Who farms the city? : a case study of urban agriculture governance in Amsterdam, the Netherlands

(1)

Who farms the city?

A case study of Urban Agriculture Governance in

Amsterdam, the Netherlands

Master thesis Political Science: Specialization track

Public Policy and Governance

Ben van der Erve (12347833) benvandererve@gmail.com Supervised by: Dr. R. Pistorius Second reader: Dr. J. Doomernik

This thesis brings together much of what I learned over the past years. It is the result of inspiration and support from my personal surroundings, for which I would like to express my gratitude. My special interest in social and political matters have been strengthened by regular conversations with my close friends and in particular my girlfriend. My parents have introduced me to inspiring cultural and political activities at a young age. They have always been supportive in my student career, for which I would like to thank them. I am very grateful to be surrounded with such supportive, stimulant and bright people.

My academic development would not have been possible without help from the study coaches, tutors and teachers. Especially the excellent accompaniment from Dr. Robin Pistorius during the writing of my thesis was exceptionally pleasant and has been of much help. Also, I would like to thank second reader Dr. J. Doomernik for proofreading my thesis.

My personal fascination for in urban environments, city gardens and local policy are instigated through living in the Randstad, and many years of observing local innovations within these cities. It is a pleasure to live in this interesting and stimulating environment. The documentary ‘Boer zoekt voedselflat’ by VPRO Tegenlicht (2017) was my primary inspiration for this piece of work. The potential of urban agriculture was excellently depicted and quickly led to my choice to devote writing my thesis to this subject.

(4)

Cities worldwide are growing quickly. In order to tackle threats of climate change and ecosystem depletion, they must change to environmentally and socially sustainable systems. Urban agriculture is an example of a civil society-born response that contributes to this transition, as it enhances social integration and ensures environmentally friendly food production. Detailed academic insight is required with regard to the implementation and the institutional arrangements that are necessary to enable social-ecological innovations such as urban agriculture. This study presents a case study of adaptive governance in the urban agricultural network of Amsterdam. The dynamics between local legislation and urban agriculture-related organizations is examined to gain deeper understanding of the role of adaptive governance in the development of these projects. Empirical evidence derived from interviews, field visits and a policy review and it reveals a substantial level of self-governance among projects, but a lack of fit between local legislation and its targets. The conditions for adaptive governance to emerge are partially created, but they are not consistently executed for it to ensure enhancement of the urban agricultural network in the future.

(5)

1.1 Urbanization

The world is undergoing a period of rapid urbanisation, with most of the future population growth expected to take place in cities (UN, 2014). As Deelstra and Girardet (2000) state it dramatically:

“At the end of the 20th_{century, humanity is involved in an unprecedented}

experiment: we are turning ourselves into an urban species.”

The global trend of urbanisation comes at a time when cities face many other challenges. Ecosystem depletion, exploitation of natural resources and climate change are among the threats that are affecting communities worldwide. Densely populated areas such as cities are inherently more vulnerable to the effects of climate change such as floods and droughts. This underlines the need for cities to transform towards more environmentally and socially sustainable systems. At the same time, cities are gaining more power in national affairs and are more capable to autonomously govern these issues. The need for evincive sustainable urban governance is therefore of utmost urgency.

1.2 Urban Agriculture and governance

The recent emergence of urban agriculture (UA) is often praised for its potential to tackle these overarching goals of environmental and social sustainability (Crush, Hovorka & Tevera, 2011). A growing number of municipalities and governments worldwide are attempting to implement UA as an adaptive response in the repurposing of vacant land (Pothukuchi & Kaufman, 1999). In post-industrial settings for example, urban space is being repurposed to provide sustainable food production. Cities often provide an array of green spaces such as gardens, parks, vacant lots, rooftops and brownfields (Kattwinkel et al. 2011). These urban green spaces are part of large social-ecological systems (Odom Green et. al., 2016) and can provide all sorts of ecosystem services and social services that can be of high importance to the resilience of urban areas.

Implementing UA is a relatively new policy mission and rightfully, questions are posed in what way it should be governed in order to stimulate further growth (Seyfang, 2006). Governance interventions are often too homogeneous and not well specified to local

(6)

(Prove, 2016), and traditional government practices tend to not be competent in stimulating context dependent systems. Therefore, scholars call for more flexible types of governance, based on independence and self-organizing capacity of local communities (Ostrom & Cox, 2007). There is an academic consensus on the need for adaptive, and less rigid governance arrangements (Odom Green et. al., 2016; Ostrom & Cox, 2010). Traditional governance models are deemed ineffective in such increasingly complex and interconnected issues (Correa de Faria et al., 2013). Nevertheless, flexible arrangements prove rather difficult to apply in existing governance patterns and political contexts. Additionally, there is a lack of academic insight in the deliberate implementation of UA and its perceived effects on local communities (Lohrberg et. al., 2016; Odom green et. al. 2016).

1.3 Research question

This study aims to contribute to this literature gap by exploring the implementation of UA and the governance practices that are in place. Since adaptive governance arrangements are more suitable to support UA, the presence of these arrangements and its perceived effects on the UA will be examined. This examination is executed through a qualitative case study of the urban agricultural network in the Amsterdam area, the Netherlands. The Amsterdam municipality is positively oriented towards new UA-projects is occupied with housing and supporting a growing number of initiatives (Gemeente Amsterdam, 2015). The analysis will take on a social-ecological perspective, focusing on the relation between communities and the ecosystems they interact with. All is brought together in the following research question and accessory sub questions:

What is the role of adaptive governance in the recent development of urban agriculture in the Amsterdam area?

(7)

1) What are the properties of adaptive governance in relation to social-ecological systems?

2) To what extent are conditions for adaptive governance addressed through local legislation?

3) To what extent can adaptive institutional arrangements be identified within the UA network?

This paper commences with an introduction to UA, including a contextualisation in contemporary literature. The following section aims provide a theoretical connection between UA governance and the adaptive management of urban social-ecological systems. Next, the properties and conditions for adaptive management are elaborated operationalized into an integrative assessment scheme. The chapters that follow cover the research methods, succeeded by an overview of all the findings. Results are contextualized along the lines of the theoretical guidelines in order to answer the research question.

(8)

2.1 Defining urban agriculture

One of the key elements in interpreting UA is the distinction between urban and regular farming practices. The demarcation can be set between ‘land’ and ‘urban’ environment, or city and countryside. The FAO (2007) further elaborates this distinction by proposing the concept ‘peri-urban agriculture’, which covers the areas directly surrounding the city as well. Another boundary in the definition is for the production of these farms to be predominantly aimed at local consumption. Only UA production for local consumption is covered in this study. Initially adopted by scholars, the media and governmental organizations such as the UN (2011) are now recognizing the notion of urban agriculture (UA) too. Projects can vary between small-scale (community) gardens, urban allotment gardens, (indoor) vertical farms and rooftop gardens (Moustier & Laurent (2010). Small, micro and large commercial scales are all equally recognized.

2.2 Contextualizing urban agriculture

Over the past decades, a growing global demand for a broad variety of food products has forced agricultural sectors to modernise and shift to industrial scales of production. Food production, trade and distribution are increasingly globalised and cities draw upon resources from far distances (Deelstra & Girardet, 2000). Urban environments are thus highly dependent on the international food market. Urban Agriculture has come up as an antidote to this development. In the global north there is an ongoing trend of urban space repurposing, often with UA as one of the new innovations to reuse these spaces. Vacant lots, rooftops, empty buildings and parks are converted to community gardens, private farms and vegetable gardens. Although UA has been a part of urban communities for centuries, the past decades have shown a sharp rise of projects globally (Pearson et al., 2010; Masi et. al., 2014). Agriculture integrated within urban fabric often makes a significant contribution to cities’ food self-sufficiency and social integration. Multiple studies have shown that UA can have notable influence on local community interaction through integrating food production and local consumption. Additionally, local food production has an effect on reducing transport movements, thereby reducing the ecological footprints of the food supply chain (Hovorka & Tevera, 2011). Urban greening projects also offer multiple social-ecological and educational benefits, which will be elaborated later in this study. Apart from appropriate governance, the success of UA

(9)

depends on several circumstances such as soil structure, nutrients and steady water supply (Zhang et. al., 2007). Mcklintock et. al., (2018) point at the political and social-economic dimension of UA in modern urban communities. They argue it is part of a trend of gentrification, merely favouring people of higher social-economic class. This critical note is important to keep in mind considering the social integrative potential of UA, as it theoretically should favour members of urban communities equally. The focal point of this thesis is on the relation between adaptive management and the UA system. As mentioned in the introduction, both policy makers and academics are searching for the optimal governance strategies for UA to further develop. Adaptive governance favours civil innovations such as urban farms, but is proven to be difficult to blend with rigid management practices. Case studies of Urban Agriculture in Warsaw (Poland) and Ghent (Belgium) show that the implementation of local UA networks strongly differs and is heavily context dependent (Prové, 2016). Often there is a lack of fit between policy programmes and their targets, but it has not yet lead to a drastic change in policy approaches. Understandably there must be more practical insight in the dynamics between local policy and UA systems. The next chapter provides a literature review of adaptive governance and a new theoretical approach that can add to existing literature on UA governance.

(10)

3. Adaptive Governance of Social-Ecological

Systems

3.1 Social-ecological systems

In the face of climate change, environmental degradation and urban social challenges, the academic world is increasingly occupied with finding new urban governance strategies (Kern & Alber, 2008). Our urban societies keep growing and as a result of the capitalist economic system, modern society is leaving a growing ecological footprint. Ideally, human interaction with ecological resource systems must be rearranged in a way that retains ecological systems. In order to create these new restorative institutions, society must radically integrate new and more sustainable practices, norms and rules (Brunkhorst, 2012). Sustainable urban governance can be connected to the intellectual domain of social-ecological systems (SES). The term ‘social-ecological’ refers to the delineation between social and ecological systems, which would be ‘artificial and arbitrary’ according to Berkes and Folke (2004). Underlining only the social dimension of resource management without proper notion of the ecosystem dynamics would not be competent to address the modern challenges of sustainable resource governance. A high focus on the socially desirable and economically sensible can lead to overharvesting of stocks, pollution etc. For example, fishing stocks are often depleted as a result of overharvesting and there is not much attention for restocking and regeneration of the ecosystem for stable future provision. This phenomenon of individual interest standing in the way of common benefit is also referred to as the ‘tragedy of the commons’ (Ostrom, 1999). To conclude from the SES perspective, urban sustainable development is dependent on a system of resource governance that mediates the relationship between society and the ecosystem it is part of (Brunkhorst, 2002).

Social-ecological systems can encompass different types of landscapes, natural areas, human living places and other land uses. Urban organic food systems and urban green spaces are part of such systems and yield a mixture of social and ecological services to the community (Walker et al. 2002, Lebel 2004). Hence, urban gardens can be regarded as part of ecological systems (Barthel et al., 2010) and part of (larger) coupled social-ecological systems (Hinkel et. Al., 2015; Fernandez Andres, 2017; Odom Green et. al.,

(11)

2016). Although SES literature and its adjacent analytical tools have yet seen little application on cases of UA, it has been used to interpret multiple cases of agro-sectors such as fishery and forestry (Lescourret et al., 2015, McGinnis and Ostrom 2014). This new perspective on UA could help to get more theoretical insight in the governance practices aimed to support UA. The SES perspective has been connected to forms of ecosystem governance and adaptive governance of urban green spaces (Barthel et al., 2010; Odom Green et. al., 2015) and the next paragraph will elaborate on this adaptive governance strategy.

3.2 Adaptive governance of social-ecological systems

Scholars argue that the past decades urban governance is shifting to from the traditional government-centred model to more networked arrangements. Governments and other institutions increasingly seem to operate in an ‘institutional void’ (Hajer, 2003), in which governing entities and municipalities are ‘retracting’ and thereby offering space for citizens, institutions and businesses outside of the traditional governmental realm to shape policy outcomes. Although there are critical side notes to the benefits of this development for society (Jessop, 2002), there is large consensus on the positive effect of a facilitating governance structure for innovations from civil society, especially in the context of urban environmental governance. It is becoming clear that communities have complex motivational structures and a ‘blueprint’ approaches to resource governance wont trigger the desired developments. Korten (1980) identified the danger of blueprint approaches to the governance of tough social–ecological problems and urged that policy makers adopt a learning process rather than imposing final solutions. Ostrom & Cox (2007) describe this as the ‘panacea problem’, in which a government may ‘fail by homogenizing the diversity of contexts to which it applies its policies and management practices’. Municipal agencies risk becoming trapped in rigid, maladaptive systems that are unable to provide sustainable responses to effectively retain ecological resources and give space to initiatives that rise from civil society (Gunderson & Holling, 2002). Tight control of systems has even regularly lead to and increase of challenges for local communities (Holling and Meffe 1996). Unfortunately, the preference for simple solutions to complex governance problems continues to be strong (Ostrom, 2007).

(12)

By reducing top-down meddling and regulation, responsibility is supposedly shifted towards the innovative and participative capacity of civil society. This is also referred to as ‘adaptive governance’. Although adaptive governance is widely promoted as the main basis for urban sustainable development, it has received some criticism. Existing governance structures seem to stand in the way of successful adaptive management, since it is required in all phases of the adaptive cycle (Walker et. al., 2004; Walters, 1997). Political processes that are ill suited to swiftly respond to social change can add to the weakness of adaptive governance (Healy, 2004). Besides, the vagueness and complexity of the concept makes it difficult to conceptualize and measure. Therefore specific indicators have been research. The next paragraph elaborates on these factors.

3.3 What conditions enable adaptive governance and how is it observable?

The potential benefits of adaptive governance to social-ecological systems (and thereby also UA systems) have been elaborated. In order to answer the research question, we must further dive into the factors that characterise adaptive governance and the conditions that enable for it to emerge within civil society. First of all, the conditions for adaptive governance to emerge within civil society are deemed ‘good parameters for assessing governmental actions’ (Folke et. al, 2015). The authors appoint the following conditions to contribute to the emergence of adaptive arrangements: enabling legislation, flexible institutions/organizational learning, and the recognition of bridging initiatives. Enabling legislation involves beneficial policy measures such as subsidies and permits. Learning is referred to by Carpenter & Gunderson (2000) as ‘the adjustment and monitoring of management actions, adapting to changing circumstances. Finally, bridging organizations could for example be NGO’s or community-led business organizations and can provide opportunities by spreading knowledge other incentives for resource management. The next section elaborates the different institutional arrangements that could emerge as a result of these conditions.

Next, the attributes of adaptive arrangements are covered. In the context of social-ecological systems, adaptive governance generally involves polycentric institutional arrangements (Ostrom, 2009). These are built up of multiple decision-making centres across organizational levels and domains, which ideally balance centralized control. In the conceptualization of polycentric governance, Ostrom (1990) envisions locally managed

(13)

resource systems by the agents that use them through ‘inter-organisational arrangements’. Ostrom argues that local communities are often best equipped to deal with issues of local resource distribution, since they have been used to their geography, climate, culture and local resources (Ostrom, 2010). This is also being referred to as ‘institutional fit’. Empowering local participants to get involved rather than forcing them into certain practices should enable the formation of polycentric governance, by opening up opportunities for bottom-up local organizational arrangements. This can be achieved through the conditional factors, as mentioned in the previous paragraph. Polycentric institutions have proven to be effective in creating opportunities and ‘servicing needs in spatially heterogeneous contexts’ (McGinnis 1999). Practices of polycentric governance would supposedly strengthen resilience of the local food system, which implies the ability of a system to undergo change while remaining stable, possible through adaptive capacities (Carpenter et al. 2001, Holling 2001, Walker et al. 2002). This strengthens resilience and adaptive capacity, which should in turn increase the efficiency of local-resource use. Thus, local involvement should allow the possibility for independent governance interventions without obstructing the organizing capacity of the network itself. In this way local communities can develop to better match the multiple social and ecological contexts and dynamics at different locations. Academic criticism on polycentrism is based on the assumption that there is insufficient overlap of coordination and administrative responsibilities (Lebel et. al. 2006). Involvement of more than one administrative level is seen as obstructive to policy implementation in some cases, because each level of administration increases the probability of wrong interpretation of the original policy programme and opposing interests. Furthermore, the existence of multiple autonomous decision-making centres characterizes a polycentric governance system, yet it does not guarantee that there will be sufficient coordination between the governance centres for it to be a proper functioning polycentric governance system (Pahl, Wostl & Knieper, 2014).

The self-organizing process of polycentric decision-making centres can emerge in ‘systems of adaptive co-management’, which are flexible communities occupied with local resource management, linked to specific geographical areas (Folke et. al., 2015). Institutional arrangements and ecological knowledge are tested and revised in processes of learning by doing. Therefore it incorporates the learning characteristic of adaptive

(14)

management. They have adopted the flexible structure that allows the to respond to changing circumstances (Folke et al., 2002). To ensure the local community to be active participants in ecosystem management, governments should transfer power to local authorities and other local decision-makers. It could offer local users the “independence to make and enforce rules within adaptive co-management for a specified geographical area” (Ostrom 1998). Monitoring for social and ecological factors informs for adaptation, and offers other actors to contribute during the adaptive process (Robards et al., 2011). The conditions for adaptive governance and its attributes have now been set out. The following paragraph proposes integration with another theoretical framework.

3.4 Integrating the Social-Ecological Systems Framework

All previous paragraphs have elaborated on the contributing factors and attributes of adaptive governance. When studying the governance of UA, a wider diagnostic approach could provide a meaningful addition to the analysis of adaptive governance. Therefore a careful attempt towards the integration with another approach on social-ecological systems is made. The purpose of this integration is to investigate the utility of adding a variable to potentially enhance the insight in the functioning of adaptive governance in the context of UA development. Ostrom and Cox (2010) illustrate the need for additional analytical frameworks to diagnose social-ecological systems and the management actions they are built on in a detailed and structured manner. As Ostrom (2009) states, ‘A framework is thus useful in providing a common set of potentially relevant variables and their subcomponents to use in the design of data collection instruments, the conduct of fieldwork, and the analysis of findings about the sustainability of complex SES’s’. She argues that a much wider array of factors lie at the basis of describing sustainable local natural resource management. In this light, a model of interlinked biological, social and institutional elements is presented: the Social-Ecological Systems Framework (SESF). The SESF is presented as a ‘common language’ for scholars assessing sustainability of resource systems in the context of maladaptive management (Ostrom, 2009). It is based on the same notion of an SES, in which community output is produced by social and ecological factors combined (Berkes et al., 2000). The SESF seeks to provide a basis for institutional crafting in the context of social-ecological systems. Taking a local community and its interactions with the environment as a starting point, the SESF offers bottom up, community focused perspective. Over the past years there have been numerous

(15)

applications of the SESF in academic studies of local resource management. Research on the usefulness has shown that it not yet developed enough to provide exclusive answers on all resource governance policy-related issues, but it may well be suited as a diagnostic tool to inform policy-making decisions. The SESF scheme is depicted in figure 1.

Fig. 1: The social-ecological systems framework (Ostrom, 2009)

Where adaptive governance literature revolves around actors actions and interactions the SESF has a broader scope, focusing on governance of common resources, covering the governance network, political context, historical background, culture, and geographic characteristics of the whole SES. The SESF is a therefore system-wide approach and can be very useful for diagnosing urban and rural ecosystem management at multiple scales (Ostrom, 2010). It helps identify factors that may affect the likelihood of particular

(16)

policies enhancing sustainability in one type and size of resource system and not in others.’

3.5 Synthesis

One variable from the SESF will be added to the assessment. The variable will be integrated with the conditions that enable adaptive governance and self-organization (Folke et. al., 2015), in order to provide the base for a more extensive assessment of the case study. Considering the scope and aim of this research, a variable is selected from the SESF that is most relevant to the case. The selection is based on the amount of information available and the relevance to the goal of this study. The diagnosis of Nagendra & Ostrom (2014) shows a detailed example and explanation on the selection of indicators in the application to a specific case of an SES in India. This study follows a comparable selection method. The variables under the header “Actors” refer to be associated with self-organization, which is one of the key characteristics of adaptive management. The variable ‘trust-reciprocity / social capital’ (A6 in figure 1) is chosen. For adaptive arrangements to emerge in the UA system, there must be a high level of trust between organizations, agents and the municipality. Trust and social capital have a strong influence on the independence of self-organization of UA projects, since users of resource systems who have sufficient trust in one another to keep agreements face lower transaction costs in reaching agreements (Ostrom, 2009). In figure 2, a schematic depiction of the integrative assessment model is given.

(17)

Fig. 2: Integrated model to assess the role of adaptive governance in the development of UA

The model in figure 2 shows the variable ‘trust-reciprocity / social capital (originally from SESF) in the light blue box, integrated with the ‘conditions for adaptive governance (Folke et. al., 2015). The grey box below shows the two attributes of adaptive government: polycentric governance and adaptive co-management. All variables combined provide insight in the role of adaptive governance in the development of UA (green). The next chapter provides a detailed description on the methods utilized to examine the variables in this model.

(18)

4. Methodology

4.1 Research design

The aforementioned integration of two theoretical paradigms has not been executed in this specific arrangement before. Therefore, the assessment has an explorative nature. The main source for qualitative data gathering will be derived from interviewing participants, policy makers and representatives of involved organisations. The variables presented in figure 2. form the themes of the interviews. The guiding interview questions can be found in the appendix (9.2). The complexity of an SES also asks for a mixture of different forms of data gathering. In order to make a deep assessment, a variety of qualitative data methods must be applied to the management practices and the perceived effect of these measures on involved participants within the UA network. In the light of data triangulation, validity is enhanced by bringing together various qualitative methods (Kawulich, 2005). The further contextualization of the results is done through reviewing literature and policy documents.

4.2 Case study

To investigate the role of adaptive governance in the development of Urban Agriculture in the Amsterdam area, a case study of the Amsterdam urban food network will be executed. When investigating the contextual factors that shape policy in a specific area, a case study is a logical methodological approach (Baxter and Jack, 2008). Insight in the ways local policy has shaped the urban agricultural system demands a focused case study into the interactions between policy makers, urban farmers and others involved. Amsterdam is the most populous city in the Netherlands, and can provide more data when it comes to UA initiatives in the Netherlands than smaller cities. The authors’ personal network is concentrated around Amsterdam, which makes those most accessible sources, especially considering the difficulty of reaching out to institutions and organizations over a short period of time.

(19)

4.3 Interviews

Over the course of four weeks, 8 semi-structured interviews were conducted with UA practitioners, policy makers and one expert in Amsterdam. Semi-structured interviews are distinct from other quantitative approaches as they still retain a useful level of flexibility (Bryman, 2016: 471). Bryman notes that, while unstructured interview methods allow the interviewee to intervene more freely, ‘with the interviewer responding to points that seem worthy of being followed,’ an interview will be the most effective in achieving the aim of this research, as it helps to steer the interview to the specific circumstances of the interviewee and the organization(s) he or she is affiliated with. To get a more complete overview of the UA network in Amsterdam a broad range of different respondents was interviewed. Policy makers and civil servants are responsible for policy design and implementation and can therefore provide information about the intentions and performance of past and present policy matters. Urban farmers and affiliates on the other hand can give notion on the other side of the policy process, in order to receive more complete understanding on the effects of current governance practices on the recent development of the network. On request of the interviewees, their anonymity will be preserved. The interview questions are established alongside the theoretical concepts elaborated in the assessment scheme. All interviews were conducted in April, May and June 2019. After the conduction, interviews were recorded and transcribed. These texts were closely studied and inductively coded. Quotes are highlighted along the theoretical lines of the grounded theory of qualitative data analysis (Bryman, 2016). After the initial coding, the texts were scanned again for specific recurring concepts such as polycentric arrangements, trust and social capital between agents, institutional flexibility, organizational learning and recognition of bridging organizations.

4.4 Participant observation

Alongside the interview analysis, participant observations were conducted as an additional method of data gathering. Participant observation provides the researcher with the tools to determine interactions, communication and the time and type of activities (Schmuck, 1997). This way, events and communications could be observed which can provide information that might not be shared during interviews. Additionally, it helps to develop a ‘feel for how things are organized and prioritized’ (Bernard, 1994). It allows forging a greater understanding of the context of the daily occupations of participants and

(20)

their interrelations with other involved actors such as the municipality and other UA projects or community organizations. This could be useful considering that this study also concentrates on partnerships and independent decision-making practices between participants. The list of field visits and interviews can be found in the appendix (9.1). During the field visits, the researcher has been joining city farmers in their work on the farm, actively participating in the daily tasks such as planting and harvesting crops, or attending UA-related events such as planting-day. The researcher has been engaged in several conversations during these field visits, which were either recorded or notes were made. The texts have been transcribed and were separately analysed. When conducting interviews and field visits, the researcher is inherently part of the subject that he or she is investigating, thereby it is not excludable that this could have an effect on the data that is produced and analysed. Therefore conclusions may not be swiftly drawn, and findings are only appropriated to the case or investigated organization itself.

4.5 Document analysis

The qualitative data derived from interviews and field visits will be supplemented by document analysis and literature review in order to complete data triangulation, further adding to the reliability of the analysis (Bryman, 2016). Available online and offline policy papers, academic work and other relevant documents are examined. Special attention will go out to the policy documents published by the municipality of Amsterdam in the past years, since it provides essential information about the management actions taken with relation to UA. The following chapter will present the results of all gathered data, juxtaposed to the recurring theoretical concepts.

(21)

5. The role of Adaptive Governance

This chapter provides a detailed overview of the findings from the interviews and the document analysis. The list of interviews and field visits can be found in the appendix. The structure of the analysis follows the sequence of factors as prescribed by the integrated assessment model as shown in figure 3. Firstly the conditions for adaptive governance to emerge are covered (enabling legislation, learning and recognition of bridging organizations). This is followed by an assessment of the trust-reciprocity and social norms within the governance network, which is the integrated additive condition from the SESF (blue box). Next, this chapter will go into the characteristics of adaptive governance (grey box): polycentric governance and co-management. All assessed factors are depicted below in figure 3.

Figure 3. Integrated assessment model

5.1 Conditions for adaptive governance to emerge

This paragraph will commence with an examination of recent legislation in Amsterdam regarding urban agriculture and is then followed by institutional learning and the recognition of bridging organizations. The most influential legislative actor with regard to UA in Amsterdam is the municipality. All leasing contracts, subsidies and central networking activities are organized distributed under the responsibility of the city government, led by the aldermen and the current mayor Femke Halsema. In 2010

(22)

city in the Netherlands) had already actively incorporated UA policy (CITIES, 2011). However, since 2015 the first UA policy plan in Amsterdam was presented in the policy document ‘Agenda Groen 2015-2018’ (Gemeente Amsterdam, 2015). Here, UA is specifically brought to the attention as a development that is advantageous for the citizens of Amsterdam. Therefore it would require municipal support. The plan proposes to reduce ‘obstructive rules’ for new projects to provide guidance through stimulating overarching civil platforms and creating an informative and connecting online platform. Not unimportantly, subsidy scheme was launched specifically for UA projects. The fact that there is a general policy framework is regarded as an asset of the Amsterdam municipality in comparison to other Dutch cities:

‘At the moment, at a municipal level there is more clear policy when it comes to urban agriculture in Amsterdam. They incorporate consumption more in their policy and they are also more occupied with integrating local food production in the city than in Rotterdam.’ (Interviewee 3)

At first sight, these policy measures correspond to the conditions (such as enabling legislation and bridging organizations) for adaptive governance to emerge. On the basis of empirics from interviews and field visits, the perceived adaptive potential of these measures is investigated. The municipality has been involved in pointing out suitable locations and creating specific conditions for projects in order to provide guidance. Enabling legislation such as offering contracts, property permits and subsidies are occasionally granted. In some cases, specific prerequisites were formulated with these contracts:

“We have been here for 5 years now and got the land from the municipality. This used to be an abandoned shipyard for some years. The municipality said during the economic crisis that we could come up with a plan. The plan was to on the one hand desalinate the soil by planting special trees, remediation is what it’s called.” (Interviewee 1)

Analysis of the interviews (1,2,3,4,6,8) shows various other examples of the Amsterdam municipality offering vacant lots for interested entrepreneurs to start urban farming

(23)

projects and other types of community gardens. The degree of involvement by the municipality differs strongly per case. Some projects operate with almost no support of the government. Others are actually initiated by the municipality itself. This can be considered as a flexible instrument to quickly facilitate and enable UA projects within the city boundaries. The length of contracts can vary between 2 to 10 or more years. In some cases, projects apply for a contract and permits, but sometimes projects are invited to lease specific vacant plots:

We have a lease contract for 50 years and 45 years left. We can do everything we want with it, as long as it is urban agriculture-related, which is quite stretchable. We propose as many projects as possible. It is very open and it’s become a kind of innovation hub. They start here and grow further. After two years the municipality checks how successful it is and decides on further support. It is nice that the municipality facilitates this. (Interviewee 1)

Apart from enabling legislation, the political leadership and the priority of the ruling policy makers influence the amount of attention that goes out to UA in general. Urban agriculture is a relatively new development in many cities and the interviews clearly show that the municipal organization is struggling to find a coordinated way of managing it. The following citation from an interviewed civil servant occupied with UA (6) in the municipality illustrates this:

For many civil servants within government organizations, urban agriculture is a relatively new and unknown concept, which means that there has to be room for learning and experimenting within the municipal organization itself’. ‘Financing, spacial development and participation need to be more embedded in policy and information sharing within the municipal organisation needs to be improved.’ (Interviewee 6)

There is a relatively strong sense of institutional learning to be identified from the empirical material. The ability of the municipality to respond to local initiatives through subsidy granting is built into the current policy framework. The absence of incorporating hard targets within the framework gives the municipality officials a degree of liberty

(24)

regarding the amount of projects financially supported. However, this does not guarantee any form of institutional flexibility or learning. From data of this study there is no obvious indication of shifting policy actions as a consequence of learning. Rather, there is flexibility in selecting and stimulating specific projects or areas. The municipality is (slowly) adjusting to new UA developments from civil society. Nevertheless, according to more than half of interviewed practitioners, there is too little decisive effort yet. Apparently, there is dissatisfaction with the fact that there is no consistent steering from the municipality. This could have to do with the low level of congruence within the municipal organisation. This so-called ‘pillarization’ between departments is an issue that was drawn attention to by two respondents (1,6):

‘I notice that the civil servants are enclosed within their own ‘silo’ and are not flexible enough to quickly respond to new initiatives. I feel that is why the municipality struggles with how they want to shape the transition to ‘green’. The alderman also has his own ideas and mainly wants green roofs. The goals are not aligned and clear.’ (Interviewee 2)

Difficulties in contacting municipal officials and the complex subsidy or permit constructions were another result of this. The rules and processes not yet well enough designed to facilitate citizen initiatives. A civil initiative comes with propositions that do not align with the standards set for public space arrangements. This calls for careful monitoring and flexible support by the central legislative actor (Folke et. al., 2015). Enabling legislation for example should favour civil entrepreneurs to easily start projects. However, current policy and municipal accompaniment in the start up phase falls short. For example, the process of applying for permits and leasing contracts is too difficult for most ordinary citizens planning to start a UA initiative, illustrated by a practitioner (4):

‘The municipality does not take enough action in taking new entrepreneurs by the hand in starting up UA projects. In practice it was almost impossible to register for starting a UA project via the municipality.” (Interviewee 4)

(25)

An expert stresses that civil entrepreneurs are not skilled enough and do not have enough time to go through these difficult processes, which is why they need more guidance:

“It struck me that the success of projects was very dependent on what persons you were in contact with at the municipality. If you have citizens as co-creators of UA projects, they have to deal with pressures that they often do not have the right capacities for. They also have to balance professional and private life and they struggle to devote enough time and energy in such projects. Also, these processes of subsidy application and permits take too long for an average citizen.” (Interviewee 8)

Finally, adaptive governance theory (Folke et. al., 2015) dictates that the circumstances for adaptive institutional arrangements can be created by the recognition of bridging organizations and actively connecting initiatives. The municipality is doing this sufficiently according to one interviewee:

“The municipality is successful in bringing together parties in the network. In my eyes this is sufficient. The activating role in the sense of subsidies and investments is not sufficient. They should give the right example and stimulus by for example public procurement of local food.” (Interviewee 3)

According to others (5,8), the provision of online information and communication was insufficient. The municipality intended to actively maintain an up to date informative online platform in conjunction with organisations from civil society (Gemeente Amsterdam, 2015). Some organizations such as ‘De Gezonde Stad’ received attention and recognition from the municipality. On the other hand, the platform created by the municipality itself (‘Van Amsterdamse Bodem’) was not welcomed as much, according to an expert (8):

‘‘De Gezonde Stad’ has posted on the website to attract initiatives and support them with knowledge in exchange for help from the municipality together with citizens. This organization was recognized and their goals with regard to

(26)

greening of the city soil aligned well with the ones of the municipality. Why cant the municipality make such clear guidelines for urban farmers as De Gezonde Stad?” (Interviewee 8)

5.2 Trust-reciprocity and social capital

The long-term goal of the municipal policy framework is for the urban agricultural system to function independently (Gemeente Amsterdam, 2015). After the start-up phase of projects, the municipality would be mostly involved at a distance, in cooperation with civil organisations. This theoretically corresponds with the ideas of adaptive management and self-organization. Though, a sufficient level of trust is an important prerequisite for building such a self-organising adaptive network. According to two interviewees (6,8) the rise of UA in Amsterdam is partly caused by the economic crisis that started in 2007, resulting in many unfinished or cancelled real-estate development. The vacant lots then were attractive to urban agricultural projects and the municipality would be interested in giving it temporary repurposing. At the moment, Amsterdam is experiencing steep growth in building projects throughout the city. There is apprehension among interviewees that this is being reversed when the interest in real estate development keeps growing. Interviewees also draw attention to the temporary nature their leasing contracts. They fear that the UA projects that are now thriving will quickly be replaced for real estate development (1,5,8):

“It seems that these projects (pointing at an apartment block across the street) get the priority over ours.” (Interviewee 5)

In addition to the fact that there is uncertainty concerning the continuity of UA projects, there are more frustrations with relation to subsidies, transparency and general involvement of the municipality that have strongly reduced trust in the municipality among city farmers. For example, the subsidy application inquiries seem to be a rather difficult and time-consuming process. Multiple respondents (4,5,8) have underlined the need for a more easy and responsive registration process. ‘Representatives of beginning UA projects often do not have the knowledge or experience to go through these difficult processes.’ “Since there is a subsidy regulation and financial means are made available

(27)

especially for this, it should not difficult to apply for it.” Nevertheless, it turns out to be a rather complex task:

“We had to apply for the subsidy, but the timing was horrible. The municipality does not offer this, another company does this. Then you have eight weeks left to get the permission. The time and effort it takes to apply for a subsidy is long. We could not get this greenhouse, because it looked ugly. The municipality is not happy with the way it looks. It is a bit contradictory. They do like urban farming, but not how it looks.” (Interviewee 4)

Another cause of uncertainty is the recent jurisdictional adjustments within the municipality. When it comes to municipal decisions on subsidies and space distribution, local departments have often been the main point of contact for individual projects. The recent centralization within the municipal organization has brought uncertainty and unexpected changes in government involvement with local projects. Since 2015 there has been a reorganization of the municipal organization in the name of efficiency and transparency. Some local district jurisdictions have been centralized and moved away from city district departments. Respondents emphasized that, because of this, continuous involvement is not assured. Three interviewees have stated (3,4,8) that there can be large differences in the willingness of different municipal departments regarding accessibility and the extent of their support. One interviewee expressed special discontent towards the reorganization, which had apparently led to a drastic decrease of involvement in an UA project, originally initiated by the municipality itself:

‘Something happened within the local municipality and it has to something to do with the decentralised city departments. It has been recently fully centralised. Now we work together with the city ecologist, chief innovation etc. They have been obstructing very much recently. They have mowed down some parts that we explicitly assigned to steer clear of. These are the same people who initiated the whole project. There is a disorder within the local organisation of the neighbourhood.’ (Interviewee 5)

(28)

Another expert interviewee pointed at the lack of accessibility for new projects as well:

‘It is a pity that they ruled out the districts, cause it was a more accessible point of contact for new initiatives.’ (Interviewee 8)

We can conclude that the level of trust between the municipality and the UA-projects and affiliates is not strongly contributing to the growth of adaptive arrangements.

5.3 Reflection on the utility of trust-reciprocity in this assessment

In this study, the adaptive management paradigm to diagnose systems of UA was extended by adding a variable from the social-ecological systems framework (SESF). Trust-reciprocity was merged to the conditional factors that contribute for adaptive governance to emerge. According to Ostrom (2009), a level of trust between agents supports the self-organizing capacities of a community. Findings show that there is a lack of trust between UA participants and the municipality. It therefore contributes to our understanding of the local factors that have an effect on the functioning and the emergence of self-organizing adaptive arrangement. The presence of a self-organizing system (through polycentric arrangements and co-management) is diminished through a lack of trust. Based on this assignment, the application of this variable was useful. It was helpful in showing the relation between actors and the expectations between actors, and therefore the motives for them to act independently or not.

5.4 Polycentric governance and adaptive co-management

According to an interviewee (8) “a new layer of governance” must be created within civil society, in order for self-organization and more public participation to emerge, because it ‘originates outside of the formal legislative institutions’. This interviewee has doubts with regard the extent to which this is happening.

“It actually happens outside of the formal structures of policy-making. How can you connect this governance at a distance of the central organisation (municipality). Judicially all responsibilities are well defined. But we still had the feeling, who do they speak for? Do we have enough grip as citizen? So how does a

(29)

municipality organize a layer of governance, outside of the formal boundaries? This new entity constantly has to prove itself that there is trust. The interesting thing is that this governance layer disappears if no one nurtures it.” (Interviewee 8)

The initial stage of project development generally is accompanied with closer involvement from the local government than in later phases. During the process of maturing of the projects, most interviewees of UA project representatives (2,3,4,5,7) have said to be building contact with other projects and organizations in the area and further away to strengthen their position within the network. Some bridging organizations are well embedded within policy networks. Most urban farmers state that these organizations are in regular contact with the municipality. It seems that these bridging organizations are able to establish a more reliable contact relation with local policy makers, ensuring a more regular basis of support:

“The projects in Amsterdam-West have made a business organization and appointed an intermediary to keep regular contact with the municipality and streamline the whole decision-making process.” (Interviewee 8)

The existence of autonomously operating decision-making centres is a key component of polycentric arrangements. From this study it becomes clear that the municipality is supportive of this. The extent to which it effective in reaching its goals is not measured in this study. The empirics derived from the interviews also show that there is a substantial degree of involvement of the municipality in connecting initiatives, which is essential for creating conditions for polycentric arrangements (Ostrom, 1990).

“We are in contact with various other initiatives. Also there is an online platform created to bring these initiatives together. In this way we are also involved in policy making in the area. And subsidies here are more easily given. Here we work together with the municipality very well.” (Interviewee 7)

Connecting existing initiatives and acting as a guiding post in the development of knowledge is an important instrument to support polycentric arrangements to emerge.

(30)

This is a central pillar of the 2015-2018-policy plan. For example, the municipality offers an online platform rendering all existing and possible locations for urban agriculture (Gemeente Amsterdam, 2019). Collaborative management of public green spaces and urban gardens ideally encompasses actors from different types of sectors is better suited to embrace change and. Searching to further strengthen influence in the political arena, interviewed city farmers are in regular contact with other projects and civil organizations in the surrounding area. They get involved in local ‘green’ pressure groups and form alliances with other farms to share distribution or even merge into bigger cooperations. In the Amsterdam local food network collaborations are emerging on various levels. Questioned farmers are forming alliances with other farms and local buyers (4), and they are joining local collectives of ‘green’ pressure groups (5). This is how they attempt to mobilise groups of initiatives to gain more influence. Individuals are often involved in multiple projects too, reinforcing the shared management of projects:

‘I am part of the “…” in Amsterdam Zuid-Oost, and involved in some other groups of urban agriculture in the area. We have a harvesting club (CSA). There are different initiatives in this area, all using the same plot.’ (Interviewee 2)

‘A group of “…” women have joined our project. We are in constant contact with projects here in the neighbourhood.’ (Interviewee 5)

Independent collaborations such as business organisations and informal conjunctions predominantly emerge around hotspots in the city. There is a lot of knowledge exchange between these parties. It is a network of actors that reach out to each other around specific initiatives (1,2,3,4,5,7). Community effort can be woven into the organizational structure of city farms, since they are often maintained and founded by local residents. A prominent example of such an organizational structure is the Community Supported Agriculture (CSA) format. Two of the respondents (4,2) have implemented this method in order to run their farm in collaboration with local residents. The community has a shared responsibility for providing revenue for the main organization through shared harvesting and consumption. Several respondents (4,8,5) have underlined the importance of such cooperation, as it ensures a more regular and steady maintenance of the farm field:

(31)

“We are a CSA (community supported agriculture). People pay a set fee and they can harvest their own food. They have no fixed amount they can harvest, but it is also not overexploited. Most of it goes to the community and the surplus is sold to restaurants across the city, cause there is a pretty big demand”. (Interviewee 4)

This community supported organizational structure can be seen as a form of co-management on micro scale. Trust is built through collaboration and shared responsibility. Such innovative institutional arrangements can improve learning the ways communities deal with social-ecological complexity (Armitage et. al., 2009). Based on data from this study, we can conclude that there is a substantial level of self-organization observable through polycentric arrangements of co-management.

(32)

6. Conclusion

The aim of this study is to explore the role of adaptive governance in the development of urban agriculture, based on a case study of the Amsterdam local food network. UA-governance was approached from a unique angle, by adopting a social-ecological perspective in combination with adaptive management literature. A thorough assessment of the UA network in Amsterdam was done to reveal the adaptive arrangements within the system and the dynamics between local governance practices and the people involved with urban farming.

Results show that urban agriculture has been gaining attention in Amsterdam. It is being woven into the urban fabric of Amsterdam and adopted in the recent ‘green’ policy framework by the municipality. Generally, the municipality has a strong intent to stimulate existing projects and facilitate new ones where possible. UA-projects would be a beneficial instrument for contributing to their targets of urban food consumption, civic participation and social integration. One could argue that a form of institutional learning has taken place, in which the recognition of UA has led to policy action aiming to further support it. Interviewees express their content with the presence of such a policy framework because it ensures more stimulation than beforehand. At a first glance the policy measures also match the basic conditions for adaptive governance to come about in the network. There is a subsidy scheme for start-ups, a promise to improve guidance through an online platform and reduce obstructive legislation. A degree of institutional flexibility and learning can be observed over the past years. Nevertheless, from the perspective of urban farmers and interviewed experts, these measures are not yet well enough executed. They experience problems with the accessibility and the degree of continuous support by government officials. Although the municipality is sufficiently involved with connecting initiatives, the supportive online platform leaves something to be desired. Additionally, the process of applying for subsidies and permits is deemed too difficult for the average civil entrepreneur. The recent centralization took jurisdictions from the local departments, which further decreases accessibility of government supposedly. These inconveniences lead to a lack of trust-reciprocity between UA organizations and the municipality. There is reason to believe UA benefits from

(33)

transferring power to civil society through adaptive management, but based on these observations we can conclude that the lack of these conditions in combination with the absence of sufficient trust have a negative influence on the emergence of autonomous adaptive arrangements.

The extent to which adaptive arrangements were present within the network was also examined. The concept of polycentric governance is regarded as a valid indicator (Ostrom, 2009). Polycentric decision-making centres were identified in the form of local overarching NGO’s making decisions on local UA affairs in cooperation with the municipality. A reliable verdict on the level of autonomy that these decision-making centres have in relation to local policy makers cannot be given, since final decisions are often still made by the municipality itself. Polycentric arrangements could also emerge in systems of adaptive co-management. Various innovative projects of co-creation were found in this study, from business organizations to cooperating harvesting groups. Based on this we can conclude that there is considerable potential for these arrangements to further develop the adaptive capacity of the UA network in Amsterdam. The concept of adaptive management is a product of changing values and expectations from the (local) government. More flexible legislation and responsive urban governance should enhance the development of such social innovations, but practically it is laborious process.

All in all it becomes clear that the Amsterdam is undergoing two major changes in the light UA-governance. Firstly, the priorities of governance practices and institutional arrangements are shifting within the municipality. Attempts to empower the innovative capacity of civil society are firmly on the political agenda. At the same time the municipal organization is still attached to old structures and rules, impeding the implementation of adaptive responses. Naturally, a transition to adaptive institutions takes time, but nevertheless, there is room for improvement. The second transition happens on the other side of the governance spectrum. Civil society itself is initiating UA projects throughout the city. Cooperation between projects is emerging around UA hotspots. Without the right municipal support however, it is subsistence is uncertain.

The two transitions mentioned should ideally develop in harmony. Adaptive management by the municipality could contribute to further congruence, but it will take time and

(34)

adjustment of all involved parties. Also, it is obvious that change and adjustment is inherent to an adaptive system. Though, one must be conscious of the fragile nature of adaptive governance arrangements. A shift in priorities in both civil society and the municipality could quickly lead to the decrease of adaptive capacity.

(35)

7. Discussion

7.1 Limitations

The implications of this research are incomplete without a thorough reflection on the choices made in this study and limitations that may be involved with these choices. This research offers very much a snapshot of a specific timeframe. The flexible nature of present-day governance implies the changeability of the examined situation. The outcomes of this research are therefore to be used with care towards future implications and with reference to the period it was executed in. Furthermore, the theoretical integration this thesis attempts to effectuate is the first to be applied to a case of urban agriculture. It provides a new adjusted analytical tool for the assessment of adaptive governance, which it is not yet a well-tested approach. Therefore, the conclusions of this work are not to be generalised. This means that the results of this specific case are not directly replicable to other cities, also because the conditions between cities may vary (Prove, 2016). The qualitative nature of this study is expressed in the aim to understand a policy process from the eyes of agents, municipal actors and experts. The data from interviews and field visits illustrate the different perceived effect of management actions and the dynamics within the governance network. This study aims to provide a detailed overview of this, not a verdict on the implications of local policy on all actors. The research process came with some other challenges to overcome. The application of the social-ecological systems perspective on a case of UA opened doors for deeper diagnostic assessments, but also much to be further tested. This research was only a fraction of all possible applications. Also, the integration of the theoretical frameworks turned out to be a complicated task. Both the Social-Ecological Systems Framework and the attributes of adaptive governance share a same underlying scientific origin, but they are operationalized on different aggregate. The time frame of this study stood in the way of a more inclusive integration, but it is still highly relevant to further investigate and experiment with merging insights from both approaches.

7.2 Recommendations for further research

The theoretical explorations that are a vital part of this study reaffirm that the social-ecological perspective is useful in making assessments of UA governance practices, especially in combination with adaptive governance literature. Incorporating the variable

(36)

trust-reciprocity from the SESF creates insightful observations on the quality of adaptive responses within these networks. Therefore, it is recommended to further apply the social-ecological systems perspective on other cases of UA-development.

Regular monitoring of UA-governance is essential for the success of adaptive arrangements within the network. Hence, for both local governments and the academic world it is of high importance to regularly investigate the implications of local governance on the self-organizing capacity UA-projects in times of institutional change. This new development demands a deeper understanding of new forms of public and community governance that form the basis for UA to thrive. Adaptive arrangements are the preferred style to accommodate this transition, but the difficulty of the implementation shown in this research and in others (Lohrberg et. al., 2016) underlines the need for more locally instigated research.

Who farms the city? : a case study of urban agriculture governance in Amsterdam, the Netherlands