144 Policy Network Analysis
1978) who direct, coordinate or control material or immaterial resources in o policy field of common concern. The membership and central positions within the network are constantly negotiated and embattled among existing and potential participants. Networks gain stability over a longer period of time if stable expectations concerning calculable exchange relations are developed. The exchange of resources between formal and informal actors within the network is limited and is to a certain extent structured by institutional rules, organizational structures, contracts, informal routines, as well as the governance structure of the network. The latter is defined by political decision and describes the specific policy instruments which are used in the field, such as conunand-and-control strategies, negotiation mechanisms or market incentives.
In doing so, the governance structure of a network shapes the incentives of the participants and induces the latter to act in a specific way. Moreover, the range and nature of exchange processes, and the Hnuting and enabling role of institutions are only to be understood within the context of a specific policy field which lays open particular opportunities of action. Furthermore, interaction within the network is influenced by a shared understanding of (he adequate mode of problem solving and an adequate style of interaction among actors.
In the first section of following contribution to this volume I shall outline the main theoretical questions linked by policy network analysis. In the second part, I will discuss in which way and to what extent policy network analysis can be used in comparative political research. In a third section, the British and U.S. clean air policy innovations of 1990 will be analyzed to illustrate the network-analytical approach.
7.1 Policy Network Analysis: Linking Different Theoretical Approaches
Four different theoretical backgrounds come to bear in policy network analyse:
o) Rational choice theory in the shape of resource exchange and resource
dependence theory, as developed by Aldrich and Pfeffer (1976);
b) the new political institutional ism. as outlined by March and OlMa
(1989);

c) symbolic interaction theory, as elaborated by Bums and Flam (1987) for the organizational context;
d) public policy analysis (Windhoff-Heritier, 1987). The aforementioned four strands of theoretical and conceptual thinking are combined within the framework of policy network analyaii in order to explain the development of policies: resource dependence theory assumes that rational actors, each in a position of relative autonomy, interact in a policy field because they depend upon each other to produce a policy (decision) under an overarching collective purpose. Within their own organizations they cannot create all resources and functions required for the policy production, therefore they are required to enter into transactions with other actors in the network. The term of "generalized political exchange' has been used (Marin, 1990) to give expression to roe fact that this exchange may be multilateral and indirect (through circuits of) as welt as bilateral and, above all, that it may extend over a variety of resources (Mayntz, 1992: 25).
The actors are engaged in a 'common enterprise* (DiMaggio/Powell. 1983: 149). Membership in the network is. at all times, an object of controversy and bargaining between the dominant actors of the network and those who want to gain influence in the network, or enter the network by offering resources important for the 'problem solution' in the policy field. Each actor, public or private, is interested in maximizing his resources, material or immaterial, in order to influence the outcomes of the policy decision process, to guarantee his survival and to stabilize the relationships with other actors. On the one hand, the resource dependence model portrays the actor-organization as responding to the organizational environment, but, on the other, it also depicts it as an active entity capable of changing the terms of interaction (Aldrich/Pfeffer. 1976:
83).
Since the organizational environment does not impose strict requirements for survival of the organization, there is a range of choices and strategies available (Child, 1972). Therefore, the criteria by which decisions are made within the organization also become problematic. Internal power differences are important and the influence of internal subunits, in interaction with the demands of various external groups in the network, may come to determine the outcome (Aldrich/Pfeffer. 1976:84). Contingencies in the network environment and ensuing uncertainty are absorbed by forming coalitions with

other organizational actors, through merger, the movement of personnel among organizations, ('interpenetratioii' cf. Laumann/Knoke, 1978), tacit collusion, or legal contracts. Thus, linkages are also structured by organizational interdependencieft such as interorganizational bodies or the multi-membership in various organizations of single actors (Laumann/Knoke, 1978: 463). The linkages and contracts serve to minimize transaction costs (Williamson, 1985), i.e. the information costs and costs which may be incurred in exchange relations if the interacting partner proves to be unreliable, or to be 'shirking'. For example; actor A may have invested considerably in the product to be exchanged with actor B, but the tatter is opportunistic and fails to pay the agreed price.
The existing linkages between the actors are analytically relevant because they constitute the structure of the network. This structure may be analyzed under aspects of centralization and decentralization, the number of existing sub-networks and their horizontal interdependence. Existing linkages reveal network behavior, the power structure of a network at a given time and the coherence of a network. They are characterized by different degrees of intensity, frequency, fonnalization and standardization of interaction (Laumann et al., 1978: 465), What is 'transported' in these linkages are material and immaterial resources such as financial means, information etc. (Tichy/Fombrun 1979:927). The exchange of resources is to an important degree influenced by the dominating governance structure, that is, the dominant mode of guidance and control are defined by the policy instrument used. The latter may have the nature of incentives, negotiation patterns or command-and-control strategies. A governance structure offers reference points to the individual actor which allow the decision maker to weigh up of costs and benefits of particular courses of action (D5hler, 1989: 350).
The existence of governance structures and formal interorganizational linkages indicates that within the network resources are not freely exchanged, rather than revealing that their exchange is restricted by institutional rules. That is where the new political institutionalism comes in. This approach emphasizes that not only rules and traditions, but also routinization and 'sunk costs' of customary exchange relations set limits to the rational resource exchange decisions of the network actors. So do, of course, existing formal institutions, such as given interorganizational contracts defined by the constitution, by law or administrative ruling. It was above all the new statists (Evans et al., 1985;

Krasner 1988) and some organization theorists (March/Olsen, 1989) wh( emphasized the importance of state structures, institutions and rules in the shaping of policy decisions by trying to stem the tide of exclusively economic explanations of policy decisions. They underline the importance of the 'rule o. appropriateness', according to which decisions in unstructured situations ue fitted to existing rules, traditions etc., as opposed to the rule o, 'consequentiality', where decisions strictly follow cost and benefit calculations (Olsen, 1991).
In the institutionalist view, political and policy decisions must be in part derived from political institutions as 'irretrievable sources' of political action, institutions being defined as interconnected rules and routines which define the adequate action as a relationship between a role and a situation (March/Olsen 1989). From this point of view, politics is only in part to be understood as a rational and consequence-oriented process, and is also to be seen as process oriented in itself: decision processes are just as much concerned with the attribution of status as with the definition of truth and virtue, and the maintenance of loyalty and legitimacy (March/Olsen, 1975: 12). Rules, such a.^ standard operating procedures,
"...affectthe substantive outcomes of choices by regulating the access of participants, problems, and solutions to choices, and by affecting the participants' allocation of attention, their standard of evaluation, priorities, perceptions, identities, and resources." (Olsen 1991:93)
It would be a mistake, however, to consider the resource exchange approach and the institutional approach as mutually exclusive. Rather more, they are closely linked, simultaneously enabling and limiting each other, for choosing among different courses of action according to one's preference is only possible if there are a limited number of options. This limitation is given through the institutional context restricting the range of choice of possible alternatives of action by setting rules, creating organizational hierarchies etc. Vice versa, institutions may be understood as developing from individual choice decisions:
bi-lateral or tri-lateral contracts may be drawn up, or hierarchical organizations are established in order to save transaction costs (Williamson, 1985). However, once institutions are in place, they tend to develop a certain inertia, an interest

01 themselves, and unfold their own dynamics; they go beyond the scope of individual economic decisions.
"...Institutions have such wide span in time and space that they evade the control of individual decisions' (Giddens, 1988: 78;
see also: Wmdhoff-Heritier. 1991)
1 Interactions amongst network actors are not only guided by the actors* rational
interests augmenting their resources through exchanges with other actors, as well as by the limits set to these exchanges by organizational and legal rules, but also by, if you will, sorter rules. A common understanding about the appropriate approach of problem solving is circumscribed as 'the operating ideology' (Sharpe, 1990), or as a 'belief systems* defined as a "configuration of ideals and altitudes in which elements are bound together by some form of constraint of functional interdependence" (Converse, quoted in; Czada, 1991). Bum and Flam (1987), finally, see it as the appropriate way of dealing with each other, as the 'social role system*.
These collective belief systems are, of course, closely linked to existing institutions, for institutions are created as a result of interests, dominant ideas and the perpetration of the value systems linked to them (Czada, 1991). The "oocial rule systems", if they are backed "...toa greater or lesser extent by social sanctions and networks of power and control ...are referred to as rule regimes" (Bums/Flam 1987: 13). Since social rule systems are collectively shared, they permit supra-actor descriptions and analyses of the patterning of social traiuftctionA and social structure m tlie sphere to which they apply. Thus, for example, prevention has become the dominant idea and value system in fighting juvenile crime while punishment has increasingly been questioned as an appropriate problem approach. Or, to give another example, 'compliance by consensus* has for a long time been regarded as the appropriate way of inspectors* dealing with industry in British environmental policy.
Finally, policy network analysis uses the concepts and classifying patterns of public policy analysis. Policy analysis conceives of political and policy decision making as a dynamic process which follows specific stages of the policy cycle, although not necessarily in on orderly manner. Thus, the phases of problem definition, agenda setting, policy formulation, policy implementation and policy feedback are distinguished in order to explain the emergence and

transformation of policies and the underlying political processes. Although all actors who are engaged in the policy field at one or another of those stages 'belong* to the network, their relative importance and influence may vary depending on the phase of the policy cycle; so do the dominant coalitions and cleavage structures, that is, the 'policy arena', which is structured by the expected costs and benefits of a specific policy (Windhoff-Heritier, 1987). Accordingly, the transformation of the policy network in each phase needs to be investigated.
Moreover, policy analysis assumes that the nature of the policy at hand. the specific features of the problem to be solved, is systematically related to features of the policy network. Thus, it is expected that redistributive policies tend to produce polarized cleavage structures in the policy arena of the network;
they pit those who benefit from a policy against (hose who have to bear its costs and therefore give rise to a polarized network structure, with this being so it is assumed that redistributive policies only have a chance to be implemented, il they are based on *command-and-control', i.e.: precise regulation followed b) sanctions in case of non-compliance. Hence policy strategies and the specific opportunities of action open to the network actors can only be understood in the context of a particular policy and the specific policy instruments used.
Policy networks often emerge during historical 'watershed situations' (Skowronek, 1982). When under 'external shocks' such as wars or economic crisis, specific patterns of interaction between the stale and society arise. In these situations, the state often plays an active role in creating specific network structures, or structures of interest intermediation (Lehmbruch, 1991). Under state leadership policy networks are created in new fields of activity such a*-nuclear power policy (Czada, 1991), although the state need not necessarily play o dominant role in the field, once the new developments have taken off. New networks gain stability because the actors involved with each other are regularly interacting over a longer period of time and are evolving calculable exchange relations (Dohler, 1989) which are based on past experience. Subsequent polic} changes tend to develop within the structures of the existing network ('path-dependency') which reveal a considerable resistance towards ne^ innovations.
To briefly summarize, policy network analysis explains the development of public policies in terms of complex exchange and transaction processes between network actors within given institutional restrictions. The structure and

dynamics of these interactions vary in the different phases of the policy cycle. Networks as analytical tools are conceived for specific policy areas'. The policies under discussion and the policy instruments used offer specific incentives to the concerned actors who, simultaneously, in their interaction with other network actors are guided by a specific problem solving ideology and 'social rule system'.
7.2 Policy Network Analysis in Comparative Research
How can the instruments of policy network analysis be used in comparative political research? What are its advantages and its limits? Comparative politics faces a general dilemma: it perpetually has to navigate between Scylla and Charybdis. On the one hand, it must avoid the trap of ending up with idiosyncratic, highly detailed 'thick descriptions' of single cases; yet, on the Other, it should not limit itself to collecting highly aggregated cross-national data such as the number of unemployed, or the resources spent for welfare purposes and so forth. In the first instance, much insight and information is gained about the political decision processes and policy features of a specific country, yet one becomes hopelessly entangled in infinitesimal details and all attempts of comparison between countries have to be abandoned. In the second case, comparable macro-data is at hand. However, we know very little about the processes that lie behind this data; it remains, beyond the scope of analysis, *a black box*. Clearly, policy network analysis is prone to make the first mistake. How can this error be avoided, or how can the danger of ending up with idiosyncratic case studies at least be managed?
If two or more countries are compared, one can proceed in two phases. Following "the most similar systems approach" (Teune/Przeworski 1970), the units which are compared are characterized by a large number of similar independent variables, such as the level of economic and technological development and the basic political institutions and so forth; the compared units are different by only a smaller number of variables which - it is assumed -
explain part of the cross-national differences in network processes and their policy outcomes. The independent variables may be of either short-term or long-term 'nature. Departing from the "most similar systems approach", the first phase 'takes stock' of central common and distinct features of the compared networks (standard comparative independent variables).
The policy networks of the research units may also be compared with regard to their outputs-outcomes-impacts, which are treated as dependent variables. The question is: how do differences in the structure and the operation of policy networks explain different policy results? The latter may differ on the level of policy outputs, that is political decisions, (e.g. on the allocation of resources), on the level of policy outcomes (first tangible implementation activities), or the level of long-term policy impacts (i.e. is changes in the state of the environment, human behavior or human health). Naturally, the first and the second are much easier to measure and to relate back to the explanatory variables.
The following comparative independent variables are commonly used to describe important features of the analyzed units or nation-states (Feick/Jann, 1988). Long-term features of political, economic and socio-cultural development:
the level of economic and technological development
the basic features of political institutions
long-term values
the central features of the policy problem at hand (such as the high
technical complexity of the problem to be solved, the redistributive
character of policy etc.) Short-term features of political, economic and socio-cultural developments:
changes in the world-economy: recession or boom
external shocks such as
international political crisis
crisis in the natural environment
technological innovation.
The above common long-term and short-term factors, typical for all units under investigation, facilitate the comparability of countries (or sub-national units).
By contrast, distinctive long- and short-term features, such as the following, may explain differences in policy outcomes.

Long-term country-specific political, economic, socio-cultural variables are;
specific aspects of the political institutions (presidential versus i parliamentarian democracy; bi-chamber system or one-chamber system;
unitary or federalist system, electoral system etc.)
the role of the courts in the political process
the party system (e.g. polarized or not, multi-party etc.)
the associative structure (the number and structure, e.g. centralized or
decentralized, of organized interests, such as manufacturers' and
employers' associations, trade unions, workers' organizations,
environmental organizations etc.)
(he patterns of public-private cooperation in the implementation of
policies
the policy instruments used in the policy field of research Short-term country-specific variables are:
electoral outcomes
changes of parties in government
crisis of the national economy; structural crisis? the national economy
national crisis in the natural environment
national cultural 'shocks'
Long-term versus short-term variables are used in order to distinguish the possible effects of short-term independent variables in a longitudinal analysis of policy networks. This first stage, describing structural differences and similarities of the units under comparison (standard variables of cross-national comparison), provides an impression of structural aspects and typical patterns of the policy network. The standard variables can be matched to the routine checklist of questions used for the description of networks:
a) The standard comparative variable, 'formal political and administrative institutions', corresponds with the description of the formal ttnicture of the network. It describes the particular formal powers which actors in specific positions have specifies the nature of their formal relationships. It also describes the degree of centralization or decentralization of the formal institutional structures within the network.
b) The description of all 'consequential actors' within the network includes the formal positions described under a), goes beyond this by including oII actors disposing of resources relevant for the policy problem, even though

they may have no formal powers. It is in part covered by the standard variable describing public-private patterns of cooperation in policy formulation and implementation.
c) The network feature analyzing coalition structures of the network by which actors try to influence the outcome of policy events by pooling their resources has no corresponding standard comparative variable. Neither does the description of the ceotrality of actors which is not exclusively based on formal powers, but other material and immaterial resources.
d) The network feature 'governance structure' corresponds to policy instruments used in the policy field under investigation.
e) The dominating 'operating ideology' guiding the problem solving approach in the field is, to some extent reflected in the analysis of the dominating social and cultural values in this specific field of public intervention.
A comparison of the network-analysis checklist questions with the standard comparative variables shows that the latter partially cover the first. Only partially, because one important aspect of network analysis focussing on the dynamics of interaction between formal and informal actors, their exchange processes, tlieir coalition building, their conflicts and the ensuing cleavage structures cannot be grasped by the standard variables. This is the aspect of the networks that one author had in mind when he said: "Network analysis is like trying to take a picture of the Mississippi River". Yet, the use of a grid of 'objective' network aspects (or: standard variables) may serve as a starting point for the analysis of the particularities of policy network dynamics. Often enough, the distinctive structural network aspects may be the very reasons why the dynamics of the network processes develop in a specific way and, in consequence, produce specific policy outcomes.
7.3 Policy Innovations In Clean Air Policy Nenvorks:
Comparing British and American Clean Air Policy
How can comparative policy network analysis explain why and how an important environmental policy innovation came about in both Britain and the United States, independently? In IfWO.the British Parliament passed the Environmental

Protection Act; also in 1990, the United Stales Congress enacted the new Clean Air Act. Both laws, the dependent variable to be explained in terms of policy outputs and outcomes, are comprehensive and ambitious in scope and objectives. It is not so much the details of the laws which are of interest here, as incisive as they may be to both countries, but the very fact that an innovative departure in policy making occurred in both countries.
In the policy case to be discussed, the comparative analysis - on the basis of the roost similar systems approach - shows that the network processes that unfolded in the two countries and brought about the innovation may be partially traced back to distinctive network structures. With respect to basic long-term system variables, the United States and Great Britain fit within the context of the "most similar systems approach" i.e. .both countries have a simitar level of economic and technological development and they both have democratic political institutions and similar socio-cultural value systems. However, there are significant differences in the finer structure of their political institutions as well as their problem solving approach in environmental policy-In the British system, which on account of its unitarian political and administrative structure may qualify as a centralized network, new initiatives in environmental policy making flow, as a rule, from the executive, or administrative, leadership of the central government. Initiatives also may be developed at the local level, but since the central governments legal and financial grip on local authorities tightened during the 1980s, there has been less room for action by local governments. If central government is unwilling to respond to local initiatives, it can repress the latter more easily than before, and
o policy stalemate may ensue. This was the case during the 1980s under the Thatcher government.
By contrast, the U.S. federalist and decentralized policy network offers
o multiplicity of political arenas. The single states can use their far-reaching powers of decision making to develop new policy initiatives, even if the Federal government has decided to take no action in a specific policy field. This was the case under the Reagan administration in the 1980s. Access to the court system always offers the possibility of changing the political arena.
In the British case, the question therefore arises, where did the innovation emerge from? In retrospect, it may be demonstrated that the absence of subnational, powerful political actors in Great Britain, was however

compensated by its erobeddedness in the supranational decision structure of European Community. It was mainly the European Community which broi pressure on the British government to change its environmental policy in n significant respects. As one official said: "The influence of the Europ Community on British environmental policy can hardly be overestimated in past years'(Interview. London December 1991, Her Majesty's Inspectorate Pollution). Policy objectives and policy instruments, as applied in some of other member states such as Germany, functioned as a mode) in shaping Community and in turn Great Britain's policy.
This pressure "from above' was complemented by growing polil mobilization and activities deployed at the grassroots level by environrnei organizations. Interestingly, these political actors increasingly use supranational institutions, such as the Community's Court and the Commissi as leverage points in order to push for innovations in environmental po Thus, local groups and environmental associations, such as 'Friends of Earth', have been directly addressing the Commission to secure implementation of the Community Directives by local authorities. Both fad the Community legislation and the environmental movement contributed to modernization of British environmental policy as it was enacted in the Environmental Protection Act.
In the United States the pressure from grassroots movements has been a political factor in the policy network for quite some time. Howeve gained additional weight during the 1980s in view of the stagnation environmental policy during the Reagan administration. Local actors develo their initiatives in conjunction with single states. Support was sought from Courts which were addressed by members of environmental groups. Within a years the activities unfolded by single states, California foremost among th served as examples which were followed by other states. They developed sui political momentum that the Federal government found itself under pressun engage in new environmental activities. As a result, the new administrat under President Bush decided to push for a significant amendment of the Clean Air Act.
In both countries the policy innovation was, to some extent, favore* a change in the occupation of governmental key positions. In the U.S. advent of a new Republican administration brought about a change in cen

Protection Act; also in 1990, the United States Congress enacted the new Clean Air Act. Both laws, the dependent variable to be explained in terms of policy outputs and outcomes, are comprehensive and ambitious in scope and objectives. It is not so much the details of the laws which are of interest here, as incisive as they may be to both countries, but the very fact that an innovative departure in policy making occurred in both countries.
In the policy case to be discussed, the comparative analysis o on the basis of the most similar systems approach - shows that the network processes that unfolded in the two countries and brought about the innovation may be partially traced back to distinctive network structures. With respect to basic long-term system variables, the United States and Great Britain fit within the context of the "most similar systems approach" i.e.,both countries have a similar level of economic and technological development and they both have democratic political institutions and similar socio-cultural value systems. However^ there are significant differences in the finer structure of their political institution!; as well as their problem solving approach in environmental policy.
In the British system, which on account of its unitarian political and administrative structure may qualify as a centralized network, new initiatives in environmental policy making flow, as a rule, from the executive, or administrative, leadership of the central government. Initiatives also may be developed at the local level, but since the central governments legal and financial grip on local authorities tightened during the 1980s, there has been less room for action by local governments. If central government is unwilling to respond to local initiatives, it can repress the latter more easily than before, and a policy stalemate may ensue. This was the case during the 1980s under the Thatchcr government.
By contrast, the U.S. federalist and decentralized policy network offers a multiplicity of political arenas. The single states can use their far-reaching powers of decision making to develop new policy initiatives, even if the Federal government has decided to take no action in a specific policy Held. This was the case under the Reagan administration in the 1980s. Access to the court system always offers the possibility of changing the political arena.
In the British case, the question therefore arises, where did the innovation emerge from? In retrospect, it may be demonstrated that the absence of ntmdonal. powerful political actors in Greati Britain, was however

compensated by its embeddedness in the supranational decision structure of the European Community. It was mainly the European Community which brought pressure on the British government to change its environmental policy in many significant respects. As one official said: "The influence of the European Community on British environmental policy can hardly be overestimated in the past years' (Interview, London December 1991, Her Majesty's Inspectorate of Pollution). Policy objectives and policy instruments, as applied in some of the other member states such as Germany, functioned as a model in shaping the Community and in turn Great Britain's policy.
This pressure 'from above' was complemented by growing political mobilization and activities deployed at the grassroots level by environmental organizations. Interestingly, these political actors increasingly use the supranational institutions, such as the Community's Court and the Commission, as leverage points in order to push for innovations in environmental policy. Thus, local groups and environmental associations, such as 'Friends of the Earth', have been directly addressing the Commission to secure the implementation of the Community Directives by local authorities. Both factors, the Community legislation and the environmental movement contributed to the modernization of British environmental policy as it was enacted in the new Environmeatal Protection Act.
In the United States the pressure from grassroots movements has also been a political factor in the policy network for quite some time. However, it gained additional weight during the 1980s in view of the stagnation of environmental policy during the Reagan administration. Local actors developed their initiatives in conjunction with single states. Support was sought from the Courts which were addressed by members of environmental groups. Within a few years the activities unfolded by single states, California foremost among them, served as examples which were followed by other states. They developed such a political momentum that the Federal government found itself under pressure to engage in new environmental activities. As a result, the new administration under President Bush decided to push for a significant amendment of the old Clean Air Act.
In both countries the policy innovation was, to some extent, favored by a change in the occupation of governmental key positions. In the U.S. the odvent of a new Republican administration brought about a change in central

positions of the Environmental Protection Agency, the chief bureaucratic actor in tho network, which facilitated policy innovations. What we can see, is that -due to the distinctive institutional network structures - the pressure from below (and from above in the case of Britain) is translated differently into formal political power in the two countries. In the case of the U.S., initiatives of grassroots movements, local environmental organizations, which were supported by tho courts, are channelled and shaped by single states. Subsequently, the cumulative impact of the pace-setter functions of several states brought pressure to bear on federal policies.
The events in California serve to illustrate how a modification of the governance structure changed the coalition structure and exchange processes in a subnetwork. The relative inactivity of the Reagan administration set in motion a conflict between California and the Federal government, in the course of which environmentalists from California literally forced E.P.A.to implement the Clean Air Act more strictly. A key role was played by the courts, a very important latent actor in the policy network because citizens can turn to the courts to sue the agency for inaction (Bryner, 1984; 318). In California, in 1984 a member of an environmentalist group sued E.P.A. for failing to implement the Clean Air Act in Greater Los Angeles. After three years, the court decided in favor of the litigant. This judgment sparked off a change in the regional policy network and initiated developments which had repercussions in other states, encouraging them to also undertake policy initiatives, by-passing the Federal Bovemment, and thus also changing the clean air policy in the national network.
In Great Britain, by contrast, initiatives from below used the European Community as a vehicle in order to influence national policy. Quite independently, however, the pressure from above, from the supranational level, has to be considered as a factor of influence in its own right. A closer look into the policy-network in Britain also reveals how - with the new institutional structures - the stakes in the decision process in clean air policy have changed, The establishing of 'public access* in administrative procedures opens new possibilities of access to the network and new possibilities for environmental associations to bring their resources to bear in the network exchange processes.
'Public access' means that for the first time the new Environmental Protection Act opens the administrative decision process to the public which is offered extensive information about the applications of process operators and

subsequent bureaucratic authorizations. 'Public access' means, for instance, that the application for operation is listed in a 'public register' and responses to the application can be made within twenty-eight days. Additionally, the applicant must advertize the application in the local newspaper and again the public can comment upon the application during the twenty-eight days following publication.
With 'public access', a kind of leverage is offered to environmental organizations so that they may critically scrutinize the administrative authorization procedures and play a critical role of surveillance in clean air administration. Even more so, since all industrial plants also have to publish their emission monitoring data in the 'public register'. However, after a year and a half of practical experience with the new act, public interest in using these opportunities are not yet very pronounced. Rather more the new possibilities have been taken advantage of by industrial competitors of the applicants which try to gain information about product components and the quantities produced by their market competitors. Yet, the Department of the Environment, by means of a very active information policy, attempts to increase public interest in clean air policy by daily advertizing air quality measurements in the news, by offering telephone advice for pollution problems and distributing abundant information material. Even if it will take some time for a more active participation of environmental organizations and citizen groups to materialize, the institutional possibilities of public information offered by the Environmental Protection Act is - by comparison with other countries - far-reaching indeed.
The network dynamics producing the new legislation also brought about changes in the traditional governance structures in the two countries. In the United States, where command-and-control strategies have been the central strategy since the 1970s, even stricter elements of hierarchical guidance and control were established; however, extensive market incentives were also introduced in some fields. In Britain, with its tradition of 'consensus by compliance', we find slightly more 'command-and-control' and precise standard setting, more accountability of inspectors' activities and less negotiating between inspectors and firms and. as described, the opening up of the system for 'public access'. Linked to the change in governance structure, we therefore find a partial change of the 'social rule' system in Britain. The style of interaction between the actors and the problem-solving approach have both been redefined. In the United States this change has been more moderate. The basic problem approach remain; the same, but was however enlarged by the completely contrasting governance type of market strategies.
If we relate the analysis of the two case studies back to the above set of independent variables, it becomes clear that long and short-term distinctive features of the two policy-networks go some way in explaining why innovations were brought about in the clean air policy of both the United States and Britain. In summary, the following long-term distinctive institutional features seem the most important: the federalist structure in the United States and in the British case the embeddedness into the supranational network of European environmental policy which set into motion the dynamics of innovation. In the United Kingdom, the European Court also plays an important role since national and subnational actors gained new avenues of influence by having access to it. In the United States, the courts also had an important impact in Bpeeding up the decision-process in California, which, in turn, triggered off innovations in other sub-networks and, finally, in the entire national network.
As to the short-term variables, the electoral outcomes in the United States, although not involving a change in the governing party, offered the possibility of choosing new leading personnel with new policy ideas (head of the E.P.A.).In the case of Britain, the relatively high percentage of votes obtained by the Green Party in the European election put Margaret Thalcher's government under considerable pressure to take some action in environmental policy.
Not surprisingly, the state of the national economy had, and has an important impact on network innovations in environmental policy. The new Clean Air Act, in its most ambitious version was shaped before the on-set of the recession in the United States in 1989/90, subsequently was diminished in its requirements during the economic down-turn. The British Environmental Protection Act was passed before the recession hit Britain.
'External shocks' clearly played a role in the United States spurring legislative action, e.g. the Prince William Sound accident with the oil spilling of the Exxon Valdez which sharpened public attention for environmental issues in general. Also, the rapidly deteriorating air quality in California, especially in the Los Angeles Basin ("Los Angeles is fighting for breath"), gave rise to alarmed policy reactions in the state.

Thus, going through the standard comparative variables of policy fields in the United Stated and Britain, important simil;
differences in the network structure become apparent which - to a I - explain policy innovation.
7.4 Conclusions
In summarizing we may say that a systematic comparison of polii processes can be implemented, to the extent that common and distin and short-term features in formal network structures are controlled. I framework of comparable and distinctive, mostly structural, netwo specific dynamic processes unfold which are less accessible to a c-analysis. Yet, aspects of the basic network structure to a consider! explain the specific nature of the network processes, the resul) developments and their results.
A longitudinal cross-national comparison of policy networks i insightful because it permits the identification of country-specific i and convergences of policy patterns and the effects of short-term ir variables or events. It reveals cross-national diffusion process application of social rules and technologies in problem solving over of time, as well as their impact on the underlying political patt< supranational institutional influences may be identified, such as the EC directives on national policy patterns of the member states, as impact of economic changes in the world economy. 'External shocks' innovative phases or 'watershed phases' in policy making which in ti a restructuring of the network.
Despite the aforementioned limits to comparison, polic analysis offers considerable advantages: It systematically links all 'coi actors' in a policy network, be they public or private, and - in the cas actors -on all political and administrative levels. By including private approach draws attention to non-state organizations employing prohl Btrategies which present competing strategies to those of the state. 1 o case in point, the mass media with their activities may also play an

role in clean air policy and BipuficaoUy enhance the effectiveness of state measure*; i so may alao wholesale trade in the field of waste-disposal policy.
Furthermore, by comparing different networks under the question how ors similar policy problems solved In different countries or subnatlonal units, policy netwoik analysis draw attention to equivalent functions. In the field of clean air policy, for instance, SO3 (sulphur dioxide) standards, as they are released into atmocpoere by coal-fired power plants, are set by administrative decision in Gennany. However in Great Britain, they were for a long time decided by the Quef Inspector who set them individually for each plant under his jurisdiction, Therefore, the British Chief Inspector was the functional equivalent of the SO2 standards in other countries3.,
At the end of the day, we may conclude that whomsoever expects stringent causal evidence, rendering possible precise 'technological' policy recommendations, from comparative policy network analysis will be disappointed. But whoever, more modestly, makes an attempt to answer the question: which policy features and actor patterns, which contextual variables do countries have in common, that. with reasonable success, solve specific policy problems may consider policy network analysis as a useful analytical toot of comparative politics.

• [ На главную ]
• [ Библиотека ]
• [ Reader ]
• [ Программы ]
• [ Абитуриенту ]
• [ Фотоальбом ]
• [ О ФП ]
• [ Гостевая книга ]
• [ Ссылки ]
• [ Обратная связь ]