Issue Networks Research Paper

1. Early Development

Initially, in the 1960s and 1970s, there were two distinct views of the relationships between interest groups and officials. ‘Iron triangles’ and ‘policy communities’ suggested that very closed ‘subgovernments’ of stable participants in long-term relationships effectively made policy in well-delineated policy fields such as agriculture or defense. In the USA, ‘iron triangles’ were composed of the interest groups active in a particular policy field, members of Congress (especially those sitting in the relevant committees) and officials (particularly unelected) in the relevant parts of the bureaucracy with powers in the field. These actors knew each other, were in constant interaction, and exchanged concessions, advantages, and information. They were a closed group, able to exclude outsiders. The concept of ‘policy community’ was used and developed in studies of Britain from the 1970s onwards (Jordan 1990). Policy formation was argued to be segmented into subsystems composed of policy communities of civil servants from government departments (or units within them) and interest groups. They were built on exchange relationships and a sense of ‘community,’ and possessed a set of dominant values (Richardson and Jordan 1979). Both iron triangles and policy communities paid close attention to interest groups with powerful and continuing material interests, the basis of exchange relationships.

The concept of ‘issue networks’ was developed by Hugh Heclo in opposition to the ‘iron triangles’ literature in the USA, which he argued monopolized attention on closure and subgovernments, but ignored forces for fragmentation and the development of new forms of interest groups (Heclo 1978). An issue network consists of a large number of issue-skilled ‘policy activists’ drawn from conventional interest groups and sections of the government, together with academia and certain professions but also comprising expert individuals regardless of formal training. Participants are constantly changing, and their degree of mutual commitment and interdependence varies, although any direct material interest is often secondary to emotional or intellectual commitment. No-one is in control of the policies and issues covered by a network and it is very difficult to identify clearly the dominant actors.

Neither policy communities nor issue networks offered a free-standing, developed model of policy making (Thatcher 1998). Rather, they represented enriching metaphors and useful concepts that were empirically applicable, and provided counterbalances to previous well-established views of policy making. They encouraged disaggregated analysis and drew attention to the effects of the fragmentation of government into subsystems, the interdependence of government and interest groups and increased specialization in policy making. However, issue network, iron triangle, and policy community literatures suffered from several limitations: they covered only specific categories of relations between groups and officials; they omitted key aspects of policy making, such as ideas, the distribution of power among actors, and change; and the factors explaining the emergence of different types of issue network or policy community remained general.

2. Network Typologies

In the face of these limitations, more general network approaches were developed. One path involved creating typologies of networks. The concept of ‘policy network’ or sometimes ‘policy community’ became generic, covering different forms of relationship between interest groups and officials arising from variations in network characteristics. Rhodes and Marsh (1992) offer a six-category typology covering a continuum from issue networks to policy communities and based on differences in the degree of integration, membership, and distribution of resources among members. Jordan and Schubert (1992) claim that different terms for state group relations in fact refer to variants of networks and hence they place terms such as iron triangles, clientelism, subgovernment, policy community, and various types of corporatism and pluralism in relation to each other using three ‘dimensions,’ namely number of members, whether the network is sectoral or transectoral, and the level of its stability.

In extending the scope of networks, the typologists pursue more ambitious theoretical aims. They seek to relate existing different terms and categories to each other within one overarching typology in order to avoid a jumble of overlapping and confusing terms. They claim to assist in ordering and categorizing information, permitting policy subsystems to be described, classified, and compared. Explanatory aims have ranged from specifying the key characteristics of policy networks in order to allow the systematic ordering of material, linking the ‘structural’ features of state and society and policy making (Atkinson and Coleman 1989), and offering a ‘model’ of the role of interest group intermediation based on different forms of power dependence embodied in the types of network (Rhodes and Marsh 1992).

Yet network typologies have suffered from considerable difficulties (Dowding 1995). Terminological disputes over whether the generic term should be policy network or policy community have created confusion. Placing differing forms of state-interest group relations under the heading of ‘policy network’ has not led to a common terminological currency. As the concept of policy network community became generic, definitions have become loose and frequently contradictory. The empirical applicability of network typologies has encountered important obstacles. Several of the ‘dimensions’ that form the basis of categories are difficult to operationalize, or are unlikely to generate agreement between different researchers; examples include the ‘degree of integration,’ ‘pre-eminence,’ and ‘serving the interests of producers’ (Rhodes and Marsh 1992). Categories are poorly defined and identifying a type of network in practice is difficult, since categories are not mutually exclusive and different types of policy network or aspects of several network types can exist in the same policy area.

3. Interorganizational Analyses

An alternative generalizing strategy has been to develop interorganizational analyses of networks. Drawing on social network analysis and work on interorganizational relations, they view policy making as an example of decision making by interdependent actors who engage in exchange relations (Knoke et al. 1996). Policy is seen as the product of complex interactions among actors drawn from both ‘public’ and ‘private’ sectors. Such actors are partly interdependent but also partly autonomous. They are linked horizontally without being part of a single organizational hierarchy; their relations are based on exchange, thereby producing policy networks, but nevertheless combine elements of conflict with those of cooperation (Marin and Mayntz 1991). Analysis focuses on the interaction of actors—the unit of analysis consists of the relations among network participants.

Interorganizational network analyses ‘map’ network structures, using qualitative and quantitative methods to reveal linkages between actors. Frequently, mathematical techniques derived from formal network analysis are used to establish the network position of actors and relationships amongst them, applying indicators measures such as intensity of communication, reputation among network participants, or resources. Mapping networks to provide a systematic picture of linkages serves several purposes. It draws attention to the interrelations and interdependence between public and private actors. Quantification enables complex patterns to be revealed that would be difficult to find because the events studied are too numerous, vast, and/or scattered and allows comparison of networks across sectors, countries, and time periods. The most important ambition of interorganizational approaches is to offer a ‘structural analysis’ by looking at patterns of network relations. It is insufficient merely to analyze the attributes of individual actors, as the influence of such attributes on behavior depends on the network of relations of each actor (Emirbayer and Goodwin 1994). Instead, relations amongst actors must be studied as a whole. A structural analysis of private and public actor configurations in policy formation identifies the position of actors and the ways in which networks influence actors’ capacity to use their resources. It therefore permits explicit consideration of the ways in which different structures create different mechanisms or logics in policy making.

The scope of interorganizational approaches is very broad. Thus, for example, policy network refers to ‘structures’ located ‘beyond or between policy markets and policy hierarchies’ that are composed of ‘relatively autonomous action units’ (Marin and Mayntz 1991). From an ‘organizational state’ perspective, Knoke et al. (1996) have conceived policy networks as relations linking organized interests with public officials, transforming heterogeneous communities of policy actors into webs of common benefit-seeking actors. To operationalize and apply in empirical studies, important choices must be made to narrow the scope of interorganizational networks. Similarly, applying quantitative techniques requires the choices over matters such as questions posed, categories created, the relationships to be analyzed, and variables to be controlled for. Social network analysis may embody a particular ‘theoretical orientation’ towards the structure of the social world, notably the need to study behavior as a set of relationships, but it provides a particular set of concepts and methods and a frame-work for testing arguments models rather than a substantive set of hypotheses or a specific body of substantive theory.

In response to criticisms of important omissions in the policy network literature, current research has seen diversification (Thatcher 1998). This has taken two forms. First, additional explanatory factors are incorporated into network frameworks, often with attempts to analyze changes in networks. Thus, for instance, modifications of the institutional framework within which network actors operate, the technological and economic features of a policy problem, and the ideas, values, and knowledge of network actors have been examined. Second, in search of greater explanatory power, network concepts have been linked to other models of policy making (Journal of Theoretical Politics 1998). Thus, for instance, using rational choice theories, networks have been analyzed as institutions, notably in their capacity to establish rules and long-lasting patterns of social relations for the interactions of actors in the policy process. They are argued to reduce uncertainty and transactions costs by aiding the creation of relations of trust that allow exchange. Ideas-based models, notably the ‘advocacy coalition framework’, have provided a different partner for network analyses. The effects of networks on coordination of actors under conditions of different forms of beliefs have been considered. Hence networks offer an intervening variable for the effects of ideas.

4. Conclusion: Constant Expansion

Analyses of networks have become increasingly broad: from modest origins as concepts of issue networks and iron triangles, they have expanded to overarching frameworks of relations between public officials and private interests.

Bibliography:

1. Atkinson M M, Coleman W D 1989 Strong states and weak states. British Journal of Political Science 19: 47–67
2. Dowding K 1995 Model or metaphor? A critical review of the policy network approach. Political Studies 43: 136–58
3. Emirbayer M, Goodwin J 1994 Network analysis, culture, and the problem of agency. American Journal of Sociology 99: 1411–54
4. Heclo H 1978 Issue networks and the executive establishment. In: Beer S H, King A (eds.) The New American Political System. American Enterprise Institute, Washington, DC, pp. 87–124
5. Jordan G 1990 Sub-governments, policy communities and networks: Refilling the old bottles? Journal of Theoretical Politics 2: 319–38
6. Jordan G, Schubert K 1992 A preliminary ordering of policy network labels. European Journal of Political Research 21: 7–27
7. Journal of Theoretical Politics 1998 Special issue on policy networks 10
8. Knoke D, Pappi F U, Broadbent J, Tsujinaka Y 1996 Comparing Policy Networks: Labour Politics in the U.S., Germany, and Japan. Cambridge University Press, Cambridge, UK
9. Marin B, Mayntz R 1991 Policy Networks: Empirical Evidence and Theoretical Considerations. Campus Verlag, Frankfurt am Main, Germany
10. Richardson J J, Jordan A G 1979 Governing Under Pressure: The Policy Process in a Post-Parliamentary Democracy. Robertson, Oxford, UK
11. Rhodes R A W, Marsh D 1992 Policy networks in British politics: A critique of existing approaches. In: Marsh D, Rhodes R A W (eds.) Policy Networks in British Government. Clarendon Press, Oxford, UK, pp. 1–26
12. Thatcher M 1998 The development of policy network analyses. Journal of Theoretical Politics 10: 389–416