The role of cities in good governance for food security: lessons from Madrid’s urban food strategy

ABSTRACT Cities have become increasingly involved in food security concerns that go far beyond their boundaries by rescaling food governance to a more local level. While the role of cities as new food policy actors has attracted increased interest, few studies have critically analysed urban food governance. Critical accounts of these new governance configurations are particularly necessary, and more relevant in times of crises, as good governance is crucial in tackling the root causes of food insecurity. This study contributes to the debate around urban food governance, focusing on Madrid’s urban food strategy. It applies an adaptation of the Food and Agriculture Organization’s (FAO) analytical framework for good governance for food security. The study shows that new urban food governance arrangements can create more inclusive, democratic solutions to food security. However, these arrangements do not guarantee good governance, and many challenges remain, such as the operationalization of urban food policies and contextual barriers. More broadly, this study demonstrates that critical assessments of governance for food security can identify the deficiencies in the current system and help address current barriers to good governance.


INTRODUCTION
Many countries around the world are now facing an overlapping triple burden of malnutrition: undernutrition, micronutrient deficiencies, and overweight and obesity. These are caused by inadequate economic, social and physical access to healthy food (Food and Agriculture Organization (FAO), 2018; Labadarios, 2005;Pinstrup-Andersen, 2007). In other words, food insecurity plays a pivotal role in the simultaneous coexistence of multiple food-related challenges. This problem has been increased by the 2007-09 food crisis, and more recently by the COVID-19 pandemic. These crises have revealed that food policies, developed in higher governance levels, have significant shortcomings in addressing the underlying issue of food security (Sonnino & Spayde, 2014). The failures of food policies and governance structures are related to a disregard of the underlying principles of good governance such as effectiveness, responsiveness, equality and fairness, which have been highlighted as necessary to address root causes of food insecurity and eradicate hunger (Candel, 2014).
Although efforts to ensure equitable food access have been going on for decades, food insecurity persists (Candel, 2014). The ongoing challenge of tackling food insecurity has revealed that governance is crucial in providing enabling or hindering conditions to ensure equitable access to food (Moragues-Faus, 2020). That is, there is a growing recognition that the decision-making and implementation processes conducted in food policy arenas are essential to achieve better food security outcomes.
Traditionally, food security concerns have been discussed at higher governance levels, with a narrow and self-referential agri-business sector and representatives from vulnerable sectors noticeably absent (Morgan et al., 2006). Events such as the 2007-09 food crisis revealed the shortcomings of national and global decision-making processes concerning food security and raised awareness of its complexity (Lang, 2010). The aftermath of these events has been an increased agreement that new food policies must assume relationships between the environment, social justice, and health; integrate systems thinking; be accountable and transparent; and create new governance structures that are more horizontal and inclusive (Sonnino, 2016).
The relevance of governance has also received increased attention in academiamore so with the COVID-19 pandemicas it is vital to address food security challenges. However, there is still no clarity of what governance entails or how it should be analysed. Governance is a concept that is prone to multiple interpretations and uses (Rhodes, 2012); its initial focus on state-centred modes of managing public issues is long overdue. 1 As a continuously evolving concept, governance can be understood and applied to politics and policy-making in various ways. The ambiguous nature of governance has led to its application to the notion of food securitywhat is here termed 'governance for food security'in simplistic and narrow terms. Mainstream understandings of governance for food security focus on problem-solving strategies that rely on aligning the fragmented institutional architecture (Candel, 2014). However, these often overlook stakeholder interactions' process dynamics. Although still contested conceptually, multiple studies suggest that these interactions are often why some governance arrangements fail to meet their objectives (Candel, 2014;Moragues-Faus et al., 2017).
Thus, for the purposes of this study, governance for food security is understood to draw on conceptualizations that focus on networks and interactions. These acknowledge the contested nature of decision-making processes and the need to build multi-stakeholder policy platforms and are therefore most useful in understanding the dynamics of food policy-making. These perspectives call for the inclusion of diverse societal actors to combat so-called 'wicked' problems such as food security because of their complexity, diffusing the concept beyond the state (Klijn, 2008).
Taking a network perspective, governance can be defined as modes of horizontal steering that are effected by networks of actors (Sørensen & Torfing, 2007). Thus, governance networks 'can be seen as an indication of more or less stable patterns of social relationships between mutually dependent actors, that arise and build up around complex policy issues or programs' (Koppenjan & Klijn, 2004, pp. 69-70). This resonates with Kooiman et al.'s (2008) proposal of governance, emphasizing the governing roles and interactions between the state, market and civil society. Accordingly, in modern societies, civil society and the market also have essential roles in local, regional and international governance, under the notion that 'societies are governed by a combination of governing efforts' (Kooiman et al., 2008, p. 2). Similarly, Von Braun and Birner (2017) define global governance of the agricultural and food system as 'the formal and informal institutions and organizations at the global level that aim to influence the agricultural and food system' (p. 267). McKeon (2015) also argues that an effective and equitable global governance must be built 'from the bottom up and incorporates the democratization of food provision at all levels' (p. 329). Based on these conceptual developments, key characteristics of governance for food security include an inter-organizational dimension of policy-making and the early involvement of societal actors, stakeholders, and citizens' groups in decision-making.
Nevertheless, alternative perspectives of governance tend to disregard the potential negative consequences of internal stakeholder dynamics. In many cases, the opportunity to merge these alternative conceptualizations with power and inequality perspectives is missed (Pierre & Peters, 2005;Torfing et al., 2012). This reinforces the idea that research on governance for food security needs to consider governance interactions without de-politicizing processes by omitting the analysis of power tensions and winners and losers (Moragues-Faus, 2020). Moreover, the focus on governance interactions should not preclude the analysis of essential criteria to advance good governance for food security such as those promoted in current food scholarship and with more emphasis after the crisis and the COVID-19 pandemic: horizontal and inclusive platforms, inclusion and no discrimination, and respect of democratic values such as accountability and transparency (Sonnino, 2016).
Many international organizations, such as the European Union, the Organization for Economic Co-operation and Development (OECD) and the World Bank, have proposed principles of good governance. 2 For example, the United Nations Development Program (UNDP) defines good governance as being accountable, transparent, responsive, inclusive and equitable, effective, participatory, democratic, and following the rule of law (UNDP, 2014). The use of these principles has been primarily reduced to the assessment of governments and the prevention of corruption, especially in the context of UNDPs. Nevertheless, Kickbusch and Gleicher (2012) stress that 'good governance … is an amalgam of guiding principles that transcend specific policies, sectors and actors' (p. 41). As such, good governance is better understood as a dynamic process rather than an end goal or outcomeand is increasingly acknowledged as necessary in addressing the underlying causes of food security because it ensures a generally supportive environment in which human rights are respected, and the provision of food as a public good is guaranteed (Candel, 2014;FAO, 2011). This paper recognizes that interactions between governance actors lead to divergent good governance characteristics and, thus, positive or negative food security outcomes. Therefore, and in order to consider a holistic approach, this study proposes an adaptation of the FAO's notion of good governance for food security (FAO, 2011), integrating it with the alternative governance conceptualizations explained in previous paragraphs. Building on this, good governance for food security is defined here as modes of horizontal steering (governing) carried out by networks of interdependent public and private actors in the pursuit of food security while meeting particular effectiveness and democratic criteria. This approach applies to various levels (including local) and goes beyond the government to include other essential stakeholders, incorporating the democratization of the food system proposed by McKeon (2015). Moreover, it integrates a human rights-based perspective and the assurance of the provision of basic public goods. Hence, this study proposes an unconventional understanding of governance for food security compared with other academic literature, focusing on how it should ideally look like and how governance systems are functioning at present.

CITIES AS ACTORS TOWARDS GOOD GOVERNANCE FOR FOOD SECURITY
With more than half the world's population now living in urban areas, cities have become key players in addressing the complex socioeconomic and sustainability challenges related to food security (FAO, 2019b). There has been a corresponding increase in the attention given to the role cities play in addressing the underlying causes of food insecurity, such as developing more sustainable food systems (Morgan & Sonnino, 2010). This is mainly because cities have identified themselves as new food policy players, especially through local food systems planning using novel governance instruments (Toldo et al., 2015). Compared with higher governance levels, cities have adopted a more collaborative approach using tools that promote joined-up food policies, the democratization of policy-making and knowledge exchange. The academic literature considers this new local connectivity and urban food policy integration vital for enhancing food security (Sonnino, 2016). Moreover, some scholars argue that urban food governance can facilitate greater transparency through democratic practices (Hassanein, 2003). Thus, cities are championed as spaces where good governance for food security is thriving.
From a governance perspective, cities use two main mechanisms to implement food systems change: the establishment of new places of deliberation such as food policy councils (FPCs), and the drafting of UFSs (Sonnino & Spayde, 2014). Deliberation platforms such as FPCs have been established worldwide since the 1980s and are often initiated by government actors through executive orders, public acts or joint resolutions (Harper et al., 2009). FPCs provide a space for the building of intersectoral and multi-stakeholder relationships that lead to changes in food system policy. In general, there are three models of organizing the relation between a FPC and the government: a government organization, a non-government organization and a hybrid model (Schiff, 2008), each with specific benefits and drawbacks. FPCs serve several interrelated functions, such as providing an opportunity for diverse stakeholder discussion, promoting coordination across sectors, creating new or supporting existing programmes and services that attend to local needs, and influencing, monitoring and evaluating policies (Clayton et al., 2015).
An essential part of an effective FPC is the design of an UFS. UFSs display how a city envisions a change in its food system and intends to achieve this change. UFSs take a holistic and systemic approach to the urban food system, considering horizontal and vertical dimensions (Toldo et al., 2015). Local characteristics lead to differences, but despite this, most UFSs emphasize the creation of an inclusive governance context. Although in most of the strategies the expression 'food security' never features in their titles, it is arguably the central underlying theme of UFSs. Thus, UFSs and FPCs are examples of governance for food security at the local level. UFSs tend to approach food security in holistic terms, acknowledging that it is connected to social inequalities, the environment and the local economy (Sonnino, 2016). For example, strategies can include measures to address health inequalities through dietary education or promote local and sustainable food consumption.
UFSs and FPCs are considerably different from mainstream governance configurations found at national levels. Indeed, research has primarily focused on how these urban food governance efforts reflect a change in the current governance for food security paradigm by filling the policy vacuum left by national policies (Coulson & Sonnino, 2019). Nevertheless, some scholars have suggested that these new governance arrangements can obscure unequal power relations among different stakeholders through de-politicized collaboration processes (Moragues-Faus, 2017;Sonnino et al., 2016;Swyngedouw, 2005). Some regard them as bureaucratic managerial processes based on weak participation arrangements -'token' participationand consensus decision-making mechanisms led by the most powerful actors (Moragues-Faus & Morgan, 2015). Moreover, Mansfield and Mendes (2013) highlight significant procedural and structural factors that affect the capacity of local institutions to implement UFSs. For example, local governments rely on the participation of civil society organizations and the motivation of municipal officers to implement UFSs, which poses a risk to their long-term sustainability (Moragues-Faus & Morgan, 2015). In addition, FAO's (2019b) recent Urban Agenda framework highlights the need to address shortcomings, such as social inclusion, in local food governance mechanisms. Therefore, there is a growing concern that current understandings of UFSs and FPCs do not acknowledge their potential negative consequences, following the depoliticization trend identified in alternative governance perspectives. In other words, there is an increasing body of work that implies that cities are not spaces where good governance for food security is inherently enacted.
Unlike other studies that take an optimistic view of FPCs and UFSs, this paper takes a more critical approach to uncover the shortcomings of urban food governance. The presented research concentrates on the processes that impede the achievement of principles of good governance and thus potentially lead to adverse food security outcomes. Understanding how these new urban food governance structures build spaces that follow the underpinning principles of plural inclusive participation and rights is paramount to inform policy to create impactful strategies towards food security goals.

Framework for analysis
To assess if urban food governance is advancing good governance for food security, this study uses a supportive analytical framework proposed by the FAO. It is based on Human Rights principles and the United Nations Economic and Social Commission for Asia and the Pacific's (UNESCAP) definition of good governance (FAO, 2011). The framework highlights seven principles for good governance for food security: efficiency and effectiveness; equality and fairness; accountability; responsiveness; transparency; participation; and rule of law.
This framework has been selected because it is based on principles that can improve food security and help achieve the right to food, which are overarching goals of UFSs. Moreover, it includes considerations related to this study's conceptualization of good governance for food security, such as inclusive participation and multi-stakeholder interactions. The FAO's framework has been adapted to apply to the case of urban food governance (Table 1).

Data collection and analysis
The research method adopted was a case study (Yin, 2009). 3 The general unit of analysis was the UFS of Madrid, including Madrid's food policy platform, which supervises the strategy's implementation. The case study involved the collection of multiple sources of evidence and qualitative techniques for data collection such as direct observations, semi-structured interviews with key informants and document analysis (Yin, 2009). Fieldwork was conducted from April to June 2019 in Madrid. A purposive sample was used to identify key stakeholders related to the strategy (Gray, 2018). Participants were asked to suggest other key informants with a profound insight into the case, following the snowball sampling technique (Atkinson & Flint, 2011). The selection of participants was based on their participation in the strategy's development and implementation. Moreover, participants were purposively selected to provide a representative sample of the different groups involved in the UFS. The number of interviewees was seven, which, according to some scholars, is appropriate for qualitative inquiry and saturation (Creswell, 2013). Participants included representatives of four social movements with different core areas of action (agroecology, right to food and social exclusion), one city council representative, a member of an international organization (the FAO), and an urban food policy expert. All the participants are involved in the UFS of Madrid.
The interviews were semi-structured with open-ended questions, but were conducted in a flexible manner (Gray, 2018). The interview covered the development of the UFS, aspects of good governance, and barriers and facilitators. Themes covered in the interviews related directly to the analytical framework's good governance principles, such as management mechanisms, monitoring and evaluation mechanisms, and participation in the strategy's implementation and development. Documents were accessed from the county council's website and social movements that participate in the UFS. Documents that were not accessible online were provided by the interviewed participants. The strategy itself, internal documents of the municipality and social movements, and minutes of strategy meetings and reports such as the strategy's internal evaluation were included in the analysis. Data collection from direct observations was acquired from field notes of one food policy platform meeting and the 2019 annual assembly related to the strategy's evaluation.
The overall data analysis was based on a pattern-matching technique using a previously developed idealized scenario based on the good governance for food security analytical framework (Yin, 2009). For this idealized scenario, each good governance principle was allocated a set of indicators selected from the monitoring framework of the City Region Food Systems project of the FAO's Food for the Cities Program and RUAF Foundation, and the MUFPP monitoring framework for food governance (Carey & Dubbeling, 2017;FAO, 2019a).
Separate data sets from semi-structured interviews, documents and field notes were analysed using a directed content analysis technique to group information according to the predefined good governance principles of the framework described above (Hsieh & Shannon, 2005). The use of different data collection methods served as data triangulation and convergence of results, which improves the overall quality of the study (Crowe et al., 2011).
Extracts from interviews were selected as they represented recurring themes that illustrate compliance with good governance principles. Since participants continue to play an active role in Madrid's urban food governance, their names and specific positions are not identified. To maintain anonymity and ensure confidentiality, the term 'stakeholder' (S), followed by a number (e.g., S1), is used.

CASE STUDY: URBAN FOOD STRATEGY (UFS) OF THE CITY OF MADRID
The development of the UFS of Madrid is deeply embedded in the social dynamics that resulted from the 2008 financial crisis. The higher unemployment rates and austerity measures contributed to an increase in food insecurity and dependence on food assistance services, 4 but also fostered the participation of food movements in new forums aimed at wider social transformations.
These new social dynamics impacted Madrid's city governance, promoting increased collaboration between social movements and the local government. As a result, several new governance tools, such as the strategy examined in this study, were developed.
Following the pressure from social movements such as Madrid AgroEcológico (MAE), a community forum pursuing an agroecological transition, Madrid's municipality signed the MUFPP in 2015. The food policy platform ('La Mesa de Seguimiento del Pacto de Milán') was formed shortly afterward, with the challenge of developing a sustainable food system in environmental and social terms. It is mainly coordinated by the city council, but civil society and the private sector are also directly involved. Although it is formally recognized, it is still not institutionalized and officially registered. In 2017, this platform launched a process for the development of the UFS. The development of the strategy followed three steps: a diagnosis of the urban food system of Madrid, the identification of best practices for reference for possible interventions, and a participatory process to gather the opinion and suggestions of civil society and Madrid's citizens. In July 2018, the strategy was presented and included a new food governance framework and reform the previous food policy platform to include more external actors. It is worth mentioning that during fieldwork, the UFS of Madrid was reasonably unstable due to the political changes occurring at the time. 5 The strategy is based on concepts such as food sovereignty, the right to food, responsible consumption, health promotion, food waste reduction and short distribution channels. It has six lines of action, including one specifically for governance. Currently, the food policy platform includes actors from six municipal departments, two municipal bodies, the FAO Office in Spain and representatives of six social entities. New responsibilities of the platform after the development of the UFS are the coordination between local government areas and social actors; the monitoring and evaluation of the UFS; assessment of the annual report; elaboration of studies, programmes and projects; development of municipal tools; and the coordination with other municipal plans and strategies.

RESULTS BASED ON THE GOOD GOVERNANCE ASSESSMENT FRAMEWORK
In line with the FAO's framework's adapted matrix, the seven good governance principles were assessed and matched with the previously developed idealized scenario. An overview of the main findings is provided in Table 2.

Effectiveness and efficiency
Interview data acknowledged that the food policy platform is an effective and operative space of collaboration where specific actions are devised. Moreover, network management had gradually improved due to specific management tools and a technical secretary's presence. These ensured members' involvement by developing documents, meeting agendas and minutes, moderating meetings, and monitoring the agreed actions. Furthermore, the new governance framework fostered new tools that enhanced the platform's work and created more explicit rules regarding its functions, participation terms, and decision-making mechanisms.
Nevertheless, the inclusion of new food policy platform members challenged network management mechanisms and questioned how meetings ought to be conducted to ensure deliberation. While the integration of new members increased areas of action and synergies, it also set the need for better time management as meetings typically lasted about 1.5-2 hours. During this time, members took turns to report achievements and advocate for relevant causes. Considering the increased number of stakeholders, the structure of meetings hindered in-depth debates and left no knowledge exchange opportunity. Another identified shortcoming was the absence of successful mechanisms to monitor the actions that result from the platform's agreements. The document analysis showed that the strategy's monitoring process was inefficient. Municipal officials did not deliver updates promptly. Sometimes, actions were not followed up, leaving it to the interested platform members to bring up the topics and ask about updates during meetings, without any structured monitoring framework. Similarly, there was no clear definition of members' roles and expectations, which affected collaborations, especially with the increased number of members. Some of the participants noted that assigning responsibilities between the different stakeholders was needed to avoid conflict.
Moreover, according to civil society interviewees, there was no mechanism to address internal obstacles or conflicts related to the city council's adoption of actions. This delayed the implementation of the strategy's objectives and introduced unexpected problems, frustrating civil society participants. Similarly, while coherence among stakeholders improved due to the strategy, there were still some contradictions and duplications in interventions. Some participants perceived that the strategy lacked a core idea or focus, which led to incoherence in practice: At the time of translating it into actions, what they have done is to collect many of the things that were already being done and try to give some coherence. But the absence of a matrix, was noted in the following. … Well, at one point we said: Hey, here is not the law proposal for the right to food that we understand that it should be in the strategy … and they: Oh, we missed it! and so on. Here the interest is to get a nice document, coherent, but let's say an engine to carry a car, to carry a vehicle full of people, who eats, drinks and gets fat, well, I don't see it. (S3) However, interviewees also noted that the strategy is the first step in a broader system's transformation that involves a constant learning process. Advocacy for a regulation Strategy based on rights Source: Authors' own compilation from data collected for the study (2020).
The role of cities in good governance for food security: lessons from Madrid's urban food strategy Equality and fairness Regarding equality and fairness, the analysis of documents revealed that social inequalities and vulnerable collectives' needs were considered during the strategy formulation. Civil society organizations that focused on social exclusion were invited to participate during the strategy development, and specific initiatives were derived from these interactions. Similarly, one objective of the strategy was to facilitate access to sufficient, culturally appropriate, and nutritious food-for-all citizens. Consequently, the strategy aimed to have a compelling mix of policy measures and instruments that addressed equality and fairness of food access. The interviewees highlighted a general commitment towards this principle. Nevertheless, the majority noted that this area needed increased attention. According to interviewees' responses, the implemented actions under the strategy's umbrella did not significantly increase adequate food access for vulnerable communities. Due to the influence of MAEan agroecology advocacy groupsince early developments of the strategy, there had been a stronger focus on activities that promoted agroecology. Moreover, some interviewees related this to an 'invisibilization' of food insecurity in Madrid: One weakness is that there is no perception of food or the right to food as a problem. In the end they have managed to hide the problem, that is invisible, that the only visibility that has is the queues that normally are not in main streets. … People do not perceive it. (S2) There was a general perception that the lack of regard for equality and fairness would soon change due to the inclusion of a new member specializing in addressing food poverty and advocating for dignified solutions to food access. This new stakeholder's engagement was regarded as a positive step towards increasing awareness about the seriousness of food insecurity in Madrid.
In addition, although a gender perspective was included in the elaboration of the strategy and it was stated in the final document, there was no clear evidence that the strategy's actions were implemented with embedded gender equality concepts.

Accountability
With regards to accountability, monitoring and evaluating mechanisms were still not in place. No indicator framework had been developed, despite accessible food-related indicators to provide a basis for future monitoring.
According to interviewees, the municipality's leading action to assure accountability had been the annual assembly to evaluate the strategy. In the annual assembly, the city reported to civil society about the progress of the strategy. The assembly provided the opportunity for organized civil society groups to give feedback on the city council's activities. To showcase the progress of the strategy's initiatives, a memory of actions was presented during the assembly. The memory explained the actions executed based on the lines of action that the strategy had stipulated. However, as the memory was developed through the separate input of the related departments, different municipal departments measured, monitored, and evaluated the strategy's actions differently. This suggests the absence of a collective framework with meta-indicators that systematically evaluated and monitored progress, process, and impact until then. Interview data corroborated this view, showing that the platform did not include impact indicators in the strategy's assessment: It was presented as a collection of information but not as indicators … we don't have impact indicators … we want to know what effect it really has … what part of population is affected by it because otherwise, it is not as effective as it should be. But it is pending. (S4) The fact that the development of indicators was considered in the strategy, but no actions had been taken in this regard suggests that evaluation and monitoring were not seen as a legitimate part of the strategy and platform. As one interviewee stated, this could be due to an overall culture of implementing actions that prevails over evaluating the results: Here there is no culture of results. There is a culture of actions. Then, indeed, actions are done. Sometimes with a terrible degree of inefficiency. … So, I would tell you that here we have a culture of actions, and in those actions, we are effective. I mean, we do them. (S3)

Responsiveness
In urban food governance, responsiveness relates to the existence of a food emergency/resilience plan and the integration of food in other resilience and emergency plans. It also relates to having a comprehensive vulnerability assessment of the urban food system and specific actions to improve it (FAO, 2019a).
Although there is a municipal emergency plan for Madrid that briefly mentions food provision emergencies, no food emergency/resilience management plan related to the strategy was found. 6 This was corroborated by the interviewees as no specific plan was mentioned, and some even said that this aspect was never considered during the strategy development process; thinking of the idea of a food emergency due to external shocks or stressors seemed illogical for the situation of Madrid: It is not contemplated. … It has not come out in … in the participatory process of the strategy, the issue of food emergency linked to the vulnerable population was discussed, but not as a sudden explosion … . (S4) Nevertheless, one strategy's objective was to augment food sovereignty and, therefore, acknowledge the resilience challenges of Madrid's food system. Related actions aimed to raise the importance of the correct organization of the food supply and the strategic interest to increase food autonomy within the city based on the consumption of proximity products. Moreover, an assessment of the vulnerability of the food system was made before the creation of the strategy, and remedial actions were suggested and included in the action plan. Nevertheless, this diagnostic did not assess the vulnerability of Madrid's food system in case of natural disasters or crisis management of any kind.
However, as the strategy was based on a previous diagnostic of the food system that included consultations with relevant stakeholders and civil society groups, it could be argued that it proposed actions that comply with Madrid's needs. In addition, the fact that it was developed through an extensive participation process also directed to the general public suggests that the strategy was intended to converge the perceived needs found in the diagnostic with the actual needs of the population. In line with this view, interview data showed the relevance of the strategy's participatory process to ensure the collection of information about Madrid's residents' real needs.

Transparency
Since its inception, one of this food policy platform's aims was to publicize the city council's commitment to the MUFPP, and related processes, including the strategy. This was reflected through many initiatives, such as a communication campaign during the launch of the strategy. Other actions to enhance visibility included an exhaustive communication campaign about the strategy in 2018 and involvement in advertising events, radio and universities.
The quantity, quality, and access to information had improved since the introduction of the platform due to the development of a municipal blog that displayed relevant reports, events, and the city council's commitments. During data collection, a reform of the blog to include everything related to the strategy and MUFPP in digital format and improve information availability was suggested. To support the communication strategy of the municipality, civil society members of the platform also disseminated information about the strategy in their communication channels and websites. Other efforts to make the initiatives visible included developing the memory of actions and the annual assembly.
However, while the quantity, quality and access to information had improved, and there was a transparent attitude from the local government towards civil society, relevant information did not reach the general public. Thus, the strategy and governance structures were still unknown to regular citizens. Many interview participants attributed this to disinterest on the part of citizens. regarding organized civil society with an interest. … I believe that transparency is great. Regarding other actors it is less … and well sometimes it's a little more complicated. How to reach society as a whole? … well … that's complex … it's true that a lot of the public is not worried about whether the actions of the city council are made through a strategy, if there is no strategy, the organization, that is, what sometimes interests them is that the services of the city council are of quality and that they have information about them to be able to use them. (S1) Nevertheless, as noted by some interviewees, the lack of interest of the general public did not mean that visibility efforts should be ignored. Instead, it meant that efforts were not being channelled effectively and efficiently.

Participation
The general aim of the food policy platform and participatory processes of the strategy was to reach all concerned actors of the food system and create a space where civil society could co-create public policies with the municipality. As the need to work with a broader spectrum of agents was identified through these processes, the membership of the platform was increased. The new configuration of the food policy platform formed a rich pool of knowledge from the social, private, and public sectors, including nutrition, commerce, equity, social restoration, and agroecology.
Despite the assertion that the most relevant stakeholders of Madrid's food system were represented, interviewees agreed that participation processes were adequate at incorporating the views of interested civil society groups ('the usual ones') but not to include hard-to-engage groups, the general public, and communities at risk of exclusion. Yet, interviewees acknowledged that participation is a progressive process and that new members will bring new possibilities of collaboration and increased representation of different groups. For example, interview results suggest that due to the inclusion of new members, vulnerable groups were represented through an organization that works closely with marginalized communities. While there was a high level of participation in the development of the strategy, participation in the implementation of actions was limited. The strategy was perceived as a commitment of the municipality. Thus, the social sector was not highly involved in its execution, nor did it have the same power as the municipality in final decisions: Our capacity is to propose. The decisions at the end depend on … the executor and the technicians and the structure of the city council. In fact, I believe that in the platform decisions, decisions, are not taken. Issues are raised, eh … sometimes discussions are generated and moved to the corresponding spaces, but, but it's like a dialogue table, it's not one, a space, or my perception as a social movement, it's not a decisive space. (S4) However, decisions regarding the inclusion of new agents were made through consensus, with all members having the same voice. Moreover, platform meetings and the strategy's development successfully connected actors with similar interests and provided opportunities for new alliances, creating a collaborative network of key stakeholders. In this regard, participants agreed that the platform holds a horizontal structure with no hierarchies. Social agents had a prominent voice and expert opinion, which allowed them to define the process and content of policy-making. Although the role of the social sector was perceived as proposing actions, there was a constant dialogue between the city council and civil society through bilateral coordination: The food strategy contains commitments of the municipality. So, each commitment corresponds to an area of government and is perfectly identified. So, it's easy to know who to ask for accounts, right? But in many of those eh … in the execution of many of those measures is constantly counted with social organizations. (S5)

Rule of law
As a signing city of the MUFPP, Madrid committed to using an action framework based on rights. Therefore, the strategy has assumed a right to food approach within its objectives and initiatives, leading to a specific line of action that integrates initiatives dedicated to ensuring equitable access to food. Moreover, concerning a broader human rights approach, the strategy converges with the Human Rights Plan of the municipality, which recognizes the importance of the right to food to attain an adequate standard of living. Interviewed stakeholders agreed with this view: The strategy is very much supported, in … in claiming or trying to promote social economy, and in some respects also agroecology and then there is also the right to food as an example … that is, it is a rightsbased strategy, not a strategy of, I think it is, that is something important. (S4) Madrid's commitment to a rights-based action framework was demonstrated when a legislative initiative for the guarantee of the right to food in the Community of Madrid (ILM) was proposed to the regional government in 2018. Although the Assembly of the Community of Madrid did not approve the ILM, awareness and dissemination campaigns are still being carried out to address the right to food and advocate for its recognition.
Despite these efforts, no clear legal basis for various food security activities and services concentrating on the right to food was found. This may be because the city council has limited abilities to develop legislation. As a result, it can only advocate for adopting the proposed legislative initiative, given that it is the Community of Madrid's responsibility to introduce such a law. As a result, the law is still on hold, and no advances have been made in relation to this instance.

DISCUSSION
The application of FAO's adapted framework in Madrid's case shows that having established urban food governance mechanisms such as an UFS and a food policy platform does not directly ensure good governance for food security. The analysis suggests that urban food governance arrangements do not necessarily embody inclusive and democratic practices. This highlights the importance of taking a critical approach to understanding UFSs and FPCs. Despite the local specificities, this case provides additional insights regarding the role of cities in addressing food security in terms of governance.
First, Madrid's case demonstrates the importance of having a coordinating body for successful network management because it sets conditions for effective and efficient collaboration and shapes stakeholders' interactions. However, it also demonstrates that having established urban food governance mechanisms does not directly lead to coherent and uncomplicated network collaboration. Having clearly defined roles and expectations, platform monitoring instruments, tools to address blockages in the operationalization of actions, and overall strategy coherence are vital to avoid adverse effects in the coordination of urban food governance. This analysis shows that the lack of these mechanisms can frustrate stakeholders, undermining effective collaboration and the implementation of actions, as also appointed by relevant food governance scholars (Bock & Caraher, 2014;Mansfield & Mendes, 2013;Moragues-Faus & Morgan, 2015).
Second, the present case study suggests that although UFSs aim to follow equality and fairness, this is not fully realized in practice due to a narrow focus on one particular aspect of the food system and an invizibilitation of hunger. The findings demonstrate that having a strong influence of grassroots movements that advocate for a specific cause, for example, agroecology in Madrid, may lead to a failure in recognition of social and economic determinants of food access. As stressed by Harper et al. (2009), this shifts the focus of overall efforts to one sector rather than taking a systems approach on food, disregarding the multidimensionality of food security. This lack of recognition of the multiple dimensions of food security reduces the implementation of measures that enable fair access to food across population groups.
Furthermore, these case study's findings demonstrate that aligning efforts to increase transparency is not enough to go beyond the interested civil society and reach the general public. A lack of awareness among possible beneficiaries may lead to the failure of policies and initiatives to reach those who need them the most (FAO, 2011). In the same way, groups that implement food-related initiatives may not be aware of the actions of UFSs and FPCs, which may lead to overlapping, duplicating, or competing activities within the same territory and miss the window of opportunity for synergies.
Madrid's case also highlights several accountability shortcomings within urban food governance, such as a lack of formulated meta-indicators for monitoring and evaluation and a comprehensive and coherent set of process and outcome measurements. An appropriate set of indicators could enhance the knowledge that cities have about the relationships between different elements of their food system, detecting areas that need to be addressed and internal obstacles in the operationalization of UFSs. This raises an important question: Is the absence of an evaluation and monitoring framework the reason for the failures identified in the operationalization of UFSs? This is a critical aspect of urban food governance that deserves greater focus. Without clear and precise data that monitors progress and evaluates impact, the actual effect of UFSs is unknown. Thus, limited resources might be allocated to unfruitful interventions and programmes, while other areas continue to be neglected.
Another shortcoming associated with these spaces relates to the degree to which responsiveness is integrated into them. The case study revealed that food resilience as a concept had not fully matured other than with economical food access. This is extremely concerning as cities are increasingly being affected by acute crises, such as floods and the COVID-19 pandemic, and chronic stressors such as climate change, uncontrolled urban growth, and socioeconomic crises that may impact the food supply chain and result in higher food insecurity levels. The challenge for UFSs and FPCs is to become spaces of good governance for food security that foster the development of adaptive urban food systems that can respond effectively to emerging risks.
Moreover, Madrid's case demonstrates that urban food governance is not necessarily more inclusive. Vulnerable and hard-to-engage groups were not included in the food policy platform or strategy development. Consequently, the promise of urban food governance to democratize food policy-making remains unfulfilled. For urban food governance tools to develop egalitarian governance configurations based on alternative governance conceptions (McKeon, 2015), UFSs and food policy platforms need to redistribute genuine power to marginalized groups and the interested civil society. Thus, they should be characterized by creating a space where civil society can voice concerns about food, overcome barriers of public institutional structures, and reconfiguring power imbalances and exclusion (Moragues-Faus & Morgan, 2015). Indeed, urban food governance configurations may be creating deliberation spaces where the power is skewed towards already existing elites, rather than leading to more participatory and democratic food systems.
Finally, although the assessment revealed that Madrid did have a strong commitment to embedding the UFS in the right to food legislation following the rule of law, competence constraints remained an issue in ensuring a legal basis for its actions. Moreover, the instability of the UFS and the food policy platform due to Madrid's political changes demonstrates that urban food governance is contingent on political will and quite fragile during electoral cycles. A government that does not support the core principles of UFSs and FPCs can quickly stop their implementation and functioning. This suggests that although cities are emerging as relevant policy actors, reforms need to go beyond urban scales. Therefore, governance tools should promote collaboration between multilevel governance structures, transversal awareness of the multidimensionality of food, and ensure the commitment of all political parties to secure funding and support policy implementation.

CONCLUSIONS
This study raises questions about the extent to which urban food governance goes beyond technocratic processes that replicate unequal power structures (Moragues-Faus, 2017;Swyngedouw, 2005) and how effective it is in tackling food insecurity. As exemplified by Madrid's case, the transformative potential of urban food governance is intimately linked to place-based challenges and implementation mechanisms. The analysed UFS shows the importance of developing synergies and networks between diverse stakeholders. However, several actions need to be put in place to overcome the challenges of the city's limited competences, transparency, the inclusion of vulnerable groups, food system resilience, and network management.
In the light of these findings, a more substantial right to food approach with a clear legal basis that considers human rights principles such as equity and equality, non-discrimination, universality, and participation (De Schutter & Cordes, 2011) would contribute to current approaches adopted by urban food policies. Indeed, the FAO's recent framework for the Urban Food Agenda recognizes the importance of a rights-based approach and some cross-cutting principles as social inclusion and equity, resilience and sustainability, and food system (inter)connections (FAO, 2019b). This means that urban food governance must be sensitive to the needs of the most vulnerable, enable a diversity of voices in decision-making, and address issues related to the inclusion of marginalized actors to develop inclusive and sustainable food systems. The analysis presented here reveals several limitations of urban food governance when creating a supportive environment to address these concerns. As a result, the root causes of food insecurity remain unresolved.
Considering that malnutrition, inequalities, and poverty are increasing in urban areas (FAO, 2019b), this research raises the need to foster principles of good governance as the basis to address urban food insecurity. Accordingly, constant scrutiny by both academics and food policy stakeholders is imperative in understanding the current failings, and potential, for good governance of food security. The utilised framework could serve as a guideline that could be implemented to improve governance processes to achieve better food security results based on a rights approach.

DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.

FUNDING
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.