Abstract
Local public networks can be governed in many different ways. Among scholars as well as among practitioners, there is some debate about the best approach. Based on literature, this article examines the relationship between local public network governance roles, contingency factors, and network performance in a theoretically informed way. Governance roles are positioned on a continuum from top-down (commissioner) to bottom-up (facilitator), with an intermediate area (coproducer). How governance roles influence the performance of local public networks is assumed to depend on contingency factors, which might explain the inconsistent results of studies examining this influence. An integrated model of local public network governance is presented that includes four contingency factors: the number of network participants, diversity of network members, degree of customizability of service demands, and the number of new network participants. The model can be applied to the heterogeneous contexts that local governments encounter when governing local public networks.
Introduction
Public networks and their performance have been the subject of increasing interest in the literature (Agranoff & McGuire, 2003). A special group of public networks are those at the local level. The general tendency has been to decentralize and to force local governments, like municipalities, to assume responsibility for formerly national government policies. Authorities are constantly confronted with the intricate demands of today’s society and therefore rely increasingly on relationships among organizations in public networks in local contexts (Rethemeyer & Hatmaker, 2008). Local governments are collaborating in networks in many different contexts such as education (Meier & O’Toole, 2005), social housing (Silverman, 2008), health care and welfare (Mitchell & Shortell, 2000), and social policies (Gilsing, 2007). It is presumed that local governments better understand local needs and may therefore be able to govern local networks more adequately.
Local public networks can be governed in many different ways. Local governments are often unaware of the best way to govern their network (Klijn, Edelenbos, Kort, & van Twist, 2008). In the literature the focus on network governance roles is still in its infancy (Gössling & Oerlemans, 2007). Therefore, additional research is needed to advance public management research and knowledge about the various manners in which local governments can best manage their network (Van Slyke, 2007).
Governance roles are collections of activities and rules that serve to achieve the goals of a network (Brass, Galaskiewicz, Greve, & Tsai, 2004). Insight into governance roles within a network and the effectiveness of these roles can support local managers and policy makers in determining their choice of governance.
Our research question is as follows:
Research Question: Which governance roles of local governments work best under which contingency conditions, according to literature?
This article results in a model that contributes to understanding which governance role works best under which conditions (Provan, 2009). These conditions may vary within local public networks. The propositions may moreover guide future empirical studies in the area of interorganizational networks.
First, this article elaborates the general characteristics of local public networks. Next, different governance roles are described, and finally, the ways in which local governance roles influence local network performance are presented in a model from which propositions are drawn. The model includes four contingency factors that influence the relationship between local governance roles and local performance.
Characteristics of Local Public Networks
Local public networks are a special type of interorganizational network constituted around a governmental body. Within these networks it is always a lower governmental body such as a city district or a municipality that governs the network based on a variety of roles, with the local government ultimately held accountable for the satisfactory delivery of public goods and services (McGuire, 2006). Local public networks have general characteristics that pertain to most such networks, such as the network’s nature, structure, and performance. Other characteristics act as contingency factors and may vary between networks.
Local Public Network Nature
The network nature reveals the conditions under which organizations establish a network. The dominant focus within interorganizational network research concentrates on an emerging (voluntary) nature of networks (Gössling & Oerlemans, 2007). However, local public networks are different from most other interorganizational networks in being primarily of a mandated nature. Organizations collaborating in the network with the local government participate because they are dependent on the public (monetary) resources that it provides. A mandated nature can also stem from legal directives. In that case organizations are forced by law to work within local public networks (Andrews, Boyne, Law, & Walker, 2008).
Local Public Network Structure
The structure of a network is an important determinant of network performance (Provan, Fish, & Sydow, 2007). Structural typologies use concepts such as network density and centrality to predict network performance (Gössling & Oerlemans, 2007). Local public networks typically have a lead organizational structure that is dense and central (Johnston & Romzek, 2008). The central position of the local government derives from its legal authority. Local governments provide administration for the networks and/or facilitate the activities of member organizations to achieve network goals.
Organizations in local public networks collaborate to comprehensively “treat” clients through integrated, jointly produced services. For example, an elderly person suffering age-related disabilities may simultaneously need a wheelchair, a stair lift, help with her administration and help with her housekeeping. In the Netherlands, local governments must ensure that those products and services are delivered. To that end they have to govern the organizations within their local network. The various services are provided by different organizations and only the bundle of services together fulfills the total demand of the elderly person.
Local Public Network Performance
Network studies generally aim to understand why some governance mechanisms are more effective than others (Agranoff, 2007). Unfortunately, there is little agreement among organizational and public-policy scholars or among public administrators on how public networks should be evaluated. Local public networks can be evaluated at three levels of analysis: the community, network, and participant levels. Each of these levels has its own effectiveness criteria. However, the evaluations at the different levels are related. While local public networks that are successful are likely to be effective on all three levels of analysis, the different stakeholder needs and expectations are not necessarily consistent across levels. Therefore, scholars claim that within local public network, the community value outweighs the performance criteria for the other two categories. Korssen-van Raaij (2006, p. 19) states after examining the control processes in Dutch Health Care Networks that “public networks tend to produce community-level network results which could not be realized by one of the organizations individually”. Also Provan and Milward (2001) state after examining the diferent levels of effectiveness analysis that “the rationale for local public networks is most apparent at the community level” (p. 421). The goal of most public networks is to enhance client services. The community level can best be evaluated by examining whether the needs and expectations are met of those groups within a community that have both a direct and indirect interest in seeing that client needs are adequately met. This is in line with Andrews, Boyne, Law, and Walker (2009) who also examined the effectiveness of local governments and concluded that “local public networks aim to satisfy the demands of their citizens and therefore the degree to which these demands are met constitutes the most important performance indicator” (Andrews et al., 2009).
Besides assessing whether the performance goals are achieved (effectiveness), efficiency (cost-effectiveness) should also be considered (Provan, 2009). After all, different governance roles by governments imply different costs, making it important to examine which governance role achieves the best outcome relative to costs (Boyne, 2003).
We have argued that local public networks in general have a mandated nature, a lead organizational structure, and that they aim for the maximum level of citizen satisfaction within a policy area. Next, we will describe the different local governance roles.
Governance Mechanisms at the Local Level
How a network is governed is a critical factor in predicting network performance (Klijn, 2005; Park, 1996; Provan & Kenis, 2008). Although scholars mention the importance of this relationship, empirical studies are scarce (Dekker, 2004). Governance mechanisms in local networks are often complex and local governments are not always familiar with the role forced on them by the national government (Gilsing, 2007). Moreover, they frequently do not understand how governance mechanisms affect network performance (Rethemeyer, 2005). Therefore we elaborate on different governance roles that can be distinguished when studying local public networks.
Networks are affected from below and from above (Agranoff & McGuire, 2001; Herranz, 2008; Provan et al., 2007; Provan & Kenis, 2008). We can consequently distinguish various governance mechanisms on a continuum of top-down to bottom-up processes. The notion of a continuum is important because the intermediate area between bottom-up and top-down governance may achieve the best network effects (Alvesson & Karreman, 2004).
Often, only one of the governance roles is considered. Hill and Lynn (2005), for example, focus on the top-down type of governance, Lee (2006) concentrates on bottom-up governance, and Bogason (2000) only discusses cogovernance. Kooiman (2003) presented a model that incorporates several governance roles. However, based on an analysis of the cogovernance role, Somerville and Haines (2008) concluded that the model of Kooiman (2003) was difficult to use because it does not specify which mechanisms belong to the three different roles. Span, Schalk, Luijkx, and Schols (2009) developed an overview of characteristics of governance roles in local government based on an analysis of the literature. Figure 1 depicts the model. In the model, the characteristics of three governance roles (commissioner, coproducer, facilitator) are elaborated for nine dimensions: who is the main actor, what is the steering mechanism, who sets the boundary conditions, who is dependent, who aligns, who sets goals, who is responsible, who develops the vision, and who monitors results? Organizations in a local network can be seen as executors (commissioner role), partners (coproducer role), or initiators (facilitator role). Boundary conditions are the conditions that determine whether goals are reached. This relates to quality or quantity criteria of the services delivered; for example, what is the quality of a wheelchair and what are the activities performed when delivering housekeeping? The question is who sets these criteria: the local government, the local government together with the other parties, or the other parties? Dependency is understood as being dependent on the power wielded by local governments based on rules and procedures. Alignment refers to which party coordinates the delivery of products and services. Responsibility for reaching the goals can be with the local government or the other parties, or can be acknowledged as a shared responsibility. A vision is the basis for the network goal. Monitoring means keeping an overview of all different aspects of governance. Governance roles can be positioned on each of the nine dimensions separately on a continuum of bottom-up versus top-down (Span et al., 2009).
A typology for local network governance
The Effect of Governance Roles on the Performance of Local Public Networks
Although there is consensus on the importance of governance within public networks, there is hardly any systematic investigation of how governance roles affect performance (O’Toole & Meier, 2004). As far as we know, there are no empirical tests and few conceptual papers on the effects of governance roles on network performance when the network has a mandated nature and a lead organizational structure, which is the case for local public networks (e.g., Provan & Kenis, 2008). Scholars preferring top-down governance suggest that this leads to better performance because of the greater decision speed, stronger direction and clear goals, and less risk of internal conflicts (McGuire, 2006). Scholars who opt for bottom-up governance suggest that top-down governance may harm performance because it prevents other organizations from making independent decisions, leads to inflexible rules and procedures, and endangers adequate responsiveness to changing circumstances (Andrews et al., 2009). Studies in the public sector have found contradictory results on the effects of top-down and bottom-up governance on performance. Top-down roles may be characteristic for mandated networks (Agranoff, 2007) or may be critical to encouraging autonomous organizations to act in ways that lead to achieving desired network-level goals (Provan & Milward, 1995). Top-down governance has been shown to have a positive effect on the output of U.S. manpower agencies (Whetten, 1978). However, pulling the central policy lever does not necessarily mean that something happens at the bottom. Top-down governance is often presented as the “philosopher’s stone” of modern government, ever sought after, but always just beyond reach (Rhodes, 2007).
Other research has concluded that excluding professional organizations from governance results in poor-quality public services (Ashmos, Duchon, & McDaniel, 1998; Holland, 1973; Martin & Segal, 1977). Korssen-van Raaij (2006), studying Dutch health care, found that networks using bottom-up governance achieve better network performance if organizations are forced into a network. Mandated health care networks need bottom-up mechanisms to transcend own interorganizational interests in favor of the interests of the network as a whole. Warren, Rose, and Bergunder (1974), however, examined 42 local public networks that used a bottom-up governance role to increase the quality of public services but did not find these advantages. In these networks, even in cases where all organizations started to rely on bottom-up governance, they were not willing to exchange knowledge, time, or other scarce resources. Bottom-up governance led to an ongoing inefficient struggle for authority among the network organizations. This process resulted in some organizations benefiting more from governance activities than the clients (Warren et al., 1974). Therefore, it is suggested that there may be a darker side to bottom-up local public network governance (O’Toole & Meier, 2004).
In sum, the literature is not conclusive about the variables that predict network effectiveness. Different contexts may lead to different performances. As network contexts are reflected by network contingencies, these contingencies may help explain these inconsistent results (Andrews et al., 2008). In different conditions, different governance roles will prove most effective. A context can be considered as the set of contingencies of the network. Next, we discuss the main contingency factors and formulate propositions about their moderating effects, given the relationship between governance roles and the performance of local public networks.
Network Contingencies
Lawrence and Lorsch (1967) showed that there is no universal best way to govern organizations. The same is true with respect to governing local public networks. Empirical evidence suggests that contingency factors are likely to have a significant impact on the performance of local public networks (Andrews et al., 2008; Kastelein, Attema, Blindeman Krabbenbos, Klanderman, & Beljaar Spruit, 1977).
Mintzberg (1979) concentrated on the influence of the stability and complexity of a context on performance. Based on his research he identified age, size, and diversity as important contingency factors. In addition, he noticed the importance of stability in demand for products and services. Mintzberg formulated these contingencies at the organizational level. We propose to use them at the network level as well. Network age, network size, network diversity, and customization of services can be considered as contingency factors that influence the relationship between governance roles and network performance.
Next, these contingencies are elaborated and, based on an overview of the literature, propositions are formulated regarding the influence of these contingencies on the effects of governance roles on network performance.
Network Evolution
The age of a network is an important condition when examining its performance (Agranoff & McGuire, 1999; Provan & Kenis, 2008). Borgatti and Foster (2003) relate this factor to the evolution of the network. There are many different aspects within a network that evolve over the years. The content of relations between actors can change over time as a result of external and internal processes. Perceived trust among network members, shared institutional norms and values, the subject of communication and communication intensity can all be characteristics of different aspects of evolution.
The evolution of networks can also be assessed using quantitative measures, for example, with respect to the in- and outflow of organizations in the network. How many organizations entered or left the network during a certain period? What is the proportion of new network members in one year compared to the previous year? This way of assessing network evolution demands a longitudinal measurement of network composition. Mintzberg (1979) found that organizations with a stable composition favor top-down governance. In his view collaboration becomes more predictable when parties have worked together for a longer period of time. However, Mintzberg observed what is most common practice, which is not by definition the practice associated with best performance. Organizations are different from networks and networks may evolve differently (Korssen-van Raaij, 2006). Local governments are often confronted with complex and dynamic local networks. The entrance of new network members creates uncertainty because the performance level and contributions of these new members need to be evaluated. In these situations, mutual trust has to develop (Korssen-van Raaij, 2006). Mutual trust originates from past positive experiences and accrued reputation over the years (Larson, 1992). Maturation of the pool of network members requires bottom-up governance to be effective (Jehn & Mannix, 2001). If partners are working together for the first time, control mechanisms are needed because they do not know what to expect from each other (Hite & Hesterly, 2001). Potential gains of using a bottom-up role need time to reach their full strength (Korssen-van Raaij, 2006). Before this maturation state is attained, control by the lead organization (top-down governance) is the best way to create trust (Gulati & Singh, 1998). As a network evolves, the need for top-down governance roles decreases because of an increase in trust (Human & Provan, 2000). Therefore, we propose the following:
Proposition 1: In young public networks, local governments that use the commissioner role will achieve highest performance.
Proposition 2: In medium-aged public networks, local governments that use the coproducer role will achieve highest performance.
Proposition 3: In mature public networks, local governments that use the facilitator role will achieve highest performance.
Network Size
Network size can be considered an important determinant of the effectiveness of one governance role over another (Reuer, 2001). Network size is often operationalized as the number of organizations involved in the network. There is no standard classification of network sizes. How to classify the size of a local public network depends on the size of the local public networks active within the same policy field. Research suggests that a large size is associated with top-down governance roles (Dewar & Hage, 1978; Scott, 2003); as a network becomes larger, the complexity of the division of tasks, agreements about responsibilities and the obligations between the partners become increasingly blurry because of factors such as bounded rationality. It can therefore be concluded that with a larger network size, the need for formalization increases (Mintzberg, 1979). Using top-down mechanisms increases the span of control of the local government in the network. The additional complexity created by greater size requires more procedures and documentation and a further functional decentralization of tasks. This is consistent with the claim that, when relying on top-down governance, large organizations are more effective than small organizations (Haveman, 1993). There is an upper limit to the number of organizations that can be governed within a network. If local governments lack a clear picture of how many and which organizations are involved in the network, the number of organizations that need to be governed may exceed the span of control of the governing organization. In that case, a bottom-up governance role works better (Scott, 2003). In most local public networks, local governments are obliged to identify all network members connected to the aim of the network (e.g., disaster or watershed management).
In large networks, agreeing on goals, formulating boundary conditions, and monitoring performance is very time consuming, if not impossible (Gulati & Singh, 1998). Using the coproducer role or facilitator role in large networks may lead to an inefficient and ineffective use of critical resources and processes (Faerman, McCaffrey, & Slyke, 2001). Conversely, using a facilitator role or a coproducer role in small networks facilitates a greater responsiveness toward changing demands in the environment. Evaluating processes and outcomes on a mutual basis is feasible if the number of organizations involved is small. In a small network it is also easier to maintain contact with organizations on a frequent basis as well as to evaluate the contributions of all parties involved. This mutual evaluation may be less costly and more effective than exerting power through formal rules and procedures (Andrews et al., 2009).
Therefore, we propose the following:
Proposition 4: In large public networks, local governments that use the commissioner role will achieve highest performance.
Proposition 5: In medium-sized public networks, local governments that use the coproducer role will achieve highest performance.
Proposition 6: In small public networks, local governments that use the facilitator role will achieve highest performance.
Network Member Diversity
Local public networks consist of different types of organizations differing in nature, operational values, skills, and knowledge. These differences are relevant to how the network can best be governed. What processes underlie the effects of diversity and how do local governments need to manage these processes (van Knippenberg, De Dreu, & Homan, 2004)? These questions pose major challenges to network theory and practice. In principle, any dimension of differentiation can be studied. In practice however, most attention goes to demographic differences, educational background and functional background. With respect to diversity among members of a local public network, only the differences in functional background are relevant.
Examples of functional diversity in the context of local public networks are the following. In the Netherlands, local governments govern local network organizations operating in such diverse sectors as health care, social work and housing. All these sectors have their own attributes. The role of the local government is to connect these organizations to provide integrated packages of products and services to citizens. If the contributions of different sectors are considered to be equally important, this leads to a heterogeneous network. If one sector has a leading role and is more prominent in the network, this is likely to be reflected in a more homogeneous network.
According to Lawrence and Lorsch (1967), as diversity increases, so does uncertainty. An adequate response to uncertainty cannot be foreseen at forehand. Therefore, increasing uncertainty necessitates using bottom-up governance mechanisms.
On the other hand, if a network is more homogeneous, the behavior and outcomes of the network are more predictable. In a heterogeneous network the behavior of the other members is difficult to predict. These differences of behavior can be functional if the desired outcomes are unstable. Where there is a need for different expertise, skills, and other diversity-increasing attributes, bottom-up governance seems more suitable to achieve the highest community network goals (Lawrence & Lorsch, 1967). This leads to the following propositions:
Proposition 7: In highly diverse (heterogeneous) public networks, the facilitator role will achieve highest performance.
Proposition 8: In moderately diverse public networks, the coproducer role will achieve highest performance.
Proposition 9: In slightly diverse (homogeneous) public networks, the commissioner role will achieve highest performance.
Customization of Services
Customization of services is also an important contingency factor. The questions remain whether the services that have to be delivered are complex or simple, and if the demand for these services is of a stable or dynamic nature.
Local governments need strategies to forestall, forecast, or absorb uncertainty to achieve orderly, reliable resource flows (Oliver, 1990). Stability refers to a situation of constant demand for the same products and services. For example, concerning local government, it is important to explore the differences in citizens’ demands and to take the predictability and standardization of services into account (Boyne & Chen, 2007). Frequently changing citizens’ demands are best governed by bottom-up mechanisms. Bottom-up mechanisms make the network more flexible in overcoming uncertainties (Gater, Insull, Lind, & Seglow, 1966; Woodward, 1965). Highly stable service demands can potentially be standardized. For unstable service demands, each service needs to be customized and therefore requires creativity and innovative power. Such creativity cannot be regulated by the local government and therefore seems best managed by a bottom-up governance role.
The commissioner role is most appropriate for local governments if services for citizens are predictable and standardized. By establishing standards that produce predictability (Johnston & Romzek, 2008), top-down governance creates stability. A facilitator role enables local public networks to be “light on their feet” as an advantage over hierarchies (Powell, 1990). The flexibility offered by bottom-up governance roles is essential to responding quickly and adequately to the changing demands and needs of stakeholders. If local governments are unable to forestall, forecast, or absorb uncertainty, the coproducer or facilitator role fits best. Innovative power from network partners can be used to provide new ideas and opportunities for effective service delivery (Reagans & Zuckerman, 2001). Therefore, we propose as follows:
Proposition 10: In public networks with low customizable service, local governments that use the commissioner role will achieve highest network performance.
Proposition 11: In public networks with moderately customizable services, local government that use a coproducer role achieve highest network performance.
Proposition 12: In public networks with high customizable services, local governments that use a facilitator role achieve the highest network performance.
Coexistence of Contingencies
Contingencies are associated with the stability and complexity of a local public network (Thompson, 1967). Stability refers to the predictability of the activities that have to be performed by the network. Complexity refers to the degree of coordination required to let the network perform well. Stability and complexity of local public networks are key factors for performance (Andrews et al., 2009).
Although contingencies within network research are mentioned in current research, actual studies on network contingencies were mainly conducted in the 1970s. If governance roles fit the level of uncertainty connected to stability and complexity, performance can be expected to be optimal (Galbraith, 1973). In general, stable networks benefit most from top-down governance (Neuhauser, 1971). Dynamic and complex conditions require bottom-up governance roles (Khandwalla, 1973).
Stability and complexity can be seen as two independent dimensions. Combining stability and complexity creates a typology with four categories. Table 1 highlights how stable networks can be complex and dynamic networks can be simple. We will discuss the effects of coexistence of contingencies by elaborating the four types of networks, providing examples, and discussing governance roles.
Coexistence of Contingencies
Stable and Simple
Local waste policy in the Netherlands can be considered a stable and simple network. Local governments work with a small number of organizations that dispose of waste. Waste is divided into a small number of categories, for example, paper, glass, chemical waste (batteries, medicines), organic waste (vegetables, fruit), and residual waste. Categories of waste are dealt with by one or a few waste disposal companies. Local governments have a limited pool of organizations to select from for their networks. Waste disposal companies must operate according to strict regulations set by the national government. Since waste disposal is stable and simple at the local level, governments use a commissioner governance role to ensure the effectiveness and efficiency of the local waste policy. Relying on a coproducer or even a facilitator role may lead to decreased citizen satisfaction and deficiencies because organizations and civilians will need to be consulted, which is time consuming.
Dynamic and Simple
When a local government plans to create a business park, there is usually a dynamic and simple local public network available. Creating a business park is a nonfrequent event for local governments since local communities can generally accommodate just a few business parks, and such parks have a long lifespan. To create a network to establish a business park, governments usually select organizations that are specialized and experienced in this area. This allows for a standardization of complex issues. The requirements for the business park are defined by the network organizations and local government jointly. For example, it may be decided to only allow small-scale business activities thus excluding, say, chemical production. For the greater part, the network will include organizations that are interested in participating in the business park. Since these organizations have to meet the same criteria, the network is homogeneous.
In the process of developing the business park, parties have the right to several legal objection procedures. This makes the required course of action unpredictable (dynamic). Objections may constrain the realization of the business park, but only if they arise. As they arise, local governments must follow standardized procedures as required by the national government. In this way the response to possible objections is standardized (simplified).
There is a limited variety in demands for places in the business park, and requirements set by the local government create standardization. Therefore, the best fitting governance role for the local government is that of coproducer. Setting the requirements is done in collaboration with the network organizations, whereas actual governance is a top-down governance role since the network organizations are obliged to meet predetermined requirements.
Stable and Complex
Local disaster management networks need to be stable and complex to successfully respond to challenges associated with a disaster. If a disaster strikes, local public networks need to behave according to procedures to restore order (Moynihan, 2005). There are a limited number of network members that are assigned to complex services. Beforehand, scenarios are cooperatively designed to organize the use of services that need to be provided by specialized organizations of different types (stable). The specific requirements to fulfill the demand that may arise in case of a disaster are not easy to determine and therefore these specialized organizations have discretion when their specific skills are needed to address unpredictable requirements. It is clear for the different organizations which tasks they have to fulfill (stable), but the services they provide are complex and cannot be considered as standardized solutions.
Networks for disaster management consist of organizations that have been collaborating for a long time. For example, police or the military are expected to secure safety, hospitals are considered to tend to wounded people, firemen will evacuate buildings and sites, local broadcasting companies will provide emergency information, and so on. Therefore, the network is relatively small as well as very diverse. All parties have to be very flexible toward demands that may occur in a different way than expected beforehand. Local governments are in charge of coordinating the local network. These networks tend to be most effective when a coproducer role is used in which consensus between the local government and the other parties is necessary for effectiveness and efficiency.
Dynamic and Complex
Most local social policies can be characterized as dynamic and complex. Adequate, integral health care services (Korssen-van Raaij, 2006), for example, or providing social support (Span et al., 2009), both require dynamic and complex networks.
An example is local social support in the Netherlands. Local governments govern local public networks. In 2007, following the introduction of the social support act, the Dutch national government provided local governments with policy discretion to better balance the local support demand and the local support supply (Schalk, 2006). Local governments rely on old as well as new collaborations with other organizations in order to provide this support. Social support aims to enable every person to participate in society and live independently. People who need social support are individuals suffering age or disability limitations, or a chronic psychological problem. In providing social support, the demand is very diverse. Products and services need to be customized. To achieve this, local governments use governance roles to connect organizations in the areas of living, care, and well-being. Variety in demand requires variety in connections between organizations. This makes the network less predictable (dynamic). The heterogeneous composition of organizations unites knowledge and resources necessary to deliver highly customizable and complex service demands (complex).
Since providing local social support is dynamic and complex, local governments often use a coproducer or facilitator role to enhance the effectiveness and efficiency of social support services.
These examples demonstrate that various combinations of contingencies exist. The examples given are, of course, prototypical types. Local networks in comparable contexts, for example, social support networks, will vary in network size, network diversity, network evolution, and customizability of services. For example, variability in service demands and opportunities to standardize these services depend on the needs of local citizens. If disabled senior citizens all request the same housekeeping activities, services are easy to standardize.
Toward an Integrated Model of Local Public Network Governance
This article began with the observation that local governments are frequently unsure of how to best fulfill the role forced on them. Knowing whether local governments achieve or fail to achieve the network goal is not enough. Drawing on existing literature, we proposed how different governance roles may influence public local network performance. A model was developed to help explain why performance is at a particular level and how network performance can be improved.
The model in Figure 2 summarizes our propositions and includes different contexts for local public networks in which local government takes the lead and mandates policy. In these contexts, local governments can take the roles of commissioner, coproducer, or facilitator, on a continuum from top-down to bottom-up. The effects of these roles on network performance depend on several contingency factors. The commissioner role is expected to be most effective if there are many organizations, many new network members, low network member diversity, and if services are stable. The coproducer role is expected to work best if there is a balance between young and old network members, a moderate number of network organizations, moderate network member diversity, and if services are moderately stable. Finally, the facilitator role is expected to perform well if there are many old network members, few organizations, a high network member diversity, and if services are unstable.
Contingencies conditions leading to a positive effect of governance roles on network performance
Discussion and Conclusion
This article offers a theory-based model for examining the effects of governance roles on local public networks and specifies propositions on these effects. It provides a starting point for further research into local public networks characteristics and into the contingency factors that affect performance when specific governance roles are used. The model can be applied to a variety of local contexts. This has been demonstrated with examples of disaster management, waste policy, business park development, and organizing local social assistance. However, these examples are only intended to be illustrative. In practice, different governance roles are often used simultaneously within a network. Within disaster management, for example, responsibilities and broad task descriptions are set out in advance based on a commissioner role, but all organizations may rely on a coproducer or even a facilitator role if a disaster actually occurs (Moynihan, 2005).
The approach we have chosen in looking at governance roles is fairly innovative, and more specific than previous approaches, in that it specifies characteristics on several continua.
The contingency factors included in the model are derived from contingency theory literature. Nevertheless, the list of contingencies may not be exhaustive. Future research could broaden the model by looking at the role of other possible contingencies that might be relevant; for example, the local political culture, the local government form, or specific network member values.
This article has focused primarily on local public networks. However, the model presented may also be applicable to other lead organizational networks. At the level of country networks, such as the European Union, the effects of different ways of governing individual member states in a certain policy area may be influenced by the same contingencies as in local network governance.
The question is whether our propositions also hold if the lead organization is not a local government but a school organization or a public health service. McGuire (2006) states that there is a growing interest in determining the strength and influence of collaborative management. Rather than simply documenting the existence of different governance roles, we have elaborated the influence of different governance roles on performance in specific situations.
As said, the model and propositions can be tested further in future research. It would be interesting, for example, to examine the effects of different governance roles on local governments in changing circumstances. One such example pertains to the changes in responsibilities regarding social support in the Netherlands. In 2007, local governments became responsible for coordinating local support networks. All these networks are local public networks as defined in this article. Given the differences in governance roles used, the contingency factors, and network performance, this offers an excellent opportunity to test the propositions of the model developed in this article.
Footnotes
The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
The author(s) received no financial support for the research and/or authorship of this article.