City networks’ power in global agri-food systems

Cities and local governments loom large on the sustainability agenda. Networks such as Fair Trade Towns International (FTT) and the Organic Cities Network aim to bring about global policy change from below. Given the new enthusiasm for local approaches, it seems relevant to ask to what extent local groups exercise power and in what form. City networks present their members as “ethical places” exercising power with, rather than power over others. The article provides an empirical analysis of the power of FTT and Organic Cities in Germany. In both cases, we found cities that are eager to emphasize their inclusive potential. Their willingness to compromise is demonstrated most illustratively by the fact that several cities are members of both networks: While the FTT campaign aims to address problems of international trade but does not abandon it, Organic Cities advocate for a new localism based on food supply from farmers in the same region. In both cases, city networks use their purchasing power to increase the share of certified products. By doing so, the city networks reproduce privileged positions of consumers benefitting from the global capitalist order (power over). However, our analysis revealed that networks also make citizens reflect upon agri-food challenges and allow developing alternatives for more sustainable systems (power with).

Cities and local governments loom large on the sustainability agenda. There is a growing body of literature on both individual cities and city networks, such as Fair Trade Towns International (FTT) and the Organic Cities Network (Filippini et al. 2019;Moragues-Faus and Sonnino 2019;Discetti 2020). Organic Cities encompass more than 200 municipalities in Europe, and FTT has over 2000 members worldwide. These members have committed to increasingly purchase from Fairtrade or Organic certified producers (Cittá del Bio 2021; FTT 2021) and are presented as "ethical places" (Malpass et al. 2007) exercising power with, rather than power over others. However, we can hardly neglect power asymmetries among actors, especially when local governments and non-governmental organizations (NGOs) cooperate with large agri-food corporations such as Lidl and Nestlé (Lyon 2014;Discetti et al. 2019). Contributing to the growing literature on city networks, we ask to what extent networks of local governments exercise power and in what form: Is joining networks a means for cities to become more powerful and promote a transition towards sustainability in global agri-food systems?
A "popular leitmotiv" has emerged over the last decade where "cities, not states, are best equipped to deal with complex problems such as climate change" (Bansard et al. 2017, p. 230). In this article, we are interested in the political power of the two city networks of FTT and Organic Cities in Germany (while acknowledging that some network members are towns, rather than cities). We use the concept of power with (cooperation and learning) to reflect upon members' self-perceptions of inclusive city. Power with is a concept that implies learning processes which allow actors to actively build up new awareness of themselves (Eyben et al. 2006;Partzsch 2017). From this theory angle, actors with a transformational orientation have substantial agency, if they 1 3 act in concert (Partzsch 2017). City networks state that such actions in solidarity are possible and that the spread of local alternatives can contribute to transitions in global systems (Cittá del Bio 2021;. However, local governments are at the bottom of the state hierarchy, and even if we do not perceive power as a zero-sum game, we need to acknowledge that cities are subordinate to structures and discourses (see e.g. Curtis 2016;Sassen 2016). Hence, in this article, we also assess the limits of power with by further reflecting upon the power of city networks along the lines of visible, hidden, invisible and unconscious power over (coercion and manipulation) (Gaventa 2006;Partzsch 2017).
As power with and power over are not mutually exclusive concepts (contradictory interpretations of the same phenomenon) but complementing each other (different aspects of change), our twofold analysis allows us to discuss their interrelations and opportunities of sustainability transitions, while not neglecting limits. Moreover, as research on city networks in the agri-food sector is scattered across disciplines and studies are often focused on single campaigns (e.g. Malpass et al. 2007;Lyon 2014;Discetti 2020), we aim to link this research to the much broader and still growing body of literature on urban politics of climate change (e.g. Bansard et al. 2017;Simpson et al. 2019;Bulkeley 2021). At the same time, by bringing together research on urban governance and city networks, we provide a qualitative analysis that systematically compares two networks in the agri-food sector.
Comparing perspectives on power for FTT and Organic Cities in Germany, in both cases we found networks that are eager to emphasize their inclusive potential. Their willingness to compromise is demonstrated most illustratively by the fact that several cities are members of both networks: While the FTT campaign aims to address problems of international trade but does not abandon it, the Organic Cities Network advocates for (organic) food supply from farmers in the same region (Biostaedte 2021;. In both cases, city networks use their purchasing power to increase the share of certified products. By doing so, we argue and confirm earlier research, they reproduce privileged positions of consumers benefitting from the global capitalist order (power over). However, our analysis revealed that networks also make citizens reflect upon agri-food challenges and allow developing alternatives for more sustainable systems (power with).

City networks and global agri-food governance
There is a growing body of research on city networks in diverse policy fields, including in the agri-food sector (Moragues-Faus and Sonnino 2019; Discetti 2020).
Scholars have documented the emergence of these diverse networks since the 1990s, especially in the context of climate change. Impressive networks in this field are, for example, the C40 network, which consists of 96 cities that produce 25% of global GDP, and the Global Covenant of Mayors, that engages over 9000 cities, representing nearly 800 million people or 10% of the global population (Bansard et al. 2017;Simpson et al. 2019;Nguyen et al. 2020). Studies have shown that local governments' action is driven by cities' need to reconfigure their infrastructure in the wake of climate change, to control carbon emission, and especially in the agri-food sector, by political claims in opposition to the dominant national interests (Moragues-Faus and Sonnino 2019; Bulkeley 2021).
In the agri-food sector, we can distinguish between cities that have urgent needs and thus are forced to focus on priorities, for example actions on food production to allow urban dwellers to grow their own food, and cities that have a more holistic approach, and choose to implement a wide range of food policy actions (Filippini et al. 2019). A much-cited example of the latter group is Milan, where the Milan Urban Food Policy Pact, an international agreement of Mayors, was adopted in 2015. With this Pact, more than 200 cities worldwide committed to develop more sustainable urban food systems by fostering city-to-city cooperation and best practices exchange (Filippini et al. 2019).
Sustainability challenges in the agri-food sector range from food waste and environmental pollution resulting from industrialized agriculture to the farm sector's role in the greenhouse effect and climate change mitigation. Food security continues to be the highest policy concern as seen in the formulation of Sustainable Development Goal (SDG) 2, to achieve "zero hunger". In many places, for example Milan, people respond to these challenges with a new localism, i.e. a return to regional food procurement (Bornemann and Weiland 2019;Filippini et al. 2019;Nink 2019). Such local activism is often related to the promotion of Organic production (Kalfagianni and Skordili 2018; Baldy and Kruse 2019). Organic standards require farmers to maintain and replenish soil fertility without the use of toxic, persistent pesticides and fertilizers. In addition, Organic farmers follow stricter standards of animal husbandry and reject genetically modified organisms (GMOs) (IFOAM 2021). Organic Cities (Città del Bio in Italian, Biostaedte in German) is a network that brings together municipalities "that share the choice of promoting Organic farming, not only intended as an agricultural model, but also as a cultural project" (Cittá del Bio 2021, p. 1). The network was founded in Italy in 2003 and now consists of 200 members all over Europe (in 2021) (Cittá del Bio 2021). In total, 21 German cities are part of the network, including major cities such as Berlin, Hamburg and Munich (Biostaedte 2021).
Another example of a city network in the agri-food sector is FTT. A group of Oxfam activists in Garstang, UK, announced their city to be the first 'Fairtrade Town' in 2000. Activist groups have raised North-South issues in international agri-food trade since the 1970s (Barratt Brown 2007). Volatile world prices of agricultural commodities such as coffee often force farmers, especially in the Global South, to sell their products under production costs. In addition, smallholders' lack of access to social infrastructure and services and, frequently, insecurity of land tenure lead to further disadvantages. Asymmetries persist due to subsidies for domestic agriculture especially in the European Union (EU) and United States, and trade distortions through colonial heritage at the nation-state and international levels (Barratt Brown 2007;Battersby and Crush 2016).
Fairtrade International, which was founded in 1997 as an international umbrella organization for Fairtrade certification, ensures a set of standards are met in the production of commodities in the Global South, including workers' rights, safer working conditions and fairer pay (Fairtrade International 2021). In difference to Organic farming, Fairtrade certification is not publicly regulated. Instead, the Fairtrade label is a registered trademark. In addition to 'Worldshops', where mostly volunteers sell Fairtrade products, supermarkets like Lidl increasingly offer Fairtrade products usually from conventional companies, which produce at least a segment of their products according to Fairtrade standards (Barratt Brown 2007; Raynolds 2017). The Garstang activists' motivation to announce their city to be a 'Fairtrade Town' was to spread Fairtrade "like wildfire through the whole country and beyond" (FTT 2021). Fairtrade International seized upon Fair Trade Town as a device, which was then formulized into a broader campaign (Malpass et al. 2007). Since 2000, more than 2,080 cities worldwide, including 723 cities in Germany, 1 have been awarded certification as Fair Trade Town/City and are now campaigning for greater purchase of Fairtrade-certified products (FTT 2021).
While we have not seen any academic publications on Organic Cities, there are a few studies on (individual) FFTs, of which the majority, directly or indirectly, raises questions of power. For example, in the context of urban and regional studies, Malpass et al. (2007, p. 638) apply an ethnographic approach to outline how Bristol used the FTT campaign and the positive notions of fair trade to deal with its negative iconography of past trading relations. This iconography included the slave trade upon which the city's wealth and infrastructure were built. The campaign sought to re-narrate the history of the city as well as of international trade (Malpass et al. 2007, p. 639). In this vein, adding a business and management perspective, Samuel et al. (2018, p. 759) argue that city networks have today become an essential element of branded marketing. At the same time, urban studies have shown how city networks' growth-oriented neoliberal strategies are in conflict with citizens' everyday practices and lived experience (Freytag et al. 2014;Samuel et al. 2018). In this vein, in particular, urban climate governance has been characterized as "strategic urbanism where climate action comes to be bound into the strategic priorities of economic and urban development" (Bulkeley 2021, p. 267).
In the context of Fairtrade, scholars have shown how through the creation of fair trade subjects of the 'Global North' and'Global South', Fairtrade certification has normalized and naturalized dichotomous power relations between 'helping' consumers and 'marginalized' producers (Naylor 2014). Some alternative local networks try to overcome such consumer-producer dichotomies, which are inherent to the global capitalist order (Lockie 2009). For example, in community supported agriculture (CSA), people contribute membership fees instead of paying a price per food item (Peuker 2015). Adding to this, Malpass et al. (2007, p. 642) explain how Fairtrade reproduces disparities even among consumers as it is associated with particular city sites "where there is a cluster of fairtrade cafes, clothes shops and alternative food shops alongside Organic, charity and alternative retailing" (Malpass et al. 2007, p. 642). With regard to urban climate actions, there is systematic analysis of the emergence of 'climate gating', where the development of off-grid, decentralized water and energy infrastructure allows those who can afford it to secure ecological and material privileges (e.g. Simpson et al. 2019). Curtis (2016) argues that states increasingly use global cities to secure the capitalist order, while ultimately reproducing inherent asymmetries (see also Sassen 2016). However, most urban and food governance scholars reject conducting debates in bipolar terms of "moral versus market" (Lyon 2014, p. 158). Instead, they emphasize the significance of civil society and new modes of governing at the local level, as it allows for "alternative ways in thinking about urban futures" (Bulkeley 2021, p. 268, see also Discetti et al. 2019;Discetti 2020). Lyon (2014, p. 158) highlights that FTTs offer "space (…) for critical dialogue" (Lyon 2014, p. 158). Cities are considered to have the ability to foster social change through community cohesion and political participation (Discetti et al. 2019; Moragues-Faus and Sonnino 2019; Discetti 2020).
In sum, when dealing with the power of cities, we should take account of local potential to effectively address global agri-food challenges, which is central to the self-perceptions of city networks. At the same time, we need to embed local efforts into a global order with large chains and asymmetries 1 3 among countries, which has even manifested within communities. In the following section, we outline our theoretical approach to systematically study the power of city networks.

Theoretical framework
On the one hand, scholars working on city networks confirm members' self-perception, or self-portrayal, that actions in solidarity are possible and that local innovations can contribute to sustainability transitions in the agri-food sector (Bornemann and Weiland 2019;Discetti 2020;Hughes et al. 2020). At the same time, across disciplines, there is a vibrant debate on the limits of such actions due to dominant actors, structures and/or discourses (Freytag et al. 2014;Lyon 2014;Simpson et al. 2019). Against this backdrop, our framework brings together the concepts of power with and power over to more systematically approach the potential of city networks. These concepts have in common that they are directly relational (in difference to concepts of power to and power within). 2 i.e. we are interested in how city networks exercise power in relation to other actors, such as superior government units and large corporations (Gaventa 2006;Partzsch 2017).
From power over perspectives, it is not possible for subordinate actors, such as local governments, to change the overall system, for example, the international order of agri-food trade. Transformational change would require an external shock or other disruptive event (e.g. food crisis affecting the powerful). Otherwise, the existing system serves the powerful to become more powerful, and subordinate actors continue to be in a subordinate position (Partzsch 2017). By contrast, the concept of power with allows for collective agency and hence transformational change. Here, the assumption is that actors deliberately exercise power with others, for example, when networks of local governments cooperate with large corporations, and that they purposely change the existing system together. In consequence, there is only a 'visible' dimension of power with. The concept is generally much less elaborated in Political Theory, compared to power over (Eyben et al. 2006;Partzsch 2017).
We use concepts of power with and power over below as analytical heuristics to cluster the various understandings of city networks' power. In other words, we use these different concepts, which are not mutually exclusive, to look at the same empirical phenomenon. While the concepts allow for different analytical perspectives and highlight various aspects of the city networks' potential for transitions to sustainability, these categories also mirror different mechanisms of relational power operating in reality. They overlap in theoretical terms and are interrelated in practical terms. We will explain how we deal with this methodologically at the end of this section.
The concept of power with builds upon Arendt's power theory (Eyben et al. 2006;Partzsch 2017). Arendt defines power positively as "the human ability not just to act but to act in concert" (Arendt 1970, p. 44). For Arendt, power is about finding common ground for action; according to her, power always refers to a group or to a collective of individuals. She is interested in finding agreement which becomes an end in itself and does not only serve the assertion of particular interests (Arendt 1970;Canovan 1978). Deliberative approaches build upon this concept of power (e.g. Dryzek 2000) (Table 1).
Going beyond the exchange of arguments, power with implies mutual learning processes that allow actors to question self-perceptions and to actively build up a new awareness of individuals or groups (Partzsch 2017). These processes happen as much in conceptual and cognitive terms as "through the senses, emotions, creativity, intuition and bodily forms of knowledge" (Eyben et al. 2006, p. 9). The result is a compromise or even win-win situations. Following an understanding of power as power with, there are no antagonisms such as top-bottom and winners-losers. Instead, we understand power (or similar concepts, such as leadership) as serving the common good (i.e. food security, environmental protection). As there are no subordinates from this power perspective, only inaction can explain persist problems (Partzsch 2017). When studying power with below, we should hence assess (see Appendix 1) what the narrated objectives of the city networks are regarding their relation to current agri-food systems. What was the motivation for starting the city network? Does mutual learning take place?
Many scholars have criticized that studies of power with neglect existing asymmetries (e.g. Brunnengräber 2017). In this vein, we also look at city networks from the four classical dimensions of visible, hidden, invisible and unconscious power over (Gaventa 2006;Partzsch 2017). The assumption here is that there is always domination and subordination in politics. Power is perceived to be a zero-sum game meaning that, if the city networks gain power, it happens at others' expense. From power over perspectives, there are always winners and losers (Partzsch 2017).
The first dimension describes the ability of actor A to "get B to do something that B would not otherwise do" (Dahl 1957). Any kind of municipal 'top-down' regulation means exercising direct, potentially visible power over others (Partzsch 2017). Following a broader understanding of politics, besides public authorities, we can also consider the visible power over of business and civil society actors based Power over Visible power Power is the potential of powerful actors to directly determine the actions of others FTT groups challenge conventional agri-food corporations (e.g. Nestlé) especially in larger municipalities (e.g. Hamburg) Organic Cities join forces against conventional agri-food business but are defeated by the latter's lobbying activities Hidden power Power manifests itself through some issues not making it onto the political agenda or being discarded before (observable) negotiations start Relevant Fairtrade share is only accomplished in cooperation with conventional agri-food business Public tenders often advantage market leaders who sell both Organic and conventional products Invisible power Power is exercised by means of influencing, forming and constituting ideas, norms and intentions FTT supports international trade and economic growth, although this is no primary motivation of participants Organic Cities pursue values such as openness, fair treatment and esteem of regional Organic food, in addition to economic growth Unconscious power Power is inherent (inscribed) in social constructions of subjectivity and individuality that are described in historical terms FTT reconstitutes subjectivities of privileged consumers as helpers in a global and local context (while farmers in the South are presented as those in need) Organic Cities reconstitute privileged positions, while producing a system not merely aiming for economic growth on their material and/or ideational resources under this first dimension. From this perspective, materially and ideationally disadvantaged actors have an ability to act (e.g. local groups), only if actors with more resources (e.g. superior government units, large corporations) allow them to do so. If people act in concert (power with), based on this understanding of power over, there is no exercise of 'real' power (as nobody acts against others) (Partzsch 2017). For instance, we would expect corporations give charity for local groups but not sharing their power and voluntarily giving up privilege. In our analysis below, we hence need to assess against whom city networks take action, and if this opposing party, or these parties, have more or less resources available to prevent a transition (see Appendix 1). The second dimension of hidden power over refers to Bachrach and Baratz's (1962) "two faces of power". They point to the fact that some issues never even make it onto the political agenda, for instance, in consequence to threats of corporations to shift investments. Studies also emphasize that businesses exercise structural hidden power over through private regulation, such as Fairtrade certification, that allows private market actors to actively set rules. As a result, power in the global political economy has been diffused, leaving non-state actors with considerable power including in the agri-food sector (Fuchs 2007;Bennett 2017). While scholars who share an understanding of power with, may interpret the inclusion of more actors as a chance for change, scholars analyzing power over emphasize that state actors at all levels are losing capacity for 'top down' regulation (e.g. Fuchs 2007). When studying our cases below, we therefore need to assess if the networks prioritize austerity and if their involvement with non-state actors limits their capacity to regulate the agri-food sector (see Appendix 1).
In a third dimension, invisible power over comes into play as a result of norms and ideas. Lukes (1974) emphasizes that certain issues are not simply hidden (second dimension), there also is a manipulation of the "very wants" (Lukes 1974, p. 23) of subordinate actors. Referring to approaches of international political economy, similar to perspectives of power with, most scholars deny an agent-based moralmarket antagonism (e.g. Lyon 2014). Instead, as outlined above, scholars emphasize that city networks like all other actors are subordinate to structures and discourses (Freytag et al. 2014;Curtis 2016;Sassen 2016). In this vein, we can assess below to which norms and ideas FTT and Organic Cities refer and whether they only serve dominant interests to protect the status quo (see Appendix 1).
Moreover, using Foucault (1982), we can capture links between knowledge, power and politics in a fourth dimension of unconscious power over (Digeser 1992). Power is understood here in a way that, in the final analysis, everything is socially constructed, including subjectivity or individuality. This fourth dimension of power over points to the role of accepted truth and knowledge regarding desirable developments, seen in dominant discourses about issues at stake. From this perspective, actors work to mainly stabilize and concretize systems and positions, for example, consumers of Fairtrade-certified products stabilizing North-South asymmetries (Naylor 2014). Thus, we study below if the city networks alter or reproduce systems and positions (see Appendix 1).
Methodologically, we applied the analytical framework illustratively by answering the questions derived above for the two cases of FTT and Organic Cities in Germany. The questions build the core of our qualitative comparative analysis; they bridge the theoretical and empirical parts of the research (Kelle 2015). We used the questions to code our empirical material manually with the software MAXQDA. For example, regarding power with, we derived the question "What was the motivation for starting the city network?" (see Appendix 1). Accordingly, we coded any passage, in which motivations were directly or indirectly mentioned (e.g. "There are more and more people who want to know how products here were produced, that they are not based on child labor and such things", see below).
Our research strategy was to first explore the power of the city networks based on literature review. Complementing the literature on urban governance and FTTs, we then used our own empirical material. This material comprises 71 policy documents (websites, self-publications, legal texts etc.) and transcripts from 18 semi-structured interviews with representatives of public authorities, businesses and NGOs involved in the networks. A first round of interviews was conducted with representatives of cities, which were FTT members (while some were Organic Cities in parallel), between June and September 2020. A second round of interviews focused on the Organic Cities Network in November and December 2020. Interviews were conducted in person (2) or via phone/online (16) and lasted between 20 min and 1 hour. Our main aim was to close information gaps regarding the questions derived above, but we also aimed to maximize potential diversity among interviewees in terms of their interests and roles in the networks. In order to have a balanced sample, interviewees also included fair trade and environmental activists who were aware but suspicious of city networks. It is necessary to maintain interviewee anonymity against the backdrop of ongoing controversies within city networks and municipalities. References to documents that we used are provided in the next section, in which we outline our results first for FFT and second for Organic Cities. Afterwards, we compare and discuss the two cases. Appendix 1 lists our analytical questions for each power dimension (coding guide), and Table 1 provides an illustrative overview of our results.

Case study 1: the fair trade towns international (FTT) network
Fairtrade International defines five admission criteria for candidate cities: (1) The local authority must pass a resolution supporting Fairtrade and agree to serve Fairtrade tea and coffee at its meetings, offices and canteens. (2) A range of Fairtrade products must be made available in a specified number of local shops and cafes so that it should be easy for local people to include Fairtrade products in their everyday shopping. (3) Fairtrade products must be used in a specified number of local workplaces and community organizations.
(4) Significant media coverage and popular support must be attracted for the campaign, and a strategy must be developed to keep the campaign in the news. (5) A local Fairtrade steering group must be established to ensure continued commitment to the FTT status. This steering group consists of public actors, local businesses and NGOs (FTT 2021).
We found that these steering groups' self-perception very much follows an understanding of power with. They emphasize that network membership is "the result of civil society efforts (interview with NGO representative, 1 July 2020)". 3 The incorporation of the public authority as integral to the campaign strategy means that the formal authority of political jurisdiction is combined with the voluntary engagements of individuals, NGOs and private companies. Adding to this, an interviewee explained: "We support the value of making Fairtrade better known. But for me, this (campaign) is primarily only a means of reaching out to the public (interview with NGO representative, 19 June 2020)." 4 A public representative endorsed that "more and more people […] want to know how products here were produced, that they are not based on child labor and such things (interview, 21 September 2020)". 5 Such quotes illustrate that FTT does not merely promote sells of Fairtrade-certified products. However, the campaign's central aim is the promotion of trade according to the standards defined by Fairtrade International. Learning is hence one-directional, rather than mutual, and to this effect, FFT does not necessarily illustrate power with.
FTT exercises visible power over others if their success comes at the expenses of producers without Fairtrade certification. An interviewee told us that, "there are also many critics from the agricultural production sector who see conventional agriculture as disadvantaged (interview with public representative, 6 August 2020)". 6 Local groups are able to mobilize enough resources to challenge conventional business, in particular, in larger municipalities, including those locating Nestlé branches (e.g. Düsseldorf, Hamburg). However, smaller municipalities that locate Nestlé branches are not FTT members (FTT 2021). Hence, here, conventional corporations do not share their power and visible asymmetries persist (see also Discetti et al. 2019).
At the same time, studying hidden power over reveals that conventional agri-food business is deeply intertwined with the FTT campaign. In particular, we found FTT groups to indicate their 'success' in terms of the share of Fairtrade in public procurement (e.g. Erfurt 2021), neglecting issues beyond councils' agendas (e.g. trade distortions due to EU subsidies). As mentioned above, Fairtrade certification is increasingly offered by conventional agri-food industry and retailers (Barratt Brown 2007), and in order to meet the second and third FTT admission criterion, steering groups in most municipalities include corporate retailers on their lists. A public representative told us that "especially the Worldshops, some of which have higher standards, do not like the fact that we work together with Lidl and other big companies. And they argue sometimes that this makes everything untrustworthy (interview, 26 June 2020)." 7 Hence, while FTT visibly turns against existing trade practices, the network needs conventional corporations in order to be 'successful'.
Moreover, FTT exercises invisible power over by influencing the "very wants" of citizens, i.e. making them want to live in "ethical places" (Malpass et al. 2007, p. 634), and FTT supporters themselves are also subject to dominant norms and ideas of international trade. In particular, we found Fairtrade narratives used for city marketing to be very much embedded in growth-oriented neoliberal strategies (e.g. Erfurt 2021; FTT Leipzig 2021). However, our interviewees did not confirm that participation in the city's FTT campaign was primarily driven by self-marketing interests: "To put it frankly, the campaign is not the big thing here in Freiburg (…), shops would sell their products anyway, with or without the Fair Trade Town campaign. (…) I would not say that sales increase strongly due to the campaign (interview with public representative of the city, 26 June 2020)." 8 Lastly, there is unconscious power over, if FTT supporters understand themselves as part of movements resisting an unjust global agri-food system (see e.g. Erfurt 2021; FTT Leipzig 2021), while they mainly reproduce their consumer position and hence the system as such (see also Naylor 2014). We found FTT supporters to describe themselves as in a privileged consumer position acting as helpers, whereas producers appeared to be underprivileged and in need of their help (FTT 2021). By emphasizing that not everybody is able to help, Fairtrade reconstitutes privileged positions in a global as well as a local community context (see also Malpass et al. 2007): "Basically one shifts all the responsibility to the consumers, while knowing that only a few care about where products come from after all (interview with NGO representative, 19 June 2020)." 9 In sum, we can see that the FTT network indeed helped the diffusion of Fairtrade. Developing over time, the campaign has increasingly incorporated conventional businesses. While FTT has helped raise awareness of global injustices resulting from conventional trade, the campaign does not completely turn against conventional agri-food systems and the inherent North-South divide. Hence, the network demonstrates power with, while still showing aspects of power over.

Case study 2: the organic cities network
Our second case is Organic Cities in Germany. Founding members developed a cooperation agreement, which municipalities need to sign in order to become members. This agreement specifies the objective of the network to support Organic agriculture, processing and demand for Organic food with short transportation routes and regional value creation. In the medium-term, members also want to increasingly support the use of other fairly traded Organic products, for example textiles and natural cosmetics, if possible, with short transportation routes and regional value creation (Biostaedte 2013).
In general, Organic Cities is a project-based network, i.e. either cities join existing projects by replicating initiatives, or they initiate a new project and diffuse the respective idea through the network. Such projects deal with increasing the Organic share in public procurement (including guidelines for public contracting), Organic catering in public institutions (including kindergartens, schools etc.), development of regional networks, and use of municipal land for food production. The network also helps member cities to get funding for their projects (Biostaedte 2021).
We found that an understanding of power with in the sense of acting in concert prevails in all documents (e.g. Stadt Lauf a.d. Pegnitz 2019;Augsburg 2021;Biostaedte 2021). Network members cooperate to find "common (…) strategies and approaches (Biostaedte 2013, p. 3)". 10 Interviewees confirmed processes of mutual learning among public authorities, companies and NGOs, especially, in the context of the bi-annual network meetings (interviews with public representatives, 1, 3, and 4 December 2020). The network does not frame the members' commitment in antagonisms such as winners-losers. An interviewee stated: "We do not turn against anyone; instead, we offer opportunities for participation and address those who want to cooperate with us, (…) everyone who wants to contribute is welcome, we do not exclude anyone (interview with public representative, 3 December 2020)." 11 Interviewees highlighted that benefits are not limited to members. For example, they mentioned that Organic agriculture mitigates climate change (interview with public representative, 1 December 2020) and contributes to environmental protection (interview with public representative, 8 December 2020). Several interviewees also mentioned how the network made them reflect upon their position in agri-food systems, illustrated by this statement, among others: "I have somehow started reflecting about political contexts, just that it takes so many people who move things, and (…) it takes huge efforts. One often says: 'The public authority needs to finally change these things now.' However, it's not only a task of the public authority. Instead, so many groups are involved there, and actually, all have to take action, and this changed me a bit. Also, as citizens, you can just do it (interview with public representative, 4 December 2020)." 12 10 Original in German: "gemeinsame […] Strategien und Lösungsansätze ". 11 Original in German: "Also wir wenden uns gegen niemanden, sondern wir bieten Beteiligungsmöglichkeiten an und sprechen diejenigen, die mit uns zusammenarbeiten möchten, an, […] alle, die sich aktiv einbringen möchten sind willkommen, wir grenzen da niemanden aus." 12 Original in German: "[I]ch habe irgendwie angefangen anders über […] politische Kontexte nachzudenken, einfach auch, ja, dass es einfach super viele Menschen auch braucht, die so was bewegen und […] dass es eine riesen große Leistung ist. Und man sagt immer so "ja das muss der Staat jetzt aber mal ändern" und das ist halt nicht nur Aufgabe des Staates, sondern es sind so viele Gruppen da involviert While such statements generally neglect power asymmetries preventing people from "just doing it" and, at least implicitly, inaction is seen as the only reason for the persistence of sustainability problems, we find visible power over others too. All interviewees tacitly saw the network turning against conventional agri-food business (interviews with public representatives, 10 and 11 December 2020). In particular, a public representative explained how the network helps to put "soft pressure" onto city administrations to adapt executive proposals in favor of Organic producers (interview, 10 December 2020). Another interviewee stated that, "one really needs to approach farmers and convince them to produce organically (interview with public representative, 17 December 2020)". 13 However, another interviewee made explicit that the network allows "conducting a kind of lobbying at superior levels (interview with public representative, 10 December 2020)". 14 In this vein, we were told that the city network has more "political weight" compared to a single municipality (interview with public representative, 3 December 2020), while another interviewee admitted that still, in comparison, the conventional agri-food industry has fundamentally more influence on decision-makers at superior government levels (interview with public representative, 10 December 2020).
The situation is similar with regard to hidden power over. The Organic Cities network advises members on how to formulate calls for public tender in a way that they advantage Organic suppliers "with short transportation routes" (Biostaedte 2021). In this vein, for example, a member city emphasizes the purchasing power of the public sector on its homepage, and explains how public procurement, which accounts for 19% of German GDP, can incite sustainable markets by providing consistent demand (Erfurt 2021). However, several interviewees mentioned cases in which market leaders, which often offer both Organic and non-Organic, were advantaged over small businesses focused on Organic when competing for public tenders (interviews with public representatives, 27 November 2020, 10 and 11 December 2020).
In terms of invisible power over, the network pursues values different from the conventional system, based on Organic standards (no pesticides etc.). Interviewees named values such as "openness", "fair treatment" and "esteem of regional Organic food" (interviews with public representatives, 3, 10 and 17 December 2020), and confirmed that public tenders make many farmers produce organically, despite having refused respective values before (interview with public representative, 3 December 2020). At the same time, again, we found narratives used for city marketing to be very much embedded in growth-oriented neoliberal strategies (e.g. Erfurt 2021; FTT Leipzig 2021). The network refers to conventional discourses when emphasizing that "(t)he Organic industry is a growth industry with excellent economic perspectives" (Biostaedte 2013, p. 2)." 15 The network's homepage describes the Organic sector as a "lucrative field of business development" (Biostaedte 2021, see also e.g. Bremen 2021. Interviewees agreed that network membership serves the image of their municipality and makes them more attractive to young people and families who can afford to buy Organic (interviews with public representatives, 1 and 3 December 2020). In consequence, the network helps members to entice additional economic investments (Hamburg 2016; interview with public representative, 4 December 2020). Such considerations, repeatedly voiced in our interviews, illustrate that the network does not turn against current market paradigms. Instead, like in the case of FTT, network membership has become an essential element of branded city marketing (see also Samuel et al. 2018). Applying a perspective of unconscious power over, we can see how the Organic Cities' marketing then reconstitutes privileged positions of those families able to afford living in respective municipalities (see also Simpson et al. 2019). However, as the network also promotes values others than economic growth in the agri-food sector, families actively moving to cities which are members also oppose systems in which such values do not matter. Hence, they contribute to more sustainable agrifood systems at least in some regard (e.g. regional supply).
In sum, although the Organic Cities Network emphasizes its inclusive character, it generally opposes conventional agri-food business. When studying its visible and hidden power over, we found that local supporters of Organic agriculture could at least occasionally organize respective means to overthrow conventional business. In terms of invisible and unconscious power, we can see that, although campaigns are embedded in growth-oriented neoliberal strategies and reconstitute privileged positions, they produce an alternative system based on different values.

Comparison and discussion
Our study very much demonstrates city networks' self-portrayal of exercising power with, rather than over or against others. We have seen that, while Oxfam activists founded the first FTT and the network is run by Fairtrade International, the Organic Cities network was founded by local governments. In the first case, cities are awarded membership by Fairtrade, i.e. from an outside non-state organization. The main purpose is to promote Fairtrade-certified products. In the second case, local governments are active members deciding themselves about membership conditions, and only in this second case, we could record mutual learning between public authorities, business and NGOs. However, on the ground, there is great willingness to cooperate and compromise also in the case of FTT. In this vein, the networks have much in common, and we found earlier studies confirmed emphasizing modes of cooperative governing at the municipal level (Lyon 2014;Discetti 2020;Bulkeley 2021).
The willingness to compromise is demonstrated most illustratively by the fact that several cities are members of both FTT and the Organic Cities Network: Whereas the latter advocates for Organic food supply from farmers in the same region (Biostaedte 2021), the FTT campaign aims to address problems of international trade, but does not abandon it (FTT 2021). In consequence, on the one hand, local authorities' preference of Fairtrade procurement has resulted in greater offer of imports in public institutions (see also Malpass et al. 2007). On the other hand, there are strong pressures for local authorities to give preference to regional suppliers, suggesting a contraction away from global sourcing. Such contradictions will ultimately lead to failure of implementation, the more successful city networks are in promoting their specific aims.
While local groups generally demonstrate willingness to compromise, both city networks exercise visible power against conventional agri-food business. They both have formed in opposition to dominant interests and have accomplished to mobilize considerable resources for change, especially in larger municipalities. However, conventional industry continues to have more resources available to influence decision-making. Depending on the perspective, by not preventing cities from becoming network members, large corporations are either voluntarily sharing their power (power with) or are rather doing a form of charity in response to growing demands for change (power over). From the latter perspective, city networks actually serve to prevent transformative change in global agri-food systems.
Our analysis of hidden power over reveals that local groups often depend on conventional business, offering both certified and non-certified products, to accomplish FTT admission criteria (i.e. availability and use of Fairtrade products) as well as ambitions of Organic public procurement. The fact that networks cooperate with conventional businesses supports networks' self-perception of being inclusive (power with). However, confirming earlier studies (e.g. Lyon 2014), we need to see that economic incentives provided by the steering groups' marketing and municipal tenders tend to disproportionally benefit larger corporations and hence to reproduce asymmetries in a hidden way (power over). Therefore, again, this indicates that networks prevent transformative change.
Moreover, analyzing invisible power over, we found that the city networks are subordinate to growth-oriented neoliberal strategies, which also serve dominant actors. Members use campaigns for their city marketing, as already shown by other research on city networks (Malpass et al. 2007;Samuel et al. 2018). At the same time, FTT participants told us that increase in sells was not their primary motivation to join activities, and although the Organic Cities Network is explicit about aiming for economic growth, it also pursues values such as fair treatment and esteem of regional food, which conflict with conventional approaches. Hence, in terms of unconscious power over, on the one hand, we found a reconstitution of privileged positions confirming earlier studies (Naylor 2014;Simpson et al. 2019). On the other hand, our multidimensional power analysis reveals that additional values alter agri-food systems, at least in some regard. Thus, city networks may secure a global capitalist order, as argued by Curtis (2016), but not without reform.

Conclusions and outlook
Agri-food challenges range from "zero hunger" to the farm sector's role in climate change mitigation. As local municipalities, rather than superior government units, are increasingly seen to be best equipped to deal with such complex problems (Bansard et al. 2017;Bulkeley 2021;Hughes et al. 2020), we studied FTT and Organic Cities to better understand how city networks exercise power, and whether they can contribute to global policy change. So, is joining networks a means for cities to become more powerful and promote a transition towards sustainability?
A higher share of Fairtrade-and Organic-certified produce signifies a visible change and respective power dynamics among agri-food actors (Biostaedte 2021;. At the same time, as diverse authors have argued, the conventional industry is increasingly part of the Fairtrade and Organic sector and, as we show for both networks, directly benefits from activities that were originally driven by its opponents. Moreover, both city networks focus on purchasing power, instead of exhausting other policy instruments ('top down' regulation, nudging etc.). By doing so, they confirm earlier studies (e.g. Naylor 2014; Curtis 2016) that argued that city networks reproduce a consumer-producer dichotomy and hence a global capitalist order, which again advantages large corporations. In addition, also endorsing earlier studies (Freytag et al. 2014;Samuel et al. 2018), we found local governments strategically using their network membership to attract young people and compete for investments against other cities.
Therefore, on the one hand, our findings confirm for agri-food city networks what diverse scholars have shown for urban climate networks and FTTs: These initiatives are not free from power over. This means that dominant actors, structures and/or discourses reproduce asymmetries and prevent transformational change. However, concurrently, we cannot neglect that local initiatives exercise power with and that they have a transformational impact on the ground based on their agency.
Our analysis revealed that the promotion of Fairtrade and Organic supply makes citizens reflect upon agri-food systems. In this vein, city networks might not necessarily oppose global asymmetries per se, especially if cooperating with corporations such as Lidl and Nestlé. However, as problems of the conventional agri-food systems become more and more visible, citizens face conflicts in their everyday life and question economic structures and discursive paradigms. The city networks make suggestions for reform. Both networks take a stance on what sustainability transitions can mean, for example, retaining soil fertility without the use of toxic pesticides, and empowerment of Southern producers. By discovering their purchasing power, local governments become increasingly creative and rediscover their legal scope of action, for example, by advantaging regional producers in public tenders and allocating municipal land to Organic farmers. Following research on urban climate actions, we hope that this article encourages more research on such local actions for global policy change in the agrifood sector.

Appendix 1
Power conceptions and their appliance to city networks To which norms and ideas do the city networks refer (e.g. economic growth)? Are these norms and ideas different from dominant ones in agri-food systems? Unconscious power Power is inherent (inscribed) in social constructions of subjectivity and individuality that are described in historical terms Do city networks alter or reproduce systems and positions (e.g. global capitalist order with producers in the Global North and producers in the Global South)?

Source: Authors
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.