City Districts as Sites of Experimentation and Learning for Sustainable Development : An Analysis of mid-sized Swedish Cities

(1)

Department of Thematic Studies Environmental Change

_______________________________________________________________________________

City Districts as Sites of

Experimentation and Learning for

Sustainable Development

An Analysis of mid-sized Swedish Cities

Mareike Oelrichs

MSc Thesis (30 ECTS credits) Science for Sustainable Development

(2)

ii

Upphovsrätt

Detta dokument hålls tillgängligt på Internet – eller dess framtida ersättare – under 25 år från publiceringsdatum under förutsättning att inga extraordinära omständigheter uppstår.

Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka kopior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervisning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säkerheten och tillgängligheten finns lösningar av teknisk och administrativ art.

Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsmannens litterära eller konstnärliga anseende eller egenart.

För ytterligare information om Linköping University Electronic Press se förlagets hemsida https://ep.liu.se/ .

Copyright

The publishers will keep this document online on the Internet – or its possible replacement – for a period of 25 years starting from the date of publication barring exceptional circumstances. The online availability of the document implies permanent permission for anyone to read, to download, or to print out single copies for his/hers own use and to use it unchanged for non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility.

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

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

(3)

(4)

iv

List of Figures

Figure 1 Map of Sweden showing the location of cities studied and Sweden’s three largest cities for locational reference ... 18

List of Tables

Table 1 Summary of the Internet and Literature Research on Swedish Cities ... 12 Table 2 City, Date and Position of Interviewee ... 13 Table 3 Characteristics of the six cases studied ... 20

(6)

1

1. Abstract

In Sweden, it is expected that more than 90% of the population will live in cities by 2040. Therefore, cities have an important role in sustainable development to ensure high quality of life in the face of climate change and social challenges. This thesis uses the framework of Urban Living Labs as a form of experimentation to analyse how evaluation and learning in municipalities take place. Districts identified as sites of experimentation in the six mid-sized municipalities Umeå, Gävle, Jönköping, Borås, Örebro, and Lund are studied using a multiple case study approach. Secondary data is discussed together with findings from interviews. The results show that the aims and challenges of experimentation are similar in all cases. Municipalities aim at increasing their attractivity with these extraordinary districts, but they also contribute to sustainable development through innovation in different sectors (e.g., energy, construction, social life). The importance of structured evaluation for learning is recognized, but only few have (as of now) developed a framework for this. Double-loop learning (or reflectivity) is hinted at in the interviews, but for most projects, it is too early to identify whether it will take place. Experiments are contributing to sustainable development on a larger scale mostly through interactions in networks. Personal Networks, as well as international networks (e.g., initiated by the EU), have a special role as these provide incentives for experimentation, quick informal exchange, as well as spread findings and knowledge to municipalities outside of Sweden.1

Keywords: Urban Living Lab, Experimentation, Case Study, Sweden, Learning

2. Table of Abbreviations

EnoLL European Network of Living Labs ESS European Spallation Source

EU European Union GHG Greenhouse Gas

KTH Kungliga Tekniska Högskolan ÖBO Örebro Bostäder AB

PBL Planning and Building Act PED Positive Energy District

SGBC Sweden Green Building Council TM Transition Management

ULL Urban Living Lab

3. Introduction

More than half of the worlds’ population lives in cities. In 2019 this share was already at 84% in Northern Europe. The United Nations estimates the urban population globally to reach 68% by 2050, with Sweden exceeding the 90% benchmark already in the 2040s (United Nations, 2018, 2020). The densification of urban areas allowed many to become hubs of production and wealth (Rode, 2013: 80). However, this also led the same areas to become a focal point of its negative consequences. Cities consume large amounts of resources, including almost 70 percent of the global energy, thus responsible for the same share of globally emitted Greenhouse Gases (GHGs) and large amounts of waste (Betsill and Bulkeley, 2004: 477; Bibri, 2018: 1). This can rapidly lead to a range of environmental, economic, and social problems, with congestion, pollution, and social exclusion being only a few examples (Hendriksen and Peereboom, 2013: 129). On top of this, climate change is increasingly becoming reality around the world (Kabani

1_{Referencing Style following Urban Studies, ISSN: 00420980, as described in}

(7)

2

and Kabani, 2013: 72), threatening cities’ prosperity with extreme weather events, floods, and others.

Research on municipal strategies towards climate change has been conducted since the mid-1990s, however, their contribution has only recently been recognized in international policy-making (Castán Broto and Bulkeley, 2013: 92). Similarly, neighbourhoods have been a focal point for policy-making for over half a century as they are perceived as primary building blocks for democracy with the ability to reduce inequalities and increase participation in planning and provision of services (Bailey and Pill, 2011: 927). Furthermore, they can influence purchasing decisions by a range of financial instruments (Hendriksen and Peereboom, 2013: 131). Thus it can be said that they are critical actors in climate protection (Betsill and Bulkeley, 2004: 477). The global impact of current lifestyles and consumption is increasingly being understood. Similarly, awareness is building that new practices are needed, to steer society towards sustainability (Beers et al., 2019: 229). The sustainability concept has been increasingly used in urban planning to preserve natural systems and resources while maintaining prosperity, attractiveness, and socially equitable communities, thus embracing opportunities that sustainable development provides (Dehghanmongabadi et al., 2014: 1; Kabani and Kabani, 2013: 72; McCormick et al., 2013: 6). However, the complexity of sustainability and other challenges are not reduced by tackling them at city level. Barriers of spatial, temporal, and institutional scale must be overcome. Nevens et al. (2013: 112) describes the attitude associated with these challenges as “’not my turf’, ‘not in my term’, ‘not my business’”. Scholars criticise the path-dependency of our societies, which limits the ability to find new solutions (Berglund-Snodgrass and Mukhtar-Landgren, 2020: 98). Meadowcroft (2009: 329) also describes this situation as a socio-technological lock-in, where incumbent actors and technologies have an advantage of the current systems, such as existing infrastructure and regulations, therefore making change difficult. Solutions today instead often favour the system in place (for example in the discussion about efficiency), thus only allowing for marginal change rather than radical steps towards sustainability (Burch et al., 2014: 472).

Not only politicians but also planners, institutions, and citizens need to challenge their own beliefs and “break out of the box” (Albrechts, 2010: 1115) to allow for co-evolution of behavioural change, the market, and technology in sustainable development (Albrechts, 2010: 1116; Lindhult et al., 2016: 207). Experimentation, especially within the urban context has gained importance recently as one possible alternative mode of governance for change. It differs from traditional urban planning by learning from real-world interventions (see also chapter 5) (Evans et al., 2016: 1–2). So, diverse actors – including non-scientific – come together to find, or co-produce, a path to a particular vision of the future, while actively working with the uncertainty of today (Bulkeley et al., 2019: 2; Ersoy and van Bueren, 2020: 90; Nesti, 2018: 310–311).

There are many frameworks and forms of experimentation in the urban realm, examples include urban living labs (ULL), sustainability experiments, or transition experiments. Although these all describe slightly different approaches and concepts, there are overarching characteristics of experimentation in the urban realm (Sengers et al., 2019: 161). Main similarities of experimentation frameworks include the importance of the geographical context, the embeddedness of problems, and possible solutions. Therefore actors come together to test innovations in a given space (Berglund-Snodgrass and Mukhtar-Landgren, 2020: 96–97). A second main similarity is the aspect of learning from experiments. This is considered key to be

(8)

3

able to repeat, expand or strengthen successful experiments (Evans et al., 2016: 3; Fuenfschilling et al., 2019: 225).

Current research approaches are focussing on methods to embed experiments in a more meaningful way into cities, but before this can be done successfully, a greater understanding of experiments within their urban context – how they shape the urban area and governance – is needed (Bulkeley et al., 2019: 3; Evans et al., 2016: 10). This knowledge is currently still sparse (Chronéer et al., 2018: 2). Furthermore, little focus has been given in the literature to the learning aspect within different frameworks and how this might influence an experiments’ success and possible wider-reaching influence (Cuomo et al., 2020: 3; Ersoy and van Bueren, 2020: 89). In short: sustainability researchers have to mobilize knowledge that can support transformations or transitions towards sustainability through producing evidence about the effectiveness of different solutions (Caniglia et al., 2017: 39). This is currently missing and it is unclear how local findings are incorporated in other projects (Bailey and Pill, 2011: 929).

Sweden is seen as a pioneer in urban sustainability and environmental governance, with districts such as Bo01 and Västra Hamnen in Malmö or Hammarby Sjöstad in Stockholm widely studied and discussed. Attention to urban sustainability has increased and living lab settings have emerged in many large cities in Sweden, but also in the Netherlands and the United Kingdom (Sharp and Salter, 2017: 2). The fascination in experimentation, especially in the Nordic countries, has increased mainly due to external funding possibilities and the interest of companies to get involved in innovation in real-life settings (Berglund-Snodgrass and Mukhtar-Landgren, 2020: 97). While the large population hubs in Sweden have been central in the discussions around sustainability and experimentation in Sweden, the large number of cities around 100 000 to 160 000 inhabitants has not been researched much. As will be shown below, these mid-sized cities are expecting a major population rise, therefore their contribution to sustainability is of increased importance.

This thesis aims to analyse experimentation in the form of new districts in mid-sized Swedish cities, to understand how evaluation and learning from experiments take place. By knowing this on the local scale, the effect that individual experiments have towards transitioning to sustainability should be elaborated. Urban experiments in six Swedish cities are analysed with regard to their learning patterns and their general characteristics. Specifically, the research questions to be answered are the following:

1. What are the characteristics of the area and main aims of the experimentation site in regards to sustainable development?

2. How are experimental approaches evaluated and is the evaluation used to learn from? 3. How are learnings spread/communicated to contribute to sustainable development in

municipalities with similar characteristics?

Following this introduction, a background on planning in Sweden will be presented, proceeded by the theoretical background of this thesis including two of the frameworks used by researchers to classify and analyse urban experimentation (5. Theoretical Framework). In section 6 the methodology will be described including the process of choosing the five cities discussed as case studies in this thesis, the literature research, and the interviews conducted. Lastly, the findings will be discussed in relation to the research questions to encapsulate the main outcomes.

(9)

4

4. Background

In Sweden, municipalities are responsible for urban planning and thus sustainable development. A general framework is provided by a range of top-down regulations given by the European Union EU or the State. The main document is the Planning and Building Act (PBL – Plan och bygglagen), which defines obligations of municipalities for building permits and ensures that the right of individuals and different public interests are taken into account (Boverket, 2020; Building Sweden, 2021). Environmental regulations in accordance with the EU laws are given by the Swedish Environmental Code (MB – Miljöbalken), which describes minimum measures for environmental protection and coastline protection. (Building Sweden, 2021)

In the Swedish planning system, only the detailed development plans (detajlplan) and area regulations are legally binding, although comprehensive and regional plans provide a guideline for the overall planning directions. The regional level is (except in the Stockholm region) rather weak, it mainly comes into play when several municipalities join together to tackle infrastructure or climate issues (Boverket, 2020). Consequently, the municipality has the main decision-making power in urban planning. A comprehensive plan provides an overview of the municipalities’ development and shows how national goals should be achieved. In the area regulations, specific land use or environmental protection areas are legally bindingly defined. The detailed plan is the most important document in planning, it defines housing and recreational areas as well as public spaces, schools, or planning details such as housing type or height (Boverket, 2020).

Due to municipalities planning sovereignty in Sweden, interventions towards sustainability make the most sense at local level. For this reason, the thesis will focus on the use of experimentation in the urban realm on municipal level and how municipalities are learning for sustainable development. The framework used for this thesis focuses on interventions at district level within municipalities. In the literature, there are several reasons to support experimentation for sustainability at this level. Districts are small city units with similar requirements to the whole city regarding social and environmental factors (Dehghanmongabadi et al., 2014: 5). Similarly, innovations for public space or mobility can be planned at this more manageable unit size (McCormick et al., 2013: 11). This has also been realized in Swedish planning, which has switched from planning individual sustainable buildings to a more holistic approach of a sustainable urban environment (Smart City Sweden, 2020a).

(10)

5

5. Theoretical Framework

This section will describe two frameworks focused on experimentation in the urban realm which provide the theoretical basis for this thesis. Experimentational urban planning differs from traditional urban planning in that several aspects are approached differently: the problems tackled, the methods used, relation to stakeholders, governing tools, and the priorities within the process. Often, experimentation is still entangled in traditional planning processes (Berglund-Snodgrass and Mukhtar-Landgren, 2020: 99). At this point, it should be noted that the experimental approach, using the city as a living trial area, is not new. Already in the early ´90s a city neighbourhood was used and described as “Living Lab” (Chronéer et al., 2018: 3). Today there is a range of successors of Living Labs2_{. Many other frameworks also make use of}

experimentation in real-world settings, all with a slightly different focus. These range from Strategic Niche Management to Multi-Level Governance, Co-Production Processes, or Sustainability Initiatives (Burch et al., 2014: 471; Dijk et al., 2018: 3; Lam et al., 2020: 3; Nesti, 2018: 312; Nevens et al., 2013: 112). Often it is unclear what the delimitations of certain frameworks are or how the definition provided in single studies can be applied to other contexts. For this reason, the thesis will elaborate two major frameworks namely Transition Experiments and Urban Living Labs. Both have evolved as separate frameworks, nonetheless, an increasing number of scholars are describing ULLs as one type of Transition Experiment. The two approaches will be described below, going into detail for the characteristics of ULLs which will be further used in this thesis.

5.1. Transition Experiments

Transitions are seen as a change of technologies, markets but also culture and practice that evolve in many small ways over longer-term, rather than in instant revolutions towards a desired vision (Albrechts, 2010: 1118; Burch et al., 2014: 472; Nevens et al., 2013: 113). Transition Research has its focus on steering and supporting transitions (Beers et al., 2019: 229). Within this field several frameworks have been developed, all with a systemic perspective and focus on co-evolution and the complexity of path dependency, among them Transition Management (TM) (Beers et al., 2019: 230; Köhler et al., 2019: 4). TM was first developed at the start of the century, it is seen as a policy-oriented framework focussing on the complexity of societal systems and governance studies (Köhler et al., 2019: 5; Loorbach et al., 2015: 48). It was initially developed for sectoral transitions (e.g., of the energy sector), but has been developed significantly since then. Key of TM are strategic activities in a transition arena through experiments and innovation. One type of such experiment is Transition Experiments (Köhler et al., 2019: 8–9). These experiments as real-life approaches, bringing together different actors and facilitating a process of innovation. New ways of working or thinking about a societal challenge are developed, as new actors challenge dominant values (Köhler et al., 2019: 9; Nevens et al., 2013: 114). Most important in this process is the constant monitoring and re-evaluating of progress towards a joint vision (Köhler et al., 2019: 5). Only intentional learning activities can lead to a successful niche or experiment, the diffusion of new practices, and thus in longer-term to sustainability (Beers et al., 2019: 230; Fuenfschilling et al., 2019: 220).

Some scholars see ULLs as a type of transition experiment or arena, due to their real-world setting in the urban realm which also involves multiple actors (Luederitz et al., 2017: 5; Sharp and Salter, 2017: 2). ULLs combine several experiments in one area, which co-evolve and interact. In contrast to sectoral transitions, ULLs explicitly focus on a holistic approach (Nevens

2_{Such as Co-Labs, Urban Living Labs, organic ULLs, strategic ULLs, civic ULLs, City Labs, or Urban}

(11)

6

et al., 2013: 115). The embeddedness of ULLs in Transition Studies is demonstrated in the JPI Urban Europe programme. It funded several projects in 2013 and 2014, amongst other the GUST (Governance of Urban Sustainability Transitions) project which aimed at governing transitions through ULLs in Sweden, the UK, the Netherlands, and Austria (Sharp and Salter, 2017: 2; Voytenko et al., 2016: 3–4). However, ULLs by themselves often lack the resources for long term change, therefore they can only trigger successful transition when embedded in a network where findings are spread outside their spacial and institutional borders, an aspect that will be discussed in detail below (Scanagatta, 2020: 65; Voytenko et al., 2016: 8).

5.2. Urban Living Labs

Although exact definitions differ, ULLs have been described similarly in the literature. There have been several reviews focussing on the different definitions, such as by Leminen (2015, in Santonen et al., 2017: 8) who acknowledged over 70 different definitions. Often terms such as ‘Living Laboratory’, ‘Living Lab’ or ‘Transition Lab’ are used interchangeably. For this thesis, the main characteristics of ULLs will be described, with the term ULL referring to the aggregation of similar concepts. Since many scholars see experimentation in ULLs as an essential approach in transitioning society, many of the concepts described are similar to the characteristics of TM and Transition experiments.

ULLs are seen as a framework or environment for systematic innovation and experimentation with knowledge transfer between the actors in co-creation processes (Chronéer et al., 2019: 58; Ersoy and van Bueren, 2020: 89; Nevens et al., 2013: 115). There is a range of actors involved in the process, but the most important is the combination of user and stakeholder involvement (McCormick, 2016 in Sharp and Salter, 2017: 2). This includes a range of groups from children to elderly, researchers, visitors, and companies, making it a bottom-up process of urban planning (Cuomo et al., 2020: 1; Juujärvi and Lund, 2016: 17). The city or municipality is usually the initiator of the process and/or plays an important part in it (Scholl and Kemp, 2016: 90). Decision-makers have to support ULL developments and provide leadership and management. Therefore they are responsible for finding a balance of steering and enabling flexibility (Chronéer et al., 2019: 58–59).

The actors come together to develop new processes, systems, or services which are tested and evaluated to provide creative, innovative, or new solutions in a real-world context (Ersoy and van Bueren, 2020: 89). Thus, the ULL is centered around finding a solution for a challenge, which is both contemporary and of local concern. An example of this can be climate change and the threat of flooding (Chronéer et al., 2018: 58; Nevens et al., 2013: 115). Focus is often on a holistic approach of resilience or urban sustainability, not limited to environmental challenges (Santonen et al., 2017: 8). Similar to the size of the ULL, the complexity of problems addressed can vary from new technologies or services to complex social or economic problems. Often these challenges are multidimensional and therefore solutions are sought through a process with a multitude of actors and methods (Juujärvi and Lund, 2016: 17). The experimentation is limited to local scale, embedded in a real-world context, and geographically bound by it (Nevens et al., 2013: 115; Scholl and Kemp, 2016: 90). Its location and surrounding context shapes the ULL and contributes resources, while in turn, the ULL does similar, inevitably intertwining with the real-world urban network (Frantzeskaki et al., 2018: 1046). Nonetheless, Karvonen and Van Heur in Bulkeley et al. (2019: 5) highlight that while an ULL is embedded in a local context it can still produce ‘placeless’ knowledge, such that can be applied to and that could happen anywhere. This assumption is crucial for this thesis to be able to apply findings to other cities.

(12)

7

To break free from path dependency in planning, ULLs should provide space for experimentation. This means that unconventional planning methods should be used and places for experimental activities must be created. In these experimental places, citizens can propose and actively take part in the creation of their district (Cuomo et al., 2020: 3). Approaches and instruments are mixed in a systematic way (Santonen et al., 2017: 8) to encourage sustainable innovations (Frantzeskaki et al., 2018: 1046). The innovations should be designed in a way that allows for long-term scaling (Chronéer et al., 2018: 8–9). Bulkeley et al. (2019: 6) on the other hand argue that ULLs themselves don’t serve for radical change, but rather aim to identify conditions in need of improvement and create capacities to document these. As such, their focus is on the ‘politics of improvement’.

The two aspects of focus in this thesis are the evaluation and learning of actions during the process of experimentation. It is key to evaluate actions and their impacts, this will allow learning of what is successful and consequently allow for the adaption of policies (Cuomo et al., 2020: 4; McCormick in Sharp and Salter, 2017: 2). Key for a successful ULL that reaches goals is the willingness to reconsider the design, goals, and methods; something very untypical in projects with expected outcomes (Nevens et al., 2013: 120). The following sub-chapter will discuss the importance of evaluation and learning in urban experimentation.

5.3. The Importance of Evaluation and Learning

The ultimate aim of experimentation is to find and learn about different solutions to move towards a sustainable society (Berglund-Snodgrass and Mukhtar-Landgren, 2020: 97; Evans et al., 2016: 9). To be able to share outcomes of the experimentation and trigger societal change two things are needed. The evaluation of actions taken and consequently learning from these (Ersoy and van Bueren, 2020: 91; McCormick et al., 2013: 10). Without this, ULLs will lead to local one-time innovations (Steen and van Bueren, 2017: 66). Ideally, both happen in real-time, at all stages of experimentation, leading to a constant revision of goals and future actions (Fuenfschilling et al., 2019: 225). This feedback in learning was first described as a learning-loop by Argyris and Schön (1978), later the concept was expanded to double and triple learning-loops (as cited in Scanagatta, 2020: 73).

Single-loop learning is the adaption of actions to improve the situation and be able to reach one’s goals. There is no reflection on the process or the actions themselves, instead, it is a mere removal of symptoms. Double-loop learning includes a reflection to create a deeper understanding of the assumptions & decision-making to see how goals can be achieved. Both the individual and their organisation actively ‘de-learn’ previous experience in this process. In triple loop learning then, actors reflect upon how they have learned and how they and their organisation have influenced this learning. Both double and triple-loop learning are used to understand how actions separated in time & space create the current situation or outcome (Pettersson et al., 2018: 397; Scanagatta, 2020: 74). This evaluation and learning ideally happens at two scales, the technical and conceptual, to include the functioning of the innovation but also the aims of it (Steen and van Bueren, 2017: 66).

All evaluation activities need to be steered and explicitly included in the ULL process (Steen and van Bueren, 2017: 66). Dijk et al. (2018: 2) highlight that from the start the question ‘how does the program enable learning?’ is of central importance. Whether a framework for evaluation is present and how learning takes place within this (as single, double, or triple-loop) will be of focus in research question two. The findings are then discussed together with the findings for the third research question, to see if and how learning from experimentation affects the spread or repetition of results. Ideally, learning feedbacks not only within the ULL but at

(13)

8

different scales of policy-making, institutional structures, and decision making processes can be uncovered (Cuomo et al., 2020: 4) and how they shape the process of experimentation.

There is a range of vocabulary discussed in the literature on how ULLs can trigger wider change, such as scaling out, scaling deep or amplifying. For this thesis the main concept behind all – sharing knowledge of successes and failures – is of importance, thus the details will be neglected at this point. A detailed review of scaling vocabulary can be found in Lam et al. (2020), Westley et al. (2014), or Wirth et al. (2019). Furthermore, the wording ‘test bed’ will not be used in this thesis as definitions in the literature vary widely. Some use the phrase interchangeably with ULL and Transition Experiment, others see test beds as free-standing concept. Test beds as their own concept focus primarily on the experimentation with a technology, rather than a holistic experimentation. Although they are multi-actor processes in a real-world setting, the primary focus there is put on monitoring and systematic variation of parameters to optimize a technology (Engels et al., 2019: 4). Therefore, test beds do not fall under the delimitation of the study, or when used interchangeably with ULLs, the test bed vocabulary is not needed.

(14)

9

6. Methods

The following chapter describes the methodology followed during the research for this thesis. It highlights choices made throughout the research process and discusses challenges that have arisen. Following what seems to be common practice3_{in the research field on experimentation}

in the urban realm, an internet and literature research was conducted to choose case studies as well as to provide basic information on each of the cases. These were then followed up with expert interviews of several representatives of each city. The aim of the mixed method of case studies and interviews is to provide greater insight into the research questions by collecting complementary data (Yin, 2009: 63). The focus was set on cities in Sweden to make institutional regulations, national governance, and policies similar in all cases. These top-down factors are described by Fuenfschilling et al. (2019: 223) as creating very different, difficult to compare cases.

6.1. Case Study and Secondary Data Research

In the first step, suitable projects and cities had to be identified. According to the main characteristics of experimentation, a web research was conducted to identify possible urban development projects in Sweden. Following findings by Berglund-Snodgrass and Mukhtar-Landgren (2020: 97) the keywords such as innovative or smart were used. The authors highlight, that experimentation sites are usually situated on brown or green fields and use this labelling to attract innovations to be implemented in the newly built areas. As a side-effect, these districts are attractive to residents as innovations and co-creation during the development phase create added value (Juujärvi and Pesso, 2013: 22). The case studies aim to investigate a real-life situation of a contemporary phenomenon – in this case – experimentation in urban planning. It is a common approach for complex phenomena such as neighbourhood change in urban planning (Yin, 2009: 18).

Research on Sweden’s three largest cities, Stockholm, Gothenburg, and Malmö is abundant. Districts such as Hammarby Sjöstad or the Royal Seaport4_{(Stockholm), Hyllie and Västra}

Hamnen5_{(Malmö), and Älvstaden}6_{(Gothenburg) have received significant attention both in}

research articles and thesis projects, concerning the different pillars of sustainability (social, ecological and economical), urban planning strategies and the interplay between the two. Furthermore, they differ significantly in terms of planning demand and capacity (people, funding etc.) making a comparison with smaller cities difficult. To make a contribution to the research, this thesis will focus on mid-sized Swedish cities.

3_{See for example studies conducted by: Berglund-Snodgrass and Mukhtar-Landgren (2020: 99), Castán Broto}

and Bulkeley (2013: 95), Castán Broto et al. (2019), Ersoy and van Bueren (2020: 92), Nesti (2018: 314–315), Smedby and Neij (2013: 149), , Westley et al. (2014: 239–240).

4_{For example: Ehnert et al. ; Green (2006) (2018), Goel (2013), Pan et al. (2020), Pandis Iverot and Brandt}

(2011), Schrenk et al. (2016), Westling (2019), Yin and Feng (2012).

5_{For example: Brange et al. (2016); Castello Branco et al. (2012); Isaksson and Heikkinen (2018); Jankovic}

and Gjurovska (2015); Parks (2018).

6_{For example: Bibri et al. (2020); Brorström (2015, 2017); Hulthén and Ransgård (2012); Pettersson et al.}

(15)

10

The pre-set criteria to choose cases from the next fifteen largest cities in Sweden were 1) a district must currently be in planning or development which 2) is announced to have at least one of the characteristics of urban experimentation.

• experimentative: a district aimed to be innovative, groundbreaking, smart or in other ways different to any other district in the city

• co-producing knowledge: aimed to include more partners or participants than usually during citizen consultation in planning

• learning and scaling: aimed to become a role model, or example to others

Projects focussing on one sector only, for example, the re-structuring of public transport or projects taking place on a city-wide level were not eligible for the thesis, even if they displayed one or several of the characteristics of experimentation, because they do not fall within the framework of ULLs.

The selection based on pre-defined characteristics enables suitable decision grounds without being too resource-intensive (Yin, 2009: 91). Not only the three largest cities were excluded (explained above) but also cities in the greater Stockholm area as these are expected to be largely influenced by Stockholm’s developments. The search was conducted systematically, by searching for keywords such as “experimentation”, “urban planning”, “innovation”, “smart” in combination with the city names in different literature databases. The aim of this was to find previous research on the city and possibly on certain districts or experimentation sites. As this thesis focuses on a research gap – identified for the topic and region of interest – results were sparse. The majority of initial information on development projects was obtained from municipal websites and their comprehensive and detailed plans. These were used because they highlight current and upcoming projects, how they are embedded in the overall city development, at which stage of planning the projects currently are, and any special ambitions or goals. Once a potential project was found, a deeper search was conducted collecting information from project websites or programs by Vinnova7_{or the EU.}

One limitation of this initial research was that the necessary data is only published in Swedish, for this reason, websites were studied with the Chrome Browser, which includes an automatic Google Translate function for all text on websites. For documents such as detailed plans or project reports, individual sections were selected based on promising titles and translated. For example, in the Sustainability Program (Hållbarhetsprogram) of Tomtebo Strand (Umeå) initially only the sections “how we create a sustainable district”, “The Tomtebostrand-Model” and “Tomtebo Strand – an international role model” were translated and read. The reason was, that these titles pointed at the characteristics of interest: to find if the district is more ambitious or innovative than others, a new method of planning and building (the Tomtebo-Model) and the idea to influence others in planning as role model. The same procedure was applied to all other Swedish documents studied in this first case selection step. Both the machine translation as well as the selective translation of parts are potential sources of error, which are acknowledged. However, the argument is made that for a pre-selection of cases this method gives a good enough understanding of the projects to make a decision if they are applicable for the thesis.

7_{Vinnova is the Swedish Innovation Agency. They act under the Swedish Ministry of Enterprise and}

(16)

11

The documents studied initially to find cases, were again studied in detail for the case analysis. They were used to verify, further develop or contrast the information given in the interviews to provide a holistic approach. Of course, the majority of sources is self-reported data, for this reason, it might be biased towards successes. Furthermore, only written information publicly accessible on the internet was used and this was studied with the support of translation tools. In this regard, the approach of Castán Broto and Bulkeley (2013: 96) is similar, they also acknowledge the limitations posed by accessibility, self-reporting, and language.

Table 1 shows findings condensed to a short summary; the sources portrayed in the right column provided a basis for decision. Sources that were read but did not prove to be of interest or create added value for a decision (e.g., because they deepen findings of another source) are not listed, for this reason, the list of sources is not a comprehensive collection of everything read. It is important to note, that case studies do not follow statistical sampling methods, instead, cases are chosen to allow for replication, thus the characteristics of interest have to highlight conditions of a certain outcome (literal replication) or of different outcome (theoretical replication) (Yin, 2009: 54). As such it is argued that Västerås, Jönköping, Umeå, Lund, and Borås are interesting cases to study because the districts currently under development seem most promising and show most characteristics of urban experimentation as described above. Furthermore, they are very similar in size with around 110-150 thousand inhabitants. Växjo, Sundsvall, Eskilstuna, and Helsingborg were excluded, as no district is developed with the characteristics described above. Instead, ambitions are on city-wide level. Both in Norrköping and Halmstad no information on the current district development projects was found which would show that the districts would fall within the selection criteria. Uppsala is, in comparison to the other cities a lot larger and thus excluded to have five cities of similar size.

In the next step, possible interview partners in each of the cities were contacted. At this stage, two additional cases were added to the study in this thesis because no interview partners were found for the project in Västerås and only one for Jönköping. Due to the lacking first-hand information from Västerås the city was discarded and instead the two cities of Örebro and Gävle included. Of the cities remaining from the first selection, these are closest in size to the five chosen before. This flexibility is a key attribute needed by researchers, as cases frequently need to be adapted or changed during the research process (Yin, 2009: 70).

(17)

12 Table 1 Summary of the Internet and Literature Research on Swedish Cities

City

Population (in

thousands)8 Summary of Findings Sources studied

Uppsala 230,767

Östra Sala Backe: Pilot Project in Sustainability with ambition to become most climate adapted area in Uppsala, technologies further developed in Rosendal

Rosendal: focus on crime reduction, increased equality and sustainability

South-eastern Districts: Aim to become model area for sustainable development in Sweden

Uppsala Kommun (n.d., 2019a, 2019b, 2019c, 2019d, 2019e, 2019f), Uppsala Kommun (2021), Viable Cities (2020b)

Linköping 163,051

Vallastaden: Build as part of Sweden’s housing expo, large number of actors included in the design and planning process, to provide smart and flexible solutions. New method of architectural competition.

Ebbepark: designed as test bed for new ideas and technologies. Companies, organisations and scientists collaborate to develop and try new innovations.

Ebbepark (n.d.); Linköping Kommun (n.d., 2018, 2020); Linköping Stad (2020); Smart City Sweden (2020c)

Örebro 155,696 Tamarinden: Best Practice in EU Horizon Multiply Network. District with smart solutions, development of new energy solutions with _{E.On to create an energy autarch district.} IVL Svenska Miljöinstitutet (2020); Örebro Kommun (2019a); _{Örebro Kommun et al. (2019); Örebro Kommun (2019b, 2020)} Västerås 154,049 Sätra: New district including plus-energy housing, low-temperature district heating and solar panels. Modern version of the garden city _{with the aim to make sustainable living easy, and create a safe place with community sense. Part of the EU-Multiply Network.} IVL Svenska Miljöinstitutet (2020); Västerås Stad (n.d., 2020) Helsingborg 147,734

H+: Part of the H22 city expo in 2022, where the whole city is seen as test bed. The RECO LAB will function as test bed for sewage treatment, and Öresundskraft (energy company) will provide an arena for exchange and dialogue. Smart energy grids should optimize energy use and provision.

Helsingborgs stad (2016a, 2016b)

Norrköping 143,171 Västra Staden: create a safe, sustainable and active district. City aim to be among Swedens miljöstader by 2035 Norrköping (n.d.); Samhällsbyggnadskontoret Norrköping (2020) Jönköping 141,081 Södra Munksjön: ‘Urban transformation Area’ in central location with focus on sustainable mobility and energy efficiency embedded in the Cities Planning Vision 2.0 (Stadsbyggnadsvisionen 2.0) and the Program for sustainable development.

The city is part of the European sustainable Cities, Sverige ekokommuner and others.

Jönköpings Kommun (n.d., 2012); Södra Munksjön Utvecklings AB (2020a, 2020b)

Umeå 128,901

Tomtebo Strand: Aimed to become international role model for sustainable development and conscious lifestyle, investigation how the district can become climate positive.

Universitetsstaden: Improvement to be a smart district, developed within the project RUGGEDISED (EU Horizon 2020 program). Umeå is part of Sharing Cities Sweden, a strategic innovation Program of Vinnova, Formas and the Swedish energy authority.

IVL Svenska Miljöinstitutet (2020); Smart City Sweden (2020d); Umeå Kommun (n.d.a, 2019, 2020a, 2020b, 2021b)

Lund 124,935 Brunnshög: smart. Sustainable district where plans will change as the project goes on to stay at the forefront of innovation and _{sustainability. It should be pioneer and role model in Europe.} Lunds Kommun (n.d., 2012, 2020d, 2020e); Viable Cities (2020b)

Borås 113,179 Gässlösa: strategic transformation sire to build a sustainable district. Won the EUROPAN15 competition. Follower of the Norrby _{Innovation Plattform in Borås, a Living Lab project funded by Vinnova.}

Ahlgren (2019); Arroyo (2020); Borås Stad (2019a, 2019c, 2019b); Europan Sweden et al. (2019); Europan Sweden (2019b, 2019c, 2019a); Klasén and Medin (2018); Sinfonia Smart Cities (2016); Vinnova et al. (2016)

Eskilstuna 106,859 Focus on a sustainable development of the entire city. Eskilstuna is part of the EUROTowns network and receives funds as part of the _{Project for energy efficient cities, furthermore in several other city networks.} Eskilstuna Kommun (2019a, 2019b); European Commission _(n.d.a) Halmstad 102,767 Halmstad Train station: Aim is to form a hub of mobility and activity, where people can meet. _{Ranagard: 400 new homes should be developed with focus on sustainable lifestyle and efficient land use.} Halmstads Komun (2020); Langlois et al. (n.d.)

Gävle 102,418

Näringen: aimed to become Europes most sustainable district, where innovation and testing of new solutions will take place. Part of Swedens hållbarer Städer (sustainable cities) contract.

Kungsbäck: Enlargement of the area around campus in an innovative & sustainable manner.

Arnell et al. (2017); Gävle Kommun (2017, 2020); Jonegard (2017)

Sundsvall 99,449

Development is also focussed on the city as a whole, with the aim to create and attractive and exiting place that is an economically, socially and ecologically successful growth engine. The city sees great potential in improvement for healthy living environments and decreasing GHG emissions.

Sundsvalls Kommun (2016); Tyberg (2016)

Växjö 94,129

Växjö is among the cities that signed the Climate contract. Several planning documents aim at changing future infrastructure to follow the social and climatic change. The city aims at being Europes’ greenest and most sustainable city still in 2030. It is also an EU Horizon 2020 best practice city.

Empowering Horizon 2020 (2020); Växjö Kommun (2014, 2019); Viable Cities (2020b)

(18)

13

6.2. Interviews

The interviews conducted were expert interviews, intending to gain increased insight on the development project as well as procedures and methodologies not published online. Generally experts are chosen due to their specific knowledge, status or position within a certain field, which is higher than that of laypeople (Döringer, 2020: 1; Meuser and Nagel, 2009: 467). They are seen as representatives of certain modes of action or functionaries, rather than being interviewed as their own self (Meuser and Nagel, 2009: 469). This means experts are in a position where they have knowledge of certain procedures and at the same time have decisive power over other people’s actions (Bogner et al., 2014: 35; Kruse et al., 2015: 170). For this thesis, two categories of experts were contacted. Firstly, politicians who have a leading position in the building committee or local council (in Swedish: (Samhälls-) ‘Byggnadsnämnd’ and ‘Kommunalråd’), to get insight into the overall planning strategies and how the outcomes of experimentation influence political or planning decision on city level. Secondly, urban planners or architects responsible for the respective projects were contacted, as these are seen as experts on the planning process and interactions of actors in the process. Overall, both types of experts provided information on processes not observable from the outside (Bogner et al., 2014: 19). In total, eleven interviews were conducted, one with two experts together. The role in which the interviewees were interviewed can be seen in Table 2.

Table 2 City, Date and Position of Interviewee. Reference given is used throughout the thesis.

City Interview

Date Position Reference

Umeå 25.02.2021 Politician

Politician Interview 1

Umeå 10.03.2021 Land Development Engineer Interview 2

Örebro 27.04.2021 Politician Interview 3

Örebro 08.04.2021 Project Manager Interview 4

Jönköping 25.03.2021 Politician Interview 5

Borås 12.03.2021 Strategic Urban Planner Interview 6

Borås 19.03.2021 Politician Interview 7

Lund 26.02.2021 Project Manager Interview 8

Lund 15.04.2021 Energy Strategist Interview 9

Gävle 14.04.2021 Politician Interview 10

Gävle 23.04.2021 Development Strategist and Project Manager Interview 11

It is believed that named interviewees do not contribute to answering the research questions posed. Anonymity was ensured to create a freer, more honest conversation without interviewees having to fear that troublesome information about possible shortcomings or other negative outcomes can be directly reflected back to an individual. However, due to the prominent positions of the interviewees and their public appearance as politicians and project managers, it is only a factual anonymisation, not a total anonymisation, as persons can probably be identified out with enough time and resources spent on research (Bogner et al., 2014: 88–93). By anonymizing interviewees, it is also expected to reduce the “social desirability effect”, where interviewees give an answer that is socially more acceptable or desired, possibly neglecting parts of answers, rather than telling the full truth (Diekmann, 2020: 448). This was also done by not asking judgemental questions. From the critical answers provided by the interviewees, it can be seen that answers were provided truthfully and honestly. An example is the statement below concerning the participation of citizens:

(19)

14

‘To be honest, we don't. I think we are all the time developing new ways of working with citizens, but that's not something we have been very good at.’ (Interview 6)

There are different types of knowledge that can be collected in expert interviews. Of focus in this thesis was interpretation knowledge, the main type asked for in theory generating interviews. The interviews are thematically structured, but less strict so that they can be adapted to the individual interviewee and situation. Information gathered contains ideas, interpretations, and conclusions of the expert and is thus subjective. However, this subjective viewpoint can be shared by the organisation the expert is part of (Bogner et al., 2014: 18–19). Although it is assumed that each expert provides correct information, content is validated by interviewing two to three persons on each case/district (Bogner et al., 2014: 11; Kruse et al., 2015: 177).

The interview guideline was designed using the structure proposed by Kruse et al. (2015: 227), where initially all types of questions were collected and then converted to a interview guide using the SPSS method (German for: collect (sammeln), check (prüfen), sort (sortieren), subsume (subsumieren); similarly proposed by Bogner et al. (2014: 29–30)). Hereby all questions collected are checked for suitability and relevance before being sorted by themes and subsumed into thematic ‘chapters’ of the interview. Each main question in the guide is then accompanied by follow-up and in-depth questions. For this thesis, the guide followed the three research questions. Interviewees were asked to describe the district and (possible) other innovative planning projects, continuing with an explanation of different actors involved and the type of involvement. Lastly, focus was on evaluation frameworks and the learning and communicating aspect. For example, how cities aim to make their district a role model. The Interview Guideline can be found in Appendix 1. The questions asked were generally similar for all interviews, but depending on the answers additional follow-up questions were added. Furthermore, two new questions concerning direct city-to-city collaboration and thesuccess of citizen participation were added to the guide after interesting insights were presented by the interviewees of the first interviews. Generally, the questions were not shared with the interviewees in advance, to receive spontaneous answers that reflect more truly everyday practice, rather than receiving prepared answers (Bogner et al., 2014: 30). Only in two cases, exceptions were made, where interviewees specifically asked for it, to be able to prepare key vocabulary in English needed to answer.

All possible interviewees were contacted in Swedish and English via e-mail. In case no answer was received a second e-mail was sent out after one week, while in parallel new possible interviewees were contacted. The main obstacle in finding interviewees was the language, as all interviews were conducted in English. Other obstacles included: finding a suitable interview time, first contact via secretaries, and convincing possible interview partners that their knowledge or contribution is valuable. These challenges have also been noted by Bogner et al. (2014: 37–38). On top, the COVID-19 pandemic and consequently home-office work for most city workers caused slower work processes and other hurdles (e.g., children supervision in homeschooling, suitable internet connection) which caused some people to decline participation. On the other hand, some postponed the interview to late evening (19:00 o’clock) or during lunch break to be able to provide their knowledge and support for this thesis. Participation in digital interviews outside official work hours was also noted as an advantage by Deakin and Wakefield (2014: 604). In the special case of Västerås, direct contact with possible interviewees by mail was not possible, as the city website only provides contact forms. Through these, as well as through the general email ‘kontaktcenter@vasteras.se’ no answer was received. Phone numbers from the website have proven to either be inactive or the corresponding people were on holidays or otherwise absent. For this reason, Västerås and their project Sätra was taken out of this thesis.

(20)

15

The interviews took between 35 and 45 minutes and were fully conducted online using the video-conference platform Microsoft Office Teams. While this type of interview was still heralded as the innovative frontier in 2010 (Deakin and Wakefield, 2014: 604), the last year of the COVID-19 pandemic made it a prominent day-to-day tool. The digital interviews enabled a wide range of cases and interview dates, which would not have been possible with in-presence interviews. Generally, a range of advantages have been realized in recent years, but even more with the COVID-19 pandemic. Video conferences allow an almost face-to-face conversation because facial expressions and gesticulations are seen (unlike in telephone interviews), travel time and budget are saved, a larger sample of interviewees (in number and geographic range) and digital recording is possible (Misoch, 2015: 180–182). However, there are also disadvantages in digital interviewing, including technical prerequisites (e.g. device with camera and microphone, stable internet connection), technical problems, or a lower degree of obligation (Misoch, 2015: 182–183). Advantageous for this thesis was, that at the point of conducting interviews (February to April 2020), digital working and meetings through videoconferences had become a daily routine. While scholars a few years ago still highlight hesitance, discomfort, and a bias towards technical affinity of interviewees in recorded video calls (Deakin and Wakefield, 2014: 605), this was not experienced during the interviews for this thesis. Nonetheless, technology was still an issue. Some interviewees had connection problems at home and many only gave oral consent to the interview because printing and scanning devices were not available to give written consent.

During transcription, any repetitions of partial sentences (e.g. “this project is.. is very.. the project is very different in terms of...”) were taken out, as the focus is not on communication style and semantics, but on the information given (e.g. “this project is very different in terms of..”). The loss of non-verbal communication (such as mimics) was seen as irrelevant for the analysis, only in cases where non-verbal communication changes the meaning, for example in irony or sarcasm, this was noted in the transcripts. It was recognized, that in many interviews the same grammatical errors occurred, in almost all cases linked to the difference in sentence structure of Swedish and English or to the conjugation of verbs where the third person singular form of verbs was conjugated incorrectly (e.g. instead of ‘it has’, ‘it have’). For easier readability, the verb conjugations were corrected in the transcripts, but sentence structure was kept as to not alter the data too much from the original. It is acknowledged that already the transcript is a subjective construction of the interview, influenced by the researchers’ expectations and hopes towards answering the research questions (Kruse et al., 2015: 347). The challenges that arise with foreign language interviews, their transcription, and possible losses in quality due to lacking information have only recently entered the methodology literature and there is no unified way of working proposed yet (Bogner et al., 2014: 42–44).

The interviews were coded using the Program Nvivo, to find patterns across all interviews more easily. In a first step, the interviews were coded separately, similar or same codes in different interviews were then compared. During this initial coding step, codes were chosen in close relation to the text, no theories or underlying structures were abstracted yet. Where applicable sentences or phrases were coded with several codes. The quote below demonstrates this process. The first part focuses on a national network, followed by the mention of EU projects, and lastly a statement about how learning can take place. The partial sentences were coded by three different codes.

(21)

16

“[…] We have RISE that is the national organization for scientific studies and so on. So we have a big network and we are also quite good to get projects from the EU and have a lot of projects with other countries that way, and that's a very good way to learn if you try to learn from someone else, you have to think about what you do first, but I think we are quite good to have contacts.” (Interview 1)

In the next step, passages with similar codes were grouped to find overarching themes in all interviews. Same or similar codes were grouped. To prevent a too fast generalization and consequently loss of valuable insights, questions for self-monitoring were included at this stage (Bogner et al., 2014: 79). Focussing for example on differences between interviews: Where do the interviews differ? How are opinions contrasted? What is mentioned in one case but not in another? The coding process is always shaped by the expectations and theories that a researcher has and thus never objective (Bogner et al., 2014: 82–83). Self-monitoring, therefore, is a way to keep open-minded for alternative outcomes. In the last step, the interviews were studied in relation to the theories and research questions to generalize knowledge from the cases presented in this thesis (following Meuser and Nagel (2009: 476) and Bogner et al. (2014: 78–80)). The individual elements of the interviews were abstracted to find “the whole picture”. The outcome of this is presented in the results and discussed together with previous literature to show how findings are embedded in or differ from existing knowledge in the field.

6.3. Validity and Reliability

This thesis is based on a qualitative approach. The case study methodology and interviews are producing findings from real-world settings (Golafshani, 2015: 6). Methods used in qualitative research depend upon description and interpretation by the researcher (Gillham, 2005: 10). As such, it is of utter importance for qualitative researchers to reflect upon the working process and to keep an open mind during the analysis (Gillham, 2005: 13). By describing the methodology as well as shortcomings and challenges (Yin, 2009: 40), good reliability of the research should be ensured in this thesis. Furthermore different types of data were combined in a triangulating fashion to ensure the reliability of the results (Gillham, 2005: 13). Self-reported information such as reports and websites were combined with data from sources other than the municipalities and interviews. The aim was to reduce bias and to create a multifaceted analysis. External validity of the data refers to the generalization from the cases studied in this thesis to other cases. The aim of this research to find theories that allow for analytical generalization regarding learning from experiments in the urban realm (Yin, 2009: 43). Following the replication logic, the cases chosen in this thesis were initially selected due to the number of inhabitants (mid-sized Swedish municipality) as well as due to characteristics of a district development project currently taking place. The selection was purposefully structured and not random, only like this can generalization take place (Flyvbjerg, 2006: 229). The fact that the findings from all cases are similar shows, that generalization is possible. Hence the findings from this thesis can well be applied to similar municipalities. Of course, it has to be taken into account, that all cases studied here were located in the same country and therefore influenced by similar regulations. As described above municipalities in Sweden have the major planning task, in countries where municipal planning is more regulated, results might differ.

(22)

17

7. Results and Discussion

This section will present and discuss the results in accordance with the research questions posed above. Namely, the aims of the experimentation sites, the evaluation of experimental approaches, and how learning takes place. The last section of this chapter will discuss how learnings are communicated to contribute to sustainable development.

Except for Umeå, the municipalities studied answered during the interviews that currently there is only one district this ambitious and with the characteristics of experimentation. This is in line with the previous research conducted. Individual aspects, such as new approaches to land allocation or technological innovations have been applied in other areas, but not in a holistic ULL approach. These cases will be explained in detail in section 7.3 on the spreading of knowledge. Following the preliminary research for cases as presented in Table 1 the districts Gässlösa (Borås), Näringen (Gävle), Södra Munksjön (Jönköping), Tomtebo Stand (Umeå), Brunnshög (Lund) and Tamarinden (Örebro) have been chosen as case studies. Each project starts with the development of a vision, which then is developed into more concrete plans until the detailed planning is completed and construction starts. As all projects are at different stages, their contribution to the thesis is different. This will be realized especially in chapters 7.2 and 7.3. It is acknowledged, that projects currently in early phases of development had fewer opportunities for evaluation, learning, and consequently also of spreading their learnings.

7.1. District Characteristics and Aims of the Experimentation

Before the aspects of evaluation, learning, and spreading of knowledge for sustainability can be discussed, the characteristics and aims of the experimentation must be discussed. Highlighting similarities and differences in the project processes and aims will also aid in showing how experimentation in mid-size Swedish cities occurs, allowing for generalization to other cases.

7.1.1. Characteristics of the Districts Identified as Experimentation Site

The current standpoint of the projects as well as the overarching timeline will shortly be highlighted here. Further details of the districts’ planning process and aims will then be taken up and discussed in the following paragraphs. The districts are at different stages of development. The city of Borås has only recently begun with the development process of Gässlösa after the movement of the water treatment facility was completed. Once the study on soil and water pollution is completed, a new school will be the first building of the new district, followed by residential housing in 2025 (Borås Stad, n.d., 2021). The development of Näringen is one step further, a pilot study in the area has been completed, an innovation process and detailed planning will follow until 2023 with the aim of finishing construction by 2040 (Arnell et al., 2017: 10). Södra Munksjön in Jönköping covers a quite large area. For this reason, the area has been divided into six sections, in the first one construction started in 2016, the last one should be completed by 2050. Currently Munksjökajen and Västra Skeppsbron are being built (Södra Munksjön Utvecklings AB, 2020b). In the district studied in Umeå, Tomtebo Strand, construction of municipal infrastructure will start next year. So far the sustainability program and the detailed plans have been developed (Umeå Kommun, n.d.a, 2021a). In Brunnshög in Lund construction is well advanced. The detailed plan was adopted in 2013, since then both research facilities in the area, as well as the tramline and first section (Southern Brunnshög), have been completed. By 2030 central Brunnshög should be completed (Lunds Kommun, 2020f). Lastly, for the district of Tamarinden in Örebro, the land allocation has recently been completed and construction will start in 2022. Completion of the area is aimed at 2025 to 2030 (Örebro Kommun, 2020).

(23)

18

Figure 1 Map of Sweden showing the location of cities studied and Sweden’s three largest cities for locational reference (own represenatation with base map from GADM, 2018)

There is a clear pattern in the characteristics of the districts and the initiation of experimentation in contrast to a traditional district (summarized in Table 3). Half of the cities describe the area a significantly larger than anything developed in the recent past, which creates the opportunity to do something significantly different, using a new and holistic approach rather than filling single building gaps (Interview 1; Interview 10; Interview 11; Interview 5; Interview 6; Interview 7). Furthermore, the location of the site within the existing city structure seems to be a key attribute. Except one, all of the studies sites are close to the inner city and four describe the location as strategically placed close to an existing recreational site, like a forest, river, or lake, creating higher living standards (Interview 1; Interview 10; Interview 3; Interview 5; Interview 6; Interview 7; Örebro Kommun, 2020; Umeå Kommun, 2020b: 7, 2021a). Furthermore, as suggested by Berglund-Snodgrass and Mukhtar-Landgren (2020: 97) the experimentation sites are in three of the six cases located on former industrial sites, the other three on green fields. Lastly, the decision to design a sustainable, innovative, more-than-the-usual district were in all cases political decisions, with the planners being the executive, working

(24)

19

to fulfill the planning visions. This is in contrast to findings by Sánchez Gassen et al. (2019: 43) who conclude that when working with Agenda 2030 in the Nordic context, the largest challenge was integrating these in the municipalities’ goals and to convince officials of the importance of sustainability. Figure 1 shows the location of the six cities studied, they are widely spread throughout the country. Even though their location is as diverse as possible in Sweden, there are many similarities as described above, suggesting that the district’s characteristics in terms of size and location are important factors or initiators of experimentation and thus key to show where future ULL type experimentation could take place.

Not only the districts themselves but also the planning processes are different from traditional planning processes. In Lund, a new project group was founded, joining competencies from very different fields (Interview 9). Furthermore, the municipality actively works with a sustainability agreement with the developers, that should help to keep sustainability ambitions alive (Lunds Kommun, 2012: 5). Södra Munksjön is developed by a municipality owned company – Södra Munksjön AB. They have the responsibility and are driving force behind the project (Södra Munksjön Utvecklings AB, n.d.b). In Umeå and Gävle new methods of collaboration have been started for the development of Tomtebo Strand and Näringen and lastly, in Tamarinden the process of land allocation differs from the usual processes (Interview 11; Interview 4; Umeå Kommun, n.d.d).

Other than the environmental aims described, the municipalities need to create an attractive municipality for the rising urban population (Borås Stad, n.d; Gävle Kommun, 2017: 3; Umeå Kommun, 2020c). The municipalities are experiencing a population rise of 1000-2000 citizens annually, this leads to 150-200 thousand inhabitants, which can be seen in Table 3. Thus, a major reason for the development of such large areas is the possibility they bring in terms of housing and job opportunities. By aspiring for more sustainable solutions, cities can extend their reach and put their names on the global map (Haupt et al., 2020: 144). Urban areas with good employment options and high quality of life attract more people and skilled workforce (Affolderbach and Schulz, 2017: 678; European Comission, 2010: 9), providing another reason for a district with high sustainability aims. For mid-sized Swedish cities, not only representation outside of the country, but especially competition with cities of similar size and the large hubs Stockholm, Gothenburg, and Malmö is crucial (Gävle Kommun, 2017: 3; Jönköpings Kommun, 2008: 6). Their location in Sweden (seen in Figure 1 above) greatly shapes whom they are influenced by. Cities such as Gävle are already today shaped by commuters to Stockholm, but to truly integrate into the greater Stockholm region, the municipality must offer attractive jobs, housing and education (Gävle Kommun, 2017: 7). In contrast, Umeå faces another challenge being “maybe 600km North of where the end of Sweden is for many” (Interview 1). The districts, therefore, aim not only at contributing their share to sustainable development, but also to attract visitors, skilled workforce, and industry (Jönköpings Kommun, 2008: 6; Lunds Kommun, 2020g; Umeå Kommun, 2021a). During the past decades, a discussion has arisen, whether cities are using the sustainability discourse merely as advertisement to position themselves as climate change leaders, while indeed new plans shy away from radical change (Affolderbach and Schulz, 2017: 676–677). However, for the districts where planning has already moved beyond the envisioning phase, it is clear that the districts do fulfill the ambitious goals. Nonetheless, the districts also serve as vehicles for place-making to create an urban environment by and for locals (Smart City Sweden, 2019; Wirth et al., 2019: 245). Furthermore, it cannot be neglected that indeed the high aims towards sustainability are a method to secure funding for sustainable development (Voytenko et al., 2016: 3).

City Districts as Sites of Experimentation and Learning for Sustainable Development : An Analysis of mid-sized Swedish Cities