Implementação de Políticas Públicas

Public Administration Vol. 82 No. 2, 2004 (249262) Blackwell Publishing Ltd. 2004, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street,Malden, MA 02148, USA.

IMPLEMENTATION STUDIES: TIME FOR A REVIVAL? PERSONAL REFLECTIONS ON 20 YEARS OF IMPLEMENTATION STUDIES

SUSAN M. BARRETT

This paper presents a review of three decades of implementation studies and isconstructed in the form of a personal reflection. The paper begins with a reflectionupon the context within which the book Policy and Action was written, a time whenboth governments and policy analysts were endeavouring to systematize andimprove the public decision-making process and to place such decision-makingwithin a more strategic framework. The review ends with a discussion about howpublic policy planning has changed in the light of public services reform strategies. Itis suggested that as a result of such reforms, interest in the processes of implemen-tation have perhaps been superseded by a focus upon change management andperformance targets. It is further argued that this has resulted in the reassertion ofnormative, top-down processes of policy implementation. The paper raises pointsthat are important ones and indeed are reflected throughout all four papers in thesymposium issue. These are: (1) the very real analytical difficulties of understandingthe role of bureaucratic discretion and motivation; (2) the problem of evaluatingpolicy outcomes; and (3) the need to also focus upon micro political processes thatoccur in public services organizations. In conclusion, the paper emphasizes thecontinued importance of implementation studies and the need for policy analysts tounderstand what actually happens at policy recipient level.

INTRODUCTION

When I was invited to participate in the ESRC seminar series (ESRC 2000) Iwas excited that implementation is once more on the policy studies agenda,but somewhat daunted by the request to contribute a think piece on 20years of implementation studies. It is now some time since I was activelyinvolved in research with implementation in the title, and over 20 yearssince the publication of Policy and Action (Barrett and Fudge 1981). I foundmyself asking why there now appears to be a revival of interest in the imple-mentation of policy, and whether there are situational similarities or differ-ences between when I started to get involved in this area of policy analysisand now.

I decided to approach the task by looking back at why I originally becameinterested in implementation, and to trace key factors in my own learningand development as a basis for addressing the central seminar question is

Susan M. Barrett is among the early British Scholars in the field of implementation studies. She wasformerly Reader in Policy and Organisational studies at the School for Policy Studies, University ofBristol.

250 SUSAN M. BARRETT

Blackwell Publishing Ltd. 2004

it time for a revival?. So this is very much a personal reflection which tracesthe context and motivation for my involvement in implementation studies,outlines the way my thinking developed through different strands of research,and concludes by raising issues which seem of particular importance forfurther development in the current policy and organizational context.

Policy and Action was originally conceived by Colin Fudge and myselfas a collection of essays illustrating different aspects of, and disciplinaryperspectives on public policy implementation. It was also intended to show-case the range of research studies being carried out within the, then, newlyestablished School for Advanced Urban Studies at the University of Bristol.We had no idea that the book (Barrett and Fudge 1981) in particular theintroductory review, would become a core text in the 1980s for policy stud-ies, planning schools and even public administration. I certainly did notanticipate that the issues raised and addressed would become the focus formy own research interest for the next 20 years.

While Policy and Action became an important benchmark in my academiccareer, on reflection I realize that my interest in this area goes back a lotfurther to my earlier education and practice as a professional Town Plannerin the 1960s and 1970s. This is where I shall start.

IMPETUS AND MOTIVATION FOR IMPLEMENTATION STUDIES

The late 1960s and early 1970s could be described as a period of growingconcern about the effectiveness of public policy and governance. This con-cern was addressed by a range of initiatives to enhance the policy content ofgovernment decision making, to improve public decision-making processesand the co-ordination of policy (joined-up thinking) and to streamline man-agement structures and service delivery. These days, the 1970s tend to havea bad press, but it was a period of important innovation in the field of policystudies. For example the mid-1960s saw a shift from two-dimensional landuse plans to the concept of a strategic plan as a statement of reasoned policy,and similar policy plans were subsequently promulgated for social services,housing and transport (DHSS 1972). Policy capacity was enhanced in localgovernment by the employment of research staff to review and evaluatepolicy effectiveness and to formulate alternative courses of action, and incentral government by the creation of departmental Policy Units, theCentral Policy Review Staff (CPRS) within the Cabinet Office, and the devel-opment of Programme Analysis Review (PAR) in the early 1970s. Ideas andtechniques for linking policy objectives and resource allocation derivedfrom the American Planning Programming and Budgeting System (PPBS)spawned the development of corporate planning philosophy, aimed atimproving the connectedness of government policy and co-ordination ofservice delivery.

Similar concerns were reflected in the concurrent development ofacademic policy studies as a multi-disciplinary and applied subject ofresearch (Heclo 1972) focused around three main areas:

IMPLEMENTATION STUDIES: TIME FOR A REVIVAL? 251


1. Policy analysis: concerned with understanding and explaining thesubstance of policy content and processes of decision making;

2. Evaluative studies: concerned with understanding and assessing pol-icy outcomes as a basis for evaluating effectiveness;

3. Organizational studies: concerned with understanding the operationof political and administrative organisations as behavioural systems,and prescriptions for improving performance.

This was the world I entered as a professional Town Planner in 1965. I waseducated in concepts of planning as a rational and linear process fromsurvey and analysis to plan making. I started practice in one of the newLondon Boroughs created by London government reorganization. Thoughthe Borough now had planning powers, as yet there were no plans in place.The reality of planning in practice was more about dealing with the chaos ofmerging several smaller authorities and responding to political opportun-ism than producing plans policy tended to follow and serve to justifyaction rather than the other way around.

It was also the time of the introduction of the new concept of strategicpolicy plans, influenced by systems thinking to emphasise planning andpolicy development as a process of response to changes in the environmentcharacterized by ongoing monitoring and review of policy performance. Imoved into central government to work on methodologies for actually doingthe new kind of policy planning in particular methods of evaluating bothpolicy options and outcomes, and the information systems demanded by theconcept of continuous review. The emphasis on monitoring policy effec-tiveness spawned a wide range of evaluative studies here and in the US.Such studies tended to show that in spite of the plethora of policies andplans, performance more often than not still seemed to fall short of policyexpectations. Concern shifted from the what of policy outcomes to thewhy of perceived policy failure, and to focus on the actual process of trans-lating policy into action the process of implementation as exemplified byPressman and Wildavskys classic implementation study subtitled HowGreat Expectations in Washington are Dashed in Oakland; or Why itsAmazing that Federal Programs Work at All. . . . (1984).

For myself, my first involvement in an explicit study of implementationcame after I moved to the School for Advanced Urban Studies in 1975. Thiswas a new multidisciplinary institution established to develop continuingprofessional education and public policy research. Many of the staffappointed at that time came from a professional practice background andthe emphasis was on linking theory and practice through applied studiesand research. I was contracted by the Department of the Environment tomonitor and evaluate the implementation of the governments short-livedCommunity Land Scheme under which local authorities were granted newpowers to buy and sell land required for development. From a centralgovernment perspective the scheme was not working as anticipated; not



enough income was being generated from land sales to fund the scheme. Inaddition, local authorities were perceived as either failing to implement oras using the scheme to fund expensive bottom drawer projects. However,looking at what was actually going on in a sample of local authorities, theresearch showed that local authorities perceived the scheme as essentiallydiscretionary (that is, that they were not obliged to participate), to be usedonly when the schemes provisions fitted with existing local political prior-ities, or opportunistically as a source of funding for local projects withoutputting extra costs on to their ratepayers (Barrett 1981).

This research together with my previous experience of both local and cen-tral government raised a host of questions for me about the whole concept ofimplementation as putting policy into effect, and about the importance ofinter-organizational value perspectives in policy interpretation, and the roleof discretion in shaping outcomes. These issues were addressed further inresearch funded by the then Social Science Research Council to review theconceptualization of implementation in the central-local government rela-tionship as part of a series of linked implementation studies directed byMichael Hill. My work on these projects formed the background to thethinking in the introductory review of Policy and Action.

APPROACH TO UNDERSTANDING THE POLICY-ACTION RELATIONSHIP

Much of the existing policy studies literature at the time tended to focus onthe politics of policy making, assuming implementation as an essentially top-down administrative and hierarchical follow on process. Policy, once formu-lated and legitimated at the top or centre, is handed in to the administrativesystem for execution, and successively refined and translated into operatinginstructions as it moves down the hierarchy to operatives at the bottom ofthe pyramid. With increasing attention paid to policy effectiveness, evaluativestudies were starting to highlight the problematic of implementation,and identify key factors deemed to contribute to what was perceived asimplementation failure. These factors have been researched and explicatedby Pressman and Wildavsky 1984; Gunn 1978; Sabatier and Mazmanian 1979;Dunsire 1978; Hood 1976; Hanf and Scharpf 1978. They include:

1. Lack of clear policy objectives; leaving room for differential interpreta-tion and discretion in action;

2. Multiplicity of actors and agencies involved in implementation; prob-lems of communication and co-ordination between the links in thechain;

3. Inter- and intra-organizational value and interest differences betweenactors and agencies; problems of differing perspectives and prioritiesaffecting policy interpretations and motivation for implementation;

4. Relative autonomies among implementing agencies; limits of adminis-trative control.



The core argument of Policy and Action was to challenge the traditionalpolicy-centred view of the implementation process, and a priori assumptionsabout the existence of hierarchical relations between policy making andimplementation. It was suggested that implementation should be regardedas an integral and continuing part of the political policy process rather thanan administrative follow-on, and seen as a policy-action dialectic involvingnegotiation and bargaining between those seeking to put policy into effectand those upon whom action depends. The political processes by whichpolicy is mediated, negotiated and modified during its formulation continuein the behaviour of those involved in its implementation acting to protect orpursue their own values and interests. Policy may thus be regarded as botha statement of intent by those seeking to change or control behaviour, and anegotiated output emerging from the implementation process.

This negotiative perspective shifts analytical attention away from a focuson formal organizational hierarchies, communication and control mechan-isms, to give more emphasis to the power-interest structures and relation-ships between participating actors and agencies, and the nature ofinteractions taking place in the process, as key factors shaping the policy/implementation outcomes.

This conceptualization of implementation as negotiated order, involvingbargaining and negotiation between semi-autonomous actors pursuing orprotecting their interests, was further developed with Michael Hill as part ofthe SSRC sponsored theoretical project (Barrett and Hill 1984) and influ-enced by a number of key ideas emerging in the organizational field. Theseincluded, notably, resource-dependency/exchange theory models of inter-organizational power relations being developed by Rhodes in a parallellinked project (see, for example, Rhodes 1981), the elaboration by Elmore,drawing on Allisons decision-making models, of alternative implementa-tion process models (Elmore 1978), micro-political conceptualizations oforganizational behaviour as power bargaining and negotiation (see, forexample, Strauss 1978; Bacharach and Lawler 1980). Perhaps of key signifi-cance was Strausss work conceptualizing all social order including organ-izations as the product of negotiation rather than coercion; his research inpsychiatric units pointed to the existence of discretionary power to bargaineven in the most rule-bound of environments. This, together with the grow-ing literature on so-called street-level bureaucrats (see, for example, Lipsky1980; Prottas 1979), explored the existence and nature of discretionarypower or scope for action in organizational settings. Also explored werethe ways that such discretion was being used by front-line operatives eitherto develop coping mechanisms in the absence of clear policy rules or tonegotiate policy modification in action.

During our research we became involved with the International WorkingParty on Policy Implementation, a multi-disciplinary group of US, Scandi-navian and UK academics with a common interest in developing theory andmethodology for exploring/understanding implementation processes as a



key factor in explaining outcomes. An important aspect of the debate withinthis group was to challenge the traditional policy-centred approach toconducting implementation and evaluation studies. Elmore (1980) andHjern and Porter (1981) in particular argued that, due to the complexity ofrelationships and interactions in the implementation process, action cannotnecessarily be directly related to, or evaluated against specific policy goals.

Implementation agencies are likely at any point in time to be respondingto a wide variety of policy initiatives or environmental pressures from arange of sources. Rather than asking whether and how a particular policyhas been implemented, or comparing outcomes against original policyobjectives, which assumes a priori a causal link between the policy andoutcomes observed, implementation studies needed to start with what wasactually happening at delivery/recipient level (the bottom) and explorewhy from the bottom up. In order to identify the factors influencing actionand behaviour including the role (if any) played by policy in shapingbehaviour and outcomes. Elmore (1980) coined the term backwardmapping. He regarded this as an analytical approach for improving theimplementability of policy design. He also saw it as a methodology forconducting implementation and evaluation studies. Others, such as Smithand Cantley 1985; Lincoln and Guba 1985; Means and Smith 1988, wereinvolved in parallel developments in pluralistic evaluation methodologies.Hjern and Porter (1981) talked similarly about building, bottom-up, a pictureof the particular implementation structure, the network structure of actorsand parts of organizations as well as their relationships and interactions ininfluencing outcomes.

ISSUES IN THEORY DEVELOPMENT

In the early 1980s academic debate was polarized around the apparentlycompeting claims of so-called top-down and bottom-up approaches toconceptualization of the implementation process. In some respects thepolarization of debate was associated with differing value and disciplinaryperspectives on the role of policy in governance. The top-down model wasreflected in traditional structures of governance and public sector organ-ization, emphasizing the separation of politics and administration, andco-ordination and control through authority and hierarchy. Those espous-ing or defending the top-down model saw it as a normative ideal for puttingpolicy into effect. Policy should be made at the top, and executed byagents in compliance with policy objectives. The role of implementationstudies was to identify the causes of implementation problems or failure,and suggest ways of enhancing the likelihood of obtaining compliance withpolicy objectives, generally focused on strategies for improved communica-tion of intentions, co-ordination of the links in the chain, management ofresources and control of implementing agents.

The bottom-up categorization was a somewhat misleading label, includ-ing both alternative approaches which regarded implementation as part of a



policy-making continuum in which policy evolved or was modified in theprocess of translating intentions into action, and new methodologies forimplementation/evaluation studies. The bottom-up camp (Berman 1978;Hjern, Hanf and Porter 1978) was associated with those espousing a micro-political view of intra- and inter-organizational behaviour, and included arange of models, some emphasizing consensus building, influence andexchange processes (persuasion, positive-sum negotiation and learning),and others emphasizing conflict and the exercise of power (zero-sum negoti-ations and power bargaining) in the policy-action relationship.

Although somewhat protracted and confusing, the top-down/bottom-updebate did raise a number of important questions and issues concerning thepurpose of implementation analysis, and indeed the meaning of implemen-tation. First was the question of what implementation studies are trying todo. Are they about prescription or description? Is the purpose to designbetter policy, achieve greater control over policy outcomes, and/or to seekunderstanding and explanations of what happens in practice? Top-downapproaches could be regarded as essentially prescriptive what ought tohappen, but were seen by critics as failing to provide adequate descriptionor understanding of the complexity of interactions taking place in imple-mentation processes. Bottom-up approaches tended to focus on understand-ing and explanation on the basis that it is not possible to prescribe withoutunderstanding. From a top-down perspective, bottom-up approaches toconceptualization were criticized for failing to offer any prescriptions forpractice, or for offering prescriptions tantamount to accepting that policy asexecuted would be subverted or modified to reflect the interests of the mostpowerful upon whom action depended, and could potentially be seen assubverting the proper role of governance.

Linked to the debate on prescription versus description is the question ofwhat is meant by implementation. Is implementation about achievingconformance or performance? (Barrett 1981). Policy-centred approaches toanalysis of necessity involve comparing outcomes against a priori statementsof intent or targets. Performance is thus judged in terms of achievingconformance with policy targets and standards. In practice, so-called per-formance criteria tend to operate more as conformance criteria; often theminimum level or standard deemed to constitute satisfactory performance.

For some types of regulatory policy (for example, health and safety),conformance or compliance may be an essential objective. But much publicpolicy is couched in more permissive and discretionary terms; the objectivebeing to permit and encourage innovative courses of action within a frame-work of procedural rules. Here, output targets or performance criteria areharder to specify in advance, and, as pointed out by Williams (1971), thedemands of public accountability are likely to mean that performance in thesense of potentially risky innovation, will be tempered by tight administra-tive and procedural controls, that is, conformance which becomes an end initself for judging performance.



Interactive and negotiative models of implementation tend to seeperformance as the achievement of what is possible within a particular pol-icy implementation environment (that is, the array of actors and interests,their relative bargaining power, degree of change or value conflict involved,and so on). From this perspective, judging performance is a matter of morepluralistic and bottom-up evaluation to assess outcomes in terms of who hasgained or lost what and how has this been affected or influenced by policy.

Crucial to this view of performance is a positive attitude to discretion. Akey contemporaneous debate centred on attitudes to discretion (Bacharachand Lawlor 1980; Lipsky 1980; Prottas 1979). Social and welfare policyanalysts, with a tradition of emphasis on issues of equity and commonstandards in service delivery, tended to view discretion as something to belimited as being potentially discriminatory. For example, research on streetlevel bureaucrats raised questions about whose policy was actually beingimplemented. Thus, Lipskys thesis gave rise to considerable debate regard-ing whether discretion was desirable and necessary or whether it was anti-democratic and reflected inadequate top-down control and so acted tosubvert policy (Linder and Peters 1987). On the other hand, those in disci-plines or professions where negotiated and contractual relations with clientsand consumers were the norm, tended to see discretion as both positive andnecessary, as the space within which negotiation and bargaining of positivesum outcomes can take place.

As what has been termed the new public management (Aucoin 1990;Dawson and Dargie 1999) and its associated managerial reforms, began toimpinge on public service organizations and agencies, there was increasingfocus on organizational change processes and change management, draw-ing on ideas emerging from the private sector and business schools. Thiswas reflected in implementation studies in consideration of implementationas change management, and the relationship between the degree and natureof change, conflict with existing structures and cultures, and the nature ofthe implementation process and outcomes (Wilson 1992).

The organizational culture literature also suggested that implementationprocesses needed to take account of organizational cultures as a powerfulaspect of the status quo; policy change very often involved both organ-izational and cultural change, as for example in shifting from bureaucratic toquasi-market structures, policy is only implemented when it has becomeencultured into the normal way of doing things (Hofstede 1991; Schein1985; Deal and Kennedy 1988).

During the 1980s a so-called third generation of implementation modelsemerged, which either focused on the refinement of negotiative and learn-ing conceptualizations within different policy environments, or sought tosynthesize elements of the top-down and bottom-up approaches or focus onthe dialectics of the policy-action relationship (Sabatier 1988; Goggin et al.1990; Palumbo and Calista 1990.) In my own work I was particularlyinfluenced by the Structuration Theory developed by Giddens (1994). In his



formulation, structures or rules of the game determine the status quo of powerrelations, but since these are socially constructed they are also susceptible tochange through human agency. He identifies in particular the importance ofwhat he calls knowledgeable agents, those with insider knowledge of thesystem, to use discretionary power to either challenge or defend the statusquo according to their own values and interests. This has offered a new wayof looking at concepts of power and negotiation in implementation as thedialectic between structure and agency, which reinforces a view of perform-ance, or what happens in practice, as a function of the scope or limitations ofscope for action (rules and roles), and the use made of that scope (values andinterests).

THE 1990S: IMPLEMENTATION STUDIES OUT OF FASHION?

In parallel with this interest in implementation, during the late 1970s to theearly 1980s, there was an additional agenda in the public policy literatureconcerning financial stringency and economic efficiency. Then, gradually,the political analysts joined the economists in elucidating a thesis of theNew Right during the late 1980s. The beginning of the 1990s heralded thesynthesis of previous work and gave attention to the managerial impact ofthe New Right in terms of the New Public Management and the operational-ization of Williamsons transaction cost thesis into a more fully developedexpression of quasi markets. The literature at the end of the 1990s was influ-enced by a condition to the quasi market thesis, this condition being thenew economic sociology and the idea of relational markets. An interestingnomenclature had developed as a shorthand to describe the numerouschanges which occurred in public sector welfare since 1979, namely,reformist, rolling back the state and reinventing government. It is inter-esting to take the provenance of each of these phrases which have been sothoroughly incorporated into the literature. It is also important to under-stand the ideologies which underpin the changes if we are to understandhow they are implemented.

Within the British literature the political science explanation of theseideological changes has been explored and summarized by Hood andPollitt. In particular, while acknowledging other scholars, Hood (1991)describes four administrative megatrends (p. 3): the slowdown andreversal of government growth; privatization and quasi-privatization;increased automation (including information technology expansion) and amore international agenda which generalizes issues in public managementand policy design. Importantly, Hood acknowledges the theoretical originsof the New Right, firstly, the new institutional economics of Arrow (1963)and Niskanen (1971); secondly, ideas of managerialism as defined by a UStype of business model which has been transported to public sector settings.The latter is exemplified by the influence on policy makers of Osborne andGaeblers work Reinventing Government (1992). Other writers, such asThompson (1992), have also characterized themes within public sector



reforms; she presents seven: privatization, delegation, competition, enterprise,deregulation, service quality and curtailment of trade union powers.

In the world of policy studies, business management language and ideasreplaced the discourse of public administration. Planning and policy-makingbecame imbued with concepts of strategic management, and concerns aboutthe process of implementation were superseded by emphasis on changemanagement and performance targets. The shift in ideological priorities wasreflected in the structure of public research funding and in research fundingopportunities. To a certain extent it can be argued that the apparent demiseof implementation studies represented no more than academic opportunism;using different language and labels for the same issues.

At the same time, public service agencies and managers had little choicebut to embrace the radical shifts that had occurred in policy direction. Theshake up of public service organizations involving the introduction of quasi-market competition, contracting out of professional services and performanceorientation in performance-related pay and short term contracts generatedthe search for new models of public management seeking to reconcileconcepts of public service with private business principles. The combinationof increased policy centralization and agency decentralization and contractingout reinforced the separation and distance between politics and administra-tion. This in turn served to reassert the dominance of normative, top-down,coercive process models of policy implementation or performance, and ofperformance as conformance with policy targets.

In this new policy construction there was perhaps less perceived need forstudies of implementation since there was a belief that the reforms in thepublic services associated with the New Public Management had addressedthe key problems of implementation failure which include a lack of clearunambiguous policy objectives, resource availability and control overimplementing agencies.

For example, in respect of clear and unambiguous policy objectives, therewas now an increased emphasis on specific performance targets and stand-ards. Formal contracts for service provision appeared to leave no doubtabout what was expected to be achieved and what would be regarded assatisfactory performance. In respect of resource availability, privatization,marketization and public/private partnership initiatives were aimed at bothreducing the public cost of service provision and injecting new resourcesfrom the private sector. Finally, as far as increased control over the discretion-ary power of implementing agencies was concerned, measures included: thecurtailing of professional autonomies by bringing professionals into line-management accountability structures with managerial accountability forperformance, replacing inter-agency power bargaining with market-basedcompetition, and contracting out.

Managers were now responsible for putting policy into effect and alsoto blame if things went wrong. Success or failure was judged on the basisof meeting pre-set targets for ensuring delivery on policy targets. From



this perspective there was little sympathy or concern for reaching agreementsor making compromises between competing or autonomous interests, orindeed in research aimed at understanding what was going on in theprocess.

The all-pervasiveness and cultural hegemony attained by new managerial-ism has also resulted in the suppression of dissent and policy challenge. It isdifficult to challenge without appearing to be against improving perform-ance and effectiveness. More insidious than the explicit gagging clausesintroduced into contracts of employment have been the self-censorshipeffects of this cultural hegemony: compliance, caution and risk avoidance inthe interests of survival. This can be seen too in academia, in the prioritiesaccorded to research topics and methodologies, and in a degree of retrench-ment into the safety of established disciplines and away from multidiscipli-nary working, in response to research selectivity categories and competitionfor funding.

It could be argued that the ideological and service reforms that haveoccurred in the public services over the past three decades have taught thosewho study the public policy process a number of lessons that in turn can beboth seen through the lens of implementation studies and at the same timecause that lens to refract in different ways. An important lesson has been theconceptualization of a new role for the state as a consequence of contempo-rary changes in the public sector. These vary from the managerial state, orneo-Tayloristic state (Pollitt et al. 1990); the contract state (Stewart 1993);and entrepreneurial government (Osborne and Gaebler 1992), but perhapsthe most graphic descriptions came from Hood (1995) and his HeadlessChicken State or Gridlocked Contract State.

In short, it became possible to see the state as an enabler rather than aprovider, government has been depoliticized and re-cast as business andneo-management. Equally importantly, was the lesson of public organiza-tions becoming de-coupled from the relationship between service deliveryand political control, well exemplified by the proliferation of agencies asone of the structures through which to deliver public services.

All of these descriptions of a new role for the state have considerableimplications for the implementation of policy. The Headless Chicken stateis one wherein there are no clear rules of the road or demarcation ofresponsibilities (Hood 1995, p. 112). Similarly, a disaggregated or de-coupledstate meant that the important collaborative and network links needed forimplementation was threatened. Agency proliferation without the cohesionand co-ordination of hierarchy resulted in fragmentation and the loss ofimportant implementation feedback loops in both the vertical and horizontal(Holmes 1992).

The very processes of policy implementation are themselves deeply polit-ically dependent, having both a macro and micro political context. Given theargument that there has been a changing role for the state, then the import-ance of context in policy implementation becomes even more important. It is



in this context that the globalisation and potential transnational characterof future public services could have a great impact upon implementation.Hood (1995) points out that the global paradigm of the New Public Manage-ment ignored the very different and typically path dependent local polit-ical agendas of public management (p. 106).

TIME FOR A REVIVAL?

The New Labour administration seems committed to the pattern of publicservices reforms described. Yet there are growing concerns, emerging fromaudit and evaluation processes, about the unintended consequences of thedirigiste model of policy implementation embedded within current prac-tices. Amongst these unintended consequences are a top down coercivepressure to meet prescribed targets that has led to the skewing of servicepriorities, for example, hospital waiting lists, and even the manipulation offigures for fear of the consequences of failure (performance becomes con-formance). There is then the lack of recognition of the complexity, time andresources involved in achieving the organizational capacity to achieveeffective change (implementation as the enculturation of change). Finally,we are also able to identify, a continuing tension, between the normativeexpectations of managerial control of the policy implementation processand the experienced reality of inter- and intra-organizational micro-politicsin the policy-action relationship (characterized by multiple negotiationsbetween semi-autonomous agents with often-competing interests anddivergent values).

There is a certain sense of dj vu in these issues relating to the efficacy ofthe top-down managerial model for implementing policy innovation andorganizational change. Managerialism sought to address the perceivedproblems of administrative bureaucracy, but over-emphasis on coercion andconformance has resulted in a lack of attention to the dynamics of organiza-tional process and the dialectic between structure and agency in the processof change.

I would thus argue the need for a revival of interest in implementationstudies. First, there is more than ever a need to invest in studies of imple-mentation and change processes, both conceptual and empirical; studiesaimed at both understanding and explaining the dynamics of the policy-action relationship, and seeking to develop more appropriate prescriptionsthan currently demonstrated by the experience of managerialism for thenegotiation of change. There is still a lack of attention to process in govern-ance theory and practice, in particular explicit attention to the appropriate-ness of differing conceptualizations of the policy-action relationship todesired outcomes (means and ends). What are the relative benefits andfeasibility of negotiation/learning strategies versus more coercive strategiesin differing policy environments? The understanding of process is also anessential part of capacity building, addressing questions such as: Is thisdoable? How might it work? What would it take?



Second, there is a need for renewed emphasis on multi-disciplinary workingin policy studies. No one discipline can claim to be the exclusive home forpolicy studies, and there are substantial benefits for theory development insynthesizing ideas from a plurality of disciplines addressing similar issuesfrom different perspectives (for example, the bringing together of ideasdrawn from recent developments in evaluation, or change literature in busi-ness management, with the literature on implementation). Multi-disciplinaryworking also helps to develop a common language for inter-disciplinarydiscourse.

Third is the renewed need to address the central paradox of control andautonomy in achieving desired performance/outcomes. How to balance therequirement for public accountability with consumer responsiveness, respectfor difference and local autonomies, creativity, and so on? How to avoidperformance becoming conformance with targets at the expense of broadergoals?

Last but not least I would argue the need for a new emphasis in researchand practice on the relationship between ethics, social responsibility, publicaccountability and control in implementation. Increased attention to ethicsand social responsibility in the policy process is overdue, and would bringinto the implementation debate issues such as the value conflict betweenprofessional principles and codes of ethical practice versus the managementperformance imperatives. There is undoubtedly a rich seam of data to bemined by implementation researchers on the impact of adopting the HumanRights Convention on employment and its consequences for policy, practiceand the delivery of public services. Finally, a much neglected area withinpublic service agencies has been the whole arena of social audit and demo-cratic accountability with the consequent attention to the role of consulta-tion, participation and advocacy in the implementation process.

REFERENCES

Allison, G. 1971. Essence of Decision. Little, Brown and Co.Arrow, I.L.J. 1963. Social Change and Individual Values. New Haven, CT: Yale University Press.Aston Business School. 2000. Implementing Public Policy: Learning from Each Other, ESRC Seminar Series,

Aston Business School.Aucoin, P. 1990. Administrative Reform in Public Management: Paradigms, Principles, Paradoxes and Pen-

dulums, Governance, 3, 3745.Bacharach, S. and E. Lawler. 1980. Power and Politics in Organisations. San Francisco, CA: Jossey Bass.Barrett, S. 1981. Local Authorities and the Community Land Scheme, in S. Barrett and C. Fudge (eds), Policy

and Action. London: Methuen.Barrett, S. and C. Fudge (eds). 1981. Policy and Action. London: Methuen.Barrett, S. and C. Fudge (eds). 1981. Understanding the Policy-action Relationship, in S. Barrett and

C. Fudge (eds), Performance and Conformance: Some Implications for Studying the Policy-actionRelationship, Paper presented to the ECPR Lancaster Joint Sessions, Planning Session, University of Lancaster.

Barrett, S. and M. Hill. 1984. Policy, Bargaining and Structure in Implementation Theory: Towards anIntegrated Perspective, Policy and Politics, 12, 3, 21940.

Berman, P. 1978. The Study of Macro and Micro Implementation. Public Policy, 27, 15784.



Dawson, S. and C. Dargie. 1999. New Public Management: an Assessment and Evaluation with SpecialReference to UK Health, Public Management, 1, 4, 45982.

Deal, T. and A. Kennedy. 1988. Corporate Cultures: The Rites and Rituals of Corporate Life. Harmondsworth: Penguin.DHSS. 1972. Town and Country Planning Act, circular 35/72.Dunleavy, P. 1994. The Globalization of Public Services Production: Can Government be Best in World?,

Public Policy and Administration,. 9, 2, Summer, 3665.Dunsire, A. 1978. Implementation in a Bureaucracy: the Execution Process: Vol. 1. Oxford: Martin Robertson.Elmore, R. 1978. Organisational Models of Social Program Implementation, Public Policy, 26, 2, 185228.Elmore, R. 1980. Backward Mapping: Implementation Research and Policy Decisions, in W. Williams et al.

(eds), Studying Implementation. New York: Chatham House.Giddens, A. 1994. The Constitution of Society. Oxford: Polity Press.Goggin, M.L., A OM. Bowman, J.P. Lester and L.J. OToole. 1990. Implementation Theory and Practice: Towards

a Third Generation. Upper Saddle River, NJ: Scott Foresman.Gunn, L. 1978. Why is Implementation so Difficult?, Management Services in Government, November.Hanf, K. and F.W. Scharpf (eds). 1978. Interorganisational Policy Making: Limits to Co-ordination and Central

Control. London: Sage.Heclo, H. 1972. Review Article: Policy Analysis, British Journal of Political Science, 2, 1 (January 1972).Hjern, B., K. Hanf and D. Porter. 1978. Local Networks of Manpower Training in the Federal Republics of

German and Sweden, in K. Hanf and F.W. Scharpf (eds), Interorganisational Policy Making. London: Sage.Hjern, B. and D.O. Porter. 1981. Implementation Structures: a New Unit of Administrative Analysis, Organ-

isation Studies, 2, 21127.Hofstede G. 1991. Cultures and Organisations. New York: McGraw Hill.Holmes, M. 1992. Public Sector Management Reforms: Convergence or Divergence?, Governance, 5, 4, 47283.Hood, C. 1991. A Public Management for All Seasons, Public Administration, 69, 1 (Spring), 319.Hood, C. 1995. Contemporary Public Management: a New Global Paradigm?, Public Policy and Administra-

tion, 10, 2 (Summer), 10417.Hood, C.C. 1976. The Limits of Administration. London: John Wiley.Lincoln, Y. and E. Guba. 1985. Naturalistic Inquiry. London: Sage.Linder, S.H. and B. Guy Peters. 1987. Relativism, Contingency, and the Definition of Success in Implementa-

tion Research, Policy Studies Review, 7, 1, 1627.Lipsky, M. 1980. Street Level Bureaucracy. New York: Russell Sage.Means, R. and R. Smith. 1988. Implementing a Pluralistic Approach to Evaluation in Health Care, Policy and

Politics, Jan.Niskanen, W.A. 1971. Bureaucracy and Representative Government. Chicago, IL: Aldine-Atherton.Osborne, D. and T. Gaebler. 1992. Reinventing Government. Reading, MA., Addison-Wesley.Palumbo, D.J. and D.J. Calista (eds). 1990. Implementation and the Policy Process. Policy Studies Association/

Greenwood Press.Pollitt, C., S. Harrison, D.J. Hunter and G. Marnoch. 1990. No Hiding Place: On the Discomforts of Research-

ing the Contemporary Policy Process, Journal of Social Policy, 19, 2, 16990.Pressman, J. and A. Wildavsky. 1984. Implementation, 3rd edn, Berkeley, CA: University of California Press.Prottas, J. 1979. People Processing. Boston, MA: D.C. Heath.Rhodes, R.A.W. 1981. Control and Power in Central-Local Government Relations. Aldershot: Gower.Sabatier, P. 1988. An Advocacy Coalition Framework of Policy Change and the Role of Policy-oriented

Learning Therein, Policy Sciences, 2, 1, 12968.Sabatier, P. and D. Mazmanian. 1979. The Conditions of Effective Implementation, Policy Analysis, Fall.Schein, E. 1985. Organisational Culture and Leadership. San Francisco, CA: Jossey Bass.Smith, G. and C. Cantley. 1985. Assessing Health Care. Buckingham: Open University Press.Stewart, J. 1993. Rational Choice Theory, Public Policy and the Liberal State, Policy Sciences, 26, 4, 31730.Strauss, A. 1978. Negotiations: Varieties, Contexts, Processes and Social Order. San Francisco, CA: Jossey Bass.Thomson, P. 1992. Public Sector Management in a Period of Radical Change: 19791992, Public Money and

Management, July-Sept.Williams, W. 1971. Social Policy Research and Analysis. New York: Elsevier.Wilson, D. 1992. A Strategy of Change: Concepts and Controversies in the Management of Change. London:

Routledge.

Documents

Implementação de Políticas Públicas