Energizing a transformation to a circular bioeconomy: mechanisms to spread, deepen and broaden initiatives

Transformative change is necessary to reach a circular bioeconomy. In this context, a variety of societal and business initiatives have emerged, which in their everyday practices aim to increase their circular impact. These emerging circular initiatives, so-called small wins, continuously spread, deepen and broaden and as such contribute to transformative change. A small win spreads by becoming larger and more numerous, deepens by becoming more radical and circular, and broadens by connecting with other themes and domains. This paper explores how mechanisms drive these processes and, more in particular, how circular initiatives stay or become more transformative while developing. By building on existing literature about mechanisms that contribute to the development of initiatives—we extend the small wins framework by focusing on the interplay between mechanisms and spreading, deepening and broadening. We applied this framework to two illustrative cases in the Netherlands: a circular laying hen farm, Kipster, and the community farming initiative, Herenboeren. We empirically explored how the interplay of mechanisms constitutes continuous transformative change. The results indicate that deepening the small win is mainly driven by learning by doing; spreading takes place through professionalization and broadening by partnering. Both case studies indicate that the energizing and logic of attraction mechanisms are key in continuous transformative change.


Introduction
The current linear economy is putting the Earth's ecosystems and human societies that depend on them at risk (Poore and Nemecek 2018;Díaz et al. 2019;Rockström et al. 2009). A transition toward a circular bioeconomy (CBE) is increasingly recognized as a promising way forward: minimizing the use of finite resources, encouraging the use of regenerative ones and preventing losses and waste (Muscat et al 2021;Desing et al. 2020;Jurgilevich et al. 2016;Korhonen et al. 2018). The importance of such a transition is also increasingly reflected in the agendas and ambitions set by the European Union such as the Green Deal, The Farm to Fork Strategy and the Circular Economy Action Plan (European Commission 2019, 2020a. The CBE combines the concepts of circular economy and bioeconomy. The concept circular economy is contested (Kircherr et al. 2017), but is most frequently depicted as a combination of reduce, reuse and recycle activities. The contribution of the bioeconomy to the circular economy is the explicit focus to replace fossil resources by renewable resources and biomass from land and water (Mak et al. 2019). Realizing a circular bioeconomy requires a rewiring of societies and their economies that goes beyond technological fixes, as these often offer only temporary and partial solutions that can even stimulate lock-ins of existing practices (Farla et al. 2012;Loorbach et al. 2017). Hence, it calls for transformative change: fundamental changes of policies, technologies, organizations, social behavior and markets (Mak et al. 2019;Muscat et al. 2021;Tura et al. 2019).

3
The promising concept of a circular bioeconomy and the urgency to fundamentally change unsustainable practices has inspired and energized many citizens, companies, NGOs or public organizations to develop innovative practices. This is reflected in the growth of circular initiatives, relatively small-scale initiatives which in their everyday practices aim to contribute to a circular bioeconomy. These emerging practices include societal initiatives, such as emerging networks on sustainable food production; business initiatives, such as plant based meat; and policy initiatives, such as the ban on plastic bags (Grin 2020;Pel et al. 2020;Potjer 2019;Termeer and Metze 2019). This variety of initiatives can be considered seeds of transformative change and positive futures (Mcphearson et al. 2021). They have the potential to create impact by transferring ideas, creating narratives or showing new practices that might influence economic paradigms or policy agenda-setting Wittmayer et al. 2019). Furthermore, initiatives themselves often aim to transcend their local context and contribute to a systemwide transformation (Bours et al. 2021;Loorbach et al. 2020).
These seeds of transformative change are also referred to as the small wins, initiatives showing concrete in-depth change (Termeer and Metze 2019). We approach transformative change as a continuous process through which emerging circular initiatives, so-called small wins, develop (Bours et al. 2021;Connor and Keil 2017;Termeer and Metze 2019;Termeer and Dewulf 2019;Termeer et al. 2017). This approach is rooted in organizational science theories on small wins (Weick 1995), continuous change (Weick and Quinn 1999) and appreciative inquiry (Bushe 2011). Small wins are defined as "concrete, complete, implemented outcomes of moderate importance" (Weick 1984, p. 43). In addition, in this paper, small wins are also characterized by in-depth change that alters current practices; positive contribution to a transformative ambition (in our case a circular bioeconomy); having overcome barriers and resistance; and connecting technical and societal change (Termeer and Metze 2019;Termeer and Dewulf 2019). In contrast to episodic change, which is defined as short periods of disrupting fundamental changes situated within long periods of stability; continuous change assumes that organizations are continuously adapting, learning and improvising through small steps (Weick and Quinn 1999). The perspective of continuous transformation is also a positive approach to change: it considers the appreciation of initiatives by their environment as a key engine of change (Bushe 2011).
A singular small win is not sufficient to constitute the desired transformation to a circular bioeconomy. Next to requiring many different small wins, transformation is achieved through the way these small wins develop. The main motivation of this research is to understand how initiatives can generate more impact by continuously re-inventing and adjusting themselves. Especially within the challenges and knowledge uncertainties of the CBE, initiatives need to continuously adapt to remain transformative. The novelty of this paper lies in studying initiatives from this perspective of continuous change. This can overcome three complexities addressed by literature which are at the root cause of why the contribution of initiatives to transformative change still remains a black box.
The first complexity is that the development of smallscale initiatives are often seen in the light of upscaling (Augenstein et al. 2020). However, spreading, deepening and broadening of initiatives are equally important as upscaling (Moore et al. 2015;Termeer and Metze 2019;Van Den Bosch and Rotmans 2008). The second complexity is that initiatives develop in a non-linear dynamic which does not neatly follow certain phases but processes such as initiating, acceleration and stagnation are taking place concurrently (Ehnert et al. 2018;Voß et al. 2006). The third complexity is the risk of initiatives losing their radicality and novelty as they scale up, limiting their contribution to transformative change (Blok and Lemmens 2015;Avelino and Wittmayer 2016;Loorbach et al. 2020). Therefore, we look into the transformative character of initiatives in relation to their development.
The concept of mechanisms offers an entry point for studying how initiatives continuously develop in a transformative way. Mechanisms comprise the multiple processes that together constitute an explanation for change or stagnation (Gerring 2008). Previous research has identified key mechanisms (von Wirth et al. 2019;Gorissen et al. 2018;Lam et al. 2020;Loorbach et al. 2020;Termeer and Metze 2019). For example, some mechanisms relate to the space for experimentation in initiatives, whereas others focus on how to institutionalize innovative practices (Termeer and Metze 2019;von Wirth et al. 2019). The main promise of mechanisms is the possibility of creating a propelling effect, which means the potential of mutual reinforcement of mechanisms that can drive change towards increasingly favorable, in this case, circular outcomes (Termeer and Dewulf 2019). The main literature gap that this paper fills is understanding how this mutual reinforcement takes place.
To address this gap we further develop the small wins framework to conceptually and empirically explore this dynamic interplay of mechanisms in spreading, broadening and deepening of small wins. Insights into these mechanisms is of interest for actors who aim to facilitate or accelerate transformative change, assuming that, if the right mechanisms get activated, small wins can deepen, broaden and spread. We explored this extended framework in two illustrative cases-the business initiative of a circular laying hen farm, Kipster, and the societal initiative of community farming, Herenboeren. We selected Kipster and Herenboeren for an in-depth study, as these initiatives both have a strong circular ambition, comply to the indicators of a small win and are rapidly developing in their own way. To analyze the development of these cases, case study descriptions were developed based on two in-depth interviews with the founders of the initiatives and complemented with documents produced by the initiatives, their websites, and two policy documents. For the analysis, we focused on the development of the initiatives, and more specifically on how and whether mechanisms shaped the deepening, broadening and spreading of the initiatives. In an iterative process, a preliminary framework of mechanisms as well as case study descriptions were discussed within an interdisciplinary research team. This led to both a further understanding of recognizing mechanisms in real-life cases as well as further conceptualization of the framework of mechanisms for transformative change.
The paper has an explorative character, both theoretically and empirically, and offers two illustrative cases. We unraveled the interactions between mechanisms and deepening, broadening and spreading to identify three main patterns in the development of initiatives. Using mechanisms as a lens to study the development of the initiatives contributes to our understanding of how the mutual reinforcement of mechanisms in initiatives can help to overcome challenges and generate more impact. In this paper, we will further conceptualize the small wins framework, and then illustrate it with the case studies, followed by a reflection in the discussion and conclusion.

Conceptual framework
The small wins framework prescribes three steps for steering transformative change: setting a provocative ambition; identifying and appreciating small wins; and activating mechanisms to upscale, broaden and deepen small wins. We further develop this last step by focusing explicitly on the dynamic interplay among mechanisms in how they drive small wins to broaden, deepen and spread as this currently lacks clarity. We do so by reviewing and incorporating additional mechanisms from other literature (see Sect. 3.2).

Spreading, deepening and broadening of small wins
In transition and innovation literature, initiatives with potential to scale up and extend beyond the local setting are commonly referred to as niches (e.g., Geels 2002; Rip and Kemp 1987). Niches are alternative practices developing within a protected space before being confronted with the dominant unsustainable practices they want to challenge. Small wins, opposed to niches, are already exposed, even though shielded by their scale, and have already overcome various legal, financial, technical or cultural barriers (Bours et al. 2021). However, not all initiatives automatically qualify as small wins and thus do not form seeds for transformative change. Furthermore, one small win by itself may seem unimportant, but it might trigger series of wins through continuous processes of experimenting, adapting, learning and reflecting (Weick and Quinn 1999;Weick 1984).
We state that through these continuous change processes small wins also spread, deepen and broaden. The pathway of upscaling initiatives is dominant in literature, but is only one way of increasing effects of a single initiative. Though upscaling is the most known way to describe the patterns of becoming larger and more numerous, we prefer the term spreading as this emphasizes the expansion of the initiative, but not the linear growth in economic or spatial scaling. Spreading implies that the initiative becomes more numerous. The initiative itself does not change but is expanded to other places or grows in members. Deepening and broadening of initiatives is just as important, but have not received much attention so far (Moore et al. 2015; Termeer and Metze 2019; Van Den Bosch and Rotmans 2008). Deepening does not focus on expanding but on offering better and more indepth solutions for the circular bioeconomy. Deepening means that the initiative intensifies and becomes more radical and more transformative. It might evolve from first to second or third order change, going beyond change of structures and cultures, and also involving change in values or identities (Termeer et al. 2017). Through broadening an initiative escalates its consequences and effects to other fields. It becomes integrated in other related sectors and agendas, which will alter the initiative itself. Hence, to stay or become more transformative, initiatives must not only spread, but also deepen, and broaden, although the order may vary.

Mechanisms
The way transformations and transitions develop and generate outcomes is considered unpredictable (Loorbach et al. 2017). Nonetheless, to analyze the development of transformations, the concept of mechanisms has been receiving increasing attention in transition and transformative change literature (Köhler et al. 2019;Zolfagharian et al. 2019). These causal mechanisms are propositions of the processes that explain certain outcomes within 'concrete systems' (Gerring 2008) such as the linear and circular bioeconomy. Much research on the role of mechanisms has focused on historical longitudinal dynamics of change, disruptive events and external shocks, and dynamics in markets and technology for innovation (Geels and Raven 2006;Bergek et al. 2008;Loorbach et al. 2017;Schot and Kanger 2018;De Oliveira et al. 2020), with much attention to so-called lock-in mechanisms, such as vested interests and past investments, which have been studied to explain sectors' path dependence (Klitkou et al. 2015).
The literature on how mechanisms trigger the development of initiatives into systemwide transformative change is growing and has resulted in various lists of identified mechanisms (von Wirth et al. 2019;Gorissen et al. 2018;Lam et al. 2020;Loorbach et al. 2020; Metze 2019) (see Appendix 1). These mechanisms are also referred to as 'propelling mechanisms', defined as "non-linear chains of events that reinforce themselves through positive feedback loops to create increasingly favourable outcomes" (Termeer and Dewulf 2019). But the working of this positive feedback needs clarification. Understanding the positive feedback can assist in grasping the dynamics and understanding how mechanisms mutually reinforce each other, as identified by various scholars (Ehnert et al. 2018;Gorissen et al. 2018;Termeer and Metze 2019).
The spreading, deepening, and broadening of small wins are not the final stages and are never completed, as learning involves continuous development by reinforcing mechanisms through positive feedback loops. Deriving from a brief literature analysis on mechanisms we arrived at a set of eight distinct mechanisms (see overview in Appendix 2). We classified these mechanisms by: (1) removing overlap between mechanisms through merging and relabeling some categories that use different words for the same mechanism; (2) removing those mechanisms that belong to development pathways, such as spreading, to have a clear conceptual distinction between mechanisms that trigger the development process and the development of an initiative itself. This resulted in the following eight mechanisms that steer the development of initiatives and their potential impact (short definitions in Appendix 2).
Learning by doing is experimentation in and through initiatives. Experimentation is considered a prerequisite for broader change von Wirth et al 2019;Fuenfschilling et al. 2019). Experimental learning and experimental governance are considered important drivers of transitions (Frantzeskaki et al. 2012;Köhler et al. 2019). Through learning by doing, opportunities and challenges are recognized and feedback is received on current strategies (Termeer and Metze 2019). Learning by doing encourages reflection on, and adaptation of, personal and other belief systems; this is important not only at the start of an initiative but throughout the whole development process (Lam et al. 2020;Loorbach et al. 2020;Termeer and Metze 2019).
Partnering is the building of networks and the coupling through which combinations of ambitions and topics across policy systems, businesses, and scales take place (Termeer and Metze 2019;Weick and Quinn 1999). Engaging in these networks makes it possible to share resources, capacities, and competences Loorbach et al. 2020). To remain transformative while partnering, it is important that partnering takes place between various relevant actors who define shared norms and interests, preferably in long-term cooperation, as otherwise the partnering could come at the cost of an initiative's initial values (Ehnert et al. 2018).
Embedding an initiative implies the adoption and integration of its design, approach, or outcomes into existing local structures (institutions, regulations, planning, agenda and ambitions) and/or communities of practice (Ehnert et al. 2018;Gorissen et al. 2018;von Wirth et al. 2019). The institutionalization and routinization of innovative practices developed by the initiatives leads to increased impact by transforming the rules and logic of higher institutional levels (Lam et al. 2020). To remain transformative, it is important that initiatives do not become co-opted and too dependent for example on specific intermediaries that translate between the public, private and civil society (Ehnert et al. 2018). This creates a risk for the ownership, continuity and resilience of initiatives (Ehnert et al. 2018).
Stabilizing is the internalization of an initiative's approach among the involved actors and the materialization of this in a stable flow of resources. This means that an initiative is becoming resilient and its continuity is secured (Lam et al. 2020). In other words, the initiative becomes robust (Termeer and Metze 2019). This may result from embedding, but it can also be that initiatives find ways to inspire stronger belief in their purposes and approaches, are able to find a stable flow of income, have human resources attached to them and so on.
Professionalization is the process of becoming better at acquiring resources and members important for further development, like developing fund-raising skills (Ehnert et al. 2018;Gorissen et al. 2018). It can also refer to the process of becoming recognized as a serious actor (instead of a temporary experiment) and attaining legitimacy and authority. This is distinct from instrumentalizing and anchoring in regulation, physical space, and funding schemes (which we consider to be part of embedding), which has also been used to define professionalization .
Logic of attraction also relates to acquiring resources, members, partners, and new opportunities for an initiative. This mechanism concerns the process of financial and human resources flowing toward winners, in this case circular initiatives considered successful (Termeer and Dewulf 2019;Weick 1984). Furthermore, this process discourages opponents and lowers political and societal resistance (Termeer and Dewulf 2019; Weick 1984). This can incentivize processes such as embedding and replicating. Unlike professionalization, this process is less intentional but derives from other positive developments of the initiative.
Replicating is the process of others starting to copy the initiative's approach and/or practices. Replicating is based on a combination of other conceptualizations describing the translation of ideas, models, and practices outside the initiative's context Lam et al. 2020;Loorbach et al. 2020;Moore et al. 2015;von Wirth et al. 2019). Termeer and Metze (2019) address the underlying psychological process involved, in which people start copying practices because others are doing this (bandwagon). We distinguish between replicating as a mechanism and spreading as an accumulation pathway. The difference lies in the first being external to the initial initiative and the latter is the scaling of the same initiative in terms of locations, members or size.
Energizing takes place when concrete and visible results inspire actors to believe that these are attainable, thereby encouraging them to look ahead for the next step (Termeer and Metze 2019;Weick 1984). It is a reassuring process of commitment, optimism, and trust in the actors involved and the results (Termeer and Metze 2019;Weick 1984). Internal motivation is vital for setting up initiatives, but it is also important to maintain motivation as initiatives face many obstacles that can damage their trust in the process.

Dynamic interplay: interacting mechanisms and positive feedback loops
Various scholars mention that mechanisms mutually reinforce each other (Ehnert et al. 2018;Gorissen et al. 2018;Termeer and Metze 2019). For example, positive results of an experiment (learning by doing) might inspire and encourage actors to develop new experiments (energizing). However, the interplay between mechanisms and their possible reinforcement through feedback loops have largely remained a black box. It also remains unclear how mechanisms constitute the spreading, deepening, and broadening of initiatives. For example, the spreading of initiatives-becoming more numerous in members or locations-can be considered an outcome derived from mechanisms such as partnering (pooling resources and knowledge) or embedding (opportunities opened by a change in the institutional context, for example in regulations) . Spreading, however, also feeds into mechanisms that lead to further development. Spreading can lead to increased visibility and recognition, thereby opening up opportunities for embedding (Ehnert et al. 2018) and the logic of attraction (more resources and partners) (Termeer and Metze 2019). The interplay of mechanisms leads to an initiative spreading, deepening, and broadening, potentially generating positive feedback through which mechanisms are strengthened.
Positive feedback loops (self-reinforcing) tend to reinforce change, while negative feedback loops (self-correcting, limiting or balancing) tend to oppose change (Senge 1991). Transformative change is about developing through, and maintaining, positive loops. One loop lays the foundation for the following one, either in spreading, deepening, or broadening of which the order may vary. This non-linear continuous change is visualized in Fig. 1, showing how interacting mechanisms constitute spreading, deepening or broadening and are reactivated by the deriving positive feedback. However, mechanisms do not necessarily lead to transformative change of initiatives. For example, it is recognized that partnering can provide new resources to an initiative, but it can also make an initiative dependent on external funding (Ehnert et al. 2018;Gorissen et al. 2018). Hence, several tensions can be identified between mechanisms and their influence on small wins. This implies that the opposite is also true: non-transformative change happens when spreading, deepening, and broadening does not provide positive feedback to the mechanisms. For example, this occurs when an initiative becomes more radical (deepening), but as a consequence the trust and commitment of the involved actors decreases (energizing), and partners start to withdraw (partnering), which endangers the continuity of the small win (stabilizing).

Transformative change in circular agriculture: the Kipster and Herenboeren cases
In this section, we present the cases of the circular laying hen farm (Kipster) and community farming (Herenboeren) as small wins and we illustrate how the dynamics of deepening, broadening and spreading manifest in practice.

Kipster as a small win
Kipster is an innovative egg production concept which started in 2013, and opened its first farm in 2017 (van Hal et al. 2019). The concept can be identified as a small win as it advocates a clear narrative for the CBE, shows concrete visible results, combines technical and social innovation and has overcome barriers and resistance. Kipster's ambition is to produce eggs with respect for animals, the farmer, and the planet, introducing 'the world's most sustainable egg'. Kipster aims to address feed-food competition by prioritizing food for human consumption and feeding its hens with feed inedible by humans, consisting mainly of leftovers from bakeries (van Hal et al. 2019). It thereby challenges the current idea of feed efficiency within the poultry sector (producing as much as possible with few inputs) and reformulates efficiency as using available resources most effectively following the concept of circularity (van Hal et al. 2019;).
Kipster combines a high level of animal welfare (the highest level, 3 stars, of the Better Life label) with producing its own electricity using solar panels, reducing particulate matter emissions to a minimum, and raising the roosters (which are normally killed as one-day-old chicks) and valorizing them in the food system. Furthermore, the farm is unique through its direct contract with a retailer to ensure a good price. Through this it demonstrates concrete results that indicate in-depth changes in which it integrates social and technological aspects. Kipster's four founders have different backgrounds and expertise. This played a crucial role in the development of the concept and overcoming barriers and resistance in terms of funding and partnering with a feed producer to develop the required feed based on leftovers. One founder is a former poultry farmer and former director of Rondeel (an earlier novel housing concept) (Klerkx et al. 2012), the second is a fourth-generation poultry farmer and responsible for daily operations, the third is an expert in communications and marketing, and the fourth brings in a wide network and experience with sustainability and social responsibility business concepts.

Spreading, deepening and broadening Kipster
Spreading The accumulation pathway of spreading can be discerned in Kipster's development: a second farm in the Netherlands has been realized. For the extension to the USA, Kipster signed a seven-year contract with Kroger to cover the demand side, and these eggs are planned to be available at the end of 2022. Further active plans are to expand to Germany, Belgium, France and the UK.
Deepening Deepening can also be recognized as an accumulation pathway in Kipster's development. For example, it is continuously learning more about animal welfare and intelligence, specifically hens, making it question whether laying hens should be kept for egg production. Consequently, Kipster is now investing in the development of a vegan egg. Moreover, reflexivity on what circularity in a globalized food system entails is part of the deepening pathway: 'What we do; feeding waste streams to animals is of course not the most efficient, that would be to prevent these waste streams in the food sector in the first place or being able to convert them directly into food for humans' (Founder, Kipster). Through recycling, Kipster creates awareness of the presence of these leftovers and draws attention to the challenges of the linear economy.
Broadening Kipster is extending circularity at the farm level with the broader chain of which it is part; this can be recognized as broadening. Kipster is involved in organizing higher levels of circularity within the food chain and retail. In its feed-producer partnerships, Kipster follows the food chain to see whether the manure of the hens can be returned to where the leftovers were produced.

Mechanisms visible in Kipster
Several mechanisms can be discerned in Kipster's development. Firstly, learning by doing and partnering can be recognized as key processes. In its design, it is actively influenced by cooperation with parties like the Dutch Society for the Protection of Animals and Wageningen University & Research (WUR). 'These are not our own ideas. We are simply really good listeners to what societal organizations and science are saying' (Founder, Kipster). Designing the concept went through trial and error, for example in finding a good feed composition and deciding on the breed, as new insights are continuously developing.
As described above, the four founders have brought in different expertise: hands-on farming, setting up sustainable businesses, organizing marketing and communications, and a broad network in the food chain. This is expressed in high levels of professionalization from the start in which Kipster projected itself as the most sustainable poultry farm. Moreover, the Kipster concept is an integrated farm with attention to key environmental, economic, and social challenges in the laying hen sector, such as circularity, feed-food competition, farm income and animal welfare. The start and development of the Kipster farms were actively picked up by the media. This was also done for strategic marketing purposes; 'We are also creating it ourselves… When people are searching online for a certain aspect, such as sustainable henhouses or white eggs, it will lead them to us' (Founder, Kipster).
Furthermore, it tries to connect with the public and advocates its story through media, its retailer, a visitor center, and the organization of tours. Through its network, partnering was possible from the beginning. That Kipster was able to do this shows how its professionalization was already developed and embodied by the founders. This resulted in signing a long-term agreement with the supermarket Lidl to cover the demand side before the farm was built. As Lidl is one of Europe's widest reaching supermarket chains, the founders wanted to cooperate with it, with an eye on future spreading possibilities. This demand-side security and a commitment within the partnership for several years reflect the stabilizing mechanism. 'First secure the demand side and from there you can start organizing the other parts of the chain' (Founder, Kipster).
This process of achieving a first step, in which commitment, energy, and trust in the concept and partners is reinforced, is captured by the mechanism of energizing. The realization and opening of the first and second farm were covered by international and national media, and Kipster won several prizes for entrepreneurship (e.g., Most Sustainable Company Netherlands 2020, Sprout Challenger 2019, Rabobank Sustainable Innovation Award 2018) and animal welfare (e.g., Corporate for Animals Award 2018, Good Egg Award 2018). The consequences of this success can be viewed as the logic of attraction, in which success attracts more partners and more resources and lowers initial barriers. Kipster initially tried to finance the first farm through a regular bank, but this was not possible within the timeframe of its contract with Lidl. The first farm was set up with private investors from within the founders' networks. After the success of the first farm however, Rabobank was willing to finance the second farm. Furthermore, other partners, besides Lidl, became involved in covering the demand side. All the partners were present at the opening of this second farm in 2020, including representatives from other sustainable enterprises, institutes, non-governmental organizations and the local mayor, showing how many want to be associated with Kipster and are attracted by its positivity.
The Dutch Minister of Agriculture, Nature, and Food Quality opened the second farm, confirming that Kipster is an example of practical execution of the circular agriculture vision developed by the minister (Ministry of Agriculture, Nature and Food Quality 2018). This aligning with a political agenda can be recognized through the mechanism of embedding. This vision on the future of agriculture within the Netherlands was developed in 2018, when Kipster and WUR were already defining the role of animals within circularity through the valorization of leftovers (Van Zanten et al. 2019). The farm's timing and vision have further established the ideas and approach advocated by Kipster.
The replicating mechanism describes the process of adopting Kipster's approach and practices in a different context. Replicating can be recognized for example in Kipster's focus on white hens (and eggs) as being more efficient in converting feed compared to brown hens and thus having a lower carbon footprint. Subsequently, in 2019, Lidl committed to selling only white eggs. Furthermore, Kipster has prioritized the use of leftover streams from the food system to increase circularity in livestock farming. This was established through its partnership with the feed company, Nijssen Granico, which took up the challenge to produce feed for hens based on leftovers, thereby showing that feed based on 95% food-system by-products is achievable. This is now available for other farmers. Kipster's drive to become more circular influences its partners to do so. In terms of animal welfare, Kipster is spreading further consciousness, for example, about the role of roosters in producing eggs. 'It is not our goal that everyone starts doing the same as us, or that there will be a 1000 Kipster farms, but about spreading the approach and vision' (Founder, Kipster). Figure 2 shows how professionalization, partnering and learning by doing are more directly linked to the spreading, broadening and deepening of Kipster. Professionalization seems to be strongly connected to the spreading. Having achieved positive results with the first farm strengthens the initiative to further spread as it becomes recognized as a serious player and attracts new partners. Then followed signing of the 7-year contract with Kroger making extension to the USA possible (logic of attraction). Lidl can be considered a constitutive force in the spreading, as it has a network in other European countries, showing the importance of partnering for expanding the field. Furthermore, signing a contract with Lidl helped to stabilize the initiative. Stabilizing through the first two farms helped setting the conditions for further development.

Dynamics between mechanisms and Kipster's spreading, deepening and broadening
As described earlier, Kipster is increasingly engaged in acknowledging the emotional intelligence of animals within the concept. Deepening can be mainly attributed to its reflective attitude toward its own practices (learning by doing), but also through doing this in cooperation and consultation with other parties from business, education, and catering (partnering). This led to its interest in developing a vegan egg. Through these steps, an energizing effect is feeding back into the other mechanisms. By creating new visible steps, Kipster can continuously inspire itself and its partners to take further steps, strengthening the commitment and optimism in the initiative and the partnerships. The creation of a vegan egg is attracting new (media) attention and strengthening Kipster's image as a changemaker. This is likely to activate the logic of attraction through which more resources or partners are attracted.
Kipster is involved in organizing more circularity within the food chain and retail. This can be considered broadening, as circularity from farm level is extended to the broader food and nutrients chain. The mechanism of partnering seems to be the main driver. Lidl supports Kipster by connecting them with Lidl's producers of bread, to facilitate the use of leftovers by the farm. The successful collaboration with Lidl has increased the mutual trust and commitment to each other (energizing) and the authority attributable to the initiative (professionalization). New partnerships are being explored with the producers of the leftovers to close the nutrient cycles by bringing manure to these soils.

Herenboeren as a small win
The initiative Herenboeren, a farming community, started in 2012. Within Herenboeren around 150-200 households make a financial commitment to start a farm and are organized in a local cooperative. Together, they make the decisions on how and what to produce on approximately 20 ha, and which and how many animals will be involved. A farmer is hired to execute these ideas and is also consulted about the feasibility of decisions. Production risks, for example bad harvests, are shared by the cooperative and not by the farmer, as he/she is paid a salary.
Herenboeren is a small win as it demonstrates concrete and in-depth changes, in which it combines alternative agricultural practices with a socially innovative community concept. The focus lies on local and sustainable food production with an explicit place for land and soil management, creating consciousness, community and a shared responsibility in food production and consumption, and providing a business model based on these values. This relates to the small win characteristic of having a clear contribution to a more sustainable and circular agricultural production. Herenboeren challenges the current role of consumers by not approaching citizens as consumers, but as members, investors and even partly producers, also called prosumers. The founders and members are continuously developing and learning about how this relationship should be organized.

Spreading, deepening and broadening Herenboeren
Spreading Currently, fourteen farms have been established in the Netherlands, and around 30-35 groups are seriously committed to establishing one, reflecting the spreading of the initiative. As a wider food movement, Herenboeren is growing and engaging in international relationships with France, Germany, and Sweden.
Deepening Herenboeren develops solutions for the barriers it encounters along the way, showing signs of deepening. Because a farmer within a Herenboerderij has a different Fig. 2 Dynamics of mechanisms and spreading, broadening and deepening within Kipster role and requires different capacities (more diverse farming skills), an educational program has been developed. Furthermore, Herenboeren is involved in research leading to a toolbox for nature-driven farmers and other initiatives.
Broadening Herenboeren is engaged in broadening to different aspects with its partners, such as research, but also land access by co-initiating a land-lease fund called Aardpeer. Furthermore, Herenboeren is striving to improve logistics, with for example electric vehicles, with the aim of reaching potential members who currently experience the visit to the farm and the high start investment as barriers.

Mechanisms visible in Herenboeren
A loan was used to realize the first farm, and 56 families were willing to invest 2000 euro each. Many members are driven by their concern about the current external effects of agricultural production. 'It is about one thing only; people standing up because they are following their gut feeling' (Founder, Herenboeren). Their belief in the movement and its results can be recognized as energizing. This is vital, as the responsibility for initiating new farms lies with the local organizers. The founders must think two steps ahead to maintain this positive energy and creativity: 'It is also about developing a concept that is simply fun to join' (Founder, Herenboeren).
The later farms were realized with a starting capital of 400,000 euro per farm, consisting of 200 households investing 2000 euros each. This shows how the realization of the first farm activated a logic of attraction through which more resources and members became involved. The first farm enjoyed a lot of visibility in the media, and a visit by the King of the Netherlands contributed to this. Logic of attraction can also explain the lowering of barriers. When the first farm did not fit in the local municipality's zoning plan, Herenboeren started to organize tours for visitors and invited the press. A newspaper article contributed to the municipality realizing how powerful this community already was and led to it being tolerated. Furthermore, through a Green Deal-an agreement for the farm in Boxtel to be exempted temporarily from certain legislation applicable to it-space was created for further development. For example, they could now experiment with keeping animals outside.
Herenboeren shows various signs of professionalization. It presents itself as a professional organization through its website, logo and other public relations content. It is on the advisory board on circular agriculture at the Ministry of Agriculture and Food Quality. Together with research institutions, a knowledge-sharing consortium has been set up. The consortium focuses on experiments with robots, drones, and the value of food forestry for food production. In this consortium, Herenboeren has set up an 80-ha experimental farm.
The Herenboeren concept is to strive to be independent from financial support such as subsidies and bank loans. Through this, the mechanism of stabilizing becomes visible. The business model is not based on profitability but on creating a long-term connection between people and the farm. The farms are established with the intention of lasting 200 years, so they must ensure continuity. They attract people who are very committed, as membership is for a minimum of three years, together with a high start-up investment. Stabilizing can be considered important to maintain energy. Currently, some existing and planned farms are still vulnerable, as they rely on specific members' energy and voluntary time investment.
Learning by doing is visible in the variation among the farms and the on-farm experiments. New farms are set up by local volunteers and adapted to the wishes and possibilities of the local context. The locations are adjusting and changing, so the Herenboeren concept remains flexible in its execution. 'The Herenboeren of today will not be the one of tomorrow, and if it is, we are doing something wrong. We want to create a movement…and consequently we have to accept that variations of Herenboeren farms will follow' (Founder, Herenboeren). In the consortium with research institutions and the toolbox, this learning by doing is professionalized.
Herenboeren is engaged in partnering by being part of co-learning networks that share a similar ambition for sustainable food production (e.g., Caring Farmers). In cooperation with BD Grondbeheer, Weiland, and Triodos Regenerative Money Center, Herenboeren set up the land-lease fund, Aardpeer. The aim is to reconnect citizens with the soil by letting them invest in it by buying shares. Herenboeren believes that soil and land need to be safeguarded and should not be subject to financial speculation. Furthermore, the Herenboeren concept is expensive and would lose its flexibility if the local cooperative had to buy the land. Dependence on foundations and municipalities is considered as slowing down the process. For these reasons, they set up a land cooperative to make it possible to buy land with Aardpeer, and the local cooperative or other initiatives sharing similar principles can lease the land.
The Herenboeren approach is showing signs of embedding; the concept is presented as a successful initiative of circular agriculture, including on the government website that gathers examples of circular farming: platformkringlooplandbouw.nl. The founder is on the circular agriculture advisory board at the Ministry of Agriculture and Food Quality. These embedding activities do not, however, necessarily lead to more funding to help to stabilize the initiative. There is support and recognition, but Herenboeren experiences resistance in opening new locations in certain areas because it is too alternative, despite the positive presentation.
It is difficult to identify whether replication is taking place. Herenboeren aims for citizens to act on their own responsibility and initiate change, whether it is a Herenboeren farm or another form. It is easier to identify volunteers espousing the Herenboeren farm idea, but this can be considered spreading as it is replicating the same initiative. Through the land-lease fund, it could become easier to replicate the Herenboeren approach and practices, as land could become more accessible for similar initiatives. Furthermore, creating a toolbox and engaging in a research consortium could inspire the spread of knowledge and tools for natureinclusive farming and advocate different business models.

Dynamics between mechanisms and spreading, deepening and broadening in Herenboeren
The spreading of Herenboeren is visible in the increasing number of realized farms and the many more interested groups across the Netherlands (see Herenboeren.nl). Professionalization is key for spreading by the increasing level of organization and the recognition of Herenboeren as an alternative for regular food production (Fig. 3). Realizing the first successful farms and receiving a lot of media attention attracted many new members and volunteers (energizing and logic of attraction).
In the Herenboeren concept, the role of the farmer has changed; decisions are made in collaboration with members, and the number and variety of crops and animals are higher. Aligning the wishes of the members, the possibilities of the local farm and the farmer (learning by doing) and sharing this knowledge on the Herenboeren Nederland level (professionalization), a new educational program was introduced. Reformulating the farmer's role is now formally organized in the educational program, in which knowledge and competences are defined and shared. This step increases trust and commitment among the involved actors, as the quality of the local initiative increases (energizing and logic of attraction), creating a further strengthening effect on learning by doing and professionalization which can trickle down to the other mechanisms. For example, it becomes more likely for the initiative to be replicated or for the knowledge to become embedded.
The broadening of the Herenboeren initiative through the Aardpeer land-lease fund can be explained by the interaction of a variety of mechanisms of which partnering seems the most crucial. Herenboeren recognized access to land as a barrier and reflected on strategies to overcome this (learning by doing). This, in combination with becoming recognized as a reliable partner (professionalization), created the possibility for cooperation and sharing resources, competences, and capacities with actors such as Triodos (partnering). Aardpeer functions as a pathway for initiatives to access land without high investments or dependence on subsidies, and

Discussion
The aim of this paper was to apply and further develop the small wins framework to understand how circular initiatives in society remain and become transformative through spreading, deepening and broadening. The addition this paper made to the framework is twofold. First, it explores the interactions within eight key mechanisms for spreading, deepening and broadening, the importance of which has been identified in other studies (Gorrison et al. 2018;Ehnert et al. 2018;Termeer and Metze 2019). Second, the paper explores the propelling effect associated with the mechanisms referred to in this paper as 'positive feedback'. These two components offered the entry points to study the development of initiatives from a dynamic perspective. In the paper, we therefore illustrated this dynamic in the cases of Herenboeren and Kipster.
The interplay between learning by doing, professionalization and partnering seems crucial for developing. Specifically, learning by doing seems quintessential for deepening, partnering for broadening and professionalization for spreading. The mechanism of stabilizing provides the foundation for an initiative's further development in deepening, broadening or spreading. The mechanisms of embedding and replicating are involved in integrating and spreading an initiative's approach, knowledge and practices.
Small wins can remain transformative when developing continuously through deepening, broadening and spreading. These cycles are sustained by the interactions of mechanisms that create deepening, broadening and spreading and the positive feedback of energizing and logic of attraction that derives from this. Our empirical findings indicate that energizing and logic of attraction function as positive feedback on other mechanisms through which the development of initiatives is sustained. This shows similarity with the work of Ola and Victoria (2021), who state that the nonlinear feedback interactions through transformative learning, in which perspectives are shifted, enhance legitimacy and empowerment of sustainability transformations.
However, the findings indicate that the mechanisms of energizing and logic of attraction are not enough to preserve the transformative change of initiatives. Their commitment, and a stream of resources (such as networks, members, financial means), needs to be stabilized in long-term funding and a secure belief in the initiative by the actors involved. This implies that when studying the development of initiatives, the dynamic of mechanisms needs to be considered instead of solely studying the presence of an individual mechanism. One of the main stabilizing factors is the organization of the two business models. Burch and Di (2021) argue that spreading and improving business models that operate within the planetary boundaries require mind set changes such as considering future generations (at the core of Herenboeren) and non-human nature (Kipster advocates for high animal welfare).
Further studies into additional relevant interactions between mechanisms and spreading, deepening and broadening are recommendable for three reasons. First, in this paper, we explored and illustrated the function of the most important mechanisms and their interactions through the development of two cases. A further investigation in a larger set of cases is necessary to establish if found patterns are also valid across cases and sectors, and will give more generic insights (Köhler et al. 2019;Gatto and Re 2021).
Second, we focused on identifying the positive feedback attached to mechanisms. We applied an appreciative inquiry approach, the focus of which lies on results and strengths and new possibilities derived (Cooperrider and Whitney 1999). This comes with the limitation that we did not explore possible negative feedback loops occurring in the development of initiatives or how mechanisms contribute to desired outcomes of the CBE for social justice (Ramcilovik-Suominen 2022). Studies into barriers to a transition to a circular economy indicate that institutional, technological, behavioral and economic barriers are multiple (Tura et al 2019). Future research would benefit from exploring these barriers in light of the absence of certain mechanisms, or the presence of negative counterparts that create negative feedback loops through which initiatives lose their impact and energy.
Third, by focusing primarily on the development and change of the initiative itself, we understand 'the landscape' from transitions theories-in which opportunities are referred to as 'political momentum' or 'disruptive shocks' (Loorbach et al. 2017) as external influences on the mechanisms that shape the initiatives. For example, initiatives become embedded as a result of political momentum or are adaptive to newly developed insights into the CBE. Given this unpredictability of change and feedback processes within the mechanisms, this paper does not claim to identify casual patterns. This relates to the question of to what extent mechanisms are unintended consequences or can be intentionally organized to direct change (von Wirth et al. 2019).
Consequently, activating mechanisms to steer toward spreading, deepening, and broadening cannot be understood as a matter of pushing the right buttons. Further exploration is necessary to increase insights into interaction patterns in order to understand how strategies can be derived to guide initiatives in decision making. Additionally, further research could focus on how involved actors can use these insights in specific interventions to influence the dynamics, in which initiatives develop. We need to better understand the complexities at work in processes of transformative change and develop more adaptive 'buttons' for initiatives and all involved actors to foster deepening, broadening and spreading of initiatives.

Conclusion
One of the key driving forces of transformative change for the CBE are circular initiatives with concrete outcomes that are referred to as 'small wins'. Alongside a variety of these small wins transformation is sustained through their further deepening, broadening and spreading.
In this paper, we applied and further developed the small wins framework in two cases, circular laying hen farm Kipster and community farming initiative Herenboeren, to illustrate how the interaction of mechanisms constitutes the spreading, deepening and broadening of these small wins.
Our study shows that focusing on the dynamic of mechanisms visible within the development of small wins provides a better understanding on how these can maintain continuous transformative change. The development of small wins in a transformative way implies the continuous (re)activation of energizing and logic of attraction by spreading, deepening and broadening, and preserving these developments. By doing this, initiatives continue to transform themselves and challenge new aspects of the linear economy. By understanding the non-linear dynamic among mechanisms and the positive feedback better, initiatives can hopefully not only maintain their energy but also speed up the overall learning process from which the CBE and all those involved can benefit.

Appendix 1
Conceptual approaches to mechanisms

Appendix 2
Origin of definitions of the eight mechanisms. The definitions for each mechanism were either literary taken from a reference (column 3), or combined and adjusted from multiple references (column 4). Additionally, a definition can relate to similar identified mechanism by other references (column 5) Funding This work was published in the context of the strategic investment theme 'Connected Circularity' of Wageningen University and Research. This work was supported by the Dutch Ministry of Agriculture, Nature and Food Quality.

Data availability
The data that support the findings of this study are available from the corresponding author, O.M. Schagen, upon reasonable request.
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/.