PUBLIC POLICY - THE ISSUE OF IMPLEMENTATION

What is public policy?

"Policy," of course, is a term with a rich variety of definitions and nuanced elaborations. We continue with the Laswell and Kaplan (1950: 71) usage, where policy is a "projected program of goal values and practices," because this formulation includes both the initial aims and supplementary procedures. Public policies are thus rules for action which directly or indirectly affect the whole population of a country and which usually are established by statutory authorities. As the comparative study of subnational program implementation has demonstrated, public policies are continuously reinterpreted and modified, simplified, and/or restricted at successive levels of government (Altenstetter and Bj6rkman, 1978). New rules for actions may supersede or coexist with old be- havioral parameters; new goals may emerge from a change in value perspectives. In addition, changes in public policy at one level of government influence per- formance at other levels.

THE ISSUE OF IMPLEMENTATION

By implementation we mean the process by which decisions are carried out by actors and organizations. This process is conducted by the numerous public and private organizations that translate national objectives into concrete outputs, i.e., discrete goods and services. All these organizational activities are influenced by both internal and external pressures. They are subject to stresses from the outside environment which may be dynamic and turbulent (Metcalf, 1978), and they are subject to internal strains, depending on the mission of an agency, its format, and its professional, bureaucratic, or technocratic nature. Such macrostructural implementation designs and capabilities cannot be assumed. The empirical mappings of such arrangements and the interdepen- dencies among public bureaucracies and between public and private bureau- cracies are important research tasks. Conclusions about the way in which implementation capabilities are related to observed results cannot be drawn from abstract deduction but only from empirical research. Consequently, some interdependencies are probably highly stable and exclusive, while others are not. Some interdependencies may be highly dynamic as a result of changes in policy and plans that have to be applied. Coexistence among, rather than exclusiveness of, multiple actors and organizations characterizes much of the implementation in the health field. Comparative studies of planning processes in the health field and the implementation of health plans start by mapping actors, structures, and contextual environments as preconditions for generalizations. The individual country studies within the larger research project describe basic features of the political and health care systems. Then, for selected issues, those informal functional and structural macroprocesses and interdependencies will be mapped that involve decision-making on planning, financing, operating, and evaluating health care. (For detailed applications to the German case, see Altenstetter, 1 979c.) The study of social policy implementation is difficult because social services are delivered by local organizations like hospitals, clinics, and health care centers that are relatively independent of central control. Each of these two levels, central and local, has its own implementation problems, so the implementation of national policy consists of two separate classes of problems. A central government must execute its policy in order for local delivery organizations to behave in desired ways; this is the macro-implementation problem. And in response to those central government actions, the local organizations must devise and execute their own internal policies; this is the micro-implementation problem. "Essential differences between the processes of micro-implementation and macro-implementation arise from their distinct institutional settings. Whereas the institutional setting for micro-implementa- tion is a local delivery organization, the institutional setting for macro- implementation is an entire policy sector" (Berman, 1978: 164). Implementation analysis involves the study of why authoritative decisions (laws, policies, plans) do-or, more likely, do not-lead to expected results. Such analyses go beyond the content of the policy goal per se because, to paraphrase a biblical admonition, a goal ("faith") without action ("works") is dead. As Schulz and Slevin ( 1975) observe, policy, especially in the social arena, exists only as it is implemented and, therefore, the 'goodness' of most ideas cannot be tested apart from their implementation. Thus, the two relationships, the policy's technical validity and its implementation effectiveness, cannot be divorced; taken together, they constitute the domain of implementation analysis" (paraphrased by Berman, 1978: 160). In contrast to the normative theory of administrative law that continues to be widespread in Europe, theories of policy implementation have indicated that the implementation process is one of continuous reciprocal interaction between the setting of objectives and the actions geared to achieving them. According to Pressman and Wildavsky (1972), the study of implementation involves tracing the paths of decision points all the way from the national policy-making level down to the level of the service manager, who might be called upon to make the final decision. The study of implementation also involves the examination of the network of clearances at each decision point and a study of those actors involved at each decision point. Delay in implementation is viewed, in part, as a function of the number of actors involved. European research on implementation of health policies has reinforced the validity of the Pressman-Wildavsky theory (Altenstetter, 1979b). Although their conclusions dealt with implementation in a highly competitive, pluralistic, and heterogeneous political and administrative system, their theory of implementa- tion also applies when the best of all circumstances for implementation are given-when, for example, the following exist:

(1) a hierarchically structured administrative system;

(2) a uniform and codified body of legal rules and norms that are understood by civil servants in central headquarters and in field offices, and by service managers, all of whom have received the same training; and

(3) a relatively high degree of homogeneity in national, regional, and local politics.

Even under these circumstances, the delay encountered in the implementation process is directly related to the number of decision and clearance points and to the different views-to the refinements and the nuances, if you will-held by diverse actors who intervene in the implementation of a health program. The circle of actors is much wider than administrative law would suggest. Conceptual formulations based on public law and administrative law treat the governmental sector as a monolithic entity. But implementation theory shows that the governmental sector instead consists of numerous competing govern- mental segments. Bureaus and divisions are just as competitive in a highly centralized system of the "administrative state" as they are in a more open American system. And one reason there is opposition from below, from the periphery, is that field officials and field service managers have to be as attentive to the requirements of local and regional politics and politicians, and as subject to interest-group pressures, as they have to be obedient to central instruction from above.5 The conclusions to be drawn from these observations on the theory of implementation for health planning and social development should be clear. Implementation largely depends on the functioning of multiinstitutional, multiactor, and multilevel systems. Sometimes these complex systems already exist; sometimes they have to be set up by national mandate. Recent laws on health planning, for example, require such multisegmented implementation networks. The initial assumption, of course, based on normative theories of administration, is that most national programs in Europe require a unified, integrated approach to implementation. But, inevitably, a highly complex interorganizational network that includes multiple, competing actors will develop for implementing the preferred goals. Despite the location of health service managers in the health care system's lower levels, they too are part of such interorganizational and intraorganizational networks of implementation. And like the managers of other service systems, health services managers are increasingly caught in the dilemma that results from wrong assumptions made by national policy makers about the real world. For example, in most national programs, subnational and local units and delivery systems are treated as if they were identical. Also, subnational and local actors are assumed to respond uniformly to national programs and policies. As always, insufficient attention is paid to the fact that organizational structure and political contexts do make a difference for implementation. All those who have been involved in the practice of implementation know that responsible agents differ, or can differ, in their response and receptivity to national initiatives-whether they are public, quasi-public, or private actors. They also know that the particular profile of local or areawide networks of involved actors has an influence on the implementation of national policy and the performance of a health program. Finally, they know that such local and regional networks of implementation do differ from program to program, depending on methods of financing, regulations, and characteristics of clients and providers.

The subfield of intergovernmental relations explicitly recognizes that na- tional, state, and local governments emphasize different goals, even though these layers also interpenetrate in marbled fashion. They have different value orientations and traditions; they have different stakes; and they involve different mixes of actors. Therefore, there is a wide range of policies for health planning. Similarly, quasi-public and private organizations that are major actors for realizing national objectives in the subnational implementation arena are composed of multiple and interdependent organizational layers. These range from national peak associations to regional assemblies to single organizations. Each group of organizations representing the interests of providers, hospitals, sickness funds, or local agencies can rightfully be assumed to pursue relatively homogeneous and uniform goals with regard to other participants in the health system. However, when intraorganizational relations of one type of organiza- tion-such as the three-tiered organization of health insurance funds or the federated units of a medical society-are examined, the goals of the national peak associations do not necessarily coincide with those of local branches. Therefore, it is plausible to assume that one may empirically find as much diversity in goals, orientations, and traditions among different organizational layers of the same group, as we readily recognize for the public sector or for intergovernmental relations. As judged from planning documents and subsequent assessments, planning officials in all countries do not plan for implementation. They do not specify policy practices in detail. Rather, they wax eloquent on goals and preferred end- states, and assume that the means of implementation will be forthcoming. This exclusive weighting toward goal specification is understandable in liberal democracies, since the payoffs to political actors on planning boards are relatively short-term where votes are concerned. Planning is thus retained for its symbolic, rhetorical function, and is unrealistically discussed in its rational- technical mode. Also, it does not take account of the necessary implementation structures (Hjern, 1978; Hjern and Porter, 1979). Ironically, only the bureaucrats have the larger time frame within which to approach implementation. Yet, such bureaucrats are usually not entrepre- neurially farsighted, but are instead only concerned with maintaining their positions and advancing their individual careers. Indeed, attention to the hard questions of implementation-when not everyone can benefit and some costs must be imposed-is likely to jeopardize the good standing of an administrative bureaucrat as much as it would the political fortunes of an elected official. Consequently, such apolitical planning actually allows many more interests to get involved and to carve out (or defend) their own territories. A common device for implementing public programs is government-by- delegation, which spans the deliberate transfer of certain decision-making, planning, financing, and administrative responsibilities to quasi-public or private organizations by the state authority. Delegation is justified as a means to achieve better administrative efficiency or a way to tap specialized expertise, although those concerned with accountability in health services note the thin line between the delegation and the abdication of authority (Bjiorkman and Altenstetter, 1979). Illustrations of formal transfers of authority lie in the arrangements for licensing health professionals through the boards appointed by state medical societies in the United States, or the extensive delegation of rate-setting responsibilities to the German sickness funds and their organized physicians (Kasseniirztliche Vereinigung). In the FRG, corporate organizations with public law status (Korperschaft des bffentlichen Rechts) have authority to make binding decisions in those matters that are explicitly delegated to them. This legal construction conveys not only considerable authority but also serves the function of automatically placing almost any health matter on their agendas, to the exclusion of other participants. In the process of implementing programs in democratic political systems, the lines between public and private sectors, between public and private interests, and between public and private policies become increasingly blurred. These developments have occurred as a result of general societal developments, as well as particular instruments that facilitate such changes (e.g., contracting out, grants-in-aid, and regulation). In all these cases, the industries involved come to be in charge, while state authority atrophies.

This is an except from an article named "Planning and Implementation: A Comparative Perspective on Health Policy" which was published by the "International Political Science Review / Revue internationale de science politique, Vol.2, No. 1, Social Policies: A Comparison (1981), pp. 11-42" (SAGE PUBLICATIONS). The authors of the article are Christa Altenstetter and James Warner Björkman.