policy process, political science, implementation, Policy Analysis, University Press, International Regimes, development, London, International Relations, linear model, Public Administration and Development, World Development, dominant discourses, organisation, Policy Change, narratives, New York, policy makers, interest groups, Discourse Analysis, Grillo, R., International, International Organisation, Columbia University Press, Duke University Press, Power and influence, Grindle, Policy networks, indigenous groups, policy network, policy decisions, epistemic community, societal interests, social interests, Classes Interest Groups Parties, expatriate communities, implementing change, management literature, position power, American trade policy, discourses, political interests, communities, driving forces
1 Working Paper 118 THE POLICY PROCESS: AN OVERVIEW Rebecca Sutton August 1999 Overseas Development Institute
Portland House Stag Place London SW1E 5DP
2 The research on which this work is based was funded by the World Bank
, and completed under the guidance of Simon Maxwell, Director, Overseas Development Institute. Any questions should be directed to him (email [email protected]
). Responsibility is the author's. The author would like to acknowledge the work of James Keeley and Ian Scoones, Institute of Development Studies, University of Sussex. Printed by Chameleon Press Ltd, London SW18 4SG ISBN 0 85003 417 5 © Overseas Development Institute 1999 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publishers.
The Linear Model
Key Ideas from Five Disciplines
2.1 Political Science/Sociology
2.3 international relations
3.1 The dichotomy between policy-making and implementation
3.2 The management of change
3.3 The role of interest groups
in the policy process
3.4 Ownership of the policy process
3.5 The urge to simplify
3.6 The narrowing of policy alternatives
What makes policies happen
5 Summary The paper offers an introduction to analysis of the policy process. It identifies and describes theoretical approaches in political science, sociology, anthropology, international relations and management. It then reviews five cross-cutting themes: (a) the dichotomy between policy-making and implementation; (b) the management of change; (c) the role of interest groups in the policy process; (d) ownership of the policy process; and (e) the narrowing of policy alternatives. The paper concludes with a 21-point check-list of `what makes policy happen'. A glossary of Key Terms
is also provided. The key argument of the paper is that a `linear model' of policy-making, characterised by objective analysis of options and separation of policy from implementation, is inadequate. Instead, policy and policy implementation
are best understood as a `chaos of purposes and accidents'. A combination of concepts and tools from different disciplines can be deployed to put some order into the chaos, including policy narratives, policy communities, discourse analysis, regime theory, Change Management
, and the role of street-level bureaucrats in implementation.
6 Glossary Change agent An individual who sees change as an opportunity rather than a threat, who will be instrumental in managing change and taking it forward. The individual will give direction and momentum to the implementation of new policies and methods. Bridger (see Ambrose 1989) first developed the term. Development discourse A development discourse describes a way of thinking and outlook, a system of values and priorities that marginalises other possible ways of thinking. A discourse is a configuration of ideas which provides the threads from which ideologies are woven. Numerous discourses can be identified, for example the `scientific discourse', which sees development as a rational, technical and scientific process, grounded in Western expertise. Discourse analysis There are two meanings for this term depending on how `discourse' is defined: · When used to mean a particular way of thinking and arguing which involves the political activity of naming and classifying (as above), discourse analysis attempts to make explicit the implicit values and ideologies in discourses. It aims to depoliticise them and strip them of their value-laden terminology. · `Discourse' can also refer to dialogue, language, and conversation. If defined in this way, discourse analysis relates to the analysis of language used in policy-making. It relates, for example, to the use of labelling in policy discussions, such as `peasants', `the rural poor', or `landless'. Epistemic/policy community A group of technical experts who have access to privileged information and share and discuss ideas. Others do not have access to this information and are excluded. Individuals can be from the research community, NGOs, international organisations or a range of other organisations. Epistemic communities can have powerful influences on policy-making, some expressing certain political opinions and having links with governmental decision-makers. Escape hatches A term developed by Clay and Schaffer (1984) describing the way policy makers avoid responsibility for policies they make. Clay and Schaffer pinpoint the dichotomy between policymaking and implementation (see section three) as an avenue which can be used to this end. For example, policy makers who see implementation as a separate process to policy-making, may blame a poor policy outcome on inadequate political will or lack of sufficient resources in the implementation phase, rather than poor policy-making.
7 Other `escape hatches' include the way policies may be termed in a style and language which implies the policy being pursued is `obviously the only practical one' or `clearly the best option'. If a particular policy direction is apparently so obvious, it seems that everyone would make the same decision in the circumstances, and responsibility is therefore taken away from the policy maker. Force-field analysis A term from management literature
to conceptualise the forces interacting to oppose and support change. These forces act in opposition to each other to create a state of equilibrium in a system. Change happens when the balance point of the equilibrium shifts. For this to happen there needs to be substantial alteration to one or more of the restraining or driving forces which maintain the equilibrium. International Regime A set of principles, norms, rules, and procedures accepted by states, which help them to realise common interests. Regimes are institutions with explicit rules, agreed upon by governments that pertain to sets of issues in international relations. The concept explains how states, each acting in self-interest, can come together to work towards some basic common interests in the absence of a regulating authority, such as a `global government', which forces them to do so. Policy narrative A `story', having a beginning, middle and end, outlining a specific course of events which has gained the status of conventional wisdom
within the development arena. The `tragedy of the commons' is a policy narrative, for example, which outlines the series of events leading from overgrazing of common land by pastoralists to eventual desertification. The `wood fuel crisis' in Africa is another. Despite evidence which calls into question the validity of many narratives, they persist widely because they simplify complex development processes. They are an attempt to bring order to the complex multitude of interactions and processes which characterise development situations. Policy makers often base policy decisions on the stories outlined in development narratives. Policy narratives are distinct from discourses, which refer to a wider set of values and a way of thinking. A narrative can be part of a discourse if it describes a specific `story' which is in line with the broader set of values and priorities of a discourse. Policy network/coalition A group of individuals and organisations who share similar belief systems, codes of conduct and established patterns of behaviour. There is much debate about the distinction between policy networks and epistemic/policy communities in the political science literature: · In one definition (Rhodes, quoted in Atkinson and Coleman 1992), a policy community is defined as a stable, tightly-knit group of relationships, with more restrictive membership and greater insulation from other institutions than a policy network. In this definition, a policy network is a broader system of relationships, which are less stable and less restrictive. A policy
8 community can therefore be seen as a subset of a policy network. · In a second definition (Wilks and Wright, Coleman and Skogstad, quoted in Atkinson and Coleman 1992), policy communities are broader, a group of actors who share an interest in the same policy area and succeed in influencing policy over time. In this definition a policy network refers to the system of relationships that links the community together. The first definition will be used in this paper. Policy networks are often the mechanisms through which narratives and discourses develop and are sustained. Policy space/ room for manoeuvre The room within which a policy maker has to manoeuvre relates to the extent to which a policy maker is restricted in decision making by forces such as the opinions of a dominant epistemic community or narrative. If there are strong pressures to adopt a particular strategy a decision maker may not have much room to consider a wider set of options. There may be times, on the other hand, when an individual has a substantial amount of leverage over the process, able to assert his or her own preferences and mould the way policy choices are considered fairly considerably. Political technology First introduced by Foucault, this term relates to the way policy is often `depoliticized', if such depoliticization is in the interest of dominant group. A political problem is removed from the realm of political discourse and recast in the neutral language of science. It is represented as objective, neutral, value-free, and often termed in legal or scientific language to emphasise this. This reflects the `technology of politics', the way various means are used to work within a political agenda. `This masking of the political under the cloak of neutrality is a key feature of modern power' (Shore and Wright 1997). Street level bureaucracy A concept developed by Lipsky (1980) to refer to the role actors who implement policy changes have to play in the process. He emphasises that such individuals are not simply cogs in the process, but rather have substantial ability to mould policy outcomes. Street level bureaucracies are schools, welfare departments, lower courts, legal service offices etc. As a result of time constraints and other practical considerations, as well as political opinion, those who work in these bureaucracies influence the practical working out of a policy to produce an outcome which may be substantially different from that originally intended by a policy maker.
9 1. The Linear Model Variously called the linear, mainstream, common-sense or rational model, this model is the most widely-held view of the way in which policy is made. It outlines policy-making as a problemsolving process which is rational, balanced, objective and analytical. In the model, decisions are made in a series of sequential phases, starting with the identification of a problem or issue, and ending with a set of activities to solve or deal with it. The phases are: · Recognising and defining the nature of the issue to be dealt with · Identifying possible courses of action to deal with the issue · Weighing up the advantages and disadvantages of each of these alternatives · Choosing the option which offers the best solution · Implementing the policy · Possibly evaluating the outcome
The Linear Model: Agenda phase
On agenda Not on
Decision phase Decision for reform Decision against
Fortify political will
Source: Grindle and Thomas (1990)
This model assumes that policy makers approach the issues rationally, going through each logical stage of the process, and carefully considering all relevant information. If policies do not achieve what they are intended to achieve, blame is often not laid on the policy itself, but rather on political or managerial failure in implementing it (Juma and Clarke 1995). Failure can be blamed on a lack of political will, poor management or shortage of resources, for example. There is much evidence to suggest that this model is far from reality. The pages that follow review how political science, sociology, anthropology, international relations and business management consider policy-making, and attempt to build a broader picture of the process.
10 2. Key Ideas from Five Disciplines 2.1 Political Science/ Sociology There has been an ongoing debate within political science on whether policy-making is a rational, linear process or a more chaotic procedure, dominated by political, practical and socio-cultural forces. Various models have been developed to explain the process. One of the most important themes discussed in sociology and political science is that of development `narratives'. These are stories, which simplify complex development situations, often used by policy makers to guide their decision-making. They often develop the status of conventional wisdom. Another important theme within the political science and sociological literature is the importance of interest groups, power and authority. This is discussed in more detail in section four on `the role of interest groups'. These disciplines also consider the importance of development discourses, which are discussed further in the anthropological section. Different models of the policy process The incrementalist model Policy makers look at a small number of alternatives for dealing with a problem and tend to choose options that differ only marginally from existing policy. For each alternative, only the most important consequences are considered. There is no optimal policy decision - a good policy is one that all participants agree on rather than what is best to solve a problem. Incremental policy-making is essentially remedial, it focuses on small changes to existing policies rather than dramatic fundamental changes. What is feasible politically is only marginally different from the policies that exist, drastically different policies fall beyond the pale. In this model, policy-making is also serial, you have to keep coming back to problems as mistakes become apparent and are corrected, and new approaches to the issues are developed. The model suggests that major changes occur through a series of small steps, each of which does not fundamentally `rock the boat'. The `policy process is one of disjointed incrementalism or muddling through' (Lindblom 1980). The mixed-scanning model This covers the middle ground between the rational (or linear) and incrementalist models (Walt, 1994). It essentially divides decisions into a macro (fundamental) and micro (small) classification. It involves the policy maker in taking a broad view of the field of policy. The rational/ linear model implies an exhaustive consideration of all possible options in detail, and the incrementalist approach suggests looking only at options which from previous experience are known to exist. In contrast, a mixed-scanning approach suggests taking a broad view of possible options and looking further into those which require a more in-depth examination. Policy as arguments Juma and Clarke (1995) describe this approach as one in which policy reforms are presented as reasoned arguments. Policy is developed through debate between state and societal actors. Participants present claims and justifications which others review critically. Language not only depicts reality in such arguments, but also shapes the issues at hand in these debates. It is a means of communication of ideas, but also serves to reflect certain political stances, moulding social reality according to outlook and ideology.
11 Policy as social experiment This sees social change
as a process of trial and error, which involves successive hypotheses being tested against reality in an experimental manner. It is based in the experimental approach of the natural sciences. Policy as interactive learning This approach is rooted in a criticism of development policy as being `top-down', not generated from the communities in which polices are implemented. It argues for an `actor-perspective', emphasising the need to take into account the opinions of individuals, agencies and social groups that have a stake in how a system evolves. The approach promotes an interaction and sharing of ideas between those who make policy and those who are influenced most directly by the outcome. The advocacy of participatory rural appraisal methods by Chambers (1983) is an example of this. Development narratives A development narrative is a `story', having a beginning, middle and end, outlining a specific course of events which have gained the status of conventional or received wisdom within the development field (see glossary). The `tragedy of the commons' is a narrative, for example, which outlines the series of events leading from overgrazing of common land by pastoralists to eventual desertification. The African `wood fuel crisis' is another and Roe (1991) talks of an `except-Africa' narrative, a system of thought which indicates that development works `except in Africa'. Roe also refers to `crisis' narratives, which he says abound in Africa, framing situations as being in urgent need of action, with dire consequences if action is not taken. An example is the `Doomsday scenario' of rising birth rates, over-utilisation of scarce resources, population flows from the countryside to cities, and the increased unemployment, poor hygiene conditions and poverty that result from this (Roe 1991). Why narratives develop Narratives are an attempt to bring order to the complex multitude of interactions and processes which characterise development situations. They function to simplify situations, to bring clarity. `Rural development is a genuinely uncertain activity, and one of the principal ways practitioners, bureaucrats and policy makers articulate and make sense of this uncertainty is to tell stories or scenarios that simplify the ambiguity' (Roe 1991). The effect narratives have Narratives reduce the `room for manoeuvre' or `policy space' of policy makers, that is, their ability to think about new alternatives or different approaches. This is discussed further in section two, `the narrowing of policy alternatives'. How narratives function Narratives are transmitted through policy networks and communities (discussed below). Narratives develop their own `cultural paradigm' as they become influential: that is, certain types of development programmes, methods of data collection and analysis
become associated with particular narratives. The problems with narratives Narratives are criticised because it is believed they cause `blueprint' development, that is, a prescribed set of solutions to an issue used at times and in places where it may not be applicable. Narratives serve the interests of certain groups, usually the epistemic communities or policy networks that sustain them; and help to transfer ownership of the development process to members of these epistemic communities. They often serve to reduce the role and perceived expertise of
12 indigenous groups
, providing justification for the role of experts and outsiders in the policy process. (Leach and Mearns 1996, Clay and Schaffer 1984, Roe 1991, 1995). This is discussed further in section 3, `the role of interest groups'. `It is possible to show that the interests of various actors in development - government agents, officials of donor organisations, and independent `experts' - are served by the perpetuation of orthodox views, particularly those regarding the destructive role of local inhabitants', (Leach and Mearns 1996). Why narratives are so pervasive Despite debate about the way in which narratives are misleading or too simple, they are still widely used in development policy-making. Leach and Mearns (1996) shed light on why this is. They argue that what they call `received wisdom' (what we are calling `narratives') is tenacious because it is embedded in particular institutional structures or actor-network groups, and has strong cultural and historical roots. For example, narratives are sustained by the pervasive and persistent scientific theories on which they are based. Narratives of overgrazing and desertification and the tragedy of the commons, for example, are rooted in the ecological belief that environmental change occurs when there is a departure from ecological ideals and the concept of a carrying capacity. They are also pervasive because of the continued use of a particular methodological approach, which comes alongside such theories. `It can be the inheritance of methods, as well as the actual messages they generate, which explains the persistence of some received ideas' (Leach and Mearns 1996). Some narratives are embedded in socio-cultural roots, which sustain them. Leach and Mearns show (1996) how policies adopted in Kenya in colonial times served the interests of expatriate communities, and were adopted on independence to serve the interests of the ruling authorities that took over power. Policy networks and communities A policy network is a group of individuals and organisations that share similar belief systems, codes of conduct and established patterns of behaviour. It is an open and flexible system of relationships. A policy or epistemic community is a more tightly-knit group of elite experts who have access to certain information and knowledge, which excludes those who do not have such access (see glossary). Policy networks and communities are important concepts which have been useful for developing understanding about the role of interest groups in the policy process. They provide a mechanism whereby narratives and political interests can be brought together in policy development. Epistemic communities may express strong opinions about the way policy decisions should be made, and if politicians agree with these positions, they may invite the experts into the circles of power, providing an opportunity for such communities to have a substantial influence on the policy process. Network analysis Sociologists analyse the composition of epistemic communities, and their degree of integration. The most integrated networks are the so-called `iron triangle coalitions'. These evolved in the USA in the 1960's where congressional sub-committees, interest associations and government bureau's enjoyed mutually supportive relationships. This was a closed process, involving the bargaining of political and economic resources between actors. Other communities are less integrated, the degree of integration depends on the policy area. Sociologists dissect networks to find who the dominant actors are and how networks differ between policy areas. For example, it has been found that networks in the USA which deal with energy
13 issues are dominated by business groups
, while those relating to health policy are dominated by professional bodies. Networks have also been identified as being corporatist, state-directed, collaborative and pluralist. 2.2 Anthropology An important theme in the anthropology debate about the policy process is that of `development discourses', (other disciplines also focus on these, including political science and sociology). A discourse is an ensemble of ideas, concepts and categories through which meaning is given to phenomena. Discourses shape certain problems, distinguishing some aspects of a situation and marginalising others. As dominant discourses set out ways of classifying people and defining problems, they have serious material consequences on the process of policy-making. Anthropologists also work on the analysis of language in policy discussions and statements. This sheds light on the way in which policies are depoliticized and rationalised, taking responsibility away from policy makers for their decisions (Apthorpe 1986). Development discourses A development discourse is a particular way of thinking and arguing which involves the political activity of naming and classifying, and which excludes other ways of thinking. As examples, Grillo (1997) outlines three discourses that have developed over time: · The discourse of the `state-engendered order' in development, involving the intervention of experts located in UN agencies and embodied in multilateral and bilateral aid agencies. This, says Grillo, was the discourse of the post World War II era which he calls `elaborated, authoritative, interventionist ideology'. · The `market-engendered spontaneous order' discourse, developed in the 1970's, which was institutionally located in the IMF and the World Bank. · The `discourse of the public sphere', found in NGOs, research institutes and charities. It is constructed on the `optimistic', reason-informed pursuit of formal and substantive democracy' (1997). The effect of discourses on the policy process Discourses function to simplify complex development problems (see section three, `the urge to simplify'). They also serve the interests of some groups over others. The dominant interests they support define the issue about which policy is made, provide the framework in which alternatives are considered, influence the options which are chosen and impact on the process of implementation. Shore and Wright comment specifically on the initial stages of agenda-setting and the identification of alternative courses of action, `a key concern is who has the `power to define': dominant discourses work by setting up the terms of reference by disallowing or marginalising alternatives', (Shore and Wright 1997). The pervasive influence of discourses on the policy process is summarised by Grillo (1997), `discourses identify appropriate and legitimate ways of practising development as well as speaking and thinking about it'. The distinction between development discourses and development narratives The concepts of development discourses and narratives are distinct, although both imply a domination of the development process by certain interests to the exclusion of others. A discourse is
14 a broader concept than a narrative. Discourses relate to modes of thought, values and fundamental approaches to issues, whereas narratives define an approach to a specific development problem. Discourse analysis Discourse analysis is important in anthropology, sociology and political science. It is an attempt to understand, break down and deconstruct discourses so that the perspective they bring to the development process can be understood. Discourse analysis helps the search for alternative approaches to the resolution of policy problems. `Always there are alternatives, some of which may remain to be considered again, even those which have been rejected on other grounds. So to deconstruct discourse about policy serves a constructive purpose' (Apthorpe 1986). There are also ambitious attempts to analyse the historical evolution of discourses (Escobar 1995) and disentangle their social structure, shedding light on whose ideas are represented. The use of language in policy-making Anthropology also looks at the use of language in the policy process. This is called `discourse analysis' but refers to a different meaning of the term `discourse', meaning conversation, dialogue, language, and speech. The labelling of groups Development planning makes repeated use of `target group' labels such as `rural poor', `peasant' or `landless', `which are at once over-determinate and under-descriptive' (Wood 1985 in Apthorpe and Gasper 1996). Such labelling `disarms' groups, simplifying the complexity of their outlook, the range of interests they represent and the diversity of their experience. The `framing' of issues to be tackled Gasper (1996) suggests that `frame' is used to relate to the way policy problems are defined, analysing specifically what is included and excluded from consideration. Hajer suggests that framing works to `distinguish some aspects of a situation rather than others' (1993, in Apthorpe and Gasper 1996). `Policy discourse analysis must examine the framing of problems to be tackled, and its connections to the generation of answers offered' (Apthorpe and Gasper 1996). Making policy solutions seem obvious and unquestionable Apthorpe (1996) draws out another important aspect of the use of language in policy-making. He analyses written policy documents and emphasises the way policy is couched in terms of a Problem Solving
activity for which there are obvious solutions. He discusses the way in which documents set out clearly what `inescapably ought to be done', what `stands to reason', and can not be negotiated or bargained for. Policy which claims to be exemplary in some way `is presented in language chosen mainly to attract and persuade one of this. It normally neither invites nor accepts refutation, especially when it takes a high moral posture; rather, by every trick and trope in the book, its hallmark is non-refutability' (Apthorpe and Gasper 1996). Depoliticising policy decisions Foucault first used the term `political technology' to relate to the way an essentially political problem is removed from the realm of political discourse and recast in the neutral language of science. Policy is set out as objective, neutral, value-free, and is often termed in legal or scientific language, which emphasises its rationality. In this way, the political nature of the policy is hidden by the use of technical language, which emphasises rationality and objectivity. `This masking of the political under the cloak of neutrality is a key feature of modern power' (Shore and Wright 1997). An important effect of the simplification and depoliticisation of the policy process is that it creates a distance between policy makers and those affected by policy. This creates a mechanism whereby
15 policy makers are absolved from responsibility for the outcomes of a policy decision. 2.3 International Relations Much of the international relations literature looks at the issue of maintaining coherence and cooperation between groups in policy-making in the absence of an organising global authority, such as a global government. The concept of international regimes explains how norms, rules and procedures are developed to provide ways in which nation states can co-operate. Analysis of the nature of interest groups and tools, which attempt to model how they interact also provide some useful insights into the policy process. International regimes There has been much debate over the last decade on the importance of regimes in international relations. `Regimes can be defined as sets of implicit or explicit principles, norms, rules, and decision-making procedures around which actors' expectations converge in a given area of international relations. Principles are beliefs of fact, causation and rectitude. Norms are standards of behaviours defined in terms of rights and obligations. Rules are specific prescriptions or proscriptions for action. Decision-making procedures are prevailing practices for making and implementing collective choice', (Krasner 1983). There are, for example, international regimes for the prevention of nuclear proliferation, regimes which provide rules governing trade behaviour e.g. WTO, the environment, and third world debt. Regimes function to enhance the ability of nation states to co-operate on a given issue-area. A variety of norms serve to guide the behaviour of regime members in such a way as to produce collective outcomes which are in harmony with the goals and shared convictions specified in the regime principles. A host of more specific rules convert the regime norms into concrete prescriptions of behaviour. In many cases, regimes are accompanied by organisations designed or employed to support them in various ways. The power-based, interest-based and knowledge-based approaches Theorists from the schools of thought
of neo-realism, institutionalism, and cognitive approaches have provided power-based, interest-based and knowledge-based explanations of regimes respectively. Interest-based (or neoliberal) theories of regimes represent the mainstream approach. They emphasise the role of international regimes in helping states to realize common interests. In so doing, they portray states as rational egoists who care only for their own gains. The power-based approach argues that in the absence of a centralized authority, a hegemon, or the strongest country, shapes collective behaviour. Regimes are created by a dominant country, which leads Other Countries
to accept a regime it prefers. Knowledge-based theories of regimes emphasise the importance of normative and causal beliefs that decision-makers hold. They suggest that changes in belief systems can trigger changes in policy. Cognitivists argue that analysis of the way knowledge is distributed is important, as it constitutes identities, and shapes the preferences as well as the perceived options, of state actors. Epistemic communities and international regimes An important element of knowledge-based theories is the role played by epistemic communities in
16 international regimes. It is argued that epistemic communities can influence four stages of the policy process: policy innovation, policy diffusion, policy selection and policy persistence (Hasenclever et al 1997). Policy makers are more likely to heed the opinions of an epistemic community if they represent a consensus of opinion. Haas (in Rittberger 1995) looks at the role of an epistemic community in the regime relating to efforts of Mediterranean countries to protect the Mediterranean Sea from pollution. He states that the power-based and institutional-based theories accurately capture brief periods of the Med Plan, but fail to explain broader developments over time. An understanding of the knowledge controlled and transmitted by the epistemic community is important to understand the change over time in the regime's substance, strength, and effectiveness. Interest groups The distinction between `high politics' and `low politics' These terms are used to provide an insight into the groups that interact to make certain types of policy. The term `high politics' is used to relate to important policy decisions such as whether a currency should be devalued, or decisions to use military force. In these cases, the policy-making process is closed, with only a small group of influential people consulted. In issues of low politics, which are of less importance to nation states, a wider set of groups are considered, incorporating various societal bodies. A widening balance of interests A key theme in international relations is the way in which policy decisions are made in new ways in the context of the rise of global communication
s. Hocking and Smith (1997) outline the now widely-recognised trend towards the increasing importance of non-state actors in the policy-making arena, such as international organisations, research groups and NGOs. `While not denying the significance of the state, its agencies and representatives.... [they must be set] alongside a diversity of public and private actors who are increasingly able to mobilise at both domestic and international levels for the attainment of political goals, whether in support of or in opposition to governments' (Hocking and Smith 1997). This creates an added pressure on governments in the policy-making process, but also represents new resources for action. These groups have a wealth of knowledge and information, to which national government
s aim to obtain access. Such trends have resulted in a decline of one-to-one negotiations between national governments, and an increase in the management of groups and interests by governments who seek to acquire access to the resources provided by non-state actors. There is also a growing `international-domestic nexus'. This relates to the increasing importance being attributed to foreign affairs, by domestic actors, and the merging of the international and domestic boundaries in policy-making. Modelling the interaction of interest groups `Game theory' developed in the field of economics. It has been used in international relations to understand the interaction between actors who approach an issue with different vested interests, but must come to an agreement that is workable for all sides. It helps to explain how different interest groups (e.g. nation states) can develop policies together (e.g. in international negotiations) which are based on co-operation in the absence of an independent authority. These actors are assumed to have a set of preferences and interests, to share a common knowledge, and to act rationally. Models have been developed which are applicable to varying situations. The approach has been used, for
17 example, by Nelson 1992, to look at the foundations of power and interaction in American trade policy. 2.4 Management The business management literature sheds light on the complexity of the implementation of change. This provides a useful perspective in the light of the linear model, which implies that once a policy decision is made, implementation of the decision happens automatically. A number of particularly useful issues are considered in the literature, including the barriers to change, the skills required to manage change, the importance of power and influence, the value of an open-system approach to organisational function, and various models of the natural phases of change which companies go through. Barriers to change Force-field analysis This is a tool that has been used for many decades within management, and continues to be taught in Business schools
. It starts from the belief that in a given moment an organisation is in a state of equilibrium. This equilibrium results from the balance between driving forces that push for change and restraining forces that act against change. In order to make change happen, the balance of these forces must be altered so that the equilibrium moves. There either needs to be an increase in the driving forces, or a decrease in the resistant forces. The strength of forces:
18 Reactions to change People react against change for a wide range of reasons, including fear of the unknown, lack of information, threat to status, there being no perceived benefits, fear of failure, low trust in the organisation, strong peer groups
norms and being bound by custom. The following diagram shows four stages which people go through when facing change. These range from negative reactions, which see change as a threat, to positive reactions, which see it as an opportunity. The transition stages: The skills required of managers Much of the management literature focuses on the skills managers require in order to deal with the unpredictability, resistance, practical difficulties and disruption to personnel arrangements that change can bring about. Skills for Managing Change (Leigh 1988):
19 Power and Influence An important element in a manager's success in implementing change is his/her power and ability to influence others. `Power and influence make up the fine texture of organisations, and indeed of all interactions ... Organisations can be looked at as a fine weave of influence patterns whereby individuals or groups seek to influence others to think or act in particular ways', (Handy 1973). `Anyone contemplating, or involved in, a process of influence needs to reflect upon his source of power, and thence the range of methods of influence that it suggests', (Handy 1973). The possible sources of individual power which give one the ability to influence others are position power, expert power and personal power (Handy 1973). Position power has been called `legal' or `legitimate' power. It is the power that comes as a result of a role or position in an organisation, which conveys control over the assets of information, the right of access and the right to organise. Expert power is vested in someone because of his/her acknowledged expertise. It can only be given by those over whom it will be exercised and as a result is the most socially acceptable source of power and the most sought-after. Personal power, sometimes called charisma, sometimes popularity, resides in a person and their personality. These bases of power allow one to use one or more methods of influence, which can be divided into the overt and the unseen. Overt sources of influence include physical force, exchange (negotiating, bargaining, or bribing), rules and procedure, and persuasion (the power of argument and the evidence of facts). Unseen sources include `ecology' (the relationship of individuals to the physical, psychological and sociological environment), and `magnetism' (the invisible but felt pull of a stronger force, of personal power). An open system approach An open system approach to the way an organisation functions helps managers to know where to focus limited energy to bring about change. Based on the metaphor of an ecosystem, this model encourages awareness of the linkages between departments in an organisation, the way an organisation responds to outside stimuli, and how it is reliant on the inputs it receives from the external environment. The total energy in the system is finite, so any attempt to produce change must involve a re-balancing of priorities.
20 The Organisation as an Open System: Models of change A development model Linking together the ideas of Blake, Lievegoed, Greiner and Sadler and Barry, Plant (1995) states that it is possible to see three phases organisations move through in the normal course of their development. These are the autocratic, bureaucratic and democratic phases. The essential difference between them lies in the use of power and authority. The autocratic phase is the `start-up' phase, led by one individual with most control, and entrepreneurial skills. The bureaucratic phase sees power spread more evenly. The most distinctive feature of this phase is the way activities have a set of rules by which they should be carried out, all is spelled out clearly. The final phase is of horizontal organisation, where influence is assigned meritocratically, a structure that promotes collaboration and is highly complex. A crisis model Healthy and successful organisations undergo change as a natural part of their development, sometimes called the `pioneer phase', the `phase of differentiation' and the `phase of integration'. The transition between phases is marked by a period of crisis, a different crisis being the cause in each stage.
21 The Five Phases of Growth and Crisis:
22 3. Cross-cutting Themes 3.1 The dichotomy between policy-making and implementation There is a tendency to split policy-making and implementation in the linear model. There is a notion of a `divided, dichotomous and linear sequence from policy to implementation', (Clay and Schaffer 1984). In general, the divorce between decision-making and implementation can be `ascribed to decision makers' sense that politics surrounds decision-making activities while implementation is an administrative activity', (Grindle and Thomas 1990). This is a major flaw in the linear model because policies often change as they move through bureaucracies to the local level where they are implemented. `Implementation always makes or changes policy to some degree', (Lindblom 1980). `Policy implementers interact with policymakers by adapting new policies, co-opting the embodied project designs or simply ignoring new policies, hence underscoring the fact that implementers are crucial actors whose actions determine the success or failure of policy initiatives', (Juma and Clarke 1985). Brickenhoff states that `it is important to ..... develop both a wider and better understanding of implementation factors and the processes linking policy goals to outcomes' (1996). How does policy implementation change a policy that is made? The interactive model (Grindle and Thomas 1991) Grindle and Thomas are against the idea that `the decision to reform is the critical one and what follows is merely a mechanical process' (1991). This model stresses that the process of policymaking is interactive, not linear. A central element in the model is that a policy reform initiative may be altered or reversed at any stage in its life cycle by pressures and reactions from those who oppose it. `Unlike the linear model, the interactive model views policy reform as a process, one in which interested parties can exert pressure for change at many points.... Understanding the location, strength and stakes involved in these attempts to promote, alter, or reverse policy reform initiatives is central to understanding the outcomes' (1991). Street level bureaucracies (Lipsky 1980) Street level bureaucracies are schools, the police, welfare departments, lower courts and legal service offices, for example. Lipsky emphasises that actors who work in these bureaucracies have a role to play, they are not merely cogs in an automatic transfer of policy-making to outcome in practice. Due to constraints on their time, and bureaucratic procedures at the local level, Lipsky argues that field-level workers may exercise considerable flexibility in implementing instructions. Long (1992) provides technical insight into this approach, highlighting the `agency' of individual actors (their knowledge and power), which enables them to act autonomously and mould outcomes. `The concept of intervention .... is an ongoing, socially constructed and negotiated process, not simply the execution of an already-specified plan of action of expected outcomes (Long 1992). Administrative considerations The decision to set in place a new 55 mph speed limit on roads gives an example of how administrative issues can affect the outcome of a policy (quoted in Lindblom 1980). Given such a new directive, a Police Commissioner must decide: · whether to allow a 5 or 10 mph leeway over the 55mph limit, or none at all, · whether to enforce the limit on all roads or just the two lane highways which are most dangerous or
23 · whether to arrest a few violators or draw officers from other tasks in order to make a large number of arrests. Given the Police Commissioner's decisions, each patrol officer must subsequently decide whether to hold tightly or interpret loosely these decisions. It is clear from this that the process of trying to implement a policy can change it, a change that results purely from administrative factors rather than any political motivations or resistance to the policy. The consequences of the dichotomy between policy-making and implementation One of the most important effects of the division between policy-making and implementation is the possibility for policy makers to avoid responsibility. `The dichotomy between policy-making and implementation is dangerous. That is because it separates the `decision' from the `implementation' and thus opens up `escape hatches' through which policy makers can avoid responsibility (for example, the oft-heard problems of `bad implementation') (Gillespie and McNeill, 1992, discussing Clay and Schaffer 1984). 3.2 The Management of Change A major drawback of the linear model is its failure to consider the complexities of the implementation process. As outlined in the previous section, this underestimates the complexity of the implementation process and develops a situation in which the practical working out of the policy may be very different from the policy originally planned. Grindle and Thomas state that `the role of implementation in the [policy] process is substantially different from the linear model' (1990). The need to manage the implementation process Policy implementation is an ongoing, non-linear process that must be managed (Grindle and Thomas 1991). It requires consensus building, participation of key stakeholders, conflict resolution, compromise, contingency planning, resource mobilisation and adaptation. `New policies often reconfigure roles, structures, and incentives, thus changing the array of costs and benefits to implementers, direct beneficiaries, and other stakeholders. As a result, policy implementation is often very difficult. Experience has shown that an inwardly focussed, `business as usual' approach will fall short of achieving intended results' (Brinkerhoff 1996). The management of change is an important area within management literature, and also arises in political science (Grindle and Thomas 1990, Crosby 1996).
24 Managing change Plant (1995) sets out six key activities for successful implementation:
Developing a plan for change This involves data collection and analysis, reflection, developing a vision and concept building. Identifying `change agents' Bridger (in Ambrose 1989) emphasises the importance of identifying individuals who will lead change. These people, he believes, will give direction and momentum to the implementation of new policies and methods. Crosby (1996) suggests that in some situations it is difficult to identify a single individual or agency to lead the change. In such circumstances reform leadership may be embodied in special task forces, commissions or co-ordinating committees. Recognising barriers to change It is important to predict the reaction of individuals and groups to proposed changes.
Who are winners?
Who has information?
Who has power? Source: Plant (1995)
Who are losers?
The reasons for resistance to change are discussed in `management', section two. Grindle and Thomas (1990) discuss how the strength of public resistance varies according to four factors: · Location: if villages are far apart, it is harder to generate joint opposition · Organisation: groups that are already organised around common interests will resist more effectively
25 · Socio-economic group: which affects the capacity to get information quickly · Literacy: which determines the capacity to use information. Peters (1987) states that one should `seek to minimize potentially paralysing fears, despite the uncertainty which makes fearfulness legitimate'. Building support for reform Plant (1995) states that `Recognizing is one thing. Energizing people to do something is another'. The solution, according to Plant is to `communicate like you have never communicated before', explaining the need for change and the ways in which individuals will benefit. Leigh (1988) emphasises the need to generate commitment to reforms by involving people in activities early and encouraging their participation. Leigh also emphasises the importance of a broad base of support `so that when you need help it can be obtained from a number of sources' (1988). Reforming organisational structure Organisational restructuring may well be necessary to deal with a new way of working. `As new tasks are developed, new procedures will be created, responsibilities will shift, some divisions or departments will gain importance while others may even be abolished, and new patterns of internal resource allocation will emerge in accordance with the demands of the new policies' (Crosby 1996). A central problem is fitting new policy objectives to old public sector organisations. Crosby (1996) suggests that one well-used response to this is to establish task forces, which cut across the traditional organisational structure, pushing forward the policy reform agenda. Mobilising resources `In the implementation process, political, financial, managerial, and technical resources are likely to be needed to sustain reform. Mobilising these is part of the challenge to decision-makers and policy makers. Those opposing the policy change may attempt to block access to the necessary resources, thus stalling the reform and returning it to the policy agenda' (Grindle and Thomas 1991). In discussing financial resources, Crosby (1996) states, `The inability of governments to redistribute resources to new priorities is frequently the cause of program or project shutdowns once donor resources have been exhausted'. In developing countries
, it is also difficult to accumulate the required skilled human resources, with emigration, war, repression or disease causing problems in some cases. Consolidating change Cotter (1996) discusses the importance of getting the `easy, quick wins' early on in the process so that motivation for change is maintained. He also talks about consolidating the gains of the change process later, main-streaming the new way of working so that it is part of normal procedure. 3.3 The role of interest groups in the policy process The policy process is influenced by a range of interest groups that exert power and authority over policy-making. These influences affect each stage of the process from agenda setting, to the identification of alternatives, weighing up the options, choosing the most favourable and implementing it. `Policy practices are not in fact just a rational search. No truths or decisions ...... are unproblematic. A crucial aspect of all policy practice is actually and specifically what and who is included. The style of policy discourses is overwhelmingly to talk as though that were not so; but as though the data were inclusive, the processes rational and the remedy simply knowledge- or research- based (Apthorpe 1986).
26 Grindle and Thomas (1991) summarise the wide-ranging debate within political science on interest groups and the exertion of power and influence. They divide the interest groups into society-centred and state-centred. Interests in the policy formation process (Keeley 1997 adapted from Meier 1991):
Advisors' predictions and prescriptions
Society-centred forces Classes Interest Groups Parties and Voters
State-centred forces Technocrats Bureaucrats State interests
Society-centred models Class analytic models Based in the Marxist approach, these argue that the policy process is influenced by opinions that divide along class lines, with the interests of the bourgeoisie dominating the process and acting against those of other classes. Pluralist models This approach presents policy as primarily reflecting the interests of groups within society. The role of government is to provide a playing field for the expression of social interests, and to allow these to shape policy. In this model, policy change simply reflects changes in the balance of power between interest groups in society. There are concerns over the applicability of these models, however, to developing countries, where it is harder for groups to co-ordinate their activities and positions than it is in the developed world. They are also criticised for not reflecting the influence politicians have on the process. `There is a general recognition that .... images of responsive politicians and compliant bureaucrats need to be amended' (Atkinson and Coleman, 1992). State-centred models As a result of this lack of emphasis on actors within the state, there was an effort to `bring the state back in'. There are two groups of state-centred models. One group is the bureaucratic politics models that focus on conflict and negotiation between actors within the state machinery. The contests are driven by individual career incentives, and `turf wars' between Ministries trying to maintain control over policy arenas. A further important area of conflict is between the bureaucracy and the executive. Grindle and Thomas (1991) state that `players' compete over preferred options and use the resources available to them through their positions hierarchy, control over information, access to key decision makers, for example to achieve their goals'.
27 A second group is the state-interests approach. This focuses on the specific interests the state has in policy outcomes, such as the interests of regime authorities to remain in power and the maintenance of its own hegemony vis-а-vis societal actors. These interests may or may not correspond to interests of particular classes or groups in society. `The state is considerably more than an arena for societal conflict or an instrument of domination employed by the dominant class or class alliance. It is potentially a powerful actor in its own right' (Grindle and Thomas 1991). The criticism of this model is that in some cases states are weak, and are dominated by societal interests. They would not have the authority to make decisions that reflected their own interests. Epistemic communities and policy networks In reality, of course, policy is influenced by a range of the above actors (Keeley 1997). The concept of a policy network and community provides a framework that allows for this. `Both state and society actors cohere around key policy principles in policy communities ... This recognition of a coincidence of views helps avoid the state-society dichotomies that beleaguer other political science frameworks' (Keeley 1997). Fischer's analysis of think tanks as epistemic communities in the USA Fischer (1993) provides an analysis of how think tanks in the USA evolved to become discourse coalitions, playing an important role in policy-making between the 1960's - 1980's. He looks at organisations such as the Brookings Institution
, the Heritage Foundation and the American Enterprise Institute for Public Policy research. Initially acting as objective experts who offered advice to government policy makers, Fischer shows that over this period there was a move towards subjectivity, a `politicization of elite think tanks and their experts'. The Heritage Foundation was said to state `Unlike other institutions that pretend ideological neutrality, we're conservatives, no bones about it' (quoted in Fischer 1993). These experts had a significant impact on the policy process. `The experts' role in defining problems... is more than an analytical activity. It is also the ability to bring to political consciousness problems, such as poverty, that would otherwise be accorded little attention by either politicians or the public... the fact that experts tell people a problem exists sets up a `social disequilibrium' which can be translated by politicians into a political demand for compensatory action'. In this way, a number of think tanks developed to become epistemic communities, or discourse coalitions, which had a substantial influence over policy-making. 3.4 Ownership of the policy process Much of the anthropological, political science and sociological literature discuss the ways that ownership of the development policy process tends to be drawn away from local and indigenous groups to policy experts or outsiders. `Policy-making tends to become the mystique of elites. And these elites are separated from [local] people ... these mysteries and separations put policy-making processes in rural and agricultural development
into a privileged position', (Clay and Schaffer 1984). Development narratives Crisis narratives are the primary means whereby development experts and the institutions for which they work claim rights to stewardship over land and resources they do not own. `By generating and appealing to crisis narratives, technical experts and managers assert their rights as `stakeholders' in
28 the land and resources they say are under crisis' (Leach and Mearns 1996). Roe argues that outsiders tend to claim that `not only are insiders, specifically local residents, not stewarding their resources, but those who really know how to sustain those resources are outsiders, specifically development experts and professionally trained resource managers in the country concerned' (Roe 1995). Accordingly, so this argument goes, local people must be guided by the stewardship of techno-managerial elites, be they experts in host country governments, international donor agencies, or transnational NGOs. `Whether right or wrong, the claims, counter-claims and changing claims of experts serve principally to reinforce and widen the belief that what they, the experts, have to say really matters and solely by virtue of their expertise' (Roe 1995). `The power of development narratives is enhanced through the incorporation of dominant symbols, ideologies, and the real or imagined historical experience of their adherents. In this way they are culturally constructed and reflect the hegemony of Western discourse' (Hoben 1995, in Leach and Mearns 1996). Development discourses Discourses `can be taken as an example of the capture and exercise of power by some sorts of people, arguments and organisations against others through specific happenings, in particular arenas, over various periods of time' (Apthorpe 1986). `Discourses that become hegemonic, `totalising discourses', relegate the position of indigenous groups who represent a different view, to the role of resistors to the projects imposed. Little head is paid to their perceptions' (Leach and Mearns 1996). The use of labelling An important tool by which narratives and discourses `control' or marginalise the interests of indigenous groups is by labelling and categorising them. Target groups are labelled (such as `landless' or `women') and as a result can be conceived as passive objects of policy rather than as active subjects with projects and agendas of their own. This has been referred to as the `disarming of labelling'. The response Develop an actor-oriented approach One response has been to encourage an `actor-oriented' approach to development. This illuminates the cultural aspects of development, and stresses the value and importance of Indigenous knowledge
(Grillo 1997). Develop self-awareness Clay and Schaffer (1984) state that `There is an absolute need for self-awareness and self-criticism in policy-making processes .... all is to be questioned. Nothing is to be taken for granted. Nothing is innocuous.' The task of highlighting and making explicit the implicit values and belief systems which privilege the ideas of some over others, is not considered easy, however. `The privileging of policy-making in agricultural and rural development is a force to be reckoned with. The therapy and the science involved enjoy peculiarly strong constructions. Potential alternatives and challenges are remote, inconvenient, hidden and excluded' (Clay and Schaffer 1984).
29 3.5 The urge to simplify When policy makers think about alternative policy approaches they are observed to simplify issues in order to understand a situation better. This is often an attempt to develop some order out of chaos, to weed out some threads of causation from very complex situations (Roe 1991). While often necessary, the main drawback of this is that it can go too far, misrepresenting a situation and producing false information upon which decisions are based. Leach and Mearns (1996) state that `conventional wisdom obscures a plurality of other possible views and often leads to misguided or even fundamentally flawed development policy'. Development narratives `Narratives simplify complex issues to create a more compelling account likely to generate action, narratives are often based on shaky scientific claims', (Roe 1991). Policy communities take hold of these narratives, they provide an oasis of certainty, `proven' case studies
or development issues which are `understood'. Development discourses These simplify by setting up a way of thinking that helps to mould outlook and outline a course of action based on those principles. `The stabilizing assumptions of policy makers ... substitute for the rich diversity of people's historical interactions with particular environments. Even when they embrace debate, such debates often reduce the world to two dimensions in a simplified and ultimately unhelpful way' (Leach and Mearns 1996). The response One response is to carry out research to show how they simplify and where they are wrong. Another response is to develop `counter-narratives'. This is the approach endorsed by Roe (1991). He argues that policy makers tend towards simplification when making decisions, so making complex models of how narratives are wrong may not be the answer. Instead, he suggests developing counternarratives, which reverse the thinking of narratives, providing a balance to the ideas of the original narrative. `If project designers are to reject the blueprint they must have another story whose design implications are equally obvious to them'.... `Blueprint development can be improved by better manipulating the narratives on which they are based', (Roe 1991). 3.6 Narrowing down the options The linear model suggests that when policy is being made a range of options are reviewed which represent possible solutions to a problem. It implies that all possible options are considered, with an exhaustive amount of information reviewed in each case, and one alternative chosen on merit. In contrast to this, there is a range of literature, particularly from political science, which suggests that policy makers only consider a narrow range of options, not the full range that is theoretically possible.
30 Practical considerations It is unrealistic to think policy makers have the time, imagination and information required to make comprehensive predictions about the costs and benefits of each possible alternative option, the complexity of this task is too great. The incrementalist model of policy change Lindblom is the main proponent of this (see section two, `political science and sociology'). The main emphasis of the model and its relevance here is that policies tend to be only marginally different from those that have gone before. Policy makers do not consider options that would lead to radical change. This is because usually, though not always, what is feasible politically is only incrementally or marginally different from existing policies. If there is a change in policy stance, it occurs by a series of small steps rather than one radical change. Narratives and discourses By outlining a way of thinking or approach that represents and serves certain interests, narratives and discourse marginalise other interests and approaches. `The effect of narratives is to close down policy space, policy space being understood as the room to pursue different approaches to policy', (Keeley 1997). The response Clay and Schaffer state `There are always other choices in policy. The need, but also the difficulty, is to reveal them' (1984). There need to be `hard searches for alternatives within highly reiterative practices and apparently unquestionable agenda, data and strategies', (Clay and Schaffer 1984).
31 4. Conclusion What makes policies happen? Drawing on the literature from all the disciplines discussed, a list of issues which facilitate the development of policy has been put together. These are not mutually exclusive, and any one policy innovation will include some and not others. The points are intended primarily as a guide to factors which contribute the development of policy. Some are good motivations for change, others, as have been discussed, may not be. Thus, a policy innovation happens when.... · A new ground-breaking piece of research is completed which defines a problem and clarifies appropriate courses of action to remedy it. · There are good links between and within agencies whereby lessons learned from practical experience can be shared and acted upon. · A development problem is analysed in a scientific, technical way, producing tangible data that offer something concrete to act on. · A person in authority has a particular interest in a certain issue and as a result those around him/her are influenced to work on it and develop policy in that area. · Events are timed in such a way that a person who is particularly interested in pushing forward an agenda is working at a time when a powerful political authority has reason to be interested in the same agenda. · Similarly, timing is such that the publication of research work happens when a policy-making organisation is particularly interested in the issue being researched. · A situation develops which is represented in a widely accepted scenario or narrative as a `crisis', requiring rapid and dramatic action to avoid catastrophe. · There are good connections between interested parties such as aid organisations, the research community, and government (making a `network') through which ideas are exchanged and thoughts clarified about possible policy directions. · There is a dominant epistemic community, a particularly influential group that has close links with policy makers, and forces an issue on to the agenda and shapes policy-making. · There is a general consensus within an organisation or wider network (which may include the general public) that change is needed, a new policy direction is required, and that old strategies are not working as well as they could. · A development problem is turned into a `story' which simplifies it and sets out an agenda for action. · A dominant discourse or way of thinking becomes established which makes clear certain priorities, thereby simplifying a situation and providing guidance towards certain policy directions.
32 · There is a code of conduct or best practice regarding a particular issue, creating guidelines as to how to act. · An organisation and the individuals within it are open-minded and consider it important to adapt to new ideas from the external world, rather than seeing these as a threat. · An organisation fosters innovation. People are encouraged to develop new ways of doing things and are confident their ideas will be considered with an open mind by others. · There is an individual or a group of people who have an idea for a new policy direction. These `change agents' carry the idea forward, explaining it to others and building a consensus towards the new position. · There is a network of people around the `change agents' who will respond to them and help them carry the process forward. · An organisation has a sufficiently flexible organisational structure to enable the development of new groups or units, which will be effective in seeing a policy change through. · Policy-making and implementing bodies have sufficient authority to push a new policy through even if it is not widely supported. · Resources within an organisation exist, or can be gathered together, to respond to a new way of working. · There is the required motivation and energy to use and mobilise these resources to achieve the goals of a policy innovation. These points show that in reality the process of policy-making is very different from that outlined in the linear model. Policy-making must be understood as a political process as much as an analytical or problem solving one. `The policy-making process is by no means the rational activity that it is often held up to be in much of the standard literature. Indeed, the metaphors that have guided policy research over recent years suggest that it is actually rather messy, with outcomes occurring as a result of complicated political, social and institutional processes which are best described as `evolutionary'', (Juma and Clarke 1995). This is summarised by Clay and Schaffer (1984), `The whole life of policy is a chaos of purposes and accidents. It is not at all a matter of the rational implementation of the so-called decisions through selected strategies.'
33 Bibliography Atkinson, M. M. and Coleman, W. D. (1992) `Policy Networks, Policy Communities and the Problems of Governance'. Governance: An International Journal of Policy and Administration Vol. 5 No. 2. Ambrose, A. (1989) `Key Concepts of the Transitional Approach to Managing Change' in Klein, L. (ed.) Working With Organisations, Papers to Celebrate the 80th Birthday of Harold Bridger. Loxwood: Kestrel Print. Apthorpe, R. (1986) `Development policy discourse'. Public Administration and Development. Vol. 6, pp377389. Apthorpe, R. and Gasper, D. (eds) (1996) Arguing Development Policy: Frames and Discourses. Introduction: Discourse Analysis and Policy Discourse. London: Frank Cass. Axelrod, R. and Keohane, R. O. (1986) `Achieving Co-operation Under Anarchy: Strategies and Institutions' in Oye, K. A. (ed.) Co-operation Under Anarchy. Princeton: University Press. Belasco, J. A. (1992) Teaching the Elephant to Dance: Empowering Change in your Organisation. London: Century Business. Brinkerhoff, D. W. (1996) `Process Perspectives on Policy Change: Highlighting Implementation'. World Development. Vol. 24 No. 9. Clay, E. J. and Schaffer, B. B (eds.) (1986) Room for Manoeuvre, An Explanation of Public Policy in Agriculture and Rural Development. London: Heinemann. Coghlan, D. (1993) `A Person-Centred Approach to Dealing with Resistance to Change'. Leadership and Organisation Development Journal. Vol. 14 (4). Cox, R. W. (1986) `Social Forces, States, and World Orders: Beyond International Relations Theory' in Keohane, K. O. (ed.) Neorealism and its critics. Columbia University Press: New York. Crosby, B. (1996) `Policy implementation: The Organisational Challenge'. World Development Vol. 24, No. 9. Dunn, W. N. (1993) `Policy Reforms as Arguments' in Fischer F. and Forester, J. (eds) The Argumentative Turn in Policy Analysis and Planning. UCL Press: London. Escobar, A. (1995) Encountering Development: The Making and Unmaking of the Third World. Princeton: Princeton University
Press. Fischer, F. (1993) `Policy Discourse and the Politics of Washington Think Tanks' in Fisher, J. and Forester, J. (eds) The Argumentative Turn in Policy Analysis and Planning. Durham, NC: DUKE UNIVERSITY
Press. Fischer, F. and Forester, J. (eds) (1993) The Argumentative Turn in Policy Analysis and Planning. Durham, NC: Duke University Press. Grillo, R. (1997) `Discourses of Development: The View from Anthropology', introduction of Grillo, R. and Stirrat, R. L. (eds) Discourses of Development: Anthropological Perspectives. Oxford: Oxford International Publications. Grindle, M. and Thomas, J. (1990) `After the Decision: Implementing Policy Reforms in Developing Countries'. World Development. Vol. 18 (8). Grindle, M. and Thomas, J. (1991) Public Choices and Policy Change: The Political Economy of Reform in Developing Countries. Baltimore: John Hopkins University Press. Haas, P. M. (1992) `Introduction: Epistemic Communities and international policy
Co-ordination'. International Organisation. Vol. 46, No.1. Haas, P. M. (1975) `Is there a Hole in the Whole? Knowledge, Technology, Interdependence and the Construction of International Regimes'. International Organisation. 29: 827-76. Hajer, M. (1993) `Discourse Coalitions and the Institutionalisation of Practice' in Fischer, J. and Forester, J. The Argumentative Turn in Policy Analysis and Planning. Durham, NC: Duke University Press. Hajer, M. (1995) The Politics of Environmental Discourse: Ecological Modernization and the
34 Policy Process. Oxford: Clarendon. Handy, C. (1976) Understanding Organisations. Harmondsworth: Penguin. Hasenclever, A., Mayer, P. and Rittberger, V. (1997) Theories of International Relations. Cambridge Studies in International Relations. Cambridge: Cambridge University Press. Hocking, B. and Smith, M. (1997) `Beyond Foreign Economic Policy' in The United States, the Single European Market and the Changing World Economy. London: Pinter. Juma, C. and Clark, N. (1995) `Policy research in sub-Saharan Africa: An Emploration'. Public Administration and Development, Vol 15, pp121137. Keeley, J. (1997) `Conceptualising the policy process' in Reconceptualising Policy Processes, The Dynamics of Natural Resource Management and Agricultural Intensification Policy-making in Ethiopia, 198497. MPhil Dissertation, Chapter 3. Brighton: IDS, University of Sussex. Keohane, R. O. (1986) Neorealism and its Critics. New York: Columbia University Press. Keohane, R. O. (1995) `The Analysis of International Regimes: Towards a European-American research programme
' in Rittberger, V. (ed.) Regime Theory and International Relations. Oxford University Press: New York. Kotter, J. (1996) Leading Change. Boston: Harvard Business School Press. KPMG Management Consulting (1993) The Challenge of Change. London: KPMG Management Consulting. Krasner, S. (1983) `Structural Causes and Regime Consequences: Regimes as Intervening Variables' in Krasner, S. (ed.) International Regimes. Ithaca, New York: Cornell University Press. Krasner, S. (1983) `Regimes and the Limits of Realism: Regimes as Autonomous Variables' in Krasner, S. (ed.) International Regimes. Ithaca: New York: Cornell University Press. Krasner, S. (ed) (1983) International Regimes. Ithaca, New York: Cornell University Press. Leach, M. and Mearns, R. (eds) (1996) The Lie of the Land:Challenging Received Wisdom on the African Environment. Oxford: James Currey. Leigh, A. (1988) Effective Change: Twenty Ways to make it Happen. Institute of Personnel Management. Lindblom, C. E. The Policy-Making Process. Englewood Cliffs, New Jersey: Prentice Hall
. Lipsky, M. (1993) `Street-Level Bureaucracy: dilemmas of the individual in public services' in Hill, M. The Policy Process: a reader. pp38992. London: Prentice Hall/Harvester Wheatsheaf. Mayer, P., Rittberger, V. and Zurn, M. (1993) `Regime Theory: State of the Art and Perspectives' in Rittberger, V. (ed.) Regime Theory and International Relations. Oxford: Clarendon Press. Nelson, D. (1991) `Trade Policy Games' in Murphy C. and Tooze, R. (eds) The New International Political Economy. Basingstoke: Macmillan Education. Plant, R. (1987) Managing Change and Making it Stick. London: Fontana/Collins. Peters, T. (1987) Thriving on Chaos, Handbook for a Management Revolution. USA: Alfred A Knopf, Inc. Roe, E. (1991) `Development Narratives, Or Making the Best of Blueprint Development'. World Development. Vol. 19 No. 4. Roe, E. (1994) Narrative Policy Analysis, Theory and Practice. Durham; London: Duke University Press. Roe, E. (1995) `Except-Africa: Postscript to a Special Section on Development Narratives'. World Development. Vol. 23, No. 6. Rosenau, J. N. (1986) `Before Co-operation; Hegemons, Regimes and Habit-Driven Actors in World Politics'. International Organisation Vol. 40 No. 4. Shore, C. and Wright, S. (eds) (1997) Anthropology of Policy, Critical Perspectives on Governance and Power. Introduction: pp334. London: Routledge. Strange, S. (1983) `Cave! Hic Dragones: A Critique of Regime Analysis' in Krasner (ed.) International Regimes. Ithaka, New York: Cornell University Press. Gill, W. (1994) Health Policy: An Introduction to Process and Power. pp40-52, London: Zed Books
35 White, L. (1994) `Policy Analysis as Discourse'. Journal of Policy Analysis and Management. Vol. 13, No. 3. Young, O. R. (1983) `Regime Dynamics: The Rise and Fall of International Regimes' in Krasner, S. (ed.) International Regimes. Ithaka, New York: Cornell University Press.