The (In)Visibility of Gender in Scandinavian Climate Policy-Making

Abstract This article explores the link between gender representation and climate policy-making in Scandinavia. We ask to what extent equal descriptive representation (critical mass) results in substantive representation (critical acts). Our study shows that women and men are equally represented in administrative and political units involved in climate policy-making, and in some units women are in the majority. However, a text analysis of the outcomes, that is, the Scandinavian climate strategies, reveals a silence regarding gender, further confirmed through interviews. Accordingly, a critical mass of women does not automatically result in gender-sensitive climate policy-making, recognizing established gender differences in material conditions and in attitudes toward climate issues. In interviews, we also note that policy-makers are largely unaware of gender differences on climate issues in the Scandinavian context. We discuss why a critical mass of women in climate policy-making has not led to critical acts and offer alternative explanations informed by feminist IR theory. For example, poststructural feminism claims that masculine norms are deeply institutionalized in climate institutions; hence, policy-makers adapt their actions to the masculinized institutional environment. Thus, substantive representation should be understood in relation to gendered institutional processes.


INTRODUCTION
This article explores the link between climate governance and gender representation in Scandinavia. Research on political representation has advanced the understanding of women's impact on policy-making, showing that women's presence affects both substance and emphasis in particular policies (Lovenduski 2005;Wängnerud 2009). This project studies gender and climate change by analyzing the possible effects of equal gender representation on climate policy-making. Through a quantitative analysis of crossnational data, Ergas and York (2012) established that carbon dioxide (CO 2) emissions per capita are lower in nations where women have higher political status. 1 This effect is significant when controlled for other factors, for example, world system position or industrialization. Their work suggests that gender equity is a relevant associated issue with climate change. Our ambition is to explore this link in Scandinavia.
Based on Ergas and York's (2012) research, it is reasonable to assume that in more gender equitable nations such as the Scandinavian states, there will not only be more efforts geared toward CO 2 emissions reduction, but also that climate governance strategies will show signs of gender sensitivity. This assumption is based on representation research that has advanced the understanding of women's impact on policy-making and constitutes an important theoretical foundation of the study. It is also founded in feminist international relations (IR) theory; liberal, standpoint and poststructural theory is used to analyze the results. The Scandinavian states are particularly interesting as they have an international image as forerunners in environmental and gender politics (Ingebritsen 2006;Magnú sdó ttir 2010Magnú sdó ttir , 2012 and have promoted the image of being "climate-and gender-friendly" (e.g. via the Nordic Council 2009). We explore how gender is relevant to Scandinavian climate policy-making by analyzing the relationship between descriptive and substantive representation.
Climate change cannot be reduced to just a matter of CO 2 emissions. While it may be a measurement for evaluating whether climate efforts have been successful, there is a whole range of other greenhouse gases (GHG) to be considered (IPCC 2013). Most importantly, climate change is the result of human activities over a long period of industrialization and related to the dependency on fossil fuels in most societies and sectors. The transport and energy sectors are particularly implicated and these policy areas demand action from policy-makers at international, regional, national and local levels. 2 Levin et al. (2012, 128) characterize climate change as a "super wicked" problem, a characterization that implies challenges for policymakers due to uncertainty, complexity and time dimensions (Levin et al. 2012, 128). The climate problem's wickedness is tied to human activities, and to a need to question current behaviors and lifestyles and their great dependence on fossil fuels (Soper 2009). However, the climate change solutions proposed so far by policy-makers have primarily focused on technical fixes and economic measures. This article builds on the assumption put forth by the scientific community through the Intergovernmental Panel on Climate Change (IPCC) that the stipulated reductions in greenhouse gas emissions as well as necessary societal changes must be radical in order to reach the goals. The efforts to deal with climate change will pose challenges to political institutions and call for societal re-organization. Thus, in our view, it is of utmost importance that equal representation is assured so that a multitude of voices is heard and diverse interests and experiences allowed to influence climate policy. The aim of this article is to analyze whether this has been the case in Scandinavia.
The article begins by arguing the rationale of the cases described and explaining the methodology. We then develop the article's theoretical framework based on feminist and gender research on representation. This helps explicate the relationship between women's involvement in climate governance and their impact on the substance of climate strategies. It concerns how representation can move from critical mass to critical acts. The remainder of the article will analyze the empirical findings using this framework. We begin with a survey of descriptive representation, and then advance to study and problematize substantive representation.

Case Selection, Methods and Material
In this explorative comparative case study, we followed the most similar case design, analyzing states with many commonalities apart from their gender and environmental profile. The Nordic states in question are interesting cases due to their reputation as pioneers in gender politics 3 as well as climate change issues, 4 but also since they have already taken steps in connecting gender and climate change. In the run-up to the Copenhagen COP 15 meeting in 2009, the Nordic Council emphasized the necessity of taking gender aspects seriously by promoting a "Nordic gender climate regime" (Nordic Council 2009). This was followed by the web portal "Equal Climate" that launched during the 2011 Nordic Council session, themed "The Green North, a Gender-equal Actor." Our ambition was to enquire into women's institutional representation beyond the top political level by focusing on committees and administrative bodies where policy initiatives are designed and detailed strategies developed. We did this mainly through the following qualitative methods: (1) mapping of representation in eleven key state bodies dealing with climate change in Sweden, Norway and Denmark; 5 (2) text analysis of relevant climate strategies; and (3) interviews with policy-makers in the aforementioned key institutions. Through the text analysis, we explored whether and how equal descriptive representation of women and men had made gender visible in key documents through a search for explicit as well as implicit references to what we stipulated as substantive representation in climate policy. In fact, in the end, simple references to keywords or any indications that the population was not considered a homogeneous group were noted in our analysis. Thirteen key documents were selected 6 for the text analysis, including climate change roadmaps and strategies from 2007 to 2012. The purpose of the interviews was to explore the policy-makers' perceptions of the proposed connection between gender and climate change issues, and if they were aware of any particular policies or policy measures dealing with gender and climate issues. We conducted a total of sixty-nine short semi-structured interviews (in person or via telephone) with officials in the state units in question. 7 The interviews were gender-balanced, evenly distributed between the three states in question and anonymous. The main selection criteria for the interviewees were that they should preferably be middle-ranking officials and unit heads involved in the day-to-day formation of Scandinavian climate policies.

WHY SHOULD WE EXPECT A LINK BETWEEN WOMEN, GENDER AND LOWER CO 2 EMISSIONS?
The theoretical problem addressed in this article is whether and how women's representation matters in climate issues. Representation in politics or the lack thereof has been an important topic in gender studies and for the women's movement since early suffrage debates (Lovenduski 2005, 1). It is also central to debates on democracy because concerns about representation usually signal a deficiency and imply making something present when in some sense it is not. If gender representation is lacking in democratic institutions, then it is a sign of an ill-functioning democracy or a democratic deficit.
Feminist international relations theory hinges on a marked critique of women's marginalization from global politics (e.g. Enloe 1989;Pettman 1996) and invisibility in international relations theories (e.g. Tickner 1992; Sylvester 1994), but has not explored representation conceptually. Most studies have been empirical, analyzing women's role in international organizations (Jain 2005;Caglar, Prü gl, and Zwingel 2013) and in transnational activism (Meyer and Prü gl 1999;Hawkesworth 2006;Confortini 2012), often with a focus on particular issues of concern to women (e.g. Joachim 2007).
Feminist perspectives in international relations (IR) (e.g. Tickner 1992;Sylvester 2002, 159 -181;Youngs 2004;Peterson and Runyan 2010) are useful to discuss the results of the analysis. Descriptive representation has been analyzed most extensively within the empiricist and liberal feminist studies elaborated below. Whether women are represented in international bodies often falls back on national institutions, policies and actions (Peterson and Runyan 2010, 103 -142), and may be why it has been discussed less frequently in feminist IR. This study thus makes a contribution in bridging feminist research on representation with feminist IR.
The meaning of the concept of representation has been actively debated. The interest of gender and feminist scholars has been to increase the engagement of women in politics, that is, to increase gender parity. Political theorists are inclined to use several different categories of representation (Pitkin 1967;Mansbridge 2003). Phillips (1995) warns that it is not sufficient to rely only on the representation of common political ideas, through political parties. A representative has to be responsive to the interests and concerns of the electorate while including simultaneously appreciation of difference in those concerns and interests (Phillips 2010). Taking a starting point in Phillips' (1995) politics of presence, two main categories have been developed while exploring women's representation in public service (Lovenduski and Norris 2003). The first is descriptive representation of women and men in elected bodies and concerns the relationship and gender balance in political institutions. The share of seats in government, parliament and in the nomination lists of political parties is commonly used as a way to indicate inclusion of women (Wängnerud 2009, 53). There is a fair amount of empirical research on descriptive representation. Without going into much detail here, the main focus of that research has been on the presence of more women in politics and how the integration of women in politics is affected by various different structural, cultural and political factors. Important factors that favor gender parity include a representative democratic system and the presence of leftist parties, shown to facilitate the election of women. Also, parties that favor environmental and social values over economic ones tend to have a higher representation of women (Wängnerud 2009, 55).
As shown in gender research, when women are represented in greater numbers, it is often the result of formal or informal quotas (Krook 2009;Dahlerup and Freidenvall 2010). We note that this approach qualifies descriptive representation with the argument that a few women in politics will not make a real difference. A link between descriptive and substantive representation is established with the "critical mass" argument originally formulated for the business sector by Moss Kanter (1977). Moss Kanter argued that a certain number of women need to be present for their presence to be felt. Dahlerup (1988) developed the critical mass concept for politics, suggesting that the number of women determine the magnitude of the effect. While no exact number has been established, informal gender equality quotas used today aim for the 40 to 60 percent representation range (Niskanen 2011). Concepts like critical mass and gender quotas have been useful and important in establishing a female presence in politics; however, a sole focus on numbers is problematic. Instead, Dahlerup (2006, 517) argues that it is "critical acts" that lead to substantive representation rather than a critical mass. This proposition begs the questions: will women make a difference once they have been included in the polity and if so, what kind of difference will that be?
The democratic quality of the polity clearly improves when there is a larger community of female politicians and decision-makers (Phillips 2010). The much trickier question regards what substantive effects, if any, female representatives will bring to policy activities. According to Wängnerud (2009, 59), the link between descriptive and substantive representation is more of a probability than determined empirically. Such links have nevertheless been established. With evidence from New Zealand, Grey (2006) states that descriptive representation makes it far easier to introduce feminist policies. Franceschet (2011, 60) explains that the difficulty in finding evidence and clear linkage is due to the complexity of women's substantive representation, "being best conceived as a multifaceted and dynamic process that includes both legislative activities and outcomes." Childs and Webb (2012) who write on female representatives in British politics also point out the multifaceted character of substantive representation. They conclude that it "is not just about the attitudes and behavior of individual representatives. Nor is it just about putting women's issues on the table. It is also about evaluating the extent to which legislation and public policy is changed" (Childs and Webb 2012, 221).
This study will focus particularly on policy outcomes in the climate issue area, paying attention specifically to critical acts. Critical acts are expected to have implications for policy processes as well as decisions and policies because the (political) representative thinks and acts in a deliberative manner. While we recognize, as does Franceshet (2011, 59), that these acts can have lasting implications for policy institutions, this article is limited to addressing substantive representation as outcome only. We capture this through our focus on policy documents and interviews.
Substantive representation demands attention to the question of whether women's contributions to political decisions are in any way different from men's contributions (Christiansen and Raum 1999). Liberal feminism is uneasy about the link between descriptive and substantive representation due to this implicit assumption, particularly as it lacks full empirical verification in research. According to standpoint feminism in IR (Zalewski 1993;Kronsell 2005), women's presence and representation in policy-making is expected to have substantive impacts based on women's different interests, views and knowledge. Standpoint theory posits that women's position in society shapes ways of knowing and provides other experiences than those lived by the dominant male elites (Hartsock 1985;Harding 1991). Poststructural feminist IR scholars tend to address this in terms of women's "crises of representation." Women are not part of the political elites that exercise global power, due to the presence of exclusive gendered and masculinist norms of international relations (Peterson and Runyan 2010, 104). Thus, both descriptive and substantive representation would reflect global power structures in the absence of structural change.
An interesting point to note from the Nordic context is that Nordic gender quotas (especially in Norway) are not merely perceived as a matter of equal democratic representation and social justice as is the case in Germany and France (Freidenvall and Krook 2010), but are viewed as tools to include women's experiences and views, presumed to differ from those of men (Christiansen and Raum 1999). Accordingly, complete representation is also about acknowledging that women may have different views and experiences that need to be included as important contributions to gendered climate policy-making. Thus, the effects of women's representation in the Nordic context have made democracy more robust, but significant links between female representation and actual gender equality have been difficult to establish here (Wängnerud and Sundell 2012, 119). In the analysis that follows, we will map and discuss descriptive representation in climate policy-making ministries and agencies in Scandinavia.

DESCRIPTIVE REPRESENTATION IN CLIMATE POLICY-MAKING BODIES
Our intention in this section is not to give a complete account of women's involvement in climate policy-making in the Nordic states but to map women's and men's representations in the climate units selected for the study. They are deemed well suited to give us sufficient indications about descriptive representation in climate governance in Sweden, Denmark and Norway. There are favorable statistics regarding gender equality in Nordic politics compared to the fact that on average, women only hold 23 percent of seats in national parliaments in other European states (Interparliamentary Union 2012). Women hold 30 to 45 percent of seats in all the Nordic national parliaments, which factors into the positive image of Nordic gender equality (Interparliamentary Union 2012).
Our findings reveal that a large majority of all the state units explored are either gender-balanced or have a descriptive overrepresentation of women (see Figure 1). Although this is well in line with the Scandinavian gender profile based upon a higher level of institutionalization of gender equality than in most other industrialized states, the fact that women were overrepresented in several units was somewhat surprising.
As demonstrated in Figure 1, every Scandinavian energy agency explored, is gender-balanced. This is also bewildering since energy issues in other contexts are male-dominated (Clancy and Roehr 2003, 19; Carlsson-Kanyama, Ripa, and Rö hr 2010), mainly in terms of how industries, entrepreneurs (cf. Carlsson-Kanyama, Ripa, and Rö hr 2010), and labor are organized and how transport patterns are gendered (Uteng and Cresswell 2008). Officials interviewed had also noted this. A Norwegian official stated: "The technical side of climate politics is worldwide quite male-dominated but that is not the case in Norway." 8 The only units explored that are not gender-balanced (or overrepresented by women) are the Scandinavian transport agencies. The proportion of officials in the Danish Transport Authority is 66 percent male; the Norwegian transport agency, Transnova, is 70 percent male, although the management board is gender-balanced, with 60 percent women and 40 percent men. The Swedish Transport Agency is the most gender-balanced of the three transport agencies, with 36 percent of female officials, but reaches the gender equity limit at the management level with 40 percent of the management being women (TrA 2012, 85 -86). Even with the overrepresentation of women in some climate units, there is still a strong sense that transport issues are masculine coded (Polk 2009a).
It was not surprising that the majority of the Scandinavian state units explored were gender-balanced, but it was unexpected that some even exceeded the general equality norm of 40 to 60 percent female. For liberal feminists, equal descriptive representation would be a satisfactory sign of gender equality in Nordic climate change policy-making, but we side with those feminist scholars who claim that participation in politics is not limited to equal descriptive representation in policy and decision-making. Participation also concerns the content and quality of decisions. In order to find comprehensive answers about the possible effects of equal descriptive representation, we explore substantive representation through the presence of gender in policy documents.

SUBSTANTIVE REPRESENTATION: HAS A CRITICAL MASS TRANSFORMED INTO CRITICAL ACTS?
We have established that the majority of the Scandinavian units explored are gender-equal but as Dahlerup (1988) suggested, when discussing equal representation, one needs to move beyond any discussion of a critical mass (in this case, the proportion of female officials) and focus on critical acts. Critical acts in climate policy-making in Scandinavia would be those acts that have come as a result of equal representation and have led to policy changes and institutional changes. As a first step, we suggest that evidence of critical acts in climate policy-making should be found in the most important climate policy documents.
In order to understand what substantive representation in climate policy entails, neither feminist work on representation nor feminist IR were sufficient on their own. We propose that substantive representation in climate policy should reflect what has been argued in various literatures on gender and climate change (Rö hr et al. 2008;Alaimo 2009;Hemmati and Rö hr 2009;Mac-Gregor 2010;Kronsell 2013), in feminist analyses on women's relation to nature (Merchant 1980;Plumwood 1993;Mellor 1997;Agarwal 2010;Gaard 2011) and literature that also highlight the necessity of understanding gender and climate as intersecting power relations (Sandilands 1999;Cudworth 2005;Salleh 2009;Mallory 2010;Kaijser and Kronsell 2013). We used a loose framework based on these literatures for the text analysis.
The enquiry was conducted through a text analysis of selected documents on the climate goals of the governments in question. 9 The documents from all three countries included in the study completely lacked any recognition of gender. Hence, we were not able to develop the framework for substantive representation through our empirical findings as we had hoped. Instead, we ended up with a text analysis that used as a baseline an awareness of the gender implications of climate issues. The documents from gender-balanced institutions compared with those from the transport agencies, which are male-dominated, did not reveal any difference in regards to gender recognition. The identical invisibility of gender and lack of gender awareness in all the documents explored was indeed puzzling as it indicated the lack of a link between descriptive and substantive representation in climate policymaking.
We probed the material in more depth. Inspired by Bacchi's work (1999) on policy analysis, we suspected that there were certain "silences" on gender (Ackerly, Stern, and True 2006) and climate issues. Hence, most references to gender relevant issues would be implicit, meaning that we had somehow missed this in our first analysis. Bacchi's approach not only focuses on what is included, but also on what is excluded in problem definitions and knowledge presentation, thus on what is perhaps silent and invisible.
Using a more sensitive approach in comparing the content of the Norwegian, Swedish and Danish climate roadmaps (Danish Energy Agency 2007;KLIF 2010;NOU 2010:10;Naturvårdsverket 2012) revealed a subtle difference in approaches and in ways of thinking about what climate change strategies should entail. However, we cannot claim that this is an effect of gender. In brief, the Norwegian documents had a more holistic and inclusive approach. The Norwegian roadmap was also grounded in social sciences, focusing on socioeconomic consequences and societal vulnerability. On the other hand, the Danish roadmap did not appear to take socioeconomic factors into consideration and focused primarily on new technology for energy efficiency as a strategic measure. In the Swedish Roadmap for 2050 (Naturvårdsverket 2012), 10 there was no mention of gender aspects at all and the focus was mainly on general policy instruments emphasizing CO 2 taxes and certain investments. While the roadmap recognized in a broad and sweeping way that behavioral changes will be both necessary and likely, this was not clearly spelled out. The emphasis was on the need for societal planning but scant attention was paid to issues that relate to people's lived lives, like cycling or public transportation. The more holistic approach of the Norwegian roadmap appeared to call for more extensive changes in different layers of society than the Danish or Swedish strategies, which took a narrower and more technical approach often associated with masculinized climate governance (Plumwood 1993;Salleh 2009;Kronsell 2013). There is no explicit reference to gender in the Norwegian strategy either, but it still appears to acknowledge to some extent that the Norwegian population is not a homogeneous group and we are not all affected by climate change in the same way, although this refers to ethnic minorities only (the Sami people), not gender, age or class (KLIF 2010).
To better understand the (in)visibility of gender in Nordic climate change policies, it was deemed necessary to interview the officials involved in the production of the documents and/or in the everyday policy-making. In 2011 and 2012, we contacted a total of sixty-nine male and female officials 11 at the aforementioned ministries and agencies in person or by telephone and asked whether they considered the gender perspective to be relevant for climate governance. The majority of Nordic officials, approximately two-thirds of Norwegian and Danish officials interviewed and three-quarters of Swedish officials interviewed did not deny the relevancy of gender but were very uncertain about what gender in climate change entailed in the Scandinavian context. Thus they appeared to lack knowledge or insight rather than the will to include gender as a relevant issue for Scandinavian climate policy-making. Typical expressions were: "I don't know if gender is relevant, I don't know enough to speak about it with any credibility." 12 Or as a Danish official explained: "Gender might be relevant when it comes to transport since women use public transportation more, but it is not visible in my work or within my unit." 13 Several interviewees pointed to gender as significant in climate policymaking in developing states but then claimed that due to different living standards in the North and due to the Nordic gender equality model, gender was not relevant here. A Norwegian official claimed: "Gender does not play any role at all in the Norwegian context where we have a very gender-balanced environment but it is definitely important in developing states." 14 A Swedish official expressed similar sentiments: "It feels like gender is perhaps not so relevant nationally but that it can be globally, in developing countries and for the women there." 15 They are correctly informed. The average differences regarding their ecological impact, their vulnerability to climate change and their contribution to greenhouse gas emissions is larger between the women in the global North and the global South, than between women and men in the North (Johnsson-Latham 2007). However, there are considerable and relevant gender differences regarding these factors in the Scandinavian countries as well, particularly when class, age, place and ethnicity are considered (Polk 2009b;Räty and Carlsson-Kanyama 2010). Räty and Carlsson-Kanyama's study on energy consumption by gender in Europe from 2010 for example is interesting in this context and reveals that the total energy consumption of Norwegian and Swedish single men is considerably higher than the energy consumption of Norwegian and Swedish single women (Räty and Carlsson-Kanyama 2010).
In sum, Nordic officials were not necessarily indifferent, but rather unable to see the link between gender and climate change in the Nordic context, while some recognized the relevance when it came to poor countries in the global South. It should be noted that while more studies on gender and climate change have been conducted on developing countries, there is ample evidence that climate issues in the North have gendered implications both regarding causes and effects (Polk 2009b;MacGregor 2010;Räty and Carlsson-Kanyama 2010;Nagel 2012;Kaijser and Kronsell 2013). Also worth mentioning is the fact that the views and standpoints of the policy-makers at the Scandinavian transport agencies, the only units where women were underrepresented, did not differ from the views of other interviewees, and there was no detectable gender difference in the overall views of the respondents. It can thus be concluded from this exploration that equal descriptive representation does not automatically result in substantive representation where gender is a visible factor.

DISCUSSION
Our analysis of women's representation in climate institutions in Scandinavia confirmed the presence of women in various institutions dealing with climate strategies and policies. There were variations between countries and types of institutions but also cases where descriptive representation was astonishingly high. However, our assumption that women's presence as the result of equal descriptive representation would also lead to impacts on climate strategies was not verified. How can this be explained? Below we interpret the results using feminist perspectives while also suggesting future research paths.
The results of our analysis confirmed previous research on representation, that the connection between descriptive and substantive representation is far from straightforward. A critical mass does not necessarily lead to critical acts. This result can be argued to support liberal feminist views on equal representation. Equal representation is important for democratic reasons but as the results of this study show, there is no real difference between women and men.
Once obstacles to equal participation (like discriminatory legislation) are removed and women make decisions on climate issues, the policy content is expected to be the same regardless of sex, according to the liberal interpretation.
There was a profound silence regarding gender in all the documents. Yet, we have noted evidence of gender differences regarding both the material conditions relevant for climate policy and the attitudes on climate change issues. Applying standpoint theory from IR, the data suggested that women have different interests and bring different experiences and knowledge to climate policy. The literature on intersectionality urges us to probe this further. The material conditions related to climate issues vary greatly along the lines of class, age, place and ethnicity. Hence, female policy-makers in Scandinavia may be part of the high-consuming, high-GHG-emitting elite, with interest and views far from the working class, immigrants or elderly women. These policy-makers, albeit female, act according to their own interests and views which are very different from the interests and knowledge held by groups who consume and travel less, and who live, for example, in dense urban areas under poor conditions. To study this proposition, it would be relevant to follow up this study with a close investigation on how the Scandinavian political elites position themselves in relation to climate issues and to compare this with studies of the interests, experiences and views held by different groups and movements in society.
Finally, the results of the study can be explained using poststructural feminist IR theorizing. It proposes that masculine norms and power are so deeply institutionalized in climate-relevant institutions that policy-makers regardless of their sex, accept and adapt their views to the masculinized institutional environment wherein Scandinavian climate policies are made. The tendency in institutions dominated by masculinity is that gender is made invisible. The presence of women in policy-making even when in the majority does not have the expected impact, as institutional path dependency privileges masculinized ways of understanding climate issues, with the tendency to prioritize technical fixes rather than behavioral change. Accordingly, substantive representation can only be understood in relation to gendered institutional processes in the context of climate policy-making. For future research it would be relevant to investigate the dynamics of climate institutions in terms of embedded norms and practices that may demonstrate gendered path dependence in climate issues.
Finally, what we do note about the Scandinavian countries in a comparative perspective is that they are forerunners on climate issues, promoting far-reaching policies, and also being fairly successful in reducing GHG emission levels in line with stipulated goals. Noteworthy is that the dedication to the climate issue is shared among the population (Eurobarometer 2009). Considering this, we propose that masculine power may be deeply institutionalized in climate policy-making in Scandinavia but that this is a different gender climate regime than what we would find in other regions, following what welfare state theorists have argued about differences in gender regimes (e.g. Walby 2004). What we may be witnessing is the expression of a Scandinavian gender climate regime, the result of a long period of women's descriptive representation that has not led to explicit substantive representation but has had important indirect effects on masculinity and men. The Scandinavian gender climate regime may build on institutionalized gendered norms based on an alternative expression of masculinity, one that is more climate-sensitive. This should be further studied through comparative analysis that juxtaposes different types of gender regimes in relation to how they perform on climate objectives. We have suggested a number of research paths to further explore the link between descriptive and substantive representation in climate policy in Scandinavia, which has turned out to be far more multifaceted and complex than we anticipated.

Annica Kronsell Department of Political Science
Lund University, Box 52 S-221 00, Lund, Sweden Email: Annica.kronsell@svet.lu.se  (1), directly followed by Finland (2), Norway (3) and Sweden (4) and then Denmark in the seventh seat out of 135 countries measured (Hausmann, Tyson, and Zahidi 2012). 4 There is evidence of high climate awareness and willingness to take action also among the population in Denmark and Sweden. In the Eurobarometer (2009) these countries score high with Sweden at the very top. 5 The state units analyzed are: in Norway: the Environmental Ministry; the Climate and Pollution Agency; Transnova -the Transport Agency, and Enova -the Energy Agency; in Sweden: the Environmental Ministry; the Environmental Protection Agency and the Transport and Energy Agencies; in Denmark: the Ministry

Acknowledgments
Annica Kronsell acknowledges the research support of Lund University Excellence Center for the integration of the social and natural dimensions of sustainability (LUCID).

Notes on contributors
Gunnhildur Lily Magnusdottir is Assistant Professor of Political Science at the Department of Global Political Studies at Malmö University. She has been teaching extensively in the fields of international relations, European studies, environmental politics, and public administration. Her current research is on gender perspectives of climate politics in the Scandinavian and European context. Other research areas include small states in the EU, EU negotiations and environmental policy-making, and the Nordic EU member states' power resources within the EU.