Analysis of the Policy Process

Introduction

It was shown in Chapters One and Two that the development of the Agri-

environmental Regulation and subsequent schemes in the UK involved a series of complex political processes through which various actors sought to influence policy outcomes. Although it is argued that the Agri-environmental Regulation remains outside the 'core' of European agricultural policy, its achievement is a milestone and, as a new initiative, its subsequent implementation is almost certain to raise other issues about the nature of the policy process and the evolution of policy. In an attempt to make sense of this particular process, this chapter examines the ideas and methods that have been put forward in the academic literature to understand the policy process. The review highlights the difficulty of defining and evaluating any one approach, but concludes with ideas about policy networks and the concept of

'enrolment' to suggest how actors attempt to strengthen the linkages within policy networks to their own advantage, even if all the actors have vested interests, albeit different ones, in making the Regulation work successfully.

In the Introduction it was noted that problems associated with the outcomes of policy are related to 'gaps' that are perceived to appear between policy intent, delivery and outcome. This chapter explores the approaches that have been developed to address these 'gaps' and questions the applicability of the term 'gap' to describe the problems that arise. The concept of 'gap' is often focused around particular issues within the policy process, and is frequently framed in terms of break down or failure of

communication between specific actors. It is shown that the adoption of this approach fails to appreciate the complex exercise of negotiation, bargaining and exchange that

occurs throughout the policy process or to acknowledge that the process is not reducible to single events.

In drawing on the concepts of enrolment and the development of networks formed around certain policies, the approach outlined below attempts to move away from the idea that policy is produced according to a set process of development,

implementation and evaluation. Instead, the process is viewed as cyclical with no tightly defined boundaries or sequences connected to particular phases. It is

suggested that policy is made reactively and recursively, which requires the adoption of a more holistic approach to the study o f the policy process so as to understand the mechanisms at work.

The inclusion of a wide range of actors into the policy process marks the

development of a new style of policy making. Under these conditions, interested parties, both public and private, are drawn into partnerships. This marks a radical shift away from earlier public policy making which sought to control the process through legislation, and towards policies centred on stakeholder arrangements which depend on a 'contractual' relationship between the interested parties. This shift in the structure of governance has altered the position of some interest groups within the policy process, and has become more important as policy makers have been faced with increasingly complex issues that cut across traditional areas of policy or introduce new ones altogether. These changes have allowed interest groups to position themselves so that they can provide decision makers with information (resources) which they hope to exchange for influencing the direction of policy. Majone has noted that:

"this development is closely related to the expansion of the area [i.e. of public policy making] where science, technology and public policy meet, the

consequent growth o f "regulatory science", and the emergence of powerful movements demanding a social assessment of new technologies and greater environmental and consumer protection. Under such conditions, obtaining and co-ordinating expert opinion in fields like nuclear power, environmental protection, or occupational health and safety has proved to be a much more difficult problem than the early advocates o f government regulation could imagine" (1991, p.455).

The chapter will discuss how analysis of the policy process has moved away from the assumption that the production of ideal policy models can lead to the production of efficient and effective policy to the development of policy networks, and the roles individuals and organisations play in the policy process. This shift echoes changes advocated more generally in government where it is now accepted that policies will only be successful when interested parties are not only given the opportunity to influence the development of policy but also to have some input into its delivery and evaluation. The chapter will discuss five approaches that have been developed for policy analysis. These are;

• top-down,

• bottom-up or behavioural, • hybrid or adjoint,

• governance and the role of central-local relations, and • network analysis.

None of these perspectives alone are capable of explaining how policy is made and implemented. The conceptual framework (Figure 3.1) illustrates how elements from each of the policy analysis methods reviewed have been combined so as to provide an enhanced approach for investigating the agri-environmental policy process in

England. The examination of the different approaches for policy analysis has stressed that while each has valuable merits, none of these approaches alone is capable of explaining the key issues of this study. The conceptual framework depicts how elements from each approach can be used to develop a refined analysis of the policy process, and how the principal weaknesses identified in each can be strengthened by reference to other models. Coupling together these various approaches enables a more detailed and thorough investigation of the policy process to be developed. This has formed the approach adopted in the thesis to study the design and implementation of the AESs and the network of actors involved in the policy process.

Figure 3.1 Conceptual Framework for Understanding how Different Policy Analysis Approaches can be Combined for the Study of the Agri-environmental Policy Network in England

Policy Themes

I

1

E xam ining ‘g ap s’ between policy intent and delivery.

Top-down Main tenet

Seeks to identify ‘g ap s’ between policy intent & outcom e (C loke, 1987; G ilg & Kelly, 1997; M arsh & Rhodes, 1992a).

Principal W eakness Separates policy developm ent from im plementation. Botto Main tenet m-up policy does not hase of lion ynn, eakness Process o f f form ulation stop at the p im plem enta (Lew is & F 1979). Princioal w Links between those w ho form ulate' and im plem ent’ policy are not established.

Central-local Relations Main tenet

Stresses im portance o f negotiation between central and local actors (R hodes, 1981; Y oung, 1981).

Principal weakness Does not take account o f the lobbyists involved. > ^

G overnance M ain tenet

Policy is developed and im plem ented through a range o f actors from both the public and private sector (G oodw in & Painter, 1996; R hodes, 1997).

T he policy process is dynam ic rather than top- down.

Top-down M ain tenet

Policy is m ade through a causal chain originating from the ‘top’ (Pressm an & W ildavsky, 1973). Principal W eakness Fails to take account o f policy ’history’ or future, and the role o f im plem enters in shaping policy.

Hvbrid Model M ain tenet

Policy is m ade though a process o f interaction and negotiation between those who im plem ent’ and those w rite’ policy leading to the developm ent o f networks o f decision (B arrett & Fudge, 1981 ; H am & Hill, 1993).

Network Relationships W ithin the networks actors seeks to define

passage points’ where there is common agreem ent (Gallon,

1986) thereby allow ing the policy process to progress.

N egotiation is a key elem ent o f the policy process. Boltom -up Main tenet N egotiation between those engaged in im plem entation is essential for successful policy delivery (H jem & Porter, 1981). Principal weakness The approach concentrates on negotiation between implementers. Botto M ain tenet m-up ues to be jring the lementation ynn, 1979). akness Policy contir developed di phase of imp (Lew is & FI Princioal we Does not establish how implementers feed their experiences back to the ‘policy m akers’. Hybrid Main tenet Model fixed but a round otiate ttle, 1990). akness Policy is not focal point a which to ncg (Gloke & Li Principal we The approach does not suggest w ho is engaged in the process of negotiation. Policy Networks Main tenet Attempts to explain interactions between governm ent and interest groups (Jordan & R ichardson, 1997; Kenis & Schnieder, 1991).

G overnance Main tenet Actors from both the private and public sector are linked to all stages o f the policy process (M urdoch & M arsden,

1995; M unton, 1995). Principal weakness The approach does not consider how the interests work together.

Netw ork relationships Actors seek to enrol others so as to increase/m aintain their position in the netw ork (Gallon, 1986).

T

G hanging pattern o f the policy process.

Top-dow n. Bottom-up. Hybrid M odel and Gentral-local Relations M ain tenets

Each approach suggests that policy is developed and im plem ented through

com m and and control’ ideologies.

Principal weaknesses Public policy m aking in the UK has significantly changed since the developm ent o f these approaches.

G overnance Main tenet

Process o f governing has changed tow ards a

differentiated polity’ (Goodwin & Painter, 1996; Rhodes, 1997). Principal W eakness Does not consider how actors gain access to the policy process.

Policy N etw orks M ain tenet

Seeks to identify how actors attem pt to influence policy through the lobbying o f government and the form ation o f policy com m unities (H aas, 1992; M arsh & Rhodes, 1992b; Sabatier, 1991).

Netw ork R elationships Main tenet

W ithin the policy process actors may be drawn from and influence different levels o f the network (M unton, 1995; M urdoch & W ard, 1997; W ard & M cN icholas, 1998).

W hich organisations are involved in the policy process?

Top-dow n, Bottom-up. Hvbrid Model & Gentral-

local P Main tenets relations hes identify ental actors he policy aknesses T he approac the governm involved in process. Principal we The approaches do not consider the role o f lobby groups in the policy process.

Policy Networks Main tenet Explains interactions between government and interest groups (Haas,

1992; Jordan & R ichardson, 1987; Sabatier, 1991). Principal weakness Does not consider the inter-organisational relationships between interest groups.

T

Process o f governing has changed from control & com m and’ ideologies tow ards a ‘differentiated polity’ with actors from public and private sectors w orking together (G oodw in & Painter, 1996; Rhodes, 1997).

Netw ork Relationships Main tenet

The different actors involved provide resources for the functioning o f the policy process (Gallon, 1986; Peterson, 1995; Richardson, 1996; Sabenius, 1992). £ A <

Approach Adopted for Studying the Policy Process

T o identify ‘g ap s’ betw een policy intent and delivery it is necessary to take a holistic view o f the netw ork and the actors involved. The policy process does not proceed in a linear fashion but is reactive and recursive with actors from the national to the local levels feeding into the process.

Negotiation is an essential element o f the policy process. The actors use negotiation and consensus building to set and shape policy. The policy process is dependent on a netw ork o f actors draw n from both the private and public sector. T he actors attem pt to enrol others to their objectives and views so as to strengthen their position and to increase their influence in the policy process. The pattern o f the policy process in the UK has significantly altered. This affects the form ulation and im plem entation of public

Top-down or Sequential Approach

It is widely accepted that many policies fail to deliver stated objectives and that this represents a 'gap' between intent and final outcomes. This failure is often referred to as an 'implementation gap' (Cloke, 1987; Dunsire, 1978; Gilg and Kelly, 1997; Marsh and Rhodes, 1992a). In an attempt to explain why these problems arise and to offer solutions, a top-down or sequential approach to policy implementation was put forward (Ham and Hill, 1993; Sabatier, 1991).

Pressman and Wildavsky's book 'Implementation' published in 1973 was one of the first studies to attempt to address the 'gap' in the policy process. The authors were aware that the study of policy implementation, as opposed to policy formulation, had been overlooked and that rather than the policies themselves being the cause of failure it was often the implementation phase of the process that was defective. The authors were keen to separate policy formulation from implementation. They stressed that it was often the mechanisms of putting the policy objectives into effect that were at fault. Building on this claim. Pressman and Wildavsky suggest that:

"Policies imply theories. Whether stated explicitly or not, policies point to a chain of causation between initial conditions and future consequences.... Implementation, then, is the ability to forge subsequent links in the causal chain so as to obtain the desired results" (1973, p.xv).

Such an approach suggests that the 'causal chain' is formed by different sections of government and those involved at ground level. This view fits well with a theory of the state which presumes that the state is in a position to 'command and control' civil society and the market. The success of the policy in achieving its objectives therefore rests with the ability of the various actors within government to work together so as to avoid divisions and to develop the policy, if necessary, in ways that allow initial aims to be fulfilled. The approach presumes that the policy process is wholly or very largely dependent on government departments and agencies, with the state itself being responsible for policy outcomes. It largely ignores the involvement of other interest groups, including the consumers of the policy outcomes. W hilst such an approach may be appropriate to a state which adopts a regulatory style of government, the

move towards a more open, consultative method of policy making which has been a feature of many advanced democratic societies in the last two decades (see Rhodes,

1997) has reduced the value and relevance of this type of analysis. In addition, the approach does not place much emphasis on the process of evaluation, or specifically who decides if a particular policy has failed and the reasons that underlie this.

In an attempt to reveal the flaws in this line of thinking, Hogwood and Gunn (1984) developed ten preconditions for establishing where problems may be encountered during implementation through a command and control style of policy development: • That circumstances external to the implementing agency do not impose crippling

constraints.

• That adequate time and sufficient resources are made available to the programme. • That not only are there no constraints in terms of overall resources but also that at

each stage in the implementation process the required combination of resources is actually available.

• That the policy to be implemented is based upon a valid theory o f cause and effect.

• That the relationship between cause and effect is direct and that there are few, if any, intervening links.

• That there is a single implementing agency which need not depend upon other agencies for success or, if other agencies must be involved, that the dependency relationships are minimal in number and importance.

• That there is complete understanding of, and agreement upon, the objectives to be achieved; and that these conditions persist throughout the implementation

process.

• That in moving towards agreed objectives it is possible to specify, in complete detail and perfect sequence, the tasks to be performed by each participant. • That there is perfect communication among, and co-ordination of, those various

elements involved in the programme.

Hogwood and Gunn argue that such an approach would be extremely mechanistic and far removed from the reality of the policy process. They stress that it is not expected that the policy context will meet the pre-conditions, but these can be helpful in explaining why policies may fail and at which points failure has occurred and, in doing so, they provide a tool for the policy analyst to define certain points which require further refinement or examination. Furthermore, to formulate policy using these pre-conditions relies on a number of factors which by the very nature of policy making in a complex world are not attainable. For instance, most policies require the involvement of a wide range of interests, nor is it be expected that 'perfect obedience' would occur in a democratic society. Additionally, the conditions can only be met if each individual policy works in a vacuum removed from previous antecedents and will not form the basis of future policies.

Using this description of the top-down model assumes that the objectives to be followed have been agreed and finalised, and that clear goal statements have been produced. The decision maker has only to select the best course o f action to solve the problem. The objectives are then formalised into documents and plans which can then be followed as they move through a pre conceived process that describes the further development of the policy and its implementation (Pressman and Wildavsky, 1973; Van Meter and Van Horn, 1975). Such an approach fails to take account o f the fact that:

"decisions are not discrete events; they are part of a process that extends over time, that is affected by many variables, and that no single individual or even group of individuals can actually be identified as having made" (Palumbo and W right 1981, p.26).

Furthermore, decisions are rarely made with the luxury of an unspecified time limit which allows each option to be thoroughly investigated. Even if time is not a limiting factor, policy makers knowledge of the outcomes o f the various alternatives is necessarily incomplete, and evaluating the effect of policy involves considerable uncertainties (Ham and Hill 1993). Additionally, the 'unintended consequences' of policy or 'policy mess' that arise because o f the development of new issues and

generation of new information through time highlights the need to develop a recursive and, in some cases, flexible approach to policy making if objectives are to be met