Sustainable Welfare in Swedish Cities: Challenges of Eco-Social Integration in Urban Sustainability Governance

: In this paper, we study the integration of ecological sustainability and social welfare concerns in cities. E ﬀ orts to handle ecological challenges risk having negative impacts on equality and social welfare. While current levels of consumption and material welfare are unsustainable, there is a need for more sustainable approaches to welfare and wellbeing. Still, ecological and social concerns in urban governance are treated as separated topics. Based on text analysis of policy documents and qualitative interviews, we study how ecological and social welfare concerns are being addressed and integrated into urban planning in three Swedish cities (Stockholm, Göteborg, Malmö). Theoretically, the paper draws on conceptualizations of sustainable welfare, social and ecological sustainability, and policy integration. We ﬁnd ecological and social welfare concerns being acknowledged as interconnected and we see signs of an emerging sustainable welfare agenda in the cities, e.g., around Agenda 2030. However, in practice, eco-social policy integration is only established to a limited degree, for instance in neighborhood development, transport planning, and green city planning. Issues of ecological justice and equity and the relationship between socioeconomic factors and consumption-related environmental impacts are hardly addressed. Thus, much remains to be done for eco-social policy integration to materialize at the urban


Introduction
Social inclusion and ecological sustainability are recognized as two main challenges for today´s societies. Cities are key sites at which such social, economic, and ecological challenges are handled. Limited scholarly attention has, however, been paid to how cities cope with conflicts and synergies between policies directed towards ecologically sustainable and socially inclusive societies in practice.
Decentralization of social policies and urbanization of social problems are general tendencies in Sweden as well as in a European and international context. National governments have provided local governments with greater responsibilities to shoulder the social risks that local populations are facing and to develop suitable policy solutions applicable to particular social problems [1]. Such downscaling is driven by demands for more flexibility because local governments are assumed to be in possession of information and knowledge on local conditions. However, the decentralization of social policies is also a response to economic constraints and austerity measures, as national governments seek to mobilize resources to meet the needs of various groups. In addition, issues such as unemployment, poverty, and social exclusion are increasingly seen as localized social problems, somewhat beyond the reach and responsibilities of national governments [2].
Cities also face a series of ecological challenges due to high consumption of energy and natural resources, use of water and land as well as problems of climate change, air pollution and congestion (see e.g., [3]), while rapid urbanization creates pressure to use resources in smarter and more efficient ways. Investments in low-carbon energy, waste management and recycling, greener modes of transportation, and eco-designed housing are attractive solutions and geared towards reconciling ecological concerns with economic ones. Urban sustainability has to a large extent sought smart technologies to urban problems and been governed through urban experimentation, e.g., 'urban living labs', 'strategic urbanism', and 'smart cities' [4,5]. Though ambitions and efforts to reconcile ecological and economic objectives are identifiable, urban planning strategies are often hampered by the neglect of social concerns and over-reliance on technological solutions. Missing in these debates is how sustainability governance relates to and may cause social segregation and/or aggravate social inequalities.
In the emerging research field of 'sustainable welfare' [6], scholars have started to explore how global and intergenerational ecological concerns could be accounted for in the conceptualization of welfare and how policies towards the establishment of corresponding welfare systems-'eco-social policies' [7]-might look like. Given that a growing share of the world´s population lives in cities and that urban populations use an increasing share of finite resources, it is urgent to take the debate on sustainable welfare to the urban level.
In this paper, we study the integration of ecological sustainability and social welfare in cities-what we term eco-social integration-with an interest in exploring whether we can see traces of a sustainable welfare agenda emerging. Our aim is to analyze to what extent policy integration between environmental policy and social welfare in urban governance is occurring in the three metropolitan cities of Sweden: Stockholm, Göteborg, and Malmö. We investigate what shape eco-social integration takes and where it takes place, as well as the challenges of integrating ecological and social welfare concerns at the urban level. This is done through a comparative case study of the three cities based on interviews with civil servants in the environmental, social, and planning departments along with text analysis of central planning strategies in the cities.
Sweden is used as a case study because it is a typical example of an advanced welfare state while at the same time being a progressive case of contemporary environmental and climate governance. As a progressive environmental state, Sweden has successfully addressed environmental management and sustainability objectives [8,9], including ambitious climate policy targets [10]. Sweden has a comparably low-carbon intensity and has cut national carbon emissions by over 25% since 1990 [11]. In Sweden, cities and municipalities have rather far-reaching autonomy and self-governance at the local level and have significant responsibility and authority regarding social welfare, environmental regulation, and urban planning. Thus, this case can shed light on the potentials and problems of eco-social integration in urban planning and policies.
The outline of the paper is as follows. First, we introduce previous research on sustainable welfare and eco-social integration and relate this to urban governance and planning. We also discuss the literature on environmental policy integration which helps us understand the challenges of integrating ecological concerns in other policy areas. Second, we present the method and material of the study. Third, we present a comparative analysis of the three case cities. Finally, we conclude with a discussion about the potential for eco-social policy integration at the urban level as well as the challenges associated with reconciling ecological and social concerns.

The Links Between Ecological and Social Dimensions of Sustainability
Since the introduction of the concept of sustainable development, environmental, social, and economic dimensions of sustainability have been present. The social dimension is clear in the well-known definition of the Brundtland Report from 1987 by its emphasis on human needs: "Sustainable development is a development that meets the needs of the present without compromising the ability of future generations to meet their own needs" [12]. The principal goal is to fulfill the needs of all humans living today, implying that needs have priority over wants, but within the planet's carrying capacity (ecological limits) so that the needs of future generations are not jeopardized. Still, in its practical application, there has been a one-sided focus on ecological aspects of sustainability often in connection to economic activities. While this has generated a debate on whether economic development and environmental protection are in conflict or can be aligned, most works on social sustainability have studied social aspects in solitude and seldom considered the links to ecological aspects [13].
Increasingly, scholars have argued for paying closer attention to the links between ecological and social sustainability and analyzing these challenges in an integrated manner. Three important links should be highlighted. First, negative environmental impacts often affect poor and marginalized groups and individuals harder, something extensively studied in the literature on environmental justice. This is true both for direct impacts of, for instance, infrastructure, energy facilities, waste plants and industries, and indirect effects related to global environmental pressures such as climate change, desertification, and deforestation [14]. The reasons are complex but include that poor people tend to live in more remote and marginalized areas and that their political voice is weaker. Second, poor households and individuals face the risk of being relatively more affected by policies and measures that aim to reduce environmental harm [15,16]. One example is carbon taxation that makes fuel and heating costs more expensive, while poorer households spend a larger proportion of their income on these goods. Thus, environmental policies need to be carefully designed so that they do not exacerbate existing inequalities or do not even contribute to reducing them. For instance, carbon pricing policies could be introduced in policy packages that include measures for redistributing tax revenues in a revenue-neutral or progressive manner. Alternatively, eco-taxation could be balanced with social welfare investments funded by the revenues collected. Third, rich countries or richer households and individuals within countries contribute a larger share to environmental problems such as climate change and local air pollution since their lifestyles are more material, energy and travel intensive [16]. In relation to climate change, Gough [7] refers to a triple injustice where poor households and individuals are impacted harder both by the effects of climate change and the measures addressed to combat climate change, while they have contributed comparatively very little to the problem itself.
In order to comprehensively study different connections between ecological and social sustainability, some scholars have started to use the concept of sustainable welfare. This concept puts emphasis on overarching questions about how to provide basic social welfare for all people on Earth while respecting ecological limits and retaining ecological integrity in an intergenerational and global perspective [6,7,17]. This also reflects an ambition to join two strands of research that have hitherto been quite separated: environmental studies and social welfare research [6,18].

Eco-Social Integration
That the objectives of ecological and social sustainability are connected at the urban level and need to be approached in an integrated manner is also attracting increased attention. Relating to urban planning, Campbell [19,20] introduced the concept of "the planner's triangle" representing three fundamental priorities of planning (environmental protection, economic development, and social justice), which correspond to the three pillars of sustainability. The relationship between these priorities creates conflicts in property, resource use, and developmental issues. While the social welfare state and environmental regulation are institutional responses to the first two of these conflicts, Campbell argues that the development conflict between environmental protection and social justice is both the least understood and the most important to handle in the long run. This challenge boils down to questions about environmental justice: "How do we simultaneously protect the natural environment and reduce poverty and human injustice?" [20] (p. 392). Drilling [18] identifies three major steps in the planning process where environmental and social concerns are actualized and can come in conflict or reinforce each other: the decision where to locate new urban developments; the urban design competition that defines the concrete goals of the project; and the legally binding development schemes that guide implementation of the projects. He argues that social sustainability in urban planning needs to include attention to mixed social structure, mixed-use, accessibility for different groups, short routes for daily life, and broad public participation.
Murphy [21] has contributed with an important framework to analyze the links between ecological and social sustainability. He addresses four dimensions by which environmental policies can be analyzed in terms of their social content. These are equity, awareness of sustainability, participation and social cohesion. Equity relates to issues such as the export of pollution, how vulnerable groups are affected by environmental impacts and measures, and the environmental effects of social welfare provision and consumption. The last dimension concerns how to promote social cohesion and environmental objectives simultaneously, e.g., in infrastructure planning or through promoting social activities aimed at environmental goals. Although Murphy discusses these links in general terms it is easy to apply his framework to the urban level where many of the policy efforts he refers to are addressed and implemented.
Gough [7] suggests the introduction of eco-social policies in order to further both ecological and social goals simultaneously. Gough´s interest lies in ways to advance a low-carbon transition of society while maintaining goals of social justice and equality. He argues this to necessitate a rethinking of the concept of wellbeing towards less material consumption, resource use, and throughput. Gough proposes three forms of eco-social policies that target different aspects of the relationship between environment and welfare. First, eco-social policies can ameliorate the negative social effects of current environmental policies and eco-efficiency measures, while designing social welfare provision with low environmental impacts. One example is house retrofitting schemes with both social and environmental components. Second, eco-social policies can be designed to reduce emissions from consumption and redirect consumption in a less carbon-intensive direction. Examples here could include taxing of high-carbon luxury consumption and regulation of advertising, especially to children. Third, and more radically, eco-social policies can be applied to reduce consumption levels and question the strong focus on economic growth in our societies. Examples of such policies are reductions in working hours and the development of local economies. Gough discusses eco-social policies in general terms mainly with a view at the national level. However, it is possible to think of several policy areas at the urban level where eco-social policies are relevant, and there is a need to further explore this concept for the urban level both theoretically and in empirical terms. Other authors have discussed the links between ecological and social sustainability in areas such as urban transportation [22,23], urban greening [24,25], and energy retrofitting in housing [26,27].

The Challenges of Policy Integration
A key issue in developing more sustainable welfare arrangements is related to the potential and problems associated with integrating ecological and social concerns in urban policy and governance. In general, the integration of ecological, social and economic dimensions of societal development has been a pronounced ambition in sustainability development discourse and sustainability governance ever since the Brundtland Report [12]. Various kinds of efforts have been addressed to foster policy integration and cross-sectoral collaboration at both the national and urban level. Currently, this is perhaps most clearly demonstrated by Agenda 2030 and the implementation of the Sustainable Development Goals (SDGs) as well as the New Urban Agenda, two processes related to each other and coinciding in urban practices of governance and planning.
This ambition is also reflected in sustainability governance, for instance in Sweden, the context of our empirical analysis where commitments to policy integration have been expressed in national policy strategies on the implementation of sustainable development since the late 1980s (see e.g., [8,9]). At the national level, public agencies were given sector responsibility for ecological sustainability concerns in 'their' policy sector, an ambition later operationalized in the environmental objectives reform [28,29] and nowadays also in the Swedish Agenda 2030 strategy [30]. At the urban level, local sustainability strategies have been informed by similar ambitions to promote cross-sectoral cooperation, stakeholder participation and the inclusion of various societal groups and interests to a varying extent. In particular policy areas, policy integration has even been codified, not the least in urban planning where the Planning and Building Code stipulates the integration and balancing of various interests as a legal requirement.
Sustainability governance always contains some form of a balancing act between ecological, economic and social considerations and, thus, in the words of Underdal [31], results in some kind of 'integrated policy'. Building further on Underdal's work, the literature on environmental policy integration (EPI) has conceptualized policy integration at various levels and stages of policy development. It is well-established to differentiate between horizontal and vertical policy integration [32], as policy integration cuts across both policy sectors (horizontal) and administrative levels (vertical). A prominent conceptualization is found in Lafferty and Hovden [32], who define EPI as "the incorporation of environmental objectives into all stages of policymaking in non-environmental policy sectors, with a specific recognition of this goal as a guiding principle for the planning and execution of policy; . . . and a commitment to minimize contradictions between environmental and sectoral policies by giving principled priority to the former over the latter" [32] (p. 9). This clearly represents a normative approach to policy integration, while other approaches emphasize policy integration as an issue about coordination (policy coherence) and harmonization (balancing) across policy areas [33]. More analytical accounts are reflected in, for instance, Lenschow [34], Nilsson and Eckerberg [29] and Jordan and Lenschow [35]. In parallel to Lafferty and Hovden [32], Nilsson and Persson [36] and Nilsson and Eckerberg [29] developed a framework for analyzing EPI in practice at conceptual and operational levels. While finding strong support for EPI at the conceptual level, they identified few examples of integration in operational practices. Jordan and Lenschow [37] emphasize how the political commitment to policy integration has been widespread and led to new policy approaches, while the practical application is lagging far behind and that evidence on policy outcomes is sparse.
So, while policy integration might be a well-established ambition in policy rhetoric, in practice it is a challenging and notoriously difficult task to accomplish. As public policy at the urban level struggles with similar challenges as at other levels of jurisdiction, urban governance suffers from implementation deficits due to a lack of resources, gaps in competencies, institutional inertia and turf wars in public administration. This is related to the organization of public administration, characterized by administrative separation, fragmented sectorization, and high degrees of specialization. For instance, van Stigt [38] discusses five factors causing barriers to policy integration in urban governance and planning. First of all, the multidimensional nature of the urban sustainability concept leads to 'qualitative multiplicity' in the assessment of environmental quality among actors with varying interests. Second, the bounded rationality of decision-making makes policy integration subject to processes of reframing and linking policy issues to each other. Third, the use of expert knowledge is limited and sometimes even ad hoc. The disciplines and professions in different policy sectors add to this complexity and cause separation and fragmentation. Fourth, while local authorities have acquired more responsibilities, they still have limited maneuvres to act and influence the behavior of local actors and structures under which they operate. Fifth, the devolution of authority is often inadequate and blurred across levels of jurisdiction. Thus, while it is hard to imagine any normatively justifiable rationale for giving certain objectives "principled priority" over others, as suggested by Lafferty and Hovden [32], the integration of ecological, social and developmental concerns will remain a politically charged issue subject to disputes and debate, political compromises and balancing acts. Thus, in practice, efforts addressed to foster policy integration are expected to be laden by conflicting values and interests that have to be balanced somehow in the practical arrangements of urban governance and planning.
Our study supplements the literature by investigating how ecological and social welfare concerns are viewed as interrelated to each other, how cities are addressing efforts to handle them in an integrated fashion (or not), by which measures and across which policy areas. In our empirical analysis, we investigate the ways in which ecological and social welfare concerns are addressed and the extent to which they are integrated into practical arrangements of urban policy and planning in three Swedish cities. For this purpose, we pose three questions to our empirical material: • Where does eco-social integration take place? • What are the characteristics of eco-social integration? • What are the challenges of eco-social integration?

Method and Material
The article is based on a qualitative comparative analysis of case studies in the three metropolitan cities in Sweden: Stockholm, Göteborg, and Malmö. Stockholm is the capital city and the main politico-administrative and financial center of Sweden, while Göteborg is the second most populated city located at the west coast with the largest harbor, and Malmö is the third-largest city located in the southwest, which over the latest two decades transformed itself from a manufacturing city into a cosmopolitan and multi-cultural urban center. We have used two sets of material for the study which have been analyzed separately in order to increase triangulation and the validity of the results.
First, we made a text analysis of key policy documents in the three cities relating to environmental policy, social issues, and urban planning. A thorough search in the three cities resulted in a large list of policy documents (more than 30 for each city). We decided to limit the scope of the analysis by focusing on those documents deemed to be of more strategic relevance, which resulted in 6-7 documents for each city. The documents chosen are, for instance, environmental strategies, climate strategies, sustainability reports, social welfare reports, and the cities' comprehensive master plans (for a complete list of the strategies analyzed, see the reference list). The main focus of the analysis of the policy documents was on integration between ecological and social concerns, both in terms of policies and measures described and the more general reasoning and rhetoric in the documents. Each document was closely read and coded according to connections between the two policy areas. For each city, a synthesizing summary was produced describing the general results.
Secondly, we conducted semi-structured qualitative interviews with civil servants in the three cities. In each of the cities, we interviewed one strategically placed civil servant in the environmental office, the social office, and the planning office, respectively, in order to provide a general picture of how eco-social integration was handled in different parts of the administration. Malmö had, at the time of the case study, a centrally placed sustainability office and one person working there was also interviewed. This resulted in 10 interviews in the three cities, each interview lasting approximately 60 min (see reference list). The same interview guide was used for all interviews and the questions focused on issues such as the perceived connection between environmental and social issues (synergy or conflict), the extent to which the areas are integrated in city planning, where integration occurs and where it is lacking, the level and forms of cooperation between environmental and social offices, the awareness of the concept of triple injustice, and the extent to which a social analysis is made when environmental policies are designed and applied. All interviews were recorded, transcribed and coded based on the different themes in the interview guide. The interviews for each city were analyzed together followed by a comparison between the cities. The combined analysis of strategic documents and interviews with key civil servants has provided a more comprehensive picture of the situation in the three cities.

Where Does Eco-Social Integration Take Place?
It is clear from our material that ecological and social welfare concerns are seen as two of the major challenges facing the metropolitan cities of Sweden. Ecological sustainability is addressed as a central concern in all cities. In particular, climate change is high on the agenda and it is discussed both how climate mitigation can be achieved and how the cities need to become more robust and resilient to the impacts of climate change. On the social side, the main issues discussed are social segregation, inequality, and differences in public health. The interviews indicate that the cities have been actively promoting climate and environmental policies for some time and have profiled themselves in this area, particularly in the case of Malmö but also in the other two cities. While welfare services, including schools, child care, elderly care, and social service, have for long been a core function for city administrations, social issues have not been acknowledged as a strategic policy issue at the city level until recently [1]. In all three cities, there has been a renewed focus on problems of social segregation and inequalities in public health, with special commission inquiries focusing on this aspect (2012 in Malmö, 2014in Göteborg, 2015. At the time of the interviews, a quite uniform picture emerges about a strong focus on both environmental and social concerns in the three cities. In the material, we also observe a growing awareness about how these two challenges are interlinked and needed to be handled in an integrated manner. Still, our analysis of the three cities clearly shows a lack of integration between ecological and social aspects of sustainability in the main policy strategies and in day-to-day planning. This is most evident in the policy documents that are related to either the environment (environmental programmes, climate strategies) or social welfare (social sustainability reports). The environmental programmes and the climate strategies in the three cities give little attention to social issues, if any, either in terms of social effects of environmental measures or differences in ecological footprints between various social groups. In the introduction to these documents, the connection is mentioned in general terms, which represents indications of a general ambition to integrate ecological, social and developmental concerns in urban governance. However, when specific measures are discussed there are few explicit connections made, for example in Malmö regarding noise, air pollution, health effects from bicycling, and climate adaptation for vulnerable groups; in Stockholm regarding energy efficiency and the risk of increased rents; and in Göteborg in relation to green spaces and food production. One interpretation of the lack of attention to social issues is that the environmental programmes explicitly focus on the environmental impacts while leaving social issues to other arenas. This is confirmed by the interviews. Environmental officers stated a wish to protect their core area of expertise and were reluctant to take into account too many other aspects for fear of losing sight of the environmental objectives. As expressed by one environmental officer, "after all, it is our mission to represent the ecological dimension" (Interview 5).
The social sustainability reports in the three cities focus exclusively on social issues and there are very few connections to environmental aspects of social justice or inequality. In Malmö, the social sustainability report from 2013 had two main recommendations, to establish a social investment policy to reduce differences in health and living conditions and to increase democratic governance and participation at the local level. In Göteborg and Stockholm, the social sustainability reports have a similar focus on health and living conditions and do not connect this to environmental issues.
In 2015, the United Nations adopted the Agenda 2030 and the 17 Sustainable Development Goals (SDGs), and since 2017, they are incorporated into Swedish national policy. In all three cities, local Agenda 2030 work has been initiated and the SDGs are used as tools for addressing ecological, economic and social dimensions of sustainability in an integrated and cross-sectional manner. Malmö has come farthest in this process and in 2017 formed a Sustainability Office within the city council office responsible for coordinating the Agenda 2030 strategy and for facilitating processes of integration, but that organization was recently restructured after the latest local elections in 2018. In Stockholm in 2019, the city council board formed a new Agenda 2030 council with representatives from political parties, local stakeholders, and experts. In Göteborg, a process of establishing a similar Agenda 2030 body in the local city administration is ongoing. The local implementation of Agenda 2030 can prove an important tool to integrate different aspects of sustainable development, but it is yet too early to assess what the effects will be in practice.
When it comes to urban planning, the comprehensive master plans represent more deliberate attempts to integrate social and environmental issues. This is not surprising since the function of the comprehensive plan is to outline the developmental plans for future land use in the city taking into account all relevant aspects. Environmental and social challenges are discussed in conjunction with each other and there is an awareness that they are interrelated, as seen in the following quotes: Göteborg shall develop into a viable and long-term sustainable city with a balance between social, economic and ecological factors. A holistic view is essential for the possibility of a sustainable development in the Göteborg region. There are given connections between human welfare, economic growth and environmental sustainability (Göteborg 2009b, p

. 48). [own translation]
Measures for ecological sustainability can work as a motor for social sustainability, while these in turn in some cases can be a precondition for ecological sustainability (Malmö 2014b, p. 15).

[own translation]
In all three comprehensive plans, both environmental and social issues are prevalent when strategies and specific measures are discussed. Environmental goals relate to, for instance, increased use of renewable energies, shifts towards more sustainable modes of transport, preservation of green areas and urban ecosystems, and improved air quality. Social goals concern, for instance, planning for less segregation, increased house building, safety, and liveability in the built environment. In most cases, however, environmental and social aspects are discussed separately even though apparent connections exist, e.g., regarding a sustainable and just transport system, functional mix in areas and mixes in types of dwellings, equal access to green areas, and social impacts of energy efficiency measures in rental apartments. These types of connections are sometimes mentioned in the comprehensive plans but are not detailed to any great extent, and solutions and measures are not addressed as a package or in an integrated fashion.
The interviews confirm the perspective of urban planning as an arena with the potential for integrating ecological, social, and developmental concerns. The planning offices seem to constitute a hub where both environmental and social issues are raised and coincide. A social officer in Malmö stated that they do not have much direct cooperation with the environmental office. Rather they get in contact with them through their involvement in the planning process, which is coordinated by the planning office (Interview 2). Environmental officers in Malmö and Göteborg (Interviews 1, 5) stress the specific role of the planning office to collect viewpoints from different departments and emphasize the importance of them providing ecological perspectives in the remittal processes, while other departments reflect other aspects and dimensions.
While the general picture in the three cities is a lack of integration between ecological and social aspects of sustainability, we find examples of quite developed integration in specific projects and policy areas. This confirms the impression that eco-social integration takes place either in an ad hoc fashion or is experimented with in a project-based manner. When respondents were asked to reflect on where integration occurs, a recurring answer was urban development projects either in socially problematic areas or in the development of new neighborhoods. In general, respondents were of the opinion that integration is much more feasible in specific projects since there are often additional resources devoted to cooperation and since participants get time to work focused on one project and have the opportunity to get to know each other and build trust among participants. In all three cities, there are examples of urban regeneration projects in suburbs that are characterized by social problems such as poverty, segregation, unemployment, and criminality. These projects usually have a strong social focus, and in some cases, an important feature has been the measures that seek to combine both social and environmental goals, e.g., in Järva (Stockholm), Lindängen (Malmö), and Backa (Göteborg). Examples of eco-social integration in these areas are energy efficiency measures with a social profile, participatory renewable energy projects, improved bicycle infrastructure, bicycle courses for women from various national backgrounds, urban gardening, and improved access to green spaces and parks.
In our analysis, we also find examples of specific policy areas where the integration between social and environmental aspects has come a long way. In Malmö, the 2016 Transport and Mobility plan has an equal focus on a transport system with lower environmental impact and that contributes to a more inclusive and less segregated city (Malmö 2016b). The second chapter, titled "Människan" (the Human Dimension), focuses on equality, gender aspects, traffic safety and involvement in planning. Special attention is paid to how to reduce segregation by linking different parts of the city in the planning of the transport system. The third chapter is about how to develop more environmentally sustainable transport patterns with a focus on walking, cycling, and public transport. In the fourth chapter, titled "Staden" (the City), the two dimensions are merged in terms of, for example, ideas on how to design city streets in order to both provide a good living environment for the inhabitants and promote more sustainable modes of transport. The links between ecological and social concerns in the transport sector have been explored in previous research. For instance, Gössling [23] shows that there are often considerable injustices in urban transport systems in terms of, for example, exposure to noise, air pollution and accidents, and he argues for a transformation to more sustainable transport systems to be motivated also from a justice perspective. Martens [39], however, warns that sustainable transport planning risks perpetuating existing inequalities since much policy attention will be devoted to persuading car drivers, who are more often higher-income earners, to shift to public transport and cycling, while neglecting the transport needs of poorer communities.
In Göteborg, the 2014 Green strategy contains both social and ecological goals that are given equal standing (Göteborg 2014c). The social goals include equality in access to green areas and parks, green areas as a way to connect different parts of the city, and positive health impacts from recreation. Although the green strategy is mainly geared towards ecological objectives, social aspects are addressed throughout the document. Previous research has shown negative equity impacts of greening projects in cities if social aspects are not considered since they tend to disproportionately benefit affluent communities [25] and cause distributional injustices [40].
The plans adopted in these two cases are fairly recent and it is still too early to determine what effect they will have on actual planning practices. However, they show how a more far-reaching integration between ecological and social aspects is possible and is starting to take place in city planning.

What Are the Characteristics of Eco-Social Integration?
In the analysis of the characteristics of eco-social integration, we focus on whether the relationship between ecological and social concerns is seen as synergistic or in conflict, and which types of social and environmental issues are considered.
Our respondents mainly conceptualize the connection between ecological and social concerns as a synergetic relationship, i.e., that environmental measures can reinforce and improve social conditions, and this is also confirmed in the policy documents. Examples include positive impacts of public transport and bicycling on both health and decreased climate emissions; building regulations and inspection requirements promoting both energy-efficient housing and good living conditions; and urban green areas that are positive for both urban biodiversity and improved health effects. The respondents also state that it is important to strive for synergy and win-win solutions in urban planning. At the same time, several of the respondents acknowledge that there can be conflicts between environmental and social objectives and view these as important to take into consideration. Examples of potential conflicts are gentrification effects of energy efficiency improvements (due to higher rents); environmental standards for kindergartens vs the need for new kindergartens; and the construction of new housing blocks vs the will to preserve green areas. Many respondents saw economic restrictions as a key challenge and stated that some of the conflicts between social and environmental priorities could be solved with more economic resources.
In general, the types of social effects that are addressed in relation to environmental aspects are very much linked to health issues, e.g, from transport habits and access to green areas and improved housing conditions. However, we did not find much consideration of distributive and equity effects of local environmental impacts or of urban environmental policies and measures, neither in the policy documents nor in the interviews. In all three cities, there are tools to make environmental and social impact assessments as a part of the physical planning process, for instance, when new area plans are developed. The environmental impact assessment is mandated by legislation, while the social impact assessment is a municipal initiative. However, the connection between environmental aspects and social aspects is not considered and there is no systematic analysis of the possible social impact of environmental measures, e.g., regarding negative distributional effects or which groups are targeted or affected. Notably, the concepts of double or triple injustice are not in use in relation to climate change or other environmental effects, and it was mainly social policy officers who addressed a more reflective view on the social issues related to environmental management in the cities (Interviews 2, 6,9). They were clearly aware of potentially negative effects of environmental measures and were also critical that measures for more sustainable practices are most often directed to wealthier neighborhoods. One example that was mentioned is that municipal rental of electric bikes has been offered in the city center but not in the poorer suburbs where the need is greater (Interview 9). This shows that social workers can have an important role in environmental work at the urban level, something that has been highlighted by others [41].
The literature on sustainable welfare and eco-social integration has mainly emphasized more general and global environmental issues, most notably climate change, in discussing the links between environment and social welfare [6,7,17]. In our study, we, on the contrary, find that eco-social integration at the urban level is mainly concerned with localized environmental issues and their relations to social welfare and public health. It is striking that we find the most developed eco-social integration in urban regeneration projects where social problems are prevalent alongside local environmental problems. Examples of localized environmental issues are how to take care of rainwater, access to green space, improvement of local living environments, and experiments with urban gardening. Measures relating to these issues often have a strong social element and social objectives are the driving component while positive environmental effects are more of an add-on, side-effect. To be sure, there are other examples of eco-social integration where climate measures are also included relating to, for example, transport and changes in travel habits and energy efficiency measures in housing, but these are not the most common. Many climate-related measures tend to be driven mainly from the environmental or technical offices where the connection to social aspects is often quite loose. Thus, when it comes to, for example, renewable energy, energy efficiency, transport, and waste and electricity use in households, there is not much focus on social aspects.
One area with particularly little integration relates to sustainable patterns of consumption and lifestyles, which is surprising considering its strong ties to both environmental and equity aspects. Research shows that high-income and middle-class households and neighborhoods tend to have much higher ecological footprints than poorer neighborhoods, for instance in relation to transportation [42,43]. At the same time, the interest for sustainable practices is often stronger among higher-income groups, and the infrastructure to facilitate more sustainable practices tends to be more developed in affluent communities e.g., regarding cycle infrastructure, recycling and ecologically responsible consumption. However, the current conceptions of welfare and wellbeing could be questioned as being linked to material consumption, and scholars are exploring new views on wellbeing that are less carbon and resource-intensive. For instance, Andersson et al. [44] examine the link between greenhouse gas emissions and subjective well-being and show that practices that individuals view as providing meaning and happiness are seldom carbon-intensive, arguing that it is possible to live a good life with low carbon footprints. Gough [7] argues from a justice perspective to distinguish between necessary emissions (related to needs) and luxury emissions (related to wants) and to give priority for the former over the latter.
Reducing ecological impacts from consumption is something that all three cities work with. All cities have initiated specific projects, campaigns, and guidelines that target consumption-related emissions and aim to influence citizens to change habits and make choices to reduce their emissions in daily life regarding e.g., food consumption, travel habits, and domestic energy use. However, the focus of these activities is on changing to more climate-smart consumption and little attention is paid to the need for drastic reductions in consumption levels or changes in lifestyle. There is also limited attention to the differences in consumption patterns between income groups or geographical areas in the city. In Göteborg, the city has partly focused on differences in climate impact between different households and carried out a study in cooperation with researchers at the Chalmers University of Technology, which showed that higher-income households emit more carbon emission than lowerincome families due mainly to a higher degree of car and flight travel [45]. In Malmö, transport planners have shown differences in car use between both gender (men drive more than women) and income (people with high-income drive more than those with low-income), and that these differences also hold between geographical areas (Malmö 2014b). Such studies are, however, not part of the core strategic approach of the cities and have (as of yet) not spurred specific policy measures targeting socio-economic differences in ecological impacts between households. In the interviews, few respondents reflected on environmental effects related to consumption and lifestyle or the social dimensions of such issues. In Göteborg, the environmental officer showed awareness of differences in impacts across different groups (Interview 5), but otherwise, it was mainly the social policy officers who discussed this topic. For instance, in Malmö, the social officer brought up the issue that environmental choices, e.g., regarding cycling and recycling, is dependent of the local context in which you live (Interview 2), while considering it a municipal responsibility to give everybody the same opportunities to lead sustainable ways of life. He was also critical of national policies directed more towards the middle class and specifically mentioned the subsidies for electric bikes. The social officer in Stockholm reasoned in a similar way and argued that there is much more infrastructure for bicycles in the central parts of the city than in the poorer suburbs (Interview 9).

What Are the Challenges to Eco-Social Integration?
The general impression from the analysis, both of the policy documents and the interviews, is that policy integration in general and eco-social integration in particular, is being addressed in an ad hoc or project-based fashion. Two quotes from the interviews in Malmö illustrate this: There are pilot and forerunner projects where it is happening, but in relation to our gigantic production factory, it is really on the margin (Interview 3).

[own translation]
Yes, I would say that so far, we are better in projects than in the ordinary planning, because cooperation is a challenge, it takes more time to go outside yourself and manage cooperation (Interview 2).
[own translation] The challenges and barriers to policy integration discussed in the literature are confirmed by our study. All three cities struggle with administrative separation and fragmentation of policy measures. This could in part be explained by professional specialization, but is also related to organizational structures. For instance, a respondent in Göteborg reveals how the organization of the Social Equality strategy has been decentralized at a distance from the central city administration (Interview 6). In a similar vein, the social welfare policy in Stockholm is delegated to the city district administration, a lower organizational level than the city council office, which generates fragmentation and separation between policy officials within the city administration and creates hurdles for collaboration and policy integration across policy areas and departments. In Malmö, the Agenda 2030 work and responsibilities for implementing the SDGs have recently been decentralized with similar effects. In the interviews, some respondents emphasize how the city administration lacks comprehensive systems of governance and monitoring for counteracting such tendencies towards fragmentation (e.g., Interview 5).
A related barrier is the differences in professional language and culture between civil servants in the environmental and social administrations. In Malmö, the representative from the social office stated that they sometimes had difficulties in understanding the language and concepts used by the environmental administration making it difficult to give meaningful input, and gave a concrete example of the meaning of "green and blue areas" (Interview 2). In contrast, environmental officers revealed a wary attitude towards including too many other aspects in their assessments for fear of losing focus on environmental aspects. Additionally, in Göteborg and Stockholm, the representatives from the social offices were more of the opinion that integration was lacking.
Another key barrier to policy integration is the lack of resources. While most resources and competencies are bound up in the ordinary day-to-day operations of city administrations, new tasks (such as eco-social measures) are often organized in an ad hoc fashion in extraordinary units or projects alongside the standard operating procedures. This is one explanation as to why integration occurs mainly in projects where dedicated resources are allocated to foster cooperation and collaboration between different stakeholders and administrations, both in terms of time and money. In the ordinary planning and day-to-day work, it is much more difficult to allocate time to develop and nurture relations necessary for deeper integration (Interviews 3, 5, 8,9). On the other hand, the developments and learning taking place in urban experiments or dedicated projects seldom feedback into the ordinary planning, due to lack of procedures for systemic learning within city administrations and in urban governance.
The discussion above relates to the problems in the cities to find an appropriate arena for cooperation and integration across departments. As we have seen, the planning office has the potential to function as such an arena but they mainly deal with issues directly related to urban planning and the physical structure of the city. Additionally, such arenas could be established in specific projects facilitating integration. However, these projects are of a temporary character and mechanisms to translate such local instances of eco-social integration into more permanent cooperation and integration in the cities seem to be lacking.
The local implementation of Agenda 2030 can be seen as a way for the cities to address the integration of sustainability concerns such as expressed in the SDGs. Malmö has come farthest in this process, where, during the previous electoral period, the city council addressed a commitment to integrate the implementation of SDGs in all parts of the city administration [46]. A special Sustainability Office was formed at the central city council office responsible for coordinating the Agenda 2030 strategy and for facilitating processes of integration. Göteborg has taken steps to introduce a similar organization but it has not yet been done. Stockholm commissioned expert assessments during the previous electoral period, while the newly elected local government in 2019 introduced a local Agenda 2030 council with political representatives as well as non-governmental organizations and interest groups and a sustainability office in a similar way to Malmö. While the SDGs and the New Urban Agenda are ambitious in addressing both a socially progressive agenda and strategies for policy integration, implementing Agenda 2030 in an integrated fashion at the local level is challenging in several respects. Valencia et al. [46] show that successful integration requires among other things strong political support, a mandate to promote policy coherence, the inclusion of local administration as well as stakeholders, and sufficient resources. In our study, the challenges are witnessed, for instance, by the reorganization of the Agenda 2030 work in Malmö; first, the city council commissioned a centrally placed Sustainability Office to coordinate the implementation, however, after the latest local elections, this office was decommissioned while responsibilities for integrating the SDGs are delegated to all sector departments in the city administration. This shift represents two distinctly different approaches to policy integration, which has consequences for both the priority given to Agenda 2030 and the coordination of work with the SDGs within the city. This points to the difficulties in balancing between needs for policy coherence and coordination of such comprehensive strategies on the one hand, and for specialization and delegated responsibilities in implementing eco-social integration measures in practice on the other.
So, while the overarching ambition might turn out to be more or less well-integrated into policy rhetoric, there is an impending risk that its implementation gets fragmented and trapped by various interests in the sectorized city administration. In other words, what Nilsson and Eckerberg [29] conceptualized as the difference between policy integration at conceptual and operational levels seems to be relevant also for understanding the challenges of eco-social integration at the urban level.

Conclusions
Our study has clearly shown that both environmental and social issues are high on the political agendas in the three metropolitan cities of Sweden. Climate change and social segregation are particularly portrayed as two major challenges. Still, despite decades of work with local sustainable development, most recently through the Sustainable Development Goals and Agenda 2030, we find that environmental and social welfare concerns continue to be managed as two quite separate topics while apparent connections between them are not addressed in a systematic way. We do find that the need to integrate ecological and social dimensions of sustainability is acknowledged in both policy documents and in our interviews, and it is possible to speak about an emerging sustainable welfare agenda in terms of eco-social integration, at least in the official policy rhetoric.
In practice, however, the main impression from our analysis is that eco-social integration tends to take place either in an ad hoc fashion or is being experimented with in a project-based manner. More specific measures of eco-social integration do typically occur in delimited projects around e.g., neighborhood development and sometimes in specific policy areas, e.g., mobility planning in Malmö or green area planning in Göteborg, while more comprehensive strategies for integration are, as of yet, lacking. Rather, in relation to the "gigantic production factory", as one respondent named the day-to-day planning in the cities, eco-social integration is a marginalized issue and departmental separation is the norm. This indicates that the challenges of policy integration across established policy areas, professions, and organizational cultures are relevant also at the urban level. While ambitions to integrate ecological and social welfare concerns are emerging on a rhetorical and conceptual level in the three cities, the practical efforts of eco-social integration seem to be torn between the extraordinary and ordinary arrangements of urban planning and governance. On the one hand, we find examples of eco-social measures in extraordinary projects and practices that give room for additional resources, time to build new relations and forms of cooperation as well as providing leeway for the courage to experiment with and try out new practices. On the other hand, it is challenging to get such eco-social measures accepted in the ordinary operations of city administrations where eco-social integration, as of yet, mainly exists as an idea but not as an established practice.
In our analysis, we highlighted findings similar between the cities and we do argue that in general, the similarities are more significant than the differences. Still, there are variations worth mentioning. In general, eco-social integration seems more developed in Malmö than in the two other cities. Malmö launched a commission on social sustainability a few years before Stockholm and Göteborg and has also had a centrally placed office responsible for coordinating sustainable development and the Agenda 2030 work.
An important finding of our study is that the type of eco-social integration we observe in the three cities relates primarily to local environmental issues such as access to green areas, rainwater protection, urban gardening, and improved local living environments. This is in contrast to conceptual work on sustainable welfare, which emphasizes global issues such as climate change, its social implications, and eco-social policy integration as a way forward [6,7]. At the urban level, the most apparent connections between ecological and social concerns seem to be those that are tangible for local citizens and communities. This has implications for sustainable welfare research that needs to be more aware of how eco-social integration is constituted at the local level and how it can develop further. At the same time, our findings point to a gap in urban planning where the challenges related to climate change, considered of prime importance to all three cities, are handled mainly as an environmental problem while its social implications are to a large extent neglected. This is visible both in terms of who is given responsibility for the issue i.e., the environmental management office (partly also the technical office, the planning office, and the traffic office), and how it is framed i.e., mainly as a technical issue of reducing carbon emissions and climate impacts, while it is striking how representatives from the social work offices are hardly involved at all in local climate change strategies.
Likewise, issues of justice and equity are quite absent in urban sustainability planning. The concepts of double or triple injustice do not seem to be familiar and questions such as who is responsible for most emissions or who is most affected by environmental harm are not yet guiding principles for planning. The environmental side of city administrations has not yet addressed issues of justice while the social office representatives, who do raise such concerns, are seldom part of the process. We also find that eco-social integration is limited in relation to addressing the ecological impacts of consumption and lifestyles. To the extent that the three cities address such issues, it is mostly in the form of advice and information to households on how to lower their individual impacts, but there is not much attention directed towards variations in ecological footprints depending on socio-economic differences. This points to a critical area of improvement for more sustainable welfare arrangements, especially as sustainable consumption holds a huge potential for leading more sustainable and equitable lives at the urban level.