Sustainability Transitions at the Frontline. Lock-in and Potential for Change in the Local Planning Arena

: This paper explores challenges and possibilities for integrating goals of long-term sustainable development into urban planning practice, with a speciﬁc focus on local institutional conditions for sustainability transitions. The analysis is based on a qualitative single case study of a large urban development process: the development of a new city district in Hyllie in the city of Malmö, Sweden. Hyllie was branded as a ﬂagship project for sustainable urban development, with particularly high ambitions on climate neutrality and sustainable energy consumption. Several innovative elements were initiated in the development process, for instance the “climate contract” between the municipality and large energy companies. In the paper, this climate contract is discussed as an initiative with a promising potential for sustainability transitions. In practice, however, the outcome of the development in Hyllie in terms of sustainable development is ambiguous, since the district is also framed around luxury shopping, entertainment, and an ambition to attract visitors from a long distance. The Hyllie development illustrates pre-requisites for work on sustainable development in a decentralized and market-oriented planning context. Theoretically, the analysis is inspired by the multi-level perspective (MLP) and institutional theory. The results illustrate how the development process was shaped by a complex interplay between actors with differing agendas and targets at different stages in the process. These results are applied in a general discussion of challenges and possibilities for urban planning to contribute substantially to a transition to long-term sustainable development. Overall, the analysis demonstrates the importance of considering speciﬁc local institutional conditions in strategic work for long-term sustainability.


Introduction
Current research on sustainability transitions (ST) is being driven by the increasing urgency of tackling challenges relating to climate change, biodiversity loss, and resource depletion, where neither traditional top-down planning nor incrementalism has proven capable of enabling the structural change necessary to pull back from unsustainable paths [1]. Research on ST aims to explore and provide input to substantial, radical change of society and social relations so that long-term sustainability targets can be reached.
The ST literature provides a rich body of knowledge regarding sustainability transitions in a range of specific sectors like energy, transport, and food production. In this paper, we examine conditions to work for sustainability transitions in an urban planning context. Urban planning is widely understood as an important arena for realizing goals of long-term sustainability [2][3][4][5][6]. However, it has not yet developed into a major theme in the ST research field, although the number of studies on urban

•
The framing of the development project and the way in which sustainability dimensions were integrated in the plans for the area • The interplay between specific actors involved in the planning and development process • The creation of arenas and collaborations with the potential to challenge the status quo and bring new perspectives and priorities to the process.
The study is theoretically based on ST literature, with particular focus on the multi-level perspective (MLP) [15], which provides a comprehensive framework for analyzing this complex local planning case. The MLP framework is complemented here by the theory of institutional capacity [16], which provides a basis for exploring more specific dimensions of the local planning process, particularly knowledge resources, relational resources, and mobilization capacity.
The paper is organized as follows: Section 2 presents the materials and methods, while Section 3 describes the theoretical framework. Section 4 provides a detailed background to the empirical case study and a qualitative analysis of the case. Section 5 presents a concluding discussion.

Materials and Methods
The study is an exploration of a large urban development project in the city of Malmö, the third largest city in Sweden and located in the very south, just beside Öresund and the bridge to Denmark. Malmö is an interesting case to explore since, over the past decade, its city council has placed much emphasis on sustainable development. Today, Malmö is viewed as one of the most ambitious cities in Sweden in this respect [17]. Malmö can thus be expected to provide a 'favorable' context for sustainable urban development.
The development project studied was the ongoing construction of a new city district in Hyllie, south-west Malmö. The development includes housing, transport infrastructure, a new sports and entertainment arena, and a shopping mall. The district has been promoted as Malmö's new flagship of sustainable development based on goals of climate neutrality [18,19]. However, Hyllie has simultaneously been framed as a mobility hub, shopping center, and entertainment area, which has caused tensions. This illustrates goal conflicts which often occur, to a greater or a lesser degree, in urban planning in general. Hyllie can be viewed as a "critical case" [20], suitable for exploration and for gaining a better understanding of some of the specific conflicts, complexities, and challenges that influence transition processes, depending on the local context. It should be noted that since the development process is still not completed, the present analysis could not reach any clear conclusions on whether Hyllie is a 'successful' case from a sustainability transitions perspective. Instead, the main point of the analysis was to explore the challenges and complexities involved in trying to enact sustainability transitions in the urban planning arena. The analysis sought to identify specific institutional dimensions that affect integration of sustainability targets in local planning practice. While it built on one specific case, it sought to produce a few more general reflections on the conditions for municipalities to work for sustainability transitions in decentralized and market-oriented institutional contexts.
The analysis was based on qualitative interviews together with written documents from the policy and planning process. A total of 10 qualitative interviews were conducted during March 2014-October 2014 (for an overview, please see Table 1). We started by interviewing three key city officials in the planning and development process: one from the City of Malmö's Real Estate department; one from the Planning department; and one from the Environment department. These interviews provided much information about the case and the interviewees also recommended other individuals for us to interview to get more details. In total, we interviewed five officials from the City of Malmö, one private entrepreneur and one representative from the municipal housing association. In addition, three short, specific interviews were conducted over the phone with city officials who had knowledge of specific details of the process.

Analytical Framework
The analysis took its theoretical inspiration in two strands of literature: (1) Transition science, where we found inspiration in the MLP framework and (2) institutional theory, with specific focus on institutional capacity. In addition, we considered the specific conditions for local planning in Sweden.

Sustainability Transitions and the Multi-Level Perspective
Sustainability transitions (ST) are understood here as goal-oriented processes aimed at achieving sustainability targets. ST are typically matters of "structural transformation", "systemic change", or "radical change" of socio-technical systems [1,21]. Furthermore, they should be seen as "complex and long-term processes comprising multiple actors" [1] (p. 24). Despite (or perhaps because of) the complexity that tends to characterize transition processes in practice, ST research is mainly focused on the "big picture" of sustainability transitions. It has provided several concepts and frameworks that explain and prescribe the general elements and dynamics of transition processes, for instance, MLP, Strategic Niche Management (SNM), and Transition Management (TM) [15,22]. In this study, we chose to apply MLP, which is a framework that provides an "overall view of the multi-dimensional complexity of changes in socio-technical systems" [15] (p. 495). It supports exploration of the dynamics of change processes with the focus on three analytical levels: landscape, regime, and niche.
The landscape level refers to slowly changing external factors that cannot usually be directly influenced by individual actors [21,23]. It has been described as "the surroundings of a particular societal system under study, where one sees trends with a relatively slow progress and/or developments with a high autonomous character" [24] (p. 545).
The regime level refers to "the 'dominant' configuration of actors, structures and practices" that preserves the functioning of a societal system [24] (p. 545). A regime usually provides stability, but may also cause lock-ins by defending the status quo. For a transition process to be successful, it needs to involve a transformation of the regime [23,24].
The niche level refers to a more narrow, micro-level setting where "non-conformism and innovation can develop" [24] (p. 545). In the literature, niches are often described in terms of "protected spaces" where (radical) innovation occurs, evolves, and eventually starts to affect the dominant regime and ultimately perhaps even overturn it [21,23].
According to the MLP, transitions are dependent on developments on all three levels, and there may also be interactions between the levels in specific transition phases [23,25]. Transition processes are often associated with a "power struggle between the current regime, upcoming niches, and landscape pressures." [24] (p. 545). For instance, when niche actors with new ambitions and goals start to challenge an existing regime, that regime might respond in a way that absorbs and "locks in" the niche and prevents radical ideas from evolving and threatening the status quo. Alternatively, the niche might be able to resist this response and start to challenge the regime more seriously, which is a necessary condition for a transition to continue.

Institutional Conditions and Institutional Capacity
A strength of the MLP framework is the way it provides a comprehensive idea of the dynamics of change processes. However, in our analysis of the new urban district in Hyllie, we needed a framework that could support exploration of more specific aspects of the planning process. We turned to institutional theory, which is one of the most prominent theories for explaining stability and inertia in social systems and processes [26,27]. The term institution refers to "humanly devised constraints that structure political, economic and social interaction" [27] (p. 97). These constraints may be both formal and informal; formal institutional dimensions comprise explicit procedures, formal rules, and regulations, whereas informal dimensions are related to norms, traditions, discourses, codes, and conducts that are not explicit, but still powerful dimensions of social interaction [27].
Institutional perspectives have already been introduced in previous ST work [28]. We set out to continue this work by introducing the concept of institutional capacity, which we considered suitable for analyzing policy and planning processes that aim for a certain outcome, but meet some challenges along the way. Based upon the work by Magalhães et al. (2002), we chose to focus attention on three key dimensions of institutional capacity: knowledge resources, relating to formal, informal, and tacit knowledge, frames of reference and the meaning and understanding of a certain issue; and relational resources, relating to the presence of social networks that link various actors together. Relational resources also relate to the range, the bonds, and the level of trust and reciprocity in and between the networks, as well as power relations; and mobilization capacity, which is the ability of actors to use knowledge and networks in order to challenge a prevailing power structure and to pursue a process of change. As part of this, Magalhães et al. (2002) direct specific attention to critical change agents [16,29].
In the empirical analysis, we combined the MLP with the theory of institutional capacity. The MLP provided a structure and overview of key events and elements in the planning process, while we applied the three dimensions of institutional capacity to identify and discuss more specific aspects that shaped the outcome of the process in this specific case.

Local Planning in Sweden
In addition to the MLP framework and the concept of institutional capacity, the analysis needed to be informed by key characteristics of land use planning in Sweden. As noted above, land use planning in Sweden is highly decentralized, which means that local municipalities hold a key position in achieving national environmental quality objectives and social and economic ambitions. However, although Swedish municipalities have a powerful position in land use development, they cannot act independently. Several studies have shown that private interests have gained an increasingly influential role in land use development in Sweden over recent decades, due to generally weak municipal finances [30].
Swedish municipalities involve a number of administrative units with different roles in the land use planning process. Departments working with business development, economic development, and real estate issues are often involved in the part of the process that involves establishing contacts with private actors who are interested in buying land and investing in new buildings. Other parts of the planning process (i.e., formal stages of developing plans and programs) are managed by the city planning department. In addition, there is normally a specific department that manages the planning and management of technical infrastructure, such as roads, sewers, etc. Most Swedish municipalities also have a special environment department that has specific competence in matters related to the environment and sustainable development. Altogether, Swedish municipal planning is a complex activity, which involves a variety of professions and perspectives. This has consequences for the management of sustainability. Previous research has shown that sustainability targets often lead to conflicts and tensions in the planning process, since different departments make different priorities [31]. This aspect is important to understand in the analysis of municipal planning as a potential arena for working towards sustainability.

Background
Hyllie is located in southwest Malmö, in an area for which there were plans for development in the 1960s. However, an economic downturn in the 1970s put an end to expansion and the area remained a green wedge (farmland) close to the city. New plans to develop the area were drawn up during the 1990s as a consequence of the City Tunnel project, which aimed to link Malmö Central Station with the Öresund bridge and Denmark. The project included construction of a new railway station in Hyllie and the area was re-zoned for development due to its strategic location in both the city and the region.
At that time, Malmö city council mainly expected Hyllie to be a new communications node with park-and-ride facilities and a site where new workplaces and services could be located [32]. However, in the mid-1990s private developers started to discuss the greater potential of the area. They envisioned developing Hyllie into a place for entertainment, shopping, and other activities, and eventually the city council started a planning process that built on these ideas, as further illustrated in the section 'Regime level' below.
The main focus at the early stage of the planning process was a number of large-scale buildings around the new railway station: a hotel, a sports and entertainment arena, and a large shopping center. Eventually, sustainability also became a key issue for consideration [33]. When the city council adopted a new environment program in 2009 [18], it defined "climate neutrality" as one (of four) main environmental targets for the city. The city started to frame Hyllie as a key project in the realization of its vision for Malmö as an attractive and sustainable city, and as a pilot project to try out solutions for the target of climate neutrality. A milestone was reached in February 2011, when the City of Malmö signed an agreement with the energy distributor E.ON and VA Syd, an organization that manages water, waste, and sewage in Malmö municipality. The agreement, which was called the "climate contract", confirmed their joint commitment to turning Hyllie into the most climate-smart city district in the region, with a target that by 2020 (at the latest), the energy supply would consist of 100% renewable or recycled energy [34].
Meanwhile, the actual development of the area had continued, but more slowly than expected, due to the economic downturn of 2007-2008. Malmö Arena, which seats 15,000 spectators, was opened in 2008. The new railway station opened in 2010 and the exhibition and conference center and the Emporia shopping center in 2012. The housing projects were somewhat delayed and the first occupants moved into their new houses in early 2014. More housing projects are underway, together with a water park, a "sustainability house", schools, a new public indoor swimming pool, and more workplaces. In addition, there will be allotments for urban gardening and public green spaces.
During the later stages of the development process, the City of Malmö has promoted Hyllie as a flagship project for sustainable urban development with a specific focus on the target of climate neutrality, which has gained a key position as a central target for the city as a whole. At the same time, however, Hyllie is a key project in the city's ambitions to achieve goals of attractiveness and economic growth and this dual focus has led to tensions within the project. As we see it, it is precisely these internal tensions between different goals and priorities, and the changes over time, that make the case of Hyllie an interesting example to explore.

Landscape Dimensions
In the following, we explore key events in the planning and development process of Hyllie and, based upon the frameworks of MLP and institutional capacity, identify and critically discuss the specific local factors that had a major impact on the actual development of the area.
The landscape dimension in the MLP framework is targeted at slowly changing external factors that affect a certain transition process. Through interviews with public officials, scrutiny of planning documents, and earlier research on Malmö, we identified three such factors that were influential for development of the Hyllie district: (1) Transformation of Malmö from an industrial to a post-industrial city. This is a long-term trend that dates back to the 1970s, when Malmö was hit hard by a general economic downturn that led to the widespread closure of local industries. Another economic downturn in the early 1990s reinforced the view of Malmö as a city in decline [17,35]. At the same time, however, it paved the way for efforts to move Malmö in a more 'post-industrial' direction during the late 1990s and early 2000s. The transformation of Malmö's identity and image was brought about by, among other things, conversion of the old shipyard area into a university and the national housing exhibition 'Bo01 , which included a new 'sustainable' residential area in the former Western Harbor [35,36]. The Öresund bridge, completed in 2000, reinforced the view of Malmö as a hub in the emerging region of Öresund [37,38]. The development of Hyllie, with its strategic location on the railway line between Malmö city, Copenhagen's international airport Kastrup, and the city of Copenhagen, appeared to be a logical continuation of this development path for the city.
(2) A general shift to more market-oriented planning and network governance in local urban development in Sweden. Today, local development processes are often driven by complex governance networks of public and private actors [39,40]. "Place marketing" has become a key issue in local urban development strategies in Sweden [41]. One of the clearest illustrations of the market-oriented approach in the development of Hyllie was the way in which private entrepreneurs influenced the development process at an initial stage (see section 'Regime level' below).
(3) An overall increased focus on sustainable urban planning and climate neutrality, which has been evident in Swedish policy during the last 15-20 years. In Sweden in general, sustainability was given a central position at national policy level during the late 1990s-early 2000s [42,43]. In particular, Malmö has become known for its high ambitions regarding sustainable urban development [17] (p. 19. It was also Sweden's first Fairtrade City, in 2006, and was ranked as Grist's fourth greenest city in the world in 2007. Sustainable development became prominent in the renewed comprehensive plan for the city in 2005, which emphasized all three dimensions of sustainable development and set out Malmö's ambition of becoming an attractive city for residents, enterprises, and visitors [40,44]. Since then, attractiveness and sustainability have continued to be the two key features of Malmö's overall vision [45], and key dimensions in the development of Hyllie since the formulation of the plan program for the area in 2003 [33].
Malmö's focus on sustainability increased around 2008/2009, when the city decided on a new environment program. This program states that Malmö will become Sweden's "most climate-smart city" and that the City Council will be climate-neutral by 2020, with the whole city running on 100% renewable energy by 2030 [35]. All the while, however, the City has maintained its focus on economic growth and attractiveness, without any explicit discussion about the potential conflicts between these aims and the more specific target of climate neutrality.

Regime Level
The regime dimension in the MLP permits exploration of 'dominant' configurations of actors, structures, and practices. In the development of Hyllie, the City of Malmö held a central position, with its Real Estate, City Planning, Environment, Streets and Traffic departments playing key roles throughout the process, alongside other public bodies at regional and national level. In addition, a range of private actors, such as real estate developers, construction companies, and others, became involved. Within this broad constellation, some actors were more influential than others and their roles shifted throughout the process.

The Initial Framing of the Development Area
Interviews with key individuals indicated that a limited number of private entrepreneurs were very active in the initial and partly informal stage of the development process (late 1990s), in which they made an effort to influence the overall framing of the project.
In terms of institutional capacity, the events taking place at this stage illustrate the importance of knowledge resources, in this case expressed by the ambitions to develop and establish joint ideas, visions, and frames of reference about the type of place Hyllie could and should be. As noted in the 'Background' section, the City wanted to develop Hyllie as a communications node with some workplaces and services. However, private entrepreneurs had other ideas. For instance, a local real estate developer produced a film in 1997 to visualize what Hyllie could be like in the future, with the focus on mobility, shopping, and entertainment [46]. When it comes to relational resources and mobilization capacity, individuals working in the city administration described how private actors contacted leading councilors and officials in the City Council to convince them about their ideas, with the Planning department eventually including these ideas in the formal planning documents for the area (Interviews 1, 2). In the comprehensive plan for Malmö from 2001, Hyllie is presented as a new communications node with the right conditions for setting up: "hotels, conference centers, exhibition and sporting facilities, tourist attractions, a secondary school, etc., as well as workplaces and housing in both urban and rural environments" [32] (p. 78).
The intentions were taken further in the plans for the central development area released in 2003, which framed the development of Hyllie as a "unique location" which should be developed into "a meeting place with a creative environment in the middle of the Öresund region" [33] (p. 11).
The success of private entrepreneurs in influencing the planning process owed much to their knowledge resources and relational resources, but also to their capacity to provide funding for new investments. An official from the City of Malmö's Planning department summarized his reflections on the matter thus: "To approach a city council on the verge of bankruptcy and begin talking about these kinds of investments-well, there is just no politician on earth who would say no" (Interview 2). In terms of institutional capacity, private entrepreneurs with their own funding thus had a good opportunity to act as critical change agents and influence the direction of the land use development process.

Sustainability Added as an Additional Goal
After 2003, planning process documents show that the development process for Hyllie was characterized by an increased focus on sustainability [33,40,47,48]. For instance, a document from 2007 presenting key principles for land allocation includes a whole chapter on sustainability and states that sustainability shall be the key rationale of the development process [48] (p. 16). This can be seen as a reflection of the increased emphasis on sustainability developing in Swedish policy and planning in general during these years (see section 'Landscape dimensions' above). In the case of Hyllie, it can also be seen as institutional capacity building, with city officials working on environmental issues using their knowledge resources to strengthen the sustainability profile of the ongoing development (Interview 3). In practice, however, it was unclear how to interpret and operationalize the general objectives and ambitions regarding sustainable development. When the detailed planning process started, the link to sustainability was still not clarified. Instead, there was a remaining emphasis on accomplishing the large-scale buildings which were primarily aimed at promoting the attractiveness to visitors, thereby strengthening economic growth.

The Place-Marketing Process Starts
When we asked our interviewees what the increased focus on sustainability came to mean in practice, several of them remembered these years as a period characterized by uncertainty. A general economic downturn at some time around 2007/2008 meant that the development process slowed down. The communications manager for Hyllie reported that she was approached by private entrepreneurs and real estate developers querying the City's plans for the area, since they did not consider the existing goal formulations to be clear enough for them to market their investments. A defining stage in the process occurred in 2008, when the City's Real Estate department initiated a visionary process for Hyllie in order to project a clearer vision and so stimulate investment in the area (Interview 1).
The visionary process, which was carried out as a place-marketing process, was led by the Real Estate department. It involved officials from various City departments, as well as private developers and lobby organizations working for developers in the Öresund region [49][50][51]. The aim was to identify and develop a joint vision that all actors involved in the area could share. In autumn 2008, the process led to the identification of three key values that were intended to guide further development [52]:

•
International flavor • Self-confident identity ("assertiveness") • Sustainable experience The communications manager explained that in the first outline of the key values for Hyllie, sustainability was not included at all, which led to questions from some of the participants in the process. In the end, it was decided to add 'sustainability' to 'experience' and so the term 'sustainable experience' was created. However, according to the communications manager, it remained unclear what 'sustainable experience' actually meant: "'Sustainable'-I don't know what it means [in this context] but 'experience'-well it was because we knew that there would be an arena and a shopping mall, we knew that there would be activities here." (Interview 1).
After agreeing on the key values, an overall vision was formulated that included sustainability: "With communication at the center, Hyllie will be a driving force for growth and sustainable development of the new Malmö-an outward-looking place that cuts across borders and has the whole world as its arena" [52] (p. 9).

A Network of Actors Focusing on Economic Growth and Attractiveness
As the process continued, the place-marketing process evolved into a closer collaboration between the Real Estate department and commercial interests. In 2009, they formed a network named "Samsyn Hyllie" (literal English translation "A joint vision for Hyllie", later given the English title "Hyllie Harmony"). However, some actors, for instance the Environment department, withdrew from the process and did not participate in this network. Eventually, Samsyn Hyllie evolved into a key actor in efforts to attract consumers, new residents, companies, etc. to the area [53].
It is clear that, even at an early stage, actors, goals, and practices related to economic growth and attractiveness were highly influential for the development process. Throughout the process, new goals and ambitions were added (relating e.g., to sustainability), but the actors focusing on shopping, high mobility, economic growth and attractiveness, for instance the city's Real Estate department and the network of actors in Samsyn Hyllie, had the strongest influence. In terms of MLP, they evolved into a powerful regime with a significant influence over how the project was framed and how it developed over time. Specific explanations can be found in the way these actors managed to develop knowledge resources (establishing joint key values for the development of the area) and relational resources (linking a broad range of actors together). Together, this provided them with significant mobilization capacity and power to influence the development of the area significantly. Even though sustainability was added as one of the project's central ambitions, the concept remained vague and unspecific and did not lead to any substantial change in the established development agenda.

Niche Developments
In MLP, the niche level is the analytical dimension that gives emphasis to specific initiatives, agents, and arenas that may not seem to be of any significant importance, but can eventually pose significant challenges to a dominant regime (unless the situation leads to lock-in, where the status quo remains) [21,23]. In the case of Hyllie, events relating to the climate contract provide an example of such niche development. The climate contract was signed by the City of Malmö and the energy companies E.ON and VA Syd in 2011. The intention was to enact a joint development process that would lead to achievement of climate neutrality in Hyllie. As mentioned in the background section, this was perceived as an important initiative for realizing the goal of climate neutrality for the whole city. This was not because such a contract, in one specific urban district, was considered able to solve the entire climate ambition for the city. Rather, it was seen as an innovative and promising approach that would hopefully prove effective in other neighborhoods and contexts [34,54]. As explained by an official from the Environment department, Hyllie was defined as a test area where practices and innovations for climate neutrality should be developed and tested and then later implemented for the whole city. The ambition was that Hyllie should be climate-neutral a decade before the rest of the city, i.e., by 2020 (Interview 3).
The City's Environment department was a key actor in operationalizing the goal of climate neutrality in Hyllie. To understand this specific part of the development process, it is relevant to reflect upon the specific role of the Environment department, compared with that of other departments involved in the planning and development process. Unlike the other departments, the Environment department has no formal mandate in the land use planning process. Its role is to coordinate the City's efforts on the environment and sustainability, and to contribute relevant expertise to ongoing and planned projects. Since Hyllie was the next major development project for the city when the new environment program was formulated, it naturally became a main object of the Environment department's attention (Interviews 2, 3, 8). One interviewee from the Environment department talked about various challenges they were facing once the environment program had been settled. He referred to notable resistance among private developers involved in Hyllie and recalled an event at which they presented the new environment program during a collaborative dialogue (not part of Samsyn Hyllie). According to the interviewee, the representatives at the meeting, who came from other City departments and private developers, responded to the goal of climate neutrality with a certain amount of skepticism: "The reaction was something like: 'Is the municipality serious about this?'" (Interview 3).
The target of climate neutrality, which meant that energy consumption among residents in Malmö should be reduced by at least 20% per person by 2020 and by a further 20% by 2030, and that the whole municipality will run on 100% renewable energy by 2030, was perceived as difficult and highly demanding, both by the private developers and by the other City departments. According to the interviewee from the Environment department, the skepticism was based upon a belief that the City did not have the full power to meet this target on its own (ibid.). For a city like Malmö, with poor municipal finances, it is not possible to drive all development by itself. Thus, the skepticism was caused by the insight that, if the intended goal was to be realized, a number of other actors involved in the urban development process must also be willing to work for its realization. This, in turn, led the representatives from the Environment department to initiate new collaborations. They contacted the large energy companies in the area, E-ON and VA Syd, and eventually the dialogue between these parties resulted in a joint agreement to work together to achieve the goals of the environment program and to supply what was needed to enable developers to contribute towards achieving those goals (ibid.).
The climate contract for Hyllie was signed in February 2011. It defines the direction for development in Hyllie as: "Hyllie will show the way for Malmö's future development as a sustainable city and it will be in the frontline when it comes to innovation and the ability to tie supply to consumption and patterns of usage. Energy solutions are to be based on loop models and have a high degree of self-sufficiency. We shall see to it that the Hyllie energy concept has a positive balance of energy, that is, that energy production exceeds energy consumption. Hyllie is to develop into the Öresund region's most climate-smart district and become a global role model for sustainable urban development." [34].
The climate contract identified a range of specific goals and priorities, for instance, that the energy supply must consist of 100% renewable or recycled energy by 2020; that integrated infrastructure systems must be developed for electricity, gas, heating, and cooling to enable optimal combination of local and central production; and that the buildings in Hyllie must be energy-efficient and have the facilities needed to realize the potential of smart systems. Moreover, walking and cycling must become the main focus of transport planning, with public transport seen as supplementary. Furthermore, smart solutions for gas and electric vehicles must be created [34]. Another of the contract's stated aims was to involve property owners and other actors working in property development through an ancillary contract. Thus, the intention was for the climate contract to function iteratively and increasingly involve more actors.
The climate contract was founded upon the ambitious goal of climate neutrality. Based on the formulations in the environment program, it is clear that this goal is not intended to be merely about technical progress and efficiency, but also about people's lifestyles and consumption habits. However, the climate contract between the city and the energy companies had a significant focus on technological development, and other issues were not given any particular attention (see below). The question is thus whether the climate contract is a failure, from a ST perspective, or whether it may be seen as a promising initiative that paves the way for a more radical transition.
As stated by Avelino and Rotmans [24] (p. 560) when writing about different phases of transition processes, the pre-development of a transition process requires "innovative power" to be developed at the niche level, for instance through the creation or discovery of new resources. As we see it, the climate contract reflects such a potential. Initially, there was skepticism among private developers about the new goals for climate neutrality. This skepticism can be seen as a barrier to transition. However, it also motivated the Environment department to initiate a creative dialogue with major energy companies and, eventually, formulation of a climate contract that set out a joint agenda for how to work towards the goal of climate neutrality.
Seen through the lens of institutional capacity, the climate contract is thus a clear example of rapid development of mobilization capacity. The events may also be viewed in terms of what Avelino and Rotmans call "innovative power", where the climate contract acted as an organizational innovation that promoted a new type of cooperation between actors in the public and private sectors. It meant e.g., that a substantial part of the responsibility for reaching the new climate goals was taken by private actors. Overall, the climate contract, which assumes that all involved parties have a genuine interest in long-term sustainability goals, appears to be something quite different to e.g., Samsyn Hyllie, where companies were seen as actors geared towards individual economic targets and competition rather than with a genuine interest in long-term sustainability.
In this case, the climate contract became a unifying force for various efforts to achieve ecological sustainability. However, as the contract gained ground, it also lost some of its radical nature. One example relates to the issue of energy efficiency, which the contract stated should be handled in the construction of buildings. According to our interviews, there was a resistance among some developers to meet such tough requirements (Interview 1). This led to compromises and lowered demands on the actors involved. In the end, the City introduced different "levels of ambition", which meant that developers could commit to greater or lesser levels of climate neutrality (Interview 1).
As mentioned above, there were also other limitations in the climate contract and in the degree to which it could function as a radical force for sustainability transitions, not the least in the way that it kept more difficult and challenging issues from the agenda. As regards transport, clear demarcations were made in the contract and only private transport within the area was included when considering energy consumption, while goods transport and private transport to and from the area were not mentioned. Another difficult issue is consumption, which was critical for the development of Hyllie since the Emporia shopping mall was to be one of its main attractions. Reconciling this large-scale, luxury shopping center with the overarching goals for sustainability emerged as a complicated theme in interviews. Throughout the interviews, it was clear to us that staff at the City Council did not feel very comfortable criticizing the role of consumption in relation to sustainable development (see e.g., Interviews 2 and 3).
In summary, the actual significance of the climate contract in the development of Hyllie is debatable. On the one hand, it can be taken as a sign of an emerging and ongoing transition, where actors engaged in meeting the goal of climate neutrality managed to take steps forward to challenge a former dominant regime. On the other hand, it can be interpreted as an attempt that failed. According to the latter view, the climate contract was a promising and potentially powerful initiative that was eventually absorbed and locked in by a dominant, market-oriented regime. Looking at the more recent development of the area, we note an increased emphasis on climate targets [55], but the actual development of the area is as contradictory as ever. In terms of institutional capacity, the situation appear as a continuous struggle between different development ideas, where knowledge resources, relational resources, and mobilization capacity are of continuing importance. A critical issue in this process is the existence of critical change agents with power, resources, and a political mandate to pursue a more substantial transition towards climate neutrality and long-term sustainability. In the case of Hyllie, the actors with a clear focus on sustainability goals have so far not been strong enough to actually enact a more substantial transition where critical issues related to consumption and lifestyle are also considered. One of the explanations for this relates to the internal organization and funding within the City Council, where e.g., the Environment department has a limited position in the land use planning process, being dependent on external funding or initiatives from others in order to initiate radical innovation.

Concluding Discussion
This case study of Hyllie provides a rich illustration of the pre-requisites for working with sustainable development targets in a decentralized, market-oriented land use planning context, where different actors and departments within the municipality organization have different roles, perspectives, priorities and power resources.
As regards our research task, which focused on the overall framing of the Hyllie project, we noted that the development process was influenced by a number of external conditions, as described in the 'Landscape dimensions' section of this paper. One such condition relates to the general development agenda for Malmö. Even though sustainability targets were high on this agenda, the role and status of these targets were never clarified in relation to other goals, such as economic growth and attractiveness. Thus, when sustainable development was eventually added as a key dimension for the development of Hyllie, this was done without any concrete decisions about substantial changes to the general planning direction for the area. Planning documents and interviews bear witness to a widespread interpretation of sustainability as a component that could be added on fairly easily to the existing development agenda. Altogether, this approach resembles the type of "post-political" view discussed in previous studies, sometimes with reference to the term "ecological modernization" [43,56,57]. This interpretation of sustainability served as a type of knowledge resource and 'frame of reference' that caused problems with integration of sustainability goals in practice. In the planning process, it led to the pursuit of a vague and unspecific approach to sustainability, which counteracted any deeper and/or critical reflections on goal conflicts or other difficulties that the sustainability objectives raised for the development process.
The vague and unspecific approach to sustainability was, however, not only caused by a general, external policy discourse or frame of reference. As regards our second research task, which focused on the interplay between different actors in the process, we identified a network of powerful actors who became highly influential in the development process. These actors, who collaborated in the place-marketing process and through the network Samsyn Hyllie, maintained a primary focus on economic growth and attractiveness, even after adoption of the new environment program. Their actions can be explained in terms of institutional capacity building, where the place-marketing process became an arena for establishing and developing knowledge resources (ideas, visions, joint understandings) and relational resources (networking, establishing trust between actors). Together, this generated significant mobilization capacity for these actors to influence the subsequent development process.
Our third and last research task focused on the creation of arenas and collaborations with the potential to challenge the status quo of local land use planning. According to the MLP, transition processes are always dependent on developments on all three levels (landscape, regime, niche). Previous research has drawn attention to the dynamics between these dimensions and emphasizes that the type of substantial change addressed by the ST literature involves challenging, breaking down, and finally replacing an established regime with a new one [24,58]. However, challenges to a regime do not always lead to its breakdown and replacement, since a transition initiative can be absorbed and "locked in" by the existing regime [23,24].
The ST literature proposes that the type of innovative power required for challenging and breaking down an existing regime normally comes from the niche level [24]. In this case, we identified the climate contract for Hyllie as an illustration of a niche development with radical transition potential. However, the question is whether this climate contract will actually lead to any long-term, substantial transition.
In the empirical analysis, we presented two possible interpretations of the events surrounding the climate contract so far. One is a hopeful interpretation, according to which the climate contract is a sign of an emerging and ongoing transition towards climate neutrality, which is a key dimension of a wider sustainability transition.
However, Hyllie is still an area with a strong emphasis on shopping, entertainment, and high mobility, and is characterized by large buildings, motorways, and proximity to an international airport. The climate contract has so far not led to any radical new agenda for ST-this would require clear indications of a more structural transformation. Thus, our second interpretation suggests that the climate contract is an attempt that failed, i.e., led to "lock-in" and reinforcement of the status quo. According to this interpretation, the climate contract has been used by the dominant regime to promote the attractiveness of the area, without actually changing anything in a more substantial sense.
The long-term results of the development in Hyllie remain to be seen. Based on our analysis, the greatest potential of the climate contract lies not in its direct impact on reducing energy demand and climate emissions, but in the institutional innovation that it represents. There is an important difference between the collaboration established in the climate contract and that in Samsyn Hyllie. In the latter case, the focus of the public actors involved was on adapting and adjusting local planning to commercial perspectives and interests. In contrast, the climate contract collaboration was characterized by an ambition to find new joint development paths and ways of working. The climate contract process was far from uncomplicated and unproblematic, but it was based on consensus among public and private actors about the need to start doing things differently. Thus, the climate contract is an interesting example of a new type of collaboration between public and private actors, which carries an important transition potential. The type of collaboration that the climate contract represents is likely to be important in decentralized, market-oriented planning contexts elsewhere.
In the present analysis, the MLP framework proved to be a fruitful approach for sorting and discussing key features of the planning process in general. However, it did not provide much help in identifying the specific aspects of the planning process. Combining MLP with the theory of institutional capacity enabled us to identify and discuss specific elements and events of the planning process, particularly knowledge resources, relational resources, and mobilization capacity. A general lesson from our work is that it is important to pay attention to specific local institutional conditions as key features of sustainability transition processes. When it comes to the situation in Malmö, we found that it is a city with high ambitions regarding the environment and sustainability. However, we also found that the sustainability targets proved to be difficult to pursue when they led to tensions in relation to other long-term goals for the city's development. In this case, the city's ability to give priority to sustainability targets was particularly compromised due to weak public finances. As a response to the recession in 2007/2008, the city adjusted the aims for the development of the area, and developed a vision where sustainability was given lower priority than before. The experiences from Malmö testify to the need for clear political priorities and well-considered strategies for persisting with environmental and sustainability goals even under uncertain economic conditions. The case shows, in particular, the importance of maintaining knowledge resources and relational resources, for instance through networking among actors within the local administration.
A general reflection arising from our work is that it is difficult for a single city, in a decentralized, market-oriented context, to enact a radical transition on its own. Malmö is one of Sweden's most ambitious cities as regards sustainability, but our analysis revealed a range of tensions and difficulties that hampered the work for sustainability in urban planning practice. In this case, rather conventional priorities related to shopping and high mobility persistently permeated the development of Hyllie, even when new ambitions related to sustainability were added to the development agenda. This example shows that an individual city, however large, may have a weak position vis-à-vis private entrepreneurs. Urban planning is-and will continue to be-an important arena in the work for wider sustainability transitions. Cities have an important role to play, especially in decentralized contexts where they have great freedom to initiate new projects and ways of working, with innovative potential. However, relying heavily on the local arena to manage initiatives for sustainability transitions might also be a risky strategy if there is too strong a dependence on private entrepreneurs to make things happen. A key conclusion from this analysis is that if cities are to be able to work purposefully towards sustainability transitions, they would benefit from receiving stronger and more explicit support from other public actors, such as regional and national authorities, at least within the market-oriented planning context represented by the present case. While the institutional conditions described in this paper are specific to the situation in Malmö and Hyllie, they are probably not unique to that location. However, there is a need for more comparative studies of local conditions, so that general lessons can be learned from obstacles and opportunities in local planning for sustainability transitions.