Top-down versus bottom-up pathways to collaboration between governments and citizens: reﬂecting on different participation traps

versus pathways between


Introduction
Collaboration in policy-making and service delivery between citizens and government is influenced and can be provoked by both citizens and government. In this chapter, the focus is explicitly on the interaction with citizens, who are either individually participating or in more or less organized groups (e.g. in associative forms). An explicit distinction between bottom-up and top-down pathways to collaboration is made. While there is a lot of literature on top-down, government initiated forms of collaboration, literature on bottom-up pathways is more diffused and fragmented, while it receives increasing attention (Brandsen et al., 2017;Edelenbos and Van Meerkerk, 2016;Igalla et al., forthcoming). The aim of this chapter is to systematically reflect on these pathways: bringing together these different strands in the literature. It is argued that these pathways have their own dynamics and often reflect different objectives. Therefore, it is crucial to take notice of these different pathways as they give rise to different kind of implications and challenges for both governments and citizens in collaborative efforts in public policy-making and service delivery. Typical participation traps are pointed out for each of these pathways (cf. Howlett and Ramesh, 2016).
At the same time it should be stressed that both pathways can be highly related: for instance, consulting citizens on policy proposals (as a form of top-down initiated collaboration) can provoke citizens to join efforts and to come up with their own initiative, challenging governmental-led policymaking. For analytical purposes it is useful though to make this distinction, to better capture different ways towards collaboration and their specific challenges, both from a governmental perspective as well as from the perspective of citizens. They give different meaning to the concept and practice of collaboration.
Top-down pathways of collaboration between governments and citizens can take different forms of which several are discussed in the introduction chapter of this book, like consultation, citizen participation and co-production. Crucial is here is the initiating role and dominance of governmental actors in involving citizens in policy making or service delivery. While not necessarily the case, citizens are often involved at an individual level. Especially in co-production when individual service beneficiaries are involved in individual transactions yielding mostly private value (Bovaird et al., 2015). Bottom-up pathways, in contrast, are characterized by citizens taking the lead in formulating and/or generating community goods and services. Crucial here is the self-organizing and initiating role of citizens (see section 3). These bottom-up initiatives have a group or collective nature and are (often) initiated by groups of citizens in the local community. Both forms can lead to more or less intensive ways of collaboration with government institutions. For instance, a bottom-up initiative of citizens to develop and manage community assets in urban regeneration can result in comanagement (see chapter 1) of these assets and of related public spaces (e.g. Van Meerkerk et al., 2013). However they can also lead to forms of counter-production in which collaboration is hampered or in which interaction between governments and citizens is characterized by conflict or competing policy proposals Gofen, 2015;Kleinhans, 2017).
In the next section these two pathways are fleshed out, one by one. First, both forms will be more elaborated, taking notice of different drivers, forms and characteristics. Next, different issues and challenges in relation to both pathways will be discussed. Differences and similarities are explored between these pathways. The discussion is structured by going deeper into the why, how and who questions of bottom-up and top-down participatory efforts of governments and citizens, as these pathways are shaped by partly overlapping, but also differentiating motives (why?), crucial capacities (how?) and trigger different types of citizens (who?). Key participation traps in the pathways are distinguished. In the last section these two pathways will be compared and conclusions will be drawn, going deeper into the implications for future research.

Top-down pathways to collaboration between governments and citizens
There are different top-down pathways to collaboration between governments and citizens. Top-down pathways refer to arrangements which are organized by governments to include citizens directly in the formulation or implementation of public policy and public service delivery. Following the typology in chapter 1, this include consulting and co-production arrangements (though co-production can also be an outcome of bottom-up pathways). Although the notion of collaborative arrangements might suggest some clear institutional design, in practice collaboration between governments and citizens is not necessarily experienced as such, comes about in different ways and forms and is driven by different motives. In this section, we go deeper into specific characteristics of top-down pathways to collaboration, driving forces and typical governance issues in managing these arrangements.
To dig deeper into top-down pathways to collaboration between governments and citizens we will build on different streams of literature. A first stream of literature in this respect concerns the vast amount of literature on citizen participation and related literature on stakeholder participation. Citizen participation is approached as a process in which individuals take part in decision making in the institutions and programs that affect them (Florin & Wandersman, 1990). Efforts of governments to include citizens more directly in the formulation or implementation of public policy and service delivery can take various forms, including citizen panels, citizen juries, citizen charters, and participatory planning (e.g., Dryzek, 2010;Lowndes, Pratchett, & Stoker, 2001;Roberts, 2004). Such government-led organized forms of participation has become a significant policy strategy in many Western countries. It has been put, for example, at the heart of spatial planning, environmental decision-making, social care, regeneration, housing and education policies (e.g. Irvin and Stansbury, 2004;Marinetto, 2003;Roberts, 2004;Reed, 2008).
A second stream of literature relevant for top-down pathways concerns the literature on coproduction. Co-production of public goods and services come in different shapes and sizes. Voorberg et al (2015) have found that in general three different types can be distinguished: (1) citizen as co-implementer, (2) citizen as co-designer, and (3) citizen as initiator of public services. The first type is the most frequently found, whereas the third type is reported the least (and which is part of the bottom-up pathway). Some authors, like Bovaird et al. (2015: 3), define co-production broadly as "making better use of each other's assets, resources and contributions to achieve better outcomes or improved efficiency". This includes collective and community initiated forms of co-production (the third type) in which long term relationships with professionalized service providers are developed (Bovaird, 2007). Other co-production literature define co-production in a more narrow way. Based upon recurring elements in the literature, Brandsen and Honingh (2016) come with the following definition: "Coproduction is a relationship between a paid employee of an organization and (groups of) individual citizens that requires a direct and active contribution from these citizens to the work of the organization" (Brandsen and Honingh, 2016: 431). The last part indicates that much literature on coproduction mainly emphasizes the engagement from citizens to contribute to the work of already initiated and institutionalized forms and projects of public service delivery. Much literature on co-production therefore concerns top-down pathways of collaboration: the implementation of programs and projects are mostly induced and controlled by governments and other professionalized institutions.
A core characteristic of top-down pathways to collaboration is that governments decide when, who and how people get involved . At certain moments in decision-making governments initiate participation procedures in which citizens get a chance to respond to or provide input for decision-making highly structured by rules set by the government. With regard to the how and who, typical issues emerge in top-down pathways. Before going deeper into these 'participation traps', it is first important to address why governments want to invest in such collaborative efforts.

Reflecting on different top-down participation traps
Why Different motives for starting (top-down) citizen participation can be found in the literature. These motives are important to consider as they can explain strategies used as well as issues emerging in top-down pathways. For governments, key motives are increasing support for policies and governmental decisions, enhancing the quality of services and decisions made, breaking through deadlocks, avoiding litigation costs and building strategic alliances (Irvin and Stansbury, 2007;Roberts, 2004;Reed, 2008).
While the idea of enhancing the quality of decisions is an important and frequently mentioned motive, the need for obtaining public acceptance is often a more powerful motive for governments to invest in participatory efforts. Irvin and Stansbury put it even stronger (2004: 57): "Some citizen-participation programs primarily serve a marketing purpose, where the participation process consists of government representatives guiding citizens toward decisions the administrator would have made in the first place". This is an important participation trap, as such participatory efforts easily run the risk of becoming symbolic forms of participation in which citizens don't have real influence. Inviting citizens to contribute to the formulations and implementation of public policy raise expectations about taking input and efforts seriously. Participating citizens expect that their input (values, interest, worries, etc.) are taken seriously into account. When these expectations are not met this can negatively impact on trust in government (Irvin and Stansbury, 2004;Edelenbos and Van Meerkerk, 2016). Some evaluative studies on relatively new participatory efforts (such as mini-publics and citizen juries) show that the impact of citizen input on policy-and decisionmaking is often not clear, but rather diffusive (Jacquet, 2017;Font and Blanco, 2007). We come back to this in the how question.
For citizens, important motives to collaborate in participatory efforts are also diverse, which are likely to vary given the specific form and context of engagement (e.g. advisory council, citizen jury, community planning), although empirical research in comparing such motivations is scarce. In general, an important motivation for citizens to engage is the opportunity offered to influence service design, delivery and/or other types of decision-making outcomes. Specifically, the stakeholder literature on participation stress the influence on policy and implementation decisions as an important motivation (see Reed, 2008). Also more altruistic motives are mentioned in the literature, such as contributing to the common good (Alford, 2012;Van Eijk and Steen, 2016).
Important reasons for non-participation are negative perceptions about self-efficacy and capacities (e.g. perceived lack of political competence and expertise regarding the specific issue), but also negative perceptions about the meaning and impact of participatory efforts (Lowndes et al., 2006;Jacquet, 2017). This underlines the importance of making clear and transparent what the aim of co-production or other participatory efforts are and how input is processed. Furthermore, it stresses the need for providing good information and support, making participation convenient and participants better able to participate. Both aspects relate strongly to the 'how' question of top-down participatory efforts.

How
Recurrent issues with regard to the management of participatory efforts are related with how citizens are involved. Important here is to consider the extent in which citizens get room for participation and collaboration and how this is organized. The famous ladder of Arnstein (1969) is often used to label the extent of participation, ranging from manipulation (as symbolic participation) to one-way consultation (as a form of tokenism) towards two-way collaboration and citizen control (at the other side of the spectrum). Authorities often lack experience and abilities with participatory efforts and/or fear losing power, leading to government-citizen collaboration often remained limited to providing information or consultation (Leach and Pelkey, 2001;Videira et al, 2006;Tatenhove et al, 2010). In this respect, top-down pathways generally include the lower parts of Arnstein's rung, ranging from manipulation to forms of one-way consultation, while partnerships, delegated power and citizen control (the top-rungs of Arnstein's ladder) are much harder to achieve in practice (Reed, 2008).
While the ladder of Arnstein is static and disregards the complexities of collaborative and interactive governance processes it points at the issue of whether citizens are truly empowered in top-down participation by which meaningful interaction can emerge (see also chapter Johnson and Howsam). Various literature on citizen participation stresses the importance of matching expectations of participatory efforts and the methods and strategies used to put participation into practice (Tritter and McCallum, 2006). Other literature goes a step further and stresses the point that participants need to be empowered before meaningful participation can occur at all. This empowerment takes two forms (Reed, 2008): 1) ensuring that participants have the power to really influence the decision and 2) ensuring that participants have the capability to engage effectively with the decision.
While not having the room to be exhaustive, several key factors and traps in fulfilling expectations can be mentioned. Key factors in empowering participants are how governments organize participatory processes and how responsive they are to citizens' input. Research shows that especially this response is one of the biggest deterrents (Lowndes et al., 2006). A lack of institutional linkages between existing power structures and participatory spaces is often pinpointed as a crucial barrier here (Edelenbos, 2005;Healey, 2006). Participatory forums often run parallel to formal decision-making processes hampering real meaning and impact. Political and managerial competences (see chapter 1) are particularly needed to prevent this participation trap. Particular, at the organizational capabilities, like creating communicative and procedural linkages between formal decision-making structures (e.g. within the local Council, within administrative project and expert groups) and participatory spaces. A crucial task for managers organizing participation efforts is to make participation transparent and to ensure feedback mechanisms about how input is processed.
For the more 'ambitious' participatory efforts aimed at two-way communication and deliberation between governments and citizens, the idea of developing an authentic dialogue and due deliberation is stressed by scholars often building on deliberative democracy theory (Dryzek, 2010;Van Meerkerk et al., 2015). This includes the idea that problem definitions and goals of participation are deliberatively discussed with participants. This requires important managerial competences, like skilful facilitation of informal group discussions, including maintenance of "…positive group dynamics, handling dominating or offensive individuals, encourage participants to question assumptions and re-evaluate entrenched positions" (Reed, 2008(Reed, : 2425. According to the review of Reed (2008) among stakeholder participation literature in environmental governance, citizens typically get involved in decision-making at the implementation phase of the project cycle, and not in earlier project identification and preparation phases. Engagement with stakeholders as early as possible in decision-making has been frequently cited as essential if participatory processes are to lead to high quality and more meaningful decisions (ibid). Typically, stakeholders only get involved in decision-making at the implementation phase of the project cycle, and not in earlier project identification and preparation phases, hampering meaningful influence. This may make it a challenge to motivate participants. Furthermore, it places participants in a reactive position, where they are asked to respond to proposals that they perceive to have been finalised already (Chess and Purcell, 1999).
Next to ensuring influence and facilitating participatory processes, participants must also be able to participate. The capabilities of participants is an often mentioned factor in the participation literature (e.g. Lowndes et al., 2006;Reed, 2008;Verba et al., 1995). Governments can invest in developing the knowledge needed to create a more equal level playing field during participation trajectories. Furthermore, they can invest in capacity building aimed at ensuring that citizens are given the support to develop the skills and resources needed to engage (Lowndes et al., 2006).
Another view on the 'how' question is to make different participation options for different target groups. This can also provide a better matching of expectations in terms of motivation for participation (the motive of participants), its form and the costs of participation. As Lowndes et al. (2006: 288) put it: "The variety of participation options for engagement is important because some people are more comfortable with some forms of engagement such as a public meeting, while others would prefer, for example, to engage through on-line discussions. Some people want to talk about the experiences of their community or neighbourhood, while others want to engage based on their knowledge of a particular service as a user".
A related issue to empowerment concerns scale: the 'appropriate level' of participation (Roberts, 2004). As organizing meaningful two-way forms of communication between public officials and citizens by which authentic dialogue and deliberation can occur, group size becomes an important parameter in the participation equation. It is clear that for due deliberation groups should not be too large. Group-level participation involves citizens in groups ranging in size from 3 to 75 (Roberts, 2004: 332). Due deliberation requires thoughtful examination of issues and communicative processes of opinion and will-formation are of central concern rather than aggregation of interests and opinions (Dryzek, 2010). Due deliberation is grounded in an assumption that preferences of individuals are not fixed, but can change in debate and political dialogue (Held, 2006). "It requires information about an issue, knowledge of the basic elements of a problem, as well as an understanding of the relationships among the elements and the consequences and tradeoffs associated with different policies. The larger the number of people involved in direct participation, the harder it is for public judgment to emerge" (Roberts, 2004: 332).
The how question also points at the dynamics in participatory efforts. Next to motivations, competences and capacities of both governments and citizens this refers to the quality of the interaction process. Various literature on participation and co-production stress the need for developing trustful relationships between government officials and participants and among different groups of stakeholders or participants. In chapter 1 this is also discussed referring to the managerial and political competences at the organizational level. Trustful relationships enhance the quality of information exchange and mutual learning, contributing to both the satisfaction of the participatory process as perceptions about the quality of the outcomes (Irvin and Stansbury, 2004;Van Meerkerk and Edelenbos, 2014). Summarizing one of the main findings of a comparative case study analysis on the role of process design of participatory processes on sustainable land management De Vente et al. 2016 (p. 24) note: "If information exchange occurred through face-to-face contact between participants, there was also a significantly greater likelihood that the process would lead to sustainable solutions; conflict resolution; increased trust among nonstate actors and between nonstate actors and researchers".

Who
Which groups of citizens are and should be involved remains a highly debated issue in the participation literature. Many experiences in collaborative efforts with citizens show that the 'usual suspects', the highly educated, resourceful and political active citizens, are overrepresented as compared to members from lower socio-economic groups (e.g. Irvin and Stansbury, 2004;Koehler and Koontz, 2008).
To avoid the issue of 'usual suspects' lots are increasingly used as a way to recruit participants. For instance, lots are part of the design of mini-publics. Mini-publics are participatory forums gathered for deliberation on a specific topic lasting one or more days. Various forms and designs exist, but the most standardized are citizens' juries, deliberative polls and citizens' assemblies (Grönlund et al. 2014;Jacquet, 2017). One of the most important reasons to use lots is to increase inclusiveness and representation as random selection ensures that each person in the sampled population has an equal chance of being selected. However, experience shows that due to self-selection mechanisms still the better educated and politically active (and more politically interested) citizens participate (Goidel et al., 2008;Michels, 2017).
Furthermore, critical concerns are often raised with regard to the capture of participatory spaces by specific interest groups (Irvin and Stansbury, 2004). In this respect, another perspective on the 'who' question is to ensure that relevant stakeholders and groups holding different views need to be approached and included. This requires analytical competences in particular. For instance, careful ex-ante stakeholder analysis is an often mentioned method for identifying stakeholders, categorizing and differentiating between stakeholders (preferable with stakeholders) and -as a more advanced form -investigating relationships between stakeholders (Reed, 2008).
Thus, recurring issues with top-down pathways to collaboration is that expectations of influence are raised, but not met in practice. Moreover, continuous motivation and commitment of governments and citizens is challenging in top-down pathways. Citizens are often not satisfied with the strictly conditioned rules of engagement, particularly when their degree of influence in decision making turns out to be low (Irvin and Stansbury, 2004;Lowndes et al., 2001). Furthermore, critical concerns are raised about the usual suspects participating in these top-down constellations. As an alternative to top-down pathways, bottom-up pathways emerge in which citizens themselves set the rules for their engagement. This may lead to government participation. In the following section we go deeper into this pathway to (potential) collaboration between citizens and governments.

Bottom-up pathways to collaboration between governments and citizens
Collaboration between governments and citizens can be provoked bottom-up: by citizens themselves. Denters (2016) refers to created spaces of participation by citizens versus invited spaces of participation referring to government-led forms of participation. In this respect, there seems to be a growing trend of community-based initiatives in which citizens organize themselves to deliver public services for their community, such as the maintenance of public spaces, running a community facility, setting up a cooperative or charity to provide community-led care services for the elderly in the area or environmental initiatives aiming to provide local renewable energy (Brandsen et al., 2017;Edelenbos and Van Meerkerk, 2016;Healey, 2015). These community-based initiatives are increasingly considered to be valuable agents in the creation of public value by policymakers, since governments have implemented austerity measures and cuts in policy programs, alongside longer trends of welfare retrenchment (Kleinhans, 2017). Moreover, various campaigns and policy initiatives in Western countries have increasingly been aimed at shifting responsibilities and engaging civil society in the production of public services (e.g. Brandsen et al., 2017).
In this chapter we use the concept of community based initiatives (CBIs) to refer to these bottom-up citizen activities in which citizens mobilize capacities and resources to collectively define and carry out actions aimed at providing public goods or services for their community; citizens control the aims, means, and actual implementation of their activities (Healey, 2015). A systematic literature review by Igalla et al (forthcoming) reveals main features that characterize community based initiatives: 1. they are often locally oriented, which means that a. local residents, often collectives of residents, are the initial driving force behind the initiatives; b. they mobilize volunteers from within the community; and c. they focus on community needs; 2. they provide and maintain an alternative form of traditional governmental public services, facilities, and/or goods themselves; 3. they strive for autonomy, ownership, and control regarding decision making; 4. they are often linked to formal institutions, such as local authority, governmental agencies, and NGOs, especially for facilitation and public funding; 5. they often use market-based approaches to increase financial stability, but not aim at profitmaking.
For a number of reasons interdependent relationships between government organizations and community-based initiatives exist. For instance community based initiatives need resources (subsidies, knowledge, etc.) and legitimacy from governments to make their initiative viable or simple to execute their plans (Healey, 2015;Korosec and Berman, 2006). Government support or allowance, especially in dense regulated policy areas, is often important for community-based initiatives to grow or succeed ).
However, these initiatives arise from different motives and collaboration with governments is not self-evident. Just like top-down pathways to government-citizen collaboration, bottomup pathways come along with their own participation traps.

Bottom-up participation traps
Why CBIs can arise from dissatisfaction or complaints about governmental policy and actions or emerge in spaces that governments withdraw from due to budget cuts (e.g. Edelenbos et al., 2018;Gofen, 2015). For example, in the field of environmental planning, citizens may develop alternatives to government proposals. This can be an explicit reaction to government-led participatory efforts in which citizens do not feel sufficiently heard or in which they are, to their opinion, involved too late in the decision-making process. As a response and to avoid being marginalized as exclusively negative and driven by NIMBYism (Not In My Back-Yard), they can be triggered to develop their own initiative .
Similar motivations are found in the literature on citizen-initiated co-production. For example, Van Eijk and Steen (2016) found that dissatisfaction was an important motivation for citizens participating in neighbourhood watches. This type of motivation is also referred to as the salience of the issue at stake. They note that dissatisfaction about the current situationperceptions about an unsafe environment and beliefs about the inability of the police to respond in time -was an important drive for citizens to start a neighbourhood watch. In contrast with the top-down forms of co-production they studied (e.g. client councils in health care organisations), dissatisfaction was a more salient motivation in this bottom-up form of co-production.
Community-based initiatives also emerge in spaces that governments withdraw from due to budget cuts, and in domains that 'slipped' governmental attention. For example, in the field of Dutch health care cooperative emerge which are partly a response to the downsizing and upscaling of health care organizations, especially in rural areas. Also, in the field of urban regeneration, both in the UK (for quite some time now) and the Netherlands (more recently), community enterprises have emerged which are partly a response to government retrenchment and budget cuts in this field (Bailey, 2012;Kleinhans, 2017;. Coming back to the example of the neighbourhood watch studied by Van Eijk and Steen (2016), austerity measures were also an important cause for citizens to start their own initiative. As they note, reporting on their focus group findings (p. 39): "Because of austerity in public finances, the police and citizens must work together. 'Pressure on government is increasing and then you want to take on your responsibility' and 'you have own responsibilities as well for the safety in your own environment'." Another type of motivation are perceptions about the effectiveness of participation. This is referred to as external efficacy (Van Eijk and Steen, 2016;Alford, 2002). For instance, in the field of urban regeneration, CBIs emerge motivated by the idea to realize a more integrative and continuous approach to community issues as a response to more mono-centric, projectoriented government-led approaches. For instance, community enterprises (as a specific form of CBI) aim to integrate social, environmental and economic benefits in their approaches to neighbourhood development and fighting deprivation (Bailey, 2012). An important motivator for such CBIs is the idea that they can make a difference by developing a more communityled, anchored and holistic approach.
Building on the large amount of literature on volunteers' motivations several other, more personal factors to engage in community based initiatives can be mentioned. For instance, perceived personal rewards such as developing competences and enhancing self-esteem (see Alford, 2002;Van Eijk and Steen, 2016). Moreover, opportunities for connecting to others in relation with enhancing social cohesion in the neighbourhood is also a particular motive for participants in CBIs. Various studies have shown that the link with the neighbourhood is important. Although these motivations are also mentioned in the literature on top-down participation, they are likely to be stronger present in bottom-up participation (see next section).
Lastly, a more broad type of motive mentioned in the literature are so-called communitycentered motivations (Alford, 2002;Van Eijk and Steen, 2016). In this respect, this can refer to beliefs about the importance of contributing to the public good. Comparable with public service motivation. The idea of realizing empowerment, citizen self-control or 'doing democracy' can be a powerful motivator for bottom-up forms of engagement.
Resuming the motivations of citizens to initiative CBIs, both collective outcomes as personal rewards play a role, as also pointed out in the broader literature on civic participation (Scholzman et al., 1995). CBIs seem to be characterized by this balance in which there is both room for collective rewards and personal connections (Denters, 2016). Crucial is that citizens have the idea that they can make a more direct difference for their living environment (Bang, 1999).
Participation traps may lie in the perceived legitimacy of CBI's motivations (cf. . CBIs might become a vehicle for those stakeholders with much more resources (information, money, time, skills, etc.) to exert influence on policy-making and service provision that serves their private interests under the label of community and joint interests (Taylor, 2007). Furthermore, governance challenges in keeping CBIs running and keeping participants committed and motivated over time is another potential trap for CBI initiators.
For governments, the existence and effectiveness of these initiatives can be essential and of strategic interest (Korosec and Berman, 2006), making it worthwhile for governments to develop supportive relationships with community-based initiatives. Firstly, community based initiatives develop resources (people, activities, knowledge, implementation power, etc.) which may be useful for cost-effective service provision. For example, financial restraints and budget cutbacks may stimulate governments to seek and activate new forms of civic engagement, like social entrepreneurship and community self-organization, in order to maintain a certain level of service delivery in the fields of retreat (for example health care, sustainable energy, etc.) (Edelenbos and Van Meerkerk, 2016). In this respect, some "analysts may see [community-based] initiatives as an expression of a 'neo-liberal' downsizing of the state, and to some extent they are filling gaps left by a retreating state" (Healey, 2015: 117). Next, involvement of governments might be approached as normal given the significance of government in current society; its support is important and often crucial for communitybased initiatives to succeed, especially in institutionally dense policy areas (Brandsen, 2016). Moreover, community-based initiatives and social entrepreneurs bring experimentation and innovation, which is often problematic to achieve in bureaucratic organizations like governments (Korosec and Berman, 2006).
An important dilemma for governments in whether to support CBIs in one form or the other is the question of how representative the initiative is for the local community Denters, 2016). Furthermore, when CBIs are supported or are provided a greater role in the production of public services, the issue of formal accountability arises. In this respect, enhancing CBIs or giving them more influence on public issues (e.g. the maintenance of public spaces) can come at the price of government control. Next, collaboration with CBIs can be quite time-, resource-, and skill-consuming for governments. Specifically as this often requires customization: tailor-made solutions for a specific project. This might conflict with key values such as the state's legality, reliability and impartiality (Grotenbreg and Altamirano, 2017). Fourthly, politicians and executives may consider community-based initiatives, which often rely on crucial efforts of volunteer behavior, as unreliable, because of unpredictable behavior or commitments of citizens (Roberts et al. 2004). In this respect, a 'risk-averse, conservative administrative culture' may explain why citizens are not considered as a reliable resource providing partners (Voorberg et al., 2015). Community-based initiatives are often engaged in new ways of delivering services, which had hitherto been regarded as strictly professional, thereby changing the rules of the game or deviating from established policy and practices (Gofen, 2015). Next, political and professional reluctance to lose status and control can also explain the unwillingness to support community-based initiatives (Voorberg et al., 2015;Kleinhans, 2017).

How
Resources and mobilization capacities are key factors to consider in determining how CBIs develop and which participation traps can be distinguished Denters, 2016). Key resources for CBIs to emerge and sustain are different forms of social capital (bonding, bridging and linking capital), entrepreneurial community leadership and government support Igalla et al., forthcoming).
Social capital facilitates the mobilization of resources (e.g. knowledge, information and experience), co-ordination of actions , and creation of safety nets that reduce risks for individual community members (Peredo and Chrisman, 2006). CBIs "can harness social capital in local communities and use it to achieve positive outcomes through mobilizing volunteers, board members and paid staff" (Bailey, 2012: 30). The literature on social capital generally distinguishes three types: bonding, bridging and linking social capital (Putnam, 2000;Woolcock, 2001). In the context of CBIs bonding social capital is derived from strong ties between neighbours, friends and association members. These ties help to explain the formation of CBIs and dynamics in board management and continuity (Bailey, 2012;Somerville and McElwee, 2011). Bridging capital refers to the ties of the CBIs in the wider neighbourhood, ties to other networks and other community organizations. Bridging capital can generate support and a base for recruiting volunteers, thereby potentially enhancing the organizational capacity and legitimacy of CBIs (Somerville and McElwee, 2011;Bailey, 2012;Healey, 2015;Kleinhans, 2017). Linking capital refers to ties that CEs have with institutional key players, which can be crucial for getting resources and support.
Another crucial and typical factor for CBIs and bottom-up pathways to collaboration is community leadership. This plays an important role in the mobilization of resources and development of organizational capacities. Community leadership includes mostly managerial and political competences as community-building and mobilizing capacities, skills to build collaborative and strategic alliances and skills for identifying and exploiting new entrepreneurial opportunities (e.g. Selsky and Smith, 1994;Purdue, 2001).
From a critical stance, CBIs might give shape to new divides between those citizens who are capable to organize themselves and those who are not. Both community leadership skills and social capital, key factors in the development and durability of CBIs , is unequally dispersed among neighbourhoods. Furthermore, just because selforganizing citizen groups are connected around a particular kind of identity, they can be exclusive. Moreover, in order to function effectively, homogeneity and exclusion can be helpful, if not necessary in the context of self-organization (Dietz et al., 2003). That is, developing commitment and trust is far more easy in smaller, more defined groups where participants can interact in a face-to-face mode (Pestoff, 2014;McLennan, 2018). Hence, from a democratic point of view, keeping CBIs inclusive and open to different groups of citizens is a challenge.
Entrepreneurial oriented forms of CBIs, such as community enterprises, strive for considerable autonomy and own income sources and are characterized by hybridity: referring to the dual mission of realizing financial sustainability and a social purpose (e.g. Bailey, 2012;Doherty et al., 2014). Generating own income (through a business model) and simultaneously adding social value, are crucial for those types of CBIs. This hybridity may cause difficult tradeoffs. On the one hand, financial considerations may 'squeeze out' social goals, which is also a risk for the legitimacy of CBIs. On the other hand, mainly emphasising social goals endangers financial aspects of the business and thus threaten its survival. Various studies suggest that entrepreneurial forms of CBIs face considerable governance challenges to keep their organization running and effective over time, i.e. delivering particular services and social benefits to the target community (e.g. Bailey, 2012;. Key governance challenges are management , in particular recruiting board members with the right skills and experience, involving local people and other partners through participatory decision-making processes, and balancing the dual mission of social benefits and financial sustainability (ibid). Achieving sufficient staff capacity is a recurring challenge for CEs, which commonly depend on volunteers and a few (if any) paid staff members.
Government support or facilitation may come in different forms. Some authors roughly indicate that the supportive action of governments is important for the potential future of community based initiatives as this can boost their start and growth in scale and scope (Healey, 2015). This can be in general moral or material support, for example by providing grants, technical and managerial expertise, and advisory services in order to improve the management and operational capacity of CBIs. Next to material support, several scholars stress that community-based initiatives benefit from guidance through bureaucratic tangles and by enabling legal and policy environments (Healey, 2015;Van Meerkerk et al., 2013). Korosec and Berman (2006) distinguish several ways in which public officials and civil servants can support leaders and members of community-based initiatives (they focus on social entrepreneurship in particular). They distinguish different types of government support, covering in particular political and managerial competences:  Raising awareness. This can enhance legitimacy and interest to the efforts of community-based initiatives. Moreover, governments can collect and provide data on important social issues and educate beginning civic entrepreneurs in planning, programming and resource-development strategies (grant writing or facilities planning);  Providing assistance in acquiring resources. Government officials can join communitybased initiatives in submitting grants and funding proposals and pledge to match funds from private sources. Next, governments can also directly support communitybased initiatives by providing in cash (money) or in kind (time, attention) resources, regardless of proposals to other funding sources. Government can also get leaders of community-based initiatives in contact with organizations and community groups with space and leasing opportunities.
 Assistance in coordination and implementation. Governmental organizations can initiate and facilitate interaction and cooperation between members of communitybased initiatives and other private, nonprofit, community and public organizations. They can also play a key role in fostering connections and collaborations between projects and creating portfolios of inter-connected initiatives. Furthermore, public managers "…can also help (…) by assisting in matters that make program implementation easier, such as expediting permitting or approval for a project" (Korosec and Berman, 2006: 450).
Although governments may consider community based initiatives interesting, e.g. for enhancing service delivery, or even stimulate their development, the willingness and capability of governments to actually provide support is not that straightforward (Healey, 2015). According to the literature on co-production of public services (including citizen initiated forms of co-production), government organizations are often not compatible with such initiatives. They often lack the presence of inviting organizational structures and procedures within the public organization' and infrastructures to communicate with citizens or to collaborate with community based initiatives (Voorberg et al., 2015;Kleinhans, 2017). Furthermore, public servants might find it too much of work, which intervenes with their daily business or they may simply lack incentives to invest in collaboration with community-based initiatives. As Voorberg et al. (2015Voorberg et al. ( : 1342 note: for public officials, "it is often unclear to what extent public services can be improved by incorporating citizens or how co-creation creates budgetary benefits". Thirdly, government support and interventions may come at the price of turning CBIs into something quite different (Brandsen, 2016). Forcing them to comply with regulations, pushing them towards formalization and professionalization in order to receive financial support may crowd out motivation and enthusiasm among CBI's participants.

Who
The question of how representative the organisation is for the wider community pose challenges in requiring resources (e.g. community assets, funding) or acceptance from public bodies . Although the body of evidence on the representativeness of CBI participants and initiators is not that big, several Dutch studies show that CBI are able to attract a diverse pool of participants (Igalla and Van Meerkerk, 2015;Bakker et al., 2011). In comparison with top-down forms of participation they do not have a strong bias for the usual suspects, although it is important to make distinctions between different roles here. The various literature on CBIs stresses the role of community leaders with mobilizing, inspiring and boundary spanning qualities (Purdue, 2001;. These initiators and 'managers' of CBIs generally are higher educated and often have previous experience in public sector jobs (Igalla and Van Meerkerk, 2015;Denters et al., 2013). Other types of participants, for instance working as volunteer for specific projects or activities, have a far more diverse background. Furthermore, from their study on Dutch CBIs Bakker et al. (2011) came to the conclusions that despite biases in descriptive representation, views of participants and non-participants on the problems in their neighbourhood were strikingly similar.

Discussing key participation traps and crucial competences in top-down and bottom-up pathways
This chapter has reviewed bottom-up and top-down pathways to government-citizen collaboration with a particular focus on citizen participation. Various participation traps in relation to the why, how and who question. In this section the two pathways will be compared.
Wrapping up the motivations for participating in top-down or bottom-up pathways (the why question), we can conclude that bottom-up pathways can partly be seen as a response or alternative to the participation traps in top-down pathways. In the top-down pathway a frequently sited participation trap concerns the mismatch of expectations of participants and their level of influence on decision-making and implementation. A lack of feedback mechanisms between formal decision-making and participation processes or imbalance in motivation (more oriented towards gaining support for already made decisions) could endanger top-down participation turning into symbolic participation. Moreover, low levels of perceived efficacy among citizens can easily diminish broad or committed participation.
In this respect, bottom-up forms of participation seem to be partly a response to these issues of top-down participation. Perceived levels of external efficacy (having real influence, 'seeing' direct results), dissatisfaction with government-led participation or with perceived levels and quality of service provision are strong motivators for citizens to start their own initiative. However, more research is needed into the (long-term) motivations of participants in CBIs and to compare these types and level of motivation with top-down forms of participation (e.g. government-initiated co-production, consultation). Moreover, an intriguing question for future research is when and how citizens get triggered to start their initiative. In particular as governments in several Western welfare states have adopted measures to stimulate 'active citizenship' (Brandsen et al., 2017;Verhoeven and Tonkens, 2013). Though the literature on co-production has developed quite some knowledge on different types of motivations (e.g. Alford, 2002;Van Eijk and Steen, 2016), less attention has been paid to how external interventions impact on the intrinsic motivation of citizens to start collective forms of co-production and to keep their initiative running. This would mean more research on the effects of, for instance, government strategies and government responsiveness on citizens' intrinsic motivation. Psychological theories, such as Self-Determination Theory and Motivation crowding Theory, offer promising frameworks to go deeper into different psychological components of intrinsic motivation and how external interventions may impact upon these.
Concerning the how question various competences were pointed out in line with those discussed in chapter 1. In particular the managerial and political types of competences are key for both pathways, though they get a somewhat different shape. In top-down pathways political and managerial competences to effectively link participatory spaces with formal power structures seems crucial to overcome the participation trap of loosely coupled participatory processes (consultation or more deliberative forums) lacking responsiveness. However, the issue of responsiveness seems to be even more salient for bottom-up pathways to collaboration as they are more challenging for existing roles and structures vested in governmental institutions (Edelenbos et al., 2017;cf. Torfing et al., 2012). In this respect, various literature point out that facilitative organizational structures, autonomy of public managers and their boundary spanning capabilities are particular important for building collaborative and innovative governance arrangements with community based initiatives McLennan, 2018).
Furthermore, the individual and organizational capabilities addressed in chapter 1 are not only government's capabilities, but in bottom-up pathways they are particularly also about citizens' capabilities. Managerial and political competences of initiators and leaders of CBIs, such as their mobilising capacities, entrepreneurial skills and abilities to build strategic alliances are highly important for their performance and durability, as well as impacting on the relationship with governments Igalla et al., forthcoming).
An important question for future research in this respect is under which conditions governments and citizens in bottom-up pathways complement each other (cf. McLennan, 2018). Citizens can add or bring in different resources, competences and capabilities (e.g. local knowledge, networks, local commitment, creativity and community leadership skills).
In the search for enhancing governance capacity, a recommendation for governments would be to carefully scan for, trigger and mobilise existing initiatives and to look for ways to complement these. Moreover, governments can play an important role in on the one hand safeguarding certain democratic values, while at the same time providing external authority which enhances the impact of CBIs.
Concerning the who question, a clear critique on top-down participation is the mobilisation of usual suspects. For bottom-up participation, there are indications that more diverse groups of citizens participate, though initiators and also the more large or professional CBIs are often run by 'usual suspects'. However, this is not that clear yet and provides another avenue for future research to find out.