Governance without we. Wicked problems and collaborative governance

Abstract

In this introductory article to our PPA special issue we argue that the relation between wicked problems and collaborative governance is fraught with difficulties. Doing so, we take four steps: we maintain that wicked problems are better understood as wicked actor constellations; we delve into the collaborative governance-approach and posit that it is mainly concerned with wicked problems; we then argue that collaborative governance might not be the most feasible response to wicked problems; and after that we argue that an assumption instead of actors being engaged in actor positioning processes allows a clearer understanding of policy processes in wicked actor constellations. Following this discussion, we introduce the articles featured in the special issue and discuss implications for research.

Keywords

Collaborative governance new modes of governance wicked problems wicked actor constellations

Introduction

This is the introductory article to the special issue on wicked problems and collaborative governance. Before introducing our argumentation, we first want to clearly set out what the main subject of interest is of our special issue. It is this: a collaborative governance-approach often starts from the assumption that collaborative governance is needed because we are faced with wicked problems. This set of ideas, and this is our main argument, is inconsistent: wicked problems are wicked because relevant actors disagree. Through the Special Issue we want to further the debate by making the argument that a focus on collaborative governance ideals may be a distraction from the reality of power inequalities, information and educational differences, politics, different cultures and aims. Collaborative governance assumes agreement; not direct agreement, but agreement among involved actors is assumed that their collaboration in the end helps everyone because it addresses wicked problems. And although this is in part true because collaboration might address reconstructions of wicked problems, this conceptualisation of collaboration underestimates that wicked problems are problems that actors disagree about.

In this introductory article to our PPA special issue we take four steps in order to discuss the problematical relation between wicked problems and collaborative governance. We first make a methodological argument: we think wicked problems can better be understood as wicked actor constellations (step 1, §2). We then delve into the collaborative governance approach on the basis of a number of key publications and argue that it is particularly concerned with wicked problems (step 2, §3). We argue that a collaborative governance approach might not always be the most feasible response to wicked problems (step 3, §4). We then again make another methodological argument: we argue that an assumption that actors are engaged in positioning processes allows us a better understanding of the policy process in the face of wicked problems (or, as we would phrase it, in wicked actor constellations) than the assumption of collaborative action (step 4, §5). After this discussion, we briefly present the contributions to our special issue and discuss implications for research (§6).

Wicked problems as actor constellations

In their seminal, well-known 1973-article, Rittel and Webber point to the importance of wicked problems for policy sciences. They argue that problem definitions and definitions of the public good are disputed and ‘it makes no sense to talk about “optimal solutions”’ (Rittel and Webber 1973: 155). Rittel and Webber maintain ‘that one of the most intractable problems is that of defining problems (of knowing what distinguishes an observed condition from a desired condition) and of locating problems (finding where in the complex causal network the trouble really lies)’ (ibid., 159). Planning problems, therefore, are ’wicked’ problems, not ‘tame’ ones. Rittel and Webber mention a set of 10 characteristics of wicked problems, for example that they prevent ultimate definitions and have no stopping-rule.

To further explain this, we turn to Alford and Head (2017), Hoppe (2010) and Levin et al. (2012). Alford and Head’s (2017: 397) notion of wicked problems varies with respect to ‘their cognitive complexity or the diversity and irreconcilability of the actors or institutions involved.’ Wicked problems show high complexity and diversity. Wicked problems, Hoppe (2010: 73) argues, are characterised by a lack of certainty on required and available knowledge and a lack of agreement on norms and values at stake. Levin et al.’s (2012: 124) conception of ‘super wicked problems’ is comprised of ‘four key features: time is running out; those who cause the problem also seek to provide a solution; the central authority needed to address them is weak or non-existent; and irrational discounting occurs that pushes responses into the future.’ A central element in all these conceptions is that different actors have different understandings of the factual basis of an alleged problem and different normative orientations that support different definitions of a specific state of the world as a problem.

We consider that these insights were already captured by Rittel and Webber’s ninth characteristic: ‘the existence of a discrepancy representing a wicked problem can be explained in numerous ways,’ so that ‘the choice of explanation determines the nature of the problem’s solution’ (ibid., 166). We consider this ninth characteristic centrally important for our understanding of the wicked nature of problems. A wicked problem is wicked because actors do not agree on problem definitions and policy preferences; they not agree on facts and they do not agree on values.

‘Problem’ as a reified concept

This main characteristic of wicked problems causes that we cannot take an analysis of the weaker and stronger aspects of policies in relation to a predefined problem as the point of departure. Understandings of the problem do not exist apart from the actors involved. Wicked problems point to the variety of different interpretations of both policies and problems by different actors. Wicked problems are considered wicked because their interpretation differs among the actors involved in policy making. This implies that, strictly speaking, we cannot treat the issue as a single problem. We do not deny that a realist analysis of problems and related policies is possible. In the case of wicked problems, however, we observe that different actors emphasise different intersections of reality. These in turn invoke different normative appreciations of this reality and different responses to it. Problem definitions and policy preferences are characteristics of the actor and wickedness is a characteristic of the actor constellation.

Relatedly, Bevir and Rhodes argue in their book ‘The 3 R’s in rethinking governance’ (2016) that ‘structure’ is a reified concept. Structure is the emerged constellation of actor activities regarding the 3R’s in Bevir and Rhodes’ book title, ‘rulings, rationalities and resistance.’ Actors’ attempts to ‘rule,’ inspired by their own ‘rationalities’ and possibly in resistance to other actors attempting the same, constitutes ‘structure.’ In the same vein, one might argue that ‘problem’ is a reified concept. Actors’ attempts to define a problem, inspired by their own ‘rationalities’ and possibly in resistance to other actors attempting the same, constitutes ‘problem.’ The ‘problem’ actors face is not a single conceptualisation that functions as the input of governance processes in which actors transcend their differences and work towards a response. Instead of a shared conceptualisation of ‘problem,’ various conceptualisations of ‘problem’ are inserted into the interaction between actors, who each act departing from their own view. While doing so, governance actors cannot fully distance themselves from their own understanding of a problem. They look at the world ‘reflexively,’ not ‘reflectively,’ as Lash (2003) might argue.

The governance of wicked problems as a dependent variable-problem

When wicked problems are concerned, the involvement of a variety of actors is required to build a more complete understanding of the ‘problem.’ The multi-dimensionality of problems requires multi-dimensional input of actors. Their involvement is required, because these actors bring different understandings of a problem. However, this difference of understandings, one would expect, is exactly the characteristic that complicates the emergence of a shared understanding of problems. Multi-dimensional input is required to produce an output of a policy process that is composed of different, actor-level understandings of a problem. The analytical problem here is that a shared output of the policy process cannot be assumed to exist before the process in which multi-dimensional inputs are aggregated.

What does exist is a constellation of actors, each with their own understandings of the world around them. This is what makes ‘problems’ wicked. A problem is wicked, when actors do not agree on its definition. Therefore, they do not necessarily organize around a presumed problem. They act based on their own interpretations. They might, but they also might not come to a shared understanding of the problem.

There is an infinite set of potential policy responses to an infinite set of problem definitions, of which only very few actually emerge. In these responses, the problem of shared policy formation is solved. Research into these shared policy responses can then easily assume that there is shared recognition of network partners of some ‘problem,’ of course partially contested, but still lying at the centre of the policy response. In this sense, the problem conceptualisation is reified from the effective formation of the network of policy actors that defined it. We face a problem related to selection on the basis of the dependent variable. If we study collaborative governance-processes in the face of wicked problems, we might develop the impression that collaborative governance is the feasible response to wicked problems.

If, on the other hand, we start our analysis in the actor constellation, we face a problem of case selection, admittedly, but our methodology allows an understanding of political struggles, non-inclusion of problem definitions, competition between problem definitions, etc.

In the next section, we turn to the collaborative governance-approach, briefly discuss a number of examples from this approach and make two key arguments: firstly, that the collaborative governance-approach understands itself as an approach aimed at understanding how responses to wicked problems emerge; and secondly, it does so by re-ifing ‘problem’ from collaboration.

Collaborative governance as a response to wicked problems

A collaborative governance-approach often starts from the assumption that collaborative governance is needed because we are faced with wicked problems (e.g. George et al., 2016; Waardenburg et al., 2020; Weber and Khademian, 2008).

A recent article by Bartels and Turnbull (2020) shows that the Public Administration-literature contains a broad variety of ‘relational approaches.’ They present a two-by-two categorization, crossing the two dimensions ‘individualist’ – ‘holist’ and ‘critical-reflexive’ – ‘instrumental-strategic,’ and as such leading to four broad categories of assumptions on actor relations in the Public Administration-literature. They categorize the ‘collaborative governance’ and ‘new public governance’-approaches as halfway the individualist and holist position on the first dimension and more instrumental-strategic than critical-reflexive on the other dimension. In these approaches, Bartels and Turnbull (p. 1335) explain:

‘some degree of coordination is beneficial to the performance of individual organizations and can generate public value. Hence, “boundary spanning” (Williams, 2002) and creating “collaborative advantage” (Huxham and Vangen, 2003) involves building and sustaining relationships of mutual trust, understanding, commitment and reciprocity. While the strength and shape of relationships varies according to network types, in more collaborative networks “the traditional commitment to producing goods and services must give way to a commitment to improved relationships and forming a new whole” (Mandell and Keast, 2007: 593).’

We reconstruct the main argument in collaborative governance and new public governance-approaches as follows:

• because of high complexity and multi-dimensionality of social problems, collaborative governance aimed at a shared understanding of public value is required;

• successful collaborations show such shared understanding and are able to align the individual and collective level;

• if actors in policy networks have such shared understandings, successful collaboration develops.

This argumentative structure can indeed be observed in some of the core publications of the approach.

A wide variety of definitions of governance exists (Van Kersbergen and Van Waarden, 2004), but a general element of the term is that the formal procedure of government is not taken to fully define the actual unfolding of the processes of governance (e.g. Bevir and Rhodes, 2016; Jessop, 2011). Governance is the action by actors involved in a process of governing and these actors might be public actors, civil society actors or private actors; a wide variety of actors. Governance modes can be understood as the means of ‘ruling’ (cf. Bevir and Rhodes, 2016) various actors take to attempt to define a collective response to a problem and attempt to organize its implementation.

We take ‘governance’ to refer to the actual, material action of societal decision-making, as opposed to the concept of ‘government,’ which then refers to the formalized structures and procedures of societal decision-making. The ‘governance’-concept is important, we consider, because it makes us understand that we cannot simply assume actors’ capacity to organize the world and we cannot simply assume the organization of the process of conceiving a plan and implementing it. A formalised structure (noun) does not actually structure (verb) action, the set of actors involved, nor their power positions. Decision making, planning and organisation are all governance processes in which a variety of actors is involved, all on the basis of their own preferences and understanding attempting to affect the process.

This in itself descriptive theory of governance is in the literature often also applied as a normative theory. It is often argued that the ‘hybridity’ of decision-making (Brandsen et al., 2005), better than ‘government,’ describes actual decision-making process (descriptive) and fits into a world-view in which people (both in the meaning of individual people, persons, as in the meaning of a people, a society) can organize their lives bottom-up (normative).

A very prominent example of this application as a both descriptive and normative theory is Osborne (2006, 2010). He wrote a short history of modes of ‘public administration and management’ (abbreviated as ‘PAM’). He distinguishes between three modes of PAM: ‘public administration’ (PA), ‘new public management’ (NPM) and ‘new public governance’ (NPG). Osborne conceived of the changes that he observed in modes of PAM as an ‘evolution’ (emphasis added) towards increased functionality of PAM: ‘the NPM has actually been a transitory stage in the evolution from traditional PA to what is here called the New Public Governance (NPG)’ (Osborne, 2006: 377; see also the discussion in Bannink, 2019).

Osborne presents ‘Public Administration’ as a politically neutral system, created to implement policies or perform tasks that were defined in an organisationally distinct non-neutral political system.¹ Public Administration is a Weberian bureaucracy, in which officials employ their expertise, but are assumed to do so in a normatively neutral manner: they employ their expertise to implement policies defined elsewhere.

‘New Public Management,’ different from PA, is based on the premise that one specific characteristic structures the actual functioning of policy implementation systems: bureaucratic actor preferences. A systematic conflict of preferences between the political system and the administrative system required that implementation systems be designed in such way that administrative preferences – being neutral and rational – are aligned to and protected by political preferences through the market-based co-ordination mechanisms of New Public Management, where alignment is bought through incentives rewarding specific policy implementation outcomes.

‘New Public Governance,’ then, different from both PA and NPM, is based on the premise that not only varying actor preferences but also dispersed actor expertise affects the actual functioning of policy implementation systems. A systematic conflict of preferences among political actors and between political actors, implementation officials and the citizenry, required that policy and policy implementation be designed in such way that preferences are aligned. At the same time, and additionally, policy formation and implementation are and need be formed by a broad variety of expertises.

In Osborne’s terms, new public governance ‘posits both a plural state, where multiple inter-dependent actors contribute to the delivery of public services and a pluralist state, where multiple processes inform the policy making system’ (Osborne, 2006: 384; emphasis in original). Actors have their own interdependent preferences that need be included (as in Osborne’s pluralist state) and more or less legitimate claims to expertise that need to inform policy implementation (as in Osborne’s plural state). Pluralist policy making is linked to plural policy delivery.

If NPM can cope with interest interdependency (coordinating these interests through the market mechanism), NPG, then, is presented as a new evolutionary stage of PAM that can potentially cope with both pluralist interests and a plurality of expertises.

The structure of argumentation is: in a ‘complex’ society in which actors have different ideas of what is true and different preferences regarding what is right (a world of wicked problems), we can overcome our differences (both dispersed resources and different policy views) by applying a ‘relational approach’ to governance (collaboration enables the identification of problems and supports the formulation and implementation of a response).

New Public Governance is closely related to the collaborative governance-approach. Bartels and Turnbull (2020: 1334) locate NPG and collaborative governance next to each other and present them as ‘relational approaches.’ Following Bartels and Turnbull’s (ibid.: 1335) classification of relational public administration approaches, ‘Their point of departure usually is that individual public managers face complex problems which should be addressed through interagency networks. Some degree of coordination is beneficial to the performance of individual organizations and can generate public value.’

Boundary spanning, for example, creates collaborative advantage. Emerson et al. (2011: 2) define ‘collaborative governance’ as ‘the processes and structures of public policy decision making and management that engage people constructively across the boundaries of public agencies, levels of government, and/or the public, private and civic spheres in order to carry out a public purpose that could not otherwise be accomplished.’

We consider it important to understand how this definition excludes non-collaborative modes of governance. In itself, after all, any procedure of governing structures public policy decision making and engages people to carry out a public purpose. New Public Management, for example, engages people to contribute to a public purpose by incentivizing them. Old Public Administration, as it is often called, engages people by regulating actions.

In contrast, Emerson et al. (2011: 3–4) explicitly refer to deliberative democracy as a core component of their approach. What exactly is the collaborative nature of engaging people in collaborative governance? The core is captured by the phrase ‘constructively across the boundaries.’ A collaborative governance-approach builds a new whole that aggregates the elements required to build the integrated purpose. This aggregation is done in a specific way that differs from aggregation efforts in NPM and Old PA. The distinction between Old PA, NPM and collaborative governance lies in the nature of the procedure aggregating the input required to carry out a public purpose. Furthermore, the normative claim of the collaborative governance-approach is that an improved, that is collaborative, input generates an improved, that is also collaborative, output of the governing process. Again, a similar line of argumentation is applied: we are faced with public purposes that could not otherwise be accomplished; and the mode of governance re-ifes ‘problem’ from collaboration.

To further clarify what we see as the core conceptualization of collaborative governance, we argue that the collaborative mechanism that is being suggested by authors like Osborne (2006) and Emerson et al. (2011) differs from the collaborative mechanism Huxham and Vangen observed (Huxham and Vangen, 2003; also mentioned in Bartels and Turnbull, 2020). Their ‘suppression of hatred in pursuit of money’ (Huxham and Vangen, 2003: 26) connotes this other mechanism. This statement is a citation from one of their respondents and it underscores Huxham and Vangen’s argument that actors do not engage in collective endeavours, but instead engage in co-ordination efforts with other actors in order to ‘assess uncertainty and reduce fear of opportunistic behavior’ (ibid., 26). Huxham and Vangen do not conceptualise ‘collaborative advantage’ as the advantage of being able ‘to carry out a public purpose,’ but instead as the actor’s advantage that can only be attained through some form of co-ordination with others.

Osborne (2006, 2010) and Emerson et al. (2011) presume a social structure of actors that engage in collaboration in order to build a collective good that cannot otherwise be built, a higher-order goal. Huxham and Vangen, on the other hand, presume a social structure of actors that pursue their own goals and while doing so engage in co-ordination with others, who pursue their own goals in turn. Engaging in interaction with others in order to co-ordinate action is not necessarily ‘collaboration’ and the other way around, ‘collaboration’ is a specific type of inter-actor co-ordination. Also the Old PA- and NPM-approaches and Huxham and Vangen’s ‘suppression of hatred in pursuit of money’ describe modes of inter-actor co-ordination.

It is precisely the reference to a higher-order goal, a goal located at the inter-actor level, which forms the core of the collaborative approach. This reference subsequently is related to the presumption that actors might waive (parts of) their own goals for the higher-order goal they cannot reach separately. This mechanism is visible in in Bryson et al.’s approach.

Bryson et al. (2017) see as the main theoretical challenge of collaborative governance-approaches the connection between the beliefs and preferences of actors and the understanding of reality and the formation of preferences at the level of collaboration. In their article ‘Towards a multi-actor theory of public value co-creation’ on Moore’s public value approach, Bryson et al. (2017: 646) phrase this issue as follows:

One of the limitations of the public value governance triangle is the fact that each actor in the relevant policy field faces the challenges posed by his/her/its own strategic triangle. In other words, in any given multi-actor situation, there are multiple strategic triangles actually or potentially in play. When one realizes that fields also include multiple tiers and spheres, the result is triangles ‘all the way up, all the way out, and all the way down.’

Here, we clearly detect the core of the wicked problem-concept: different actors have different beliefs and work towards different ends. The conclusion of the article, however, is a research agenda aimed at the mechanisms that might move actors to engage in a bigger triangle, by, for example, studying ‘the conditions required for different but needed actors to: Gain agreement on what public value is to be co-produced’ (p. 649). We observe that Bryson et al., like Osborne and Emerson, assume a higher-order goal (the agreed upon public value) that derives from the actors’ own triangles up, out and down. They presume actor-level willingness to engage in this collaborative understanding of problems. This implies that the co-produced public value is re-ified through the collaborative effort; the effort defines, not follows the problem.

Nabatchi et al. (2011) place this argument in the context of societal complexity. They state that ‘public administration has failed to consider what it means to govern in dark times’ (Nabatchi et al., 2011: i29). ‘Darkness’ is related to ‘systemic domestic policy failures’ (ibid., i32), including ‘the impoverishment of educational access […]; disparities in access to and the quality of health care; […] reductions in the supply of and access to energy; […] business failure, unemployment, and the rise in the need for welfare and social services’ at the national level, while globally ‘event and policy forms of dark times are even more discouraging. Climate change, […] the growing fragility of the global economic system, […] threaten not just security but humanity as a whole.’ (ibid., i32–i33). ‘At the root of these events and policy problems’ lies a deeper problem, Nabatchi et al. argue (ibid., i33): ‘The disappearance of the public realm has eroded our abilities to discuss and make sense of shared problems and concerns.’ ‘[W]hen the deliberative functions of civil society are contrasted with more rational modes of problem solving based in neoclassical economics,’ Nabatchi et al. (ibid., i33) conclude, ‘the cycling between public cynicism and policy failure is perpetuated, making the work of public administration increasingly intractable and wicked.’

Nabatchi et al. (i34) argue that ‘the field is rendered impotent to govern in dark times because of two interrelated and mutually reinforcing issues. First, public administration does not adequately acknowledge itself as a de facto arbiter of political conflict and therefore as a discipline capable of and responsible for shaping societal affairs. Second, public administration suffers from a bureaucratic pathology that consigns the field to the pursuit of narrow administrative goals and limits its abilities to address complex policy problems.’ The response Nabatchi et al. put forward is: ‘We suggest that a reinvigoration of democratic ethos may help public administration improve its governance capacities amidst such troubling circumstances and realities’ (ibid., i30).

We see this as the core idea of the approach. When society is complex, collaborative governance is required and what it does is improving the output of governance processes (collaborative advantages, public value) by improving the input of these same processes (reinvigorated democratic ethos, boundary spanning).

We return to the issue of democracy in the section ‘Governance Without We’ below. We see two problems: in dark times, we disagree on public value and the ways to attain it, so the input into democratic processes might not yield social alignment; and, in dark times, because we disagree on public value and the ways to attain it, also the outputs of democratic processes are not likely to encompass the various preferences and understandings.

In the next section, we argue that actors engaged in collaborative governance processes should not be assumed to collaborate.

Actor constellations, problem types and governance modes

We concede, indeed, that wicked problems can be addressed when there is shared understanding, or, the other way around, if there is shared understanding, then successful collaboration in response to wicked problems might easily emerge. We would argue however, contrarily, that shared understanding is an often-seen characteristic of successful collaborations, but – the other way around – shared understanding is not a likely result of the interaction of actors who move in a constellation around a wicked problem, a constellation of actors with their own intersections of relevant facts and their own evaluations of these. Research of successful collaborations shows shared understandings and aligned preferences, indeed. However, such successful collaborations are a sub-set of the wide variety of actor constellations around wicked problems, an important part of which do not produce successful collaborations. We might even put the issue in stronger terms: where understandings are shared and preferences are aligned, either the problem is not wicked, or important understandings are not included.

Below, we present our conceptualisation of ‘actor constellation’ and show how various actor constellations allow specific suitable modes of governance.

Actor constellations, as we define them here, are sets of social actors, not necessarily involved in actual governance processes on the basis of their factual understandings of social problems and the normative preferences vis-à-vis these problems.

An actor constellation with high agreement on facts and a strong belief that ‘we’ share the social problems we suffer from, allows ‘public administration’ as a mode of governance. The public administration of problems, can indeed be conceived of as neutral, because the actors in the actor constellation agree. It can solve rather simple problems, where agreement on facts is easily attained. A public bureaucracy built on input steering is an example: input steering requires a strong understanding of the facts; and it functions well if the officials working in it and the people affected by it accept policy implementation by the bureaucracy because of its ‘input legitimacy’ (Scharpf, 1999). This is in accordance with Osborne’s (2006, 2010, see above) conceptualisation of PA as one of the modes of PAM.

New public management is the response to normative differences, disagreement on preferences. It organises the response to social problems, where the factual understanding of the problem is not subject to conflict, but normative orientations are. In response to normative conflict, actors within the constellation trade values and buy other actors’ conformity to the values the buying actors pursue. NPM is aimed at incentivising officials to act in accordance with policy maker preferences. This conceptualisation is in accordance with Osborne’s conceptualisation of NPM as the second mode of PAM.

The next one puts New Public Governance as the response mainly to factual differences of understanding in the actor constellation. It does not form a response to normative differences. We can ask a variety of actors, each bearing a specific knowledge base, to work together and collaboratively solve a joint problem, under the condition that the problem is in fact experienced as a joint problem. New public governance presumes a shared normative orientation. If different actors control specific knowledge bases, we need their involvement to solve a social problem related to all these knowledge bases. Nevertheless, we cannot be sure, because we do not understand that these others work towards the same preferred state of the world.

In our Figure 1 above, we therefore present new public governance as a suitable mode of governance in an actor constellation of shared preferences and a dispersal of factual understandings, but not as a suitable mode of governance of an actor constellation characterised by disagreement on facts and preferences, a wicked problem. In the context of wicked problems, new public governance is used as a ‘magic concept,’ a concept with ‘a seeming ability to dissolve previous dilemmas’ (Pollitt and Hupe, 2011: 641). In a magic concept, the need for a specific response to a problem and the capacity to actually produce this response are conflated (Bannink, 2020).

Figure 1.

Actor constellations (as defined by agreement on facts and preferences); related problem types and suitable modes of governance.

Governance without we

An actor constellation of different understandings and different preferences poses a wicked problem, or, as we call it, a double governance challenge. We need the joint application of factual understandings to solve a joint problem, but we cannot understand, and therefore we cannot govern whether the actors involved in the governance process share an orientation to the same shared problem. We presume a ‘we’ encompassing the social problem, but we cannot produce the ‘we,’ because the ‘we’ that is to produce it, is an actor constellation characterized by disagreement on facts (which this potential collective might possibly overcome) and preferences (which causes that this same collective does not exist). We see this as an instance of ‘governance without we.’

We think a collaborative or new governance approach is not the most suitable approach to address this issue. We expect that actor preferences with respect to the inclusion of specific types or groups of actors in the policy process, reflect their preferences with respect to the substantial result of the same policy process. In other words, the pattern of inclusion and exclusion of actors in the policy process (the governance mode) more or less defines the outcome (choices already made).

This is not just a formally, but also a substantively important issue. Consider, for example, Rodrik (2011) who uses the concept of the ‘globalization paradox’ in order to argue that globalisation, national sovereignty and democracy stand in a tense relation towards each other and most likely cannot be obtained together. Given this ‘reality of society’ (Polanyi, 1947), what does an inclusive policy response mean? Is it, as Aulenbacher et al. (2019: 106) outline the issue, the local embedding of the economy to achieve inclusive prosperity (Naidu et al., 2019, cited in Aulenbacher et al., 2019: 106) or even more radical pleas for selective economic deglobalisation (Novy, 2017, cited in Aulenbacher et al., 2019: 106), inclusiveness within the limits of a presumed society? Or the other way around, working towards Beck’s (Beck and Sznaider, 2006) cosmopolitanism, inclusiveness at a global scale? Who is the ‘we’ that we define in our governance processes and who is the ‘we’ defining it? We face a ‘double governance challenge’ of needing to define a ‘we’ required to define a ‘we.’

A similar problem might be conceived with regard to Covid-19. We might arrange our societies in such ways that individuals take responsibility for other individuals, both nearby and farther away or we might arrange our societies in such way that individuals isolate themselves from others in order to contain the spreading of Covid-19. Who, whose preferences, in other words, are included in ‘we’ and whose are not? If ‘we’ disagree on this question, the same ‘we,’ that we cannot define beforehand, needs to decide who is included in the governance arrangement that defines who is included in the policy response to the virus. We consider that the analogies to other policy fields are clear: social policies, security policies, asylum, infrastructure, etc. Seen as such, the collaborative governance of a problem is not the choice procedure; it defines the choice. Hajer (2003) showed in the early 2000s how this led to the non-recognition of new understandings of social problems. The environmental issue, but also socio-cultural issues disappeared in an ‘institutional void.’

Of course, we need to acknowledge that governance, not only collaborative governance, always defines inclusion and exclusion. A collaborative governance process following a democratic ethos as Nabatchi et al. suggest (Nabatchi et al., 2011: i30) is very difficult to design, but also very difficult to conceive: what would a democratic process exactly need to do, what does democracy actually mean (Axtmann, 2013)? We might vote, but this is not contrary to the classic conceptualization of a formally democratic decision-making procedure combined with a neutral bureaucratic-regulatory (Public Administration) or incentive-based (NPM) implementation of decisions. The collaborative governance approach, in Emerson’s et al. (2011: 2) formulation, aims to ‘engage’ ‘across boundaries,’ but it is less clear who is engaging, who is to be engaged and with what aim? We argue that collaborative governance is based on the presumption that it organises a ‘we’ and as such can produce an adequate response to social challenges.

We argue, however, that the presumption of the ‘we’ that might produce an adequate response to challenges in collaborative governance-theory is systematically not fulfilled. This has to do with exactly the point of departure that Nabatchi et al., Emerson et al. and Osborne start from. We face dark times, we are divided by boundaries, we live in a plural and pluralist society – in other words we disagree on public values. Governance ‘in dark times’ can be ‘democratic’ if the social context in which it takes place is ‘democratic,’ but it is unlikely that it is, exactly because times are dark.

Actors in collaborative governance should not be assumed to collaborate

So, what do actors engaged in collaborative governance processes do? Why are they there? Not necessarily with the intention to jointly produce the as yet unknown intended outcome. The main theoretical challenge of collaborative governance-approaches, we would argue, is the connection between the actor- and collaboration-level and it is exactly the understanding of this connection that seems weakly developed. Actor-level willingness to engage in a collaborative understanding of problems seems to be assumed, not explained.

In Bartels and Turnbull’s figure of relational approaches in the public administration scholarly domain, one of the quadrants, interestingly, is almost empty. They discern only a few approaches they categorize as ‘individualist’ and ‘critical-reflexive.’ We agree with this observation and we consider this gap a problem for the public administration-field. In this quadrant, PA-scholars would analyse the position and action of the actor, applying a reflexive (Lash, 2003) approach, thus taking the interpretation of the own position of the actor as a point of departure.

This central element – correspondingly – accounts for what we see as the reification of collaborative governance as an approach to tackle wicked problems. A broad literature in the Public Administration-field, comprised of New Public Governance, Collaborative Governance, also part of the Public Value-approach, seems to reveal a disconnect between the meso-level network analysis of collaborative governance processes aimed at realising public values and addressing wicked problems, on the one hand, and the micro-level theory of action, on the other, that in our view does not adequately underpin these analyses. From our perspective, the foundational theories at the micro-level fail to fully support these meso-level analyses. One might argue that it is required in the face of wicked problems to overcome disagreement among involved actors, but the main characteristic of a wicked problem is its wickedness, which refers to disagreement on problem nature and preferred responses. Where, then, are the orientations produced that make the actors with differing orientations collaboratively respond to wicked problems?

The question is a very important one. Under what conditions would an actor leave his or her own triangle in order to get involved in the bigger triangle (Bryson et al., 2017)? Starting from a critical-reflexive perspective, such bigger triangle needs to be seen as the triangle of someone else, possibly someone claiming a higher-order, integrative, collaborative position. Is collaboration indeed the most feasible interpretation of what an actor engaged in collaborative governance-processes aimed at the production of public value in a policy field does? We think not always and not only. We argue that actors involved in collaborative governance-processes may do something different than collaborate and their action is something that we can analytically understand (e.g., Källström et al., 2021).

Actor positioning in collaborative governance processes in the face of wicked problems

Meso-level analyses of collaborative governance seem to focus on the relation between network characteristics and network outcomes with regard to collaborative action: for example, how can or should networks composed of strongly different members be managed as compared to networks of members who are more alike? A micro-level analysis would focus on the relation between actor characteristics and the nature of these networks: for example, how would an actor work towards a position of network leadership and how would other actors respond or how would actor involvement in a network emerge, if at all? The analysis then starts from the point of departure of actors being involved in their own, different Moore-type triangles. Understood as such, the assumption is left aside that there is a single process of collaboration that is to be understood, aimed at some understanding of public value, which we understand as being reified from the process. Policy making-processes, allegedly collaborative and allegedly oriented towards public value, might better be understood as mutual actor positioning-processes in which actors strive for different, actor-based understandings of value, public or otherwise, emerging in their own ‘triangles.’

We place ourselves in the tradition of the ‘strategic relational’ approach (Baker and Stoker, 2015: 40) that ‘accepts the argument that the opportunity to act is contingent on institutional circumstances but claims that individuals understand their situation differently based on their status, the resources that they control and their ability to change their circumstances.’ In a relational approach, the actor is justifiably understood as an ‘inter-actor’ (Bartels and Turnbull, 2020). This does not imply, however, that the orientation of the actor becomes integrative or collaborative in nature. Perceptions and preferences are informed by and respond to the perceptions and evaluations of others, but remain differentiated still, based in this specific actor. In our understanding of the interaction of actors in actor constellations around differentially defined problems, actors are embedded in a context of other actors and institutions, indeed, but the differential embeddedness of actors in this context creates differential perceptions and evaluations of this context, including the estimation of the perception and evaluation of the position of the other.

Observed at a distance from such actor constellations, one might consider the ‘problem’ a ‘wicked’ one, but this consideration is not fully accurate. When a wicked problem is actually wicked, the problem definitions (the intersections of facts and their evaluation) of the actors in the constellation differ, so any statement on what the ‘problem’ might be is re-ified from the analysis of the result of the interaction.

From this it follows that it is also very difficult to understand what collaborative governance might entail. Referring to Bryson et al. mentioned above: it is ‘triangles all the way up, all the way down,’ and because of that one cannot assume that ‘different but needed actors’ in some way make these differences disappear and ‘gain agreement on what public value is to be co-produced.’ Actors, differentially embedded as they are, make their own intersections of facts and formulate their evaluations of those and estimate those of others. Actor positioning processes start from these estimations, not from the assumption that agreement on the co-production of public value is to be gained.

The core message of our article is that wicked problems ‘are actually wicked’ (Bannink and Trommel, 2019: 210), so that it becomes to delineate what problem we study. The loss of governability associated with wicked problems also concerns collaborative governance. Collaborative governance is needed in the face of wicked problems, but it is also unlikely, precisely because these problems are wicked. It becomes difficult to understand actors’ interaction in terms of collaborative governance and we consider actor constellations instead. In these constellations, actors position themselves and while doing so do not necessarily (but may) collaborate and do not necessarily (but may) perceive similar intersections of facts and do not necessarily (but may) evaluate these in a similar way.

In an article ‘Actor positioning in collaborative governance: financial negotiations on elderly care through a situated actor lens,’ La Grouw and Bannink (under review) follow a rather strict methodology with respect to the assumptions on actors’ action. La Grouw and Bannink start from actor positioning processes and show how these generate a specific problem definition with respect to their subject of elderly care and a specific policy response. This definition and corresponding policy response were shared by a specific set of actors, excluding other actors. The actors supported this shared definition, because it supported their own policy implementation processes.

The actors who were successful in this constellation built their shared definition through a process La Grouw and Bannink called ‘empathic positioning’: they aimed at a thorough understanding of the other, to be able to align their own interest to the interest of the other. La Grouw and Bannink show how this is different from what they call ‘sympathic positioning,’ which they consider being aimed at the mutual acceptance of the value of the preference of the other. Empathic positioning refers to factual estimation of the other actor’s position, while sympathic positioning refers to normative support. La Grouw and Bannink observed that the normative support for the selectively shared problem definition resulted from an actor’s normative support of the own preference and the factual estimation (not the normative evaluation) of the other’s preference and the factual estimation that some overlap exists.

What La Grouw and Bannink aim to do, is showing how the normative appreciation of a partially shared problem definition follows from the initially separated normative orientations of the (in their case) two actors, combined with their factual estimations of each other’s position.

PPA-special issue on wicked problems and collaborative governance

In this Public Policy & Administration-special issue on wicked problems and collaborative governance, we present a selection of research articles that build upon the radical appreciation of wicked problems as wicked and start from the perspective of actor positioning processes instead of collaborative governance. We asked the authors to evade assumptions on shared problem definitions and shared normative evaluations on policy responses that analytically precede the governance process. As such, the special issue is a first attempt to build a research agenda we think better grasps the intricacies of wicked problems and collaborative governance.

We received a high number of about 30 submitted article proposals of which 18 were selected for a next round of development. We held a 1-day online conference with the authors on 28 March, 2022 to further discuss the approach. After the regular double blind-review process, we were able to include eight articles in our special issue on wicked problems and collaborative governance. Below, we briefly introduce the articles included in our issue and after that, we discuss implications for Public Administration-research and analysis.

A multitude of analyses of collaborative governance processes in the face of wicked problems

The articles in this special issue can be grouped in two different sets. In the first set, authors (Martinius, Kuitert et al., Joosse, Zonneveld et al.) argue that inter-organisational collaboration in the face of wicked problems lead to intra-organisational tensions. The second set (La Grouw et al., Yalcin, Hoppe and Turnbull, Dudink and Veenswijk, Castelnovo and Sorrentino) points to the strategic ‘positioning’ that actors in collaborative endeavours undertake.

All articles point to a central problem of the collaborative approach to wicked problems: it is difficult to really understand the interaction processes as ‘collaborative’ and to understand the policy process as working towards ‘policy integration.’ To borrow from Joosse and van Buuren (2024: 327), who refers to Stead and Meijers (2009): ‘while policy coordination seeks to fill the gaps between policy fields and remove inefficient overlap, policy integration searches for a joint and new policy in which sectoral interests and perspectives are combined to a richer whole.’ Collaboration then, (Joosse: 327, refers to Ansell and Gash, 2008 and Emerson et al., 2011) is influenced by factors like dialogue, trust, commitment, shared motivation, and principled engagement. Actors do not build a richer whole but elaborately struggle to combine demands from the intra- and inter-organisational level of interaction; and actors do not engage in dialogue, but instead position themselves and try to affect the positioning by others in strategic inter-actor processes (cf. Ysa and Greve, 2023).

Not the problem is wicked, but the actor constellation.

Martinius (2024: 359) in her case study on water management networks notes that ‘boundary work flourishes at the fringes of organizations’ involved in the organisational network she studied and that this boundary work was ‘largely invisible to outsiders,’ including other members of the organisations. This, she argues, might limit ‘the potential of boundary work to withstand wicked challenges.’

Organisations in Martinius’ case study requested that boundary work realizes collaboration on the basis of high moral values, while, as a result, frictions and tensions between the own values of water authorities and utility owners were not reduced. The invisibility of the boundary work mission stands in the way of this: Unless boundary spanners are asked to report back, to the centre stages of their organizations, not only about the solutions but also about the down to earth, every-day delicacies in their ne- gotiations, it is easy to confuse collaborative outcomes as the solution to boundary work, when its true potential lies in the negotiation of differences.

Kuitert et al. (2024) discuss the ‘wickedness of combined value systems’ as their title announces. In a line of argumentation similar to that by Martinius, they argue that hybridity following from collaboration between organizations creates a problem that ‘especially manifests within an […] organisation’ (p. 367, emphasis in original). Kuitert et al. ask: ‘How does the implementation of New Public Governance confront civil servants in municipal organizations with internal hybridity and value tensions?’ (p. 367). By posing this question, Kuitert et al. point to an important complication to the collaborative governance-literature: hybridity not only creates an inter-organisational mix of governance modes, but also an intra-organisational one.

In their case study, Kuitert et al. found differences in the degree of NPG adoption, top-down translation of values causing implementation problems related to tensions between formalisation and flexibilisation, misalignment between intra-organisational levels and remaining differences between pillars, professions and value interpretations (p. 379). Kuitert et al.’s findings suggest ‘a scenario […] where value conflicts will result in trade-off value systems rather than collaboratively striving towards balancing public values’ (p. 380).

Also Joosse and van Buuren (2024) ‘unravels the intra-organizational dynamics of collaborating on policy integration’ (p. 324). Joosse defines policy integration as distinct from policy coordination: ‘while policy coordination seeks to fill the gaps between policy fields and remove inefficient overlap, policy integration searches for a joint and new policy in which sectoral interests and perspectives are combined to a richer whole’ (p. 327).

However, Joosse argues, ‘intra-organizational policy integration is equally important here as it is between different organizations; however, this may bring about its own dynamics’ (p. 327) and it is these dynamics he studied. In his case study, ‘policy integration was developed in temporary program teams at the edge of public organizations and easily volatilized after time because of a lack of foothold and ownership in the line organization. Ironically, policy integration became marginalized: a differentiated, stand-alone task executed in the periphery of bureaucracy’ (p. 341). ‘Organisational reflexes and forces’ may ‘easily result in a marginalization of policy integration,’ Joosse concludes (p. 342).

Zonneveld et al. (2024) place their article in the context of Ansell and Gash (2008) who argue that the identification of common values is an important step in the formation of multi-party networks. Zonneveld et al. instead ask (p. 386) what values the different actors in the network find important, what value tensions arise and how the actors in the network cope with these value tensions? They take a decentred actor perspective and focus on the values of actors in an interorganisational network, ‘without assuming a priori that shared values are necessary for successful collaboration’ (p. 386).

Zonneveld et al.’s findings are important for the understanding of collective action: instead of working towards a shared understanding of value, Zonneveld et al. argue that ‘value differentiation within a collaborative network should not be seen as a problem only, but also as a vital component of working together’ (p. 405) and: ‘Public professionals need to be aware of this and learn to resist the urge to reduce value differentiation’ (p. 405).

La Grouw et al. (2024) refer to Bannink and Trommel (2019) and Bartels and Turnbull (2020) and argue that it is plausible that actors with different perspectives and interests cannot collaborate easily, and concurrently that it is important to analyze ongoing actor-positioning processes in which actors strive for different, actor-based understandings of the nature of the issues, the purpose of the collaboration, and its rules of engagement as an important part of understanding collaborative governance for wicked problems.

These actor-based understandings inspire actor-positioning processes. They discuss ‘mundane dynamics’ in order to broaden the comprehension of actor positioning processes, because such an analysis ‘can give insight into the dynamic, contested, and provisional nature of governing and positioning; in the invisible work such as building trust or catching up that do not make it into formal documentation; and into hidden aspects of actor positioning, acts of resistance, or concealed agreements, laying bare harsh and hidden realities’ (p. 417), which in turn can increase the understanding of ‘hidden aspects of actor-positioning, acts of resistance or concealed agreements, laying bare harsh and hidden realities’ (p. 10).

Muftugil-Yalcin and Mooijman (2024) points to a similar pattern. She discussed social entrepreneurship. She argues that social entrepreneurs are often seen as ‘being predominantly motivated by a social purpose, […] as being trustworthy providers of public services making the collaboration between social enterprises and government bodies possible.’ Following this perception, ‘we see that in many places around world, social entrepreneurship policy is translated into programs that focus on strengthening an ecosystem in which social entrepreneurship can flourish’ (p. 437).

In her case study on local social entrepreneurship, however, Muftugil-Yalcin and Mooijman (p. 436) found that social entrepreneurs and municipalities follow different logics: ‘The results of this study confirm that in the governance of wicked problems, there is an ongoing actor positioning process where, in this case, social enterprises and municipalities, hold on to their actor-based understanding of the nature of issues and collaboration.’ The analysis showed, Muftugil-Yalcin and Mooijman argue (p. 451), ‘how in the governance of wicked problems we encounter an actor-based understanding that dominates the nature of collaboration’.

A similar understanding of different logics is present in Castelnovo and Sorrentino’s (2024) article on a complexity-embracing approach to public policy. Local variations, they argue (p. 462), ‘should be considered as the result of the resourcefulness of the policy implementers.’ Castelnovo and Sorrentino maintain that Lipsky’s coping mechanisms might be understood as an illustration of local policy entrepreneurship. Local variations thus are ‘the emergent result of processes of self-organisation’ of local implementers, who mix elements of national policies with local requirements and specific contextual dynamics. Their understanding of local actors is that of stewards who are motivated by content-related values and by ego-related values like self-realisation. ‘Local implementers’ resistance’ then is a factor that triggers local self-organisation.

This emphasis on the nature of actors and the policy network is also present in Hoppe and Turnbull (2024). They point to what they call the ‘wrong problem-problem.’ When confronted with wicked problems, not so much the substantive policy problem, but instead the ‘structure and dynamics of the policy network’ (p. 475) is the main cause of policy lock-in. In such case, Hoppe and Turnbull argue (p. 476), it is unfeasible to plead for collaborative governance as a metagovernance solution.

Hoppe and Turnbull studied EU-level governance processes and noted that the European Commission (EC), instead of following a participatory logic and instead of formulating an integrative solution to the substantive policy problem, facilitated the continuation of interaction among policy actors. Referring to Bannink and Trommel (2019), they argued (p. 491): ‘Instead of pushing gentech and agro-ecology complementarity, this is no more than a gentle nudge, accepting the need to manage entrenched distances. It is clearly imperfect and incomplete. But one cannot deny it is sound and intelligent strategic analysis,’ and as such the EC solved the wrong problem-problem through an intelligent actor positioning effort.

Dudink, Taminiau and Veenswijk (2024) address a public-private-partnership (PPP) as a mode of collaborative governance in which ‘a voluntary, mutually beneficial collaboration [is] aimed at creating social or economic value combing complementary skills of public and private partners’ (p. 498). Collaborative relationships like PPP’s are expected to enhance the understanding and addressing of wicked problems, because a variety of perspectives is combined. Therefore, PPP’s are sometimes considered a ‘panacea for the crisis-ridden world […] by providing more effective and efficient structures to carry out public tasks and goals’ (p. 498).

In their case study on PPP in financial crime, however, Dudink and Veenswijk show that collaboration is a questionable partnership because the collaboration is imposed on banks by the government. Their analysis questions whether the PPP is justly defined as such. The presentation of the collaboration as a joint partnership, is contrasted to ‘[t]he private actor’s motivation to use a reputable and collaborative relationship as a guise to conceal the negative aspects and power imbalances of the collaboration’ (p. 514).

Governance without we: a double paradox of collaboration

We conclude our analysis by presenting a double paradox of collaborative governance when faced with wicked problems. A first paradox concerns the input and another concerning the output of collaborative governance processes. The input-paradox of collaborative governance arises because those required to collaborate may not do so because they disagree with the alleged purpose of the collaboration as it is presented by a sub-set of the actors seeking other actors’ contribution. The output-paradox of collaboration holds that those engaging in collaboration produce a sub-set of possible understandings of the problem the collaboration might be aimed at and therefore produce only a sub-set of possible responses to a wider set of potentially relevant problem definitions.

This would not be so much of a problem, if we might assume that the pattern of inclusion and exclusion in the governance process is ‘inclusive.’ But in the face of wicked problems, this is exactly what we cannot do.

We are faced with a condition of ‘governance without we.’ The conceptualization of a problem required to define inclusion and exclusion in the governance of this problem, required to define an inclusive solution is itself subject to contestation. The ‘we’ required to define the ‘we’ our policies address is itself the result of governance processes that include specific actors with specific normative orientations and exclude others.

Understood as such, the set of collaborative networks around recognized problems is a small sub-set of the set of wicked problems, or – as we understand them – actor constellations around differentially recognised and evaluated conditions. In these actor constellations we cannot assume an orientation of actors to jointly produce an as yet unknown intended outcome. We consider that further research into the collaborative governance of wicked problems thus needs to address two methodological issues.

The first concerns the selection of relevant cases: what, in the endless set of actor constellations are interesting research cases? Acknowledging this problem forces us, PA-scholars, to make explicit what remains implicit in a wicked problem and collaborative governance-approach: a case starts with the recognition and evaluation of some actor that a specific condition is a problem. This actor might be a policy actor, a stakeholder identifying a problem, a detached researcher in some substantive field of concern, or the researcher of the case itself. What is important, though, is that we consider this actor as a representative of a specific subset of actors in the constellation proposing a specific definition of the condition and the evaluation, not as an actor identifying a wicked problem. When a problem is wicked, it is not the problem that we can identify, but the actor orientations towards it.

The second issue concerns the orientation of actors that are included in these cases. When we define our cases as wicked actor constellations, the second methodological issue is more or less solved, although we do consider it important to make it explicit. The orientation of actors in these constellations is our point of departure and this orientation naturally differs. Therefore, research concerns the actor positioning processes that occur in the actor constellation. This enables us to do research into governance processes in which different actors aim to pursue their own, different understanding of public value. In other words, it means opening up research on the micro-foundations of public governance, acknowledging the political and open nature of governance in a multi-actor world where collaboration is a subjective choice rather than a new whole. Furthermore, it allows us to understand that other actors might not be concerned with this same problem, although their action might be relevant for the first actor’s attainment of their pursuit.

A broad research question might then be: how do actors position themselves within ‘wicked’ actor constellations, where different actors have their own, different understandings and evaluations of conditions? Such an approach allows us to refrain from what we consider an implicit need for closure that we recognize in collaborative approaches to wicked problems and to take wicked problems for what they are: wicked.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Duco Bannink

Alessandro Sancino

Note

References

Alford

Head

(2017) Wicked and less wicked problems. A typology and a contingency framework. Policy and Society 36(3): 397–413.

Ansell

Gash

(2008) Collaborative governance in theory and practice. Journal of Public Administration Research & Theory 18(4): 543–571.

Aulenbacher

Bärnthaler

Novy

(2019) Karl polanyi, the great transformation, and contemporary capitalism. Österreichische Zeitschrift für Soziologie 44: 105–113.

Axtmann

(2013) Conceptualising democracy. Australian Journal of Political Science 48(1): 118–128.

Baker

Stoker

(2015) Nuclear Power and Energy Policy. The Limits to Governance. Basingstoke, UK: Palgrave Macmillan.

Bannink

(2019) Besturen Zonder Wij. Amsterdam, Netherlands: VU University Press.

Bannink

(2020) Proeftuinen zonder wij. Bestuurskunde 29(3): 66–72.

Bannink

Trommel

(2019) Intelligent modes of imperfect governance. Policy and Society 38(2): 198–217.

Bartels

Turnbull

(2020) Relational public administration: a synthesis and heuristic classification of relational approaches. Public Management Review 22(9): 1324–1346.

10.

Beck

Sznaider

(2006) Unpacking cosmopolitanism for the social sciences: a research agenda. British Journal of Sociology 57(1): 1–23.

11.

Bevir

Rhodes

(2016) Rethinking Governance: Ruling, Rationalities and Resistance. Abingdon, UK: Routledge.

12.

Brandsen

van de Donk

Putters

(2005) Griffins or chameleons? Hybridity as a permanent and inevitable characteristic of the third sector. International Journal of Public Administration 28(9–10): 749–765.

13.

Bryson

Sancino

Benington

, et al. (2017) Towards a multi-actor theory of public value co-creation. Public Management Review 19(5): 640–654.

14.

Castelnovo

Sorrentino

(2024) Agency in multi-level governance systems: The implementation puzzle and the role of ‘intelligent’ local implementers. Public Policy and Administration.

15.

Dudink

Taminiau

Veenswijk

(2023) The dark side of public-private partnerships: Enforced hybridity and power dynamics in fighting financial crime. Public Policy and Administration.

16.

Emerson

Nabatchi

Balogh

(2011) An integrative framework for collaborative governance. Journal of Public Administration Research and Theory 22: 1–29.

17.

George

Howard-Grenville

Joshi

, et al. (2016) Understanding and tackling societal grand challenges through management research. Academy of Management Journal 59(6): 1880–1895.

18.

Hajer

(2003) Policy without polity? Policy analysis and the institutional void. Policy Sciences 36(2): 175–195.

19.

Hoppe

(2010) The Governance of Problems: Puzzling, Powering and Participation. Bristol, UK: Policy Press.

20.

Hoppe

Turnbull

(2024) Problem structuring, wrong-problem problems and meta-governance as the strategic management of untractable positions: The case of the EU GM Crop Regulatory Framework controversy. Public Policy and Administration.

21.

Jessop

(2011) Metagovernance. In: Bevir

(ed) The Sage Handbook of Governance: Newcastle upon Tyne, UK: Sage, pp. 106–123.

22.

Joosse

van Buuren

(2024) The marginalisation of policy integration: Dynamics of integrated policymaking in the periphery of bureaucracy. Public Policy and Administration.

23.

Källström

Mauro

Sancino

, et al. (2021) The governance games of citizens and stakeholders’ engagement: longitudinal narratives. Local Government Studies 47(5): 859–885.

24.

Kersbergen

Waarden

(2004) ‘Governance’ as a bridge between disciplines: cross-disciplinary inspiration regarding shifts in governance and problems of governability, accountability and legitimacy. European Journal of Political Research 43: 143–171.

25.

Kuitert

Volker

Grandia

(2024) Facing NPG implementation problems in municipal organizations: The wickedness of combined value systems. Public Policy and Administration.

26.

La Grouw

Bannink

(2024) Actor positioning in collaborative governance. Financial negotiations on elderly care through an actor positioning lens.

27.

La Grouw Y, Van der Woerd O, Visser, EL, and Blijleven W (2024) Mundane dynamics: Understanding collaborative governance approaches to ‘big’ problems through studying ‘small’ practices. Public Policy and Administration.

28.

Lash

(2003) Reflexivity as non-linearity. Theory, Culture & Society 20(2): 49–57.

29.

Levin

Cashore

Bernstein

Auld

(2012) Overcoming the tragedy of super wicked problems. Constraining our future selves to ameliorate global climate change. Policy Science 45(2): 123–152.

30.

Lipsky

(1980) Street-Level Bureaucracy. Dilemmas of the Individual in Public Services. New York, NY: Russell Sage Foundation.

31.

Mandell

Keast

(2007) Evaluating network arrangements. Toward revised performance measures. Public Performance & Management Review 30(4): 574–597.

32.

Martinius

(2023) Mission invisible? Managing boundaries for flood protection. Public Policy and Administration.

33.

Muftugil-Yalcin

Mooijman

(2024) Making sense of each other: Relations between social enterprises and the municipality. Public Policy and Administration.

34.

Nabatchi

Goerdel

Peffer

(2011) Public administration in dark times: some questions for the future of the field. Journal of Public Administration Research and Theory 21: i29–i43.

35.

Naidu

Rodrik

Zucman

(2019) Economics for inclusive prosperity: an introduction. Economics for Inclusive Prosperity: An Introduction. Economics for Inclusive Prosperity (EfiP) Network, 1–9.

36.

Novy

(2017) Emancipatory economic deglobalisation: a polanyian perspective | desglobalização econômica emancipatória: uma perspectiva a partir de Polanyi. Revista Brasileira de Estudos Urbanos e Regionais 19(3): 554–575.

37.

Osborne

(2006) The new public governance? Public Management Review 8(3): 377–387.

38.

Osborne

(2010) The New Public Governance? Emerging Perspectives on the Theory and Practice of Public Governance. London, UK: Routledge.

39.

Polanyi

([1947] 2001) The Great Transformation: The Political and Economic Origins of Our Time. 2nd edition. Boston, MA: Beacon Press.

40.

Pollitt

Hupe

(2011) Talking about government: the role of magic concepts. Public Management Review 13(5): 641–658.

41.

Pressman

Wildavsky

(1973) Implementation. How Great Expectations in Washington Are Dashed in Oakland; or, Why It’s Amazing that Federal Programs Work at All, This Being a Saga of the Economic Development Administration as Told by Two Sympathetic Observers Who Seek to Build Morals on a Foundation. Berkeley, CA: University of California Press.

42.

Rittel

Webber

(1973) Dilemmas in a general theory of planning. Policy Sciences 4: 155–169.

43.

Roberts

(2020) Bearing the white man’s burden. American Empire and the origin of public administration. Perspectives on Public Management and Governance 3(3): 185–196.

44.

Rodrik

(2011) The Globalization Paradox. New York, NY: Norton.

45.

Scharpf

(1999) Governing in Europe: Effective and Democratic? Oxford, UK: Oxford University Press.

46.

Stead

Meijers

(2009) Spatial planning and policy integration. Concepts, facilitators and inhibitors, Planning Theory & Practice, 10(3), 317–332.

47.

Vangen

Huxham

(2003) Nurturing collaborative relations: building trust in interorganizational collaboration. The Journal of Applied Behavioral Science 39(1): 5–31.

48.

Waardenburg

Groenleer

De Jong

(2020) Designing environments for experimentation, learning and innovation in public policy and governance. Policy & Politics 48(1): 67–87.

49.

Weber

Khademian

(2008) Wicked problems, knowledge challenges, and collaborative capacity builders in network settings. Public Administration Review 68(2): 334–349.

50.

Williams

(2002) The competent boundary spanner. Public Administration 80(1): 103–124.

51.

Ysa

Greve

(2023) Emerging ideas for strategic public management: strategizing collaborative governance. In: Handbook on Strategic Public Management. Cheltenham, UK: Edward Elgar Publishing, Vol. 280.

52.

Zonneveld

Raab

Kenis

Minkman

(2024) The role of values in the interorganizational network response to wicked problems. Public Policy and Administration.