secondly, following the inductive read, a read through of the material with a set of analytical questions was performed (Table 1). Asking questions of the text is not uncommon in discourse analysis, and helps the researcher conduct a focused reading (Esaiasson, 2017; Boréus & Bergström, 2017). Each document of the empirical material was read with these questions in mind. What are the assumptions and themes in this debate, how do they legitimize carbon offsets and how do these connect to previous discursive categories? The answers for each text read was put into a common document where supporting quotes and similar supporting functions in the text were also included. • Thirdly, the results in step two were analyzed through commonly appearing themes and

how Hajer defines storylines and coalitions (Table 2). What themes can be found from step one and two? What main ideas support these themes? What function do these ideas and arguments serve? Who says them? These themes were later collected and grouped to provide an overview of the different reappearing concepts and themes, culminating in the storylines (see Table 3). As according to Hajer (1995) actors will likely subscribe to different ideas and storylines, the different texts in step one-two could exhibit several contradictory concepts and themes. These different concepts found were therefore grouped with similar concepts and themes found in the other texts. Inevitably, this process was also influenced by the history and technical issues of the empirical subject matter (see chapter 6). For example: If cost-efficiency is an idea brought forward, does it reflect ideas found under Kyoto (such as the two-pronged win) or other ones reflecting the Paris-era? Do different actors bringing forward cost-effectiveness share the same assumptions or do they differ? How do they then correspond to the discursive categories of Bäckstrand and Lövbrand (2006)?

• Fourth, the prevalence of the storylines were determined, or how dominant they are. This was drawn from the above steps. How often did these themes and ideas occur and how were they presented?

(18)

16

These four steps were mainly conducted in order and separately, but overlap inevitably occurred as the material was read through many times and the different analytical questions were tested.

Table 1. Analytical Questions.

Storylines

1, What main assumptions, concepts and themes are presented for carbon offsets as a way to reach domestic and global emissions reductions?

2, How are carbon offsets legitimized, justified and contested in the policy process? Is it a moral pursuit? Is it legal? Is it economic? 3, How and to what extent do these assumptions relate to the overarching environmental discourses of ecological modernization, green governmentality and civic environmentalism?

Discourse

Coalitions 1, Which actors collect around these storylines?

The analytical questions (Table 1) were made for two main functions in the process of reading. First, they work according to Hajer to find the storylines and their broad coalitions. Storylines presents snippets of ideas, arguments, assumptions and other factors that influence their construction and understanding of an issue. Also, as mentioned previously, storylines and discourse coalitions are much narrower concepts than that of discourse overall, making them manageable as analytical tools. Therefore, the storylines and discourse coalitions were the first step in the questions. This allowed for the storylines to develop organically, without necessarily being affected by the broader environmental discourses. Second, more focused questions regarding the specific case and the justification of the storylines were asked. Lastly, a question introducing and tying into the theoretical understanding of the broader policy discourses presented in the previous chapter, and how (if at all) these storylines sustain and relate to them. These questions that have informed my discourse analysis are largely interpretive. They look for patterns, arguments and norms. This as prescribing completely to premade theoretical concepts may eliminate the possibility of finding other ways of representing the issue studied (Esaiasson, 2017).

Common for studies who use argumentative discourse analysis (or other forms of discourse analysis) is to interpret if the storylines or discourses found are dominating. Looking for patterns and concepts in text as in the steps above not represented in quantifiable terms, and later representing them in quantifiable terms, is not free from challenges. Making a clear distinction of where one argument begins and ends, or if an idea is frequently appearing is not a clear cut task and a question of definition. What is important is that the process of selection and motivation for the conclusions drawn by the material is clearly defined.

Different studies approach this conundrum in different ways. As discussed in the previous chapter, Hajer (1995, p.60-61) approaches it through discourse structuration and the

(19)

17

institutionalization of a discourse. While not possible to fully claim the level of structuration or institutionalization described by Hajer in this study, the prevalence of ideas and arguments can indicate their direction. The prevalence and dominance of storylines in this study was determined in two main ways. 1) Reoccurrence of the argument/idea, and 2) The amount of actors raising the idea/argument. In some cases an actor may only briefly use a phrase or argument and use it often, while others offer in-depth and complex analysis but less frequently. Both these measurement are admittedly imperfect, but provide a way to categorize and structure the analysis.

5.2.1 Structuring the Results

Previous studies inspired by Hajer’s methods or the storyline concept often times present their results in matrix or table. These present the main points in the storylines or discourses they are looking for (Bäckstrand & Lövbrand, 2006; Stevenson & Dryzek, 2012; Zannakis, 2015; Williams & Sovacool, 2019). Reoccurring key concepts used are; function of the storylines, main assumptions/narratives/metaphors, basis of legitimacy or contestation, frequency/structure of the discourse or storyline as well as the actors and discourse coalitions collecting around them. Others are simpler, presenting how storylines connect to larger discourses. For this study, main inspiration will be taken from Zannaikis (2015) and Teräväinen (2010) due to the focus on storylines rather than larger discourses. The results will be presented through the factors below.

Table 2. Matrix for presentation of results.

Storyline

Carbon offsets as part of national emissions reductions

How does the storylines approach the use of carbon offsets in the future of Sweden’s national emissions reductions? Positive, negative, agnostic?

Main function of the storyline What are the main ideas, arguments, assumptions and themes attached to carbon offsets? What do the storylines promote for the use of carbon offsets? Legitimacy What is legitimizing the storylines? Is it a moral

pursuit? Is it legal? Is it economic? Connection to overarching

discourses

How does the storyline connect to the earlier discussed discourses of ecological modernization, green governmentality and civic environmentalism?

Main actors collecting around the storylines

Who are the main discursive actors sustaining and creating the storylines?

Prevalence of the storyline With what frequency do the ideas and arguments show in the text?

5.3 Material

To answer the research questions and provide understanding of the issue a wide variety of material was needed. Three main strands of material have been collected for use in this study;

(20)

18

1) Published material such as official policy documents 2) Academic and grey literature, and 3) Interviews with actors involved at different stages of the Swedish Climate Policy Framework.

5.3.1 Primary Material

As the full nature and use of supplementary measures role in the Climate Policy Framework is not yet set, a specific set of sources were needed for the empirical analysis. Primary empirical material for this study consists of the inquiry SOU 2020:4, requested by the Swedish government regarding the viability of supplementary measures. To broaden the view, and further be able to evaluate how different actors see the proposed use of verified emissions reductions, additional material was needed. The responses to SOU 2020:4 – The Road to a Climate Positive Future (my translation), therefore became especially important. The committee report gives preliminary guidance for Sweden on how the supplementary measures should be utilized. As it is a committee inquiry, it also invited a wide variety of Swedish actors to respond to its suggestions. In total, 118 different actors responded to this inquiry with written comments. Actors important to this process: such as local governments, state government agencies, business associations, NGOs, civil society, academic institutions and others entities are all represented in the material, as is necessitated by the research questions. While the actors represented in the material may be biased towards different levels of government institutions, sufficient variety of other type of actors was found in the material. These provide a base for the discourse around verified emissions reductions.

In total, 27 responses to this enquiry were included in the study alongside the official report. This disparity from the total response rate occurred as not all written responses had detailed comments on verified emissions reductions, and only the other two supplementary measures. The responses not containing comments on verified emissions reductions were sorted out. While these different responses and documents may not always contain new proposals or ideas to shift discourse, this is not a problem according to Hajer (1995). Rather, as actors repeat the same ideas, they are constantly refining and sustaining discourses, making it key to study. Additionally, semi-structured interviews were conducted with four Swedish actors related to the SOU report to triangulate with the findings. These actors were selected due to their past and potential future involvement of carbon offsets. The function of the interviews is explained further in section 3.4.2.

This material is by no means exhaustive but is able to present a fair representation of the current situation on offsets in the policy process of the Climate Policy Framework. In total, 28 documents and reports were reviewed, and four interviews were conducted.

5.4 Data Collection

5.4.1 Documents and Published Material

The three main types of materials were collected in different ways. The secondary material in form of academic articles and grey literature was collected primarily through different academic search engines, such as Google Scholar, but also through reference tracing of already collected material. The policy documents, reports and other primary material were found on government, parliament and official agency websites.

For SOU 2020:4 and its responses, the main empirical material for this study, all currently available inquiry responses were collected on the Government’s website. Actors may also

(21)

19

respond to inquiries outside of invite from the government. These responses are not available until a final decision of the inquiry has been taken by the Government. Some of these responses were however still used in this study, due to organizations publishing them on their own websites. They may also have been named in a summary of all inquiry responses put together by the Swedish Ministry of the Environment (Miljödepartementet, 2020).

Limiting the amount of material for a discourse analysis is difficult, as there is no specific set of sampling technique as in quantitative studies. When saturation is reached through the material is also difficult to decide, as there are always further competing interpretations out there to take account of. As the main focus of this study is verified emissions reductions as part of the climate policy framework and SOU 2020:4, a natural limit for collection of the empirical material occurred and became the inquiry in itself and its responses.

5.4.2 Interviews

To broaden the finding found in text, interviews were conducted with four actors in the policy process for the climate policy framework. The actors interviewed are representatives of government agencies, researchers and independent environmental consultants. All have a role in the implementation and creation of the framework. Unfortunately no invited civil society organization were able to participate within the time frame of the in the study, though attempts were made to contact several. Their perspectives are therefore accounted through their response to SOU 2020:4 in the empirical material.

For this study a decisions was made that the interviews were not to be used as part of the empirical material or discourse analysis. Instead, the interviews are triangulated with the results found. This was done for three main reasons. 1) An interviewer will by design steer an interview and decide frames for what the participant can say, making it less valuable (Kvale et al., 2014). 2) A question of power and ethics in the interview situation occurs. A “normal” interview situation expects the interviewer to document the participants actual answers, emotions or accounting of an issue. But discourse analysis is more interested in how these things are said and how it establishes construction in the conversation itself. Some argue this is a breach of trust and unethical way to approach the power dynamics in an interview (Kvale et al., 2014). 3) The time constraints and the amount of interviews needed provide a meaningful material were also part of this decision.

The interview guide (Appendix A) shows a few main open-ended interview questions following the research problem, conducted in a semi-structured fashion. These overarching questions allowed for the interviewed actors to prioritize what they conceived as important of carbon market without much steering from the interviewer (Newcomer et al., 2015). More specific follow-up questions were also included, to aid if the interview strayed or did not fully contribute to the research questions. Depending on the nature of the interview, these were not always necessary to include (Newcomer et al., 2015; Esaiasson, 2017). No personal data or sensitive materials were collected through the interviews. Nevertheless, due to the employment of some participants some questions asked could be more or less sensitive. To ensure the participants felt comfortable answering the questions candidly, full anonymity was provided.

The participants were provided a consent form (Appendix D) prior to their interview where they were informed of how their data would be handled and processed. The participants were also informed of their right to withdraw consent at any time. Three of the interviews were recorded over video link. One participant declined to be recorded, and detailed notes were instead taken

(22)

20

during the interview. This of course limits the way the content can be used, but the interview still provided valuable context and background to the study.

5.5 Considerations and Limitations of the Method

Interpreting text and presenting the results is by design challenging. Can the method answer the proposed question in a credibly way? Discourse analysis as a method brings with it several challenges. Reading and reviewing text, the bias of the researcher should not be underestimated (Boréus & Bergström, 2017). What parts of the text are seen as important, how it is understood and interpreted will by design be carried by the knowledge and pre-understanding of the researcher. Critics of these interpretive methods raise how they by design can become idealistic pursuits, and cannot produce scientific results (Breeze, 2011). However, this may also be argued true for all research ventures, as science is not inherently unbiased and value-free (Sismondo, 2010). For those within interpretive, critical fields, part of the research assumption is moreover that competing interpretations and storylines will always be present. This is then not necessarily a problem, but simply the nature of the approach to science and something to be aware of (Fairclough, 2003).

This raises the question about reliability. As Hajer (1995) approaches discourse from a perspective in social constructivism, where there is no one truth to found – rather many competing truths with actors trying make sense of discourse. The same must hold true for the researchers. Two researchers looking at the same material will likely not find the same nuances, storylines and ideas. This as they carry with them different pre-knowledge, bias and general understanding of the world. Therefore different researchers coming to the same conclusion is not necessarily possible or desirable for discourse analysis – or a sign of reliability. What is however crucial for discourse analysis in terms of reliability, is that the step-by-step process is clear and appropriately motivated and accounts for the researchers potential bias (Boréus & Bergström, 2017). Reliability for discourse analysis thus becomes less about the actual similarities in results, as it does not expect the same conclusions drawn by different researchers. Rather, reliability is found in the transparency and ability to follow the method step-by-step. With a theoretical background in social construction and discourse, this is especially important, due to the interpretive nature of these approaches. In this study, the assumption is that competing discourses and storylines exist outside what is found in the material. This is only natural, as the background I as a researcher provide can only interpret them from my background and experiences.

In terms of the material, the challenge in using the responses to the inquiry SOU 2020:4 was how they are framed in the context of negative emissions in a Swedish perspective. As such, part of the framing and assumption of what carbon offsets could be had already been made by the material. The actors responding to the inquiry were also most often invited to reply – making the choice of actors represented in the study random. Other perspectives not found in this selection of actors could be possible and skew the result. However, this again ties in to the above discussion, that other interpretations of both research problem and the material used always exists. Which is not necessarily a problem, but something that needs to be considered and acknowledged when conducting the analysis.

(23)

21

6. A Brief Introduction to Carbon and Offset Markets

6.1 The History of Carbon Markets

International market mechanisms for emission reductions have become a common feature of climate policy – for states but also for individuals and private actors. Carbon offsetting is one such market mechanism. While there are different markets with different standards, carbon offsetting boils down to a (theoretically) simple transaction. A buyer of a carbon credit (or offset) can use the fruit of mitigation efforts elsewhere to meet a domestic/internal emissions reduction target (Lovell, 2010). The calculus behind carbon markets overall comes down to a simple point: As carbon is global, the location of carbon emissions reductions does not matter, just that they occur somewhere. Therefore, it is most efficient to reduce emissions where it is most inexpensive (Andersson & Carton, 2017; Lohmann, 2011).

While carbon markets have been theorized previously, they first became an international reality through the negotiations of the Kyoto Protocol in 1997. In the Kyoto context countries from the Global North (Annex 1 countries) had targets for emissions reductions, while developing countries were exempt from emissions reductions commitments (Michaelowa et al., 2019b). This difference in commitments was understood as a way to honor the different historic responsibilities in causing the emissions crisis. Certain Annex 1 economies raised the argument for implementing market mechanisms because of their ability to lower the cost of compliance to mitigation targets. Further arguments of how the market would be a driver for innovation, development and growth were raised (Lohmann, 2011).

As a result of the negotiations in the United Nations Framework Convention on Climate Change (UNFCCC), three main market mechanisms for carbon were produced under the Kyoto Protocol. One emissions trading system and two mechanisms based on projects for offsets. The offset mechanisms were called Joint Implementation (JI) and the Clean Development Mechanism (CDM). JI produced projects by countries with emissions reductions under the Kyoto protocol, while the CDM produced projects by countries with no emissions reductions commitments under Kyoto. However, over the history of carbon offset market, two main markets became dominant for the use of offsets. The compliance offset market as represented by the CDM and the voluntary carbon offset market, outside the UN system.

The process for these markets goes as follows. A carbon offset is represented through tons of CO2 equivalents (tCO2e). One ton acquired in offset credits represents the equivalent emissions reduction not yet released into the atmosphere. These credits are then sold or traded on the international carbon offset market (Bumpus & Liverman, 2008). In simple terms, an offset project is produced in the Global South creating an emissions reduction. A state or actor from the Global North buys this offset credit, and accounts for the emissions reduction towards their own mitigation commitments. The projects financed through offsetting projects can be efforts such as (re)forestry, investments in renewable energy, energy efficiency and land- and soil management (World Bank, 2020).

6.1.1 The Carbon Offset Markets

The Clean Development Mechanism (CDM) was created to be under international oversight with standard set and approved by actors within the UNFCCC. Here, only actors signed to the Kyoto Protocol could become buyers or host projects for offsets. To govern the compliance

(24)

22

market the UNFCCC created multiple layers of panels, committees and boards for project accreditations. Subsequently, the process of certification and verification in the compliance market was complex, resulting in high barriers of entry for both buyers and project developers (Lovell, 2010; Bumpus & Liverman, 2008). While states are the signing parties and the primary intended actors in this market, private actors were eventually able to purchase credits from the CDM. Through the EU emissions trading system (EU ETS) for example, different linking measures made it possible for companies within the EU ETS system to utilize these credits towards their own emissions reduction targets.

Projects produced under the CDM must ensure environmental integrity, meaning they should not jeopardize the natural system further. For a project to become an offset credit, a long line of actors are involved before verification. The complexity made rigorous governance and standards important to avoid fraud and to ensure the co-benefits for the actors hosting projects (Lovell, 2010). It also ensured the overall goal of environmental integrity through a few key terms:

• Projects must prove they are additional. I.e. the emissions reduction relating to the project would not have happened independent of it. Without the criteria of additionality fulfilled, credits can be created as an afterthought – inflating the market (Michaelowa, 2019b).

• Projects and their additionality must be calculated in accordance to a correct baseline. Baselines show the trajectory of emissions that would occur in the absence of the project. Practically, this means that projects should show results of emissions below the Business as Usual (BAU) baseline. However calculations of baselines are complex, and their accuracy questioned (Michaelowa, 2019b).

• A credit must only be issued once or be purchased by one buyer. Otherwise a credit may be subject to double counting, inflating the emissions reduction made.

While the CDM had a golden period of operation, especially after the EU ETS was introduced, by 2012 it had started a slow decline. Falling demand for credits combined with increasing supply of credits and falling prices created an unstable environment for actors resulting in a weakened market (Michaelowa et al. 2019b).

The voluntary carbon offset market is still growing today, accounting for more than half the credits issued in 2019 (World Bank, 2020). The voluntary offset market is a parallel system born out of private initiatives, and not under international supervision. The voluntary market has become particularly popular for individuals and corporations – used to compensate all from flights to internal corporate sustainability targets. With no international oversight, the voluntary market operates under standards brought forward by private actors and NGOs (Lovell, 2010). Notable standards used are Gold Standard and Verified Carbon Standard. Here projects developers, intermediates such as retailers and carbon brokers and the end buyer are the main actors in the project. Credits in this market are known as VCOs or Verified Emissions Reductions, VER. While the voluntary market is independent, it is still intrinsically tied to the CDM, as it was created in part to mirror its functions while bypassing perceived bureaucratic bottlenecks (Green, 2013; Hoffman 2011). Due to the rigorous and often lengthy process of certification through the CDM, the voluntary market became an alternative where actors could rapidly receive credits (Lovell, 2010).

(25)

23

6.1.2 Main Criticism of Carbon Offset Markets Under Kyoto

While carbon offsets have become common practice, they are deeply polarizing as a policy arrangement. In the academic debate many have questioned the counterfactual quantification that carbon offsetting rests upon, and suggested that it produces a false understandings of emissions reductions (Lohmann, 2005; Gunderson et al. 2018; Blum, 2020; Lane & Newell, 2016). That is, one ton of CO2 equivalent exactly corresponding to the planting of a certain amount of forest in the Global South. Or the planting of forests and building renewable energy capacity becoming equivalent in climate benefits through quantified credits (Lohmann, 2008; Bumpus & Liverman, 2008). This process of quantification is important, as it is a way to separate carbon from its natural context, into one of economic and social value. As it transforms through construction, carbon becomes an entity of political value. Thus it can be used to exert power (Lövbrand & Stripple, 2011). Similarly, some argue that the market structure of carbon markets makes them difficult to remove from climate policy. Any market failure is seen as technological or based on efficiency – with the only remedy being further technological and market measures (Leonardi, 2017; Stuart et al, 2019).

After 20 years of use and evaluation some have also argued the carbon offsetting mechanism simply has not produced any tangible proof of reduced emissions (Hulme, 2019; McAfee, 2016). Regardless of if the practice works practically as a decarbonization tool, there are structural and procedural worries in the market. The carbon offset markets under the Kyoto Protocol have been particularly scrutinized, due to issues of accounting and verification with false inflation of actual emissions reductions (Michaelowa et al. 2019a).

Supporters have long advocated for carbon offsets as a measure with comparatively low mitigation costs and flexibility as a decarbonization tool – with the ability to fill ambition gaps left by states. Markets are seen as providing the greatest amount of good possible with the lowest amount of cost (Kaswan, 2011; Bumpus & Liverman, 2011; McAfee, 2016). Sceptics however raise criticism of a zero-sum game allowing for lowered ambitions from states and other actors – with scarce evidence of co-benefits for project countries, and sometimes direct harm (Okereke & Coventry, 2016; Turnhout et al., 2017). The line from credit to emissions reduction is long, with the distance creating difficulty in verification of a, emissions reductions overall but also b, the effect of projects or mitigation efforts locally.

While carbon as a physical entity is not necessarily harmful or unjust – the mechanisms for emissions reduction or distribution of carbon rights can create disproportionate burden and harm (Kaswan, 2011). A growing body of literature presents how offsetting projects - while claiming benefits in terms of climate; displaces populations, erodes livelihoods and erases local technology and knowledge making. Examples of this can be land grabbing for reforestation, building dams that changes water supply or failure to adapt projects to local needs - introducing the thought pattern of the Global South as a carbon sink to be utilized by the Global North. (Cavanagh & Benjaminsen, 2014; Lohmann, 2008; Okereke & Coventry, 2016). Others however see that projects have created benefits in different ways, but comes with challenges, where reform of the market is key for success (Simon et al. 2012; Michaelowa, 2011).

6.1.3 The Future of Carbon Offsets Markets Under Article 6

For carbon offset markets under Article 6 of the Paris Agreement, a whole new landscape for governance is made possible – including challenges and opportunities. Article 6 of the Paris Agreement opens up for the continued use of carbon markets, however the structure and form remains pending as negotiations are ongoing (with a rulebook expected after COP26 in Glasgow

Carbon Storylines : The discursive struggle over carbon offsets as a decarbonization pathway in the Swedish Climate Policy Framework

Sandra Ideskär

Carbon Storylines

The discursive struggle over carbon offsets as a

decarbonization pathway in the Swedish Climate

Policy Framework

Table of Contents

1. Abstract

2. List of Abbreviations

3. Introduction

3.1

Aim and Research Questions

3.2

Thesis structure

3.3

Delimitations

4. Theoretical Background

4.1

Discourse and the Making of Meaning

4.2

Argumentative Discourse Analysis

4.3

Policy Discourses

4.4

Theoretical framework

5. Method and Material

5.1

Operationalizing Argumentative Discourse Analysis

Storyline

5.3

Material

5.4

Data Collection

5.5

Considerations and Limitations of the Method

6. A Brief Introduction to Carbon and Offset Markets

6.1

The History of Carbon Markets