Frame Disputes or Frame Consensus? “Environment” or “Welfare” First Amongst Climate Strike Protesters

Abstract

Present debates suppose a close linkage between economic, social, and environmental sustainability and suggest that individual wellbeing and living standards need to be understood as directly linked to environmental concerns. Because social movements are often seen as an avant-garde in pushing for change, this article analyzes climate protesters’ support for three key frames in current periods of social transformation, i.e., an “environmental”, an “economic growth”, and a “welfare” frame. The analyzed data material consists of survey responses from over 900 participants in six Global Climate Strikes held in Sweden during 2019. The article investigates the explanatory relevance of three factors: (a) political and ideological orientation, (b) movement involvement, and (c) social characteristics. The results indicate that climate protesters to a large degree support an environmental frame before an economic growth-oriented frame, whereas the situation is more complex regarding support for a welfare frame vis-á-vis an environmental frame. The strongest factors explaining frame support include social characteristics (gender) and protestors’ political and ideological orientation. Movement involvement has limited significance. The article shows how these frames form a fragment of the complexity of these issues, and instances of frame distinctions, hierarchies, and disputes emerge within the most current forms of climate change demonstrations.

1. Introduction

Current debates about climate change and sustainability raise a series of questions on how we organize our societies and the degree to which human activities, living standards, and consumption patterns are (un)sustainable and affect climate change (and vice versa). It is common to think of and discuss sustainability in terms of triangular concepts or as three interrelated systems, domains, or spheres in relation to the environment, society, and the economy [1]. This way of understanding sustainability stems from the Brundtland Report from late 1980s [2] and the proposal of a triple bottom line (People, Planet, Profit) model. This view has had significant impacts in both academic and public debates and has been consolidated once again under the recent United Nations’ Sustainable Development Agenda, i.e., Agenda 2030 [3]. The model builds on the assumption that sustainability requires environmental protection, individual wellbeing, and economic growth simultaneously. The recently proposed notion of sustainable welfare [4], however, further stresses the interconnectedness between welfare arrangements to promote good living standards among the population and environmental sustainability in terms of the use of limited natural resources. Thus, we cannot treat welfare systems or their outputs and the environment as two separate spheres. Through welfare state arrangements, a welfare sphere is largely about the provision and achievement of adequate or even substantial forms of material living standards through the (re)distribution of resources, but an environmental sphere is about respecting the earth’s system processes and staying within planetary boundaries. Whereas academics have explored the conflict between economic growth and the use of natural resources, less discussed is the complex relation between arrangements that seek to promote welfare among the public and how to satisfy such needs within ecological limits. This is a pressing issue because most European welfare states build upon a structure that is at odds with notions of both sustainable development and sustainable welfare, and we can observe a strong correlation between high socio-economic living standards and large ecological footprints [5].

Even though there is an increasingly widespread understanding that we need to change our societies in terms of issues of individual living standards and wellbeing, it is much more disputed how such paths of transition can take place and whether welfare states are part of the problem or part of the solution. While welfare states (at least some of them) have played a key role in promoting good living standards and redistributed economic resources across the population, such welfare arrangements can potentially be at odds with environmental sustainability as more people have more money to spend. Present mobilizations by environmental and climate movements play a key role in capturing the relationship between social welfare and sustainability, not only because movements can be seen as an avant-garde pushing for social change, but even more so because recent protests have managed to mobilize extensive numbers of protestors on a global scale and thus have highlighted the urgency of environmental concerns amongst significant sections of the global population. In August 2018, the Swedish primary school pupil Greta Thunberg decided to demand forcible climate action from the politicians through a school strike outside of the Swedish parliament during the national election campaign. After the election in September, she started to have school strikes only on Fridays, a protest tactic that since then has been adopted by millions of school students worldwide under the banner “Fridays For Future” (FFF). Note, however, that the FFF slogan is not used in every country taking part in the protests. But since the protests have the school strike in common this article, in line with Wahlström and his colleagues [6], proceeds from the conceptualization of FFF when referring to the global climate protests. Apart from the weekly school strikes taking place all over the world, global climate strikes have in the last year gathered millions of people, peaking during the Global Week for Future (20–27 September 2019) with approximately 6 million participants in the streets across the world [7].

It is certainly beyond doubt that these demonstrations push the public and political agendas on climate change, but how do protesters capture the relation between promoting people’s welfare compared to protecting the environment and dealing with climate change? Or to put it sharply, do protesters put the “environment” first or “welfare” first in their views on societal change? Or do protestors see them as mutually supportive and potentially compatible and not politically and ideologically contradictory? Moreover, what are their views on environmental protection versus economic growth, which Western welfare states have built upon since the mid-20th century and which have been debated within the sustainable welfare literature (e.g., [8,9,10])? This article analyzes climate protesters’ support for three key frames, namely an “environmental” frame (protecting the environment and reducing CO₂ emissions), an “economic growth-oriented” frame, and a “welfare” frame (welfare arrangements that promote good living standards for the population). While we find ample studies into environmental movements and climate justice movements, we find fewer studies that direct attention to the activists’ views on and support for these frames when they are contrasted to each other.

Based on an original data set with protesters in the most recent FFF demonstrations and Global Strikes, we furthermore contribute to current debates on what explains frame support among climate activists as we analyze why certain protesters support one particular frame and whether protestors express support for multiple frames. Our article thus offers an original approach to current debates on sustainable development and sustainable welfare solutions as social movements are often seen as an avant-garde in pushing for social change.

This article draws upon a recent protest survey study of participants in six FFF demonstrations (Global Climate Strikes) in Sweden from March to September 2019. During these protests, we performed more than 700 face-to-face interviews and obtained over 900 web survey responses from protest participants. Sweden is a particularly ample context to study welfare and environmental frame support. First, Sweden is often seen as a key model of an advanced welfare state with strong state and public support for providing material welfare and well-being for “all” based on notions of social citizenship and social rights [11]. Previous studies on attitudes towards sustainable welfare have shown that Swedish people in general tend to hold both pro-environmental and pro-welfare attitudes [12]. Second, Sweden has a strong and long-lasting environmental movement—the movement’s most recent expression in the form of Global Climate Strikes or FFF originated in Sweden before it took off globally.

2. Environmental Movements, Climate Justice, and FFF

Environmental movements have been, and continue to be, fundamental in bringing about change in public opinion and media reporting, and they have played a key role in the establishment of new scientific disciplines and research institutes. They have also paved the way for many NGOs to be recognized as reliable partners in international negotiations. As such, environmental movements are some of the most debated and well-known forms of social movements.

Academics have put much focus on environmental movements and have studied differences between countries and across continents, their connections to political parties, their influence, and their values and social basis [13,14,15,16,17]. Throughout the last decades, environmental movement research has come to focus on issues of climate change, environmental justice, and climate justice [18,19,20]. In their analysis of the environmental justice movement, Brulle and Pellows [21] show how environmental justice concerns have gained significance, often as grass-root mobilizations with local origins protesting against the use of pesticides and local waste dumps and expressing claims of unjust environmental risks (see also [22]). According to Rootes and Brulle, such an environmental justice frame has increasingly accommodated not only traditional environmental organizations, but also the environmental movement more generally [18]. Recent developments also point towards the emergence of what Schlosberg and Collins express as a form of climate justice movement [20] (see also [14]), illustrating a growing tide of groups and activists taking up issues of North–South divides and the unjust distribution of the risks and burdens associated with climate change. Cassegård and Thörn [23] argue that the climate justice agenda grew stronger as a result of an internal critique towards established environmental groups as being too institutionalized. Instead, claims of more radical solutions were brought forward. Hadden [24] suggests that the climate justice agenda was also the result of spillover effects from the global justice movement with clear anti-capitalist expressions [25]. The formation of a climate justice movement has also gained ground thanks to the opportunities opening up at international climate summits like the climate summit in Bali (2008, COP 13), the summit in Copenhagen (2009, COP15), and more recent ones [24,26]. Currently the movement is accommodated by several leading networks and organizations working with climate change and climate justice (e.g., Extinction Rebellion).

The most recent forms of climate change protests certainly build upon, but also differ from, previous ones. In sharp contrast to other activities, which occurred in direct correspondence to international climate summits, the recent protests emanate from the school strike developed by the Swedish pupil Greta Thunberg, which in its turn has evolved into an international phenomenon with strikes and demonstrations on a global scale. When Thunberg started her school strike on a daily basis outside the Swedish parliament in Stockholm in August 2018, three weeks before the Swedish general election, it was a protest against politicians not paying enough attention to the climate issue. Quite soon other students started to join her in their effort to get politicians to acknowledge the emergency and crisis of climate change and other environmental problems. After the Swedish parliamentary election on 9 September 2018, Thunberg and other protesters decided to continue the school strikes on a weekly basis every Friday [27,28,29]. Under banners like “Fridays For Future”, “School Strike 4 Climate”, “Youth for Climate”, and “Youth Strike for Climate”, millions of people have since entered the streets to express their frustration and anxiety over climate change and what they perceive as a lack of political leadership to accommodate such changes in a sufficient way. In a research report on the first Global Climate Strike on 15 March 2019, organized through the global FFF network, it is described as “a historical turn in climate activism” [6] (p. 6). At the end of 2018, the school strikes spread not only all around Sweden, but also around the entire world [29]. In one of the most recent Global Climate Strike demonstrations in Stockholm, it was reported that 60,000 protesters took part [30], making it one of the largest protests in post-war Sweden. Demonstrations in other countries also attracted very large numbers of protesters, ranging from 100,000 protesters respectively in Melbourne and London to around 200,000 in Rome and Berlin, and to more than 300,000 protesters in Montreal [31,32,33,34]. It is beyond doubt that this has served as an invigoration for the environmental movement as such and a broadening of its social base as new groups have joined the movement and participated in its demonstrations.

3. Frames, Frame Disputes, and Frame Support

The framing perspective in social movement studies highlights how movements produce and maintain the meaning of relevant events and circumstances (e.g., climate change and social injustices) in order to both mobilize supporters and demobilize antagonists [35], but also as way to gain favorable media coverage and win political victories [36]. Frames thus have key significance for movements, and it is easy to agree with Polletta and Kai Ho who argue that “frames matter”. They draw attention to what is relevant and irrelevant, and they tie different elements—arguments, beliefs, emotions, and experiences—into a (potentially) coherent message that makes (plausible) sense for participants. They might reconstitute or alter the meaning of an issue, or issues, at stake [35,37]. Frames thus not only matter, they also have a transformative capacity by turning “… routine grievances or misfortunes into injustices or mobilizing grievances in the context of collective action” [35] (p. 384). What come out of these framing activities are referred to as collective action frames that include “… innovative articulations and elaborations of existing ideologies or sets of beliefs and ideas, and thus function as extensions of or antidotes to them” [35] (p. 401). Central to collective action frames are, for instance, the kinds of framing activities that have an action-oriented function in the sense of negotiating and defining a shared understanding of the problem or situation in need of change, but also who or what to blame, and so forth. Prognostic framing lies at the center of this paper because it “... involves the articulation of a proposed solution to the problem” [37] (p. 616). Which prognostic frames protesters support (e.g., an environmental or a welfare-oriented frame, as in the case of this paper) is of course an empirical question. Moreover, frames are embedded into wider discursive fields [35] (p. 401) that demarcate the distinct cultural and structural contexts (encompassing beliefs, values, and ideologies) that in turn have an impact on framing processes.

Much theorizing on frames and framing processes also emphasizes relations between frames. Such relations can be seen as conflicting by addressing the potential tensions within movements over how to articulate the wider ambitions and aims of the movement (e.g., [38,39]). The notion of a frame dispute [40] demonstrates that frames are plural, complex, and might build or even cause fractions and rifts within movements. Movements that attract a wide collection of participants, like the most recent forms of climate strikes, plausibly embody different and varied forms of frame support. That is, framing processes contain elements of ordering and sorting and emphasizing and highlighting, which make certain events or issues more important than others [37]. In this way, one idea or belief might “... become more salient in an array or hierarchy of movement-relevant topics or issues” [37] (p. 398), which will become apparent further below.

This article furthermore draws on another strand of the framing literature that pays much more interest to individual frame support. Ketelaars, Walgrave, and Wouters, for instance, have declared “an individualistic and explanatory turn” in framing studies [41] (p. 342). Along these lines, our review suggests three broad types of factors that can explain why certain groups of participants tend to support one frame over another: (a) political and ideological orientation, (b) movement involvement, and (c) social characteristics.

Studies into political and ideological orientation emphasize that personal values and individual attitudes have key significance for individuals’ support of movements and particular frames. Wahlström, Wennerhag, and Rootes, for instance, found that climate protesters identifying as “right” on the left–right scale were less likely to formulate their prognostic frames as demands for “system change” or “global justice” in comparison to left-oriented climate protesters [42]. Left-leaning tendencies among climate protesters have been reported in other studies on the environmental movement [43]. This ties into wider debates on support for different political positions. Linde [44], for instance, showed a positive correlation between identification with left and green parties and support for environmental policies. Moreover, the left–right division has long been associated with socio-economic conflicts regarding the degree to which the state should regulate markets, engage in economic redistribution, and organize common welfare systems, where a left position has meant a positive attitude towards state regulation, redistribution, and welfare [45,46]. The correlation between left-wing orientation and support for economic redistribution and redistributive social policies is further confirmed by numerous studies (e.g., [47,48,49,50]). Note, however, that the correlation of individual interest and egalitarian ideology with welfare state attitudes has been questioned [51]; thus, researchers have included another set of subjective factors, such as multiculturalism, merit, authoritarianism, gender traditionalism, and generalized trust. Even though all of them correlated with welfare attitudes, none of them had a consistent effect across the 26 European countries that were included in the study [49].

When it comes to movement involvement, studies on social movements emphasize social movement organizations as key to the mobilization of participants (e.g., [52]) and for the understanding of their frame support. Participants’ organizational affiliation and previous experiences of participating in different demonstrations can thus be seen as expressions of a movement identification and thus as an inclination to express support for a particular issue. Membership in particular organizations can, hence, be seen as providing exposure to and potential support for a particular frame [53]. It should be noted, however, that there might be a varying degree of commitment to social movement organizations or movements among protest participants [41]. Nonetheless, we assume that individuals’ affiliation with a specific type of movement organization is positively correlated with their support for the specific frames that are central to the movement. Along the same lines, it can be assumed that there is an association between previous experiences of participating in different demonstrations for various causes (e.g., environmental causes) and individual participants’ framings (e.g., support for an environmental frame).

Lastly, research also stresses that different social characteristics such as gender, education, and so forth, might influence one’s support for specific frames. A study of climate activists found that women and people aged 30–49 were more likely than men and younger people to advocate solutions based on changes in individual behavior and that women were less likely to advocate system-oriented prognostic frames [42]. Previous studies about the social composition of environmental protestors in Western Europe have shown that well-educated individuals with middle-class occupations are overrepresented, a pattern quite typical of protests of new social movements [42,54] (see also reference [41] for social compositions in different kinds of protests, and not only environmental protests). Although previous studies have not found any overrepresentation of young people or women in environmental protests, the media coverage of recent environmental protests seems to suggest that the climate strikes under the banner of FFF are more dominated by young school students and women. As we will see in the descriptive data of this article, this is also the case in the surveyed climate strike demonstrations in Sweden. This has significance for the interpretation of our data because studies show that young women (compared to other parts of the population) tend to be in favor of various kinds of climate policies [55]. Moreover, studies of attitudes towards economic redistribution and welfare policies show that attitudes vary due to individuals’ class, age, gender, income, occupation, and education (e.g., [47,48,49,50]). Often, these socio-demographic variables have been related to various social groups’ self-interest in the economic gains from welfare programs. Studies show, for instance, that education has a negative effect on welfare support, but that women are more prone to support the welfare state compared to men (e.g., [48,49]). Regarding women, different theories have been used to explain welfare support. While some theories point to self-interest, others argue that it is the socialization process that makes women more concerned about the welfare of others [56] (p. 400). In a similar way, pro-environmental attitudes among women have been explained in relation to self-interest, but also to women’s concern for the maintenance of life and relationships because of women’s traditional role as caretakers [57]. Thus, the specific social composition of climate strikers will be of relevance for our analysis of support for environmental frames.

The theoretical discussions together with previous research and literature can be summarized in Figure 1.

Each relation in the figure contains a particular set of hypotheses:

Hypothesis 1:

Protesters who place themselves on the left of the political spectrum or hold economic egalitarian values prefer a welfare frame before an environmental frame.

Hypothesis 2:

Protesters who identify with green and radical left parties prefer an environmental frame before an economic growth frame.

Hypothesis 3:

Protesters who are part of the environmental movement context prefer an environmental frame before an economic growth frame and a welfare frame.

Hypothesis 4:

Protesters who are part of social equality-oriented movement context prefer a welfare frame before an environmental frame.

4. Method, Data, and Study Design

This article employed a quantitative approach for analyzing protest participants’ support for specific collective action frames, which is rare when it comes to studies of frames within social movements (see, however, [41,42]). In studying the support among Swedish climate protesters for certain frames concerning priorities in relation to the environment, welfare, and economic growth, a survey-based quantitative method allowed us to analyze more general patterns of potential frame disputes and whether ideological and socio-demographic factors as well as individuals’ movement affiliation affect the support for certain frames.

4.1. Sampling

The data collection followed the well-established and standardized protest survey method of the research program Caught in the act of protest—Contextualizing Contestation (CCC) [58]. This procedure can be regarded as a systematic random sampling method because it aims at giving all participants of a protest an equal chance of being sampled and thus minimizes sampling bias. During each demonstration, protest participants were systematically selected according to a common protocol. For instance, if the demonstration was estimated to have approximately 10,000 participants, every 10th demonstrator was asked to complete the survey. Two or more members of the research team were assigned as “pointers”, which, following this systematic selection, determined which protesters the interviewers of the research team should approach. In Malmö, on March 15, a total population sampling strategy was employed, however, since the number of survey flyers were more than the number of protestors and since the demonstration was static. The selected respondents were given a flyer with basic information about the survey, the web address to the survey (also in the form of a QR code), and a unique numerical code to get access to the online survey. In addition, every fifth respondent was interviewed face-to-face. Through these interviews, which followed a short, single-sheet survey, basic data were collected to make it possible to control for non-response bias. Because almost all respondents replied to the face-to-face interviews, the data generated in this way could then be contrasted with the data from the online survey, which had a much lower response rate. Respondents under 15 years old were informed that they needed to send us a signed parental consent document to be allowed to participate in the web survey. If they were in the company of their parents at the demonstration, the parents could sign a parental consent document on the spot. See

Table 1. Response rates of the surveyed Global Climate Strikes in Sweden, 2019.

	15 March Malmö	15 March Stockholm	24 May Stockholm	27 September Malmö	27 September Stockholm	27 September Gothenburg	Total
Estimated number of participants	600–650	3000–5000	6000–10,000	1500	40,000–60,000	5000–10,000	56,100–87,150
Number of survey flyers distributed	528	588	914	633	658	211	3532
Number of face-to-face interviews	95	108	160	174	138	41	716
Number of web survey responses	119	179	254	184	132	60	928
Response rate, web survey (%)	23	30	28	29	20	28	26

for the numbers of interviews and response rates in Stockholm, Gothenburg, and Malmö regarding the surveyed Global Climate Strikes in Sweden during 2019.

4.2. Operationalization of the Dependent Variables

In order to capture support for, as well as potential tensions or disputes between, an environmental frame, a welfare frame, and an economic growth frame, our dependent variables functioned as a proxy of predefined frames. The environmental frame was operationalized through statements about protection of the environment and reducing CO₂ emissions (for a detailed description of all survey questions analyzed in this article, see Appendix A). The welfare frame was operationalized through statements about social welfare arrangements, and the economic growth frame was operationalized through statements about economic growth and jobs. We measured support for these three frames through the following two statements: (1) “Protecting the environment should be given priority, even if it causes slower economic growth and some loss of jobs”, and (2) “Measures to decrease CO₂ emissions cannot be allowed to make social welfare arrangements worse”. In the first statement the environmental frame is contrasted to the economic growth frame, and in the second statement the environmental frame is contrasted to the welfare frame. In the Swedish context—being an advanced welfare state with extensive public welfare arrangements to provide adequate and good living conditions—social welfare arrangements generally connote tax-funded social insurance systems and welfare services that provide individual welfare and wellbeing. Moreover, the environmental frame is operationalized slightly differently in the statements, and even though it could be argued that it should be operationalized in a similar way, we argue that “measures to decrease CO₂ emissions” is closely linked to understandings of protecting the environment, and especially because this statement follows directly after the statement that contains the direct phrase “protecting the environment”. Respondents were asked to rate their agreement with these statements on a five-point scale from “strongly disagree” (1) to “strongly agree” (5).

4.3. Operationalization of the Independent Variables

The variables measuring movement involvement included the respondents’ organizational affiliation and previous participation in demonstrations (see Appendix A for a detailed description). Organizational affiliation was operationalized through active membership in environmental, social equality-oriented, or political party organizations. Respondents were asked to indicate whether they had been active or passive members, or not members at all, in different types of organizations during the previous 12 months. We measured previous participation in demonstrations through questions about how many times (ever) the respondents had participated in environmental or climate change marches, May Day marches, and 8 March (International Women’s Day) demonstrations.

The protesters’ ideological stance was operationalized through self-placement on the left–right political spectrum, political party identification, and degree of support for economic egalitarian values. To measure self-placement of the political spectrum, respondents were asked to place themselves on an 11-point scale from left (0) to right (10) or to choose the alternative “to me, this categorization is meaningless”. Political party identification was measured through the question “With which political party do you most closely identify right now?” with a text box where respondents could fill in the name of the party or choose the alternative “I don’t identify with any political party”. To measure economic egalitarian values, respondents were asked to rate their agreement with the statement “Government should redistribute income from the better off to the those who are less well off” on a five-point scale from “strongly disagree” (1) to “strongly agree” (5). In line with previous literature on public attitudes and protesters’ framings as discussed above, the following control variables were used: age, gender, education, and class identity.

5. Results

5.1. Descriptive Statistics—Independent and Control Variables

The socio-demographic characteristics of the protesters in the Swedish Global Climate Strikes deviated to some extent from previous research about who participates in demonstrations. This was especially the case when it came to gender. While much previous research has shown equal numbers of women and men participating in demonstrations (e.g., [25,41,42]), women were overrepresented in the Global Climate Strikes with around 63% of the total participants (see

Table 2. Socio-demographic characteristics of climate protesters (percent and total numbers).

	Total (%)	Total (N)
Gender
Women	62.8	514
Men	35.3	289
Other gender identity	2.0	16
Age
Under 15 years	3.8	35
15–19 years	20.2	187
20–35 years	24.5	227
36–65 years	35.6	330
Older than 65 years	16.1	149
Education
No university degree	31.5	258
University degree/studying at university	68.5	560
Class Identity
Upper class	1.7	14
Upper middle class	52.1	424
Lower middle class	32.3	263
Working class	4.9	40
Lower class	0.5	4
None	3.4	28

). In terms of age, almost half of the respondents were younger than 35 years old. These figures furthermore show that not only school students participated in the climate strike demonstrations, in contrast to how these protests are sometimes portrayed in the media. In line with previous research about protesters’ educational level (e.g., [25,41,42]), which is often regarded as a proxy for social class, a large share of the participants included current or former university students. This is once again reflected in the question about the respondents’ class identity (their subjective identification with certain class strata), which showed that around 85% of the participants identified as middle class.

When it comes to movement involvement, the numbers in

Table 3. Climate protesters’ organizational affiliation (percent and total numbers).

Member in Organization (during Last 12 Months)	Not Member (%)	Passive Member (%)	Active Member (%)	Total (%)	Total (N)
Environmental	63.3	27.4	9.3	100.0	842
Charity or humanitarian	66.0	29.1	4.9	100.0	842
Trade union	66.2	28.6	5.2	100.0	842
Political party	78.6	12.9	8.4	100.0	842
Women’s	93.5	4.6	1.9	100.0	842
LGBTI	95.4	3.8	0.8	100.0	842

below indicated that a large share of the participants were not members in any environmental or social equality-oriented organizations. Almost one in ten of the climate protestors were active in environmental organizations, and slightly fewer were active in political parties. Around 5% of the participants were active in charity or humanitarian organizations and trade unions, respectively. A very small part was active in women’s and LGBTI organizations. A large proportion of the participants, however, had demonstration experiences from various kinds of protests, including Pride parades and 8 March demonstrations. Almost three in four had previous experiences of climate change or environmental demonstrations, and a slight majority of May Day marches, while somewhat fewer had taken part in Pride parades and 8 March demonstrations (see

Table 4. Climate protesters’ demonstration experiences (percent and total numbers).

Previous Protest Participation (Ever)	Never (%)	1–5 (%)	6–10 (%)	11–20 (%)	21 or More (%)	Total (%)	Total (N)
Any type of demonstration	13.8	36.6	13.9	10.3	25.4	100.0	826
Climate change or environmental marches	27.1	56.1	10.6	3.1	3.1	100.0	818
May Day marches	47.2	28.4	12.1	5.6	6.7	100.0	803
8 March demonstration (International Women’s Day)	59.9	29.8	6.3	2.8	1.3	100.0	798
Pride parades	57.6	36.2	4.6	1.4	0.2	100.0	802

) For a comparison with the general Swedish population regarding their participation in May Day marches and Pride parades, see Wennerhag [59].

Concerning the variables measuring ideology—including identification with political party, left–right self-placement, and support for egalitarian economic values—protesters primarily identified with green and left-wing parties and placed themselves to the left of the political spectrum (see

Table 5. Climate protesters’ political orientation and economic egalitarian values (percent and total numbers).

	Total (%)	Total (N)
Identification with Political Parties
Green parties (incl. animal rights parties)	31.1	251
Democratic socialist parties (radical left)	29.0	234
Social Democratic parties (center-left)	5.9	48
Liberal parties (center-right)	5.7	46
Feminist parties	3.8	31
Christian Democrat and Liberal-conservative parties	1.0	8
Other political parties	1.6	13
Do not identify with any political party	21.8	176
Left–Right Self-Placement
Left (0–3)	66.1	542
Center (4–6)	24.9	204
Right (7–10)	3.7	30
No Left–Right position	5.4	44
“Government should Redistribute Income from the Better off to Those Who are Less Well off”
Strongly disagree (1)	2.4	20
Disagree (2)	4.9	40
Neither disagree nor agree (3)	23.1	189
Agree (4)	27.5	225
Strongly agree (5)	42.1	344

). Also, almost 70% of the participants believed that governments should redistribute income from the better off to those who are less well off. Thus, a large share held egalitarian economic values. The left-leaning tendencies among the climate protesters were in line with previous research on the environmental movement (e.g., [43]).

5.2. Descriptive Statistics—the Environmental Frame, the Welfare Frame, and the Economic Growth Frame

When the climate protesters had to choose between the environmental frame and the economic growth frame, they tended to support the former before the latter (see

Table 6. Climate protesters’ frame support.

“To What Extent Do You Agree or Disagree with the Following Statements?”	Strongly Disagree (%)	Disagree (%)	Neither Disagree nor Agree (%)	Agree (%)	Strongly Agree (%)	Total (%)	Total (N)
“Protecting the environment should be given priority, even if it causes slower economic growth and some loss of jobs”	0.6	0.7	4.3	19.6	74.9	100.0	891
“Measures to decrease CO2 emissions cannot be allowed to make social welfare arrangements worse”	12.2	20.8	37.5	19.1	10.4	100.0	888

). As much as 74.9% strongly agreed with the statement “Protecting the environment should be given priority, even if it causes slower economic growth and some loss of jobs”, and only around 1% disagreed with this statement. The result differed when it came to indicating support for the welfare frame or the environmental frame. When the welfare frame was contrasted to the environmental frame through the statement “Measures to decrease CO₂ emissions cannot be allowed to make social welfare arrangements worse”, protesters’ support leaned less towards one frame compared to another. It was instead more evenly distributed, and a fair number of protesters, 37.5%, selected the mid-option in terms of neither disagreeing nor agreeing with the statement.

5.3. Explaining Climate Protesters’ Frame Support

In order to scrutinize what factors most strongly affected climate protesters’ support for environmental, welfare, and economic growth-oriented frames, we performed a regression analysis of the two variables discussed above. In different regression models, we used some social characteristics (age, gender, educational level, and class identity) as control variables in combination with relevant variables for the protestors’ political and ideological orientation and their movement involvement. The movement involvement variables focused on both active membership in, and previous participation in, demonstrations staged by environmental organizations and social equality-oriented organizations (trade unions and women’s organizations). The variables for political and ideological orientation focused on the respondents’ left–right self-placement, their support for economic redistribution, and what kinds of political parties they identified with. The results of these regression models are shown in

Table 7. Binary logistic regression of predictors for the environmental frame versus the economic frame.

	Model 1A		Model 1B		Model 1C		Model 1D		Model 1E
Variable	Fixed Effects		Organizational and Movement Involvement		Left–Right Position and Redistribution Support		Party Identification		Accumulated Model
Control variables
Age (>20 years = ref.)
≤20 years	0.120	*	0.116	*	0.100	*	0.137	**	0.130	*
Gender (man = ref.)
Woman	0.097	**	0.111	**	0.094	**	0.085	*	0.112	**
Other gender identity	0.066	†	0.069	†	0.047		0.049		0.044
University degree/studying at university	0.073		0.057		0.052		0.030		0.025
Subjective class (lower classes = ref.)
Upper classes	−0.027		−0.025		0.026		0.007		0.033
No class identification	−0.003		0.003		0.026		0.013		0.021
Active member in organization
Environmental organization			0.057						0.061	†
Trade union			−0.070	†					−0.068	†
Women’s organization			−0.061	†					−0.072	*
Previous protest participation
Environmental or climate march			0.040						0.026
May Day march			0.020						−0.035
8 March demonstration			−0.014						−0.076	†
Left–Right self-placement (right = ref.)
Left					0.111				0.047
Center					−0.051				−0.119
No Left–Right position					0.093	†			0.049
Support for economic redistribution
Strongly disagree (1)–Strongly agree (5)					0.092	*			0.120	**
Party identification (ref. = center–right liberal and conservative parties)
Social Democrat parties (center–left)							0.003		−0.014
Democratic Socialist parties (radical left)							0.241	***	0.129
Green parties							0.240	***	0.151	†
Other parties							0.133	**	0.083
Do not identify with any party							0.197	**	0.142	*
Observations	770		745		763		753		724
Adjusted R2	0.032		0.037		0.070		0.055		0.094

Note. Linear regression. Standardized beta-coefficients are shown in the columns. † 10% significance. * 5% significance. ** 1% significance. *** 0.1% significance.

and

Table 8. Binary logistic regression of predictors for the welfare frame versus the environmental frame.

	Model 2A		Model 2B		Model 2C		Model 2D		Model 2E
Variable	Fixed Effects		Organizational and Movement Involvement		Left–Right Position and Redistribution Support		Party Identification		Accumulated Model
Control variables
Age (>20 years = ref.)
≤20 years	0.008		−0.035		−0.013		0.019		−0.033
Gender (man = ref.)
Woman	0.091	*	0.077	*	0.102	**	0.109	**	0.110	**
Other gender identity	−0.032		−0.040		−0.037		−0.033		−0.035
University degree/studying at university	−0.120	*	−0.119	*	−0.125	*	−0.111	*	−0.100	†
Subjective class (lower classes = ref.)
Upper classes	−0.028		0.001		−0.012		−0.013		0.003
No class identification	−0.020		0.007		−0.006		−0.011		0.012
Active member in organization
Environmental organization			−0.031						−0.020
Trade union			0.023						0.014
Women’s organization			0.006						0.002
Previous protest participation
Environmental or climate march			0.013						0.035
May Day march			0.093	*					0.049
8 March demonstration			0.058						0.033
Left–Right self-placement (right = ref.)
Left					−0.117				−0.170
Center					−0.113				−0.119
No Left–Right position					−0.041				−0.070
Support for economic redistribution
Strongly disagree (1)–Strongly agree (5)					0.110	**			0.087	†
Party identification (ref. = center–right liberal and conservative parties)
Social Democrat parties (center–left)							0.095	*	0.106	†
Democratic Socialist parties (radical left)							0.108		0.092
Green parties							−0.054		−0.029
Other parties							0.018		0.015
Do not identify with any party							0.082		0.094
Observations	767		741		760		750		720
Adjusted R2	0.017		0.023		0.024		0.039		0.037

Note. Linear regression. Standardized beta-coefficients are shown in the columns. † 10% significance. * 5% significance. ** 1% significance. *** 0.1% significance.

.

When it comes to the role of social characteristics for support for the environmental frame before the economic growth-oriented frame, Table 7 shows that this was stronger among women and respondents aged 20 years or older, whereas educational level or class identity did not affect this. Regarding movement involvement, the analysis showed higher support among active members of environmental organizations and lower support among active members in trade unions and women’s organizations and among those having participated in 8 March demonstrations. The protestors’ political and ideological orientations, however, had a greater impact on their support for the environmental frame than their degree of movement involvement did. Being positive to economic redistribution, identification with a green party, or not identifying with any political party, led to greater support for the environmental frame (vis-à-vis the economic growth-oriented frame), whereas the respondents’ left–right self-placement did not have any effect on this.

When it comes to social characteristics’ role in the support for the welfare frame before the environmental frame, Table 8 shows that this was stronger among women and weaker among the most educated. Neither age nor class identity affected the degree of support for this frame. Movement involvement only had an effect for those who had taken part in a May Day march and who supported the welfare frame. A related pattern was also seen for party identification, where those identifying with Social Democratic parties expressed more support for the welfare frame (vis-à-vis the environmental frame), which was also the case for those being positive to economic redistribution.

For both regressions, the strongest effects contributing to frame support were related to social characteristics (and in both cases, gender) and to the protestors’ political and ideological orientation, whereas the effect of their movement involvement was not so strong. One can also note that the figures for explained variance (adjusted R²) were quite small, especially regarding the welfare frame, which showed that there were other, yet undiscovered, factors that contributed more to the variation in the support for these frames.

6. Discussion

This article tests three broad sets of factors to explain support for environmental, welfare, and economic growth-oriented frames within the most recent wave of protests in the environmental movement, namely the Global Climate Strike demonstrations that were primarily organized by the FFF network. Our explanatory variables include political and ideological orientation, movement involvement, and social characteristics.

Our results indicate that climate protesters to a large degree hold opinions that are “pro-environment” rather than “pro-economy”, and there is wide support for putting “the environment” and concerns for the environment before “the economy” and economic growth. This is hardly surprising because the demonstrations, per se, mobilize regarding climate change, and engagement in this issue is one of the main reasons for participating in the protests. A growth-centered market economy can be seen as an adversary for those who are environmentally concerned and engaged in the environmental movement. The divide between concerns for the environment and economic growth to a large extent follows along the lines of protesters’ political and ideological orientation, and those identifying with green and radical left parties tend to emphasize an environmental frame before an economic growth frame. Another strong predictor for putting the environment before economic growth is holding egalitarian values. The support for one of these frames instead of the other is thus related to the individual’s wider ideological orientation rather than just their identification with particular political parties. These results are in line with previous literature and research on the environmental movement (e.g., [42,43]), but also more generally with research on public support for environmental policies (e.g., [44]).

However, the protesters seem to be partially split regarding their relative priority between the environmental frame and the economic growth-oriented frame. Participants active in trade unions serve as a key illustration in this respect. Here we find that trade unionists are less likely to put “the environment” first compared to environmentalists. Arguably, even though the protesters are united against a growth-centered economy and in acting for the environment, we find more fine-tuned differences among participants. Factors related to movement involvement explain less than factors related to political and ideological orientation. This might be an indication that views on environmental, welfare, and economic growth-oriented concerns, as illustrated through frame support, are more deeply embedded in personal values than movement involvement.

Whereas the support for the environmental frame vis-á-vis the economic growth-oriented frame was straightforward, the situation is much more complex regarding support for the welfare frame vis-á-vis the environmental frame. It is apparent that climate protesters hold more diverse views and opinions as to whether to prioritize the welfare frame or the environmental frame when they are contrasted with each other. Our results show that there is no unanimous support for one of these frames over the other among the protesters. This is further expressed by the large proportion of respondents who “neither disagree/nor agree” with the statements. This can be interpreted in different ways. On the one hand, this could indicate indecisiveness among respondents because they might not “have made up their mind”. On the other hand, this might reflect a more profound puzzle for protest participants because they are confronted with two frames that both come with positive connotations—being pro-environment and pro-welfare. Compared to taking a stance towards the environmental frame or the economic growth frame, this seems to be more difficult in relation to the welfare frame when contrasted to the environmental frame. This seems to be even more difficult when it comes to sorting and ranking them into a hierarchy of good and bad frames. The degree to which this is an expression of frame disputes or frame dilemmas is a matter for further research.

Once again, political and ideological orientation, but still only economic egalitarian values and not self-placement on the left–right scale, have a larger impact on frame support than movement involvement does. When it comes to social characteristics, our results show that those with higher education tend to support the environmental frame before the welfare frame. Potentially, this illustrates a tension over whose welfare is at stake here, someone else’s or mine. People with higher education in general have higher incomes and thus have potentially less to lose from poorer welfare arrangements; thus, it does not come as a surprise that they support the environmental frame before the welfare frame. Debates within the field of welfare attitudes find strong links between self-interest and support for redistributive policies [48,49,56]. This also seems to be relevant here, but much more expressed as a climate concern rather than a concern for another individuals’ welfare.

It is interesting to note that when it comes to ideological orientation, holding economic egalitarian values is a central explanatory factor when it comes to support for the environmental frame (before an economic growth-oriented frame) and for the welfare frame (before an environmental frame). That egalitarian values are positively correlated with support for the welfare frame should come as no surprise, but it is not obvious that economic egalitarianism should lead to support for environmental concerns at the expense of economic growth and full employment. Our results for the more traditional indicator of ideological orientation, left–right self-placement, are even more puzzling. While previous research on welfare attitudes has found strong support between left orientation and support for welfare policies (e.g., [45,46,47,48,49]), our study shows no correlation between left orientation and support for the welfare frame. Left orientation is, however, positively correlated with support for the environmental frame when this is contrasted with the economic growth-oriented frame. When left–right orientation is analyzed in the regression together with support for economic redistribution, the explanatory effect of left orientation more or less disappears, showing that for this frame the left–right dimension in the end boils down to holding economic egalitarian values. Why the protesters’ left–right orientation matters for their attitudes in value conflicts between environment and economic growth, but not in value conflicts between welfare and environment, certainly deserves more attention in future research. Perhaps this is due to a changed meaning of the left–right divide in contemporary societies in which “left” has become more associated with green, alternative, and libertarian values (what is today usually abbreviated as GAL and contrasted with TAN, i.e., traditional, authoritarian, and nationalist values) instead of the traditional meaning of “left” as being in favor of economic redistribution and market regulation [60] (pp. 8–9).

Another key social characteristic that has explanatory power concerning frame support regards gender and being a woman. Women support the environmental frame before the economic growth-oriented frame, but also the welfare frame before the environmental frame. The fact that women are more pro-environment and pro-welfare compared to men has been shown in previous research [48,49,55,57]. Thus, our results concerning the environmental frame vis-á-vis the economic growth frame in relation to gender do not come as a surprise. A more puzzling result is women’s support for the welfare frame before the environmental frame. Even though previous research stresses both culturally embedded forms of social responsibility and self-interest, it is still unclear why women, when having to choose between the welfare frame and the environmental frame, prefer welfare concerns before environmental concerns.

While much literature on social movements and political participation stresses that active involvement in social movement organizations contributes to activists’ frame support and political attitudes, such factors play little or no role for explaining variation in this study. Previous movement involvement does not matter much, neither in terms of active membership (some support) nor in terms of protest participation (almost no support). This does not need to be at odds with previous studies and theorizing, and perhaps this is due to the fact that the movement-involved activists found among the protesters of these recent climate change protests have not yet been active long enough to become socialized into supporting specific frames. Another interpretation of the lack of significance of these factors is that the kinds of frame support studied in this article go beyond organizational activity or previous protest participation and instead relate more to personal ideological convictions. The frames in question concern more profound ideological questions about movement adversaries (i.e., economic growth) and what needs to be protected (i.e., welfare or the environment). Especially regarding the latter, there are clear forms of ideological disagreements among movement participants that potentially constitute key divides among participants.

The Figure 2 summarizes our results and our hypotheses concerning frame support and/or frame disputes for environmental, welfare, and economic growth-oriented frames.

We find strong support for egalitarian values, but not for left–right orientation, when it comes to putting the welfare frame before the environmental frame (H1). We find limited support for the relevance of political party identification (H2). We also find some support for our third hypothesis (H3), as active membership in environmental organizations has a somewhat positive impact on emphasizing the environmental frame before the economic growth-oriented frame. However, we find no support for our fourth hypothesis (H4), as protesters who are part of a social equality-oriented movement context do not emphasize the welfare frame before the environmental frame. A limitation in our analyses is that our regressions only seem to explain a small part of the variation in frame support, especially when it comes to preference for the welfare frame over the environmental frame. This points to the need for further research focusing on other factors that might contribute more strongly to this variation. In addition, it would be interesting to explore frame support and potential frame disputes or dilemmas concerning the environment, welfare, and economic growth among climate activists in other welfare states because the connotations and understandings of these frames, especially the welfare frame, might vary between different welfare states.

7. Concluding Remarks

To sum up, this article shows instances of frame distinctions, hierarchies, and disputes within the most current forms of climate change demonstrations. The demarcation between environmental concerns and economic growth serves as a key for most protesters. This is a frame distinction that follows the long history of environmental movements mobilizing against a liberal market economy, and more recently against economic growth as a central feature of how we have organized our economies and our societies more broadly. The distinction against economic growth can be interpreted as a form of “adversary making” in the present climate change movement. It orders beliefs, ideas, and topics into a hierarchy of (non-)relevant movement-related frames. Beyond doubt, movement participants put the environment first when contrasted to economic growth. The simple ordering and hierarchy of issues and topics is, however, less evident when addressing welfare or environmental concerns. It is far from evident that protest participants in the most recent forms of climate change mobilization put the environment before welfare concerns, illustrating wider disputes and ideological rifts among participants about how to actually handle climate change and whether individuals’ welfare needs should be subordinate to the present pressure of climate change.

Finally, just as recent discussions on sustainable development and sustainable welfare address the interconnectedness between the environmental, social, and economic domains, there is a widespread argument that these spheres or domains need to be integrated in order to promote the societal change necessary to handle current pressures on the planet and to provide welfare and wellbeing within planetary boundaries. This article only captures a fragment of the complexity of these issues, yet the tensions we find among climate protesters most likely are tensions that are even more widely expressed among the population at large, i.e., that present forms of climate change should not have too great an impact on our welfare.